EDITORIAL NOTE

National Center for Education Statistics

The National Center for Education Statistics (NCES) fulfills a congressional
mandate to collect and report “statistics and information showing the con-
dition and progress of education in the United States and other nations in
order to promote and accelerate the improvement of American education.”

EDUCATION STATISTICS QUARTERLY

Purpose and goals
At NCES, we are convinced that good data lead to good decisions about
education. The Education Statistics Quarterly is part of an overall effort to
make reliable data more accessible. Goals include providing a quick way to
■ identify information of interest;
■ review key facts, figures, and summary information; and
■ obtain references to detailed data and analyses.

Content
The Quarterly gives a comprehensive overview of work done across all
parts of NCES. Each issue includes short publications, summaries, and
descriptions that cover all NCES publications and data products released
during a 3-month period. To further stimulate ideas and discussion, each
issue also incorporates
■ a message from NCES on an important and timely subject in
education statistics; and
■ a featured topic of enduring importance with invited commentary.

A complete annual index of NCES publications appears in the Winter issue

(published each January). Publications in the Quarterly have been technically
reviewed for content and statistical accuracy.

General note about the data and interpretations

Many NCES publications present data that are based nonsampling errors. In the design, conduct, and
on representative samples and thus are subject to data processing of NCES surveys, efforts are made to
sampling variability. In these cases, tests for statistical minimize the effects of nonsampling errors, such as
significance take both the study design and the number item nonresponse, measurement error, data processing
of comparisons into account. NCES publications only error, and other systematic error.
discuss differences that are significant at the 95 percent
confidence level or higher. Because of variations in For complete technical details about data and meth-
study design, differences of roughly the same magnitude odology, including sample sizes, response rates, and
can be statistically significant in some cases but not in other indicators of survey quality, we encourage readers
others. In addition, results from surveys are subject to to examine the detailed reports referenced in each article.

E D U C AT I O N S TAT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , F A L L 2 0 0 1 1
TABLE OF CONTENTS

Note From NCES Overview of Public Elementary and Secondary Schools

Peggy G. Carr, Associate Commissioner, and Districts: School Year 1999–2000
Assessment Division ............................................................................. 4 Lee M. Hoffman ................................................................................. 42
Discusses the value of National Assessment of Educational Summarizes universe data on public schools and local
Progress (NAEP) state-level results and compares student education agencies (primarily school districts). Includes
progress in mathematics and reading over the past decade. national and state data on the number, type, and size
of schools and districts, as well as selected student
characteristics.
Featured Topic: NAEP 2000 Mathematics
Assessment Revenues and Expenditures for Public Elementary and
The Nation’s Report Card: Mathematics 2000 Secondary Education: School Year 1998–99
James S. Braswell, Anthony D. Lutkus, Wendy S. Grigg, Frank Johnson .................................................................................... 55
Shari L. Santapau, Brenda Tay-Lim, and Matthew Johnson ................. 7 Presents universe data on public education finances, includ-
Describes the National Assessment of Educational Progress ing national and state data on revenues by source, current
(NAEP) 2000 Mathematics Assessment, presenting results for expenditures by function, and current expenditures per pupil.
the nation and the states, for demographic subgroups, and for
students in a variety of school and home contexts. Includes Postsecondary Education
comparisons with the results of previous assessments. High School Academic Curriculum and the Persistence Path
Through College: Persistence and Transfer Behavior of
Invited Commentary: Policy Implications of Findings From
Undergraduates 3 Years After Entering 4-Year Institutions
The Nation’s Report Card: Mathematics 2000
Laura Horn and Lawrence K. Kojaku ................................................ 65
Debra Paulson, Eighth-Grade Mathematics Teacher,
Dr. Manuel Hornedo Middle School, El Paso, Texas, and Member, Examines the relationship between the rigor of students’
National Assessment Governing Board (NAGB) ................................ 18 academic curricula in high school and the likelihood that
students will persist in their pursuit of a bachelor’s degree.
Elementary and Secondary Education Bridging the Gap: Academic Preparation and Postsecondary
The Next Generation of Citizens: NAEP Civics Success of First-Generation Students
Assessments: 1988 and 1998 Edward C. Warburton, Rosio Bugarin, and Anne-Marie Nuñez ......... 73
Andrew R. Weiss, Anthony D. Lutkus, Wendy S. Grigg, Focuses on how the high school preparation of first-genera-
and Richard G. Niemi ........................................................................ 21 tion students (i.e., students whose parents did not attend
Summarizes trends in civics education and achievement over college) relates to these students’ performance and persistence
a 10-year period. Compares 1988 and 1998 data on students’ at 4-year colleges. Compares these students with their peers
civics knowledge, classroom coverage of civics topics, and whose parents graduated from college.
classroom instructional activities.
Middle Income Undergraduates: Where They Enroll and
Homeschooling in the United States: 1999 How They Pay for Their Education
Stacey Bielick, Kathryn Chandler, and Stephen P. Broughman ........... 25 Jennifer B. Presley and Suzanne B. Clery ........................................... 78
Estimates the number of homeschoolers, describes the Profiles middle income undergraduates in comparison to their
characteristics of homeschoolers, and documents parents’ lower income and higher income counterparts, examines
reasons for homeschooling as well as their reports of public where middle income undergraduates enroll by price of
school support for homeschoolers. attendance, and discusses how they pay for postsecondary
education, including the role of financial aid.
Teacher Preparation and Professional Development: 2000
Basmat Parsad, Laurie Lewis, and Elizabeth Farris .......................... 33 National Postsecondary Student Aid Study: Student
Provides a national profile on teachers’ education, participa- Financial Aid Estimates for 1999–2000
tion in formal professional development and collaboration, Andrew G. Malizio ............................................................................. 82
and feelings of preparedness for various classroom demands. Presents key data on the percentages of undergraduate,
graduate, and first-professional students receiving different
Private School Universe Survey: 1999–2000 types of financial aid and the average amounts of aid received.
Stephen P. Broughman and Lenore A. Colaciello ................................ 37
Presents universe data on private schools, teachers, and
students by school size, school level, religious orientation,
geographic region, community type, and program emphasis.

2 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
Competing Choices: Men’s and Women’s Paths After Methodology
Earning a Bachelor’s Degree Community College Transfer Rates to 4-Year Institutions
Michael S. Clune, Anne-Marie Nuñez, and Susan P. Choy ................. 88
Using Alternative Definitions of Transfer
Provides a context for understanding the paths that women Ellen M. Bradburn and David G. Hurst ........................................... 119
and men take during the first 4 years after earning their
Examines and compares community college transfer rates
bachelor’s degree and seeks to identify how decisions about
using various definitions of potential transfer students.
graduate school, employment, marriage, and parenthood are
Includes discussion of how these definitions relate to student
interrelated.
background characteristics.
Postsecondary Institutions in the United States: 1993–94 A Classification System for 2-Year Postsecondary
and 1998–99 Institutions
Patricia Q. Brown .............................................................................. 93
Ronald A. Phipps, Jessica M. Shedd, and Jamie P. Merisotis ............. 126
Presents universe data on the characteristics of postsecondary
Outlines a proposed classification system for 2-year institu-
institutions for the 1998–99 academic year. Also compares
tions that is based on cluster analysis and could provide a
1993–94 and 1998–99 tuition, required fees, and room and
framework for policy-relevant research.
board charges.

Other Publications and Funding Opportunities

Libraries
Public Libraries in the United States: Fiscal Year 1998 Other Publications
Adrienne Chute and P. Elaine Kroe .................................................... 99 Programs and Plans of the National Center for Education
Summarizes universe data on public libraries. Includes Statistics: 2001
information about public libraries’ administrative structure Celestine J. Davis (editor) ................................................................ 129
and legal basis, operating income and expenditures, staff,
collections, and services. The Nation’s Report Card: Mathematics Highlights 2000
Shari L. Santapau ............................................................................ 129
Academic Libraries: 1998
Margaret W. Cahalan and Natalie M. Justh ..................................... 102 The Nation’s Report Card: State Mathematics 2000 Reports
Summarizes universe data on libraries in 2-year and 4-year Charlotte Solomon, Laura Jerry, and Anthony Lutkus ..................... 130
degree-granting postsecondary institutions. Includes informa-
tion about services, staff, collections, and expenditures. Funding Opportunities
The AERA Grants Program ............................................... 130
Crosscutting Statistics
The NAEP Secondary Analysis Grant Program ............... 131
Educational Achievement and Black-White Inequality
Jonathan Jacobson, Cara Olsen, Jennifer King Rice,
Stephen Sweetland, and John Ralph ................................................. 105
Explores the relationship between Black-White differences in
educational achievement and Black-White differences in a
variety of educational and economic outcomes.

Features of Occupational Programs at the Secondary and

Postsecondary Education Levels
Richard P. Phelps, Basmat Parsad, Elizabeth Farris,
and Lisa Hudson .............................................................................. 114
Provides national data on occupational program offerings, the
use of skill competencies, and other occupational program
characteristics in public secondary schools and less-than-4-
year postsecondary institutions.

E D U C AT I O N S TAT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , F A L L 2 0 0 1 3
N O T E F R O M NCES
Peggy G. Carr, Associate Commissioner, Assessment Division

A Decade of Student Achievement:

State and National Profiles of Performance
For more than 30 years, the National Assessment of Educational Progress (NAEP) has
documented the achievement of America’s students. This year, NAEP will provide a
comprehensive profile of what students who were assessed in 2000 know and can do in
the key subjects of reading, mathematics, and science. National reading results at grade 4
were published in April. National science results at grades 4, 8, and 12, as well as state-
level science results at grades 4 and 8, will be published in November. Featured in this
issue of the Education Statistics Quarterly are mathematics results that were published in
August (national results at grades 4, 8, and 12 and state-level results at grades 4 and 8).
Reports of the 2000 results include comparisons with results from assessments conducted
during the 1990s.

The release of the 2000 results highlights two noteworthy points. First, the 2000 results
mark the first decade of NAEP’s unique contribution to the body of information on
student academic performance at the state level. This is an accomplishment that many
thought unachievable when 1988 legislation first authorized state NAEP on a trial basis.
(The same legislation established the National Assessment Governing Board—NAGB—
to set policy for both state and national NAEP.) The second point is a more substantive
one, which has emerged from the results themselves. Over the past decade, differential
progress has been made by students in the key subjects of reading and mathematics.

State NAEP Proves Its Value

At the end of its first decade, state NAEP—no longer considered a “trial” since 1996—
is doing well as the nation’s only ongoing independent measure of student achievement
at the state level in the key subjects of reading, writing, mathematics, and science. State
NAEP is the only assessment that allows states to compare their students’ performance to
that of students in other states using a common assessment instrument. This capability
has made state NAEP a valuable commodity for the state education policy, research, and
assessment communities.
■ Every state and jurisdiction, with the exception of one, has participated in at least
one of the state assessments.
■ An average of 40 states and jurisdictions volunteer to participate in each state
assessment cycle.
■ As many as 15,800 schools and about 400,000 students volunteer to participate in
the now typical two-subject, two-grade state assessment program.
■ As NAGB has laid new ground for more contemporary content frameworks to
guide development of the NAEP assessments, both the state and national assess-
ments have become progressively more challenging in the knowledge, skills, and
abilities that they assess as well as in their assessment specifications (e.g., the
number of multiple-choice items has been reduced, while the number of con-
structed-response items has been increased).

4 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
NAEP Results Show More Improvement in Mathematics
Than in Reading
NAEP national and state-level results show that students across the nation are making
marked progress in mathematics but very little progress in reading. In general, national
reading scores have remained about the same since 1992 except for a small increase for
eighth-graders between 1992 and 1998, when the most recent eighth-grade reading
assessment was conducted. Consistent with these national results, the results of state-level
reading assessments at grade 4 show that relatively few states had significant score in-
creases or declines between 1992 and 1998. Mathematics results, on the other hand, have
shown progress over the past decade for nearly every subgroup of the population and
across almost all states and jurisdictions that participated in the assessments. Over the past
decade, Whites, Blacks, Hispanics, males, females, and students in the fourth, eighth, and
twelfth grades have all shown increases in their mathematics scores. Improvement has also
occurred at all percentiles. Thus, although the achievement gap in mathematics between
Whites and minorities has not changed over the past decade, students scoring in the two
lowest percentiles (the 10th and 25th), in which minorities are disproportionately repre-
sented, have shown score increases since 1990 at all three grades. Out of the 36 states and
jurisdictions that participated in both 2000 and the first state assessment at grade 4 in
1992, 26 showed increases from 1992. At grade 8, of the 31 states and jurisdictions that
participated in both 2000 and the first state assessment in 1990, 27 showed increases over
the decade. The achievement level results in reading and mathematics—that is, the
percentages of students attaining the Basic, Proficient, and Advanced levels adopted by
NAGB—showed similar patterns.

Thus, the data described in this issue of the Quarterly are a fitting example of the substan-
tive value of a decade of measuring student achievement. As intended when NAEP was
first mandated, the NAEP 2000 Mathematics Assessment provides an excellent basis for
dialogue among curriculum experts and practitioners concerning “what students know
and can do.”

E D U C AT I O N S TAT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , F A L L 2 0 0 1 5
6 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
 F E AT U R E D T O P I C : N A E P 2 0 0 0
M AT H E M AT I C S A S S E S S M E N T

The Nation’s Report Card: Mathematics 2000

James S. Braswell, Anthony D. Lutkus, Wendy S. Grigg, Shari L. Santapau,
Brenda Tay-Lim, and Matthew Johnson ................................................................... 7
Invited Commentary: Policy Implications of Findings From The Nation’s
Report Card: Mathematics 2000
Debra Paulson, Eighth-Grade Mathematics Teacher, Dr. Manuel Hornedo
Middle School, El Paso, Texas, and Member, National Assessment Governing
Board (NAGB) ...................................................................................................... 18

Mathematics 2000
The Nation’s Report Card: Mathematics 2000
——————————————————————————————————
James S. Braswell, Anthony D. Lutkus, Wendy S. Grigg, Shari L. Santapau,
Brenda Tay-Lim, and Matthew Johnson
This article was excerpted from The Nation’s Report Card: Mathematics Highlights 2000. The sample survey data are from the National Assessment of
Educational Progress (NAEP) 1990, 1992, 1996, and 2000 Mathematics Assessments.

Introduction performance standards adopted by the National Assessment

The National Assessment of Educational Progress (NAEP) Governing Board (NAGB) as part of its statutory responsi-
is the nation’s only ongoing representative sample survey of bilities. The achievement levels are collective judgments of
student achievement in core subject areas. Authorized by what students should know and be able to do:
Congress and administered by the National Center for ■ Basic denotes partial mastery of prerequisite knowl-
Education Statistics (NCES) in the U.S. Department of edge and skills that are fundamental for proficient
Education, NAEP regularly reports to the public on the work at each grade.
educational progress of students in grades 4, 8, and 12.
■ Proficient represents solid academic performance for
each grade assessed. Students reaching this level have
In 2000, NAEP conducted a national mathematics assess-
demonstrated competency over challenging subject
ment of fourth-, eighth-, and twelfth-grade students. State-
matter, including subject-matter knowledge, applica-
level results were also collected at the fourth and eighth
tion of such knowledge to real-world situations, and
grades within participating states and jurisdictions. This
analytical skills appropriate to the subject matter.
article presents highlights from the NAEP 2000 Mathemat-
ics Assessment for the nation and the states. Results in 2000 ■ Advanced signifies superior performance.
are compared to results in 1990, 1992, and 1996. Following
As provided by law, the Acting Commissioner of Education
the performance results are several sample questions and
Statistics, upon review of a congressionally mandated
student responses typical of those from recent NAEP
evaluation of NAEP, determined that the achievement levels
mathematics assessments.
are to be considered developmental and should be inter-
Students’ performance on the assessment is described in preted and used with caution. However, both the Acting
terms of average scores on a 0–500 scale and in terms of the Commissioner and NAGB believe these performance
percentages of students attaining three achievement levels: standards are useful for understanding trends in student
Basic, Proficient, and Advanced. The achievement levels are achievement. They have been widely used by national and

E D U C AT I O N S TAT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , F A L L 2 0 0 1 7
Featured Topic: NAEP 2000 Mathematics Assessment

state officials, including the National Education Goals Major Findings for the Nation
Panel, as a common yardstick of academic performance. National results are for students attending both public and
nonpublic schools.
In addition to providing average scores and achievement
level performance at the national and state levels, this National average scores
article includes national results for selected subgroups of Results for the NAEP 2000 Mathematics Assessment show
students as well as a discussion of home and school con- overall gains in fourth-, eighth-, and twelfth-graders’
texts for mathematics performance. However, this article national average scores since 1990, the first year in which
does not include results for a second sample of students the current mathematics assessment was administered
assessed at both the national and state levels—one in which (figure A). Fourth- and eighth-graders made steady
testing accommodations were provided to students with progress, with higher average scores in 2000 than in 1996,
special needs (i.e., students with disabilities or students 1992, or 1990. However, this was not the case for twelfth-
with limited English proficiency). For results that include graders. Although twelfth-graders’ average score was higher
the performance of special-needs students who were in 2000 than in 1990, it was lower in 2000 than in 1996.
assessed with accommodations, see the complete report,
The Nation’s Report Card: Mathematics 2000. Such results National achievement level results
were omitted from the highlights presented in this article in The percentages of fourth- and eighth-graders at or above
order to allow comparisons with past assessment results, Basic and at or above Proficient increased across the decade,
which did not include accommodated students. reaching their highest levels in both grades in 2000 (figure B).

Figure A.—Average mathematics scores, grades 4, 8, and 12: 1990–2000

Scores
500

325

304*
299 301
300 Grade 12
294*

275
275 272* Grade 8
268*
263*

250

228
224* Grade 4
225 220*
213*

200

0
1990 1992 1996 2000

*Significantly different from 2000.

NOTE: The average scores are based on the NAEP mathematics scale, which ranges from 0 to 500.
SOURCE: National Center for Education Statistics, National Assessment of Educational Progress
(NAEP), 1990, 1992, 1996, and 2000 Mathematics Assessments. (Previously published on p. 1 of The
Nation’s Report Card: Mathematics Highlights 2000.)

8 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
 The Nation’s Report Card: Mathematics 2000

Figure B.—Percentage of students within and at or above the mathematics achievement levels, grades 4, 8, and 12: 1990–2000

Grade 4 Advanced 1%* 2%* 2% 3%

Proficient 12%* 13%* 16%* 19%* 23%
18%* 21%* At or above
26% Proficient
Basic 37%*
41%
43%
50%* 43%
59%* At or above
64%*
69% Basic
Below Basic
50%*
41%* 36%* 31%

1990 1992 1996 2000

Grade 8 Advanced 2%* 3%* 4% 5%

Proficient 13%* 15%* 18%* 20%*
21%* 24%* 22% At or above
27% Proficient
Basic 37%
37% 39%
52%* 38%
58%* At or above
62%*
66% Basic
Below Basic 48%* 42%* 38%* 34%

1990 1992 1996 2000

Grade 12 Advanced 1% 2% 2% 2%
Proficient 10%* 12%* 13% 15% 14% At or above
16% 14%
17% Proficient

Basic
46%
49% 53%* 48%
58%* At or above
64% 65% Basic
69%*
Below Basic
42%* 36% 35%
31%*

1990 1992 1996 2000

*Significantly different from 2000.

NOTE: Percentages within each mathematics achievement level range may not add to 100, or to the exact percentages at or above achieve-
ment levels, because of rounding. Basic denotes partial mastery of prerequisite knowledge and skills that are fundamental for proficient work
at each grade. Proficient represents solid academic performance for each grade assessed. Students reaching this level have demonstrated
competency over challenging subject matter, including subject-matter knowledge, application of such knowledge to real-world situations,
and analytical skills appropriate to the subject matter. Advanced signifies superior performance.
HOW TO READ THIS FIGURE:
• The italicized percentages to the right of the shaded bars represent the percentages of students at or above Basic and Proficient.
• The percentages in the shaded bars represent the percentages of students within each achievement level.
SOURCE: National Center for Education Statistics, National Assessment of Educational Progress (NAEP), 1990, 1992, 1996, and 2000 Mathematics
Assessments. (Previously published on p. 2 of The Nation’s Report Card: Mathematics Highlights 2000.)

E D U C AT I O N S TAT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , F A L L 2 0 0 1 9
Featured Topic: NAEP 2000 Mathematics Assessment

At grade 12, the results are mixed. From 1996 to 2000, Major Findings for the States and Other
there was a decrease in the percentage at or above Basic. Jurisdictions
However, the percentage of twelfth-graders at or above both In addition to national results on students’ mathematics
Basic and Proficient was higher in 2000 than in 1990. performance, the 2000 assessment collected performance
data for fourth- and eighth-graders who attended public
National average scores at different percentiles schools in states and other jurisdictions that volunteered to
The gains in average mathematics scores at all three grades participate. State-level data have been collected since 1992
since 1990 are reflected in students’ performance across the at grade 4 and since 1990 at grade 8. In 2000, 40 states and
score distribution. Lower-, middle-, and higher-performing 6 other jurisdictions participated at grade 4, and 39 states
students had higher scores in 2000 than in 1990 (figure C). and 5 other jurisdictions participated at grade 8. The results
This finding is the result of analyzing scores at percentiles— of the state assessment are only for students attending
or points across the score distribution—on the NAEP public schools.
mathematics scale.
State average scores
The score increases seen since 1990 for fourth-, eighth-, and Of the 36 states and jurisdictions that participated in both
twelfth-graders were evident across the score distribution 2000 and the first state assessment at grade 4 in 1992, 26
(at the 10th, 25th, 50th, 75th, and 90th percentiles). had higher average scores in 2000 than in 1992. Of the 31
However, the decline at grade 12 since 1996 occurred at the states and jurisdictions that participated in both 2000 and
lower and middle points of the distribution (at the 10th, the first state assessment at grade 8 in 1990, 27 had higher
25th, and 50th percentiles). average scores in 2000 than in 1990.

Figure C.—Average mathematics scores by percentile, grades 4, 8, and 12: 1990–2000

500 Grade 4 500 Grade 8 500 Grade 12

Percentile Percentile Percentile

275 325 321 350 345 346

266 90th 317 90th 343 90th
315* 339*
262*
259* 307*
253* 301 327 326
250 298 324
250 246* 75th 300 294* 75th 325 319* 75th
242*
288*
235*
230 305*
226* 277 301 302
221* 50th 273* 50th 296* 50th
225 275 269* 300
214* 264*
208 282*
204* 252 277
199* 25th 25th 276 25th
248
200 193* 250 243* 275 270*
239*
186 261*
182* 255
177* 227 10th 254
10th 224 10th
175 171* 225 221* 250 247*
215*

150 200 225

0 0 0

1990 1992 1996 2000 1990 1992 1996 2000 1990 1992 1996 2000

*Significantly different from 2000.

SOURCE: National Center for Education Statistics, National Assessment of Educational Progress (NAEP), 1990, 1992, 1996, and 2000 Mathematics Assessments. (Previously published
on p. 3 of The Nation’s Report Card: Mathematics Highlights 2000.)

10 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
 The Nation’s Report Card: Mathematics 2000

In 2000, no state scored higher at grade 4 than these nine: other jurisdictions had higher percentages of students at or
Connecticut, Indiana, Iowa, Kansas, Massachusetts, Minne- above Proficient than did the nation, 12 had percentages that
sota, North Carolina, Texas, and Vermont. Figure D shows were not different from the percentage for the nation, and
states’ and other jurisdictions’ 2000 average score perfor- 19 had percentages that were lower than that for the nation.
mance in comparison to the national average score for
public schools. Of the 46 states and jurisdictions that National Results for Student Subgroups
participated in the 2000 assessment at grade 4, 14 had In addition to presenting information about all students’
scores that were higher than the national average score, performance, NAEP also looks at the achievement of
14 had scores that were not different from the national various subgroups of students. The performance of various
average, and 18 had scores that were lower than the racial/ethnic subgroups and of males and females reveals
national average. how these students did in comparison to each other in the
year 2000 and whether they progressed over the past
In 2000, no state scored higher at grade 8 than these three: decade. While the complete report describes the perfor-
Kansas, Minnesota, and Montana. Figure E shows that of mance of student subgroups at both the state and national
the 44 states and other jurisdictions that participated in the levels, the highlights in this article are for the nation only.
2000 assessment at grade 8, 16 had scores that were higher
than the national average score, 13 had scores that were not When reading these results, it is important to keep in mind
different from the national average, and 15 had scores that that there is no simple, causal relationship between mem-
were lower than the national average. bership in a subgroup and mathematics achievement. A
complex mix of educational and socioeconomic factors may
State achievement level results interact to affect student performance.
At grade 4, 4 states and other jurisdictions had higher
percentages of students at or above Proficient than did the Average scores for different racial/ethnic subgroups
nation, 23 had percentages that were not different from the Of the five racial/ethnic subgroups of students identified in
percentage for the nation, and 19 had percentages that were the 2000 mathematics assessment, three—White, Black, and
lower than that for the nation. At grade 8, 13 states and Hispanic—had average scores that showed overall gains

Figure D.—State versus national average mathematics scores, grade 4 public schools: 2000

WA
State has higher average scale score
AK than nation.
MT ND ME

OR MN VT State has average scale score that

NH
ID SD
WI
NY MA
is not significantly different from
WY MI CT nation.
RI
IA NJ
PA
NV
NE State has lower average scale score
OH DE
UT IL IN than nation.
CA CO WV MD
KS VA
MO
KY DC State did not meet the minimum
NC participation rate guidelines.
TN
AZ OK
NM AR SC DDESS
State did not participate in the NAEP
HI
AL GA 2000 Mathematics State Assessment.
MS
DoDDS
TX NOTE: Caution should be exercised when
LA
interpreting comparisons among states and
Guam other jurisdictions. NAEP performance
FL
estimates are not adjusted to account for
VI the socioeconomic, demographic, or
Samoa
geographic differences among states and
jurisdictions.

DDESS: Department of Defense Domestic Dependent Elementary and Secondary Schools. DoDDS: Department of Defense Dependents Schools (Overseas).
NOTE: National results are based on the national sample, not on aggregated state assessment samples.
SOURCE: National Center for Education Statistics, National Assessment of Educational Progress (NAEP), 2000 Mathematics Assessment. (Previously published as figure A on p. 4 of The
Nation’s Report Card: Mathematics Highlights 2000.)

E D U C AT I O N S TAT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , F A L L 2 0 0 1 11
Featured Topic: NAEP 2000 Mathematics Assessment

Figure E.—State versus national average mathematics scores, grade 8 public schools: 2000

WA State has higher average scale score

AK than nation.
MT ND ME

OR VT
MN NH State has average scale score that
ID SD
WI NY MA is not significantly different from
CT nation.
WY MI
RI
IA PA
NE NJ
NV OH DE State has lower average scale score
UT IN
IL
MD than nation.
CA CO WV
KS VA
MO
KY DC
State did not meet the minimum
TN NC participation rate guidelines.
AZ OK
NM AR SC DDESS
HI State did not participate in the NAEP
AL GA
MS
DoDDS
2000 Mathematics State Assessment.
TX LA
NOTE: Caution should be exercised when
Guam
interpreting comparisons among states and
FL other jurisdictions. NAEP performance
VI estimates are not adjusted to account
Samoa for the socioeconomic, demographic, or
geographic differences among states and
jurisdictions.

DDESS: Department of Defense Domestic Dependent Elementary and Secondary Schools. DoDDS: Department of Defense Dependents Schools (Overseas).
NOTE: National results are based on the national sample, not on aggregated state assessment samples.
SOURCE: National Center for Education Statistics, National Assessment of Educational Progress (NAEP), 2000 Mathematics Assessment. (Previously published as figure B on p. 5 of The
Nation’s Report Card: Mathematics Highlights 2000.)

since 1990. While White students were the only subgroup Indian students performed at or above the Proficient level in
whose average scores were higher in 2000 than in 1990 at 2000 than in 1990. There were also higher percentages of
all three grades, Black and Hispanic students’ average scores White, Black, and Hispanic students at or above the Basic
were higher than in 1990 at grades 4 and 8. level in 2000 than in 1990 or 1992.

Comparing performance across the subgroups of students in At grade 8, more White and Hispanic students were at or
2000 shows that White and Asian/Pacific Islander students above Proficient in 2000 than in 1990, and more White,
scored higher, on average, than Black, Hispanic, and Ameri- Black, and Hispanic students were at or above Proficient in
can Indian students at grades 8 and 12. Asian/Pacific Islander 2000 than in 1992. At or above the Basic level, there were
students scored higher than White students at grade 12. higher percentages of White, Black, and Hispanic eighth-
graders in 2000 than in 1990 or 1992.
Trends in average score gaps between selected racial/
ethnic subgroups There were few changes over the decade for twelfth-graders;
Across the assessments from 1990 to 2000, the score gaps only White students had higher percentages at or above the
between White and Black students and between White and Proficient level in 2000 than in 1990. There were also higher
Hispanic students were large at every grade. There was no percentages of White students at or above the Basic level in
evidence in the 2000 assessment of any narrowing of the 2000 than in 1990.
racial/ethnic group score gaps since 1990.
Comparing the subgroups’ 2000 performance shows that, in
Achievement level results for different racial/ethnic general, the percentages at or above the Basic achievement
subgroups level were higher for White and Asian/Pacific Islander
The mathematics achievement of students in the racial/ students than for the other subgroups of students.
ethnic subgroups was similar to their average score perfor-
mance—while there were improvements over the past 10 Average scores for males and females
years, not all groups improved at all grades. At grade 4, At all three grades, both males and females had higher
higher percentages of White, Black, Hispanic, and American scores in 2000 than they did in 1990 and, at grade 4, they

12 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
 The Nation’s Report Card: Mathematics 2000

both showed relatively steady improvement across the four Calculator use for classwork
assessments from 1990 to 2000. Results from the 2000 mathematics assessment suggest a
relationship between student-reported calculator use for
In 2000, males outperformed females in mathematics at classwork and mathematics performance that is markedly
grades 8 and 12. There was no significant difference different at grade 4 than at grades 8 and 12. At grade 4,
between males’ and females’ average scores at grade 4. more frequent calculator use was associated with lower
scores, while at grades 8 and 12 the opposite was generally
Trends in average score gaps between males and true: students who said they use calculators more often
females
tended to score higher than their peers who reported using
The gap between the average scores of males and females them less frequently (figure F).
was quite small at all three grades and fluctuated only
slightly across the assessments from 1990 to 2000. Time spent on homework
In mathematics, as in other subjects assessed by NAEP,
Achievement level results for males and females
most students who spent time doing homework every day
At grade 4, there were higher percentages of both males and scored higher than those who did not do homework. Only
females at or above Proficient and at or above Basic in 2000 at grade 4, where homework demands are light in com-
than in 1990, 1992, or 1996. parison to higher grades, did students who reported
spending an hour or more on homework score lower than
At grade 8, there were higher percentages of both males and their peers who did not do homework. How much time in
females at or above Proficient in 2000 than in 1990 and 1992, general is associated with higher mathematics perfor-
and a higher percentage of males at or above Proficient than mance on NAEP? Results from the 2000 mathematics
in 1996. There were also more male and female eighth- assessment suggest that at grades 4 and 8, a moderate
graders at or above Basic in 2000 than in 1990 or 1992, and amount of time—between 15 and 45 minutes depending
more male eighth-graders at or above Basic than in 1996. on grade level—is associated with a higher average score
on NAEP than a longer time of 1 hour or more. This was
At grade 12, there were higher percentages of males and
not the case at grade 12, where there was no statistically
females at or above Proficient in 2000 than in 1990. There was
significant difference in the performance of students
a decline in the percentage of both male and female twelfth-
spending any time between 15 minutes and 1 hour or
graders at or above Basic in 2000 compared to 1996, although
more on mathematics homework.
both groups’ percentages were up in 2000 over 1990.
Attitudes about mathematics
A comparison of males’ and females’ results in 2000 shows
The attitudes of students who took the NAEP mathemat-
that there were higher percentages of males at or above
ics assessment were strongly related to their performance.
Proficient at grades 4, 8, and 12.
Students who participated in the 2000 assessment were
asked to consider several statements about mathematics
Home and School Contexts for Mathematics
designed to gauge their attitudes toward the subject.
Performance
The results for two of those statements are presented
Many factors influence students’ learning. Activities that
here: At all three grade levels, students who agreed that
take place while students are either at school or at home as
they like math and that math is useful for solving prob-
well as the attitudes they develop about learning mathemat-
lems scored higher than students who disagreed with
ics may enhance or detract from their ability to do math.
these statements.
The NAEP 2000 Mathematics Assessment focused on
students’ performance in light of responses to questions Sample Mathematics Questions and Student
about mathematics activities at school and at home and Responses
attitudes toward mathematics. While these findings may
Sample questions from the 2000 assessment have not been
suggest a positive or negative relationship between perfor-
released to the public so that they can be used again in a
mance on the mathematics assessment and certain activities
future assessment. Therefore, the questions shown here are
or attitudes, it is important to remember that the relation-
taken from the NAEP 1996 Mathematics Assessment. They
ships are not necessarily causal—there are many factors that
are similar to the questions used for the 2000 assessment
play a role in mathematics performance.
because the same framework was used to develop questions

E D U C AT I O N S TAT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , F A L L 2 0 0 1 13
Featured Topic: NAEP 2000 Mathematics Assessment

Figure F.—Average mathematics scores by frequency of calculator use for classwork, grades 4, 8, and 12: 2000

Scores
500 Grade 4

Grade 8
325 Grade 12
308

300
292
283 286
279
275 276
275 268

250
240
235
230
225
210

200

0
Never/hardly ever Monthly Weekly Everyday

SOURCE: National Center for Education Statistics, National Assessment of Educational Progress (NAEP), 2000 Mathematics Assessment.
(Previously published on p. 11 of The Nation’s Report Card: Mathematics Highlights 2000.)

in 1996 and 2000. The framework provides the theoretical Grade 4 sample questions and responses
basis for the assessment, as well as directions for what kinds Getting ready for algebra. Young students are prepared for
of questions should be included in the assessment, how the abstract world of algebra by early exposure to concepts
these questions should be designed, and how student that help them make the transition from concrete numbers
responses should be scored. For details about the frame- to abstract expressions. The following multiple-choice
work, see the complete report. question, which required students to recognize that N
stands for the total number of stamps John had, puts the
Each student assessed at grades 4, 8, and 12 received a concept of a variable in a setting that fourth-graders can
booklet that contained three 15-minute sections of math- understand.
ematics questions. These questions were presented in two
formats: multiple choice and constructed response. The
Sample multiple-choice question for grade 4
constructed-response questions were either short (requiring
N stands for the number of stamps John had. He gave
students to provide answers to computation problems or
12 stamps to his sister. Which expression tells how
describe solutions in one or two sentences) or extended
many stamps John has now?
(requiring students to provide longer answers).
A N + 12
For each grade, two sample questions are presented here.
N – 12
Additional sample questions from the 1996 mathematics
assessment, as well as sample questions from the 1992
C 12 – N
and 1990 assessments, are available at the NAEP Web Site D 12 X N
(http://nces.ed.gov/nationsreportcard).

14 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
 The Nation’s Report Card: Mathematics 2000

Solving a multistep problem. Responses to the following

short constructed-response question were scored on a three- Sample multiple-choice question for grade 8
level scale: unsatisfactory, partial, or satisfactory. To answer A plumber charges customers $48 for each hour
the question satisfactorily, students needed to complete worked plus an additional $9 for travel. If h represents
three steps: (1) add the three amounts shown to get the the number of hours worked, which of the following
total spent each day, (2) multiply by 5 to get the total expressions could be used to calculate the plumber’s
needed for 5 days ($8.75), and (3) understand that nine total charge in dollars?
$1.00 bills would be needed.
A 48  9  h
B 48  9  h
Sample short constructed-response question for
grade 4
C 48  (9  h)
Sam can purchase his lunch at school. Each day he D (48  9)  h
wants to have juice that costs 50¢, a sandwich that (48  h)  9
costs 90¢, and fruit that costs 35¢. His mother has only
$1.00 bills. What is the least number of $1.00 bills that Reading and interpreting data. The following extended
his mother should give him so he will have enough constructed-response question, one of the more difficult
money to buy lunch for 5 days? eighth-grade questions used in 1996, required students to
demonstrate skills that are an important part of the junior
Sample satisfactory response
high school mathematics curriculum. It shows two accu-
A satisfactory response to this question gives the correct rately drawn graphs that appear to present very different
answer of nine $1.00 bills. results. Responses to the question were scored on a four-
level scale: unsatisfactory, partial, satisfactory, or complete.
A complete response indicates ability to critically evaluate
information presented in a graph.

Sample extended constructed-response question for

grade 8
This question requires you to show your work and
explain your reasoning. You may use drawings, words,
and numbers in your explanation. Your answer should
be clear enough so that another person could read it
and understand your think-
Grade 8 sample questions and responses METRO RAIL COMPANY
ing. It is important that you
Understanding an algebraic expression. The following Month Daily Ridership
show all of your work.
October 14,000
multiple-choice question required students to translate a
November 14,100
word problem into an algebraic expression. In a formal The data in the table to the December 14,100
algebra class, students are expected to set up equations with right has been correctly January 14,200
expressions like the one in choice E (the correct answer) February 14,300
represented by both graphs
March 14,600
and then determine, for example, the value of h if the shown below.
plumber’s total charge was $297.

E D U C AT I O N S TAT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , F A L L 2 0 0 1 15
Featured Topic: NAEP 2000 Mathematics Assessment

Grade 12 sample questions and responses

Which graph would be best to help convince others
Finding a missing value. The following multiple-choice
that the Metro Rail Company made a lot more money
question, a fairly easy one for twelfth-graders, required
from ticket sales in March than in October?
students to find a value that would make both equations
Explain your reason for making this selection. true. To solve the problem, students could either use a
formal algebraic solution process or simply substitute each
Why might people who thought that there was little of the choices until they found the correct answer.
difference between October and March ticket sales
consider the graph you chose to be misleading? Sample multiple-choice question for grade 12

Sample complete response 4  = and  3 =

A complete response to this question gives the correct
response, graph B, and provides a complete explanation. What number if placed in each box above
would make both equations true?
0
B 1
C 2
D 3
E 4

Measuring an angle. Responses to the following short

constructed-response question were scored on a two-level
Sample satisfactory response scale: unsatisfactory or satisfactory. In order to find the
A satisfactory response to the question gives the correct solution to the question, students needed to draw a line
response, graph B, and provides an incomplete but perpendicular to a given line, and then measure one of the
partially correct explanation. angles. This is an example of a NAEP question that requires
students to use a tool, such as a protractor or ruler. These
tools are provided to students during the assessment.

16 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
 The Nation’s Report Card: Mathematics 2000

Sample short constructed-response question for Data source: The National Assessment of Educational Progress (NAEP)
1990, 1992, 1996, and 2000 Mathematics Assessments.
grade 12
For technical information, see the complete report:
In the figure below, use the protractor to draw a line m
Braswell, J.S., Lutkus, A.D., Grigg, W.S., Santapau, S.L., Tay-Lim, B., and
through point P perpendicular to segment AP. In the Johnson, M. (2001). The Nation’s Report Card: Mathematics 2000
answer space provided, give the measure of the smaller (NCES 2001–517).
angle formed by lines l and m. Author affiliations: J.S. Braswell, A.D. Lutkus, W.S. Grigg, S.L. Santapau,
B. Tay-Lim, and M. Johnson, Educational Testing Service.
For questions about content, contact Arnold Goldstein
(arnold.goldstein@ed.gov).
To obtain the complete report (NCES 2001–517), call the toll-free
ED Pubs number (877–433–7827), visit the NCES Web Site
(http://nces.ed.gov), or contact GPO (202–512–1800).
To obtain the Highlights publication from which this article is
excerpted (NCES 2001–518), call the toll-free ED Pubs number
(877–433–7827) or visit the NCES Web Site (http://nces.ed.gov).

Sample satisfactory response

The following student’s response received the highest
score, satisfactory. Both line m and the degree measure
of the smaller angle are correct.

E D U C AT I O N S TAT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , F A L L 2 0 0 1 17
Featured Topic: NAEP 2000 Mathematics Assessment

Mathematics Policies
Invited Commentary: Policy Implications of Findings From The Nation’s
Report Card: Mathematics 2000
——————————————————————————————————
Debra Paulson, Eighth-Grade Mathematics Teacher, Dr. Manuel Hornedo Middle
School, El Paso, Texas, and Member, National Assessment Governing Board (NAGB)
This commentary represents the opinions of the author and does not necessarily reflect the views of the National Center for Education Statistics or NAGB.

On August 2, 2001, The Nation’s Report Card: Mathematics inclusion and accommodation, perhaps this process will
2000 was released. This report is an important landmark for affirm the value of including special-needs students and
the National Assessment of Educational Progress (NAEP). dispel the uncertainty about how this inclusion will affect
It provides findings from the NAEP 2000 Mathematics the reported results.
Assessment, the fourth national test in a 10-year period to
assess the same mathematics content and standards. In Implications of Information About
addition to results from national tests at grades 4, 8, and 12, Technology Use
the report also provides results from state-level tests that Many implications can be drawn from the information
began in 1992 at grade 4 and in 1990 at grade 8. gained from questionnaires completed by students, teach-
ers, and school administrators. “NAEP examines the
The Mathematics 2000 report provides an abundance of relationship between selected contextual variables drawn
statistics and information. The National Assessment from these questionnaires and students’ average scores on
Governing Board (NAGB) and the National Center for the mathematics assessment. Readers are cautioned that a
Education Statistics (NCES) have worked diligently to make relationship between a contextual variable . . . and student
reports like this one more accessible and understandable to mathematics performance is not necessarily causal.” In
more people. Questions that beg to be asked and answered other words, such a relationship may or may not indicate
include the following: How does all the information that a particular factor directly affects students’ mathematics
presented in this report relate to the mathematics that achievement. I would like to examine several relationships
students learn in school? How are students best taught this between questionnaire responses and student performance
mathematics content? and Who should be conducting the in the light of my experiences as an eighth-grade mathemat-
mathematics instruction? I will be exploring and highlight- ics teacher. My focus will be on the use of technology.
ing some of the implications not just for policymakers, but
also for parents, schools, and teachers. Use of computers
As computers become more and more a part of our work
Inclusion of Students Who Use Testing and personal lives, questions arise as to how much comput-
Accommodations ers should be used in school, in what ways computers
I believe that one of the most important findings emerges should be used, and whether computer use has an impact
from comparing the two sets of results provided by this on student learning. Certainly, computers are increasingly
assessment: scores for students who were not permitted to available in classrooms at each grade level. The Mathematics
use any testing accommodations and scores that include the 2000 report states that the availability of computers in
performance of special-needs students who were provided classrooms increased by at least 20 percentage points from
with testing accommodations. In both 1996 and 2000, the 1996 to 2000, although the availability of computer labs did
NAEP mathematics assessment collected the two sets of not change significantly during this period. Unfortunately,
results separately. At grades 4 and 8, there were no signifi- comparing the increase of computers in the classroom to
cant differences between the two sets in either 1996 or student scores is not encouraging. There is not a direct
2000. At grade 12, there was no significant difference relationship between the availability of computers in stu-
between the two sets in 2000. dents’ classrooms and increases in mathematics achievement.

States, districts, and even schools are all grappling with the I concur with these findings. Yes, I have two computers in
issue of including more special-needs students in assess- my classroom, compared to none in 1996. But I have little
ments as well as in regular classrooms. Everyone knows this mathematical software and relatively unreliable Internet
is a good idea, but the issue of accountability in assessing access. Without a projector, it is almost impossible to
these students is complex. As NAEP continues to model effectively use two computers in a classroom with 30

18 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
 Invited Commentary: Policy Implications of Findings From The Nation’s Report Card: Mathematics 2000

students. Also, with only 44 minutes for each class, it is had eighth-graders with higher average scores. Even the
difficult to give students time to use the computer. Not type of calculator that students reported using was directly
surprisingly, teachers responding to the NAEP question- related to how they performed on the mathematics assess-
naires quite often reported that computers were either not ment. Eighth-graders who used a scientific calculator
used at all or used primarily for math learning games or drill. scored higher than their peers who did not use one, and
the same was true of eighth-graders who used a graphing
What are the implications? By themselves, computers in calculator compared to their peers who did not. Between
classrooms or labs are not going to make a difference in the 1996 and 2000, the percentages of eighth-graders who
amount or type of mathematics learned. Teachers need reported using scientific and graphing calculators in-
ongoing training and support in using the computer as an creased. Many states do allow some calculator use on grade
instructional tool. They need software and hardware, which 8 state assessments. Again, it is important for teachers not
unfortunately are often expensive, hard to find, and difficult only to have access to calculators, but also to have training
to use. In this high-tech world, it is imperative to give in how to effectively use them. The key is teaching stu-
students the opportunity to use computers in school. The dents to use calculators as a tool and giving students
issue is, how can computers be used to increase students’ calculator tasks and assessments. I know that using
achievement in mathematics? I believe that access to and graphing calculators with my eighth-grade students is
effective use of computers in schools is essential in closing extremely motivating and really works best for exploring
the gap between those students who use technology patterns or making predictions.
efficiently and those students who are technologically
deficient or deprived. At grade 12, daily use of calculators was again associated
with the highest scores. The type of calculator used was
Use of calculators important, with those twelfth-graders who reported using a
Regarding the issue of how calculator use in the classroom graphing calculator scoring an average of 25 scale-score
relates to student performance, the results of the 2000 points higher than those who did not. Though it could be
mathematics assessment are more encouraging and clearer argued that twelfth-graders who use graphing calculators
as to what works and what doesn’t. The proper role of have higher scores because they have taken more advanced
calculators in the K–12 mathematics curriculum has been mathematics courses, I contend that being able to effi-
and continues to be debated. Calculator-use policies vary ciently use a graphing calculator could make the advanced
across districts and schools; even within the same school, mathematics courses more accessible to all students.
teachers have different opinions about how calculators
should be integrated with instruction. States are also There are a couple of implications regarding graphing
deciding if, how, and when calculators should be allowed calculators. Allowing or even requiring the use of graphing
on state assessments. calculators on state assessments has a direct effect on the
number of graphing calculators in the classroom and the
At grade 4, more frequent use of calculators for mathemat- amount of time that they are used in classroom instruction.
ics activities, as reported by students, was linked to lower I have seen this happen in Texas, where the state’s end-of-
scores. This information seems to confirm the need for course exam in Algebra I requires the use of graphing
caution in the use of calculators at grade 4. Since students calculators. These calculators are expensive, however, and
in elementary school are still becoming fluent in comput- states or districts need to provide funding for purchasing
ing whole numbers, calculators need to be used more for these calculators and for training teachers to effectively use
exploring and deepening the understanding of number them.
sense.
Conclusions
At grades 8 and 12, the implications are much clearer. For In conclusion, there is much to be learned from the
example, using a calculator in the eighth grade appears to results of the NAEP 2000 Mathematics Assessment and
benefit mathematics achievement. At grade 8, daily calcula- from comparing these results to those of past assessments.
tor use for mathematics activities, as reported by both Lots of people, especially local administrators and
students and teachers, was associated with the highest teachers, are not knowledgeable about NAEP. I believe
scores. In fact, teachers who permitted unrestricted use of that since districts and schools do not receive individual
calculators and those who permitted calculator use on tests student scores, many educators conclude that the

E D U C AT I O N S TAT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , F A L L 2 0 0 1 19
Featured Topic: NAEP 2000 Mathematics Assessment

information is not relevant. I beg to differ. In addition Using state-level results from 1990 or 1992 to 2000, states
to the implications that I have already discussed, the can track their own progress or look at other states that
information in NAEP reports has many other important have shown dramatic increases in mathematics achieve-
implications for state, district, and school policies. How ment. Mathematics learning and achievement can be
much homework to assign, what types of mathematics affected by state policies on recommended textbooks, state
courses to offer or require, and what courses teachers curriculum guidelines, assessments, course requirements for
need for certification—these are all examples of policies students, and teacher certification requirements. Investigat-
for which NAEP could have implications. The results of ing a state’s policies and the implementation of these
the NAEP assessments can help educators and policy- policies over the past 8 to 10 years may provide insight
makers make better decisions. about what it takes to improve mathematics achievement.

20 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
 E L E M E N TA RY AND S E C O N D A RY E D U C AT I O N

The Next Generation of Citizens: NAEP Civics Assessments: 1988 and 1998
Andrew R. Weiss, Anthony D. Lutkus, Wendy S. Grigg, and Richard G. Niemi ..... 21
Homeschooling in the United States: 1999
Stacey Bielick, Kathryn Chandler, and Stephen P. Broughman ............................. 25
Teacher Preparation and Professional Development: 2000
Basmat Parsad, Laurie Lewis, and Elizabeth Farris .............................................. 33
Private School Universe Survey: 1999–2000
Stephen P. Broughman and Lenore A. Colaciello .................................................... 37
Overview of Public Elementary and Secondary Schools and Districts: School
Year 1999–2000
Lee M. Hoffman ..................................................................................................... 42
Revenues and Expenditures for Public Elementary and Secondary
Education: School Year 1998–99
Frank Johnson ...................................................................................................... 55

NAEP Civics Assessments

The Next Generation of Citizens: NAEP Civics Assessments: 1988 and 1998
——————————————————————————————————
Andrew R. Weiss, Anthony D. Lutkus, Wendy S. Grigg, and Richard G. Niemi
This article was originally published as the Executive Summary of the Statistical Analysis Report of the same name. The sample survey data are from the
National Assessment of Educational Progress (NAEP) 1988 and 1998 Civics Assessments.

Introduction students about the structures, functions, and processes of

As we move into the 21st century, our nation looks to its government and about the meaningful ways in which
youth for confirmation that the government established citizens can make decisions about public issues and
over 200 years ago will remain relevant, vital, and strong. participate in governance. This report, based on findings
We expect that today’s students are being prepared to from the National Assessment of Educational Progress
understand and maintain the values of our democratic (NAEP), provides a view of students’ achievement in civics
society. Civics education in our nation’s schools informs over the 10-year period from 1988 to 1998. The data and

E D U C AT I O N S TAT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , FA L L 2 0 0 1 21
Elementary and Secondary Education

information provided give some indication as to whether the major findings from the NAEP 1998 special study is
there have been changes in students’ understanding of presented below.
civics and whether civics education has changed during the
last decade of the 20th century. Overall Assessment Results
■ In both 1988 and 1998, students at each of the three
NAEP is the nation’s only ongoing survey of what students grade levels answered about two-thirds of the
know and can do in various academic subject areas. assessment questions correctly.
Authorized by Congress and administered by the National
■ Fourth-grade students in 1998 answered more
Center for Education Statistics (NCES) in the U.S. Depart-
questions correctly, on average, than did fourth-grade
ment of Education, NAEP regularly reports to the public on
students in 1988.
the educational progress of students in grades 4, 8, and 12.
■ Eighth-grade students in 1998 answered fewer
In 1998, NAEP conducted two national assessments of
questions correctly, on average, than their counter-
students’ civics knowledge in each of these grades.
parts in 1988.
One of the 1998 civics assessments was entirely new, ■ The performance of 12th-grade students in 1998 was
employing a new set of test specifications (or “framework”). not significantly different from that of their counter-
The results of this new assessment were not comparable to parts in 1988.
those of the 1988 assessment. The other 1998 civics
assessment (based on the 1988 test objectives) was a special Results for Student Subgroups
study that repeated a number of the multiple-choice test Gender
questions used in 1988. ■ Fourth-grade males had a higher percentage of
correct responses in 1998 than in 1988, while 12th-
The results of the new assessment were reported in the fall grade males had a lower percentage correct in 1998.
of 1999 as the NAEP 1998 Civics Report Card for the Nation
■ The percentage of correct responses for female
(Lutkus et al. 1999). The results of the special study are
students at grades 4, 8, and 12 did not change
reported here as a summary of trends in students’ knowl-
significantly between 1988 and 1998.
edge and teachers’ classroom practices over the 10-year
period. Race/ethnicity
■ In 1998, the percentage of correct responses in-
The results are based on the assessment of a sample of
creased for White students in grade 4 and decreased
students at each grade that is statistically representative of
for Hispanic students in grade 12.
the entire nation. Students’ performance is described in
terms of average percentage correct, rather than the tradi- ■ At all three grades in both 1988 and 1998, White
tional NAEP scale scores. The reason for this departure is students consistently achieved a higher percentage
that the relatively small set of test questions repeated from correct than either Black or Hispanic students.
1988 in grades 8 and 12 did not allow comprehensive
coverage of the 1988 test objectives, nor did it allow the Trends in Civics Topics Studied
reliable development of scale scores parallel to those used ■ A trend was noted toward less frequent social studies
in 1988. classes in grade 4, with 49 percent of students in
1988 reporting daily classes compared to 39 percent
This report provides results for subgroups of students in 1998.
defined by various background and contextual characteris- ■ The percentage of eighth-graders who reported
tics. The analyses focus on differences between 1988 and having studied civics or American government in
1998, rather than differences among groups within each grades 5, 6, and 7 rose between 1988 and 1998.
year. To illustrate the civics knowledge that was assessed, ■ The percentage of 12th-graders who said they were
the report provides numerous samples of the test questions. studying civics or American government in their
The report also explores trends in the classroom coverage of current grade rose between 1988 (61 percent) and
civics topics from 1988 to 1998, as well as trends in class- 1998 (71 percent).
room instructional activities over the decade. A summary of

22 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
 The Next Generation of Citizens: NAEP Civics Assessments: 1988 and 1998

■ The percentage of fourth-graders who reported ■ The amount of time that 12th-graders reported
spending “a lot” of time studying three of the six spending on civics homework did not change
civics curriculum topics surveyed—elections and significantly between 1988 and 1998.
voting, President and leaders of our country, and judges ■ Between 1988 and 1998, the percentage of eighth-
and courts—increased between 1988 and 1998 graders reporting that their families regularly got a
(table A). newspaper fell from 77 percent to 71 percent. For
■ For both 8th- and 12th-graders, the U.S. Constitution 12th-graders, the percentage dropped from 82
and Bill of Rights is the one curriculum topic, of the percent to 75 percent.
10 topics surveyed at these grade levels, that was
studied “a lot” by the majority of students in both Reference
1988 and 1998. Lutkus, A.D., Weiss, A.R., Campbell, J.R., Mazzeo, J., and Lazer, S.
(1999). NAEP 1998 Civics Report Card for the Nation (NCES
■ The amount of time spent studying the various civics
2000–457). U.S. Department of Education, National Center for
topics surveyed at grades 8 and 12 was similar and Education Statistics. Washington, DC: U.S. Government
has not changed between 1988 and 1998. Printing Office.

Trends in Contexts for Learning Civics

■ The frequency with which students at grades 8 and
12 were assigned extra reading material by their Data sources: The National Assessment of Educational Progress
civics or American government teachers increased (NAEP) 1988 and 1998 Civics Assessments.
between 1988 and 1998 (figure A). For technical information, see the complete report:
Weiss, A.R., Lutkus, A.D., Grigg, W.S., and Niemi, R.G. (2001). The Next
■ The percentage of 8th- and 12th-graders who Generation of Citizens: NAEP Civics Assessments: 1988 and 1998
reported being assigned to work on group projects at (NCES 2001–452).
least once or twice a week rose substantially from Author affiliations: A.R. Weiss, A.D. Lutkus, and W.S. Grigg, Educational
Testing Service; R.G. Niemi, University of Rochester.
1988 to 1998 (figure B).
For questions about content, contact Patricia Dabbs
■ The percentage of fourth-graders who reported (patricia.dabbs@ed.gov).
discussing current events in social studies class once To obtain the complete report (NCES 2001–452), call the toll-free
ED Pubs number (877–433–7827) or visit the NCES Web Site
or twice a week increased from 29 percent in 1988 to (http://nces.ed.gov).
39 percent in 1998.

Table A.—Percentage of students who reported how much they studied selected civics topics, grade 4:
1988 and 1998

Percentage of students
How much have you studied the following A lot Some Not at all
topics in American government or civics? 1988 1998 1988 1998 1988 1998

How laws are made 21 24 52 54 27 23

About judges and courts 7 < 13 33 < 38 60 > 49
President and leaders of our country 33 < 37 48 49 19 > 14
Elections and voting 13 < 19 47 < 54 39 > 27
Your community 47 48 40 42 12 10
The rights and responsibilities of citizens 27 33 47 48 25 > 20

> 1988 significantly greater than 1998.

< 1988 significantly less than 1998.
NOTE: Percentages may not add to 100 because of rounding.
SOURCE: U.S. Department of Education, National Center for Education Statistics, National Assessment of Educational Progress
(NAEP) 1988 and 1998 Civics Assessments. (Originally published as table 2.3 on p. 28 of the complete report from which this
article is excerpted.)

E D U C AT I O N S TAT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , FA L L 2 0 0 1 23
Elementary and Secondary Education

Figure A.—Percentage of 8th- and 12th-graders who reported being assigned extra reading material almost every day or
once or twice a week: 1988 and 1998

1988

1998
Percent
40
8th-graders 12th-graders 37
34*
33

30 28

21*
20

13*
11
10
10

0
Almost every day Once or twice a week Almost every day Once or twice a week

*Significantly greater than 1988.

NOTE: “Extra reading material” means reading material not in regular textbook.
SOURCE: U.S. Department of Education, National Center for Education Statistics, National Assessment of Educational Progress (NAEP) 1988
and 1998 Civics Assessments. (Based on tables 3.1 and 3.2 on pp. 34 and 35 of the complete report from which this article is excerpted.)

Figure B.—Percentage of 8th- and 12th-graders who reported working on group projects almost every day or once or
twice a week: 1988 and 1998

Percent
40 1988

1998

30
8th-graders 12th-graders
25*
23*

20

13
11*
10
10 8*
5
3

0
Almost every day Once or twice a week Almost every day Once or twice a week

*Significantly greater than 1988.

SOURCE: U.S. Department of Education, National Center for Education Statistics, National Assessment of Educational Progress (NAEP) 1988
and 1998 Civics Assessments. (Based on tables 3.1 and 3.2 on pp. 34 and 35 of the complete report from which this article is excerpted.)

24 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
Homeschooling
Homeschooling in the United States: 1999
——————————————————————————————————
Stacey Bielick, Kathryn Chandler, and Stephen P. Broughman
This article was originally published as the Statistical Analysis Report of the same name. The Highlights and the Methodology and Technical Notes from
the original report have been omitted. The sample survey data are from the NCES National Household Education Surveys Program (NHES).

Background also be problematic. The first two efforts to estimate

Past estimates of the number of homeschoolers vary by homeschoolers at NCES—through the October supplement
hundreds of thousands of children. Brian Ray, president of to the 1994 Current Population Survey (CPS:Oct94) and
the National Home Education Research Institute, estimated through the Parent and Family Involvement in Education/
the number of homeschoolers to be around 1.15 million Civic Involvement Survey of the National Household
during the 1996–97 school year, and predicted that the Education Surveys Program, 1996 (PFI/CI-NHES:1996)—
number would grow to at least 1.3 million during 1999–2000 produced very different estimates. One problem that may
(Ray 1997). Patricia M. Lines, through her research at the have contributed to the varying estimates was the difference
U.S. Department of Education, National Institute on in how the two surveys identified students who were both
Student Achievement, Curriculum, and Assessment, homeschooled and enrolled in school part time. Neither
estimated the number of homeschoolers to be around survey collected precise data on these part-time home-
700,000 during 1995–96, possibly growing to 1 million by schoolers. An NCES technical report, Issues Related to
1997–98 (Lines 1999). Both Ray and Lines grant that their Estimating the Home-Schooled Population in the United States
estimates probably anchor the range within which the With National Household Survey Data, explores in detail the
actual number of homeschoolers could fall. differences in survey design and execution that may have
contributed to the disparity between the CPS:Oct94 and
The methods used by Ray and Lines in the development of PFI/CI-NHES:1996 estimates (Henke et al. 2000).
their estimates varied. Ray derived his most recent estimate
of the number of homeschoolers using his own 1995 survey In this report, the Parent Survey of the National Household
of homeschoolers and their use of curricular packages as his Education Surveys Program, 1999 (Parent-NHES:1999) is
base and sales of homeschooling curricular packages to used to estimate the number of homeschoolers in the
adjust for growth over time. Ray applied the ratio of users of United States, to describe the characteristics of home-
curricular packages and nonusers identified in the 1995 schoolers, and to document parents’ reasons for
survey to more recent sales of homeschool curricular homeschooling and parents’ reports of public school
packages to obtain his 1999–2000 estimate. Lines collected support for homeschoolers. Students were considered to be
data from all states that obtained records on homeschooling homeschooled if their parents reported them being schooled
children in both the 1990–91 and 1995–96 school years at home instead of at a public or private school, if their
(32 states and the District of Columbia). Using the 12 states enrollment in public or private schools did not exceed
with high record-collection rates for homeschoolers, based 25 hours a week, and if they were not being homeschooled
on Ray’s estimates of the percentage of homeschoolers who solely because of a temporary illness. The unweighted
reported filing in their state, Lines estimated the percentage number of homeschooled students used in this analysis is
of school-aged children who were homeschooling in those 275 and the unweighted number of nonhomeschooled
12 states. Lines estimated the number of children students is 16,833. Students are defined in this report as
homeschooled nationally by applying the percentage children ages 5 to 17 with a grade equivalent of kindergar-
distribution of homeschoolers from the state sample to ten through grade 12.
national totals of school-aged children.
Estimated Number of Homeschooled
The National Center for Education Statistics (NCES) was Students in the United States
the first organization to attempt to estimate the number of Approximately 850,000 students were being home-
homeschoolers in the United States using a rigorous sample schooled during the spring of 1999, according to data
survey of households. A household sampling frame circum- from the Parent-NHES:1999 (table 1). Homeschoolers
vents problems inherent in the use of incomplete sample accounted for 1.7 percent of students nationwide, ages 5
frames, such as customers of curricular providers and to 17, with a grade equivalent of kindergarten through
administrative records. Attempts to develop estimates of grade 12. The estimate includes students who were
homeschoolers through household surveys, however, can homeschooled while also enrolled in school for 25 hours

E D U C AT I O N S TAT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , FA L L 2 0 0 1 25
Elementary and Secondary Education

or less per week, and excludes students who were sample survey might produce a different estimate. A 95 per-
homeschooled due to a temporary illness. cent confidence interval defines a range of values such that
95 percent of the estimates from other similar surveys will fall
As with all sample surveys, the numbers and percentages in within the range of values. The 95 percent confidence interval
this report are estimates of the numbers and percentages in for the number of homeschoolers is 709,000 to 992,000. The
the population. Although 850,000 is the best population estimate provided here—850,000—is the midpoint of the
estimate available from this sample survey, another similar range. Figure 1 illustrates the confidence interval.

Table 1.—Number and percentage of homeschooled students, ages 5–17, with a grade equivalent of
kindergarten to grade 12, by school enrollment status: 1999

Number of
School enrollment status homeschooled students Percent s.e.

Total 850,000 100.0 0.14

Only homeschooled 697,000 82.0 2.94
Enrolled in school part time 153,000 18.0 2.94
Enrolled in school for less than 9 hours a week 107,000 12.6 2.81
Enrolled in school for 9 to 25 hours a week 46,000 5.4 1.50

NOTE: s.e. is standard error. Excludes students who were enrolled in school for more than 25 hours and students who were
homeschooled due to a temporary illness. Detail may not add to totals because of rounding.
SOURCE: U.S. Department of Education, National Center for Education Statistics, Parent Survey of the National Household
Education Surveys Program, 1999 (Parent-NHES:1999).

Figure 1.—Point estimate and 95 percent confidence interval for number

of homeschooled students, ages 5–17, with a grade equivalent
of kindergarten to grade 12: 1999

Number of home-
schooled students
1,100,000

1,000,000 992,000
900,000
850,000
800,000

700,000 709,000

600,000

500,000

400,000

300,000

200,000

100,000

0
1999

SOURCE: U.S. Department of Education, National Center for Education Statistics,

Parent Survey of the National Household Education Surveys Program, 1999
(Parent-NHES:1999).

26 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
 Homeschooling in the United States: 1999

Characteristics of Homeschooled Students between homeschoolers and nonhomeschoolers are dis-

and Their Families cussed in detail below.
Despite research limitations on documenting the number of
homeschoolers, recent research on homeschooling helps Grade equivalent of homeschooled students
suggest some characteristics of students and families who Homeschoolers distribute over the grade groupings in much
homeschool. An extensive 1998 study of homeschoolers, the same way as nonhomeschoolers (table 3). While it may
although based on a convenience sample, suggests that appear that a higher percentage of homeschoolers were in
homeschoolers differ from the general population in parents’ kindergarten compared to nonhomeschoolers, the differ-
educational attainment, household income, parents’ marital ence was not statistically significant.
status, and family size (Rudner 1999).1 Other research
suggests that although homeschooling in the United States Students’ race/ethnicity and sex
may once have been primarily a trend within a homoge- A greater percentage of homeschoolers compared to
neous subgroup of White, middle-class, Christian families, nonhomeschoolers were White, non-Hispanic—75 percent
growth in homeschooling may be reaching a broader range compared to 65 percent. At the same time, a smaller
of American families and values (McDowell, Sanchez, and percentage of homeschoolers were Black, non-Hispanic
Jones 2000; Lines 2000a; Welner and Welner 1999). students and a smaller percentage were Hispanic students.
Girls and boys were equally represented among home-
The Parent-NHES:1999 provides descriptive data about the schoolers and nonhomeschoolers.
characteristics of homeschoolers in the United States and
their families. This report includes students who were Number of children living in the household
homeschooled only and students who were homeschooled A much greater percentage of homeschoolers than non-
and enrolled in school for 25 hours or less per week. As homeschoolers came from families with three or more
shown in table 1, about four out of five homeschoolers were children—62 percent of homeschooled students were part
homeschooled only (82 percent) and one out of five of families with three or more children compared to 44
homeschoolers were enrolled in public or private schools percent of nonhomeschoolers. Homeschoolers were just as
part time (18 percent). likely as nonhomeschoolers to be an only child and were
less likely than nonhomeschoolers to have just one sibling.
Table 2 shows the number of all students by selected
characteristics, the number of homeschooled students by Number of parents living in the household and labor
those same characteristics, and for each characteristic the force participation
percentage of students who are homeschooled. As shown in In order to homeschool, parents may need to dedicate a
table 2, the percentage of students who were homeschooled significant amount of time to schooling their children.
in 1999 differed based on various characteristics of students Because of the time required, homeschooling usually
and their families. Depending on these student and family involves two parents—one who participates in the labor
characteristics, the percentage of homeschoolers among force and one who homeschools. Rudner (1999) found that
students ranged from 0.7 to 4.6 percent. Characteristics that 97 percent of homeschooling parents were married couples.
distinguished high percentages of homeschooling were two- The Parent-NHES:1999 shows the percentage of home-
parent families, especially when only one parent partici- schooled students living in two-parent households was
pated in the labor force; large family size; and parents’ high much higher than the percentage for nonhomeschoolers—
educational attainment. The percentage of students who 80 percent of homeschooled students lived in two-parent
were homeschooled was similar for both boys and girls; families compared to 66 percent for nonhomeschoolers.
across elementary, middle, and high school grades; and In addition, 52 percent of homeschoolers came from
across the four income ranges used in the analysis. two-parent families where only one parent was partici-
pating in the labor force compared to 19 percent for
Table 3 further explores the characteristics that distinguish nonhomeschoolers.
homeschoolers by comparing the characteristics of home-
schoolers to those of all students and to the characteristics Household income
of nonhomeschoolers. The similarities and differences Although Rudner found that the median household income
of homeschooling families was higher than the median
1
Rudner’s study is based on a survey administered by Bob Jones University to a household income of families with children nationwide, the
sample drawn from parents who used the university’s standardized testing program
(Welner and Welner 1999). Parent-NHES:1999 indicates that the household income of

E D U C AT I O N S TAT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , FA L L 2 0 0 1 27
Elementary and Secondary Education

Table 2.—Number of students and number and percent of homeschooled students, ages 5–17, with a grade equivalent of
kindergarten to grade 12, by selected characteristics: 1999

Homeschooled students
Number of
Characteristic students Number Percent of all students s.e.

Total 50,188,000 850,000 1.7 0.14

Grade equivalent1
K–5 24,428,000 428,000 1.8 0.20
Kindergarten 3,790,000 92,000 2.4 0.52
Grades 1–3 12,692,000 199,000 1.6 0.29
Grades 4–5 7,946,000 136,000 1.7 0.28
Grades 6–8 11,788,000 186,000 1.6 0.24
Grades 9–12 13,954,000 235,000 1.7 0.24

Race/ethnicity
White, non-Hispanic 32,474,000 640,000 2.0 0.19
Black, non-Hispanic 8,047,000 84,000 1.0 0.31
Hispanic 7,043,000 77,000 1.1 0.25
Other 2,623,000 49,000 1.9 0.65

Sex
Female 24,673,000 434,000 1.8 0.19
Male 25,515,000 417,000 1.6 0.17

Number of children in the household

One child 8,226,000 120,000 1.5 0.24
Two children 19,883,000 207,000 1.0 0.14
Three or more children 22,078,000 523,000 2.4 0.30

Number of parents in the household

Two parents 33,007,000 683,000 2.1 0.21
One parent 15,454,000 142,000 0.9 0.16
Nonparental guardians 1,727,000 25,000 1.4 0.82

Parents’ participation in the labor force

Two parents—one in labor force 9,628,000 444,000 4.6 0.55
Two parents—both in labor force 22,880,000 237,000 1.0 0.17
One parent—in labor force 13,907,000 98,000 0.7 0.16
No parent in labor force 3,773,000 71,000 1.9 0.48

Household income
$25,000 or less 16,776,000 262,000 1.6 0.27
$25,001–50,000 15,220,000 278,000 1.8 0.24
$50,001–75,000 8,576,000 162,000 1.9 0.30
$75,001 or more 9,615,000 148,000 1.5 0.28

Parents’ highest educational attainment

High school diploma or less 18,334,000 160,000 0.9 0.15
Voc/tech degree or some college 15,177,000 287,000 1.9 0.25
Bachelor’s degree 8,269,000 213,000 2.6 0.42
Graduate/professional school 8,407,000 190,000 2.3 0.46

Urbanicity2
City 31,178,000 455,000 1.5 0.16
Town 6,237,000 120,000 1.9 0.39
Rural 12,773,000 275,000 2.2 0.31

1
Students whose grade equivalent was “ungraded” were excluded from the grade analysis.
2
Urbanicity is based on a U.S. Census classification of places as urban or rural. City is a place that is urban, inside an urban area; town is a
place that is urban, outside an urban area; rural is a place not classifed as urban.
NOTE: s.e. is standard error. Detail may not add to totals because of rounding. Number and percent of homeschoolers excludes students
who were enrolled in school for more than 25 hours and students who were homeschooled due to a temporary illness.
SOURCE: U.S. Department of Education, National Center for Education Statistics, Parent Survey of the National Household Education
Surveys Program, 1999 (Parent-NHES:1999).

28 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
 Homeschooling in the United States: 1999

Table 3.—Distribution of all students, homeschooled students, and nonhomeschooled students, ages 5–17, with a grade
equivalent of kindergarten to grade 12, by selected characteristics: 1999

Number of All students Homeschoolers1 Nonhomeschoolers

Characteristic students Percent s.e. Percent s.e. Percent s.e.

Total 50,188,000 100.0 100.0 100.0

Grade equivalent2
K–5 24,428,000 48.7 0.07 50.4 3.75 48.7 0.09
Kindergarten 3,790,000 7.6 0.04 10.8 2.31 7.5 0.05
Grades 1–3 12,692,000 25.3 0.04 23.5 3.61 25.3 0.07
Grades 4–5 7,946,000 15.8 0.02 16.0 2.34 15.8 0.05
Grades 6–8 11,788,000 23.5 0.04 21.9 2.83 23.5 0.06
Grades 9–12 13,954,000 27.8 0.10 27.7 3.21 27.8 0.11

Race/ethnicity
White, non-Hispanic 32,474,000 64.7 0.32 75.3 3.36 64.5 0.33
Black, non-Hispanic 8,047,000 16.0 0.20 9.9 2.80 16.1 0.21
Hispanic 7,043,000 14.0 0.17 9.1 2.06 14.1 0.17
Other 2,623,000 5.2 0.23 5.8 2.01 5.2 0.23

Sex
Female 24,673,000 49.2 0.47 51.0 3.27 49.1 0.47
Male 25,515,000 50.8 0.47 49.0 3.27 50.9 0.47

Number of children in the household

One child 8,226,000 16.4 0.30 14.1 2.53 16.4 0.30
Two children 19,883,000 39.6 0.42 24.4 3.06 39.9 0.42
Three or more children 22,078,000 44.0 0.48 61.6 3.97 43.7 0.49

Number of parents in the household

Two parents 33,007,000 65.8 0.41 80.4 3.26 65.5 0.42
One parent 15,454,000 30.8 0.41 16.7 2.91 31.0 0.42
Nonparental guardians 1,727,000 3.4 0.17 2.9 1.70 3.5 0.17

Parents’ participation in the labor force

Two parents—one in labor force 9,628,000 19.2 0.39 52.2 4.27 18.6 0.39
Two parents—both in labor force 22,880,000 45.6 0.48 27.9 3.92 45.9 0.48
One parent—in labor force 13,907,000 27.7 0.44 11.6 2.53 28.0 0.44
No parent in labor force 3,773,000 7.5 0.32 8.3 2.21 7.5 0.32

Household income
$25,000 or less 16,776,000 33.4 0.22 30.9 4.31 33.5 0.22
$25,001–50,000 15,220,000 30.3 0.47 32.7 4.00 30.3 0.47
$50,001–75,000 8,576,000 17.1 0.38 19.1 2.62 17.1 0.38
$75,001 or more 9,615,000 19.2 0.42 17.4 2.65 19.2 0.42

Parents’ highest educational attainment

High school diploma or less 18,334,000 36.5 0.43 18.9 2.88 36.8 0.43
Voc/tech degree or some college 15,177,000 30.2 0.43 33.7 3.85 30.2 0.44
Bachelor’s degree 8,269,000 16.5 0.36 25.1 3.49 16.3 0.35
Graduate/professional school 8,407,000 16.8 0.41 22.3 4.17 16.7 0.40

Urbanicity3
City 31,178,000 62.1 0.36 53.5 4.13 62.3 0.36
Town 6,237,000 12.4 0.34 14.2 2.59 12.4 0.35
Rural 12,773,000 25.5 0.23 32.4 3.81 25.3 0.23

1
Excludes students who were enrolled in school for more than 25 hours and students who were homeschooled due to a temporary illness.
2
Students whose grade equivalent was “ungraded” were excluded from the grade analysis.
3
Urbanicity is based on a U.S. Census classification of places as urban or rural. City is a place that is urban, inside an urban area; town is a place that
is urban, outside an urban area; rural is a place not classifed as urban.
NOTE: s.e. is standard error. Detail may not add to totals because of rounding.
SOURCE: U.S. Department of Education, National Center for Education Statistics, Parent Survey of the National Household Education Surveys
Program, 1999 (Parent-NHES:1999).

E D U C AT I O N S TAT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , FA L L 2 0 0 1 29
Elementary and Secondary Education

homeschoolers, reported in ranges from less than $25,000 reasons, and poor school environment. Figure 2 shows 10
to over $75,000, is the same as the household income of reasons cited by at least 5 percent of students’ parents.
nonhomeschoolers. The same percentage of homeschooled Additional reasons are listed in table 4.
and nonhomeschooled students lived in households with
annual incomes of $50,000 or less (64 percent).2 Public School Support for Homeschooled
Students
Parents’ highest educational attainment Public schools or school districts sometimes offer support for
Parents’ highest educational attainment, however, was homeschoolers by providing parents with a curriculum,
clearly associated with homeschooling. Parents of home- books and materials, places to meet, and the opportunity for
schoolers had higher levels of educational attainment than homeschooled children to attend classes and participate in
did parents of nonhomeschoolers. Table 3 shows that 37 per- extracurricular activities at the school. Previous research
cent of parents of nonhomeschoolers did not complete any found that only a small percentage of homeschoolers enrolled
schooling beyond high school, compared to 19 percent of in classes, used textbooks, or used libraries when they were
parents of homeschoolers. Conversely, 25 percent of parents made available by public schools and that many home-
of homeschoolers attained bachelor’s degrees as their schoolers express antipathy toward using public school
highest degree, compared to 16 percent of parents of support (Lines 2000b; Yeager 1999; Mayberry et al. 1995).
nonhomeschoolers.
Table 5 shows the different types of public school support
Urbanicity for homeschoolers asked about in the Parent-NHES:1999.
Urbanicity refers to the classification of households as Parents of homeschoolers were asked whether their child’s
urban or rural. There are two classifications of urban, which assigned school or district offered any of the eight pre-
are referred to in this report as cities and towns. Places not specified types of support shown in table 5. The estimates
classified as urban are rural. The percentage of home- are based on parents’ reports of public school support and
schoolers living in a city was about 9 percentage points use, not what schools or districts may actually offer. Be-
lower than the percentage for nonhomeschoolers (53 and tween 15 and 38 percent of homeschoolers’ parents did not
62 percent, respectively). There were no statistically signifi- know whether various types of support were offered.
cant differences between the percentages of homeschoolers
and nonhomeschoolers living in towns or rural areas. The first two columns of estimates in table 5 show that,
altogether, 28 percent of homeschoolers’ parents reported that
Parents’ Reasons for Homeschooling public schools or districts offered extracurricular activities,
Parents may homeschool their children for a number of 21 percent reported curriculum support, and 23 percent
reasons. Previous studies suggest that the most common reported books and materials. Between about 3 and 11 percent
reasons that parents give for homeschooling their children of homeschoolers’ parents said that support was available
are moral or religious reasons, a desire for high educational and that they used the support, and between about 5 and
achievement, dissatisfaction with public schools’ instruc- 22 percent said that the support was available but they did not
tional program, and concerns about school environment, use it. For example, as table 5 shows, about 6 percent of
including safety, drugs, and peer pressure (Lines 2000a; homeschoolers’ parents reported that they had the chance to
Grubb 1998; Mayberry 1991). attend extracurricular activities and used this type of support,
and 11 percent reported that schools offered books and
Parents gave a wide range of reasons for homeschooling in materials and that they used this type of public school support.
the Parent-NHES:1999.3 Parents were asked to list their
reasons for homeschooling and could provide as many Future Research Plans
reasons as applied. The reasons parents gave were coded NCES plans to collect and report data about homeschoolers
into 16 categories and included better education, religious with future Parent and Family Involvement in Education
Surveys (PFI), slated to occur on a 4-year cycle next sched-
2
An additional analysis of household income in two-parent families where only one
uled for 2003 as part of the National Household Education
parent was participating in the labor force also shows no difference between Surveys Program (NHES). Future Parent and Family Involve-
homeschoolers and nonhomeschoolers (data not shown in tables).
ment in Education Surveys will provide a comprehensive set
3
The unit of analysis in the Parent-NHES:1999 is the student, not the parent. In each
household, up to two children may have been sampled for the survey. In the Parent- of information that may be used to estimate the number and
NHES:1999, there were 30 households in which parents completed interviews about characteristics of homeschoolers in the United States. Future
two homeschooled children. In 16 of those cases, the parents gave the exact same
reasons for homeschooling for both children. areas of inquiry might also include items on homeschoolers’

30 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
 Homeschooling in the United States: 1999

Figure 2.—Ten reasons for homeschooling and the percentage of homeschooled students whose parents gave each reason: 1999

Can give child better education at home 48.9

Religious reasons 38.4

Poor learning environment at school 25.6

Family reasons 16.8

To develop character/morality 15.1

Object to what school teaches 12.1

School does not challenge child 11.6

Other problems with available schools 11.5

Student behavior problems at school 9.0

Child has special needs/disability 8.2

0.0 20.0 40.0 60.0 80.0 100.0

Percent

NOTE: Percentages do not add to 100 percent because respondents could give more than one reason.
SOURCE: U.S. Department of Education, National Center for Education Statistics, Parent Survey of the National Household Education Surveys Program, 1999
(Parent-NHES:1999).

Table 4.—Number and percentage of homeschooled students, by reason for homeschooling: 1999

Number of
Reasons for homeschooling homeschooled students Percent s.e.

Can give child better education at home 415,000 48.9 3.79

Religious reasons 327,000 38.4 4.44
Poor learning environment at school 218,000 25.6 3.44
Family reasons 143,000 16.8 2.79
To develop character/morality 128,000 15.1 3.39
Object to what school teaches 103,000 12.1 2.11
School does not challenge child 98,000 11.6 2.39
Other problems with available schools 98,000 11.5 2.20
Student behavior problems at school 76,000 9.0 2.40
Child has special needs/disability 69,000 8.2 1.89
Transportation/convenience 23,000 2.7 1.48
Child not old enough to enter school 15,000 1.8 1.13
Want private school but cannot afford it 15,000 1.7 0.77
Parent’s career 12,000 1.5 0.80
Could not get into desired school 12,000 1.5 0.99
Other reasons* 189,000 22.2 2.90

*Parents homeschool their children for many reasons that are often unique to their family situation.“Other
reasons” parents gave for homeschooling in the Parent-NHES:1999 included the following: It was the child’s
choice; to allow parents more control over what their children were learning; flexibility; and parents wanted
year-round schooling.
NOTE: s.e. is standard error. Excludes students who were enrolled in school for more than 25 hours and stu-
dents who were homeschooled due to a temporary illness. Percentages do not add to 100 percent because
respondents could choose more than one reason.
SOURCE: U.S. Department of Education, National Center for Education Statistics, Parent Survey of the National
Household Education Surveys Program, 1999 (Parent-NHES:1999).

E D U C AT I O N S TAT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , FA L L 2 0 0 1 31
Elementary and Secondary Education

Table 5.—Percentage of homeschooled students whose parents reported availability and use of support from public schools or districts: 1999

Available and not used Available and used Not available Don’t know if available
Type of support Percent s.e. Percent s.e. Percent s.e. Percent s.e.

Curriculum 12.4 2.35 8.1 2.20 49.0 3.55 30.5 3.79

Books/materials 12.2 2.22 10.6 2.47 50.0 3.92 27.3 3.95
Place for parents to meet or get information 8.9 1.77 6.4 1.57 63.5 3.57 21.2 2.97
Web site for parents # # # # 53.7 4.84 37.5 4.75
Place for students to meet 4.7 1.26 7.0 1.66 69.0 3.29 19.3 2.77
Web site for students # # # # 60.2 4.30 34.5 4.21
Extracurricular activities 21.5 2.85 6.4 1.90 56.4 3.99 15.8 2.95
Chance to attend some classes* 16.5 2.91 2.8 1.32 49.4 4.09 31.3 4.10

*Data not available for students who attended private schools part time and for students who attended public schools for less than 9 hours. Estimates are based on the
number of full-time homeschoolers reporting and the number of students “using” public schools for 9 to 25 hours.
#Too few cases for a reliable estimate.
NOTE: s.e. is standard error. Excludes students who were enrolled in school for more than 25 hours and students who were homeschooled due to a temporary illness.
SOURCE: U.S. Department of Education, National Center for Education Statistics, Parent Survey of the National Household Education Surveys Program, 1999
(Parent-NHES:1999).

use of distance learning and the Internet, information about McDowell, S.A., Sanchez, A.R., and Jones, S.S. (2000). Participa-
families’ past use of homeschooling, more information about tion and Perception: Looking at Home Schooling Through a
homeschoolers who attend school for some classes or sub- Multicultural Lens. Peabody Journal of Education, 75(1,2):
124–46.
jects, and specific information about homeschoolers’ plans
Ray, B. (1997). Strengths of Their Own: Home Schoolers Across
for postsecondary education.
America. Salem, OR: National Home Education Research
Institute Publications.
References
Rudner, L. (1999). Scholastic Achievement and Demographic
Grubb, D. (1998, November). Homeschooling: Who and Why? Characteristics of Home School Students in 1998. Education
Paper presented at the 27th annual meeting of the Mid-South Policy Analysis Archives, 7(8). Available: http://epaa.asu.edu/
Educational Research Association, New Orleans, LA. epaa/v7n8/
Henke, R.R., Kaufman, P., Broughman, S.P., and Chandler, K. Welner, K.M., and Welner, K.G. (1999). Contextualizing
(2000). Issues Related to Estimating the Home-Schooled Population Homeschooling Data: A Response to Rudner. Education Policy
in the United States With National Household Survey Data (NCES Analysis Archives, 7(13). Available: http://epaa.asu.edu/epaa/
2000–311). U.S. Department of Education, National Center for v7n13.html
Education Statistics. Washington, DC: U.S. Government
Printing Office. Yeager, E.T. (1999). A Study of Cooperation Between Home Schools
and Public and Private Schools, K–12. Unpublished doctoral
Lines, P. (2000a). Homeschooling Comes of Age. The Public dissertation, Texas A&M University–Commerce.
Interest, Summer 2000 (No. 140): 74–85.
Lines, P. (2000b). When Home Schoolers Go to School: A
Partnership Between Families and Schools. Peabody Journal of Data source: Parent Survey of the NCES National Household
Education, 75(1,2): 159–86. Education Surveys Program, 1999 (Parent-NHES:1999).
Lines, P. (1999). Homeschoolers: Estimating Numbers and Growth. For technical information, see the complete report:
Web edition. Washington, DC: U.S. Department of Education, Bielick, S., Chandler, K., and Broughman, S.P. (2001). Homeschooling in
Office of Educational Research and Improvement. Available: the United States: 1999 (NCES 2001–033).
http://www.ed.gov/offices/OERI/SAI/homeschool/homeschoolers.pdf Author affiliations: S. Bielick, Education Statistics Services Institute
(ESSI); K. Chandler and S.P. Broughman, NCES.
Mayberry, M. (1991, April). Conflict and Social Determinism: The
Reprivatization of Education. Paper presented at the American For questions about content, contact Stephen P. Broughman
(stephen.broughman@ed.gov).
Educational Research Association Meeting, Chicago, IL.
To obtain the complete report (NCES 2001–033), call the toll-free
Mayberry, M., Knowles, J.G., Ray, B., and Marlow, S. (1995). ED Pubs number (877–433–7827) or visit the NCES Web Site
Homeschooling: Parents as Educators. Thousand Oaks, CA: (http://nces.ed.gov).
Corwin Press, Inc.

32 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
Teacher Preparation
Teacher Preparation and Professional Development: 2000
——————————————————————————————————
Basmat Parsad, Laurie Lewis, and Elizabeth Farris
This article was originally published as the Executive Summary of the E.D. Tabs report of the same name. The sample survey data are from two surveys
conducted through the NCES Fast Response Survey System (FRSS): “Survey on Professional Development and Training in U.S. Public Schools: 1999–
2000” and “Teacher Survey on Professional Development and Training.”

Introduction ■ Newer teachers were less likely than more experi-

Concerns about the quality of the nation’s public education enced teachers to report having a master’s degree,
system have increased attention to key elements of teacher ranging from 20 percent of teachers with 3 or fewer
effectiveness within recent years (Darling-Hammond 2000; years of teaching experience to 54 percent of teachers
Lewis et al. 1999; Mayer, Mullens, and Moore 2001; with 10 or more years of teaching experience.
National Commission on Teaching and America’s Future
1996). While there is little consensus on what constitutes Teacher professional development
high-quality teachers, past research has emphasized two Formal professional development and collaboration with
broad dimensions of teacher effectiveness: (1) the level of other teachers are key mechanisms for providing teachers
knowledge and skills that teachers bring to the classroom, with ongoing training opportunities (Henke, Chen, and
as measured by teacher preparation and qualifications, and Geis 2000; National Commission on Teaching and
(2) classroom practices. In 1998, the National Center for America’s Future 1996; Sprinthall, Reiman, and Theis-
Education Statistics (NCES) conducted a survey through its Sprinthall 1996). Formal professional development as
Fast Response Survey System (FRSS) to provide a national commonly practiced, typically consisting of school and
profile on the first dimension of teacher quality—teacher district staff development programs, however, has been
preparation and qualifications (Lewis et al. 1999). criticized for being short term and lacking in continuity and
adequate follow-up (Fullan 1991; Lewis et al. 1999;
In 2000, NCES conducted a second FRSS survey to revisit Mullens et al. 1996). Results of the 2000 survey indicate
the issue of teacher preparation and qualifications and that during the 12 months preceding the survey
measure change since 1998. The sample for the 2000 survey ■ Public school teachers were most likely to have
consisted of 5,253 full- and part-time teachers in regular participated in professional development that focused
elementary, middle, and high schools in the 50 states and on state or district curriculum and performance
the District of Columbia. The survey repeated some of the standards (80 percent; table A). More than one-half
indicators of teacher quality examined in the 1998 survey, participated in professional development programs
in addition to exploring issues such as follow-up to profes- focused on the integration of educational technology
sional development. Specifically, this survey provides a into the grade or subject taught (74 percent), in-
national profile on (1) teacher education, (2) teacher depth study in the subject area of the main teaching
participation in formal professional development and assignment (72 percent), implementing new methods
collaborative activities related to teaching, and (3) teachers’ of teaching (72 percent), and student performance
feelings of preparedness for various classroom demands. assessment (62 percent). Teachers were less likely to
This report summarizes key findings from the 2000 survey have participated in professional development that
and also makes comparisons with the 1998 data. focused on addressing the needs of students with
disabilities (49 percent); encouraging parent and
Key Findings community involvement (46 percent); classroom
Teacher education management, including student discipline (45 per-
One measure of teacher education is the type of degree cent); and addressing the needs of students from
held, including advanced degrees. Findings from the 2000 diverse cultural backgrounds (41 percent). The
survey indicate that professional development area in which teachers
■ Virtually all public school teachers had a bachelor’s were least likely to participate was addressing the
degree, and 45 percent held a master’s degree. One needs of students with limited English proficiency
percent each held a doctorate or some other degree, (26 percent).*
and 18 percent reported having other certificates. *The estimate for teacher participation in professional development on addressing
the needs of students with limited English proficiency was based on all public school
teachers rather than teachers who taught students with those needs.

E D U C AT I O N S TAT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , FA L L 2 0 0 1 33
Elementary and Secondary Education

Table A.—Percent of public school teachers who participated in professional development activities during the last 12 months that focused on various
content areas, by number of hours spent on the activity: 2000

Total hours spent

Content area Participated in activity 1 to 8 9 to 32 More than 32

State or district curriculum and performance standards 80 57 31 12

Integration of educational technology into the grade or subject taught 74 61 28 11
In-depth study in the subject area of main teaching assignment 72 43 34 23
New methods of teaching (e.g., cooperative learning) 72 59 29 11
Student performance assessment (e.g., methods of testing,
applying results to modify instruction) 62 67 25 8
Addressing the needs of students with disabilities 49 72 19 8
Encouraging parent and community involvement 46 75 18 8
Classroom management, including student discipline 45 73 20 7
Addressing the needs of students from diverse cultural backgrounds 41 71 20 9
Addressing the needs of students with limited English proficiency 26 68 20 12

NOTE: Percentages for total hours spent in the activity are based on public school teachers who participated in professional development over the 12 months preceding the
survey. Percents are computed across each row, but may not add to 100 because of rounding.
SOURCE: U.S. Department of Education, National Center for Education Statistics, Fast Response Survey System, “Survey on Professional Development and Training in U.S.
Public Schools, 1999–2000,” FRSS 74, 2000. (Originally published as table 2 on p. 15 of the complete report from which this article is excerpted.)

■ For all but one content area of professional develop- interest (52 percent). Teachers were least likely to
ment, teachers typically reported that they had spent mentor another teacher in a formal relationship
1 to 8 hours, or the equivalent of 1 day or less, on the (26 percent) or to be mentored by another teacher
activity during the 12 months preceding the survey (23 percent).
(table A). In-depth study in the subject area of the ■ Frequency of participation in a collaborative activity
main teaching assignment was the only area of was generally positively related to teachers’ beliefs
professional development in which participation about the extent to which the activity improved their
typically lasted more than 8 hours. classroom teaching. For example, teachers who
■ The number of hours teachers spent in professional engaged in regularly scheduled collaboration with
development activities was related to the extent to other teachers at least once a week were more likely
which they believed that participation improved their to believe that participation had improved their
teaching. For every content area examined in the teaching a lot (45 percent), compared with teachers
survey, teachers who participated for more than 8 who participated two to three times a month
hours were more likely than those who spent 1 to (23 percent), once a month (15 percent), or a few
8 hours to report that participation improved their times a year (7 percent).
teaching a lot.
Teachers’ feelings of preparedness
Teacher collaboration Teachers in the 2000 survey reported the extent to which
Collaboration with other teachers may revolve around joint they felt prepared for the overall demands of their teaching
work (e.g., team teaching and mentoring) and teacher assignments and for eight specific classroom activities. The
networks (e.g., school-to-school and school-university survey data indicate that
partnerships). The 2000 survey findings indicate that ■ Sixty-one percent of public school teachers felt very
■ The most frequently attended collaborative activity well prepared to meet the overall demands of their
among public school teachers was collaboration with teaching assignments. Thirty-five percent felt moder-
other teachers (69 percent). This activity was ately well prepared, and 4 percent felt somewhat well
followed by networking with teachers outside their prepared.
school (62 percent), a common planning period for ■ Teachers most often reported feeling very well
team teachers (53 percent), and individual or prepared to maintain order and discipline in the
collaborative research on a topic of professional

34 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
 Teacher Preparation and Professional Development: 2000

classroom (71 percent). They were less likely to ■ The proportion of regular full-time public school
report feeling very well prepared to implement new teachers indicating that they participated in profes-
methods of teaching (45 percent), implement state or sional development was lower in 2000 than in 1998
district curriculum (44 percent), use student perfor- for three of the seven content areas that were compa-
mance assessment (37 percent), address the needs rable across years—new methods of teaching
of students from diverse cultural backgrounds (73 vs. 77 percent), student performance assessment
(32 percent), and integrate educational technology (62 vs. 67 percent), and classroom management,
into the grade or subject taught (27 percent). including student discipline (43 vs. 49 percent).
■ Among teachers who taught students with special ■ In 1998 and 2000, participation of regular full-time
needs, relatively few felt very well prepared to public school teachers in professional development
address those students’ needs. Twenty-seven percent was likely to be short term, typically lasting for 1 to
of teachers indicated that they felt very well prepared 8 hours. This pattern held for every content area of
to address the needs of students with limited English professional development examined in the surveys
proficiency, and 32 percent of the teachers who except for programs on in-depth study in the subject
taught students with disabilities felt very well area of the main teaching assignment, where partici-
prepared to address those students’ needs. pation typically lasted more than 8 hours.
■ The extent to which teachers felt very well prepared ■ In 1998 and 2000, regular full-time public school
for most classroom activities varied with the amount teachers most often reported that they felt very well
of time spent in recent professional development in prepared to maintain order and discipline in the
those activities. With two exceptions (classroom classroom (71 and 72 percent, respectively). In both
management and state or district curriculum and years, teachers were least likely to report feeling very
performance standards), teachers who spent over well prepared to integrate educational technology
8 hours in professional development on the activity into the grade or subject taught (20 and 27 percent,
were more likely than those who spent 1 to 8 hours respectively) and address the needs of students with
or those who did not participate at all to indicate that disabilities (21 and 29 percent, respectively).
they felt very well prepared for that activity. ■ For all but one classroom activity examined in the
■ For three collaborative activities related to teaching— surveys, regular full-time public school teachers in
regularly scheduled collaboration with other teach- 2000 were more likely than those in 1998 to report
ers, networking with teachers outside the school, and that they felt very well prepared. The exception was
mentoring another teacher in a formal relationship— maintaining order and discipline in the classroom.
teachers who participated in the activity were more
likely than those who did not participate to report References
feeling very well prepared for the overall demands of Darling-Hammond, L. (2000). Teacher Quality and Student
their classroom assignments. Achievement: A Review of State Policy Evidence. Education
Policy Analysis Archives, 8(1). Available: http://olam.ed.asu.edu/
epaa/v8n1/
Selected comparisons with the 1998 survey
Fullan, M. (with Steigelbauer, S.). (1991). The New Meaning of
The 2000 survey was designed to provide trend data that
Educational Change. New York: Teacher’s College Press.
would allow an examination of change since 1998 along
Henke, R., Chen, X., and Geis, S. (2000). Progress Through the
two key dimensions—teacher participation in professional
Teacher Pipeline: 1992–93 College Graduates and Elementary/
development and collaborative activities, and teachers’ Secondary School Teaching as of 1997 (NCES 2000–152). U.S.
feelings of preparedness. For these analyses, a subset of Department of Education, National Center for Education
teachers was selected from the 2000 survey that was similar Statistics. Washington, DC: U.S. Government Printing Office.
to the teachers sampled for the 1998 survey—that is, Lewis, L., Parsad, B., Carey, N., Bartfai, N., Farris, E., and
regular full-time public school teachers in grades 1 through Smerdon, B. (1999). Teacher Quality: A Report on the Preparation
12 whose main teaching assignment was in English, and Qualifications of Public School Teachers (NCES 1999–080).
mathematics, social studies, foreign languages, or science, U.S. Department of Education, National Center for Education
Statistics. Washington, DC: U.S. Government Printing Office.
or who taught in a self-contained classroom. Findings from
the 1998 and 2000 surveys indicate that

E D U C AT I O N S TAT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , FA L L 2 0 0 1 35
Elementary and Secondary Education

Mayer, D., Mullens, J., and Moore, M. (2001). Monitoring School

Data sources: The NCES Fast Response Survey System: “Survey on
Quality (NCES 2001–030). U.S. Department of Education, Professional Development and Training in U.S. Public Schools: 1999–
National Center for Education Statistics. Washington, DC: U.S. 2000,” FRSS 74, 2000; and “Teacher Survey on Professional
Government Printing Office. Development and Training,” FRSS 65, 1998.
Mullens, J., Leighton, M., Laguarda, K., and O’Brien, E. (1996). For technical information, see the complete report:
Student Learning, Teaching Quality, and Professional Development: Parsad, B., Lewis, L., and Farris, E. (2001). Teacher Preparation and
Theoretical Linkages, Current Measurement, and Recommendations Professional Development: 2000 (NCES 2001–088).
for Future Data Collection (NCES 96–28). U.S. Department of Author affiliations: B. Parsad, L. Lewis, and E. Farris, Westat.
Education. Washington, DC: National Center for Education For questions about content, contact Bernard Greene
Statistics Working Paper. (bernard.greene@ed.gov).
To obtain the complete report (NCES 2001–088), call the toll-free
National Commission on Teaching and America’s Future. (1996). ED Pubs number (877–433–7827) or visit the NCES Web Site
What Matters Most: Teaching for America’s Future. New York: (http://nces.ed.gov).
NCTAF.
Sprinthall, N., Reiman, A., and Theis-Sprinthall, L. (1996).
Teacher Professional Development. In J. Sikula, T. Buttery, and
E. Guyton (Eds.), Handbook of Research on Teacher Education.
New York: Simon & Schuster Macmillan.

36 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
Private School Universe
Private School Universe Survey: 1999–2000
—————————————————————————————————
Stephen P. Broughman and Lenore A. Colaciello
This article was originally published as the Introduction and Selected Results of the Statistical Analysis Report of the same name. The universe data are
from the NCES Private School Survey (PSS).

Introduction statistically different from the 27,402 schools counted in the

This report on the private school universe presents data on fall of 1997 (Broughman and Colaciello 1999). Among
schools with grades kindergarten through 12 by school size, these schools there was considerable diversity as to orienta-
school level, religious orientation, geographical region, tion and affiliation. Of the three primary types of private
community type, and program emphasis. The numbers of schools—Catholic, other religious, and nonsectarian—other
students and teachers are reported in the same categories. religious schools were the most numerous, followed by
The number of students is also reported by race/ethnicity, Catholic schools and then nonsectarian schools, represent-
gender, and grade level. ing 49, 30, and 22 percent of all private schools, respec-
tively (table 1 and figure 1). Parochial schools were the
Tables in the complete report present data by three most numerous type of Catholic schools, followed by
classification schemes: private school typology, religious diocesan and then private order schools. Among the three
orientation, and association membership. The private categories of other religious schools—conservative Chris-
school nine-category typology is based on methodological tian, affiliated, and unaffiliated—there were fewer affiliated
work completed at the National Center for Education schools than conservative Christian or unaffiliated schools.
Statistics (NCES). Each of the primary divisions (Catholic, Of the nonsectarian schools, regular schools were the most
other religious, and nonsectarian) is subdivided into three numerous, followed by special emphasis schools and then
additional categories: Catholic into parochial, diocesan, and special education schools.
private order;1 other religious into conservative Christian,
affiliated with a national denomination or other religious The region with the most private schools was the South
school association, and unaffiliated; and nonsectarian into (30 percent), while the region with the fewest was the
regular program, special emphasis, and special education. West (20 percent). Ninety-one percent of private schools
offered at least some elementary grades, with 61 percent
The Private School Survey (PSS), conducted biennially by offering elementary grades only and 30 percent offering a
the U.S. Bureau of the Census for NCES, is designed to combination of elementary and secondary grades; the
collect data from all private schools in the 50 states and the remaining 9 percent offered secondary grades only. Most
District of Columbia. The PSS conducted in 1999–2000 is private schools (82 percent) emphasized a regular ele-
the data source for this report. The counts presented here mentary/secondary program. The other program emphasis
are estimates derived from a collection of state and private categories—Montessori, special emphasis, special educa-
school organization and association lists combined with tion, vocational/technical, alternative, and early child-
an area frame. (An estimate of the undercount of schools hood—each contained fewer than 10 percent of private
using this methodology is given in the complete report.) schools.
Although, beginning in 1995, the PSS definition of a school
was expanded to include those schools for which kindergar- Enrollment
ten was the highest grade, referred to as kindergarten- A total of 5,162,684 students were enrolled in the nation’s
terminal (k-terminal) schools, all estimates presented in private schools in the fall of 1999, an increase over the
this report, unless otherwise stated, are for schools (tradi- 5,076,119 students enrolled in the fall of 1997 (Broughman
tional schools) meeting the more restrictive pre-1995 PSS and Colaciello 1999). Private school students represented
definition of having at least one of grades 1 through 12. approximately 10 percent of the total elementary and
secondary enrollment in the United States.2
Selected Results
Schools The distribution of enrollment by type of private school
In the fall of 1999, there were 27,223 private elementary differed from the distribution of schools by the same
and secondary schools in the United States, a total not dimension. More students were enrolled in Catholic schools

1 2
While the tables and figures use the term “private,”“private order” is used in the text Public school enrollment source is the Common Core of Data (CCD) "State Nonfiscal
to avoid confusion with the general use of the term “private.” Survey of Public Elementary/Secondary Education: School Year 1999–2000."

E D U C AT I O N S TAT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , FA L L 2 0 0 1 37
Elementary and Secondary Education

Table 1.—Number and percentage distribution of private schools, students, and full-time-equivalent (FTE) teachers by NCES
typology and selected characteristics: United States, 1999–2000

Selected Schools Students FTE teachers

characteristics Number Percent Number Percent Number Percent

Total 27,223 100.0 5,162,684 100.0 395,317 100.0

NCES typology
Catholic 8,102 29.8 2,511,040 48.6 149,600 37.8
Parochial 4,607 16.9 1,307,461 25.3 72,497 18.3
Diocesan 2,598 9.5 835,327 16.2 49,415 12.5
Private 897 3.3 368,252 7.1 27,689 7.0

Other religious 13,232 48.6 1,843,580 35.7 152,915 38.7

Conservative Christian 4,989 18.3 773,237 15.0 60,481 15.3
Affiliated 3,531 13.0 553,530 10.7 47,433 12.0
Unaffiliated 4,712 17.3 516,813 10.0 45,001 11.4

Nonsectarian 5,889 21.6 808,063 15.7 92,801 23.5

Regular 2,494 9.2 546,649 10.6 58,279 14.7
Special emphasis 2,131 7.8 175,140 3.4 19,981 5.1
Special education 1,264 4.6 86,274 1.7 14,542 3.7

School level
Elementary 16,530 60.7 2,831,372 54.8 187,833 47.5
Secondary 2,538 9.3 806,639 15.6 62,737 15.9
Combined 8,155 30.0 1,524,673 29.5 144,746 36.6

Program emphasis
Regular elementary/secondary 22,263 81.8 4,751,634 92.0 346,300 87.6
Montessori 1,190 4.4 77,264 1.5 8,462 2.1
Special program emphasis 606 2.2 111,219 2.2 10,949 2.8
Special education 1,409 5.2 95,261 1.9 15,978 4.0
Vocational/technical — — — — — —
Alternative 1,617 5.9 120,233 2.3 13,000 3.3
Early childhood 133 0.5 5,534 0.1 532 0.1

Size
Less than 50 7,565 27.8 196,309 3.8 26,329 6.7
50–149 7,738 28.4 716,129 13.9 71,676 18.1
150–299 6,571 24.1 1,424,018 27.6 102,457 25.9
300–499 3,219 11.8 1,228,631 23.8 84,086 21.3
500–749 1,352 5.0 805,490 15.6 54,078 13.7
750 or more 778 2.9 792,106 15.3 56,691 14.3

Region
Northeast 6,452 23.7 1,294,847 25.1 103,805 26.3
Midwest 6,991 25.7 1,345,446 26.1 91,444 23.1
South 8,240 30.3 1,575,784 30.5 131,192 33.2
West 5,540 20.4 946,608 18.3 68,876 17.4

Community type
Central city 10,825 39.8 2,540,516 49.2 189,984 48.1
Urban fringe/large town 10,359 38.1 2,051,094 39.7 155,436 39.3
Rural/small town 6,040 22.2 571,074 11.1 49,897 12.6

—Too few sample cases for a reliable estimate.

NOTE: Detail may not add to totals because of rounding or missing values in cells with too few sample cases.
SOURCE: U.S. Department of Education, National Center for Education Statistics, Private School Survey (PSS), 1999–2000.

38 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
 Private School Universe Survey: 1999–2000

Figure 1.—Percentage distribution of private schools, by NCES typology

Unaffiliated Parochial
(17.3%) (16.9%) Catholic
(29.8%)
Other religious
(48.6%)

Diocesan
(9.5%)
Affiliated
(13.0%)
Private
(3.3%)

Regular
(9.2%)
Conservative
Christian
Special
(18.3%) Special Nonsectarian
emphasis
education (7.8%) (21.6%)
(4.6%)

NOTE: Detail may not add to 100.0 percent because of rounding.

SOURCE: U.S. Department of Education, National Center for Education Statistics, Private School Survey (PSS),
1999–2000.

than in other religious schools, 49 and 36 percent of total the West (18 percent). Approximately 55 percent of private
private enrollment, respectively (table 1 and figure 2). school students were enrolled in elementary schools,
Enrollment in nonsectarian schools, representing 16 per- 16 percent were enrolled in secondary schools, and
cent of all private students, was less than that of Catholic or 30 percent were enrolled in combined schools (table 1).
other religious schools. That Catholic schools represent Ninety-two percent of private school students were enrolled
approximately one-third of all private schools while in schools with a regular elementary/secondary program
containing almost half of private school students is an emphasis, while fewer than 5 percent of private school
indication that the distribution of schools by size is not students were enrolled in schools featuring any one of the
the same for the three types of schools. For example, the other categories of program emphasis.
percentage of schools that are small (fewer than 50 stu-
dents) is much greater for other religious (38 percent) and Approximately three-quarters (77 percent) of private school
nonsectarian (40 percent) schools than for Catholic schools students were White, non-Hispanic; while 9 percent were
(2 percent). Among Catholic schools, more students were Black, non-Hispanic; 8 percent were Hispanic; 4 percent
enrolled in parochial schools, followed by diocesan schools were American Indian/Alaska Native; and 5 percent were
and then private order schools. Among the three categories Asian/Pacific Islander.3 Almost half (49 percent) of all
of other religious schools, enrollment was greater in private school students attended schools that were located
conservative Christian schools than in affiliated or unaffili- in urban areas (central city), and approximately 40 percent
ated schools. Of the nonsectarian schools, regular schools attended schools that were located in an urban fringe or a
had more students, followed by special emphasis schools large town, while only 11 percent attended rural schools
and then special education schools. (table 1).

3
The region with the most private school students was the For comparisons of the racial/ethnic composition of private school enrollment with
that of public schools from the 1987–88, 1990–91, and 1993–94 Schools and Staffing
South (31 percent), while the region with the fewest was Surveys, see McLaughlin, O’Donnell, and Ries (1995) and McLaughlin (1997).

E D U C AT I O N S TAT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , FA L L 2 0 0 1 39
Elementary and Secondary Education

Figure 2.—Percentage distribution of private school students, by NCES typology

Unaffiliated
(10.0%)
Parochial
(25.3%)
Affiliated
(10.7%)
Other religious
(35.7%)
Catholic
(48.6%)
Conservative
Christian
(15.0%)

Diocesan
Special (16.2%)
education
(1.7%) Special
emphasis
Regular Private
(3.4%)
(10.6%) (7.1%)

Nonsectarian
(15.7%)

NOTE: Detail may not add to 100.0 percent because of rounding.

SOURCE: U.S. Department of Education, National Center for Education Statistics, Private School Survey (PSS),
1999–2000.

Teachers special emphasis schools and then special education

The nation’s private school students were taught by 395,317 schools.
full-time-equivalent (FTE) teachers (table 1) in the fall of
1999, representing an increase over the 376,544 FTE The region with the most private school FTE teachers was
teachers employed in private schools in the fall of 1997 the South (33 percent), while the region with the fewest
(Broughman and Colaciello 1999). The distribution of FTE was the West (17 percent). Nearly one-half of FTE teachers
teachers by type of private school differed from those of (48 percent) were teaching in elementary schools, approxi-
schools and enrollment. Catholic schools (38 percent) and mately one-third (37 percent) in combined schools, and
other religious schools (39 percent) employed approxi- about 16 percent in secondary schools. Almost 88 percent
mately the same number of FTE teachers, while both of private school FTE teachers were teaching in schools
employed more than nonsectarian schools (24 percent) with a regular elementary/secondary program emphasis. As
(table 1 and figure 3). Among Catholic schools, more FTE in the case of students, fewer than 5 percent of private
teachers were teaching in parochial schools, followed by school FTE teachers were teaching in schools featuring any
diocesan schools and then private order schools. Among the one of the other categories of program emphasis.
three categories of other religious schools, conservative
Christian schools employed more teachers than affiliated or Kindergarten-terminal schools
unaffiliated schools. Of the nonsectarian schools, more FTE Since 1995, schools for which kindergarten was the highest
teachers were employed by regular schools, followed by grade have been included in the PSS. In the fall of 1999,
there were 5,772 of these schools enrolling 91,802 students

40 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
 Private School Universe Survey: 1999–2000

Figure 3.—Percentage distribution of private school full-time-equivalent (FTE) teachers, by NCES

typology

Unaffiliated
(11.4%) Parochial
(18.3%)

Affiliated Catholic
(12.0%) (37.8%)
Other religious
(38.7%)

Diocesan
(12.5%)

Conservative
Christian
(15.3%) Private
(7.0%)
Special
education
(3.7%) Special Regular
emphasis (14.7%)
(5.1%)

Nonsectarian
(23.5%)

NOTE: Detail may not add to 100.0 percent because of rounding.

SOURCE: U.S. Department of Education, National Center for Education Statistics, Private School Survey (PSS),
1999–2000.

and employing 13,081 FTE teachers nationwide. When McLaughlin, D.H. (1997). Private Schools in the United States: A
the k-terminal schools are added to the traditional PSS Statistical Profile: 1993–94 (NCES 97–459). U.S. Department of
Education, National Center for Education Statistics. Washing-
schools, the total number of schools becomes 32,995, with
ton, DC: U.S. Government Printing Office.
5,254,485 students and 408,397 FTE teachers. Almost
McLaughlin, D.H., O’Donnell, C., and Ries, L. (1995). Private
70 percent of the k-terminal schools were nonsectarian
Schools in the United States: A Statistical Profile: 1990–91
(68 percent), 30 percent were other religious, and 2 percent (NCES 95–330). U.S. Department of Education, National Center
were Catholic. for Education Statistics. Washington, DC: U.S. Government
Printing Office.
By definition, all of these schools were classified as elemen-
tary, and most of them (97 percent) enrolled fewer than 50
students. Seventy-nine percent of these schools emphasized Data source: The NCES Private School Survey (PSS), 1999–2000.
an early childhood program, 19 percent emphasized a For technical information, see the complete report:
Montessori program, and fewer than 5 percent each empha- Broughman, S.P., and Colaciello, L.A. (2001). Private School Universe
sized any one of the other program emphases. Survey: 1999–2000 (NCES 2001–330).
Author affiliations: S.P. Broughman, NCES; L.A. Colaciello, U.S.
Bureau of the Census.
References
For questions about content, contact Stephen P. Broughman
Broughman, S.P., and Colaciello, L.A. (1999). Private School (stephen.broughman@ed.gov).
Universe Survey: 1997–98 (NCES 1999–319). U.S. Department
To obtain the complete report (NCES 2001–330), call the toll-
of Education, National Center for Education Statistics. Washing- free ED Pubs number (877–433–7827) or visit the NCES Web Site
ton, DC: U.S. Government Printing Office. (http://nces.ed.gov).

E D U C AT I O N S TAT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , FA L L 2 0 0 1 41
Elementary and Secondary Education

Schools and Districts

Overview of Public Elementary and Secondary Schools and Districts: School
Year 1999–2000
——————————————————————————————————
Lee M. Hoffman
This article was originally published as a Statistical Analysis Report. The universe data are primarily from the following two components of the NCES
Common Core of Data (CCD): “Public Elementary/Secondary School Universe Survey” and “Local Education Agency Universe Survey.” Technical notes,
definitions, and supplemental tables from the original report have been omitted.

This report summarizes information about public elemen- table A, as well as smaller proportions of other types of
tary and secondary schools and local education agencies in schools, do not appear in table 1 because no students were
the United States during the 1999–2000 school year. The reported in membership for these schools.
information is provided by state education agencies through
the Common Core of Data (CCD) survey system. In the 1998–99 school year, the CCD began reporting
schools operated by the U.S. Department of the Interior’s
Types of Public Schools and Agencies Bureau of Indian Affairs (BIA) and the domestic Depart-
States reported over 92,000 public elementary/secondary ment of Defense Dependents Schools as separate entries,
schools in the 1999–2000 school year.1 This was an increase and they are not included in the U.S. totals shown in the
of almost 7 percent over the more than 86,000 schools tables in this report. Some, but not all, of these BIA and
reported 5 years earlier, in the fall of 1994.2 Most of these Department of Defense schools previously were included in
were regular schools, those that offer a comprehensive the states within whose boundaries they were located.
curriculum and may provide other programs and services as
well. A smaller number of schools focused primarily on Most local education agencies are those that are typically
special education, vocational/technical education, or thought of as “school districts.” Operated by a local school
alternative programs. Students in these specialized schools board, they provide instructional services for students and
were often enrolled in a regular school as well, and reported comprised almost 89 percent of local agencies in 1999–2000
only with the membership of that regular school (table A). (table 2). A smaller proportion, about 8 percent, were
supervisory unions or regional education service agencies
Among the schools that reported students in membership, whose major responsibility is to offer administrative, special
almost 94 percent were regular schools (table 1). The program, testing, or other services to school districts.
second largest category with student membership was that Finally, around 4 percent of the reported agencies were
of alternative education schools (4 percent), followed by operated directly by a state or federal government or were
special education schools (about 2 percent). Note that other than any of the preceding categories. The number of
roughly two-thirds of the vocational schools identified in regular school districts increased by 1 percent from the
14,772 reported in 1994 to a total of 14,928 in 1999–2000.

1
Although the outlying areas, Bureau of Indian Affairs, and Department of Defense The governance of charter schools varies from state to state.
Dependents Schools (DoDDS) are included in the tables, national totals are limited to
the 50 states and the District of Columbia. In some cases, they are not considered under the adminis-
2
All comparisons with 1994 are based on table 90 in the Digest of Education Statistics:
tration of the regular public school district within whose
1999 (Snyder and Hoffman 2000). boundaries they operate and are reported on the CCD with

Table A.—Public elementary and secondary schools in the United States: 1999–2000
Total Regular Special Vocational Alternative

Total schools in United States 92,012 84,902 1,947 1,048 4,115

Reporting students 89,599 84,073 1,596 342 3,588
Not reporting students 2,413 829 351 706 527

NOTE: Totals include the 50 states and the District of Columbia.

SOURCE: U.S. Department of Education, National Center for Education Statistics, Common Core of Data (CCD), “Public Elementary/Secondary School
Universe Survey,” 1999–2000.

42 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
 Overview of Public Elementary and Secondary Schools and Districts: School Year 1999–2000

Table 1.—Number of public elementary and secondary schools with membership and percentage of students in membership, by type of school and by state:
School year 1999–2000
Type of school
Number of Regular Special education Vocational education Alternative education
schools having Total Number of Percentage Number of Percentage Number of Percentage Number of Percentage
State membership students schools of students schools of students schools of students schools of students
United States 89,599 46,857,321 84,073 98.2 1,596 0.4 342 0.4 3,588 1.0
Alabama 1,367 740,732 1,321 99.7 16 0.1 2 0.0 28 0.2
Alaska 497 134,391 474 98.2 1 0.2 1 0.0 21 1.7
Arizona 1,552 852,612 1,463 97.4 10 0.0 9 0.7 70 1.8
Arkansas 1,119 451,034 1,115 99.9 0 0.0 1 0.1 3 0.0
California 8,566 6,038,589 7,414 96.6 124 0.5 0 0.0 1,028 2.9
Colorado 1,561 708,109 1,479 98.7 10 0.1 3 0.1 69 1.2
Connecticut 1,073 553,993 988 96.5 24 0.6 17 1.9 44 1.0
Delaware 185 112,836 160 92.8 16 1.4 5 4.7 4 1.0
District of Columbia 189 77,194 173 95.4 10 3.4 0 0.0 6 1.2
Florida 3,131 2,381,396 2,850 98.5 128 0.7 29 0.1 124 0.7
Georgia 1,887 1,422,762 1,863 99.7 1 0.0 0 0.0 23 0.2
Hawaii 255 185,860 251 99.9 3 0.0 0 0.0 1 0.1
Idaho 658 245,331 588 98.2 12 0.1 0 0.0 58 1.7
Illinois 4,290 2,027,600 3,909 97.8 259 1.3 0 0.0 122 0.9
Indiana 1,874 988,702 1,819 99.6 8 0.1 1 0.0 46 0.4
Iowa 1,531 497,301 1,483 98.8 10 0.2 0 0.0 38 1.0
Kansas 1,440 472,188 1,435 99.9 5 0.1 0 0.0 0 0.0
Kentucky 1,364 648,180 1,298 99.5 9 0.1 1 0.0 56 0.4
Louisiana 1,513 756,579 1,390 98.1 29 0.2 5 0.1 89 1.5
Maine 691 209,253 688 100.0 3 0.0 0 0.0 0 0.0
Maryland 1,337 846,582 1,234 97.4 51 0.9 12 1.1 40 0.6
Massachusetts 1,898 971,425 1,821 96.1 1 0.0 42 3.3 34 0.5
Michigan 3,606 1,725,617 3,538 99.8 15 0.0 3 0.0 50 0.2
Minnesota 2,072 854,034 1,692 97.4 166 1.1 0 0.0 214 1.5
Mississippi 875 500,716 875 100.0 0 0.0 0 0.0 0 0.0
Missouri 2,258 914,110 2,136 98.8 56 0.7 6 0.2 60 0.2
Montana 882 157,556 875 99.9 2 0.0 0 0.0 5 0.1
Nebraska 1,312 288,261 1,255 99.4 57 0.6 0 0.0 0 0.0
Nevada 484 325,610 442 98.4 14 0.3 1 0.5 27 0.8
New Hampshire 521 206,783 521 100.0 0 0.0 0 0.0 0 0.0
New Jersey 2,383 1,289,256 2,250 97.9 85 0.7 48 1.4 0 0.0
New Mexico 755 324,495 701 97.8 17 0.7 0 0.0 37 1.5
New York 4,273 2,887,776 4,137 97.7 28 0.1 25 1.2 83 1.1
North Carolina 2,148 1,275,925 2,065 99.2 25 0.4 3 0.0 55 0.4
North Dakota 550 112,751 550 100.0 0 0.0 0 0.0 0 0.0
Ohio 3,798 1,836,554 3,673 96.7 25 0.1 72 3.0 28 0.2
Oklahoma 1,809 627,032 1,797 99.7 12 0.3 0 0.0 0 0.0
Oregon 1,277 545,033 1,174 97.9 16 0.3 0 0.0 87 1.8
Pennsylvania 3,164 1,816,716 3,125 98.4 12 1.0 14 0.6 13 0.1
Rhode Island 318 156,454 304 98.2 4 0.4 4 0.7 6 0.7
South Carolina 1,043 666,780 1,038 99.9 5 0.1 0 0.0 0 0.0
South Dakota 759 131,037 735 98.9 3 0.1 0 0.0 21 1.0
Tennessee 1,554 916,202 1,518 99.5 18 0.2 6 0.3 12 0.1
Texas 7,395 3,991,783 6,660 98.8 140 0.1 21 0.1 574 1.0
Utah 788 480,255 710 97.9 21 0.5 0 0.0 57 1.6
Vermont 359 104,559 320 98.5 38 1.4 0 0.0 1 0.0
Virginia 1,816 1,133,994 1,763 99.4 18 0.1 0 0.0 35 0.4
Washington 2,111 1,003,714 1,817 96.8 69 0.2 7 0.1 218 2.9
West Virginia 808 291,811 778 99.4 7 0.2 3 0.0 20 0.4
Wisconsin 2,118 877,753 2,041 99.3 9 0.1 1 0.0 67 0.7
Wyoming 385 92,105 367 98.7 4 0.2 0 0.0 14 1.0
Outlying areas, DOD Dependents Schools, and Bureau of Indian Affairs
DoDDS: DOD Overseas 153 73,504 153 100.0 0 0.0 0 0.0 0 0.0
DDESS: DOD Domestic 71 34,081 71 100.0 0 0.0 0 0.0 0 0.0
Bureau of Indian Affairs 188 49,076 188 100.0 0 0.0 0 0.0 0 0.0
American Samoa 31 15,477 29 97.9 1 0.3 1 1.9 0 0.0
Guam 38 32,951 38 100.0 0 0.0 0 0.0 0 0.0
Northern Marianas 25 9,732 25 100.0 0 0.0 0 0.0 0 0.0
Puerto Rico 1,523 613,019 1,465 96.2 28 1.7 12 0.9 18 1.2
Virgin Islands 35 20,866 33 99.4 0 0.0 0 0.0 2 0.6

NOTE: Table excludes 2,427 schools (14 of these in outlying areas) for which no students were reported in membership. U.S. totals include the 50 states and the District of Columbia.
Although type of school is a mutually exclusive category, many regular schools include special, vocational, or alternative education programs. Percentages are rounded to the
nearest tenth and may not add to 100. Percentages of less than 0.05 are rounded to 0.0. Total student membership is reported from the “State Nonfiscal Survey of Public Elementary/
Secondary Education.”
SOURCE: U.S. Department of Education, National Center for Education Statistics, Common Core of Data (CCD), “Public Elementary/Secondary School Universe Survey,” 1999–2000, and
“State Nonfiscal Survey of Public Elementary/Secondary Education,” 1999–2000.

E D U C AT I O N S TAT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , FA L L 2 0 0 1 43
Elementary and Secondary Education

Table 2. —Number and percentage of public elementary and secondary education agencies, by type of agency and by state: School year 1999–2000
Regional education
service agencies and
Regular supervisory union State-operated Federally operated
Total school districts administrative centers agencies and other agencies*
State agencies Number Percent Number Percent Number Percent Number Percent
United States 16,793 14,928 88.9 1,273 7.6 137 0.8 455 2.7
Alabama 131 128 97.7 0 0.0 3 2.3 0 0.0
Alaska 55 53 96.4 0 0.0 2 3.6 0 0.0
Arizona 422 413 97.9 6 1.4 2 0.5 1 0.2
Arkansas 328 310 94.5 15 4.6 3 0.9 0 0.0
California 1,057 987 93.4 58 5.5 12 1.1 0 0.0
Colorado 198 176 88.9 22 11.1 0 0.0 0 0.0
Connecticut 195 166 85.1 6 3.1 4 2.1 19 9.7
Delaware 28 19 67.9 0 0.0 3 10.7 6 21.4
District of Columbia 28 1 3.6 0 0.0 0 0.0 27 96.4
Florida 73 67 91.8 0 0.0 1 1.4 5 6.8
Georgia 180 180 100.0 0 0.0 0 0.0 0 0.0
Hawaii 1 1 100.0 0 0.0 0 0.0 0 0.0
Idaho 114 113 99.1 0 0.0 1 0.9 0 0.0
Illinois 1,055 896 84.9 154 14.6 5 0.5 0 0.0
Indiana 328 295 89.9 29 8.8 3 0.9 1 0.3
Iowa 405 375 92.6 15 3.7 15 3.7 0 0.0
Kansas 304 304 100.0 0 0.0 0 0.0 0 0.0
Kentucky 176 176 100.0 0 0.0 0 0.0 0 0.0
Louisiana 82 75 91.5 0 0.0 7 8.5 0 0.0
Maine 326 283 86.8 39 12.0 4 1.2 0 0.0
Maryland 24 24 100.0 0 0.0 0 0.0 0 0.0
Massachusetts 477 351 73.6 85 17.8 1 0.2 40 8.4
Michigan 799 737 92.2 57 7.1 4 0.5 1 0.1
Minnesota 468 407 87.0 57 12.2 4 0.9 0 0.0
Mississippi 162 152 93.8 0 0.0 10 6.2 0 0.0
Missouri 531 525 98.9 0 0.0 2 0.4 4 0.8
Montana 534 455 85.2 77 14.4 2 0.4 0 0.0
Nebraska 709 593 83.6 111 15.7 5 0.7 0 0.0
Nevada 18 17 94.4 0 0.0 1 5.6 0 0.0
New Hampshire 257 179 69.6 78 30.4 0 0.0 0 0.0
New Jersey 662 604 91.2 12 1.8 0 0.0 46 6.9
New Mexico 89 89 100.0 0 0.0 0 0.0 0 0.0
New York 745 707 94.9 38 5.1 0 0.0 0 0.0
North Carolina 201 120 59.7 0 0.0 2 1.0 79 39.3
North Dakota 272 231 84.9 38 14.0 3 1.1 0 0.0
Ohio 806 708 87.8 73 9.1 3 0.4 22 2.7
Oklahoma 556 544 97.8 0 0.0 0 0.0 12 2.2
Oregon 221 197 89.1 21 9.5 2 0.9 1 0.5
Pennsylvania 666 501 75.2 101 15.2 15 2.3 49 7.4
Rhode Island 37 36 97.3 0 0.0 1 2.7 0 0.0
South Carolina 104 90 86.5 14 13.5 0 0.0 0 0.0
South Dakota 199 176 88.4 18 9.0 5 2.5 0 0.0
Tennessee 139 139 100.0 0 0.0 0 0.0 0 0.0
Texas 1,203 1,041 86.5 20 1.7 0 0.0 142 11.8
Utah 47 40 85.1 5 10.6 2 4.3 0 0.0
Vermont 348 287 82.5 60 17.2 1 0.3 0 0.0
Virginia 169 135 79.9 34 20.1 0 0.0 0 0.0
Washington 305 296 97.0 9 3.0 0 0.0 0 0.0
West Virginia 57 55 96.5 0 0.0 2 3.5 0 0.0
Wisconsin 445 426 95.7 16 3.6 3 0.7 0 0.0
Wyoming 57 48 84.2 5 8.8 4 7.0 0 0.0
Outlying areas, DOD Dependents Schools, and Bureau of Indian Affairs
DoDDS: DOD Overseas 11 0 0.0 0 0.0 0 0.0 11 100.0
DDESS: DOD Domestic 17 0 0.0 0 0.0 0 0.0 17 100.0
Bureau of Indian Affairs 24 0 0.0 0 0.0 0 0.0 24 100.0
American Samoa 1 1 100.0 0 0.0 0 0.0 0 0.0
Guam 1 1 100.0 0 0.0 0 0.0 0 0.0
Northern Marianas 1 1 100.0 0 0.0 0 0.0 0 0.0
Puerto Rico 1 1 100.0 0 0.0 0 0.0 0 0.0
Virgin Islands 1 1 100.0 0 0.0 0 0.0 0 0.0

*States may report charter schools under the category of other agencies. For example, the District of Columbia reports each charter school as a separate agency.
NOTE: Regular school districts include those that are components of supervisory unions. Percentages may not add to 100 because of rounding. U.S. totals include the 50 states and
the District of Columbia.
SOURCE: U.S. Department of Education, National Center for Education Statistics, Common Core of Data (CCD), “Local Education Agency Universe Survey,” 1999–2000.

44 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
 Overview of Public Elementary and Secondary Schools and Districts: School Year 1999–2000

a separate education agency for each charter school. When grade spans.) A little less than 6 percent of students were in
this occurs, the districts are reported under the category of districts going no higher than grade 8, and about 2 percent
“other education agency.” For example, although not all were in secondary districts with no grade lower than 7. Less
states designate a separate agency for each charter school, in than 1 percent of students were enrolled in districts with
the District of Columbia the establishment of 27 charter some other range of grades.
schools explains why the District is shown with 28 local
education agencies in table 2. School and School District Size
Primary schools tended to be smaller than middle and high
Student Membership schools (table 5). The average number of students in a
In the 1999–2000 school year, 89,599 public schools primary school was 446 in 1999–2000. Middle schools
provided instruction to 46.9 million students in the United served, on the average, 595 students each, while the average-
States (table 1), an increase of less than 1 percent over the sized high school had 752 students. There was considerable
46.5 million students in 1998 (Hoffman 2000, table 1). Five range in school size across the states. High schools ranged
states (California, Florida, Illinois, New York, and Texas) from an average of fewer than 300 students in Montana,
enrolled more than 2 million students in their public North Dakota, and South Dakota to 1,400 students or more
schools. At the other end of the size distribution, the in Florida and Hawaii.
District of Columbia and Wyoming reported fewer than
100,000 students. Student/teacher ratios were higher in primary schools, which
had a median number of 16.2 students for each teacher, than
Most of the 1999–2000 students, 98 percent, were reported in high schools, with a median number of 14.8 students per
enrolled in regular schools. Some 1 percent were in alterna- teacher (table 6). (The median is the point at which one-half
tive schools, and special education or vocational schools of the schools had larger student/teacher ratios and one-half
each accounted for less than 1 percent of students. Missis- had smaller. Note also that student/teacher ratio is not the
sippi, New Hampshire, and North Dakota operated only same as average class size, which includes only teachers who
regular schools. are assigned to a classroom.) The median number of primary
students for each teacher ranged from a low of fewer than 13
Instructional Level in Nebraska, South Dakota, Vermont, and Wyoming to a
Schools come in all combinations of grades. To allow high of more than 20 in Kentucky and Utah.
comparisons across states, instructional level is determined
in this report by the lowest and highest grade in a school. Twenty-five school districts enrolled 100,000 or more
Among the 89,599 schools with membership during the students, while 1,809 districts served fewer than 150
1999–2000 school year, 58 percent spanned the primary students (table 7). While few in number, the larger districts
grades, beginning with prekindergarten or kindergarten and included a considerable portion of the students in America’s
going no higher than grade 8 (table 3). Middle schools, schools. Although under 2 percent of school districts
those with grade spans ranging from as low as grade 4 to as reported 25,000 or more students, almost one-third
high as grade 9, made up almost 18 percent of schools with (32 percent) of students attended school in these districts.
students. High schools (low grade of 7 or higher, high grade At the other end of the size range, more than one-third of
of 12) were another 19 percent of schools. Some 5 percent school districts had fewer than 600 students, but these
of schools had a grade configuration that did not fit into districts accounted for only 3 percent of public school
any of these three categories. enrollment.

A total of 14,571 regular school districts reported students Other School Characteristics
in membership for 1999–2000 (table 4). As with schools, The majority of schools, 57 percent, were in large or
grade span categories were assigned by the lowest and midsize cities or their accompanying urban fringe areas
highest grades offered. Approximately 73 percent of school (table 8). These schools accounted for more than two-thirds
districts included the range of grades from prekindergarten (69 percent) of all public school students. About one of
or kindergarten to 9 or higher, and they accounted for every six students was in a large city school in 1999–2000; a
92 percent of all public school students. (In fact, only in smaller proportion, about 1 in 10, attended a rural school
Arizona, Illinois, Montana, and Vermont did as many as that was not within the fringes of an urban area.
one-third of the students attend school districts with other

E D U C AT I O N S TAT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , FA L L 2 0 0 1 45
Elementary and Secondary Education

Table 3.—Percentage of public elementary and secondary schools and percentage of students in membership, by instructional level and by state: School year
1999–2000

Percentage by instructional level

Number of
schools having Primary Middle High Other
State membership Schools Students Schools Students Schools Students Schools Students
United States 89,599 58.0 49.6 17.5 19.9 19.2 27.7 5.3 2.8
Alabama 1,367 51.2 44.7 17.2 17.7 19.7 25.1 11.9 12.4
Alaska 497 36.6 43.9 6.8 12.8 14.7 25.0 41.9 18.3
Arizona 1,552 59.1 55.2 14.8 16.9 18.3 25.4 7.8 2.5
Arkansas 1,119 51.3 46.4 16.8 20.0 29.2 29.3 2.7 4.4
California 8,566 62.1 51.9 14.8 18.4 18.9 27.2 4.1 2.5
Colorado 1,561 58.3 49.6 17.9 20.6 19.7 27.3 4.1 2.4
Connecticut 1,073 61.6 50.8 17.6 20.9 16.6 27.1 4.2 1.2
Delaware 185 49.7 42.3 23.2 27.6 17.3 28.2 9.7 1.9
District of Columbia 189 63.0 63.5 12.2 13.1 16.4 18.9 8.5 4.5
Florida 3,131 53.7 49.0 15.7 21.3 13.4 25.6 17.2 4.1
Georgia 1,887 62.0 50.8 18.9 20.8 15.6 25.0 3.5 3.4
Hawaii 255 68.6 54.3 12.9 14.8 14.1 28.4 4.3 2.5
Idaho 658 51.8 47.5 17.0 21.6 25.8 28.5 5.3 2.5
Illinois 4,290 61.5 55.7 17.0 15.7 17.8 26.9 3.7 1.7
Indiana 1,874 61.5 49.7 17.4 18.9 18.8 30.2 2.2 1.2
Iowa 1,531 53.7 45.9 19.5 19.8 24.2 32.1 2.6 2.2
Kansas 1,440 57.3 48.9 17.4 19.7 24.9 31.0 0.5 0.4
Kentucky 1,364 57.5 49.5 16.9 20.1 21.7 29.6 4.0 0.8
Louisiana 1,513 52.9 48.0 19.4 20.0 16.6 25.7 11.1 6.3
Maine 691 63.7 47.1 18.1 22.3 15.6 29.0 2.6 1.7
Maryland 1,337 64.5 50.3 17.8 21.1 15.0 27.4 2.7 1.1
Massachusetts 1,898 64.5 50.2 17.1 20.3 16.0 27.2 2.4 2.3
Michigan 3,606 58.8 48.0 17.7 21.0 19.6 28.5 3.9 2.5
Minnesota 2,072 49.6 46.2 14.6 19.5 28.6 32.5 7.3 1.8
Mississippi 875 49.8 45.0 19.7 19.4 20.8 25.4 9.7 10.1
Missouri 2,258 54.7 48.6 16.4 19.6 22.2 29.5 6.7 2.3
Montana 882 53.1 47.4 26.8 20.2 20.0 31.3 0.2 1.2
Nebraska 1,312 67.8 50.5 8.1 14.8 23.1 34.2 1.1 0.6
Nevada 484 62.2 51.9 14.9 21.2 19.2 24.0 3.7 2.9
New Hampshire 521 67.0 47.4 18.0 24.0 15.0 28.5 0.0 0.0
New Jersey 2,383 62.7 52.0 18.0 19.5 13.2 25.7 6.2 2.9
New Mexico 755 57.7 47.6 20.8 22.5 18.9 28.1 2.5 1.9
New York 4,273 57.9 49.4 17.0 19.1 18.1 27.0 6.9 4.5
North Carolina 2,148 59.4 50.3 20.4 22.1 15.7 26.0 4.5 1.6
North Dakota 550 58.4 49.1 6.5 12.8 34.4 36.0 0.7 2.1
Ohio 3,798 58.2 46.5 19.8 20.6 20.2 31.9 1.8 1.0
Oklahoma 1,809 54.3 51.7 19.0 20.3 25.4 25.8 1.2 2.1
Oregon 1,277 59.2 47.3 17.1 20.8 18.8 29.7 4.9 2.1
Pennsylvania 3,164 61.1 46.7 17.8 20.7 18.9 30.0 2.1 2.6
Rhode Island 318 67.0 49.4 17.3 22.5 14.2 27.9 1.6 0.2
South Carolina 1,043 56.8 48.5 23.8 23.4 18.0 27.2 1.4 1.0
South Dakota 759 50.9 47.2 23.5 21.2 23.6 31.1 2.1 0.5
Tennessee 1,554 60.2 51.7 16.9 17.6 18.5 27.8 4.4 2.9
Texas 7,395 50.3 48.1 20.6 22.8 19.4 25.7 9.7 3.4
Utah 788 59.3 51.0 16.4 21.0 19.4 25.6 4.9 2.4
Vermont 359 74.1 52.4 7.2 9.5 13.6 31.6 5.0 6.5
Virginia 1,816 62.8 48.8 18.3 21.3 16.8 29.0 2.1 0.9
Washington 2,111 55.0 48.2 16.5 20.2 20.7 28.2 7.8 3.5
West Virginia 808 64.4 49.3 16.5 20.4 15.0 28.0 4.2 2.3
Wisconsin 2,118 57.8 47.0 17.7 19.5 21.7 31.5 2.8 2.0
Wyoming 385 58.4 46.5 19.2 22.6 18.7 28.8 3.6 2.1
Outlying areas, DOD Dependents Schools, and Bureau of Indian Affairs
DoDDS: DOD Overseas 153 56.2 58.5 12.4 12.0 24.2 22.9 7.2 6.5
DDESS: DOD Domestic 71 70.4 69.9 16.9 17.2 7.0 7.7 5.6 5.2
Bureau of Indian Affairs 188 55.9 50.4 2.1 1.8 13.3 14.6 28.7 33.2
American Samoa 31 74.2 72.0 3.2 4.9 19.4 22.8 3.2 0.3
Guam 38 71.1 50.6 18.4 22.6 10.5 26.8 0.0 0.0
Northern Marianas 25 84.0 62.6 4.0 13.0 12.0 24.4 0.0 0.0
Puerto Rico 1,523 58.4 45.5 14.8 17.1 12.0 20.5 14.8 16.9
Virgin Islands 35 65.7 53.2 20.0 17.0 11.4 28.4 2.9 1.4

NOTE: Instructional levels are primary (low grade prekindergarten to 3, high grade up to 8); middle (low grade 4 to 7, high grade 4 to 9); high (low grade 7 to 12, high grade 12 only);
other (any configuration not falling within the previous three, including ungraded schools). For states that did not provide a grade span, grade span was determined by the highest
and lowest grades in which students were reported. Table excludes 2,427 schools (14 in outlying areas) for which no students were reported in membership. U.S. totals include the
50 states and the District of Columbia. Percentages are rounded to the nearest tenth and may not add to 100.
SOURCE: U.S. Department of Education, National Center for Education Statistics, Common Core of Data (CCD), “Public Elementary/Secondary School Universe Survey,” 1999–2000.

46 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
 Overview of Public Elementary and Secondary Schools and Districts: School Year 1999–2000

Table 4.—Number of regular public school districts providing instruction and percentage of students in membership, by grade span and by state: School year
1999–2000
Grade span
PK, K, 1 to 8 or below PK, K, 1 to 9–12 7, 8, 9 to 7–12 Other
Total Number of Percentage Number of Percentage Number of Percentage Number of Percentage
State districts districts of students districts of students districts of students districts of students
United States 14,571 3,189 5.5 10,672 92.1 557 2.2 153 0.1
Alabama 128 0 0.0 128 100.0 0 0.0 0 0.0
Alaska 53 0 0.0 53 100.0 0 0.0 0 0.0
Arizona 360 175 27.1 115 62.9 51 9.5 19 0.5
Arkansas 310 0 0.0 310 100.0 0 0.0 0 0.0
California 987 558 18.9 338 72.1 88 8.8 3 0.2
Colorado 176 0 0.0 176 100.0 0 0.0 0 0.0
Connecticut 166 46 4.9 112 93.6 8 1.5 0 0.0
Delaware 19 0 0.0 15 94.3 4 5.7 0 0.0
District of Columbia 1 0 0.0 1 100.0 0 0.0 0 0.0
Florida 67 0 0.0 67 100.0 0 0.0 0 0.0
Georgia 180 7 0.2 173 99.8 0 0.0 0 0.0
Hawaii 1 0 0.0 1 100.0 0 0.0 0 0.0
Idaho 113 5 0.1 108 99.9 0 0.0 0 0.0
Illinois 896 386 25.4 410 63.8 100 10.8 0 0.0
Indiana 292 1 0.0 291 100.0 0 0.0 0 0.0
Iowa 375 0 0.0 375 100.0 0 0.0 0 0.0
Kansas 304 0 0.0 304 100.0 0 0.0 0 0.0
Kentucky 176 5 0.3 170 99.5 0 0.0 1 0.2
Louisiana 75 3 0.1 68 99.8 4 0.1 0 0.0
Maine 281 107 16.2 111 81.3 6 1.3 57 1.2
Maryland 24 0 0.0 24 100.0 0 0.0 0 0.0
Massachusetts 245 67 5.1 177 94.9 1 0.1 0 0.0
Michigan 721 138 2.0 548 97.6 21 0.2 14 0.2
Minnesota 400 35 0.7 337 98.8 17 0.2 11 0.2
Mississippi 152 1 0.1 149 99.8 2 0.2 0 0.0
Missouri 523 73 1.3 450 98.7 0 0.0 0 0.0
Montana 452 287 59.9 55 12.3 110 27.8 0 0.0
Nebraska 571 292 3.4 261 95.4 18 1.3 0 0.0
Nevada 17 1 0.0 16 100.0 0 0.0 0 0.0
New Hampshire 165 89 19.9 65 73.9 9 4.3 2 1.9
New Jersey 581 289 18.7 216 73.1 49 6.7 27 1.6
New Mexico 89 0 0.0 89 100.0 0 0.0 0 0.0
New York 704 44 1.1 641 98.2 11 0.7 8 0.1
North Carolina 120 1 0.0 118 100.0 0 0.0 1 0.0
North Dakota 229 50 2.6 171 96.3 6 0.6 2 0.6
Ohio 658 32 0.3 613 99.5 9 0.1 4 0.0
Oklahoma 544 114 3.6 429 96.4 0 0.0 1 0.0
Oregon 197 18 0.1 178 99.9 1 0.0 0 0.0
Pennsylvania 500 2 0.1 498 99.9 0 0.0 0 0.0
Rhode Island 36 4 1.5 31 97.5 0 0.0 1 1.0
South Carolina 86 0 0.0 86 100.0 0 0.0 0 0.0
South Dakota 173 3 0.8 170 99.2 0 0.0 0 0.0
Tennessee 138 12 1.9 126 98.1 0 0.0 0 0.0
Texas 1,041 64 0.2 976 99.7 0 0.0 1 0.1
Utah 40 0 0.0 40 100.0 0 0.0 0 0.0
Vermont 248 182 42.6 35 32.0 31 25.4 0 0.0
Virginia 132 0 0.0 132 100.0 0 0.0 0 0.0
Washington 296 49 1.0 246 99.0 0 0.0 1 0.0
West Virginia 55 0 0.0 55 100.0 0 0.0 0 0.0
Wisconsin 426 47 2.7 368 96.0 11 1.3 0 0.0
Wyoming 48 2 0.6 46 99.4 0 0.0 0 0.0
Outlying areas, DOD Dependents Schools, and Bureau of Indian Affairs*
DoDDS: DOD Overseas 11 0 0.0 11 100.0 0 0.0 0 0.0
DDESS: DOD Domestic 17 9 29.9 8 70.1 0 0.0 0 0.0
Bureau of Indian Affairs 24 0 0.0 24 100.0 0 0.0 0 0.0
American Samoa 1 0 0.0 1 100.0 0 0.0 0 0.0
Guam 1 0 0.0 1 100.0 0 0.0 0 0.0
Northern Marianas 1 0 0.0 1 100.0 0 0.0 0 0.0
Puerto Rico 1 0 0.0 1 100.0 0 0.0 0 0.0
Virgin Islands 1 0 0.0 1 100.0 0 0.0 0 0.0

*Table includes 28 Department of Defense and 24 Bureau of Indian Affairs school districts that are technically federally operated agencies; this is in order to report data for these
agencies in the table.
NOTE: For states that did not provide a grade span, grade span was determined by the highest and lowest grades served among all schools associated with the district. “Other”
includes all grade configurations not reported in the specified categories and includes ungraded districts. Table excludes 357 regular school districts for which no students were
reported in membership. U.S. totals include the 50 states and the District of Columbia. Percentages are rounded to the nearest tenth and may not add to 100. Percentages of less
than 0.05 are rounded to 0.0.
SOURCE: U.S. Department of Education, National Center for Education Statistics, Common Core of Data (CCD), “Public Elementary/Secondary School Universe Survey,” 1999–2000,
and “Local Education Agency Universe Survey,” 1999–2000.
E D U C AT I O N S TAT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , FA L L 2 0 0 1 47
Elementary and Secondary Education

Table 5.—Average public school size (mean number of students per school), by instructional level and by state: School year 1999–2000

Instructional level
Schools having
State membership Primary Middle High Other
United States 89,599 446 595 752 273
Alabama 1,367 467 551 682 554
Alaska 497 324 505 461 118
Arizona 1,552 515 625 766 169
Arkansas 1,119 364 480 404 657
California 8,566 581 862 999 423
Colorado 1,561 386 524 629 264
Connecticut 1,073 426 613 843 146
Delaware 185 519 724 993 122
District of Columbia 189 412 441 470 216
Florida 3,131 694 1,030 1,460 180
Georgia 1,887 618 830 1,205 735
Hawaii 255 576 836 1,468 416
Idaho 658 341 473 410 173
Illinois 4,290 428 435 713 224
Indiana 1,874 426 570 848 283
Iowa 1,531 277 330 428 263
Kansas 1,440 276 367 403 233
Kentucky 1,364 398 550 637 87
Louisiana 1,513 454 513 774 284
Maine 691 224 373 561 196
Maryland 1,337 493 752 1,161 261
Massachusetts 1,898 398 608 874 492
Michigan 3,606 378 551 714 211
Minnesota 2,072 384 552 469 100
Mississippi 875 517 566 700 595
Missouri 2,258 360 483 537 141
Montana 882 160 136 281 921
Nebraska 1,312 164 401 325 113
Nevada 484 567 957 839 465
New Hampshire 521 281 529 757 0
New Jersey 2,383 448 588 1,053 252
New Mexico 755 354 464 637 321
New York 4,273 576 758 1,005 440
North Carolina 2,148 503 644 984 214
North Dakota 550 173 401 215 586
Ohio 3,798 397 518 781 272
Oklahoma 1,809 330 371 352 605
Oregon 1,277 345 521 722 145
Pennsylvania 3,164 439 666 911 698
Rhode Island 318 363 641 969 68
South Carolina 1,043 546 628 963 459
South Dakota 759 160 156 227 43
Tennessee 1,554 497 600 868 374
Texas 7,395 516 596 715 191
Utah 788 523 780 801 291
Vermont 359 206 382 675 377
Virginia 1,816 484 728 1,076 276
Washington 2,111 416 579 647 210
West Virginia 808 277 448 676 194
Wisconsin 2,118 337 458 601 297
Wyoming 385 191 282 369 140
Outlying areas, DOD Dependents Schools, and Bureau of Indian Affairs
DoDDS: DOD Overseas 153 500 466 455 438
DDESS: DOD Domestic 71 476 489 522 447
Bureau of Indian Affairs 188 235 219 285 301
American Samoa 31 484 760 589 43
Guam 38 616 1,062 2,200 0
Northern Marianas 25 274 1,197 746 0
Puerto Rico 1,523 314 464 686 458
Virgin Islands 35 483 508 1,481 283

NOTE: Instructional levels are primary (low grade prekindergarten to 3, high grade up to 8); middle (low grade 4 to 7, high grade 4 to 9); high (low grade
7 to 12, high grade 12 only); other (any configuration not falling within the previous three, including ungraded schools). For states that did not provide
a grade span, grade span was determined by the highest and lowest grades in which students were reported. U.S. totals include the 50 states and the
District of Columbia.
SOURCE: U.S. Department of Education, National Center for Education Statistics, Common Core of Data (CCD), “Public Elementary/Secondary School
Universe Survey,” 1999–2000.

48 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
 Overview of Public Elementary and Secondary Schools and Districts: School Year 1999–2000

Table 6.—Median public school student/teacher ratio, by instructional level and by state: School year 1999–2000

Instructional level
State Primary Middle High Other
Reporting states 16.2 15.5 14.8 9.7
Alabama 14.8 17.0 15.7 15.3
Alaska 17.1 16.5 16.3 12.0
Arizona 17.8 18.2 16.9 —
Arkansas 15.2 14.4 12.2 14.8
California 19.9 22.8 21.8 17.7
Colorado 17.3 16.6 15.2 13.1
Connecticut 14.8 12.9 12.9 8.4
Delaware 16.6 16.4 15.4 4.8
District of Columbia 13.9 13.4 11.7 5.5
Florida 17.2 19.2 18.9 4.3
Georgia 15.7 15.6 16.9 15.2
Hawaii 17.5 17.0 17.5 14.1
Idaho 18.1 17.7 14.7 11.0
Illinois 17.1 15.7 14.6 7.9
Indiana 17.9 17.0 17.2 11.0
Iowa 14.7 13.5 12.9 11.7
Kansas 14.2 13.9 12.2 4.1
Kentucky 21.6 15.7 15.7 7.4
Louisiana 15.1 15.4 15.7 13.3
Maine 13.7 14.7 14.2 9.6
Maryland 17.5 15.9 17.2 6.0
Massachusetts — — — —
Michigan 18.2 17.5 18.4 13.7
Minnesota 14.8 15.7 14.9 5.0
Mississippi 17.1 16.5 16.6 15.9
Missouri 14.4 15.1 13.7 8.1
Montana 13.8 13.7 12.2 —
Nebraska 12.4 13.9 12.1 8.6
Nevada 17.5 20.8 17.6 7.0
New Hampshire 15.2 14.7 13.6 —
New Jersey 15.4 13.4 13.0 8.3
New Mexico 15.2 15.2 15.9 16.3
New York 15.6 14.6 14.5 10.7
North Carolina 15.3 14.3 14.4 6.4
North Dakota 13.0 14.6 12.7 16.0
Ohio 17.8 16.0 17.0 10.1
Oklahoma 15.6 15.2 12.6 16.9
Oregon 19.8 19.5 18.6 11.5
Pennsylvania 17.5 16.1 15.7 13.2
Rhode Island 15.4 13.2 13.7 5.3
South Carolina 14.9 15.5 15.0 12.2
South Dakota 12.6 13.1 11.7 9.0
Tennessee — — — —
Texas 15.2 14.2 12.4 9.0
Utah 20.8 19.9 19.7 12.7
Vermont 12.7 13.0 11.6 12.2
Virginia — — — —
Washington 19.4 20.1 19.9 7.0
West Virginia 14.1 14.3 15.4 8.2
Wisconsin 15.2 14.7 14.8 13.8
Wyoming 12.9 13.4 12.1 6.6
Outlying areas, DOD Dependents Schools, and Bureau of Indian Affairs
DoDDS: DOD Overseas 15.7 15.1 12.9 11.7
DDESS: DOD Domestic 15.0 13.6 12.7 12.4
Bureau of Indian Affairs — — — —
American Samoa 19.1 29.2 16.6 2.9
Guam 17.3 17.5 21.4 —
Northern Marianas 18.6 21.4 13.5 —
Puerto Rico 13.8 15.1 17.5 14.0
Virgin Islands 14.7 11.6 15.2 7.6

—Data are missing, except for Arizona, New Hampshire, Guam, and the Northern Marianas, which have no schools in the “other” category.
NOTE: Instructional levels are primary (low grade prekindergarten to 3, high grade up to 8); middle (low grade 4 to 7, high grade 4 to 9);
high (low grade 7 to 12, high grade 12 only); other (any configuration not falling within the previous three, including ungraded schools). For
states that did not provide a grade span, grade span was determined by the highest and lowest grades in which students were reported.
U.S. totals include the 50 states and the District of Columbia. If all schools were ranked by student/teacher ratio from the smallest to the
largest, half of the schools would fall below the median. For example, half of the primary schools in Alabama had a student/teacher ratio of
less than 14.8.
SOURCE: U.S. Department of Education, National Center for Education Statistics, Common Core of Data (CCD), “Public Elementary/Secondary
School Universe Survey,” 1999–2000.

E D U C AT I O N S TAT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , FA L L 2 0 0 1 49
Elementary and Secondary Education

Table 7.—Distribution of regular public school districts and students, by district membership size: School year
1999–2000

Number of Percentage Percentage

District membership size districts of districts of students
United States 14,571 100.0 100.0
100,000 or more 25 0.2 12.4
25,000–99,999 213 1.5 19.7
10,000–24,999 579 4.0 18.7
7,500–9,999 320 2.2 6.0
5,000–7,499 716 4.9 9.4
2,500–4,999 2,068 14.2 15.6
2,000–2,499 806 5.5 3.9
1,500–1,999 1,087 7.5 4.1
1,000–1,499 1,564 10.7 4.2
800–999 807 5.5 1.6
600–799 1,007 6.9 1.5
450–599 920 6.3 1.0
300–449 1,161 8.0 0.9
150–299 1,489 10.2 0.7
1–149 1,809 12.4 0.3

NOTE: Table includes the 50 states and the District of Columbia, and excludes 357 regular school districts for which no students
were reported in membership. Percentages are rounded to the nearest tenth and may not add to 100.
SOURCE: U.S. Department of Education, National Center for Education Statistics, Common Core of Data (CCD),“Local Education
Agency Universe Survey,” 1999–2000.

Table 8.—Distribution of public schools and students, by community type: School year 1999–2000

Number of Percentage Percentage

Community type schools of schools of students
United States 89,594 100.0 100.0
Large city 10,977 12.3 16.0
Midsize city 11,052 12.3 13.5
Urban fringe, large city 21,240 23.7 29.7
Urban fringe, midsize city 7,615 8.5 9.3
Large town 1,162 1.3 1.2
Small town 10,371 11.6 9.7
Rural 17,199 19.2 10.0
Rural urban fringe 9,978 11.1 10.7

NOTE: Community types classify the location of a school relative to populous areas. Table includes the 50 states and the District
of Columbia, and excludes 2,413 schools in these jurisdictions for which no students were reported in membership. Table ex-
cludes 5 schools for which no locale codes could be assigned. Percentages are rounded to the nearest tenth and may not add
to 100.
SOURCE: U.S. Department of Education, National Center for Education Statistics, Common Core of Data (CCD), “Public Elemen-
tary/Secondary School Universe Survey,” 1999–2000.

Thirty-four states and the District of Columbia recognized have magnet schools, and 18 of those states with magnet
charter schools in 1999–2000. Of this group, 30 reported schools were unable to identify them. Table 9 lists the
the number of charter schools (table 9). The number ranged number of magnet schools for the 17 states that reported
from a single charter school in Delaware, Mississippi, New this information. California and Illinois reported the
Mexico, and Oregon to more than 200 in Arizona and Califor- greatest number of magnet schools, 473 and 350, respec-
nia. In the District of Columbia, charter schools accounted tively. Illinois serves about 12 percent of its students in
for more than 8 percent of public school enrollment, more magnet schools; in California, the figure is about 9 percent.
than double the proportion reported for any state. (Note
that almost 9 percent of Puerto Rico’s public school students Table 9 shows the number of Title I eligible schools by state,
were reported to be enrolled in charter schools.) and the number of these schools that have schoolwide Title I
programs. Seven states did not identify which of their
States were asked to identify magnet schools. Fifteen states schools were eligible for Title I services. Of those that could
and the District of Columbia reported that they did not provide this information, Colorado, the District of Columbia,

50 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
 Overview of Public Elementary and Secondary Schools and Districts: School Year 1999–2000

Table 9.—Number of Title I, magnet, and charter schools and percentage of students served, by state: School year 1999–2000

Percentage of Number of Percentage of Percentage of Percentage of

Number of all students Title I all students Number of all students Number of all students
Title I eligible in these schoolwide in these magnet in these charter in these
State schools1 schools schools schools schools2 schools schools2 schools
United States — — — — — — — —
Alabama 810 54.1 570 35.7 40 2.8 0 0.0
Alaska 279 33.2 80 11.2 — — 18 1.7
Arizona — — — — 0 0.0 245 3.7
Arkansas 771 61.4 373 26.4 15 2.5 0 0.0
California 4,072 49.8 — — 473 9.2 238 1.8
Colorado 1,240 75.0 344 18.9 — — 69 2.5
Connecticut 410 35.9 91 9.1 13 0.9 16 0.4
Delaware 20 9.5 19 9.4 2 0.8 1 0.1
District of Columbia 146 81.1 120 72.9 0 0.0 27 8.3
Florida 1,127 33.5 1,020 30.4 — — 113 0.7
Georgia 930 41.6 529 22.9 63 3.5 18 0.8
Hawaii 148 51.4 124 41.9 0 0.0 2 0.4
Idaho 481 66.0 84 10.1 0 0.0 8 0.4
Illinois — — — — 350 12.2 17 0.3
Indiana 1,041 47.9 147 6.3 — — 0 0.0
Iowa 747 40.3 115 7.2 0 0.0 0 0.0
Kansas 647 34.8 — — 0 0.0 0 0.0
Kentucky 845 55.8 656 41.6 0 0.0 0 0.0
Louisiana 836 49.3 697 41.1 66 5.6 16 0.3
Maine — — — — — — — —
Maryland 418 24.5 317 18.7 0 0.0 0 0.0
Massachusetts 1,819 96.8 3 0.0 8 0.5 40 1.3
Michigan — — — — — — 193 2.8
Minnesota 969 41.0 237 8.3 40 2.1 62 0.9
Mississippi 679 70.8 568 57.9 5 0.5 1 0.1
Missouri 1,180 47.1 359 13.5 95 5.2 15 0.5
Montana 661 77.7 113 12.9 0 0.0 0 0.0
Nebraska 441 34.2 98 10.2 — — 0 0.0
Nevada 105 18.8 77 14.6 9 1.4 5 0.3
New Hampshire 404 79.0 15 2.3 0 0.0 0 0.0
New Jersey — — — — — — — —
New Mexico 543 62.8 223 26.7 — — 1 0.0
New York 1,940 36.8 1,326 25.6 — — 5 0.0
North Carolina 1,004 38.1 647 23.2 153 7.6 82 1.0
North Dakota 469 74.5 46 8.0 0 0.0 0 0.0
Ohio 2,644 63.7 1,100 26.4 0 0.0 48 0.5
Oklahoma 1,126 56.4 666 31.3 — — — —
Oregon 594 40.0 200 13.6 — — 1 0.0
Pennsylvania 1,813 50.2 470 15.3 — — 47 0.6
Rhode Island 178 50.3 63 18.0 6 2.4 2 0.3
South Carolina 509 40.2 437 33.2 — — 7 0.0
South Dakota 374 46.4 81 9.3 0 0.0 0 0.0
Tennessee — — — — 7 0.5 0 0.0
Texas 4,348 56.9 3,653 49.0 — — 176 0.6
Utah 229 21.1 112 10.0 — — 6 0.1
Vermont 209 57.8 55 15.4 0 0.0 0 0.0
Virginia 738 29.5 221 9.1 27 1.4 — —
Washington — — — — 0 0.0 0 0.0
West Virginia 422 40.8 321 29.2 0 0.0 0 0.0
Wisconsin 1,076 46.2 237 12.8 — — 45 0.4
Wyoming 154 36.1 44 10.8 — — — —
Outlying areas, DOD Dependents Schools, and Bureau of Indian Affairs
DoDDS: DOD Overseas — — — — — — — —
DDESS: DOD Domestic — — — — — — — —
Bureau of Indian Affairs — — — — — — — —
American Samoa — — — — — — — —
Guam — — — — — — — —
Northern Marianas — — — — — — — —
Puerto Rico 1,214 78.9 946 63.6 214 15.8 119 8.7
Virgin Islands 36 100.0 — — — — — —

—Data are missing.

1
Number of Title I eligible schools includes those with and without schoolwide Title I programs.
2
Zero indicates that no schools of this type operate in a state.
NOTE: Percentages are based on all schools reporting in a state. Percentages of less than 0.05 are rounded to 0.0. Numbers of schools include those not reporting students in
membership.
SOURCE: U.S. Department of Education, National Center for Education Statistics, Common Core of Data (CCD), “Public Elementary/Secondary School Universe Survey,” 1999–2000.

E D U C AT I O N S TAT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , FA L L 2 0 0 1 51
Elementary and Secondary Education

Massachusetts, Montana, and New Hampshire reported that and New Mexico were eligible for this program. The largest
three-fourths or more of all their public school students numbers of students eligible for free- or reduced-price
were in Title I eligible schools. Within the states identifying meals were in California, with more than 2.8 million
schools with schoolwide Title I programs, more than one- eligible students, and Texas, with almost 1.8 million.
half of the students were enrolled in these schools in the
District of Columbia and Mississippi. Table 11 shows the distribution of minority students across
cities, urban fringe areas, and small towns or rural commu-
Student Program Participation and Selected nities in 1999–2000. In some states, the more urban
Characteristics districts were composed primarily of minority students.
Nationally, over 12 percent of public school students had a Three-fourths or more of students were minority group
special education Individual Education Program (IEP) in members in the large or mid-sized city schools of the
1999–2000 (table 10). Among those states that did not District of Columbia, Georgia, Hawaii, Maryland, New
underreport students with IEPs, the proportion ranged from Jersey, and New York. Small town and rural schools tended
over 10 percent in Colorado to almost 19 percent in New to have smaller proportions of minority students, but this
Mexico and Rhode Island. was not the case for all states. In the small town and rural
schools of Hawaii, Mississippi, and New Mexico, one-half
Only 35 states and the District of Columbia reported the or more of the students were minority group members.
number of students receiving services for limited English
proficiency (LEP). In California, there were 1.4 million LEP References
service recipients (almost one-fourth of all students) in Hoffman, L.M. (2000). Overview of Public Elementary Schools and
1999–2000, while Texas reported more than half a million Districts: School Year 1998–99 (NCES 2000–333 [Revised]). U.S.
Department of Education. Washington, DC: National Center for
students receiving LEP services.
Education Statistics.
Snyder, T.D., and Hoffman, C.M. (2000). Digest of Education
Thirty-two states and the District of Columbia provided
Statistics: 1999 (NCES 2000–031). U.S. Department of Educa-
information about the number of migrant students who tion, National Center for Education Statistics. Washington, DC:
received appropriate services during the 1998–99 school U.S. Government Printing Office.
year or the following summer. Because a single migrant
student may enroll in several schools during the year, this
is a duplicated count of students. Therefore, table 10 Data sources: The NCES Common Core of Data (CCD): “Public
cannot estimate the proportion of students who were Elementary/Secondary School Universe Survey,” 1999–2000; “Local
Education Agency Universe Survey,” 1999–2000; and “State Nonfiscal
migrants. The greatest number of migrant students served, Survey of Public Elementary/Secondary Education,” 1999–2000.
more than 116,000, was reported by Texas, although that For technical information, see the complete report:
state did not provide information about summer school Hoffman, L.M. (2001). Overview of Public Elementary and Secondary
programs. Florida had the second-highest enrollment of Schools and Districts: School Year 1999–2000 (NCES 2001–339).
migrant students during the regular school year, more Author affiliation: L.M. Hoffman, NCES.
than 40,000 students. For questions about content, contact Lee Hoffman
(lee.hoffman@ed.gov).
To obtain the complete report (NCES 2001–339), call the toll-free
All but four states reported the number of students eligible ED Pubs number (877–433–7827) or visit the NCES Web Site
for free- or reduced-price meals. More than one-half of all (http://nces.ed.gov).
students in the District of Columbia, Louisiana, Mississippi,

52 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
 Overview of Public Elementary and Secondary Schools and Districts: School Year 1999–2000

Table 10.—Number and percentage of public school students participating in selected programs, by state: School year 1999–2000
Number of Number of Number of Percentage of all
Number Percentage students students students eligible students eligible
Number Percentage of students of students receiving receiving for free- or for free- or
of students of students receiving receiving migrant services: migrant services: reduced-price reduced-price
State with IEPs with IEPs LEP services LEP services school year1 summer meals meals
United States 5,810,658 12.4 — — — — — —
Alabama 98,835 13.5 7,911 1.1 6,322 2,846 328,183 45.0
Alaska 17,503 13.0 — — 9,856 1,258 34,503 25.7
Arizona 89,987 10.6 125,301 14.7 — — — —
Arkansas 53,632 11.9 9,102 2.0 — 1,242 204,740 45.4
California 643,243 10.8 1,442,277 24.2 — — 2,806,614 47.1
Colorado 71,895 10.2 — — 8,216 3,611 195,928 27.7
Connecticut 74,722 13.5 20,188 3.6 3,950 2,420 127,614 23.0
Delaware 16,125 14.1 1,617 1.4 — 279 37,120 32.9
District of Columbia 9,881 12.8 8,706 11.3 607 607 41,812 54.2
Florida 354,289 14.9 167,779 7.0 40,629 9,479 1,054,044 44.3
Georgia 154,586 10.9 44,393 3.1 20,255 3,924 611,014 42.9
Hawaii 21,138 11.4 12,289 6.6 0 39 72,457 39.0
Idaho 28,880 11.8 17,721 7.2 7,120 4,481 79,197 32.3
Illinois 281,028 13.9 122,365 6.0 — — — —
Indiana 150,591 15.2 — — — — 273,608 27.7
Iowa 71,066 14.2 4,821 1.0 2,508 450 130,931 26.5
Kansas 59,490 12.7 11,898 2.5 — — 149,718 32.2
Kentucky 91,307 14.1 — — — 6,209 299,613 47.6
Louisiana 94,992 12.6 9,031 1.2 4,877 3,924 442,320 58.5
Maine 31,536 14.7 — — — — 62,565 29.9
Maryland 110,742 13.1 17,055 2.0 — 323 251,165 29.7
Massachusetts 161,207 16.6 44,828 4.6 1,427 1,427 238,636 24.6
3
Michigan 81,588 4.8 — — — — 505,856 30.5
Minnesota 107,282 12.5 35,810 4.2 1,489 1,809 219,385 25.7
Mississippi 61,935 12.4 4,949 1.0 2,372 185 316,818 63.3
Missouri 134,210 14.7 8,157 0.9 2,833 517 312,863 34.2
Montana 18,978 12.0 — — — — 48,948 30.9
Nebraska 43,472 15.1 9,144 3.2 3,551 1,093 86,031 29.8
Nevada 35,867 11.0 — — 191 — 89,525 27.5
New Hampshire 27,895 13.5 1,914 0.9 — — 32,885 15.9
3
New Jersey 82,301 6.4 — — — — 364,578 28.3
New Mexico 60,739 18.7 58,174 17.9 — 723 165,172 50.9
New York 418,672 14.5 50,063 1.7 — — 1,230,162 42.6
North Carolina 172,466 13.5 37,265 2.9 — 6,371 497,886 39.0
North Dakota 13,405 11.9 — — 381 533 32,350 28.7
Ohio 226,027 12.2 322 — — — 501,121 26.6
Oklahoma 82,999 13.2 35,647 5.7 — 741 285,467 45.5
Oregon 61,723 11.3 35,027 6.4 18,245 2,780 185,854 34.1
Pennsylvania 215,329 11.9 — — — — 521,009 28.7
Rhode Island 28,993 18.5 9,220 5.9 170 170 51,474 32.9
South Carolina 91,333 13.7 3,379 0.5 — 731 307,524 46.1
South Dakota 15,980 12.2 4,659 3.6 1,997 206 36,978 28.2
Tennessee 134,581 14.8 — — — — — —
Texas 482,427 12.1 555,334 13.9 116,011 — 1,783,820 44.7
Utah 55,389 11.6 37,275 7.8 2,146 2,943 132,117 27.6
Vermont 12,348 11.8 821 0.8 726 697 23,493 22.5
Virginia 157,024 13.9 21,787 1.9 999 304 336,627 29.7
Washington 118,117 11.8 — — — — — —
West Virginia 50,314 17.2 — — — — 145,393 49.8
Wisconsin 120,598 13.7 — — — — 219,322 25.0
Wyoming 11,991 13.0 2,267 2.5 — — 25,936 28.1
Outlying areas, DOD Dependents Schools, and Bureau of Indian Affairs
DoDDS: DOD Overseas 5,683 7.7 4,666 6.4 — — — —
DDESS: DOD Domestic 31 0.1 1,927 5.7 — — — —
Bureau of Indian Affairs — — — — — — — —
2 2
American Samoa 648 4.2 15,013 97.0 — — 15,384 99.4
Guam 2,384 7.2 13,574 41.2 — — 12,448 37.9
2 2
Northern Marianas 506 5.2 — — — — 4,464 48.6
2 2
Puerto Rico 58,797 9.6 — — 467 467 497,501 81.2
Virgin Islands 1,450 6.9 1,070 5.1 — — — —

—Data are missing.

1
Migrant students include those who were enrolled at any time during the previous (1998–99) regular school year. They are reported for each school in which they enroll; because
this is a duplicated count, the table does not show migrants as a percentage of all students. Hawaii did not have a migrant education program in 1998–99.
2
American Samoa, the Northern Marianas, and Puerto Rico reported students eligible for free meals, but not those eligible for reduced-price meals.
3
Michigan and New Jersey report an undercount of students with IEPs.
NOTE: Percentages are based on schools and agencies reporting. Percentages are rounded to the nearest tenth and may not add to 100. Percentages of less than 0.05 are rounded
to 0.0. U.S. totals include the 50 states and the District of Columbia.
SOURCE: U.S. Department of Education, National Center for Education Statistics, Common Core of Data (CCD), “Public Elementary/Secondary School Universe Survey,” 1999–2000,
and “Local Education Agency Universe Survey,” 1999–2000.

E D U C AT I O N S TAT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , FA L L 2 0 0 1 53
Elementary and Secondary Education

Table 11.—Minority student enrollment, by community type and by state: School year 1999–2000

Percentage of minority students by community type

Number of
Total minority City, large Urban fringe Small town
State students students and midsize of city or rural
United States 46,857,321 17,680,852 — — —
Alabama 740,732 283,630 68.8 28.5 30.3
Alaska 134,391 50,857 35.1 0.0 40.9
Arizona 852,612 391,297 49.5 38.0 49.1
Arkansas 451,034 125,355 44.9 14.3 22.6
California 6,038,589 3,732,515 72.7 59.2 39.6
Colorado 708,109 215,653 43.9 26.8 20.1
Connecticut 553,993 163,371 68.4 19.1 7.6
Delaware 112,836 43,322 52.9 37.8 29.4
1
District of Columbia 77,194 74,043 95.9 0.0 100.0
Florida 2,381,396 1,083,654 51.4 48.7 29.0
Georgia 1,422,762 632,535 79.5 46.6 32.4
Hawaii 185,860 147,782 81.8 79.5 78.1
Idaho 245,331 32,677 — — —
Illinois 2,027,600 797,657 74.2 28.8 7.5
Indiana 988,702 155,519 39.5 10.7 3.2
Iowa 497,301 45,647 20.7 6.6 4.0
Kansas 472,188 94,042 40.5 10.8 13.1
Kentucky 648,180 72,865 29.5 15.9 4.7
Louisiana 756,579 383,916 73.4 41.9 38.4
Maine 209,253 6,303 8.2 2.8 2.3
Maryland 846,582 386,924 76.5 46.8 20.0
Massachusetts 971,425 226,841 54.7 12.5 5.4
Michigan 1,725,617 409,313 69.0 16.3 6.5
Minnesota 854,034 137,104 51.2 10.7 7.1
Mississippi 500,716 262,556 73.7 27.3 53.2
Missouri 914,110 186,010 46.5 22.6 5.9
Montana 157,556 21,311 13.1 6.8 14.2
Nebraska 288,261 46,014 27.1 16.0 9.2
Nevada 325,610 133,538 48.1 43.3 23.4
New Hampshire 206,783 8,451 11.4 3.5 1.9
New Jersey 1,289,256 505,710 78.7 36.9 15.4
New Mexico 324,495 207,028 61.3 70.7 67.1
New York 2,887,776 1,292,003 79.7 22.0 6.3
North Carolina 1,275,925 487,864 51.7 30.2 33.1
North Dakota 112,751 11,922 8.0 6.5 12.3
Ohio 1,836,554 352,900 53.1 12.2 3.1
Oklahoma 627,032 212,186 45.7 24.2 32.9
Oregon 545,033 98,603 24.7 17.8 14.4
Pennsylvania 1,816,716 384,734 64.7 12.5 4.4
Rhode Island 156,454 38,408 51.0 11.9 4.2
South Carolina 666,780 298,904 55.2 35.1 47.5
South Dakota 131,037 17,000 15.1 7.9 12.6
Tennessee 916,202 247,681 — — —
Texas 3,991,783 2,269,814 73.4 43.9 40.7
Utah 480,255 63,185 25.9 10.6 9.6
Vermont 104,559 3,302 13.3 4.6 2.5
Virginia 1,133,994 403,810 57.8 33.3 21.8
Washington 1,003,714 246,177 34.5 24.0 18.1
West Virginia 291,811 15,227 10.8 6.4 3.8
Wisconsin 877,753 162,985 43.3 9.1 5.5
Wyoming 92,105 10,707 14.2 17.4 10.4
Outlying areas, DOD Dependents Schools, and Bureau of Indian Affairs
DoDDS: DOD Overseas 73,504 23,666 — — —
DDESS: DOD Domestic 34,081 — — — —
Bureau of Indian Affairs2 49,076 48,981 100.0 100.0 100.0
American Samoa 15,477 15,477 — — —
Guam 32,951 32,244 — — —
Northern Marianas 9,732 8,693 — — —
Puerto Rico 613,019 613,019 — — —
Virgin Islands 20,866 20,670 — — —

—Data are missing.

1
Represents one school located in a small town locale outside the District of Columbia.
2
Total students reported on State Nonfiscal Survey is greater than sum of students reported on School Universe Survey.
NOTE: Percentages are based on schools reporting. National percentages were not imputed if data were missing for one or more states. U.S. totals
include the 50 states and the District of Columbia.
SOURCE: U.S. Department of Education, National Center for Education Statistics, Common Core of Data (CCD),“Public Elementary/Secondary School
Universe Survey,” 1999–2000, and “State Nonfiscal Survey of Public Elementary/Secondary Education,” 1999–2000.

54 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
Revenues and Expenditures
Revenues and Expenditures for Public Elementary and Secondary Education:
School Year 1998–99
——————————————————————————————————
Frank Johnson
This article was originally published as a Statistics in Brief report. The universe data are primarily from the “National Public Education Financial Survey”
(NPEFS), part of the NCES Common Core of Data (CCD). Technical notes and definitions from the original report have been omitted.

Over $347 billion of revenues were raised to fund public dollar in revenue; state revenues comprised 49 cents; and
education for grades prekindergarten through 12 in school the remaining 7 cents came from federal sources.
year 1998–99. Current expenditures (those excluding
construction, equipment, and debt financing) came to $303 Among states with more than one school district, revenues
billion. Three out of every five current expenditure dollars from local sources ranged from 14.0 percent (New Mexico)
were spent on teachers, textbooks, and other instructional to 87.1 percent (New Hampshire) of total revenues
services and supplies. An average of $6,508 was spent on (table 2). Hawaii and the District of Columbia have only
each student—an increase of 5.2 percent from $6,189 in one school district each and thus are not comparable to
school year 1997–98 (in unadjusted dollars). other states. Revenues from state sources also showed a
wide distribution in their share of total revenues. The state
These and other financial data on public elementary and revenue share of total revenues was less than 30 percent in
secondary education are collected and reported each year by one state—New Hampshire (8.9 percent)—and more than
the National Center for Education Statistics (NCES), U.S. 70 percent in Vermont (74.4 percent) and New Mexico
Department of Education. The data are part of the “National (72.5 percent). Federal revenues ranged from 3.7 percent in
Public Education Financial Survey” (NPEFS), one of the New Jersey to 14.0 percent in Mississippi. Federal revenues
components of the Common Core of Data (CCD) collection made up 16.5 percent of total revenues in the District of
of surveys. Columbia.

Revenues for Public Elementary and Current Expenditures for Public Elementary
Secondary Education and Secondary Education
Over $347 billion were collected for public elementary and Current expenditures for public education in 1998–99
secondary education for school year 1998–99 in the 50 totaled about $303 billion (table 3). This represents a
states and the District of Columbia (table 1). Total revenues $17 billion (6.1 percent) increase over expenditures in the
ranged from a high of around $40 billion in California, previous school year ($285 billion in unadjusted dollars).
which serves about 1 out of every 8 students in the nation, About $187 billion in current expenditures went for
to a low of about $709 million in North Dakota, which instruction. Another $103 billion were expended for a
serves about 1 out of every 405 students in the nation. cluster of services that support instruction. Over $13 billion
Nationally, revenues increased an average of 6.6 percent were spent on noninstructional services.
over previous year’s revenues of $326 billion (in unadjusted
dollars).* When expressed in terms of the typical education dollar,
instructional expenditures accounted for 62 cents of the
By far, the greatest part of education revenues came from education dollar for current expenditures (figure 2).
nonfederal sources (state, intermediate, and local govern- Instructional expenditures include teachers’ salaries and
ments), which together provided about $323 billion, or benefits, supplies (e.g., textbooks), and purchased services.
92.9 percent of all revenues. The federal government About 34 cents of the education dollar went for support
contribution to education revenues made up the remaining services, which include operation and maintenance of
$25 billion. (These numbers do not add up to $347 billion buildings, school administration, transportation, and other
due to rounding.) The relative contributions from these student and school support activities (e.g., student counsel-
levels of government can be expressed as portions of the ing, libraries, and health services). Approximately 4 cents of
typical education dollar (figure 1). For school year 1998–99, every education dollar went to noninstructional activities,
local and intermediate sources made up 44 cents of every which include school meals and enterprise activities, such
as bookstores.
*Comparisons are based on the previous edition of this report, Revenues and
Expenditures for Public Elementary and Secondary Education: School Year 1997–98
(Johnson 2000).

E D U C AT I O N S TAT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , FA L L 2 0 0 1 55
Elementary and Secondary Education

Figure 1.—The public education dollar: Revenues by source: School year 1998–99

Total revenues: $347 billion

Local and intermediate sources

(44.2%)

Federal sources
(7.1%)

State sources
(48.7%)

SOURCE: U.S. Department of Education, National Center for Education Statistics, Common Core of Data (CCD),
“National Public Education Financial Survey," 1998–99.

Figure 2.—The public education dollar: Current expenditures by function: School year 1998–99

Current expenditures: $303 billion

Instruction (teacher
salaries, textbooks, etc.)
(61.7%)

Noninstruction (food
service, bookstore, etc.)
(4.3%)

Support services (school maintenance,

nurses, administration, library, etc.)
(34.0%)

SOURCE: U.S. Department of Education, National Center for Education Statistics, Common Core of Data (CCD),
“National Public Education Financial Survey, " 1998–99.

56 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
 Revenues and Expenditures for Public Elementary and Secondary Education: School Year 1998–99

Most states were closely clustered around the national expenditures exclude expenditures for capital outlay, which
average (61.7 percent) in terms of the share of current tend to have dramatic increases and decreases. Also, the
expenditures that were spent on instruction; all but three current expenditures commonly reported are for public
states and the District of Columbia spent more than 58.0 elementary and secondary education only. School districts
percent of their current expenditures on instruction also support community services, adult education, private
(table 4). These states were Alaska, Kansas, and New education, and other programs. These programs and the
Mexico. Two states spent more than two-thirds of their extent to which they are funded by school districts vary
current expenditures on instruction. These states were New greatly both across states and within states.
York (67.8 percent) and Maine (67.3 percent).
Total expenditures for the nation totaled nearly $356 billion
Current Expenditures per Student in the 1998–99 school year (table 6). About $303 billion
In 1998–99, the 50 states and the District of Columbia of total expenditures were current expenditures for public
spent an average of $6,508 in current expenditures for every elementary and secondary education. An additional
pupil in membership (table 5). This represents a 5.2 percent $32 billion went for facilities acquisition and construction,
increase in current expenditures per student from the $8 billion for replacement equipment, and another
previous school year ($6,189 in unadjusted dollars). Three $8 billion for interest payments on debt. The remaining
states—New Jersey ($10,145), New York ($9,344), and amount ($5 billion) was spent on other programs that are
Connecticut ($9,318)—expended more than $9,000 per not part of public elementary and secondary education.
pupil. The District of Columbia, which comprises a single
urban district, spent $9,650 per pupil. Only one state, Utah, Reference
had expenditures of less than $4,500 for each pupil in Johnson, F. (2000). Revenues and Expenditures for Public
membership ($4,210). The median per pupil expenditure Elementary and Secondary Education: School Year 1997–98
(NCES 2000–348). U.S. Department of Education. Washington,
was $6,110, indicating that one-half of all states educated DC: National Center for Education Statistics.
students at a cost of less than $6,110 per student.

Data sources: The NCES Common Core of Data (CCD): “National Public
On the average, for every student in 1998–99, about $4,013 Education Financial Survey” (NPEFS), 1997–98 and 1998–99; and “State
was spent for instructional services, $2,213 was expended Nonfiscal Survey of Public Elementary/Secondary Education,” 1998–99
for support services, and $282 was spent for noninstruc- (Revised).
For technical information, see the complete report:
tional purposes.
Johnson, F. (2001). Revenues and Expenditures for Public Elementary
and Secondary Education: School Year 1998–99 (NCES 2001–321).
Total Expenditures Author affiliation: F. Johnson, NCES.
Total expenditures include all types of expenditures by For questions about content, contact Frank Johnson
school districts and other public elementary/secondary (frank.johnson@ed.gov).
education agencies. Researchers generally use current To obtain the complete report (NCES 2001–321), visit the NCES Web
Site (http://nces.ed.gov).
expenditures instead of total expenditures, when comparing
education spending between states or across time. Current

E D U C AT I O N S TAT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , FA L L 2 0 0 1 57
Elementary and Secondary Education

Table 1.—Revenues for public elementary and secondary schools, by source and state: School year 1998–99
[In thousands of dollars]

Revenues, by source
State Total Local Intermediate State Federal
United States *$347,329,664 *$152,357,106 $1,152,510 $169,298,232 $24,521,817
Alabama 4,469,278 1,298,576 11,045 2,752,647 407,011
Alaska 1,290,358 324,621 0 787,763 177,974
Arizona 5,079,076 2,242,162 134,373 2,195,345 507,196
Arkansas 2,610,267 831,216 2,980 1,509,796 266,276
California 40,002,760 12,820,245 0 23,739,295 3,443,221
Colorado 4,714,756 2,455,786 13,963 2,002,525 242,481
Connecticut *5,607,014 *3,199,294 0 2,184,637 223,083
Delaware 959,482 271,043 0 617,023 71,416
District of Columbia 760,592 635,102 0 0 125,490
Florida 16,460,206 6,879,069 0 8,279,709 1,301,428
Georgia 10,263,338 4,529,629 0 5,044,094 689,614
Hawaii 1,328,572 31,085 0 1,166,940 130,547
Idaho 1,420,902 446,213 0 874,429 100,260
Illinois 15,338,740 9,624,278 0 4,610,903 1,103,559
Indiana 7,980,582 3,339,936 51,712 4,190,124 398,810
Iowa 3,516,165 1,536,252 8,288 1,775,997 195,628
Kansas 3,282,779 966,393 94,253 2,021,434 200,698
Kentucky 4,210,793 1,221,089 0 2,600,631 389,074
Louisiana *4,697,639 *1,789,893 0 2,366,566 541,180
Maine 1,703,252 794,721 0 781,035 127,495
Maryland 6,806,086 3,746,220 0 2,687,925 371,941
Massachusetts 8,534,080 4,517,905 0 3,593,252 422,923
Michigan 14,678,359 4,126,690 19,233 9,493,662 1,038,773
Minnesota 6,785,487 2,306,753 230,734 3,907,686 340,314
Mississippi 2,544,561 792,388 463 1,396,182 355,527
Missouri 6,265,697 3,382,250 31,825 2,446,271 405,351
Montana 1,047,338 362,625 96,073 470,763 117,877
Nebraska 2,168,308 1,198,453 14,915 804,817 150,123
Nevada 2,094,467 1,319,998 0 678,951 95,518
New Hampshire 1,441,115 1,255,189 0 127,702 58,224
New Jersey 14,192,543 7,796,638 83 5,868,487 527,334
New Mexico 2,098,648 294,395 0 1,522,000 282,253
New York 29,874,220 15,364,660 111,126 12,599,176 1,799,258
North Carolina 8,137,116 1,987,916 0 5,590,644 558,556
North Dakota 709,427 324,743 6,823 285,772 92,088
Ohio 14,399,472 7,479,793 32,198 6,057,593 829,887
Oklahoma 3,652,130 1,049,459 69,982 2,200,010 332,679
Oregon 4,047,900 1,391,092 74,785 2,297,679 284,344
Pennsylvania 15,525,301 8,641,043 13,771 5,933,154 937,333
Rhode Island 1,319,597 696,951 0 548,776 73,870
South Carolina 4,398,145 1,744,791 0 2,291,942 361,412
South Dakota 829,028 434,449 10,046 297,347 87,186
Tennessee 5,089,341 2,239,663 0 2,404,133 445,545
Texas 25,647,339 12,540,382 65,345 10,873,810 2,167,802
Utah 2,449,890 781,250 0 1,496,345 172,295
Vermont 908,146 179,360 0 675,896 52,890
Virginia *8,358,036 *5,094,040 0 2,825,340 438,656
Washington 7,212,175 2,061,181 84 4,659,490 491,420
West Virginia 2,229,692 639,239 1,414 1,398,405 190,635
Wisconsin 7,409,485 3,114,260 0 3,955,854 339,371
Wyoming 779,985 256,725 56,996 408,271 57,993

Outlying areas
American Samoa 57,667 3,642 50 11,139 42,836
Guam — — — — —
Northern Marianas 53,720 252 0 37,730 15,737
Puerto Rico 2,121,183 357 0 1,532,799 588,027
Virgin Islands 160,253 130,533 0 0 29,720

—Data not available.

*Value contains imputation for missing data. Imputed value is less than 2 percent of total revenues in any one state.
NOTE: Detail may not add to totals due to rounding. National figures do not include outlying areas.
SOURCE: U.S. Department of Education, National Center for Education Statistics, Common Core of Data (CCD), “National Public Education Financial
Survey,” 1998–99.

58 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
 Revenues and Expenditures for Public Elementary and Secondary Education: School Year 1998–99

Table 2.—Percentage distribution of revenues for public elementary and secondary schools, by source and state:
School year 1998–99

Within-state percentage distribution

State Total Local Intermediate State Federal
United States* 100.0 43.9 0.3 48.7 7.1
Alabama 100.0 29.1 0.2 61.6 9.1
Alaska 100.0 25.2 0.0 61.0 13.8
Arizona 100.0 44.1 2.6 43.2 10.0
Arkansas 100.0 31.8 0.1 57.8 10.2
California 100.0 32.0 0.0 59.3 8.6
Colorado 100.0 52.1 0.3 42.5 5.1
Connecticut* 100.0 57.1 0.0 39.0 4.0
Delaware 100.0 28.2 0.0 64.3 7.4
District of Columbia 100.0 83.5 0.0 0.0 16.5
Florida 100.0 41.8 0.0 50.3 7.9
Georgia 100.0 44.1 0.0 49.1 6.7
Hawaii 100.0 2.3 0.0 87.8 9.8
Idaho 100.0 31.4 0.0 61.5 7.1
Illinois 100.0 62.7 0.0 30.1 7.2
Indiana 100.0 41.9 0.6 52.5 5.0
Iowa 100.0 43.7 0.2 50.5 5.6
Kansas 100.0 29.4 2.9 61.6 6.1
Kentucky 100.0 29.0 0.0 61.8 9.2
Louisiana* 100.0 38.1 0.0 50.4 11.5
Maine 100.0 46.7 0.0 45.9 7.5
Maryland 100.0 55.0 0.0 39.5 5.5
Massachusetts 100.0 52.9 0.0 42.1 5.0
Michigan 100.0 28.1 0.1 64.7 7.1
Minnesota 100.0 34.0 3.4 57.6 5.0
Mississippi 100.0 31.1 0.0 54.9 14.0
Missouri 100.0 54.0 0.5 39.0 6.5
Montana 100.0 34.6 9.2 44.9 11.3
Nebraska 100.0 55.3 0.7 37.1 6.9
Nevada 100.0 63.0 0.0 32.4 4.6
New Hampshire 100.0 87.1 0.0 8.9 4.0
New Jersey 100.0 54.9 0.0 41.3 3.7
New Mexico 100.0 14.0 0.0 72.5 13.4
New York 100.0 51.4 0.4 42.2 6.0
North Carolina 100.0 24.4 0.0 68.7 6.9
North Dakota 100.0 45.8 1.0 40.3 13.0
Ohio 100.0 51.9 0.2 42.1 5.8
Oklahoma 100.0 28.7 1.9 60.2 9.1
Oregon 100.0 34.4 1.8 56.8 7.0
Pennsylvania 100.0 55.7 0.1 38.2 6.0
Rhode Island 100.0 52.8 0.0 41.6 5.6
South Carolina 100.0 39.7 0.0 52.1 8.2
South Dakota 100.0 52.4 1.2 35.9 10.5
Tennessee 100.0 44.0 0.0 47.2 8.8
Texas 100.0 48.9 0.3 42.4 8.5
Utah 100.0 31.9 0.0 61.1 7.0
Vermont 100.0 19.8 0.0 74.4 5.8
Virginia* 100.0 60.9 0.0 33.8 5.2
Washington 100.0 28.6 0.0 64.6 6.8
West Virginia 100.0 28.7 0.1 62.7 8.5
Wisconsin 100.0 42.0 0.0 53.4 4.6
Wyoming 100.0 32.9 7.3 52.3 7.4

Outlying areas
American Samoa 100.0 6.3 0.1 19.3 74.3
Guam — — — — —
Northern Marianas 100.0 0.5 0.0 70.2 29.3
Puerto Rico 100.0 0.0 0.0 72.3 27.7
Virgin Islands 100.0 81.5 0.0 0.0 18.5

—Data not available.

*Value contains imputation for missing data. Imputed value is less than 2 percent of total revenues in any one state.
NOTE: Detail may not add to totals due to rounding. National figures do not include outlying areas.
SOURCE: U.S. Department of Education, National Center for Education Statistics, Common Core of Data (CCD), “National Public
Education Financial Survey,” 1998–99.

E D U C AT I O N S TAT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , FA L L 2 0 0 1 59
Elementary and Secondary Education

Table 3.—Current expenditures for public elementary and secondary schools, by function and state: School year 1998–99
[In thousands of dollars]

Current expenditures, by function

State Total Instruction Support services Noninstruction
1 1 2 1
United States $302,873,955 $186,756,251 $102,984,961 $13,132,743
Alabama 3,880,188 2,400,979 1,210,422 268,786
2 2
Alaska 1,137,610 643,949 456,525 37,136
2 2
Arizona 3,963,428 2,380,620 1,385,289 197,519
Arkansas 2,241,244 1,436,486 649,836 154,921
California 34,379,878 20,972,433 12,034,782 1,372,663
Colorado 4,140,699 2,410,400 1,581,647 148,652
1 1
Connecticut 5,075,581 3,225,862 1,602,041 247,678
Delaware 872,786 534,174 298,325 40,287
1 1 2
District of Columbia 693,716 314,468 351,290 27,958
Florida 13,534,374 7,925,242 4,945,464 663,668
Georgia 8,537,177 5,339,447 2,726,325 471,404
Hawaii 1,143,713 718,107 350,077 75,529
Idaho 1,239,755 767,909 417,679 54,167
Illinois 13,602,965 8,240,926 4,889,281 472,759
Indiana 6,697,468 4,183,169 2,229,545 284,755
Iowa 3,110,585 1,828,806 1,048,543 233,236
Kansas 2,841,147 1,632,547 1,066,511 142,090
Kentucky 3,645,631 2,249,562 1,190,541 205,528
1 1
Louisiana 4,264,981 2,576,840 1,343,988 344,153
Maine 1,510,024 1,015,728 441,736 52,560
Maryland 6,165,934 3,791,102 2,092,027 282,805
Massachusetts 7,948,502 5,295,485 2,406,210 246,807
Michigan 12,785,480 7,432,224 4,980,257 372,999
Minnesota 5,816,329 3,647,489 1,929,556 239,284
Mississippi 2,293,188 1,384,027 746,304 162,857
Missouri 5,348,366 3,293,321 1,826,265 228,780
Montana 955,695 597,772 318,994 38,929
2 2
Nebraska 1,821,310 1,145,752 538,485 137,073
Nevada 1,738,009 1,040,680 643,224 54,105
2 2 2
New Hampshire 1,316,946 858,974 412,145 45,827
New Jersey 12,874,579 7,705,505 4,790,297 378,778
New Mexico 1,788,382 1,015,367 686,625 86,390
New York 26,885,444 18,227,087 7,930,135 728,221
North Carolina 7,097,882 4,458,396 2,220,444 419,042
North Dakota 625,428 382,138 192,382 50,908
Ohio 12,207,147 7,198,914 4,565,168 443,064
Oklahoma 3,332,697 2,000,616 1,116,721 215,360
Oregon 3,706,044 2,195,694 1,383,787 126,563
Pennsylvania 13,532,211 8,575,108 4,453,339 503,763
Rhode Island 1,283,859 851,859 396,871 35,129
South Carolina 3,759,042 2,242,844 1,292,082 224,116
South Dakota 696,785 424,211 235,257 37,317
Tennessee 4,638,924 3,032,953 1,372,663 233,308
Texas 22,430,153 13,571,196 7,671,526 1,187,430
Utah 2,025,714 1,328,807 574,738 122,168
Vermont 792,664 512,432 258,060 22,172
1 1
Virginia 7,137,421 4,340,680 2,423,409 373,333
2 2
Washington 6,098,036 3,646,974 2,154,875 296,187
West Virginia 1,986,562 1,230,925 638,769 116,868
Wisconsin 6,620,653 4,138,977 2,277,038 204,639
Wyoming 651,622 391,091 237,459 23,072

Outlying areas
American Samoa 35,092 15,014 13,238 6,840
Guam — — — —
Northern Marianas 50,450 40,667 6,456 3,327
Puerto Rico 2,024,499 1,386,049 406,614 231,835
Virgin Islands 146,474 89,700 49,828 6,946

—Data not available.

1
Value contains imputation for missing data. Imputed value is less than 2 percent of total current expenditures in any one state.
2
Value affected by redistribution of reported values for missing data items.
NOTE: Detail may not add to totals due to rounding. National figures do not include outlying areas.
SOURCE: U.S. Department of Education, National Center for Education Statistics, Common Core of Data (CCD), “National Public Education
Financial Survey,” 1998–99.

60 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
 Revenues and Expenditures for Public Elementary and Secondary Education: School Year 1998–99

Table 4.—Percentage distribution of current expenditures for public elementary and secondary schools,
by function and state: School year 1998–99

Within-state percentage distribution

State Total Instruction Support services Noninstruction
United States* 100.0 61.7 34.0 4.3
Alabama 100.0 61.9 31.2 6.9
Alaska* 100.0 56.6 40.1 3.3
Arizona* 100.0 60.1 35.0 5.0
Arkansas 100.0 64.1 29.0 6.9
California 100.0 61.0 35.0 4.0
Colorado 100.0 58.2 38.2 3.6
Connecticut* 100.0 63.6 31.6 4.9
Delaware 100.0 61.2 34.2 4.6
District of Columbia* 100.0 45.3 50.6 4.0
Florida 100.0 58.6 36.5 4.9
Georgia 100.0 62.5 31.9 5.5
Hawaii 100.0 62.8 30.6 6.6
Idaho 100.0 61.9 33.7 4.4
Illinois 100.0 60.6 35.9 3.5
Indiana 100.0 62.5 33.3 4.3
Iowa 100.0 58.8 33.7 7.5
Kansas 100.0 57.5 37.5 5.0
Kentucky 100.0 61.7 32.7 5.6
Louisiana* 100.0 60.4 31.5 8.1
Maine 100.0 67.3 29.3 3.5
Maryland 100.0 61.5 33.9 4.6
Massachusetts 100.0 66.6 30.3 3.1
Michigan 100.0 58.1 39.0 2.9
Minnesota 100.0 62.7 33.2 4.1
Mississippi 100.0 60.4 32.5 7.1
Missouri 100.0 61.6 34.1 4.3
Montana 100.0 62.5 33.4 4.1
Nebraska* 100.0 62.9 29.6 7.5
Nevada 100.0 59.9 37.0 3.1
New Hampshire* 100.0 65.2 31.3 3.5
New Jersey 100.0 59.9 37.2 2.9
New Mexico 100.0 56.8 38.4 4.8
New York 100.0 67.8 29.5 2.7
North Carolina 100.0 62.8 31.3 5.9
North Dakota 100.0 61.1 30.8 8.1
Ohio 100.0 59.0 37.4 3.6
Oklahoma 100.0 60.0 33.5 6.5
Oregon 100.0 59.2 37.3 3.4
Pennsylvania 100.0 63.4 32.9 3.7
Rhode Island 100.0 66.4 30.9 2.7
South Carolina 100.0 59.7 34.4 6.0
South Dakota 100.0 60.9 33.8 5.4
Tennessee 100.0 65.4 29.6 5.0
Texas 100.0 60.5 34.2 5.3
Utah 100.0 65.6 28.4 6.0
Vermont 100.0 64.6 32.6 2.8
Virginia* 100.0 60.8 34.0 5.2
Washington* 100.0 59.8 35.3 4.9
West Virginia 100.0 62.0 32.2 5.9
Wisconsin 100.0 62.5 34.4 3.1
Wyoming 100.0 60.0 36.4 3.5
Outlying areas
American Samoa 100.0 42.8 37.7 19.5
Guam — — — —
Northern Marianas 100.0 80.6 12.8 6.6
Puerto Rico 100.0 68.5 20.1 11.5
Virgin Islands 100.0 61.2 34.0 4.7

—Data not available.

*Distribution affected by imputations and redistribution of reported values to correct for missing items.
NOTE: Detail may not add to totals due to rounding. National figures do not include outlying areas.
SOURCE: U.S. Department of Education, National Center for Education Statistics, Common Core of Data (CCD), “National
Public Education Financial Survey,” 1998–99.

E D U C AT I O N S TAT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , FA L L 2 0 0 1 61
Elementary and Secondary Education

Table 5.—Student membership and current expenditures per pupil in membership for public elementary and
secondary schools, by function and state: School year 1998–99

Current expenditures per pupil in membership

Fall 1998
student Support
State membership Total Instruction services Noninstruction
1 1 1 1 1
United States 46,538,585 $6,508 $4,013 $2,213 $282
1 1 1 1 1
Alabama 747,980 5,188 3,210 1,618 359
2
Alaska 135,373 8,404 4,757 3,372 274
2
Arizona 848,262 4,672 2,806 1,633 233
Arkansas 452,256 4,956 3,176 1,437 343
1 1 1 1 1
California 5,926,037 5,801 3,539 2,031 232
Colorado 699,135 5,923 3,448 2,262 213
1 1
Connecticut 544,698 9,318 5,922 2,941 455
Delaware 113,262 7,706 4,716 2,634 356
1 1 2
District of Columbia 71,889 9,650 4,374 4,887 389
Florida 2,337,633 5,790 3,390 2,116 284
Georgia 1,401,291 6,092 3,810 1,946 336
Hawaii 188,069 6,081 3,818 1,861 402
Idaho 244,722 5,066 3,138 1,707 221
Illinois 2,011,530 6,762 4,097 2,431 235
Indiana 989,001 6,772 4,230 2,254 288
Iowa 498,214 6,243 3,671 2,105 468
Kansas 472,353 6,015 3,456 2,258 301
Kentucky 655,687 5,560 3,431 1,816 313
1 1
Louisiana 768,734 5,548 3,352 1,748 448
Maine 211,051 7,155 4,813 2,093 249
Maryland 841,671 7,326 4,504 2,486 336
Massachusetts 962,317 8,260 5,503 2,500 256
Michigan 1,720,287 7,432 4,320 2,895 217
Minnesota 856,455 6,791 4,259 2,253 279
Mississippi 502,379 4,565 2,755 1,486 324
Missouri 913,494 5,855 3,605 1,999 250
Montana 159,988 5,974 3,736 1,994 243
2 2
Nebraska 291,140 6,256 3,935 1,850 471
Nevada 311,061 5,587 3,346 2,068 174
2 2 2
New Hampshire 204,713 6,433 4,196 2,013 224
New Jersey 1,268,996 10,145 6,072 3,775 298
New Mexico 328,753 5,440 3,089 2,089 263
New York 2,877,143 9,344 6,335 2,756 253
North Carolina 1,254,821 5,656 3,553 1,770 334
North Dakota 114,927 5,442 3,325 1,674 443
Ohio 1,842,163 6,627 3,908 2,478 241
Oklahoma 628,492 5,303 3,183 1,777 343
Oregon 542,809 6,828 4,045 2,549 233
Pennsylvania 1,816,414 7,450 4,721 2,452 277
Rhode Island 154,785 8,294 5,503 2,564 227
1 1 1 1 1
South Carolina 664,600 5,656 3,375 1,944 337
South Dakota 132,495 5,259 3,202 1,776 282
1 1 1 1 1
Tennessee 905,454 5,123 3,350 1,516 258
Texas 3,945,367 5,685 3,440 1,944 301
Utah 481,176 4,210 2,762 1,194 254
Vermont 105,120 7,541 4,875 2,455 211
1 1
Virginia 1,124,022 6,350 3,862 2,156 332
2 2
Washington 998,053 6,110 3,654 2,159 297
West Virginia 297,530 6,677 4,137 2,147 393
Wisconsin 879,542 7,527 4,706 2,589 233
Wyoming 95,241 6,842 4,106 2,493 242
Outlying areas
American Samoa 15,372 2,283 977 861 445
Guam 32,222 — — — —
Northern Marianas 9,498 5,312 4,282 680 350
Puerto Rico 613,862 3,298 2,258 662 378
Virgin Islands 20,976 6,983 4,276 2,375 331

—Data not available.

1
Value contains imputation for missing data.
2
Value affected by redistribution of reported expenditure values for missing data items.
NOTE: Detail may not add to totals due to rounding. National figures do not include outlying areas.
SOURCE: U.S. Department of Education, National Center for Education Statistics, Common Core of Data (CCD): “National Public
Education Financial Survey,” 1998–99; and “State Nonfiscal Survey of Public Elementary/Secondary Education,” 1998–99 (Revised).

62 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
 Revenues and Expenditures for Public Elementary and Secondary Education: School Year 1998–99

Table 6.—Total expenditures, including interest on debt, for public elementary and secondary education and other related programs
supported by public school districts: School year 1998–99
[In thousands of dollars]

Public elementary and secondary education

Total expenditures Facilities
(including interest acquisition and Replacement Other Interest
State on debt) Current construction equipment programs on debt
1 1 2 1 1
United States $355,859,449 $302,873,955 $31,606,994 $7,919,652 $5,262,613 $8,196,236
Alabama 4,626,884 3,880,188 397,740 121,041 171,898 56,016
Alaska 1,264,525 1,137,610 84,199 20,582 4,925 17,210
Arizona 5,341,305 3,963,428 1,061,066 281,668 28,041 7,103
Arkansas 2,480,435 2,241,244 124,257 50,789 15,724 48,421
California 39,973,279 34,379,878 3,779,490 909,009 663,236 241,666
Colorado 5,024,543 4,140,699 508,227 120,565 38,260 216,791
1 1 1
Connecticut 5,827,005 5,075,581 447,754 85,247 96,314 122,109
Delaware 972,747 872,786 56,322 17,189 14,516 11,934
1 1 1
District of Columbia 755,576 693,716 30,086 13,756 1,784 16,235
Florida 16,615,626 13,534,374 1,889,232 210,164 459,476 522,379
Georgia 10,113,816 8,537,177 1,106,237 248,728 44,952 176,722
Hawaii 1,400,063 1,143,713 157,855 25,309 32,952 40,235
Idaho 1,420,733 1,239,755 114,393 34,530 3,507 28,549
Illinois 16,255,656 13,602,965 1,701,432 540,037 121,406 289,817
Indiana 8,111,363 6,697,468 682,366 130,852 53,115 547,562
Iowa 3,521,011 3,110,585 226,933 113,256 21,400 48,837
Kansas 3,155,418 2,841,147 89,727 136,783 3,089 84,673
Kentucky 3,948,832 3,645,631 54,693 124,920 44,661 78,926
1 1
Louisiana 4,801,137 4,264,981 289,931 128,729 17,627 99,868
Maine 1,642,951 1,510,024 51,338 32,080 16,550 32,961
Maryland 7,182,797 6,165,934 814,071 113,982 20,903 67,906
Massachusetts 8,368,976 7,948,502 39,412 115,038 83,754 182,270
Michigan 15,604,340 12,785,480 1,688,999 345,938 320,671 463,254
Minnesota 7,353,993 5,816,329 782,357 203,468 279,907 271,932
Mississippi 2,830,349 2,293,188 355,383 111,877 16,001 53,901
Missouri 6,242,701 5,348,366 420,587 194,477 105,770 173,501
Montana 1,052,773 955,695 55,463 23,714 6,426 11,475
Nebraska 2,086,536 1,821,310 140,745 88,165 2,631 33,686
2 2
Nevada 2,301,747 1,738,009 381,022 68,182 11,776 102,758
New Hampshire 1,430,462 1,316,946 63,095 21,083 3,305 26,033
New Jersey 14,250,489 12,874,579 865,793 137,544 165,775 206,797
New Mexico 2,116,905 1,788,382 264,233 22,650 10,971 30,669
New York 30,357,364 26,885,444 1,407,188 379,959 1,010,219 674,555
1 1
North Carolina 8,712,151 7,097,882 1,209,573 158,241 54,456 191,999
North Dakota 711,740 625,428 50,176 23,608 4,800 7,728
Ohio 14,006,167 12,207,147 752,439 455,709 369,125 221,747
Oklahoma 3,668,577 3,332,697 182,998 107,926 10,464 34,491
Oregon 4,173,853 3,706,044 260,634 81,878 25,653 99,643
Pennsylvania 16,343,516 13,532,211 1,602,282 259,832 325,740 623,451
Rhode Island 1,341,549 1,283,859 8,958 20,068 7,885 20,779
South Carolina 4,612,674 3,759,042 608,673 99,603 45,706 99,649
South Dakota 840,515 696,785 89,145 36,822 1,454 16,309
Tennessee 5,432,316 4,638,924 553,330 116,811 24,753 98,497
Texas 27,496,479 22,430,153 3,323,470 631,757 157,648 953,452
Utah 2,536,260 2,025,714 324,244 48,275 59,763 78,265
Vermont 853,705 792,664 24,687 19,455 2,272 14,626
1 1
Virginia 8,376,912 7,137,421 699,794 220,651 131,782 187,264
2 2
Washington 7,449,610 6,098,036 912,915 133,876 33,210 271,573
West Virginia 2,207,537 1,986,562 98,113 79,550 29,321 13,992
Wisconsin 7,928,144 6,620,653 740,573 216,222 85,286 265,409
Wyoming 735,411 651,622 33,364 38,060 1,753 10,612
Outlying areas
American Samoa 43,864 35,092 2,597 3,210 2,965 0
Guam — — — — — —
Northern Marianas 55,463 50,450 4,466 318 230 0
Puerto Rico 2,090,530 2,024,499 43 3,826 43,518 18,645
Virgin Islands 152,211 146,474 1,017 2,983 1,737 0

—Data not available.

1
Value contains imputation for missing data. Imputed value is less than 2 percent of total expenditures in any one state.
2
Value affected by redistribution of reported values for missing data items.
NOTE: Detail may not add to totals due to rounding. National figures do not include outlying areas.
SOURCE: U.S. Department of Education, National Center for Education Statistics, Common Core of Data (CCD),“National Public Education Financial Survey,” 1998–99.

E D U C AT I O N S TAT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , FA L L 2 0 0 1 63
64 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
 P O S T S E C O N D A RY E D U C AT I O N

High School Academic Curriculum and the Persistence Path Through

College: Persistence and Transfer Behavior of Undergraduates 3 Years After
Entering 4-Year Institutions
Laura Horn and Lawrence K. Kojaku .................................................................... 65
Bridging the Gap: Academic Preparation and Postsecondary Success of
First-Generation Students
Edward C. Warburton, Rosio Bugarin, and Anne-Marie Nuñez ............................ 73
Middle Income Undergraduates: Where They Enroll and How They Pay
for Their Education
Jennifer B. Presley and Suzanne B. Clery .............................................................. 78
National Postsecondary Student Aid Study: Student Financial Aid Estimates
for 1999–2000
Andrew G. Malizio ................................................................................................ 82
Competing Choices: Men’s and Women’s Paths After Earning a Bachelor’s
Degree
Michael S. Clune, Anne-Marie Nuñez, and Susan P. Choy .................................... 88
Postsecondary Institutions in the United States: 1993–94 and 1998–99
Patricia Q. Brown .................................................................................................. 93

High School Curriculum

High School Academic Curriculum and the Persistence Path Through
College: Persistence and Transfer Behavior of Undergraduates 3 Years After
Entering 4-Year Institutions
——————————————————————————————————
Laura Horn and Lawrence K. Kojaku
This article was originally published as the Executive Summary of the Statistical Analysis Report of the same name. The sample survey data are from the
Beginning Postsecondary Students Longitudinal Study (BPS).

This report examines the relationship between high school preparation are based on academic courses taken in high
academic curricula and students’ persistence path through school as reported by students on their college entrance
college, approximately 3 years after first enrolling. The data exam applications.
are drawn from the 1996 Beginning Postsecondary Students
Longitudinal Study (BPS:96/98), a study of beginning The high school academic curriculum measure identifies
postsecondary students who first enrolled in a 4-year three levels of coursetaking: (1) core curriculum or below,
college in 1995–96. Measures of high school academic (2) mid-level, and (3) rigorous. The lowest threshold is

E D U C AT I O N S T AT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , FA L L 2 0 0 1 65
Postsecondary Education

based on the core New Basics curriculum first recom- Level of High School Academic Curriculum
mended by the National Commission on Excellence in Completed
Education in A Nation at Risk (1983). Core curriculum The distribution of beginning students enrolled in 4-year
includes 4 years of English, 3 years of mathematics, 3 years institutions across the three levels of coursetaking is
of science, and 3 years of social studies.1 The highest displayed in figure A. Approximately one-third (31 percent)
threshold, or rigorous curriculum, identified in the current reported completing course work no higher than core
study, includes 4 years of English, 3 years of a foreign curricula, one-half completed mid-level curricula, and the
language, 3 years of social studies, 4 years of mathematics remaining one-fifth (19 percent) completed rigorous
(including precalculus or higher), 3 years of science curricula.
(including biology, chemistry, and physics), and at least one
Advanced Placement (AP) course or test taken. Mid-level The level of high school academic curriculum completed by
covers curricula between core and rigorous curricula, but at beginning 4-year college students was associated with their
a minimum must include algebra I, geometry, at least 1 year demographic and socioeconomic characteristics and also
of a foreign language, and two science classes from the with the economic status of the student body in their high
combination of biology, chemistry, and physics.2 schools. Specifically, students from low-income families,
students whose parents had no more than a high school
education, and students who graduated from high schools
in which 25 percent or more of the students were eligible
1
It also included courses in computer science, but students did not report on such for free or reduced-price lunches were less likely than their
courses on their entrance exam applications.
2 more advantaged counterparts to report completing rigor-
The research of Burkam, Lee, and Smerdon (1996) and Adelman (1999) was used
extensively in guiding the development of the academic curriculum variable. ous high school curricula.

Figure A.—Percentage distribution of 1995–96 beginning postsecondary students

who enrolled in 4-year institutions, by high school academic curriculum

Core curriculum Rigorous3 (19%)

or less1 (31%)

Mid-level2 (50%)

1
Core curriculum includes 4 years of English, 3 years of social studies, 3 years of mathematics,
and 3 years of science.
2
Mid-level curriculum exceeds core curriculum, but is less than rigorous. Includes at a
minimum 1 year of a foreign language, geometry, algebra I, and 3 years of science including
two of the following courses: biology, chemistry, or physics.
3
Rigorous curriculum includes 4 years of English, 4 years of mathematics (including
precalculus or higher), 3 years of a foreign language, 3 years of social studies, 3 years of
science (including biology, chemistry, and physics), and at least one Advanced Placement (AP)
class or test taken.
SOURCE: U.S. Department of Education, National Center for Education Statistics, 1996
Beginning Postsecondary Students Longitudinal Study, “First Follow-up” (BPS:96/98).

66 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
 High School Academic Curriculum and the Persistence Path Through College

Related in part to family income, racial/ethnic group differ- likelihood of attending selective 4-year colleges or univer-
ences also were apparent. Black students were much less sities.3 For example, 71 percent of students who com-
likely than either White or Asian/Pacific Islander students to pleted rigorous curricula enrolled in a selective college or
complete rigorous curricula (8 percent vs. 20 and 31 percent, university, compared with 40 percent who completed mid-
respectively) and more likely to complete programs no higher level curricula and 32 percent who completed core
than the core curriculum (42 percent vs. 29 and 27 percent). curricula or lower.
Asian/Pacific Islander students were the most likely to
complete rigorous curricula (31 percent). While it appears Postsecondary Persistence 3 Years After
as though White students were more likely than Hispanic Enrolling
students to complete rigorous curricula (20 percent vs. As of 1998, roughly two-thirds of students who had first
16 percent) and less likely to complete no higher than core enrolled in a 4-year college in 1995–96 were still enrolled in
curricula (29 percent vs. 34 percent), there was not enough the same college (including 6 percent who had left and
statistical evidence to draw this conclusion. returned; figure B). One-fifth had transferred to another
institution, and 13 percent had left and not returned.
High school academic curriculum also had an obvious
association with where students first enrolled in college.
As the level of curricula increased, so did students’ 3
Selectivity was determined by admitted students’ entrance exam scores.

Figure B.—Postsecondary persistence path of 1995–96 beginning students who enrolled in 4-year institutions, 3 years after first enrollment (as of 1998)

Retention at first institution Status in 1998 Status of all in 1998

Earned bachelor's degree (0.7%)

No transfer
Left without
return (16.4%)

Off bachelor's
Continuously enrolled (59.9%) track3 (8.0%)
Left without
return (14.0%)
Bachelor's
track1
(75.6%)
Off
Stopped out and returned (6.2%) bachelor's
Transferred track3
Left without return (12.7%) Bachelor's (18.0%)
Left without track4
Stayed on return (4.6%) (68.1%)
bachelor's
track2
(38.9%)
Transferred (20.4%)

Off bachelor's
track3 (56.6%)

1
Maintained continuous enrollment in first institution or earned a bachelor’s degree.
2
Transferred to a 4-year institution with no break in enrollment.
3
Stopped out (break of more than 4 months) and/or enrolled in a less-than-4-year institution.
4
Maintained continuous enrollment in any 4-year institution.
NOTE: Detail may not add to 100 because of rounding.
SOURCE: U.S. Department of Education, National Center for Education Statistics, 1996 Beginning Postsecondary Students Longitudinal Study, “First Follow-up” (BPS:96/98).

E D U C AT I O N S T AT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , FA L L 2 0 0 1 67
Postsecondary Education

The level of college students’ high school curricula was The difference between levels of academic curricula was
strongly related to their persistence in postsecondary especially notable with respect to staying on track to a
education. This was true both for maintaining enrollment at bachelor’s degree (i.e., continuous enrollment in any 4-year
their initial institution (institutional retention) and, if they institution). As the level of academic curricula increased, so
transferred, staying on track to a bachelor’s degree.4 For did the proportion of undergraduates who stayed on track
example, 79 percent of students who had participated in (figure D). As of 1998, the vast majority (87 percent) of
rigorous high school academic curricula were continuously those who had participated in rigorous high school aca-
enrolled in their initial institution (including 1 percent who demic curricula were still on track to a bachelor’s degree,
had attained a bachelor’s degree; figure C). In contrast, 62 compared with 71 percent of those in mid-level curricula
percent and 55 percent, respectively, of those in mid-level and 62 percent of those who completed core curricula or
curricula or core curricula or lower were continuously lower. Correspondingly, the proportion of those who had
enrolled in their initial institution. Students in rigorous left postsecondary education and did not return declined
curricula also were less likely to transfer from their first with each successive level of academic curriculum (from
institution (13 percent) than those who participated in less 17 percent to 10 percent to 4 percent).
than rigorous curricula, whether in mid-level or core or
lower curricula (23 percent of both groups transferred). Patterns of Transfer
4
The rate of transfer for those who completed no higher
The “persistence track” to a bachelor’s degree is defined as maintaining uninterrupted
enrollment in a 4-year institution toward a bachelor’s degree objective. It includes than a core curriculum in high school was the same as for
those who transfer between 4-year institutions without a break in enrollment. The those who completed mid-level curricula. Roughly one-
persistence track concept was first developed by Carroll (1989) to demonstrate the
optimal path to a bachelor’s degree and the adverse effects of straying from the path. quarter of these students had transferred from their first

Figure C.—Percentage distribution of 1995–96 beginning students’ 1998 enrollment status at their first institution for those who began in a 4-year institution,
by high school academic curriculum

Status at first institution

Bachelor's degree (0.7%) Bachelor's degree (0.2%) Bachelor's degree (1.2%)

Continuously enrolled (54.6%) Continuously enrolled (61.8%)

Continuously enrolled (78.1%)

Stopped out and returned (6.3%)

Stopped out and returned (5.6%)
Left without return (15.4%)
Left without return (9.1%)

Stopped out and returned (4.2%)

Left without return (3.1%)
Transferred (23.1%) Transferred (23.2%)
Transferred (13.4%)

Core curriculum or less1 Mid-level2 Rigorous3

1
Core curriculum includes 4 years of English, 3 years of social studies, 3 years of mathematics, and 3 years of science.
2
Mid-level curriculum exceeds core curriculum, but is less than rigorous. Includes at a minimum 1 year of a foreign language, geometry, algebra I, and 3 years of science including
two of the following courses: biology, chemistry, or physics.
3
Rigorous curriculum includes 4 years of English, 4 years of mathematics (including precalculus or higher), 3 years of a foreign language, 3 years of social studies, 3 years of science
(including biology, chemistry, and physics), and at least one Advanced Placement (AP) class or test taken.
NOTE: Detail may not add to 100 because of rounding.
SOURCE: U.S. Department of Education, National Center for Education Statistics, 1996 Beginning Postsecondary Students Longitudinal Study, “First Follow-up” (BPS:96/98).

68 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
 High School Academic Curriculum and the Persistence Path Through College

Figure D.—Enrollment status in 1998 of 1995–96 beginning students who first enrolled in 4-year institutions, by high school academic curriculum

Enrollment status of all in 1998

Off bachelor's
Off bachelor's Off bachelor's track2 (9.2%)
track2 (21.2%) track2 (19.0%)
Left without
return (4.2%)

Left without
Bachelor's return (10.3%)
track1 Left without Bachelor's
(61.9%) return (16.9%) track1
(70.8%) Bachelor's
track1
(86.6%)
Core curriculum or less3 Mid-level4 Rigorous5

1
Continuously enrolled in any 4-year institution.
2
Stopped out (break of 4 or more months) and/or enrolled in a less-than-4-year institution.
3
Core curriculum includes 4 years of English, 3 years of social studies, 3 years of mathematics, and 3 years of science.
4
Mid-level curriculum exceeds core curriculum, but is less than rigorous. Includes at a minimum 1 year of a foreign language, geometry, algebra I, and 3 years of science including
two of the following courses: biology, chemistry, or physics.
5
Rigorous curriculum includes 4 years of English, 4 years of mathematics (including precalculus or higher), 3 years of a foreign language, 3 years of social studies, 3 years of science
(including biology, chemistry, and physics), and at least one Advanced Placement (AP) class or test taken.
NOTE: Detail may not add to 100 because of rounding.
SOURCE: U.S. Department of Education, National Center for Education Statistics, 1996 Beginning Postsecondary Students Longitudinal Study, “First Follow-up” (BPS:96/98).

institution, while just 13 percent in rigorous high school were much less likely to transfer either to a less-than-
curricula had done so (figure C). However, as shown in 4-year or a for-profit institution (21 percent) than their
figure E, among students who transferred, as their level of counterparts in mid-level (40 percent) or core curricula or
academic curricula increased, so did the likelihood of lower (46 percent).
staying on track to a bachelor’s degree (i.e., they trans-
ferred to another 4-year institution without a break in Examining the selectivity of students’ first institution
enrollment). For those who had completed core curricula compared with their second, the likelihood of making a
or lower, 31 percent of transfers stayed on the bachelor’s lateral transfer from selective to selective institution was
degree track, as did 39 percent of transfers who had much higher for those completing rigorous high school
completed mid-level curricula and 60 percent of those curricula (49 percent) than for those completing either mid-
who had completed rigorous high school curricula. level curricula (33 percent) or no higher than core curricula
(22 percent). Similarly, the likelihood of transferring from
These differences were clearly evident when examining the a selective institution to a less-than-4-year or for-profit
destination of transfers. Staying on track to a bachelor’s institution was much lower for those completing rigorous
degree implies staying enrolled in a 4-year institution, so curricula (19 percent) than for their counterparts complet-
all transfers who stayed on track transferred to 4-year ing mid-level curricula (40 percent) or no higher than core
institutions. However, there were also differences across curricula (47 percent).
high school academic curricula with respect to the selectiv-
ity of the 4-year institutions where students transferred While the patterns of transfer appear to be similar among
(table A). Among all transfers who completed rigorous those who began in less-selective institutions (i.e., those
high school curricula, 40 percent transferred to selective completing rigorous high school curricula were more likely
institutions, compared with just 21 percent of their than those in less rigorous curricula to transfer to selective
counterparts who completed mid-level curricula and institutions or to transfer laterally, and less likely to transfer
17 percent who completed core curricula or lower. Corre- to less-than-4-year or for-profit institutions), there is not
spondingly, transfers who completed rigorous curricula enough statistical evidence to draw this conclusion.

E D U C AT I O N S T AT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , FA L L 2 0 0 1 69
Postsecondary Education

Figure E.—For 1995–96 beginning students who transferred from a 4-year institution, percentage distribution of enrollment status in 1998 with respect to
earning a bachelor’s degree, by high school academic curriculum

Enrollment status of transfers in 1998

Left without Left without Left without

return (5.4%) return (4.5%) return (7.4%)
Bachelor's
track1 Bachelor's
(31.1%) track1
(38.5%)
Off bachelor's
Bachelor's track2 (32.3%)
track1
Off bachelor's (60.4%)
Off bachelor's track2 (57.0%)
track2 (63.5%)
Core curriculum or less3 Mid-level4 Rigorous5

1
Continuously enrolled in any 4-year institution.
2
Stopped out (break of 4 or more months) and/or enrolled in a less-than-4-year institution.
3
Core curriculum includes 4 years of English, 3 years of social studies, 3 years of mathematics, and 3 years of science.
4
Mid-level curriculum exceeds core curriculum, but is less than rigorous. Includes at a minimum 1 year of a foreign language, geometry, algebra I, and 3 years of science including
two of the following courses: biology, chemistry, or physics.
5
Rigorous curriculum includes 4 years of English, 4 years of mathematics (including precalculus or higher), 3 years of a foreign language, 3 years of social studies, 3 years of science
(including biology, chemistry, and physics), and at least one Advanced Placement (AP) class or test taken.
NOTE: Detail may not add to 100 because of rounding.
SOURCE: U.S. Department of Education, National Center for Education Statistics, 1996 Beginning Postsecondary Students Longitudinal Study, “First Follow-up” (BPS:96/98).

Transfers left postsecondary education at similar rates no including the type of postsecondary institution, full- or
matter what their high school academic curricula. As shown part-time attendance, and work status. In addition,
in figure E, 5 to 7 percent of transfers left postsecondary because previous research has shown first-year grade-point
education altogether as of 1998. average (GPA) to be a strong predictor of success in
college (e.g., Pascarella and Terenzini 1991), both analyses
Controlling for Related Variables were run twice, once without GPA, and once including
In addition to high school academic curricula, many other GPA as an independent variable.
variables can influence postsecondary education outcomes.
Therefore, it is necessary to use multivariate analysis High school academic curricula
techniques to disentangle the net influence of related The results indicated a strong association between high
variables on the outcome of interest. school academic curricula and both measures of persis-
tence. Students who participated in rigorous high school
In this study, covariance adjustment techniques based on curricula were at a distinct advantage over those who
simple linear regression models were used to analyze two completed no higher than core curricula (the comparison
persistence outcomes: (1) continuous enrollment at the group). In addition, there was some evidence that complet-
initial institution, and (2) staying on track to a bachelor’s ing mid-level curricula also was associated with higher
degree. Independent variables reflected students’ academic rates of staying on track to a bachelor’s degree when
experience in high school (academic curricula and college compared to completing programs no higher than core
entrance exam scores), demographic characteristics curricula. However, the difference did not reach statistical
(gender, race/ethnicity, age), socioeconomic characteristics significance after adjusting for the design effect of the
(income and parents’ education), and the economic status dependent variable.5
of their high schools (the proportion of students eligible
for free or reduced-price lunches). Other variables re-
flected students’ experiences in their first year in college, 5
After adjusting for the design effect, the t-value of the coefficient was 1.90.

70 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
 High School Academic Curriculum and the Persistence Path Through College

Table A.—Among 1995–96 beginning students who transferred from a 4-year institution, percentage
distribution according to their transfer institution, by high school academic curriculum and
selectivity of first institution

Transfer destination
4-year institution Less-than-4-year
Selective1 Less-selective2 or for-profit 4-year
All transfers from 4-year institutions
Total 21.3 39.5 39.3
Core curriculum or less3 17.4 36.9 45.7
Mid-level4 20.9 39.0 40.1
Rigorous5 40.4 38.9 20.8
Began in selective institution
Total 34.1 30.0 35.9
Core curriculum or less3 21.5 31.4 47.1
Mid-level4 32.8 27.6 39.6
Rigorous5 48.5 32.4 19.1
Began in less-selective institution
Total 14.7 44.3 41.0
Core curriculum or less3 16.2 38.4 45.4
Mid-level4 13.3 46.2 40.5
Rigorous5 25.0 50.9 24.0

1
Selective institutions are public and private not-for-profit institutions in which students’ average SAT scores
exceeded 1000 or Carnegie classifications in which a majority of students were enrolled in very selective
institutions.
2
Less-selective institutions are all others not identified in the selective groups.
3
Core curriculum includes 4 years of English, 3 years of social studies, 3 years of mathematics, and 3 years of science.
4
Mid-level curriculum exceeds core curriculum, but is less than rigorous. Includes at a minimum 1 year of a foreign
language, geometry, algebra I, and 3 years of science including two of the following courses: biology, chemistry, or
physics.
5
Rigorous curriculum includes 4 years of English, 4 years of mathematics (including precalculus or higher), 3 years
of a foreign language, 3 years of social studies, 3 years of science (including biology, chemistry, and physics), and at
least one Advanced Placement (AP) class or test taken.
NOTE: Detail may not add to 100 because of rounding.
SOURCE: U.S. Department of Education, National Center for Education Statistics, 1996 Beginning Postsecondary
Students Longitudinal Study, “First Follow-up” (BPS:96/98).

SAT scores and other variables Other variables were also significantly associated with
In both persistence analyses, prior to including first-year both measures of persistence after holding related vari-
college GPA as an independent variable, SAT composite test ables consistent. For example, students whose parents did
score levels and high school academic curriculum levels not attend college were less likely to persist than those
were significantly associated with the outcome. However, whose parents were college educated.6 In addition,
once GPA was taken into account, high school academic students who started college attending part time and/or
curriculum remained a significant predictor of persistence working full time were less likely to persist than their
(specifically, completing a rigorous versus core or lower counterparts, as were those who first enrolled in less-
curriculum), but the association between levels of SAT selective 4-year institutions compared with those in
scores and persistence disappeared. In other words, once selective institutions.
all related variables were taken into consideration, includ-
ing college GPA, entrance exam scores were no longer
associated with the likelihood of persisting, either with
6
respect to institutional retention or staying on track to a A recent NCES report provides a detailed analysis of the experiences of “first-
generation” college students—those whose parents did not attend college
bachelor’s degree. (Warburton, Bugarin, and Nuñez 2001).

E D U C AT I O N S T AT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , FA L L 2 0 0 1 71
Postsecondary Education

Conclusions Taken together, the results suggest that completing a

The findings of this study demonstrated a consistent rigorous academic curriculum in high school may help
advantage experienced by students who completed rigorous students overcome socioeconomic disadvantages such as
high school curricula, and to a lesser extent by those low family income and parents with no college experience,
completing mid-level curricula, over their peers completing as well as helping those who get a poor start in college
core curricula or lower. (whether academic or social) and decide to transfer.

However, the level of high school curricula students reported References

completing also was related to their family background Adelman, C. (1999). Answers in the Toolbox: Academic Intensity,
characteristics and indicators of socioeconomic status, Attendance Patterns, and Bachelor’s Degree Attainment. U.S.
Department of Education, Office of Educational Research and
including family income, parents’ education, race/ethnicity,
Improvement. Washington, DC: U.S. Government Printing
and the economic status of their high school’s student body. Office.
All of these factors relate to whether or not students have
Burkam, D.T., Lee, V.E., and Smerdon, B.A. (1996). Mathematics
the opportunities to participate in and complete rigorous Coursetaking and the NELS:88 Transcript Data. Ann Arbor, MI:
curricula. Moreover, students’ success in staying in college University of Michigan.
was also related to where they first enrolled and how well Carroll, C.D. (1989). College Persistence and Degree Attainment for
they did in their first year. Yet, even when all these factors 1980 High School Graduates: Hazards for Transfers, Stopouts, and
were taken into consideration, the advantage of completing Part-Timers (NCES 89–302). U.S. Department of Education,
a rigorous high school academic curriculum remained. National Center for Education Statistics. Washington, DC: U.S.
Government Printing Office.
The same was not observed for levels of SAT scores. Similar National Commission on Excellence in Education. (1983).
to the findings for curriculum levels, SAT scores were related A Nation at Risk: The Imperative for Educational Reform.
Washington, DC: Author.
to persistence when first-year college GPA was not included
Pascarella, E.T., and Terenzini, P.T. (1991). How College Affects
in the regression. However, after GPA was added, high school
Students: Findings and Insights From Twenty Years of Research.
curriculum remained a significant factor, but SAT scores did San Francisco: Jossey-Bass.
not. These findings are consistent with recent research based
Warburton, E.C., Bugarin, R., and Nuñez, A.-M. (2001). Bridging
on high school transcripts for a cohort of 1980 high school the Gap: Academic Preparation and Postsecondary Success of First-
sophomores (Adelman 1999); this study demonstrated that Generation Students (NCES 2001–153). U.S. Department of
high school curriculum was a stronger predictor of bachelor’s Education, National Center for Education Statistics. Washing-
degree attainment than standardized test scores or other ton, DC: U.S. Government Printing Office.
measures of high school academic performance.

Perhaps most notable in the current study is the apparent

Data source: The NCES 1996 Beginning Postsecondary Students
benefit of a strong high school academic curriculum for Longitudinal Study, “First Follow-up” (BPS:96/98).
transfer students. Students who transfer from their initial For technical information, see the complete report:
4-year college may do so because they are struggling either Horn, L., and Kojaku, L.K. (2001). High School Academic Curriculum and
academically or socially, and attempting to find a better fit the Persistence Path Through College: Persistence and Transfer
Behavior of Undergraduates 3 Years After Entering 4-Year Institutions
in another institution. One-fifth of 1995–96 beginning (NCES 2001–163).
undergraduates enrolled in 4-year colleges had transferred Author affiliations: L. Horn and L.K. Kojaku, MPR Associates, Inc.
from their first institution by 1998. For these students in For questions about content, contact Aurora D’Amico
particular, as their level of high school academic curriculum (aurora.d’amico@ed.gov).
increased, so did their likelihood of staying on track to a To obtain the complete report (NCES 2001–163), call the toll-free
ED Pubs number (877–433–7827), visit the NCES Web Site
bachelor’s degree (by transferring to another 4-year institu- (http://nces.ed.gov), or contact GPO (202–512–1800).
tion without a break in enrollment).

72 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
Academic Preparation
Bridging the Gap: Academic Preparation and Postsecondary Success of
First-Generation Students
——————————————————————————————————
Edward C. Warburton, Rosio Bugarin, and Anne-Marie Nuñez
This article was originally published as the Executive Summary of the Statistical Analysis Report of the same name. The sample survey data are from the
Beginning Postsecondary Students Longitudinal Study (BPS).

This report examines the high school preparation and with the amount of remedial coursework they took, and
postsecondary persistence of first-generation students— with their rates of persistence and attainment. As overall
those students whose parents had no education beyond high school academic rigor increased, so did students’
high school—and compares them with students whose GPA. Students who did not exceed the requirements of the
parents went to college. Previous research has demonstrated core New Basics curriculum had a lower GPA than did
that first-generation students exhibit different college those who exceeded them (2.5 points vs. 3.1 points). The
enrollment and persistence behaviors than their counter- rigor of students’ high school curriculum was also related
parts whose parents have more education. Such studies to the number of remedial courses they took during their
found that first-generation students were less likely than first year of postsecondary education. As the rigor of the
their peers to complete advanced mathematics classes in secondary curriculum increased, the proportion of stu-
high school. Even among those qualified for college, first- dents who took one or more remedial courses decreased
generation students were less likely to enroll in 4-year from 21 percent to 3 percent.
institutions (Horn and Nuñez 2000). Independent of other
relevant demographic, enrollment, and college involvement High school academic preparation was also related to
factors, first-generation status was also found to be nega- students’ likelihood of remaining enrolled in postsecondary
tively associated with students’ persistence and attainment education. In general, the more rigorous their high school
(Nuñez and Cuccaro-Alamin 1998). curriculum, the more likely students were to persist (or to
attain a degree) at the initial postsecondary institution in
What has not been well understood, however, is the extent which they enrolled. While 62 percent of students who
to which the academic preparation of first-generation did not exceed the core New Basics requirements were
students in high school affects their persistence and still enrolled or had attained a degree as of spring 1998,
attainment in postsecondary education. The purpose of this 84 percent of students who exceeded the requirements did
report is to examine whether first-generation students who so. Likewise, the more rigorous the students’ high school
were otherwise equally prepared academically were compa- curriculum, the higher their likelihood of staying on the
rable to students whose parents went to college in terms of persistence track to a bachelor’s degree: 87 percent of
their grade-point averages (GPAs), number of remedial students who took rigorous academic coursework in high
courses in postsecondary education, and rates of persistence school stayed on the persistence track, compared with
(i.e., whether they were retained at their first institution, 62 percent of students who did not take such coursework.
had stayed on a persistence track toward the bachelor’s Finally, students whose curriculum was rigorous were more
degree,1 or had attained a degree). This analysis focuses on likely to still be enrolled and working for a degree than
a subset of 1995–96 beginning postsecondary students who students who did not exceed the core New Basics require-
started their postsecondary education in 4-year institutions. ments (93 percent vs. 75 percent).

High School Coursetaking Effect of First-Generation Status

2
The academic rigor of students’ high school curriculum This study found a relationship between parents’ education
was strongly associated with their postsecondary GPA, level and the likelihood that students would undertake a
more rigorous high school curriculum and, consequently,
1
Students’ status with respect to the “persistence track to a bachelor’s degree” is enroll, perform well, and persist in 4-year postsecondary
defined by three values: stayed on the persistence track (i.e., stayed in the same
4-year institution or made a lateral transfer to a different 4-year institution), left the institutions. Overall, first-generation status was shown to
persistence track (“stopped out” for more than 4 months or made an immediate or have a negative association with students’ academic prepa-
delayed downward transfer), or left postsecondary education (was neither still enrolled
at the initial institution nor had transferred to another postsecondary institution). ration and persistence.
2
“Academic rigor” is defined by four variables that describe the overall difficulty of
students’ high school coursework: core New Basics or below, beyond New Basics I
(somewhat exceeded core New Basics), beyond New Basics II (substantially exceeded
core New Basics), and rigorous (maximally exceeded core New Basics).

E D U C AT I O N S T AT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , FA L L 2 0 0 1 73
Postsecondary Education

Preparation for postsecondary education counterparts whose parents were college graduates to
Compared with students whose parents were college score below 550 points and less likely to score 651 points
graduates,3 first-generation students were less likely to have or more. Finally, a lower proportion of first-generation
taken calculus in high school (20 percent vs. 34 percent). students reported taking any Advanced Placement (AP)
A full 40 percent of first-generation students did not exceed tests than did both students whose parents had attended
the core New Basics curriculum (figure A). Furthermore, college and those whose parents had graduated (8 percent
while about one-fifth of students whose parents had a vs. 14 percent and 22 percent, respectively).
bachelor’s degree took rigorous courses in high school, just
9 percent of first-generation students did so. Postsecondary enrollment and performance
In this study, students’ patterns of postsecondary enrollment
First-generation students were also less likely to take and academic performance confirmed previous research
college entrance examinations, and when they did, they showing differential behaviors between first-generation
were more likely than their peers to have lower scores. students and their peers whose parents were college
Whereas 15 percent of students whose parents were college educated. Of the students who attended 4-year institutions,
graduates scored 790 points or lower on the SAT or ACT first-generation students were much more likely to attend
(i.e., the lowest quartile), almost 40 percent of first- public comprehensive institutions instead of research
generation students scored in this quartile. On the SAT II, universities than those with at least one parent who had a
first-generation students were more likely than their bachelor’s degree (41 percent vs. 26 percent). More than
one-quarter (27 percent) of first-generation students
3
attended part time in the 1997–98 academic year, and these
Whenever the term “college graduates” is used, it means that at least one parent had
attained a bachelor’s degree. students were much more likely to work full time compared

Figure A.—Percentage distribution of 1995–96 beginning postsecondary students according to academic rigor of
secondary school curriculum, by first-generation status

Core New Basics 40 First-generation student

or below1 28
Non-first-generation student5

Beyond 30
New Basics I2 28

Beyond 21
New Basics II3 23

9
Rigorous4
22

0 25 50 75 100

Percent

1
Core New Basics curriculum includes 4 years of English, 3 years of mathematics, and 3 years of science and social studies.
2
Beyond New Basics I includes core New Basics and at least two of three science courses (biology, chemistry, or physics), and
algebra I and geometry, plus 1 year of foreign language.
3
Beyond New Basics II includes core New Basics, advanced science (biology, chemistry, and physics), and advanced math (including
algebra I, geometry, algebra II), plus 2 years of foreign language.
4
Rigorous includes core New Basics, advanced science (biology, chemistry, and physics), and 4 years of math (including algebra I,
geometry, algebra II, precalculus), plus 3 years of foreign language and one honors/Advanced Placement (AP) course or AP test
score.
5
Non-first-generation students are those whose parents had any college experience.
NOTE: Detail may not add to 100 because of rounding. Includes public and private not-for-profit 4-year institutions. Students in
private for-profit 4-year institutions are excluded from this analysis because the sample size was too small (less than 1 percent).
SOURCE: U.S. Department of Education, National Center for Education Statistics, 1996 Beginning Postsecondary Students
Longitudinal Study, “First Follow-up”(BPS:96/98).

74 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
 Bridging the Gap: Academic Preparation and Postsecondary Success of First-Generation Students

to students whose parents had a college degree. By the end First-generation students were less likely to be enrolled
of the 1997–98 academic year, a larger proportion of first- continuously or to attain a degree at their initial post-
generation students (25 percent) had chosen business/ secondary institution than students whose parents had
management as their major field of study, compared with completed college (60 percent vs. 73 percent). They were
their non-first-generation counterparts (17 percent). also more likely to have stopped out5 or left their first
institution of enrollment than their peers whose parents had
In general, first-generation students had lower first-year a college degree (19 percent vs. 8 percent). These differences
GPAs than students whose parents had a college degree disappeared, however, among students who took rigorous
(2.6 vs. 2.8) and were more likely to have taken at least high school courses. In this case, first-generation students
one remedial course during their first year of postsecondary were as likely as students whose parents had a college degree
education (21 percent vs. 10 percent). This difference to be continuously enrolled or to have attained a degree as of
persisted even after controlling for the rigor of students’ June 1998 (87 percent vs. 86 percent) (figure B).
high school coursework and college entrance examination
scores. Among students who substantially exceeded the In this study, students who remained at the initial 4-year
core New Basics in high school, first-generation students institution or made a lateral transfer to a new 4-year
were more likely to have taken at least one remedial course institution were considered to have stayed on the persis-
during their first year of postsecondary education than tence track to a bachelor’s degree. Overall, first-generation
students whose parents had a college degree (15 percent students were less likely than students whose parents had
vs. 6 percent). Moreover, among students whose college completed a 4-year degree to stay on the persistence track
entrance examination scores were in the lowest quartile, (58 percent vs. 77 percent). Not only were first-generation
38 percent of first-generation students had taken at least students more likely than their peers whose parents finished
one remedial course during their first year, compared with college to leave the persistence track through a stopout or
29 percent of students whose parents had a college degree. downward transfer (22 percent vs. 14 percent), they also
were more than twice as likely to leave their first institution
However, among students who took rigorous high school without returning (21 percent vs. 9 percent). Moreover, even
courses or scored in the top quartile on their college among students who took rigorous coursework in high
entrance examinations, first-generation students had first- school, first-generation students were almost twice as likely
year college GPAs and remedial coursetaking patterns as students whose parents had completed college to leave the
that were not significantly different from their non-first- persistence track through a stopout or downward transfer
generation peers. For example, among students who took (14 percent vs. 8 percent).
rigorous coursework in high school, 95 percent of first-
generation students reported taking no remedial courses Though the negative relationship between first-generation
during their first year, compared to 96 percent of students status and persistence was strong and consistent, the picture
whose parents had some college and 97 percent of students was more positive when looking at those who left post-
whose parents had a bachelor’s degree. In addition, first- secondary education without returning, while controlling for
generation students’ average first-year GPA was 3.0, which the rigor of students’ secondary school curriculum and their
was no different from the average GPA (3.1) of their non- scores on college entrance examinations. Among those
first-generation counterparts with similar academic students who took a rigorous high school curriculum, first-
preparation. generation students and students whose parents completed
college had similar rates of postsecondary departure without
Postsecondary persistence and attainment return (5 percent and 3 percent). And though first-generation
With respect to postsecondary persistence and attainment, students who did not exceed the core New Basics in high
four outcomes were examined: the number of enrollment school were less likely to stay on the persistence track to a
spells,4 retention at the initial 4-year institution, persistence bachelor’s degree compared to their counterparts (55 percent
track to a bachelor’s degree, and attainment or last academic vs. 69 percent), the likelihood of staying on the persistence
year of enrollment through 1998. track for students who took rigorous coursework did not

4 5
An “enrollment spell” is defined as a period of enrollment without a break of more An enrollment spell may end either with a stopout or leaving without return. A
than 4 months. The number of enrollment spells counts the periods of continuous “stopout” is defined as a break in enrollment of more than 4 months and a return to
enrollment (at any institution), each separated by more than 4 months of non- postsecondary education. Leaving without return is no enrollment for a period of
enrollment, through June 1998. more than 4 months and no return to postsecondary education as of spring 1998.

E D U C AT I O N S T AT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , FA L L 2 0 0 1 75
Postsecondary Education

Figure B.—Percentage of 1995–96 beginning postsecondary students who were still enrolled (or had attained
bachelor’s degree) at initial institution according to academic rigor of secondary school curriculum, by
first-generation status: June 1998

Core New Basics 55 First-generation student

or below1 67
Parent had bachelor's degree

Beyond 65
New Basics I2 68

Beyond 65
New Basics II3 72

87
Rigorous4
86

0 25 50 75 100
Percent

1
Core New Basics curriculum includes 4 years of English, 3 years of mathematics, and 3 years of science and social studies.
2
Beyond New Basics I includes core New Basics and at least two of three science courses (biology, chemistry, or physics), and
algebra I and geometry, plus 1 year of foreign language.
3
Beyond New Basics II includes core New Basics, advanced science (biology, chemistry, and physics), and advanced math (including
algebra I, geometry, algebra II), plus 2 years of foreign language.
4
Rigorous includes core New Basics, advanced science (biology, chemistry, and physics), and 4 years of math (including algebra I,
geometry, algebra II, precalculus), plus 3 years of foreign language and one honors/Advanced Placement (AP) course or AP test score.
NOTE: Includes public and private not-for-profit 4-year institutions. Students in private for-profit 4-year institutions are excluded from
this analysis because the sample size was too small (less than 1 percent).
SOURCE: U.S. Department of Education, National Center for Education Statistics, 1996 Beginning Postsecondary Students Longitudi-
nal Study, “First Follow-up” (BPS:96/98).

differ meaningfully for first-generation students and students rigorous coursework in high school accounted for more than
whose parents had a bachelor’s degree (81 and 89 percent, 80 percent of those students who either stayed on the persis-
respectively) (figure C). tence track to a bachelor’s degree or were retained at their
initial institution. At the same time, parents’ levels of educa-
Finally, this study examined overall rates of persistence and tion were found to be associated with rates of students’
attainment in spring 1998, 3 years after initial enrollment. retention and persistence in college, even when controlling for
Students whose parents had a bachelor’s degree were more measures of academic preparedness (such as rigor of second-
likely than their first-generation peers to have attained a ary curriculum and college entrance examination scores).
degree or to still be enrolled 3 years after entering a 4-year
institution (88 percent vs. 73 percent). This difference was These findings hold true even when other related variables
particularly evident for first-generation students who did are held constant. That is, even after controlling for vari-
not take a rigorous curriculum in high school: they were ables such as academic preparation and postsecondary
much less likely than students whose parents completed achievement, parents’ education continued to be a signifi-
college to be enrolled 3 years after entering a 4-year institu- cant factor in determining whether students were enrolled
tion (65 percent vs. 85 percent). at their initial institution 3 years after entering or stayed on
the persistence track. Students whose parents attained a
Conclusion bachelor’s degree were more likely than first-generation
The findings from this analysis indicate that students who students to remain enrolled at their initial 4-year institu-
were well prepared for postsecondary education were very tion. Likewise, after controlling for related variables,
likely to persist in 4-year institutions. Students who took students whose parents attained a bachelor’s degree or

76 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
 Bridging the Gap: Academic Preparation and Postsecondary Success of First-Generation Students

Figure C.—Percentage of 1995–96 beginning postsecondary students who stayed on persistence track to bachelor’s
degree according to academic rigor of secondary school curriculum, by first-generation status: June 1998

Core New Basics 55 First-generation student

or below1 69
Parent had bachelor's degree

Beyond 66
New Basics I2 75

Beyond 64
New Basics II3 76

81
Rigorous4
89

0 25 50 75 100
Percent

1
Core New Basics curriculum includes 4 years of English, 3 years of mathematics, and 3 years of science and social studies.
2
Beyond New Basics I includes core New Basics and at least two of three science courses (biology, chemistry, or physics), and
algebra I and geometry, plus 1 year of foreign language.
3
Beyond New Basics II includes core New Basics, advanced science (biology, chemistry, and physics), and advanced math (including
algebra I, geometry, algebra II), plus 2 years of foreign language.
4
Rigorous includes core New Basics, advanced science (biology, chemistry, and physics), and 4 years of math (including algebra I,
geometry, algebra II, precalculus), plus 3 years of foreign language and one honors/Advanced Placement (AP) course or AP test
score.
NOTE: Students who stay on the persistence track either remain at the initial 4-year institution in which they enrolled or make
a lateral transfer to a new 4-year institution with no break in enrollment. Includes public and private not-for-profit 4-year
institutions. Students in private for-profit 4-year institutions are excluded from this analysis because the sample size was too
small (less than 1 percent).
SOURCE: U.S. Department of Education, National Center for Education Statistics, 1996 Beginning Postsecondary Students
Longitudinal Study, “First Follow-up” (BPS:96/98).

higher were more likely to stay on the persistence track to a Nuñez, A.-M., and Cuccaro-Alamin, S. (1998). First-Generation
bachelor’s degree than first-generation students. Students: Undergraduates Whose Parents Never Enrolled in
Postsecondary Education (NCES 98–082). U.S. Department of
Education, National Center for Education Statistics. Washing-
At the same time, after holding all other variables constant, ton, DC: U.S. Government Printing Office.
students who took rigorous coursework in high school
significantly increased their chances of staying on the
persistence track to a bachelor’s degree. Taken together,
these results suggest that, while first-generation status is an Data source: The NCES 1996 Beginning Postsecondary Students
important predictor of success in postsecondary education, Longitudinal Study “First Follow-up” (BPS:96/98).

rigorous preparation in high school substantially narrows For technical information, see the complete report:
Warburton, E.C., Bugarin, R., and Nuñez, A.-M. (2001). Bridging the Gap:
the gap in postsecondary outcomes between first-generation Academic Preparation and Postsecondary Success of First-Generation
students and their peers whose parents graduated from college. Students (NCES 2001–153).
Author affiliations: E.C. Warburton, R. Bugarin, and A.-M. Nuñez, MPR
References Associates, Inc.

Horn, L., and Nuñez, A.-M. (2000). Mapping the Road to College: For questions about content, contact Aurora D’Amico
(aurora.d’amico@ed.gov).
First-Generation Students’ Math Track, Planning Strategies, and
To obtain the complete report (NCES 2001–153), call the toll-
Context of Support (NCES 2000–153). U.S. Department of
free ED Pubs number (877–433–7827) or visit the NCES Web Site
Education, National Center for Education Statistics. Washing- (http://nces.ed.gov).
ton, DC: U.S. Government Printing Office.

E D U C AT I O N S T AT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , FA L L 2 0 0 1 77
Postsecondary Education

Middle Income Undergraduates

Middle Income Undergraduates: Where They Enroll and How They Pay for
Their Education
——————————————————————————————————
Jennifer B. Presley and Suzanne B. Clery
This article was originally published as the Executive Summary of the Statistical Analysis Report of the same name. The sample survey data are from the
NCES National Postsecondary Student Aid Study (NPSAS).

The need for financial aid or other financial help, such as degrees as their highest level of education completed), one-
tuition tax credits, to assist students from middle income quarter had parents with bachelor’s degrees, 16 percent had
families to attend the college of their choice has received parents with master’s degrees, and 5 percent had parents
increased attention recently. This report provides a profile of with doctoral or first-professional degrees. Thirty-seven
middle income undergraduates in comparison to their lower percent of middle income FTFY dependent undergraduates
income and higher income counterparts and examines lived on campus, while 35 percent lived with their parents
where middle income students enroll by price of attendance or relatives. Forty-five percent were first-year/freshmen.
and how they and their families pay for college, including
the role of financial aid. Price of Attendance
Price of attendance is the student budget for FTFY depen-
The source of data for this analysis was the 1995–96 dent undergraduates for 1995–96, including tuition and fees
National Postsecondary Student Aid Study (NPSAS:96). The and total nontuition costs. Of the middle income FTFY
report is limited to full-time, full-year (FTFY) dependent dependent undergraduates, 8 percent were enrolled at the
undergraduates who were enrolled during 1995–96. lowest price-of-attendance level, 21 percent at the moderate
price-of-attendance level, 49 percent at the upper price-of-
In this report, middle income undergraduates are defined as attendance level, and 23 percent at the highest price-of-
those with family incomes between $35,000 and $69,999 in attendance level (figure A).
1994, the year of the income data included in the NPSAS:96
database. Undergraduates from families with incomes below The percentages of FTFY dependent undergraduates from
$35,000 are defined as lower income, and those from lower income and middle income families enrolling at each
families with incomes of $70,000 or more are defined as price-of-attendance level were about the same, but a smaller
higher income. Thirty-seven percent of FTFY dependent percentage of each of these two income groups (20 and 23
undergraduates in the sample were middle income accord- percent) was enrolled at the highest price-of-attendance level
ing to this definition, 35 percent were lower income, and than of the higher income group (34 percent). Middle
28 percent were higher income. income FTFY dependent undergraduates with mid-range
combined SAT scores of 825–1,199 were less likely to be
Because the financing strategies that students use to pay for enrolled at the highest price-of-attendance level than were
college vary by price of attendance, results are provided for those with similar SAT scores in the higher income group.
four price-of-attendance levels: lowest price of attendance Again, the enrollment of middle income and lower income
(less than $6,000), moderate price of attendance ($6,000 to FTFY dependent undergraduates by price-of-attendance
$8,499), upper price of attendance ($8,500 to $15,999), level was about the same within the same SAT range. In all
and highest price of attendance ($16,000 or more). The three income groups, the percentages of those with SAT
price of attendance is the FTFY dependent undergraduate scores of 1,200 or more that were enrolled at the highest
budget, which represents the total expected expenses for price-of-attendance level were not statistically different,
an FTFY dependent undergraduate to attend college in standing at 54 percent overall.
1995–96, including tuition, fees, and nontuition expenses.
Multivariate analysis showed that even after controlling for
Profile of Middle Income Full-Time, Full-Year student background and family factors likely to affect
Dependent Undergraduates enrollment at the highest price-of-attendance level, the
Middle income FTFY dependent undergraduates in 1995–96 percentage enrolled at this level was still lower for middle
were 53 percent female; 81 percent White, non-Hispanic; income FTFY dependent undergraduates (23 percent) than
and nearly all from families with at least three family their higher income counterparts (30 percent). Factors
members. About one-third were first-generation college associated with enrollment at the highest price-of-attendance
students (i.e., their parents had no more than high school level, in addition to family income, included being female,

78 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
 Middle Income Undergraduates: Where They Enroll and How They Pay for Their Education

Figure A.—Percentage distribution of full-time, full-year dependent undergraduates from middle

income families, by price-of-attendance level: 1995–96

Lowest level (8%)

Highest level (23%)

Moderate level (21%)

Upper level (49%)

NOTE: Dependent undergraduates are limited to those who attended only one institution and were U.S.
citizens or permanent residents. Detail may not add to 100 percent because of rounding.
SOURCE: U.S. Department of Education, National Center for Education Statistics, 1995–96 National
Postsecondary Student Aid Study (NPSAS:96), Undergraduate Data Analysis System.

having parents whose highest level of education was a income category (table A). Middle income FTFY dependent
doctoral or first-professional degree, and having combined undergraduates with financial need had an average of
SAT scores of 1,200 or more. $7,785 of financial need.

Financial Need and Financial Aid Sixty-five percent of middle income FTFY dependent
Financial need is defined as student budget minus expected undergraduates received financial aid, 55 percent having
family contribution (EFC). EFC is the amount that the financial need and 10 percent not having financial need.
family and the student are expected to contribute toward Overall, middle income FTFY dependent undergraduates
the price of attendance, based on formulas for calculating had 31 percent of their price of attendance covered by
financial aid awards.1 Virtually all families in the middle financial aid, and more than one-half (58 percent) had
income group are expected to pay part of the price of unmet need after financial aid and EFC (table A).
attendance from their own resources. The amount that
remains after subtracting EFC is the student’s financial One-third of middle income FTFY dependent undergradu-
need. It is possible that even after all aid is awarded, some ates with unmet need had not applied for financial aid.
students will have unmet need. Reasons given included a belief that family income was too
high (32 percent) or that the family could afford to pay
In 1995–96, four out of five middle income FTFY depen- (48 percent). But for those who did apply for aid, but did
dent undergraduates (79 percent) had some financial need, not have their financial need met, it could mean that they
compared with almost all of those in the lower income and their families were making greater financial effort to
category (99 percent) and one-third of those in the higher attend their institution of choice.

1
Expected family contribution (EFC) is based on dependency, family size, income, Net price, or the difference between price of attendance
assets, and number in college. In order to calculate how much federal financial aid
students can receive, federal regulations provide a methodology that assesses how and total financial aid received, is the amount of out-of-
much the family and the student should contribute toward the price of attendance. pocket expenses that students and their families must come
See Chapter 7 of the 1999–2000 Student Financial Aid Handbook, U.S. Department of
Education, for more information (available at http://ifap.ed.gov/IFAPWebApp/ up with to attend the colleges in which they enroll. By
currentSFAHandbooksPag.jsp).

E D U C AT I O N S T AT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , FA L L 2 0 0 1 79
Postsecondary Education

Table A.—Financial status for all full-time, full-year dependent undergraduates, by family income: 1995–96

Expected Unmet
Price of family Amount of need5
attendance contribution Ratio of Percent financial Net Percent for those
at institution (EFC) at any EFC with need3 for price4 with with
where postsecondary to price of financial those with Grant Total for unmet unmet
enrolled institution1,2 attendance2 need need aid2 loan2 all2 need5 need

Total $12,663 $8,697 77 72.6% $8,921 $2,230 $2,014 $8,113 57.6% $4,797

Family income
Lower income 11,715 1,617 16 98.5 10,372 3,560 2,158 5,628 86.9 4,914
Middle income 12,284 6,865 68 78.6 7,785 1,873 2,229 7,867 58.3 4,375
Higher income 14,316 19,729 162 33.1 7,198 1,078 1,557 11,462 20.9 5,745

1
Expected family contribution (EFC) may exceed actual student budget; therefore, the average EFC reported in this table cannot be used to calculate financial need and unmet need
reported in this table.
2
Includes zero values.
3
Student budget minus EFC.
4
Student budget minus all aid.
5
Student budget minus EFC minus aid.
NOTE: Dependent undergraduates are limited to those who attended only one institution and were U.S. citizens or permanent residents.
SOURCE: U.S. Department of Education, National Center for Education Statistics, 1995–96 National Postsecondary Student Aid Study (NPSAS:96), Undergraduate Data Analysis System.

examining student earnings and the percentage of students amounts of parental contributions toward the net price of
with parents contributing to their college expenses, we can attendance are not known, but the data suggest that middle
gain an additional perspective on how students and their income FTFY dependent undergraduates who were enrolled
families cover college expenses. at the lowest and moderate price-of-attendance levels could
meet or exceed the net prices of attendance with their own
The average net price for middle income FTFY dependent earnings and a modest contribution from their parents.
undergraduates in 1995–96 was $7,867 (table A). Net price Perhaps these students attended at the lowest and moderate
increased as price-of-attendance level increased, but aver- price-of-attendance levels in order to reduce (but not elimi-
age student earnings were lowest at the highest price-of- nate) the amount of EFC that their parents needed to
attendance level. contribute, or to increase their available discretionary
income, or both.
At the lowest price-of-attendance level, student earnings for
middle income FTFY dependent undergraduates averaged With a sizable gap between average net price ($7,632) and
$4,478, the average net price of attendance was $4,581,2 average student earnings ($3,419) for the 49 percent of
and 84 percent of these students reported that their parents middle income FTFY dependent undergraduates enrolled at
contributed to their college expenses.3 At the moderate the upper price-of-attendance level, 91 percent reported that
price-of-attendance level, average student earnings for their parents contributed toward expenses. EFC for middle
middle income FTFY dependent undergraduates of $3,737 income FTFY dependent undergraduates enrolled at this
covered a considerable amount of the $5,668 net price of price-of-attendance level was $6,913. Thus, it appears that
attendance, while 88 percent of these students reported that FTFY undergraduates attending at the upper price-of-
their parents contributed to their college expenses. The attendance level can, on average, meet the net price of
attendance through student earnings and parental contribu-
2
Students are not expected to contribute all of their earnings for EFC or price of tions below EFC.
attendance. The actual amount contributed from earnings is not known, and some
may contribute more or less than calculated for their EFC.
3
For the 23 percent of middle income FTFY dependent
For students and their families in the middle income group, virtually all are expected
to make some contribution from their own resources toward paying for college. undergraduates enrolled at the highest price-of-attendance

80 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
 Middle Income Undergraduates: Where They Enroll and How They Pay for Their Education

level, the gap between net price and student earnings was the highest price-of-attendance level, about two-thirds of
$8,919, which is more than the average EFC of $7,024 for those with need in the lower and middle income groups
these undergraduates’ families. In addition, student earnings received institutional grants, compared with 46 percent of
for middle income FTFY dependent undergraduates at the those in the higher income group. In addition, the average
highest price-of-attendance level were lower than earnings for award was smaller for those in the higher income group.
those at other price-of-attendance levels. At the highest Institutional grants constituted an important component of
price-of-attendance level, 96 percent of parents were reported financial aid packages that assisted in providing access at
to contribute to the price of attendance. Again, we do not the highest price-of-attendance level, especially for lower and
know the actual amounts contributed by parents, but these middle income undergraduates.
data suggest that parents may have contributed amounts
that exceeded their EFC by several thousand dollars. This Overall, middle income FTFY dependent undergraduates
gap between net price, student earnings, and EFC may with financial need were about as likely to borrow as their
explain why a smaller percentage of middle income FTFY lower income counterparts, and the amounts they borrowed
dependent undergraduates with mid-range SAT scores were were about the same, but they were more likely to borrow
enrolled at the highest price-of-attendance level in compari- than were their higher income counterparts. The percentage
son to those in the higher income group, with more of middle income FTFY dependent undergraduates with
financial resources to meet out-of-pocket expenses. Perhaps need who borrowed was larger at each price-of-attendance
middle income parents, on average, made additional level. For those with no financial need, the percentage
financial effort to pay for their dependents who were among borrowing was larger than the percentage borrowing from
the best scholastically, but did not make additional financial the higher income group.
effort for those who were mid-range students scholastically.
Another reason may be that institutions in the highest price- Summary
of-attendance level may not practice need-blind admissions, Lower income and middle income FTFY dependent under-
but balance applicants’ academic strengths with their graduates have similar price-of-attendance enrollment
financial aid needs, resulting in fewer enrollees with mid- patterns. Those from the lower income and middle income
range scores who have higher financial need. groups use more financial aid than do those in the higher
income group. Lower income and middle income FTFY
Sources of Financial Aid dependent students do not enroll at similar price-of-
In 1995–96, one-half of middle income FTFY dependent attendance levels as their counterparts in the higher income
undergraduates with financial need received loans, account- group, with the exception of those undergraduates with SAT
ing for 42 percent of their aid, while 52 percent received scores of 1,200 or more.
grants, accounting for 44 percent of their aid, with work-
study and other types of aid providing the remaining aid.

Institutional grants constituted the most common source of Data source: The NCES 1995–96 National Postsecondary Student Aid
grant aid for middle income FTFY dependent undergradu- Study (NPSAS:96).
ates with financial need, as well as those in the higher For technical information, see the complete report:
income group. The percentages of FTFY dependent under- Presley, J.B., and Clery, S.B. (2001). Middle Income Undergraduates: Where
They Enroll and How They Pay for Their Education (NCES 2001–155).
graduates with need who received institutional grants, as
Author affiliations: J.B. Presley and S.B. Clery, JBL Associates, Inc.
well as the average institutional awards that they received,
For questions about content, contact Aurora D’Amico
were about the same for each family-income level. In each (aurora.d’amico@ed.gov).
income category, larger percentages of FTFY dependent To obtain the complete report (NCES 2001–155), call the toll-
undergraduates with need received institutional grants if free ED Pubs number (877–433–7827), visit the NCES Web Site
(http://nces.ed.gov), or contact GPO (202–512–1800).
they were enrolled at the highest price-of-attendance level. At

E D U C AT I O N S T AT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , FA L L 2 0 0 1 81
Postsecondary Education

Student Aid
National Postsecondary Student Aid Study: Student Financial Aid Estimates
for 1999–2000
——————————————————————————————————
Andrew G. Malizio
This article was originally published as the Introduction and Highlights of the E.D. Tabs report of the same name. The sample survey data are from the
National Postsecondary Student Aid Study (NPSAS).

Introduction grants, loans, and work-study; and the remaining

The National Postsecondary Student Aid Study (NPSAS) is 13 percent received other combinations of aid.
a comprehensive survey that examines how students and
their families pay for postsecondary education. The study Federal aid to undergraduates
includes nationally representative samples of undergradu- ■ Overall, about two out of five undergraduates
ate, graduate, and first-professional students; students (39 percent) received some type of federal aid,
attending less-than-2-year, 2-year, 4-year, and averaging about $5,230 (tables A and B).
doctorate-granting institutions; and students who ■ Percentages of undergraduates receiving federal aid
receive financial aid and those who do not receive aid. varied depending on family income and type of
institution. Among dependent students, percentages
This report has been prepared to provide some key esti- receiving federal aid ranged from 70 percent of
mates as policymakers, researchers, and analysts begin undergraduates from families with incomes of less
working on research for the next reauthorization of the than $20,000 to about 25 percent of undergraduates
Higher Education Act. It is based on information from from families with incomes of $100,000 or more.
about 50,000 undergraduates, 11,000 graduate students, Among independent students, 66 percent of those
and 1,000 first-professional students enrolled at approxi- with incomes less than $10,000 received some federal
mately 1,000 postsecondary institutions during the financial aid (table A).
1999–2000 academic year. The sample represents about ■ Percentages of undergraduates receiving federal aid
16.5 million undergraduates, 2.3 million graduate students, ranged from 21 percent at public 2-year institutions
and 330,000 first-professional students enrolled at any time to 80 percent at private for-profit institutions.
between July 1, 1999, and June 30, 2000. Considerably
more detail on how students finance their postsecondary Institutional aid to undergraduates
education and student background characteristics will be ■ Forty-seven percent of undergraduates at private not-
published in subsequent reports. for-profit 4-year institutions received some institu-
tional aid, averaging about $6,760 (tables A and B).
Estimates in this report focus primarily on percentages of
students receiving specified types of financial aid, and Title IV aid to undergraduates
average amounts of specific types of financial aid received.
■ About one of every eight dependent undergraduates
Unless otherwise noted, all average amounts of financial aid
(13 percent) came from families with incomes less
described here and presented in the tables reflect the
than $20,000. Among Title IV aid recipients who
weighted means and are based only on the recipients of the
were dependent on their parents for financial sup-
specified types of aid. Highlights are presented separately for
port, 21 percent had family incomes less than
undergraduates and graduate and first-professional students.
$20,000. Three of every eight independent under-
graduates (38 percent) who received Title IV aid had
Undergraduates
family incomes less than $10,000.
■ Among the 16.5 million undergraduates (including
■ Among undergraduates enrolled full time for the full
full-time and part-time students) enrolled during
year at one institution, about 30 percent received a
1999–2000, 55 percent (about 9.2 million) received
federal Pell grant, averaging $2,314.
some type of financial aid, averaging $6,206.
■ Nearly one of every four undergraduates (23 percent)
■ Among aided undergraduates, 40 percent received
received a federally subsidized Stafford loan, averag-
grants only; 13 percent received loans only; 26 per-
ing about $3,214.
cent received grants and loans; 8 percent received

82 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
 National Postsecondary Student Aid Study: Student Financial Aid Estimates for 1999–2000

■ Among the Title IV loan recipients enrolled in private $12,850. Average amounts borrowed varied consider-
not-for-profit doctorate-granting institutions, the ably by degree program. About 73 percent of first-
average loan amount was $5,161. At private not-for- professional students took out Stafford loans,
profit non-doctorate-granting 4-year institutions, the compared to 26 percent of master’s degree students
average received was $5,095. and 21 percent of doctoral students.
■ Among the Title IV loan recipients enrolled in public
Institutional aid to graduate and first-professional
doctorate-granting institutions, the average student
students
loan amount was $4,743. At public non-doctorate-
granting 4-year institutions, the average received was ■ Overall, 27 percent of graduate and first-professional
$4,225. students received some institutional aid, averaging
about $9,840, but this varied considerably depending
■ Fifty-four percent of students enrolled in private
on the type of program. For example, 22 percent of
for-profit less-than-2-year institutions received a Title
students in master’s degree programs compared to
IV loan; the average received was $4,879.
58 percent of students in doctoral degree programs
■ Among full-time, full-year undergraduates, about received some institutional aid.
40 percent of those at public 2-year institutions,
56 percent of those at public 4-year institutions, and Employer aid to graduate and first-professional
67 percent of those at private not-for-profit 4-year students
institutions received some Title IV financial aid. ■ About one in four master’s degree students received
employer aid, averaging about $3,840. About 11
Graduate and First-Professional Students percent of doctoral students and 5 percent of first-
■ Among the 2.7 million graduate and first-professional professional students received employer aid.
students (including full-time and part-time students)
enrolled during 1999–2000, about 6 of every 10
students received some financial aid, averaging
$13,255. Eight of every 10 full-time, full-year Data source: The NCES 1999–2000 National Postsecondary Student
Aid Study (NPSAS:2000).
students (82 percent) received some type of financial
For technical information, see the complete report:
aid, averaging almost $19,600. First-professional
Malizio, A.G. (2001). National Postsecondary Student Aid Study: Student
students received an average of about $21,500 Financial Aid Estimates for 1999–2000 (NCES 2001–209).
(tables C and D). Author affiliation: A.G. Malizio, NCES.
For questions about content, contact Aurora D’Amico
Stafford loans to graduate and first-professional (aurora.d’amico@ed.gov) or Andrew G. Malizio
students (andrew.malizio@ed.gov).
To obtain the complete report (NCES 2001–209), visit the NCES Web
■ Overall, 29 percent of graduate and first-professional Site (http://nces.ed.gov).
students received Stafford loans, averaging nearly

E D U C AT I O N S T AT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , FA L L 2 0 0 1 83
Postsecondary Education

Table A.—Percentage of undergraduates receiving selected types of federal, institutional, or state aid: 1999–2000

Federal aid Institutional aid State aid

Any aid Grants Loans Work-study PLUS Any aid Grants Any aid Grants

Total 39.1 23.1 27.9 4.1 2.9 17.4 16.7 14.1 13.6

Dependency and 1998 income level

Dependent 43.7 20.1 33.8 6.6 5.8 25.3 24.1 16.9 16.1
Less than $20,000 70.0 65.9 35.2 10.5 2.8 26.7 25.6 28.9 28.5
$20,000–39,999 56.3 43.6 37.9 9.9 4.4 28.1 26.9 26.1 25.6
$40,000–59,999 40.8 8.9 36.9 7.7 6.0 25.9 24.9 18.0 17.2
$60,000–79,999 36.9 1.4 35.4 5.3 8.2 24.2 23.2 11.1 10.1
$80,000–99,999 32.5 0.5 31.1 2.8 7.1 23.9 22.6 7.6 6.6
$100,000 or more 24.5 0.4 23.3 1.8 6.5 21.6 20.4 6.1 5.2
Independent 34.6 26.0 22.3 1.7 (†) 9.9 9.4 11.5 11.2
Less than $10,000 65.7 61.5 39.0 5.0 (†) 18.9 17.9 21.8 21.3
$10,000–19,999 51.5 38.6 32.6 1.9 (†) 13.4 12.8 17.4 17.2
$20,000–29,999 34.9 24.8 21.4 1.1 (†) 8.1 7.8 11.5 11.1
$30,000–49,999 20.3 11.6 14.6 0.7 (†) 6.4 6.1 6.8 6.6
$50,000 or more 7.8 0.2 7.7 0.1 (†) 3.9 3.8 2.5 2.3

Tuition and fees

$1–999 15.2 13.0 4.3 0.7 0.1 7.3 7.2 6.1 6.0
$1,000–1,999 39.6 28.0 22.5 2.5 1.0 11.1 10.6 16.0 15.4
$2,000–2,999 51.0 31.5 38.0 4.0 1.7 16.3 14.9 19.6 18.5
$3,000–3,999 54.7 30.3 43.4 5.4 4.1 20.0 18.6 22.9 22.3
$4,000–4,999 58.1 30.2 48.9 6.3 5.1 20.8 19.9 22.6 21.7
$5,000–7,499 64.6 33.1 56.4 5.5 5.7 25.4 23.7 19.7 19.3
$7,500 or more 65.4 26.5 60.7 14.9 10.7 49.7 47.9 21.1 20.1

Institution type
Public
Less-than-2-year 25.3 23.4 6.1 0.7 0.3 4.5 4.5 6.4 5.6
2-year 20.7 17.2 7.0 1.3 0.1 7.7 7.6 10.2 9.8
4-year 46.4 24.4 38.5 4.6 3.5 18.6 17.1 17.0 16.3
Non-doctorate-granting 47.6 27.8 36.7 5.1 2.2 14.8 13.8 18.4 17.8
Doctorate-granting 45.7 22.4 39.6 4.3 4.3 20.8 19.0 16.1 15.4
Private not-for-profit
Less-than-4-year 57.7 40.9 36.7 6.8 7.0 32.8 32.4 19.6 19.4
4-year 56.6 24.7 48.2 13.4 7.1 46.7 45.5 22.3 21.5
Non-doctorate-granting 58.1 27.2 48.2 11.7 6.6 44.0 43.1 24.5 23.7
Doctorate-granting 54.3 20.9 48.3 16.0 7.9 50.8 49.2 18.9 18.1
Private for-profit
Less-than-2-year 80.1 64.0 54.0 0.6 3.7 6.1 3.8 4.1 4.1
2-year or more 80.4 45.6 72.4 1.1 8.6 8.0 7.8 12.9 12.2

Attendance pattern
Full-time, full-year 56.7 30.5 43.9 8.8 5.5 31.0 29.7 23.0 22.1
Full-time, part-year 50.1 34.5 32.9 2.8 2.8 13.0 12.3 14.5 14.1
Part-time, full-year 30.0 19.1 19.2 1.8 0.9 11.3 10.7 10.7 10.3
Part-time, part-year 14.5 9.9 8.0 0.3 0.5 5.1 4.9 4.4 4.3

†Not applicable.
NOTE: Students can receive more than one type of aid. Estimates by tuition and fees categories, institution type, and attendance pattern exclude students who attended
multiple institutions.
SOURCE: U.S. Department of Education, National Center for Education Statistics, 1999–2000 National Postsecondary Student Aid Study (NPSAS:2000). (Originally published as
table 3 on p. 8 of the complete report from which this article is excerpted.) Table revised October 2001.

84 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
 National Postsecondary Student Aid Study: Student Financial Aid Estimates for 1999–2000

Table B.—Average amounts of selected types of federal, institutional, or state aid received by undergraduates: 1999–2000

Federal aid Institutional aid State aid

Any aid Grants Loans Work-study PLUS Any aid Grants Any aid Grants

Total $5,226 $2,063 $4,643 $1,534 $7,080 $3,776 $3,722 $1,797 $1,681

Dependency and 1998 income level

Dependent 5,237 2,133 3,999 1,470 7,082 4,610 4,556 2,106 1,957
Less than $20,000 4,882 2,591 4,015 1,410 5,342 3,186 3,138 1,960 1,911
$20,000–39,999 4,925 1,920 4,095 1,480 5,367 4,078 4,027 2,180 2,078
$40,000–59,999 5,014 1,188 3,950 1,505 6,183 4,936 4,892 2,090 1,941
$60,000–79,999 5,464 1,400 3,889 1,464 6,678 5,444 5,402 2,138 1,889
$80,000–99,999 5,828 (#) 4,002 1,522 8,488 4,889 4,859 1,982 1,674
$100,000 or more 6,582 (#) 4,058 1,449 9,789 5,211 5,113 2,373 1,867
Independent 5,214 2,012 5,584 1,774 (†) 1,717 1,662 1,361 1,298
Less than $10,000 5,509 2,279 5,443 1,662 (†) 1,920 1,868 1,537 1,475
$10,000–19,999 5,045 1,993 5,482 1,655 (†) 1,614 1,548 1,370 1,299
$20,000–29,999 4,674 1,755 5,501 2,106 (†) 1,623 1,559 1,150 1,103
$30,000–49,999 4,934 1,173 5,774 2,438 (†) 1,538 1,507 1,194 1,122
$50,000 or more 6,297 (#) 6,362 (#) (†) 1,579 1,531 1,085 1,017

Tuition and fees

$1–999 2,242 1,468 3,073 1,677 (#) 520 498 747 737
$1,000–1,999 3,641 1,928 3,647 1,575 4,272 1,278 1,247 1,183 1,106
$2,000–2,999 4,956 2,246 4,389 1,640 4,810 1,955 1,897 1,682 1,554
$3,000–3,999 5,587 2,366 4,654 1,419 5,820 2,440 2,345 1,826 1,725
$4,000–4,999 5,956 2,294 4,861 1,578 5,674 2,491 2,473 2,084 1,975
$5,000–7,499 6,266 2,214 5,173 1,395 5,645 3,091 3,078 2,119 2,012
$7,500 or more 7,726 2,556 5,294 1,543 8,660 7,186 7,086 3,109 2,859

Institution type
Public
Less-than-2-year 2,824 1,760 4,613 (#) (#) 667 667 972 721
2-year 2,609 1,741 3,052 1,589 (#) 607 594 988 916
4-year 5,574 2,197 4,568 1,636 6,153 2,525 2,503 1,873 1,773
Non-doctorate-granting 4,912 2,131 4,226 1,600 5,119 1,837 1,784 1,651 1,538
Doctorate-granting 5,972 2,245 4,751 1,660 6,461 2,807 2,804 2,020 1,931
Private not-for-profit
Less-than-4-year 4,780 2,091 3,943 963 5,625 2,690 2,615 2,148 2,125
4-year 6,858 2,393 5,132 1,464 8,753 6,758 6,606 2,662 2,460
Non-doctorate-granting 6,413 2,278 5,096 1,270 7,608 5,865 5,724 2,455 2,292
Doctorate-granting 7,596 2,628 5,187 1,683 10,245 7,957 7,804 3,079 2,800
Private for-profit
Less-than-2-year 5,264 2,176 4,879 (#) 4,879 1,252 740 538 526
2-year or more 6,974 2,154 5,558 (#) 6,581 1,442 1,306 2,697 2,427

Attendance pattern
Full-time, full-year 6,090 2,541 4,838 1,560 7,539 4,727 4,665 2,143 1,997
Full-time, part-year 4,095 1,715 3,876 1,006 5,596 2,326 2,295 1,237 1,192
Part-time, full-year 4,656 1,765 5,020 1,912 6,545 2,040 2,008 1,392 1,274
Part-time, part-year 3,208 1,103 3,994 1,423 5,594 851 797 889 853

#Too small to report.

†Not applicable.
NOTE: Average amounts shown above are for recipients of the specified aid. Students can receive more than one type of aid. Estimates by tuition and fees categories,
institution type, and attendance pattern exclude students who attended multiple institutions.
SOURCE: U.S. Department of Education, National Center for Education Statistics, 1999–2000 National Postsecondary Student Aid Study (NPSAS:2000). (Originally published as
table 4 on p. 9 of the complete report from which this article is excerpted.) Table revised October 2001.

E D U C AT I O N S T AT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , FA L L 2 0 0 1 85
Postsecondary Education

Table C.—Percentage of graduate and first-professional students receiving selected types of financial aid: 1999–2000

Stafford loans Institutional aid

Any Unsub- Perkins Institutional Assistant- Employer
aid Total Subsidized sidized loans aid, any ships aid

Total 59.7 29.0 26.9 22.6 3.7 27.2 13.6 19.8

Type of graduate program

Master’s and doctoral programs 60.6 25.1 22.7 18.8 2.2 28.8 16.5 22.1
Master’s degree 57.9 26.0 23.4 19.5 2.1 22.2 10.5 24.7
Doctorate 72.4 21.3 19.4 15.4 3.0 58.3 42.7 10.5
Other graduate program 36.3 12.7 10.9 8.9 0.9 12.4 2.2 21.0
Postbaccalaureate certificate 43.8 22.0 19.1 15.9 1.2 13.2 3.7 17.0
Other, including non-degree 30.5 5.6 4.6 3.6 0.7 11.8 1.0 24.2
First-professional 85.4 73.1 71.7 62.8 15.4 37.1 7.1 4.6

Institution control
Public 55.5 23.8 22.0 16.8 2.5 28.4 17.4 17.5
Private not-for-profit 65.2 35.4 32.9 29.8 5.4 27.1 9.1 22.4
Private for-profit 70.0 46.0 41.1 43.5 1.1 5.3 1.8 35.4

Graduate program and institution type

Master’s
Public non-doctorate-granting 43.8 19.5 17.4 12.7 1.4 12.4 4.8 20.6
Public doctorate-granting 58.8 23.4 21.2 15.8 1.3 28.8 17.7 21.5
Private non-doctorate-granting 55.3 26.0 21.9 20.2 1.6 11.3 3.5 31.3
Private doctorate-granting 63.8 31.4 29.1 26.0 3.8 24.5 7.4 27.4
Doctorate
Public 72.1 18.2 16.8 11.8 0.7 61.4 50.9 10.7
Private not-for-profit 75.0 26.8 24.3 22.2 7.5 56.3 32.0 10.0
First-professional
Public 87.2 77.5 76.5 62.5 17.9 37.1 7.8 3.0
Private not-for-profit 85.3 71.5 69.8 64.2 13.9 37.9 6.7 5.5
Other program
Public non-doctorate-granting 30.6 8.2 6.7 4.2 0.7 9.2 1.1 17.7
Public doctorate-granting 37.9 12.8 10.5 8.8 0.9 15.2 3.2 22.8
Other, including for-profit 45.5 20.7 18.5 17.3 1.0 10.7 2.9 23.7

Attendance pattern
Full-time, full-year 82.3 52.1 49.6 41.7 8.8 48.7 26.5 6.5
Full-time, part-year 61.8 31.6 28.4 24.3 2.8 27.4 16.9 10.5
Part-time, full-year 54.6 21.2 18.7 16.2 1.1 18.9 6.5 31.4
Part-time, part-year 35.8 7.6 6.4 5.4 0.3 9.3 3.0 27.7

Total 1998 income

Less than $5,000 83.3 62.9 61.7 46.7 15.0 45.8 18.9 2.7
$5,000–9,999 79.5 56.6 55.6 41.8 9.6 45.7 24.7 5.4
$10,000–19,999 69.8 39.7 38.4 30.0 4.8 42.4 30.6 5.7
$20,000–29,999 61.2 33.1 32.0 26.2 2.7 28.0 14.8 14.3
$30,000–49,999 51.7 20.3 18.4 16.1 1.1 21.2 10.1 23.8
$50,000 or more 48.0 11.8 8.1 10.5 0.3 14.0 4.2 33.0

NOTE: “Any aid” includes all types of financial aid except aid from parents, friends, and relatives. Students can receive more than one type of aid. Although assistantships may
include federal, state, or institutional dollars, all assistantships are counted both as “institutional aid” and as assistantships. NCES defines first-professional programs to include
the following fields of study: dentistry, medicine, optometry, osteopathic medicine, veterinary medicine, pharmacy, podiatric medicine, chiropractic, law, and theological
professions. Private master's, doctoral, and first-professional programs are private not-for-profit. All for-profit programs are included under "other program." Estimates by
type of graduate program, institution control, and attendance pattern exclude students who attended multiple institutions.
SOURCE: U.S. Department of Education, National Center for Education Statistics, 1999–2000 National Postsecondary Student Aid Study (NPSAS:2000). (Originally published as
table 13 on p. 20 of the complete report from which this article is excerpted.) Table revised October 2001.

86 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
 National Postsecondary Student Aid Study: Student Financial Aid Estimates for 1999–2000

Table D.—Average amounts of selected types of aid received by graduate and first-professional students: 1999–2000

Stafford loans Institutional aid

Any Unsub- Perkins Institutional Assistant- Employer
aid Total Subsidized sidized loans aid, any ships aid

Total $13,255 $12,849 $7,099 $8,067 $2,767 $9,839 $9,157 $3,546

Type of graduate program

Master’s and doctoral programs 12,160 11,426 6,706 7,175 2,459 10,918 9,505 3,852
Master’s degree 10,391 11,309 6,655 7,054 2,627 7,731 7,481 3,838
Doctorate 18,466 12,059 6,983 7,860 1,952 16,320 11,676 3,998
Other graduate program 6,465 9,515 5,962 6,254 (#) 4,726 (#) 1,324
Postbaccalaureate certificate 8,700 9,912 6,114 6,357 (#) 6,572 (#) 2,101
Other, including non-degree 3,994 8,320 5,477 5,903 (#) 3,126 (#) 903
First-professional 21,505 16,428 8,042 9,945 3,081 7,221 4,981 4,847

Institution control
Public 10,976 11,060 6,766 6,822 2,536 8,969 9,165 2,365
Private not-for-profit 16,245 14,624 7,467 9,091 2,979 11,342 9,391 4,756
Private for-profit 12,545 14,714 7,012 8,936 (#) (#) (#) 4,874

Graduate program and institution type

Master’s
Public non-doctorate-granting 6,561 8,849 5,867 5,529 (#) 4,095 5,560 1,734
Public doctorate-granting 9,168 9,764 6,228 6,103 2,361 7,804 7,611 2,685
Private non-doctorate-granting 7,970 10,479 6,317 6,678 (#) 4,594 (#) 3,435
Private doctorate-granting 14,086 13,521 7,349 8,086 3,126 9,393 8,955 5,987
Doctorate
Public 16,065 10,279 6,567 6,551 (#) 14,334 11,374 4,020
Private not-for-profit 23,332 14,422 7,448 9,265 1,954 20,632 12,756 4,292
First-professional
Public 18,101 14,633 7,995 8,360 2,767 4,614 5,896 (#)
Private not-for-profit 24,014 17,787 8,079 11,024 3,363 9,001 4,216 5,612
Other program
Public non-doctorate-granting 3,295 (#) (#) (#) (#) (#) (#) 700
Public doctorate-granting 6,829 9,294 6,331 5,991 (#) 5,833 (#) 1,432
Other, including for-profit 8,793 12,396 6,672 7,672 (#) 4,153 (#) 2,934

Attendance pattern
Full-time, full-year 19,589 14,426 7,711 8,873 2,881 12,354 9,871 6,034
Full-time, part-year 11,467 10,543 6,051 6,616 (#) 8,302 7,377 6,541
Part-time, full-year 8,631 11,390 6,472 7,418 2,052 5,998 7,844 3,738
Part-time, part-year 3,801 7,278 4,724 4,655 (#) 3,372 7,540 2,117

Total 1998 income

Less than $5,000 18,792 14,057 7,668 8,811 2,683 8,387 5,937 (#)
$5,000–9,999 17,291 12,841 7,537 7,378 2,727 9,783 8,092 5,507
$10,000–19,999 17,249 12,669 7,288 7,423 2,558 12,805 10,791 4,533
$20,000–29,999 14,176 12,684 6,808 7,723 3,418 11,260 10,778 2,967
$30,000–49,999 10,258 11,858 6,257 7,808 (#) 9,021 8,981 2,798
$50,000 or more 7,806 12,394 6,242 9,081 (#) 7,007 7,879 3,728

#Too small to report.

NOTE: Average amounts shown above are for recipients of the specified aid. “Any aid” includes all types of financial aid except aid from parents, friends, and relatives. Students
can receive more than one type of aid. Although assistantships may include federal, state, or institutional dollars, all assistantships are counted both as “institutional aid” and
as assistantships. NCES defines first-professional programs to include the following fields of study: dentistry, medicine, optometry, osteopathic medicine, veterinary medicine,
pharmacy, podiatric medicine, chiropractic, law, and theological professions. Private master's, doctoral, and first-professional programs are private not-for-profit. All for-profit
programs are included under "other program." Estimates by type of graduate program, institution control, and attendance pattern exclude students who attended multiple
institutions.
SOURCE: U.S. Department of Education, National Center for Education Statistics, 1999–2000 National Postsecondary Student Aid Study (NPSAS:2000). (Originally published as
table 14 on p. 21 of the complete report from which this article is excerpted.) Table revised October 2001.

E D U C AT I O N S T AT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , FA L L 2 0 0 1 87
Postsecondary Education

Competing Choices
Competing Choices: Men’s and Women’s Paths After Earning a Bachelor’s
Degree
——————————————————————————————————
Michael S. Clune, Anne-Marie Nuñez, and Susan P. Choy
This article was originally published as the Executive Summary of the Statistical Analysis Report of the same name. The sample survey data are from the
Baccalaureate and Beyond Longitudinal Study (B&B).

During the last 30 years, women have made great strides in Gender Differences
educational attainment, particularly in participating in Women’s and men’s characteristics and experiences differed
postsecondary education, where they not only enroll and both at the time they received their bachelor’s degree and
attain at higher rates than men but also do better academi- during the next 4 years.
cally and have higher educational expectations, on average
(National Center for Education Statistics 2000; Berkner, Characteristics at bachelor’s degree receipt
McCormick, and Cuccaro-Alamin 1996; McCormick et al. Among 1992–93 bachelor’s degree recipients, women
1999). However, the superior performance of women at differed from men on a number of characteristics, including
the undergraduate level has not translated into greater age, marital and parenthood status, undergraduate major,
enrollment than males at the graduate level or enrollment grade-point average (GPA), and educational aspirations.
rates equal to males in all types of graduate programs Compared with men, women were more likely to be under
(McCormick et al. 1999). age 23 (51 percent vs. 42 percent) or over age 29 (19 per-
cent vs. 13 percent). They were also more likely than men
At the same time that young adults are making decisions to have married (29 percent vs. 24 percent) and to have
about graduate study and employment after earning their children (16 percent vs. 12 percent) by the time they
bachelor’s degree, many are also facing choices about graduated.
marriage and parenthood. These latter life transitions may
play a greater role in women’s decisions about schooling With respect to their undergraduate experiences, women
and employment at this juncture because women generally were more likely than men to major in certain fields, most
marry and have children at younger ages than do men. notably education (18 percent vs. 6 percent) and health
Thus, choices about getting married and having children professions (10 percent vs. 4 percent). Men, in contrast,
may compete with choices about employment and graduate were more likely than women to major in business and
study more for women than for men. This report aims to management (26 percent vs. 19 percent) and engineering
provide a context for understanding the paths that women (12 percent vs. 2 percent). Women graduated with higher
and men take toward graduate degrees, employment, GPAs than men: 61 percent of women had GPAs of 3.0 or
marriage, and parenthood during the first 4 years after higher, compared with 49 percent of men.
earning their bachelor’s degree. In particular, the analysis
seeks to identify how these behaviors are interrelated. Finally, at the time they earned their bachelor’s degree,
women were more likely than men to expect to earn a
This analysis draws upon data from the 1993 Baccalaureate graduate degree (87 percent vs. 83 percent). Marital status
and Beyond Longitudinal Study (B&B:93/97), which as well as gender was related to educational plans, with
identified students who received their bachelor’s degree single1 women being more likely to expect to earn a graduate
during academic year 1992–93. The analysis also used degree (89 percent) than married women (83 percent) and
follow-up surveys conducted in 1994 and 1997 to trace both married and single men (82 percent and 84 percent,
changes in employment and graduate enrollment, along respectively). Among those expecting to earn a graduate
with changes in marital status and entry into parenthood. In degree, married women were less likely than single women
order to obtain complete information about graduates’ paths and both married and single men to expect to earn a first-
4 years after degree receipt, this analysis was limited to professional or doctoral degree (figure A).
graduates who responded to the second follow-up survey in
1997. The findings of the report are summarized below. 1
Throughout the report, “single” refers to individuals who have never been married.

88 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
 Competing Choices: Men’s and Women’s Paths After Earning a Bachelor’s Degree

Figure A.—Among 1992–93 bachelor’s degree recipients who expected to earn a graduate degree, percentage distribution
according to degree expected at the time of bachelor’s degree receipt, by marital status and gender

Percent
Married women
100
Single women
80
80 Married men
69
66 65 Single men

60

40
27 24 25
20 16
8 7 9
4
0
Master's degree First-professional degree Doctoral degree
Graduate degree goal at bachelor's degree receipt

NOTE: Percentages may not add to 100 because of rounding. “Single” means never been married; “married” means married at time of
bachelor’s degree receipt.
SOURCE: U.S. Department of Education, National Center for Education Statistics, 1993 Baccalaureate and Beyond Longitudinal Study, “Second
Follow-up” (B&B:93/97), Data Analysis System.

Experiences after graduation men were about equally likely to earn a graduate degree
During the first 4 years after graduation, women and men within this time frame (16 percent and 15 percent, respec-
had different experiences with respect to marriage, parent- tively), among those who did, men were more likely to earn
hood, graduate enrollment, graduate attainment, and a first-professional or doctoral degree. Among those who
employment. Among those who had not married by the earned a graduate degree, 13 percent of women and
time they graduated, women were more likely than men to 23 percent of men earned a first-professional or doctoral
have married within 4 years (32 percent vs. 28 percent) degree.
(figure B).
Women and men were about equally likely to be employed
Entry into parenthood occurred at lower rates than mar- during the first 4 years after earning their bachelor’s degree,
riage. Within 4 years, 13 percent of bachelor’s degree but among those working, men were more likely to be
recipients who were not parents at graduation became employed full time. For example, 2 years after graduation,
parents. As with marriage, women were more likely than 84 percent of women and 86 percent of men were em-
men to make this transition (15 percent vs. 11 percent). ployed; however, 92 percent of employed men were work-
ing full time, compared with 87 percent of employed
After 4 years, 29 percent of bachelor’s degree recipients had women.
enrolled in a graduate degree program. While women and
men were equally likely to enroll, women were more likely Age, major, and grade-point average
to enroll in master’s degree programs and men were more In addition to gender, several other characteristics of
likely to enroll in first-professional and doctoral programs bachelor’s degree recipients were related to their patterns
(figure C). of marriage, parenthood, graduate enrollment and attain-
ment, and employment. These characteristics include age
Fifteen percent of the 1992–93 bachelor’s degree recipients at graduation, undergraduate field of study, and under-
earned a graduate degree within 4 years. While women and graduate GPA.

E D U C AT I O N S T AT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , FA L L 2 0 0 1 89
Postsecondary Education

Figure B.—Among 1992–93 bachelor’s degree recipients who at the time of graduation had never been married,
cumulative percentage married each month for the next 4 years, by gender

Cumulative
percentage
married
100 100

80 80

60 60

40 40
Women

20 Men 20

0 0
0 6 12 18 24 30 36 42 48
Months since bachelor's degree receipt

SOURCE: U.S. Department of Education, National Center for Education Statistics, 1993 Baccalaureate and Beyond Longitudinal
Study, “Second Follow-up” (B&B:93/97), Data Analysis System.

Figure C.—Among 1992–93 bachelor’s degree recipients who enrolled in a graduate degree program within
4 years of graduation, percentage distribution by highest level of enrollment, by gender

Master's degree

Women 83 11 7 First-professional degree

Doctoral degree

Men 68 19 13

0 20 40 60 80 100
Percent

NOTE: Percentages may not add to 100 because of rounding.

SOURCE: U.S. Department of Education, National Center for Education Statistics, 1993 Baccalaureate and Beyond Longitudinal
Study, “Second Follow-up” (B&B:93/97), Data Analysis System.

90 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
 Competing Choices: Men’s and Women’s Paths After Earning a Bachelor’s Degree

First, older graduates were more likely to have married 16 percent of women who did not enroll in a graduate
before earning their bachelor’s degree. Among women who program within 4 years of bachelor’s degree receipt became
had not married by the time they graduated, women under parents during that time. In contrast, 9 percent of men and
30 were more likely than women who were older to marry 10 percent of women who enrolled did so. Women who
within the next 4 years. For both men and women, those enrolled in first-professional or doctoral programs were less
ages 25–29 at graduation were more likely than those in likely to marry and become parents than were those who
other age groups to become parents within 4 years of enrolled in master’s programs.
graduation. Age was a factor for graduate enrollment as
well, with both men and women who were age 22 or Graduate school enrollment and attainment
younger when they earned their bachelor’s degree more Marriage and parenthood are more related to graduate
likely than older graduates to enter graduate school within outcomes for women than for men. Compared with women
4 years after graduation. who did not marry before earning their bachelor’s degree,
women who did marry before earning their bachelor’s
Second, graduates who majored in professional fields2 as a degree were less likely to enroll in a graduate program or to
group were more likely to be married and to have children enroll in a first-professional or doctoral degree program.
before graduating, compared with graduates who majored Similarly, women who married before graduation were less
in the arts and sciences. Those majoring in the arts and likely to attain a graduate degree, and, among those who
sciences were more likely than those in professional fields attained, less likely to attain a first-professional or doctoral
to enroll in a graduate program. Probably due to their degree. Similar consistent negative links to graduate
higher levels of graduate enrollment, bachelor’s degree enrollment and attainment were observed among women
recipients who majored in the arts and sciences were less who became parents after graduation.
likely to be employed during the first 4 years following
graduation. Among men, marriage before earning a bachelor’s degree
was related to a lower rate of enrollment in graduate school,
Finally, graduates with higher GPAs were more likely to be but marriage within the next 4 years was not related to the
married and have children before graduating. Those with rate of enrollment. In addition, marriage after graduation
higher GPAs were also more likely to enroll in a graduate was not related to the type of degree program chosen.
program, enroll in a first-professional or doctoral program, Among men who enrolled, neither marriage nor parenthood
and attain a graduate degree within 4 years of bachelor’s was related to men’s graduate degree attainment.
degree receipt.
Employment
Interrelationships Among Transitions While men and women were about equally likely to be
In general, marriage, parenthood, graduate enrollment and employed after earning their bachelor’s degree, differences
attainment, and employment appeared to have different existed according to marital and parenthood status. Among
interrelationships for women and men. those who married before graduating, women were gener-
ally less likely than men to work after graduating. In
Marriage contrast, among graduates who did not marry within
Graduate enrollment and marriage were negatively related 4 years of graduating, women were generally more likely
for women, but not for men. Thirty-three percent of women than men to be employed. Parenthood negatively affected
who did not enroll in a graduate program within 4 years of women’s employment: women who became parents either
bachelor’s degree receipt married during that period, before or within 4 years after graduating were less likely
compared with 29 percent of those who did enroll. In than men to work.
contrast, the marriage rate for men was about the same
whether they enrolled (27 percent) or not (29 percent). Effects of Marriage and Parenthood on
Graduate Enrollment After Controlling for
Parenthood Other Variables
Graduate enrollment and parenthood were negatively For this report, multivariate analyses were conducted to
related for both men and women: 12 percent of men and examine the net effects of parenthood and marriage on
enrolling in a graduate degree program after taking into
2
Business and management, education, engineering, health professions, and public account variables other than gender that might be related to
affairs/social services.

E D U C AT I O N S T AT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , FA L L 2 0 0 1 91
Postsecondary Education

graduate enrollment—such as age, race/ethnicity, parents’ McCormick, A., Nuñez, A.-M., Shah, V., and Choy, S.P. (1999).
education, and undergraduate education (control and level Life After College: A Descriptive Summary of 1992–93 Bachelor’s
Degree Recipients in 1997 (NCES 1999–155). U.S. Department of
of institution, major, and GPA). Analyses were conducted
Education, National Center for Education Statistics. Washing-
for women and men separately. ton, DC: U.S. Government Printing Office.
National Center for Education Statistics. (2000). The Condition of
For women, marriage before bachelor’s degree receipt was Education 2000 (NCES 2000–062). U.S. Department of Educa-
negatively related to graduate enrollment. After controlling tion, National Center for Education Statistics. Washington, DC:
for other characteristics, 23 percent of women who married U.S. Government Printing Office.
before receiving their bachelor’s degree enrolled in graduate
school, compared with 33 percent of women who had not
yet married 4 years after earning their bachelor’s degree. Data source: The NCES 1993 Baccalaureate and Beyond Longitudinal
Marriage was not significantly related to graduate enroll- Study (B&B:93/97).

ment for men, however, after controlling for other For technical information, see the complete report:

characteristics. Clune, M.S., Nuñez, A.-M., and Choy, S.P. (2001). Competing Choices:
Men’s and Women’s Paths After Earning a Bachelor’s Degree
(NCES 2001–154).
References Author affiliations: M.S. Clune, A.-M. Nuñez, and S.P. Choy, MPR
Berkner, L.K., McCormick, A.C., and Cuccaro-Alamin, S. (1996). Associates, Inc.
Descriptive Summary of 1989–90 Beginning Postsecondary For questions about content, contact Aurora D’Amico
Students: 5 Years Later, With an Essay on Postsecondary Persistence (aurora.d’amico@ed.gov).
and Attainment (NCES 96–155). U.S. Department of Education, To obtain the complete report (NCES 2001–154), call the toll-
National Center for Education Statistics. Washington, DC: U.S. free ED Pubs number (877–433–7827), visit the NCES Web Site
(http://nces.ed.gov), or contact GPO (202–512–1800).
Government Printing Office.

92 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
Postsecondary Institutions
Postsecondary Institutions in the United States: 1993–94 and 1998–99
——————————————————————————————————
Patricia Q. Brown
This article was originally published as the Summary of the E.D. Tabs report of the same name. The sample survey data are from the Integrated
Postsecondary Education Data System “Institutional Characteristics Survey” (IPEDS-IC).

Introduction school. This includes programs whose purpose is academic,

This report presents detailed tabulations for the 1998–99 vocational, and continuing professional education, and
academic year that describe characteristics of the 9,653 excludes avocational (leisure) and adult basic education
postsecondary education institutions in the United States programs. For the 1998–99 academic year, 9,485 institutions
(50 states and the District of Columbia) and outlying areas in the 50 states and the District of Columbia and 168 in the
(table A).1 These characteristics include tuition and outlying areas fit the IPEDS definition (table A). IPEDS
required fees for undergraduate, graduate, and first- attempts every year to identify institutions that should be
professional programs and room and board charges at included in the universe. Because of the changing nature of
institutions providing these accommodations. Data are from the postsecondary education enterprise (“births” and
the “Institutional Characteristics Survey,” a component of “deaths” of institutions), there may be more than the 9,653
the Integrated Postsecondary Education Data System postsecondary institutions currently identified in IPEDS.
(IPEDS) of the U.S. Department of Education’s National
Center for Education Statistics (NCES). This report also In 1998–99, IPEDS collected data from over 9,600 post-
provides a comparison between 1993–94 and 1998–99 secondary institutions, with more than 6,500, or 68 per-
tuition, required fees, and room and board charges for the cent, of the institutions having a Program Participation
50 states and the District of Columbia. Agreement (PPA) with the Department of Education and
thus eligible to participate in Title IV programs. Title IV of
Postsecondary education is the provision of a formal instruc- the Higher Education Act of 1965 (as amended) established
tional program whose curriculum is designed primarily for federal financial aid programs (e.g., Pell Grants, Stafford
students who are beyond the compulsory age for high Loans) for students attending postsecondary institutions.
Students attending institutions with a PPA may be eligible
1
The outlying areas include American Samoa, the Federated States of Micronesia,
either to receive Title IV funds or to defer repayment of
Guam, the Marshall Islands, Palau, Puerto Rico, and the Virgin Islands. their loans.

Table A.—The number of postsecondary institutions, by degree-granting status, Title IV participation, and
control of institution: 50 states, District of Columbia, and outlying areas, academic year 1998–99

Private Private
Total Public not-for-profit for-profit

50 states and DC 9,485 2,245 2,777 4,463

Degree-granting 4,500 1,698 2,043 759
Non-degree-granting 4,985 547 734 3,704

Title IV participating 6,431 2,090 1,986 2,355

Non–Title IV participating 3,054 155 791 2,108

Outlying areas 168 34 51 83

Degree-granting 91 32 42 17
Non-degree-granting 77 2 9 66

Title IV participating 142 26 43 73

Non–Title IV participating 26 8 8 10

SOURCE: U.S. Department of Education, National Center for Education Statistics, 1998 Integrated Postsecondary
Education Data System, “Institutional Characteristics Survey” (IPEDS-IC:98–99).

E D U C AT I O N S T AT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , FA L L 2 0 0 1 93
Postsecondary Education

The IPEDS universe also classifies institutions by degree- tions and institutions that report tuition and fees by pro-
granting status. Institutions are considered degree-granting if gram only are not included in this report. It is important to
they offer at least one associate’s or higher degree. In 1998–99, note that tuition and required fees do not represent the total
almost 4,600, or 48 percent, of all IPEDS institutions granted cost to attend college. Excluded are costs for books and
degrees. Forty-three percent of the postsecondary institutions supplies, social activities, and room and board.
in IPEDS were Title IV participating and degree-granting in
the 50 states, the District of Columbia, and outlying areas. Public institutions
In 1998–99, there were 2,279 public postsecondary institu-
While this summary focuses on all postsecondary institu- tions in the 50 states, District of Columbia, and outlying
tions, many of the tables presented in the complete report areas (table A). Of the 2,279 institutions, 645 offered
provide information on the subsets of 6,431 Title IV par- bachelor’s or higher degrees, while 1,269 offered programs
ticipating and 4,500 degree-granting institutions in the of at least 2 but less than 4 years’ duration. Of the respond-
50 states and the District of Columbia. ing 4-year institutions, 596 institutions in the 50 states and
the District of Columbia reported tuition and required fees
Tuition and Required Fees at Postsecondary averaging $3,186 for full-time, full-year3 undergraduate in-
Institutions state students in 1998–99 (table B). The median charge was
The tuition and required fees discussed in this report $2,998 for undergraduate in-state students. The public
represent all responding institutions that offer either 4-year institutions continued to charge lower tuition and
undergraduate, graduate, or first-professional programs2 required fees for students attending schools in states where
and enroll full-time students. The nonresponding institu- they were legal residents. The full-time, full-year out-of-
2 3
The first-professional programs consist of Chiropractic; Dentistry; Law; Medicine; Full-year is an academic year, the period of time generally extending from September
Optometry; Osteopathic Medicine; Pharmacy; Podiatry; Theology; and Veterinary to June, usually equated to two semesters or trimesters, three quarters, or the period
Medicine. covered by a 4-1-4 plan.

Table B.—Average institutional charges for tuition and required fees for full-time, full-year students at all postsecondary institutions, by level and control of
institution: 50 states and the District of Columbia, academic year 1998–99

4 years and above 2 but less than 4 years Less than

Private Private 2 years
Item Total Public Not-for-profit For-profit Public Not-for-profit For-profit Public

Undergraduate tuition and required fees (in-state)

Number of institutions responding 4,277 596 1,347 166 1,173 336 501 158
Mean charge $6,180 $3,186 $11,229 $8,194 $1,697 $5,489 $7,528 $2,809
Median charge $4,628 $2,998 $11,173 $7,502 $1,430 $5,600 $7,301 $2,078

Undergraduate tuition and required fees (out-of-state)

Number of institutions responding 4,277 596 1,347 166 1,173 336 501 158
Mean charge $7,561 $8,248 $11,246 $8,194 $4,006 $5,631 $7,529 $3,479
Median charge $6,740 $8,300 $11,180 $7,502 $3,884 $5,728 $7,301 $3,000

Graduate tuition and required fees (in-state)

Number of institutions responding 1,653 511 1,035 107 (†) (†) (†) (†)
Mean charge $7,076 $3,555 $8,609 $9,059 (†) (†) (†) (†)
Median charge $5,825 $3,248 $7,614 $7,440 (†) (†) (†) (†)

Graduate tuition and required fees (out-of-state)

Number of institutions responding 1,653 511 1,035 107 (†) (†) (†) (†)
Mean charge $8,555 $8,310 $8,623 $9,059 (†) (†) (†) (†)
Median charge $7,700 $8,046 $7,630 $7,440 (†) (†) (†) (†)

†Not applicable.
NOTE: Mean and median charges are calculated using institutions as the unit of analysis, not students. Undergraduate tuition represents all responding institutions that offer
undergraduate programs and have full-time undergraduate students. Graduate tuition represents all responding institutions that offer graduate programs and have full-time
graduate students.
SOURCE: U.S. Department of Education, National Center for Education Statistics, 1998 Integrated Postsecondary Education Data System, “Institutional Characteristics Survey”
(IPEDS–IC:98–99).

94 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
 Postsecondary Institutions in the United States: 1993–94 and 1998–99

state undergraduate students at these same institutions paid state tuition, the charge for out-of-state undergraduates
an average of $8,248 for tuition and required fees. The full- averaged $11,246, just $17 more than the charge for in-state
time, full-year graduate tuition and required fees averaged students. The median charge was $11,173 for in-state and
$3,555 for in-state and $8,310 for out-of-state students at $11,180 for out-of-state undergraduate students, only a
the 511 responding public 4-year institutions. $7 difference. The median charge for out-of-state graduate
students was higher at public institutions ($8,046) than at
The 1,173 responding public 2-year institutions reported private not-for-profit institutions ($7,630) or private for-
that in-state students attending their institutions were profit institutions ($7,440). In fact, the median charge for
charged an average of $1,697 for tuition and required fees out-of-state graduate students at private for-profit institu-
(table B). Out-of-state students attending these same tions was more than $600 lower than at public institutions,
institutions were charged an average of $4,006, a difference and the median charge was more than $400 lower at private
of $2,309. There were 158 responding less-than-2-year not-for-profit institutions than at public institutions.
institutions, with average tuition and fees for in-state
students of $2,809 for 1998–99, compared to $3,479 for Room and Board Charges
out-of-state students. The room charges are reported by academic year for
rooming accommodations for a typical student sharing a
Private institutions room with one other student. In 1998–99, 433 public 4-year
For the purpose of IPEDS, private institutions are defined institutions in the 50 states and the District of Columbia
as educational institutions controlled by a private reported an average of $2,338 for room charges to students
individual(s) or by a nongovernmental agency, usually and 176 public 2-year institutions reported an average of
supported primarily by other than public funds and oper- $1,395 (table C). The 21 private for-profit 4-year institu-
ated by other than publicly elected or appointed officials. tions that reported offering dormitory facilities charged an
There are two types of private institutions in IPEDS: for- average of $3,531. Over 1,000 private not-for-profit institu-
profit and not-for-profit. This report discusses these two tions reported having dormitory facilities, with an average
types of institutions separately. charge of $2,599 at 4-year institutions and $1,918 at 2-year
institutions.
The charge for full-time, full-year undergraduate students to
attend 4-year private not-for-profit institutions in 1998–99 The board charges are reported for an academic year for a
averaged $11,229 for tuition and required fees (table B). specified number of meals per week. Of the responding
Since very few private institutions charge different out-of- institutions, 1,457 indicated they offer meals (table C). The

Table C.—Average institutional charges for room and board for full-time, full-year students at all postsecondary institutions, by level and control of institution:
50 states and the District of Columbia, academic year 1998–99

4 years and above 2 but less than 4 years Less than

Private Private 2 years
Item Total Public Not-for-profit For-profit Public Not-for-profit For-profit Public

Dormitory facilities
Number providing facilities 1,763 433 987 21 176 98 46 2
Mean charge $2,400 $2,338 $2,599 $3,531 $1,395 $1,918 $3,082 $1,920
Median charge $2,228 $2,190 $2,400 $3,792 $1,240 $1,800 $3,230 $1,920

Meal plans
Number providing facilities 1,457 403 853 7 144 39 9 2
Mean charge $2,150 $1,934 $2,363 $1,854 $1,612 $1,831 $2,000 $1,408
Median charge $2,160 $1,930 $2,400 $1,600 $1,588 $2,000 $1,884 $1,408

Mean meals per week 18 18 19 14 17 19 15 19

Median meals per week 19 19 19 15 19 20 15 19

NOTE: Mean and median charges are calculated using institutions as the unit of analysis, not students.
SOURCE: U.S. Department of Education, National Center for Education Statistics, 1998 Integrated Postsecondary Education Data System, “Institutional Characteristics Survey”
(IPEDS-IC:98–99).

E D U C AT I O N S T AT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , FA L L 2 0 0 1 95
Postsecondary Education

average charge for all institutions offering meal plans was increase of 27 percent and a median increase of 41 percent
$2,150 for an average of 18 meals per week. Charges for a from 1993–94 to 1998–99 for dormitory charges for all
student attending a 4-year private not-for-profit institution students. The private 4-year not-for-profit institutions
were an average of $2,363 for 19 meals per week. reported an average charge of $2,037 in 1993–94 and
$2,599 in 1998–99, a 28 percent increase over this period.
Changes in Institutional Charges From
1993–94 to 1998–99 Postsecondary institutions reported an average charge of
Between 1993–94 and 1998–99, there was a 33 percent $2,150 for meal plans at their institutions in 1998–99.
increase (from $4,647 to $6,180) in the average charge of This was 17 percent higher than the charges assessed in
undergraduate tuition and required fees for in-state 1993–94, when the average charge was $1,835 for meal
students at postsecondary institutions (table D). In plans. The public 4-year institutions charged an average
1998–99, in-state undergraduate students at public 4-year of $1,622 for meal plans in 1993–94, which increased
institutions paid an average of $3,186 for tuition and 19 percent to $1,934 in 1998–99. The average charge for
required fees, a 29 percent rise from 1993–94, when the meal plans at public 2-year institutions was $1,612 in
average was $2,479. Although the average tuition and fees 1998–99, up 10 percent from $1,463 in 1993–94 (table E).
charged by public 4-year institutions increased, the rate
of increase was slightly less than that for private 4-year Overall Changes From 1993–94 to 1998–99
institutions. The in-state undergraduate students at private Over the 5-year period from 1993–94 to 1998–99, the
not-for-profit 4-year institutions paid an average of $11,229 average total institutional charges (tuition, required fees,
in 1998–99, up 33 percent from $8,419 in 1993–94, and room and board) for an undergraduate student to attend
students attending for-profit 4-year institutions paid college increased 25 percent at public 4-year institutions
30 percent more in 1998–99. ($5,969 to $7,458) for those paying in-state tuition. The
average total institutional charges for out-of-state students
The largest increase impacting the in-state undergraduate increased 27 percent during this same period ($9,856 to
tuition and required fees was at 2-year private not-for-profit $12,520). At private institutions, the average total price for
institutions. The in-state students attending these institu- undergraduates increased 30 percent at 4-year not-for-profit
tions in 1998–99 paid an average of 51 percent more than institutions ($12,482 to $16,191) and approximately
those who attended in 1993–94 ($5,489 vs. $3,624). The 29 percent4 at 4-year for-profit institutions.
median charge at these institutions rose 87 percent from 4
Percent change based on tuition, fees, and room; no board charges are available.
1993–94 to 1998–99 ($3,000 to $5,600).

Tuition and fees for in-state graduate students increased

31 percent over the 5-year period, from an average of
$5,417 in 1993–94 to $7,076 in 1998–99. The for-profit Data source: The NCES 1993 and 1998 Integrated Postsecondary
institutions reported the largest increase, 43 percent during Education Data System, “Institutional Characteristics Survey”
(IPEDS-IC:93–94 and 98–99).
this period.
For technical information, see the complete report:
Brown, P.Q. (2001). Postsecondary Institutions in the United States:
Changes in Room and Board Charges From 1993–94 and 1998–99 (NCES 2001–176).
1993–94 to 1998–99 Author affiliation: P.Q. Brown, NCES.
The average dormitory charge for students attending For questions about content, contact Patricia Q. Brown
postsecondary institutions showed a 28 percent increase (patricia.brown@ed.gov).
over the 5-year period ($1,873 vs. $2,400) (table E). The To obtain the complete report (NCES 2001–176), visit the NCES
Web Site (http://nces.ed.gov).
private for-profit 4-year institutions reported an average

96 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
 Postsecondary Institutions in the United States: 1993–94 and 1998–99

Table D.—Average institutional charges for tuition and required fees for full-time, full-year students at all postsecondary institutions, by level and control of
institution: 50 states and the District of Columbia, academic years 1993–94 and 1998–99

4 years and above 2 but less than 4 years Less than

Private Private 2 years
Item Total Public Not-for-profit For-profit Public Not-for-profit For-profit Public

Undergraduate tuition and required fees (in-state)

Mean charge 1993–94 $4,647 $2,479 $8,419 $6,296 $1,372 $3,624 $6,256 $2,070
1998–99 $6,180 $3,186 $11,229 $8,194 $1,697 $5,489 $7,528 $2,809
Percent change 33 29 33 30 24 51 20 36

Median charge 1993–94 $3,330 $2,260 $8,290 $6,093 $1,121 $3,000 $5,818 $1,676
1998–99 $4,628 $2,998 $11,173 $7,502 $1,430 $5,600 $7,301 $2,078
Percent change 39 33 35 23 28 87 25 24

Undergraduate tuition and required fees (out-of-state)

Mean charge 1993–94 $5,634 $6,366 $8,435 $6,305 $3,174 $3,719 $6,256 $2,411
1998–99 $7,561 $8,248 $11,246 $8,194 $4,006 $5,631 $7,529 $3,479
Percent change 34 30 33 30 26 51 20 44

Median charge 1993–94 $4,830 $6,244 $8,310 $6,093 $3,136 $3,125 $5,818 $2,250
1998–99 $6,740 $8,300 $11,180 $7,502 $3,884 $5,728 $7,301 $3,000
Percent change 40 33 35 23 24 83 25 33

Graduate tuition and required fees (in-state)

Mean charge 1993–94 $5,417 $2,735 $6,765 $6,332 (†) (†) (†) (†)
1998–99 $7,076 $3,555 $8,609 $9,059 (†) (†) (†) (†)
Percent change 31 30 27 43 (†) (†) (†) (†)

Median charge 1993–94 $4,301 $2,453 $5,948 $5,400 (†) (†) (†) (†)
1998–99 $5,825 $3,248 $7,614 $7,440 (†) (†) (†) (†)
Percent change 35 32 28 38 (†) (†) (†) (†)

Graduate tuition and required fees (out-of-state)

Mean charge 1993–94 $6,588 $6,246 $6,782 $6,364 (†) (†) (†) (†)
1998–99 $8,555 $8,310 $8,623 $9,059 (†) (†) (†) (†)
Percent change 30 33 27 42 (†) (†) (†) (†)

Median charge 1993–94 $5,924 $5,904 $5,948 $5,402 (†) (†) (†) (†)
1998–99 $7,700 $8,046 $7,630 $7,440 (†) (†) (†) (†)
Percent change 30 36 28 38 (†) (†) (†) (†)

†Not applicable.
NOTE: Mean and median charges are based on institution and not student enrollment. Undergraduate tuition represents all responding institutions that offer undergraduate
programs and have full-time undergraduate students. Graduate tuition represents all responding institutions that offer graduate programs and have full-time graduate students.
Institutions that report tuition by program are not included.
SOURCE: U.S. Department of Education, National Center for Education Statistics, 1993 and 1998 Integrated Postsecondary Education Data System, “Institutional Characteristics
Survey” (IPEDS-IC:93–94 and 98–99).

E D U C AT I O N S T AT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , FA L L 2 0 0 1 97
Postsecondary Education

Table E.—Average institutional charges for room and board for full-time, full-year students at all postsecondary institutions, by level and control of institution:
50 states and the District of Columbia, academic years 1993–94 and 1998–99

4 years and above 2 but less than 4 years Less than

Private Private 2 years
Item Total Public Not-for-profit For-profit Public Not-for-profit For-profit Public

Dormitory facilities
Mean charge 1993–94 $1,873 $1,868 $2,037 $2,781 $1,153 $1,524 $2,327 (#)
1998–99 $2,400 $2,338 $2,599 $3,531 $1,395 $1,918 $3,082 (#)
Percent change 28 25 28 27 21 26 32 (#)

Median charge 1993–94 $1,729 $1,768 $1,850 $2,680 $1,075 $1,450 $2,385 (#)
1998–99 $2,228 $2,190 $2,400 $3,792 $1,240 $1,800 $3,230 (#)
Percent change 29 24 30 41 15 24 35 (#)

Meal plans
Mean charge 1993–94 $1,835 $1,622 $2,026 (#) $1,463 $1,558 $1,659 3
1998–99 $2,150 $1,934 $2,363 (#) $1,612 $1,831 $2,000 (#)
Percent change 17 19 17 (#) 10 18 21 (#)

Median charge 1993–94 $1,850 $1,609 $2,000 (#) $1,470 $1,599 $1,700 (#)
1998–99 $2,160 $1,930 $2,400 (#) $1,588 $2,000 $1,884 (#)
Percent change 17 20 20 (#) 8 25 11 (#)

#Too small to report.

NOTE: Mean and median charges are based on institution and not student enrollment.
SOURCE: U.S. Department of Education, National Center for Education Statistics, 1993 and 1998 Integrated Postsecondary Education Data System, “Institutional Characteristics
Survey” (IPEDS-IC:93–94 and 98–99).

98 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
 LIBRARIES

Public Libraries in the United States: Fiscal Year 1998

Adrienne Chute and P. Elaine Kroe ........................................................................ 99
Academic Libraries: 1998
Margaret W. Cahalan and Natalie M. Justh ......................................................... 102

Public Libraries
Public Libraries in the United States: Fiscal Year 1998
——————————————————————————————————
Adrienne Chute and P. Elaine Kroe
This article was originally published as the Introduction and Highlights of the E.D. Tabs report of the same name. The universe data are from the NCES
Public Libraries Survey (PLS).

Introduction dance, and circulation of children’s materials. It also

The tables in this report summarize information about includes information about size of collection, staffing,
public libraries in the 50 states and the District of Columbia operating income and expenditures, type of legal basis, and
for state fiscal year (FY) 1998.1 (Data from two outlying type of administrative structure, as well as summary
areas, Guam and the Northern Mariana Islands, are also information about the number and type of public library
included in the tables,2 but not in the table totals.) The data service outlets.3
were collected through the Public Libraries Survey (PLS),
conducted annually by the National Center for Education The following highlights are for libraries in the 50 states
Statistics (NCES) through the Federal-State Cooperative and the District of Columbia.
System (FSCS) for Public Library Data. The FY 98 survey is
the 11th in the series. Number of Public Libraries, Population of
Legal Service Area, and Service Outlets
This report includes information about service measures Number of libraries and population served
such as access to the Internet and other electronic services, There were 8,964 public libraries (administrative entities)
reference transactions, public service hours, interlibrary in the 50 states and the District of Columbia in FY 98.
loans, circulation, library visits, children’s program atten- Eleven percent of the public libraries served 72 percent of
the population of legally served areas in the United States;
1
In three states (Michigan, Pennsylvania, and Texas), some public libraries reported each of these public libraries had a legal service area
data for FY 97. Most of West Virginia’s data are for FY 97.
2
The National Center for Education Statistics is working with other outlying areas and
3
hopes to be able to include their data in future years. See the glossary in the full report for definitions of terms used in the report.

E D U C AT I O N S TAT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , F A L L 2 0 0 1 99
Libraries

population of 50,000 or more. Of the total population of the Nationwide, total per capita5 operating income for public
states and the District of Columbia, 97 percent4 had access libraries was $26.02. Of that, $20.18 was from local sources,
to public library services, and 3 percent did not. $3.28 from state sources, $.21 from federal sources, and
$2.35 from other sources. Per capita operating income from
Service outlets local sources was under $3.00 for 11 percent of public
Eighty percent of public libraries had a single direct service libraries, $3.00 to $14.99 for 43 percent of libraries, $15.00
outlet (an outlet that provides service directly to the to $29.99 for 30 percent of libraries, and $30.00 or more for
public). Twenty percent had more than one direct service 17 percent of libraries.
outlet. This report includes information about three types of
direct public library service outlets: branch library outlets, Operating expenditures
central library outlets, and bookmobile outlets. A total of Total operating expenditures for public libraries were $6.2
1,513 public libraries (17 percent) had one or more branch billion. Of this, 64 percent was expended for paid staff and
library outlets, with a total of 7,293 branches. The total 15 percent for the library collection. Thirty-five percent of
number of central library outlets was 8,887. The total public libraries had operating expenditures of less than
number of stationary outlets (central library outlets and $50,000, 39 percent expended from $50,000 to $399,999,
branch library outlets) was 16,180. Nine percent of public and 25 percent expended $400,000 or more.
libraries had one or more bookmobile outlets, with a total
of 933 bookmobiles. Expenditures for materials in electronic format were 1 per-
cent of total operating expenditures for public libraries.
Legal Basis and Interlibrary Relationships Expenditures for electronic access were 3 percent of total
In FY 98, 53 percent of public libraries were part of a operating expenditures.
municipal government, 12 percent were part of a county/
parish, 1 percent were part of a city/county, 6 percent had The average U.S. per capita operating expenditure for public
multijurisdictional legal basis under an intergovernmental libraries was $23.92. The highest average per capita operat-
agreement, 11 percent were nonprofit association or agency ing expenditure in the 50 states and the District of Colum-
libraries, 3 percent were part of a school district, and 8 per- bia was $42.31 and the lowest was $10.43.
cent were separate government units known as library
districts. Seven percent reported their legal basis as “other.” Staff and Collections
Staff
Seventy-three percent of public libraries were members of a Public libraries had a total of 123,443 paid full-time-equiva-
system, federation, or cooperative service, while 23 percent lent (FTE) staff in FY 98, or 11.9 paid FTE staff per 25,000
were not. Four percent served as the headquarters of a population. Of these, 23 percent, or 2.7 per 25,000 popula-
system, federation, or cooperative service. tion, were librarians with the ALA-MLS,6 and 10 percent were
librarians by title but did not have the ALA-MLS. Sixty-seven
Operating Income and Expenditures percent of the staff were in other positions.
Operating income
In FY 98, 78 percent of public libraries’ total operating Collections
income of about $6.7 billion came from local sources, Nationwide, public libraries had 739 million books and
13 percent from state sources, 1 percent from federal serial volumes in their collections, or 2.9 volumes per
sources, and 9 percent from other sources, such as gifts capita. By state, the number of volumes per capita ranged
and donations, service fees, and fines.
5
Per capita figures are based on the total unduplicated population of legal service
4
areas in the 50 states and the District of Columbia, not on the state total population
This percentage was derived by dividing the total unduplicated population of legal estimates.
service areas in the United States by the sum of the official state total population
6
estimates as reported by the 50 states and the District of Columbia. (Also see Data File: Librarians with master’s degrees from programs of library and information studies
Public Libraries Survey: Fiscal Year 1998, available on the NCES Web Site.) accredited by the American Library Association.

100 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
 Public Libraries in the United States: Fiscal Year 1998

from 1.6 to 5.4. In addition to printed materials, public Other service measures
libraries nationwide had collections of 28 million audio Nationwide,
materials and 17 million video materials. Nationwide,
■ 13.5 million library materials were loaned by public
public libraries provided 4.4 materials in electronic format
libraries to other libraries (an increase of 15.0 percent
(e.g., CD-ROMs, magnetic tapes, and magnetic disks) per
since FY 97);10
1,000 population.
■ reference transactions in public libraries totaled 292
Library Services million, or 1.1 reference transactions per capita; and
Electronic services and Internet access ■ library visits in public libraries totaled 1.1 billion, or
Nationwide, 74 percent of public libraries provided access 4.2 library visits per capita.
to electronic services,7 and 88 percent of public libraries
Children’s services
had access to the Internet (a 9-percentage-point increase
since FY 97).8 Almost 72 percent of all public libraries Nationwide, circulation of children’s materials was 612
made the Internet available to patrons directly or through a million, or 36 percent of total circulation. Attendance at
staff intermediary, almost 9 percent of public libraries made children’s programs was 46 million.
the Internet available to patrons through a staff intermedi-
ary only, and almost 8 percent of public libraries made the
References
Internet available only to library staff. Ninety-three percent9 Chute, A., and Kroe, P.E. (2000). Public Libraries in the United
States: FY 1997 (NCES 2000–316). U.S. Department of Educa-
of the unduplicated population of legal service areas had tion, Washington, DC: National Center for Education Statistics.
access to the Internet through their local public library.
National Center for Education Statistics. (2001). Data File: Public
Libraries Survey: Fiscal Year 1998. U.S. Department of Education.
Circulation Available: http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=2001377
Total nationwide circulation of public library materials
was 1.7 billion, or 6.6 materials circulated per capita. The 10
See the previous edition of this E.D. Tabs report, Public Libraries in the United States:
highest circulation per capita in the 50 states and the FY 1997 (Chute and Kroe 2000), table 4.

District of Columbia was 12.5 and the lowest was 2.7.

7
Access to electronic services refers to electronic services (e.g., bibliographic and full-
text databases, multimedia products) provided by the library due to subscription, Data source: The NCES FY 1998 Public Libraries Survey (PLS).
lease, license, consortial membership or agreement. It includes full-text serial
subscriptions and electronic databases received by the library or an organization For technical information, see the complete report:
associated with the library. Chute, A., and Kroe, P.E. (2001). Public Libraries in the United States:
8
See the previous edition of this E.D. Tabs report, Public Libraries in the United States: Fiscal Year 1998 (NCES 2001–307).
FY 1997 (Chute and Kroe 2000), table 6. Author affiliations: A. Chute and P.E. Kroe, NCES.
9
This percentage was derived by summing the unduplicated population of legal For questions about content, contact Adrienne Chute
service areas for (1) all public libraries in which the Internet was used by patrons (adrienne.chute@ed.gov).
through a staff intermediary only and (2) all public libraries in which the Internet was
used by patrons either directly or through a staff intermediary, and then dividing the To obtain the complete report (NCES 2001–307), visit the NCES
total by the unduplicated population of legal service areas in the United States. (Also Web Site (http://nces.ed.gov).
see Data File: Public Libraries Survey: Fiscal Year 1998, available on the NCES Web Site.)

E D U C AT I O N S TAT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , F A L L 2 0 0 1 101
Libraries

Academic Libraries
Academic Libraries: 1998
——————————————————————————————————
Margaret W. Cahalan and Natalie M. Justh
This article was originally published as the Introduction and Highlights of the E.D. Tabs report of the same name. The universe data are from the NCES
Integrated Postsecondary Education Data System “Academic Libraries Survey” (IPEDS-L).

This report presents detailed tabulations for the 1998 Interlibrary loans
“Academic Libraries Survey” (ALS). In 1998, the survey was In FY 98, academic libraries provided a total of about 9.2
conducted as part of the Integrated Postsecondary Educa- million interlibrary loans to other libraries (both academic
tion Data System (IPEDS) of the U.S. Department of libraries and other types of libraries) and received about
Education’s National Center for Education Statistics 7.7 million loans.
(NCES).1 ALS has been conducted by NCES since 1966 at
irregular intervals. Since 1990, it has been conducted on a Public service hours
2-year cycle. Overall, the largest percentage of academic libraries
(42 percent) reported having 60–79 hours of public service
The data in this report cover academic libraries in 2-year per typical week. However, 38 percent provided 80 or more
and 4-year degree-granting postsecondary institutions in service hours per typical week during the academic year.
the United States. The tables summarize library services The percentage of institutions providing 80 or more public
(including electronic services), library staff, library collec- service hours ranged from 6 percent in less-than-4-year
tions, and library expenditures for libraries in degree- institutions to 75 percent in doctorate-granting institutions.
granting postsecondary institutions in the 50 states and the Twenty libraries reported that they were open 168 hours a
District of Columbia. Library staff data are for fall 1998. week, or 24 hours 7 days a week.
Operating expenditures and library collections data are for
fiscal year (FY) 1998. Data on library circulation and Electronic services
interlibrary loans are for FY 98, and data on other library In FY 98, 84 percent of degree-granting postsecondary
services are for a typical week in the fall of 1998. FY 98 is institutions with an academic library had access from
defined as any 12-month period between July 1, 1997, and within the library to an electronic catalog of the library’s
September 30, 1998, that corresponds to the institution’s holdings, 95 percent had Internet access within the library,
fiscal year. and 54 percent had library reference service by e-mail both
within the library and elsewhere on campus. Just under
Number of Academic Libraries one-third (30 percent) had electronic document delivery by
In fall 1998, 3,658 of the 4,141 2-year and 4-year degree- the library to a patron’s account or address from within the
granting postsecondary institutions in the United States library. Ninety-two percent had instruction by library staff
reported that they had their own academic library. Of these on the use of Internet resources within the library.
3,658 academic libraries, 97 percent responded to the
survey. In FY 98, 44 percent of academic libraries had technology
within the library to assist persons with disabilities and
Services 34 percent had access to this service from elsewhere on
Circulation campus. Sixty-five percent provided services to distance
In FY 98, general collection circulation transactions in the education students.
nation’s academic libraries at degree-granting postsecondary
institutions totaled 175.4 million. Reserve collection Almost three-fourths (71 percent) of academic libraries had
circulation transactions totaled 40.7 million. computers not dedicated to library functions for patron use
inside the library. Fewer institutions with an academic
library (12 percent) had video/desktop conferencing by or
1
IPEDS is the U.S. Department of Education’s vehicle for collecting data from all for the library accessible within the library, and 19 percent
postsecondary institutions in the United States. Other topics included within IPEDS
are institutional characteristics, fall enrollment, completions, finance, faculty salaries, had access from elsewhere on campus. Seventeen percent
and fall staff. From 1988 to 1998, ALS was a part of the IPEDS system. Beginning in the had satellite broadcasting by or for the library accessible
year 2000, ALS began collecting data independent from the IPEDS data collection;
however, data from ALS can still be linked to IPEDS data. IPEDS also provides the frame within the library, and 23 percent had access from elsewhere
used for ALS.
on campus.

102 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
 Academic Libraries: 1998

Other services were other paid staff; 270 (less than one-half of 1 percent)
■ Taken together, academic libraries reported a gate were contributed services staff;3 and 28,373 (29 percent)
count of about 16.2 million visitors per typical week were student assistants.
(about 1.6 visits per full-time-equivalent [FTE]
student enrolled).2 Excluding student assistants, the institutional median
number of academic library FTE staff per 1,000 FTE
■ About 2.1 million reference transactions were
students was 5.6. The median ranged from 3.6 in less-than-
reported in a typical week.
4-year institutions to 9.1 in doctorate-granting institutions.
■ Over FY 98, about 438,000 presentations to groups
serving about 7.4 million persons were reported. Expenditures
In FY 98, expenditures for libraries at the 3,658 degree-
Collections
granting postsecondary institutions totaled $4.6 billion. The
Total number of volumes
three largest expenditure items for all academic libraries
Taken together, the nation’s 3,658 academic libraries at were salaries and wages, $2.3 billion (50 percent); current
degree-granting postsecondary institutions held a total of paper and electronic serial subscription expenditures,
878.9 million paper volumes (books, bound serials, and $974.9 million (21 percent); and paper books and bound
government documents) at the end of FY 98. serials, $514.0 million (11 percent).

Of the total paper volumes held at the end of the year, The libraries of the 570 doctorate-granting institutions
43 percent (376.0 million) were held at the 125 institutions (16 percent of the total institutions) accounted for $2.9
categorized under the Carnegie Classification as Research I billion, or 64 percent of the total expenditure dollars at all
or Research II institutions. About 55 percent of the volumes academic libraries at degree-granting postsecondary
were at those institutions classified as either Research or institutions.
Doctoral in the Carnegie Classification.
In FY 98, the median total for operating expenditures per
Median volumes per FTE student FTE student was $301.25, and the median for information
The median number of paper volumes held per FTE student resource expenditures per FTE student was $84.98.
was 53.7 volumes. Median volumes held ranged from 18.5
per FTE student in less-than-4-year institutions to 119.8 in 3
Contributed services staff are those, such as members of religious orders, whose
doctorate-granting institutions. services are valued by bookkeeping entries rather than by full cash transactions. They
do not include volunteers.

In FY 98, the median number of paper volumes added to

collections per FTE student was 1.5. The median number
added ranged from .7 per FTE student in less-than-4-year Data source: The NCES 1998 Integrated Postsecondary Education Data
institutions to 2.9 in doctorate-granting institutions. System “Academic Libraries Survey” (IPEDS-L:98).
For technical information, see the complete report:
Staff Cahalan, M.W., and Justh, N.M. (2001). Academic Libraries: 1998
(NCES 2001–341).
A total of 96,709 FTE staff were working in academic
Author affiliations: M.W. Cahalan and N.M. Justh, Mathematica Policy
libraries in fall 1998. Of these, 30,041 (31 percent) were Research, Inc.
librarians or other professional staff; 38,026 (39 percent) For questions about content, contact Jeffrey W. Williams
(jeffrey.williams@ed.gov).
2 To obtain the complete report (NCES 2001–341), visit the NCES
FTE enrollment is calculated by adding one-third of part-time enrollment to full-time
enrollment. Enrollment data are from the 1997–98 IPEDS “Fall Enrollment Survey.”
Web Site (http://nces.ed.gov).
Calculations are based on a total FTE enrollment of 10,216,653.

E D U C AT I O N S TAT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , F A L L 2 0 0 1 103
104 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
 C R O S S C U T T I N G S TAT I S T I C S

Educational Achievement and Black-White Inequality

Jonathan Jacobson, Cara Olsen, Jennifer King Rice, Stephen Sweetland,
and John Ralph .................................................................................................... 105
Features of Occupational Programs at the Secondary and Postsecondary
Education Levels
Richard P. Phelps, Basmat Parsad, Elizabeth Farris, and Lisa Hudson ................ 114

Educational Achievement
Educational Achievement and Black-White Inequality
——————————————————————————————————
Jonathan Jacobson, Cara Olsen, Jennifer King Rice, Stephen Sweetland, and John Ralph
This article was originally published as the Executive Summary of the Statistical Analysis Report of the same name. The sample survey data are from
several NCES, Bureau of Labor Statistics, and Department of Education surveys, which are listed at the end of this article.

Major Findings ■ For young adults with similar levels of prior educa-
The study reported here explored the relationship between tional achievement, Blacks were more likely to attend
Black-White differences in educational achievement and college than Whites. Among college attendees with
Black-White differences in a variety of educational and similar levels of prior educational achievement,
economic outcomes. Comparisons were made first between Blacks’ college completion rates were as high as, or
overall average outcomes for Blacks and Whites and then higher than, the college completion rates of Whites.
between average outcomes for Blacks and Whites with ■ Throughout elementary and secondary school, Blacks
similar levels of prior educational achievement.1 The major scored lower, overall, on mathematics and reading
findings of the study reveal that tests than Whites. Even for children with similar test
■ For women with similar levels of prior educational scores one or two grades earlier, Blacks generally
achievement, Blacks earned as much, or more, per scored lower in mathematics and reading than Whites.
year as Whites. For men with similar levels of prior The Black-White mathematics gap differed in size
educational achievement, Black-White gaps in annual across grades, in a manner consistent with, but not
earnings were at least two-fifths smaller than Black- necessarily demonstrating, a narrowing of the gap
White gaps for men as a whole. Black-White dispari- during elementary school, followed by a widening of
ties in employment were, for young adults with the gap during junior high school and little change
similar levels of prior educational achievement, at during senior high school. The Black-White reading
least one-half smaller than Black-White employment gap also differed in size across grades, but not in an
disparities for young adults as a whole. entirely consistent manner; it grew wider between
grades within two elementary school cohorts, but
1
Comparisons between individuals with similar levels of prior educational achieve- was narrower for cohorts observed in grades 9
ment involved (1) Whites as a whole, and (2) Blacks with prior educational achievement and 12 than for a cohort observed in grade 2.
similar to that for Whites.

E D U C AT I O N S TAT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , F A L L 2 0 0 1 105
Crosscutting Statistics

In general, the findings show that, for children and young score lower on NAEP than Whites (Snyder, Hoffman, and
adults with similar levels of prior educational achievement, Geddes 1997). And, despite the convergence in high school
the educational and economic performance of Blacks completion rates of Blacks and Whites, the Black-White gap
relative to Whites was substantially greater than the in 4-year college completion rates of high school graduates
performance of Blacks relative to Whites as a whole. While 25–29 years old has increased slightly over the past quarter-
Blacks have lower levels of educational achievement, century. Between 1975 and 1998, this gap increased from
educational attainment, and earnings than Whites, these 13 to 17 percentage points (National Center for Education
disparities are frequently smaller, and are sometimes Statistics 1999).
entirely absent, for individuals with similar levels of prior
educational achievement. Factors other than differences in In recent years, Black-White disparities have also per-
prior educational achievement may contribute to Black- sisted—and have sometimes grown larger—for labor market
White gaps in achievement, employment, and earnings; outcomes such as labor force participation,2 unemploy-
nonetheless, Blacks’ relative educational achievement ment,3 and hourly wages. Between 1973 and 1993, Black-
during elementary and secondary school appeared to be White differences in the labor force participation rates of
highly correlated with their relative success in the academy 25- through 34-year-olds widened by 3.4 percentage points
and the economy. for men and 19 percentage points for women, and the
corresponding Black-White gaps in unemployment rates
Note: This study does not attempt to isolate the causal widened by about 2.4 percentage points for both men and
relationship between educational achievement and subse- women. Over the same time period, the Black-White gap
quent educational and economic outcomes. Rather, using in hourly wages narrowed by one-third for 25- through
educational achievement as an indicator for the cognitive 34-year-old men, but more than doubled for 25- through
backgrounds of children and young adults, it investigates 34-year-old women (Bernstein 1995).
the extent to which Black-White disparities are present for
individuals with similar levels of prior educational achieve- Recent studies have revealed a strong relationship between
ment. The comparison of outcomes for Blacks and Whites differences in prior educational achievement and Black-
with similar levels of educational achievement does not White disparities in college attendance and earnings. With
indicate what outcomes for all Blacks would be if their Black-White disparities remaining in both educational and
average achievement were raised to the level for Whites. economic outcomes, it is important to understand the
Educational achievement differences are correlated with relationship between educational achievement during
many other possible sources of Black-White disparities, elementary and secondary school and subsequent academic
some measured in survey data, others unmeasured. and labor market performance.

Background The Present Study

Over the past quarter-century, Black Americans have made The study documented in this report used multiple datasets to
important gains in narrowing the gaps in educational and confirm and extend earlier findings. Specifically, this study
economic performance between themselves and Whites. included three sets of analyses designed to investigate the
Between 1973 and 1996, for example, average scores of relationship between Black-White differences in prior educa-
Black 17-year-olds on the mathematics portion of the tional achievement and a variety of subsequent outcomes:4
National Assessment of Educational Progress (NAEP) grew ■ The first set of analyses considered the extent to
by 6 percent, while average scores of White 17-year-olds which Black-White differences in labor market
remained about the same (Snyder, Hoffman, and Geddes outcomes were present for young adults as a whole
1997). Black-White gaps in NAEP reading scores also and for young adults with similar levels of prior
narrowed over this period. Further, in 1974, the high school educational achievement.
dropout rate for 15- through 24-year-old Blacks was twice
the corresponding rate for Whites; but by 1997, Blacks and 2
The labor force participation rate is defined as the percentage of noninstitutionalized
Whites in this age bracket remained in high school at similar civilians who are employed, otherwise with a job, or looking for a job.
rates (National Center for Education Statistics 1999). 3
The unemployment rate is defined as the percentage of labor force participants who
are without a job.

Despite progress in reducing Black-White gaps in math- 4

Prior educational achievement is defined as prior mathematics and/or reading
achievement. The accompanying figures indicate outcomes for all Whites, all Blacks,
ematics and reading achievement, Blacks have continued to and Blacks at Whites’ level of prior educational achievement.

106 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
 Educational Achievement and Black-White Inequality

■ The second set of analyses considered the extent to levels of prior educational achievement, Black-White gaps
which Black-White differences in educational attain- in unemployment rates were generally at least one-half
ment were present for young adults as a whole and smaller than for young adults as a whole. Among men
for young adults with similar levels of prior educa- with similar levels of prior educational achievement,
tional achievement. Black-White gaps in annual earnings were at least two-
■ The final set of analyses considered the extent to fifths smaller than for men as a whole. Black women with
which Black-White differences in mathematics and levels of prior educational achievement similar to White
reading achievement were present for children as a women earned as much as, or more than, their White
whole and for children with similar levels of prior counterparts.
educational achievement. These analyses also
Unemployment rates
considered the extent to which Black-White achieve-
ment gaps varied in size during elementary and For the samples of young adults studied, there were no
secondary school. consistent differences between Blacks and Whites in terms
of labor force participation, but Black labor force partici-
Black-White Differences in Labor Market pants were more likely to be unemployed than White labor
Outcomes force participants (figure 1). The absolute Black-White gaps
Main findings in unemployment rates ranged between 4 and 10 percentage
points, and were similar in size for men and women. These
Analyses of labor market outcomes between 1979 and
gaps were generally at least one-half smaller for young
19925 indicate that, for young adults with similar levels of
adults with similar levels of prior educational achievement
prior educational achievement, the economic performance
than for young adults as a whole.
of Blacks relative to Whites was substantially greater than
for young adults as a whole. For young adults with similar
Annual earnings
For the samples of young adults studied, Blacks generally
5
The analyses of labor market outcomes focused on four samples of young adults: earned less per year than Whites.6 Black-White gaps in annual
(1) young adults who were high school seniors in 1972 and who were observed 7 years
later through the National Longitudinal Study of the High School Class of 1972 (“the earnings for men ranged from 16 percent in the 1979 sample
1979 sample”); (2) young adults who were high school seniors between 1976 and 1982 to about 32 percent in the 1983–89 and 1986–92 samples
and who were observed 7 years later through the National Longitudinal Survey of Youth
(“the 1983–89 sample”); (3) young adults who were high school sophomores between (figure 2). In the 1979 and 1992 samples, Black women and
1974 and 1980 and who were observed 12 years later through the National Longitudinal
Survey of Youth (“the 1986–92 sample”); and (4) young adults who were high school
sophomores in 1980 and who were observed 12 years later through the High School
6
and Beyond Longitudinal Study (“the 1992 sample”). High school sophomores and The pattern of Black-White gaps in hourly wages—reported in every sample except
seniors were generally identified as of the spring of each year. Educational achievement the 1992 High School and Beyond sample—was generally similar to the pattern of
was measured in 1972 for the 1979 sample and in 1980 for the other samples. gaps in annual earnings.

Figure 1.—Unemployment rates for Black and White young adults: 1979–92

Percent
20 20 All Blacks/all Whites
18 18
(B)
16 16 Blacks with prior educational
14 14 achievement similar to Whites’
12 (B) 12
10 10
(B)
8 (B) (W) 8
6 (W) 6
4 (W) 4
(W)
2 2
0 0
1979 1983–89 1986–92 1992
Year(s)

NOTE: Samples restricted to civilian labor force participants; higher end of gray range is for Blacks.
SOURCE: U.S. Department of Education, National Center for Education Statistics: National Longitudinal Study of the High School Class of 1972, “Fourth Follow-up” (NLS:72/79)
(1979 sample, 7 years after grade 12), High School and Beyond Longitudinal Study of 1980 Sophomores (HS&B-So:80/92) (1992 sample, 12 years after grade 10); and U.S.
Department of Labor, Bureau of Labor Statistics, National Longitudinal Survey of Youth (1983–89 sample, 7 years after grade 12, and 1986–92 sample, 12 years after grade 10).

E D U C AT I O N S TAT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , F A L L 2 0 0 1 107
Crosscutting Statistics

Figure 2.—Average annual earnings for Black and White young men and women: 1979–92

Thousands Men Women

of 1992 dollars All Whites/all Blacks
(W)
22 (W) 22
Blacks with prior educational
(W) achievement similar to Whites’
20 20

18 18
(B)
(B)
16 (W) 16
(B)
(W)
(B)
14 14
(W)
(B) (W)
12 12
(W)
(B)
10 (B) 10

8 (B) 8

0 0
1979 1983–89 1986–92 1992 1979 1983–89 1986–92 1992
Year(s)

NOTE: Samples restricted to civilians reporting some earnings; higher end of gray range is for Whites, except in the case of women in 1992.
SOURCE: U.S. Department of Education, National Center for Education Statistics: National Longitudinal Study of the High School Class of 1972,“Fourth Follow-up” (NLS:72/79)
(1979 sample, 7 years after grade 12), High School and Beyond Longitudinal Study of 1980 Sophomores (HS&B-So:80/92) (1992 sample, 12 years after grade 10); and U.S. Department
of Labor, Bureau of Labor Statistics, National Longitudinal Survey of Youth (1983–89 sample, 7 years after grade 12, and 1986–92 sample, 12 years after grade 10).

White women had similar earnings. In the 1986–92 sample, remainder of these gaps include a relative shortage of jobs
the Black-White earnings gap for women was about two- in areas where Blacks live, fewer job networks for Blacks,
thirds smaller than the corresponding gap for men. and the existence of labor market discrimination against
Blacks. Unmeasured skill differences between labor force
For men with similar levels of prior educational achievement, participants of different racial backgrounds may also
the Black-White gap in annual earnings was measured contribute to the remaining Black-White disparities in
imprecisely in the 1979 sample, such that it was distinguish- employment and men’s earnings.
able neither from zero, nor from the gap for men as a whole.
In the 1983–89, 1986–92, and 1992 samples, the Black- Black-White Differences in Educational
White earnings gap for men with similar educational achieve- Attainment
ment was over two-fifths smaller than for men as a whole. Main findings
Blacks having similar levels of prior educational achieve-
For women with similar levels of prior educational achieve- ment as Whites had received a high school diploma or
ment, Blacks earned 12 percent more per year than Whites in General Educational Development (GED) certificate at an
the 1979 sample, 22 percent more per year than Whites in equal or higher rate than Whites. For young adults with
the 1992 sample, and about the same as Whites in the similar levels of prior educational achievement in the
1983–89 and 1986–92 samples. same four samples observed between 1979 and 1992, the
postsecondary educational attainment of Blacks was as high
Additional sources of disparities in labor market
as, or higher than, that of Whites.7 For such young adults,
outcomes
Since differences in educational achievement can predict
only a portion of Black-White differences in employment 7
The analyses of educational attainment outcomes focused on the same four samples
and men’s earnings, other factors must contribute to racial of young adults studied for the analyses of labor market outcomes. Black-White
differences in postsecondary educational attainment were generally similar for males
disparities in these outcomes. Possible reasons for the and females.

108 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
 Educational Achievement and Black-White Inequality

the college attendance rate was higher for Blacks than for lower rate of college attendance in the 1992 sample.9 In
Whites. Further, Black college attendees with levels of contrast, for young adults with similar levels of prior
prior educational achievement similar to those for Whites educational achievement, Blacks had a 6- to 17-percentage-
completed college at rates similar to, or higher than, the point higher rate of college attendance than Whites.
rates for White college attendees.
College completion rates
High school/GED completion rates For young adults who had attended at least some college,
Black-White differences in high school/GED completion college completion rates10 were consistently lower for
rates could be compared for every sample of young adults Blacks than for Whites (figure 4). The Black-White gap in
except the 1979 sample.8 A Black-White gap in high school/ college completion ranged from about 13 percentage points
GED completion rates (in the range of 2 to 8 percentage in the 1979 sample to about 19 percentage points in the
points) was evident in the 1983–89, 1986–92, and 1992 other three samples. Again, in contrast, among college
samples. For young adults with similar levels of prior attendees with similar levels of prior educational achieve-
educational achievement, Blacks received high school ment, the college completion rate of Blacks equaled or
diplomas or GED certificates at a rate similar to or higher exceeded that of Whites.
than Whites.
Black-White Differences in Educational
College attendance rates Achievement
Young adults observed between 1979 and 1992 generally Main findings
showed a Black-White gap in college attendance rates The analyses of educational achievement compared math-
(figure 3). Compared with Whites, Blacks had a 4- to ematics and reading levels of Black and White children at
7-percentage-point lower rate of college attendance in the
1979 and 1983–89 samples, and a 10-percentage-point 9
In the 1986–92 sample, the Black-White difference in college attendance was
significant for men (8 percent) but not for women or for young adults as a whole.
8 10
High school/GED completion status was ambiguous for individuals in the 1979 College completion is defined here as completion of at least 4 years of college or an
sample. equivalent bachelor’s degree.

Figure 3.—College attendance rates for Black and White young adults: 1979–92

Percent
82 82 All Whites/all Blacks
80 80
78 78 Blacks with prior educational
76 76 achievement similar to Whites’
74 74
72 72
70 70
68 68
66 66
64 (W) 64
(W) (W)
62 62
60 60
(B) (B)
58 58
56 (W) 56
54 (B) 54
52 52
50 (B) 50
48 48
0 0
1979 1983–89 1986–92 1992
Year(s)

NOTE: Samples restricted to former 12th-graders or high school graduates (as indicated below); higher end of gray range is for Whites.
SOURCE: U.S. Department of Education, National Center for Education Statistics: National Longitudinal Study of the High School Class of 1972, “Fourth Follow-up” (NLS:72/79)
(1979 sample, 7 years after grade 12), High School and Beyond Longitudinal Study of 1980 Sophomores (HS&B-So:80/92) (1992 sample, high school graduates 12 years after
grade 10); and U.S. Department of Labor, Bureau of Labor Statistics, National Longitudinal Survey of Youth (1983–89 sample, 7 years after grade 12, and 1986–92 sample,
high school graduates 12 years after grade 10).

E D U C AT I O N S TAT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , F A L L 2 0 0 1 109
Crosscutting Statistics

Figure 4.—College completion rates for Black and White young adults: 1979–92

Percent
54 54 All Whites/all Blacks
52 52
50 50 Blacks with prior educational
48 48 achievement similar to Whites’
46 46
44 (W) (W) 44
(W)
42 (W) 42
40 40
38 38
36 36
34 34
32 32
30 (B) 30
28 28
26 (B) 26
24 (B) (B) 24
22 22
0 0
1979 1983–89 1986–92 1992
Year(s)

NOTE: College completion is defined as completion of 4 years of college or the equivalent. Samples restricted to persons who have attended at least some college; higher
end of gray range is for Whites.
SOURCE: U.S. Department of Education, National Center for Education Statistics: National Longitudinal Study of the High School Class of 1972, “Fourth Follow-up” (NLS:72/79)
(1979 sample, 7 years after grade 12), High School and Beyond Longitudinal Study of 1980 Sophomores (HS&B-So:80/92) (1992 sample, high school graduates 12 years after
grade 10); and U.S. Department of Labor, Bureau of Labor Statistics, National Longitudinal Survey of Youth (1983–89 sample, 7 years after grade 12, and 1986–92 sample,
high school graduates 12 years after grade 10).

various points between grades 1 and 12.11 Black-White gaps grades within two elementary school cohorts, but was
in mathematics and reading achievement appeared at every narrower in cohorts observed in grades 9 and 12 than in a
grade studied. Even for children with similar levels of prior cohort observed in grade 2.
achievement one or two grades earlier,12 mathematics and
reading scores of Blacks were generally lower than the Mathematics achievement
corresponding scores of Whites. Compared with White children, Blacks scored lower on
mathematics tests at every grade level studied between
Comparisons of the size of Black-White achievement gaps grades 1 and 12 (figure 5). Black-White mathematics gaps
were possible between nearby grades within the same were usually similar in size for both boys and girls.
sample of children, as well as across different samples of
children from grades 1 to 12. The Black-White mathematics Within the same samples of children, the Black-White gap
gap differed in size across grades, in a manner consistent increased by two-fifths between grades 7 and 9, but changed
with a narrowing of the gap during elementary school, little between grades 1 and 2, grades 3 and 5, and grades 10
followed by a widening of the gap during junior high school and 12. Across different samples of children, the Black-White
and little change during senior high school. The Black- math gap was two-fifths smaller in grade 5 than in grade 2,
White reading gap also differed in size across grades, but but one-half larger in grade 9 than in grade 5, and about the
not in an entirely consistent manner; it grew wider between same size in grade 12 as in grade 9. Between the grade 2 and
grade 12 samples there was no difference in the size of the
Black-White math gap, suggesting that any narrowing of the
11
The analyses of educational achievement outcomes focused on four samples of gap between grades 2 and 5 was largely negated by the
children: (1) children between grades 1 and 2, observed from 1992 to 1993 in cohort 1
of the Chapter 1 Prospects Study; (2) children between grades 3 and 5, observed from widening of the gap between grades 5 and 9.13
1991 to 1993 in cohort 3 of the Prospects Study; (3) children between grades 7 and 9,
observed from 1991 to 1993 in cohort 7 of the Prospects Study; and (4) children
between grades 10 and 12, observed from 1990 to 1992 in the National Education 13
Longitudinal Study of 1988 Eighth-Graders. Black-White differences in educational Note that comparisons of the grade 2, grade 5, grade 9, and grade 12 gaps involve
achievement were usually similar for boys and girls. four separate samples of children, which, while generally similar in observed family
background characteristics, may differ in terms of unobserved family background and
12
Prior educational achievement was defined as the corresponding mathematics or school characteristics. For the sample of children observed between grades 10 and 12,
reading score for the earliest grade in which a sample of children was observed however, there is corroborating evidence of a widening of the Black-White mathe-
(grades 1, 3, 7, and 10, respectively). matics gap by about one-fifth between grades 8 and 10.

110 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
 Educational Achievement and Black-White Inequality

Figure 5.—Average mathematics achievement scores for Black and White children: 1990–93

8th-grade standard
deviation units All Whites/all Blacks
2 2
Blacks with prior educational
(W) achievement similar to Whites’
1 (W) 1

(W) (W) (B)

0 0
(B)
(W) (B) (B)
–1 (B) –1
(W)

–2 (B) –2

(W)
–3 –3
(B)
–4 (W) –4

–5 –5
(B)

–6 –6
1 2 3 4 5 6 7 8 9 10 11 12
Grade

NOTE: Prior mathematics achievement refers to mathematics achievement one grade earlier for the grade 2 sample and two grades earlier for the grade 5, grade 9, and grade 12
samples. Mathematics scores are normalized so the grade 8 score for children of all races has a mean of 0 and a standard deviation of 1 in the grade 9 and grade 12 samples. Higher
end of gray range is for Whites.
SOURCE: U.S. Department of Education: Chapter 1 Prospects Study (1992–93 sample of 1st- through 2nd-graders, and 1991–93 samples of 3rd- through 5th-graders and
7th- through 9th-graders), and National Center for Education Statistics, National Education Longitudinal Study of 1988 Eighth-Graders (NELS:88/92) (1990–92 sample of 10th-
through 12th-graders).

Even for children who had similar math scores one or two grade 2.14 Within the same samples of children, the Black-
grades earlier, a Black-White mathematics gap usually White reading gap increased by one-third between grades
appeared. A Black-White mathematics gap was present in 1 and 2 and by about one-fifth between grades 3 and 5,15
grade 2, even for children with similar math scores in while remaining about the same between grades 7 and 9,
grade 1; in grade 5, even for children with similar math and between grades 10 and 12. Across different samples
scores in grade 3; and in grade 9, even for children with of children, the Black-White reading gap was one-third
similar math scores in grade 7. These gaps were 59 to smaller in grade 9 than in grade 2, and two-fifths smaller
70 percent smaller than the corresponding mathematics in grade 12 than in grade 2.
gaps for children as a whole. (Black and White children
with similar math scores in grade 10 had similar math A Black-White reading gap was generally present, even for
scores in grade 12.) children with similar reading scores one or two grades
earlier. For children with similar reading scores one or
Reading achievement two grades earlier, the Black-White reading gap was 58 to
Compared with Whites, Blacks also scored lower on reading 77 percent smaller than the corresponding Black-White
tests at every grade level studied between grades 1 and 12 reading gap for children as a whole.
(figure 6). Black-White reading gaps did not differ consis-
14
Note that the comparisons of the grade 9 and 12 gaps with the grade 2 gap involve
tently for boys and girls. separate samples of children, which may differ in terms of family background and
school characteristics.
The Black-White reading gap grew wider between some 15
For the sample of children observed between grades 10 and 12, there was corrob-
orating evidence of a widening of the Black-White reading gap by about one-sixth
grades, but was narrower in grades 9 and 12 than in between grades 8 and 10.

E D U C AT I O N S TAT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , F A L L 2 0 0 1 111
Crosscutting Statistics

Figure 6.—Average reading achievement scores for Black and White children: 1990–93

8th-grade standard
deviation units All Whites/all Blacks
2 2
Blacks with prior educational
achievement similar to Whites’
1 (W) 1
(W)
(W) (W) (B)
0 0
(B)
(W) (B) (B)
–1 (W) –1
(B)

–2 (B) –2
(W)

–3 –3
(B)
(W)
–4 –4
(B)
–5 –5
1 2 3 4 5 6 7 8 9 10 11 12
Grade

NOTE: Prior reading achievement refers to reading achievement one grade earlier for the grade 2 sample and two grades earlier for the grade 5, grade 9, and grade 12 samples.
Reading scores are normalized so the grade 8 score for children of all races has a mean of 0 and a standard deviation of 1 in the grade 9 and grade 12 samples. Higher end of
gray range is for Whites.
SOURCE: U.S. Department of Education: Chapter 1 Prospects Study (1992–93 sample of 1st- through 2nd-graders, and 1991–93 samples of 3rd- through 5th-graders and
7th- through 9th-graders), and National Center for Education Statistics, National Education Longitudinal Study of 1988 Eighth-Graders (NELS:88/92) (1990–92 sample of 10th-
through 12th-graders).

While findings within the same samples of children would, Blacks generally acquired fewer reading skills than Whites,
by themselves, suggest a widening of the Black-White and usually acquired fewer mathematics skills as well.
reading gap as children progressed through school, findings These findings imply that Black-White disparities in
across different samples suggest an overall narrowing of the educational achievement can widen as students progress
Black-White reading gap between grades 2 and 9, with this through elementary or secondary school. Possible explana-
narrowing persisting through grade 12. This difference in tions for these differences in achievement growth include
findings may be consistent with the actual experiences of differences in the school or home environments of children
children as they progressed through school, or it may arise of different racial backgrounds that make it more difficult
from the use of different cohorts of children in the compari- for Blacks to acquire math or reading skills at the same pace
sons. The collection and analysis of longitudinal data as Whites.
following the same sample of children all the way from
grade 2 through grade 12 would help to further address the Conclusion
question of how the Black-White reading gap changes over The findings of this study imply that, over the past 2
the course of the school years. decades, Black-White differences in educational achieve-
ment have been strongly associated with Black-White
Additional sources of disparities in educational disparities in a variety of educational and economic out-
achievement comes. Achievement differences do not necessarily cause
On average, Blacks in grade 1 had lower mathematics and gaps in educational attainment, employment, or earnings,
reading scores than Whites, and Blacks in grade 12 also had but they reflect a set of circumstances responsible for Black-
lower mathematics and reading scores than Whites. Among White disparities in both the academy and the economy.
children with similar test scores one or two grades earlier, Addressing the contributing causes of Black-White

112 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
 Educational Achievement and Black-White Inequality

achievement differences will be important in efforts to

Data sources:
narrow Black-White gaps in educational performance, and
NCES: National Longitudinal Study of the High School Class of 1972
perhaps also in subsequent labor market outcomes. (NLS:72/79), High School and Beyond Longitudinal Study of 1980
Sophomores (HS&B-So:80/92), National Education Longitudinal Study
of 1988 Eighth-Graders (NELS:88/92), and 1992 National Adult Literacy
References Survey (NALS).
Bernstein, J. (1995). Where’s the Payoff? The Gap Between Black Other: U.S. Department of Labor, Bureau of Labor Statistics, National
Academic Progress and Economic Gains. Washington, DC: Longitudinal Survey of Youth (NLSY), 1979–92; and U.S. Department of
Economic Policy Institute. Education, Chapter 1 Prospects Study, 1991–93.
National Center for Education Statistics. (1999). The Condition For technical information, see the complete report:
of Education: 1999 (NCES 1999–022). U.S. Department of Jacobson, J., Olsen, C., Rice, J.K., Sweetland, S., and Ralph, J. (2001).
Education, National Center for Education Statistics. Washing- Educational Achievement and Black-White Inequality
(NCES 2001–061).
ton, DC: U.S. Government Printing Office.
Author affiliations: J. Jacobson, C. Olsen, J.K. Rice, and S. Sweetland,
Snyder, T.D., Hoffman, C.M., and Geddes, C.M. (1997). Digest of Mathematica Policy Research, Inc.; J. Ralph, NCES.
Education Statistics 1997 (NCES 98–015). U.S. Department of For questions about content, contact John Ralph (john.ralph@ed.gov).
Education, National Center for Education Statistics. Washing-
To obtain the complete report (NCES 2001–061), call the toll-free
ton, DC: U.S. Government Printing Office. ED Pubs number (877–433–7827), visit the NCES Web Site
(http://nces.ed.gov), or contact GPO (202–512–1800).

E D U C AT I O N S TAT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , F A L L 2 0 0 1 113
Crosscutting Statistics

Occupational Programs
Features of Occupational Programs at the Secondary and Postsecondary
Education Levels
——————————————————————————————————
Richard P. Phelps, Basmat Parsad, Elizabeth Farris, and Lisa Hudson
This article was originally published as the Executive Summary of the Statistical Analysis Report of the same name. The sample survey data are from
two NCES surveys on occupational programs, conducted through the Fast Response Survey System (FRSS) and the Postsecondary Education Quick
Information System (PEQIS).

This report presents data collected from two surveys con- Ninety percent of less-than-4-year postsecondary insti-
ducted in spring 1999: “Survey on Vocational Programs in tutions offered at least one of the listed occupational
Secondary Schools” and “Survey on Occupational Programs programs. About half of the institutions offered 1 to 5
in Postsecondary Education Institutions.” The surveys were programs, another 11 percent offered 6 to 10 programs, and
conducted to provide the U.S. Department of Education’s an additional 27 percent offered more than 10 programs. A
Office of Vocational and Adult Education (OVAE) with similar percentage of 2-year and less-than-2-year institu-
national estimates on occupational program activities. tions offered at least one listed occupational program;
91 percent of less-than-2-year institutions offered at least
The secondary school survey was conducted through the one of the listed programs, compared to 88 percent of
National Center for Education Statistics (NCES) Fast 2-year institutions. However, 2-year institutions offered
Response Survey System (FRSS), and the postsecondary more of the listed occupational programs; for example,
survey was conducted through the NCES Postsecondary 43 percent of 2-year institutions compared with 5 percent of
Education Quick Information System (PEQIS). The FRSS less-than-2-year institutions offered more than 10 programs.
survey was administered to public secondary schools that With one exception (cosmetology), each specific occupa-
include grades 11 and 12; respondents were asked about tional program was more common among 2-year institu-
program activities for 28 selected occupations within tions than among less-than-2-year institutions.
6 broad occupational areas. The PEQIS survey was adminis-
tered to less-than-4-year postsecondary institutions, and Among the public secondary schools and less-than-4-year
respondents were asked to report on program activities for postsecondary institutions that offered at least one listed
32 selected occupations in the same 6 occupational areas. occupational program, some broad program areas and
Survey findings are presented by school type (comprehen- some specific programs were more popular than others. A
sive, vocational) for the FRSS survey, and by level of majority of these public secondary schools offered at least
institution (2-year, less-than-2-year) for the PEQIS survey. one program in two of the six broad program areas—
Most findings are based on schools and institutions that business and marketing (85 percent) and technical occupa-
offered at least one of the listed occupational programs. tions (60 percent). About half of these schools offered at
least one program in each of the other broad program
Program Offerings areas—mechanical occupations, the building trades, health/
Overall, a majority of all public secondary schools offered at life sciences, and service occupations. Among less-than-
least one of the listed occupational programs: 35 percent of 4-year postsecondary institutions offering any listed
the schools offered 1 to 5 programs, 18 percent offered 6 to program, a majority offered at least one program in three
10 programs, and another 13 percent offered more than of the six broad program areas—service occupations
10 programs. However, about one-third of the schools did (64 percent), health/life sciences (61 percent), and business
not offer any of these programs. As one might expect, and marketing (60 percent). About half of these institutions
vocational schools were more likely than comprehensive offered programs for technical occupations, and fewer than
high schools to offer the listed occupational programs; half offered at least one program in mechanical occupations
98 percent of vocational schools offered at least one listed and the building trades.
program, compared to 63 percent of comprehensive
schools. On average, vocational schools also offered more Ensuring the Teaching of Relevant Job Skills
occupational programs than did comprehensive schools; for Educators responsible for occupationally specific courses
example, 44 percent of vocational schools compared with typically attempt to ensure that the content of their courses
9 percent of comprehensive schools offered more than 10 of relates well to the occupations for which they prepare
the listed occupational programs (figure A). students. Various procedures exist to ensure a match

114 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
 Features of Occupational Programs at the Secondary and Postsecondary Education Levels

Figure A.—Percentage distribution of public secondary schools and less-than-4-year postsecondary institutions that offer
various numbers of occupational programs, by type of school or institution: 1999

All public secondary 0

34 35 18 13
1 to 5

Vocational public 2 6 to 10
15 39 44
secondary
More than 10

Comprehensive 37 38 16 9
public secondary

All less-than-4-year 10 52 11 27
postsecondary

2-year 12 30 15 43
postsecondary

Less-than-2-year 9 81 5 5
postsecondary

0 10 20 30 40 50 60 70 80 90 100
Percent

NOTE: Estimates are based on public secondary schools with 11th and 12th grades and on 2-year and less-than-2-year postsecondary
institutions with Title IV eligibility.
SOURCE: U.S. Department of Education, National Center for Education Statistics: Fast Response Survey System, “Survey on Vocational Programs
in Secondary Schools,” FRSS 72, 1999; and Postsecondary Education Quick Information System (PEQIS), “Survey on Occupational Programs in
Postsecondary Education Institutions,” 1999. (Originally published as figure 1 on p. 5 of the complete report from which this article is excerpted.)

between course content and occupational skill require- least one listed occupational program by about four-fifths of
ments, five of which were included in the surveys. For less-than-4-year postsecondary institutions that offered at
public secondary schools, the five procedures listed were least one of the listed occupational programs. About half of
industry advisory committees, surveys of employers’ skill these institutions used mechanisms for faculty to get recent
needs, follow-up surveys of graduates, student work work experience.
experience (e.g., internships), and faculty externships
(occupational work experience). Except for faculty Skill Competency Lists
externships, each of these procedures was used by at least To examine the use of skill competencies in occupational
two-thirds of all public secondary schools that offered at programs, respondents in both surveys were asked whether
least one of the listed occupational programs. About half of skill competency lists had been developed or adopted for
these schools used faculty externships to ensure that each listed occupational program. Most public secondary
courses teach appropriate job skills. For less-than-4-year schools with one or more of the listed occupational programs
postsecondary institutions, the five listed procedures were had developed or adopted skill competency lists for their
industry advisory committees, surveys of employers’ skill programs; 78 percent of these schools had developed or
needs, follow-up surveys of graduates, mechanisms for adopted skill competencies for all of their offered programs
faculty to get recent work experience, and periodic internal and 95 percent had developed or adopted skill competencies
reviews. Except for mechanisms for faculty to get recent for at least one program. As with secondary schools, a large
work experience, each of these procedures was used in at proportion of 2-year and less-than-2-year postsecondary

E D U C AT I O N S TAT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , F A L L 2 0 0 1 115
Crosscutting Statistics

institutions that offered one or more of the listed occupa- Credentialing Processes
tional programs reported that skill competency lists had been Occupational programs are sometimes linked to a cre-
developed or adopted for their programs; 77 percent of these dentialing process, through which students are awarded
institutions had developed or adopted skill competencies official documentation that they have completed a pro-
for all of their programs and 93 percent had developed or gram and/or passed a skills test. At the secondary level,
adopted skill competencies for at least one program. potential credentials (other than the high school diploma)
are state or industry regulatory exams (resulting in
Secondary schools and postsecondary institutions also were registrations, licenses, or certifications) and occupational
asked to indicate the extent of educator and industry input skill certificates. The FRSS survey asked whether each
in skill competency development—that is, whether the skill occupational program prepared students to earn either of
competency lists were developed or adopted exclusively by these credentials. Seven percent of public secondary
individual course instructors or group(s) of educators, schools with listed occupational programs prepared
primarily by educators with industry input, with about students in all of their programs to take a state or industry
equal educator and industry input, or primarily or exclu- regulatory exam (leading to registration, licensing, or
sively by industry. Skill competency lists for at least one certification), while 41 percent prepared students in at
program were developed or adopted exclusively by educa- least one of their programs to do so (figure B). Thirty-one
tors, without industry involvement, in about one-third of all percent of public secondary schools with listed occupa-
public secondary schools that offered at least one listed tional programs prepared students in all of their programs
occupational program. About half of these schools reported to earn an occupational skill certificate, whereas
a minor level of industry involvement in the development or 55 percent prepared students in at least one of their
adoption of skill competency lists for at least one listed programs to do so.
program, fewer (34 percent) reported equal industry and
educator involvement, and fewer still (6 percent) reported The PEQIS survey asked less-than-4-year postsecondary
primary or exclusive industry involvement. Industry seemed institutions whether their occupational programs prepared
to have a comparable level of involvement in developing or students to earn various types of educational or occupa-
adopting skill competency lists at the postsecondary level. tional credentials. First, the survey asked about two
Almost half of less-than-4-year institutions with one or standard academic credentials—associate’s degrees and
more of the listed occupational programs reported a minor institutional certificates/diplomas. The survey also asked
level of industry input for at least one program, 36 percent about regulatory credentials—state registrations, licenses,
involved educators and industry equally, and 8 percent used or certificates—and two types of credentials offered by
primarily or exclusively industry input. industry, associations, or unions—industry/trade certifi-
cates or diplomas, and company certificates (e.g., Cisco
Defining Vocational Program Completers Certified Internetwork Expert). About half of less-than-
The FRSS survey asked secondary schools what criteria, if 4-year postsecondary institutions that offered at least one
any, they used to determine whether a student is a “voca- listed occupational program offered institutional certifi-
tional program completer.” The criteria listed were an end- cates/diplomas in all of their programs, and 87 percent
of-program exam (not a course or graduation exam), offered this type of credential for at least one of their
passage of specific vocational courses, a minimum grade- programs. Next most common were associate’s degrees
point average in the program, and passage of specific and state-awarded regulatory credentials (registrations,
academic courses other than graduation requirements. The licenses, or certificates), each offered by about half of
majority of public secondary schools with listed occupa- these institutions for at least one of their programs.
tional programs used some criteria to determine whether a Industry/trade certificates or diplomas were available for at
student was a program completer in at least one of their least one program at about one-third of these institutions,
occupational programs (89 percent) and in all of their and company certificates were offered at about one-fifth of
programs (77 percent). The most commonly used voca- these institutions.
tional completer criterion was the passage of specific
vocational courses; 75 percent of public secondary schools Relationships Among Program Characteristics
with listed occupational programs used this criterion in at Most program characteristics cited in the FRSS survey,
least one program, while only 17 to 30 percent used each of such as offering skill certificates or defining vocational
the remaining criteria in at least one program. program completion, represent program quality-control

116 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
 Features of Occupational Programs at the Secondary and Postsecondary Education Levels

structures. These quality-control structures are often

Data sources: NCES Fast Response Survey System, “Survey on
related to each other. That is, programs that used one Vocational Programs in Secondary Schools,” FRSS 72, 1999; and
quality-control structure often use another as well. NCES Postsecondary Education Quick Information System (PEQIS),
“Survey on Occupational Programs in Postsecondary Education
Looking specifically at programs that offered skill certifi- Institutions,” 1999.
cates, these programs were found to be more likely than For technical information, see the complete report:
those that did not offer skill certificates to use skill Phelps, R.P., Parsad, B., Farris, E., and Hudson, L. (2001). Features of
competency lists, to have industry input in the develop- Occupational Programs at the Secondary and Postsecondary
Education Levels (NCES 2001–018).
ment or adoption of their skill competency lists, and to
Author affiliations: R.P. Phelps, B. Parsad, and E. Farris, Westat;
define program completers. In contrast, programs that L. Hudson, NCES.
identified program completers were no more likely than For questions about content, contact Bernie Greene
programs that did not identify program completers to use (bernard.greene@ed.gov).
skill competency lists, but they were more likely to involve To obtain the complete report (NCES 2001–018), call the toll-free
ED Pubs number (877–433–7827), visit the NCES Web Site
industry in the development or adoption of competency (http://nces.ed.gov), or contact GPO (202–512–1800).
lists and to offer skill certificates.

Figure B.—Percent of public secondary schools offering listed occupational programs that prepare students to earn an
occupational skill certificate or to take a state or industry regulatory exam, by school type: 1999

Percent
100 All schools

90 Vocational schools
Skill certificate
81 Comprehensive schools
80

70 67
State or industry
regulatory exam
60
55
52
50
50
41
40 36
31
30 27

20

10 7 7 6

0
At least one listed All listed programs At least one listed program All listed programs
program prepares students prepare students prepares students prepare students

NOTE: Data are presented for public secondary schools that offer one or more of the listed occupational programs. Estimates are based on public
secondary schools with 11th and 12th grades.
SOURCE: U.S. Department of Education, National Center for Education Statistics: Fast Response Survey System, “Survey on Vocational Programs
in Secondary Schools,” FRSS 72, 1999; and Postsecondary Education Quick Information System (PEQIS), “Survey on Occupational Programs in
Postsecondary Education Institutions,” 1999. (Based on tables 10 and 11 on pp. 54 and 56 of the complete report from which this article is
excerpted.)

E D U C AT I O N S TAT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , F A L L 2 0 0 1 117
118 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
 METHODOLOGY

Community College Transfer Rates to 4-Year Institutions Using Alternative

Definitions of Transfer
Ellen M. Bradburn and David G. Hurst ............................................................... 119
A Classification System for 2-Year Postsecondary Institutions
Ronald A. Phipps, Jessica M. Shedd, and Jamie P. Merisotis ................................ 126

Transfer Rates
Community College Transfer Rates to 4-Year Institutions
Using Alternative Definitions of Transfer
——————————————————————————————————
Ellen M. Bradburn and David G. Hurst
This article was originally published as the Executive Summary of the Research and Development Report of the same name. The sample survey data are
from the Beginning Postsecondary Students Longitudinal Study (BPS).

offerings and degree programs of many community colleges

Research and Development Reports are intended to
can accommodate diverse student interests and goals,
■ share studies and research that are developmen- preparing students to transfer to a 4-year college remains a
tal in nature; central characteristic of community colleges (Brint and
■ share results of studies that are on the cutting Karabel 1989). This preparation is key to the community
edge of methodological developments; and college’s role in higher education because it affirms the
■ participate in discussions of emerging issues of community college’s claim to a collegiate, academic identity
interest to researchers. and to a role in broadening access for those historically
excluded from a college education. Moreover, transfer is a
These reports present results or discussion that do not component of most community college students’ educa-
reach definitive conclusions at this point in time, tional aspirations (Grubb 1991, pp. 195–96).
either because the data are tentative, the methodology
is new and developing, or the topic is one on which Despite, or perhaps because of, the importance of transfer
there are divergent views. Therefore, the techniques from 2- to 4-year institutions, calculating the percentage of
and inferences made from the data are tentative and community college students who transfer has proven to be
are subject to revision. somewhat problematic. At first glance, the transfer rate
seems relatively unambiguous: it is the number of students
who transfer to a 4-year college divided by the number of
Introduction potential transfer students. However, the numerator and
A large proportion of undergraduates attend community especially the denominator can both be defined in a number
colleges (public 2-year institutions) seeking a wide range of of different ways, each having a significant impact on the
services, from a place to experiment with postsecondary transfer estimate. The purpose of the present study is to
education to a structured vocational certificate or associate’s use nationally representative community college data to
degree program (Grubb 1988, 1991). Although the course examine several ways of defining the population of potential

E D U C AT I O N S T AT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , FA L L 2 0 0 1 119
Methodology

transfer students, the relationship of these definitions to denominator. They are referred to as “increasingly restric-
student background characteristics, and the relationship tive” because the total proportion of the sample that is
of each definition to the resulting transfer rate. This report included generally decreases, although the more restric-
consists of three sections. The first section describes the tive groups are not necessarily subsets of the less restrictive
data set used in the analysis and the measurement issues groups. These definitions were selected to approximate
implicated in the study of transfer. The second section measures commonly used in previous research, from
presents the selected indicators of the key concepts in the explicit student goals to behaviors often thought to indicate
study and the results of the analysis. The report concludes intent to transfer or commitment to postsecondary educa-
with a discussion of the results in the context of other tion. They are as follows:
studies of community college students and transfer. 1) Expected to complete bachelor’s degree or higher;

Data and Measurement 2) Enrolled in an academic program;

Although a considerable amount of research has investi- 3) Enrolled continuously in 1989–90;
gated community college transfer rates, many of these 4) Enrolled anytime in academic year 1990–91;
analyses have used data that are either limited to a cohort 5) Pursuing academic major or taking courses toward a
of recent high school graduates, such as the National bachelor’s degree or both;
Education Longitudinal Study of 1988 (NELS:88), or are 6) Enrolled for 12 or more credit hours;
not nationally representative. This study uses the National
7) Taking courses toward a bachelor’s degree in 1989–90;
Center for Education Statistics (NCES) 1990 Beginning
and
Postsecondary Students Longitudinal Study (BPS:90/94),
a nationally representative sample of all students who 8) Pursuing academic major and taking courses toward a
enrolled in postsecondary education for the first time bachelor’s degree.
between July 1, 1989, and June 30, 1990. Follow-up
interviews were conducted in spring 1992 and 1994. BPS The analysis begins with estimating the percentage of the
is particularly appropriate for the study of community 1989–90 cohort of beginning community college students
college students because it is representative of all begin- who meet each of these definitions. The relationship of
ning postsecondary students, not just recent high school these criteria to various other student characteristics is then
graduates. explored. The first issue to be examined is whether the
composition of the pool of potential transfer students varies
The approach of this report is similar to that used in as the definitions become more restrictive. Then, consider-
analyses of individual community colleges or districts, ation is given to whether different subgroups of students
particularly that of Spicer and Armstrong (1996). Holding are more or less likely to meet each definition. Finally, a
the numerator constant, variously restrictive definitions of transfer rate is calculated for each definition of potential
the denominator are employed based on the different transfer students, and the relationships of these definitions
approaches to specifying the transfer population found in to transfer are explored.
the literature.
Results
This report defines transfer as follows: initial enrollment at Overall, 71 percent of 1989–90 beginning community
a community college followed by subsequent enrollment college students responded that they anticipated earning a
at any 4-year institution within the 5-year study period. bachelor’s degree or higher when asked, “What is the
Potential transfer refers to being eligible for transfer or “at highest level of education you ever expect to complete?”
risk” of transfer. The broadest definition of potential transfer (figure A). Also, the majority of students were enrolled in
used in this analysis includes all first-time, beginning an academic program, enrolled continuously in 1989–90,
community college students, although students only taking and enrolled during the 1990–91 academic year. Less
courses for which they receive no credit are excluded from than half of the students met the other definitions, with
the BPS sample. The pool of potential transfer students is 11 percent of the students both having an academic major
then restricted using eight additional definitions of the and taking courses leading toward a bachelor’s degree.

120 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
 Community College Transfer Rates to 4-Year Institutions Using Alternative Definitions of Transfer

Does the composition of the group of potential transfer sample who met the most restrictive definition of potential
students change as the definitions become more transfer were Black, compared to 6 to 10 percent Black
restrictive? students in each of the other potential transfer groups.* In
Restricting the group of potential transfer students accord- general, restricting the pool of potential transfer students
ing to these definitions may alter the composition of the systematically altered the composition of the group to
group, since these educational characteristics are them- include more traditional students (younger, dependent
selves associated with other background variables (Berkner, students who do not work full time).
Cuccaro-Alamin, and McCormick 1996). Across increas-
ingly restrictive definitions of potential transfer, the per- What percentage of students with different
centage of the pool that was in the highest socioeconomic characteristics meets each potential transfer definition?
status (SES) quartile increased from 30 percent of all In addition to examining how the composition of the
beginning community college students to 51 percent of population of potential transfer students changed as the
beginning community college students with an academic
major and taking courses leading toward a bachelor’s *This does not necessarily mean, however, that there are no Black students in the
degree. Furthermore, none of the students in this particular population of community college students who would meet this definition.

Figure A.—Among 1989–90 beginning postsecondary students enrolled at public 2-year institutions, the percentage of the initial cohort meeting each
definition of potential transfer

Percent
100

90

80

71
70 68
63 62
60

50
43
40 36

30
25

20

11
10

0
Expected to Enrolled in Enrolled Enrolled Pursuing Enrolled for Taking Pursuing
complete an academic continuously anytime in academic 12 or more courses academic
bachelor’s program in 1989–90 academic major or credit hours toward major and
degree or year taking courses bachelor’s taking courses
higher 1990–91 toward degree toward
bachelor’s in 1989–90 bachelor’s
degree or both degree

NOTE: Unless otherwise specified, variables are as of 1989–90 (base-year interview, first term of postsecondary enrollment).
SOURCE: U.S. Department of Education, National Center for Education Statistics, 1990 Beginning Postsecondary Students Longitudinal Study (BPS:90/94), Data Analysis System.

E D U C AT I O N S TAT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , F A L L 2 0 0 1 121
Methodology

definitions became more restrictive, the report also com- Transfer rates for each definition of potential transfer
pares the likelihood of meeting each definition of potential Figure B shows estimated transfer rates for all community
transfer across various subgroups of students. For example, college entrants and for the eight increasingly restrictive
students 22 years or older were generally less likely than definitions of potential transfer arranged in order. The
younger students to meet the various definitions of poten- results show that, in general, the transfer rate increased
tial transfer. In general, the higher the SES, the higher for more restrictive definitions. The lowest rate of ever
the percentage of students who met the criteria for each enrolling in a 4-year institution by spring 1994, 25 percent,
specification. Students who reported taking at least 1 credit was found for all 1989–90 beginning community college
hour of remedial mathematics instruction during 1989–90 students, compared to 52 percent for students meeting the
were generally about as likely to fit each definition as most restrictive definition (both pursuing an academic
students who did not take any remedial mathematics major during 1989–90 and taking courses leading toward
instruction. Students who were enrolled full time were a bachelor’s degree). That is, the transfer rate for the most
generally more likely to meet the various specifications restrictive definition was at least twice the rate for all
than those who were enrolled less than full time. students. However, figure C demonstrates that the

Figure B.—Among 1989–90 beginning postsecondary students enrolled at public 2-year institutions, the percentage who transferred to 4-year institutions by
spring 1994, for various definitions of potential transfer

Percent
100

90

80

70

60
52
50
45
43
40
40 37 38
36 36

30
25

20

10

0
All Expected to Enrolled in Enrolled Enrolled Pursuing Enrolled for Taking Pursuing
students complete an academic continuously anytime in academic 12 or more courses academic
bachelor’s program in 1989–90 academic major or credit hours toward major and
degree or year taking courses bachelor’s taking courses
higher 1990–91 toward degree toward
bachelor’s in 1989–90 bachelor’s
degree or both degree

NOTE: Unless otherwise specified, variables are as of 1989–90 (base-year interview, first term of postsecondary enrollment).
SOURCE: U.S. Department of Education, National Center for Education Statistics, 1990 Beginning Postsecondary Students Longitudinal Study (BPS:90/94), Data Analysis System.

122 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
 Community College Transfer Rates to 4-Year Institutions Using Alternative Definitions of Transfer

percentage of actual transfer students meeting the criteria for rates generally persisted even when the analyses were
inclusion in the denominator declined significantly as the restricted to students meeting the various potential transfer
definitions of potential transfer became more restrictive. In definitions. For example, in general, the older the age
other words, attempts to include only those students most group, the lower the percentage of students who trans-
likely to transfer actually exclude a sizable proportion of ferred, regardless of the definition of potential transfer that
students who transfer anyway, without meeting those was used. Also, regardless of the potential transfer defini-
criteria. For example, four out of five actual transfer students tion used, higher SES was generally associated with a higher
did not meet the most restrictive definition considered. transfer rate.

Additional exploratory analyses examined the percentage of Current Results in the Context of Previous
beginning community college students who ever transferred Studies
to a 4-year institution for each potential transfer definition, The results of this study can be placed in the context of the
by selected student background characteristics. In several literature about two questions: what percentage of students
cases, relationships of student characteristics to transfer in community colleges have educational expectations that

Figure C.—Among 1989–90 beginning postsecondary students enrolled at public 2-year institutions who transferred to 4-year institutions by spring 1994,
the percentage who met the various definitions of potential transfer

Percent
100
95
93
89
90
84

80

69
70

60
55

50
44

40

30

21
20

10

0
Expected to Enrolled in Enrolled Enrolled Pursuing Enrolled for Taking Pursuing
complete an academic continuously anytime in academic 12 or more courses academic
bachelor’s program in 1989–90 academic major or credit hours toward major and
degree or year taking courses bachelor’s taking courses
higher 1990–91 toward degree toward
bachelor’s in 1989–90 bachelor’s
degree or both degree

NOTE: Unless otherwise specified, variables are as of 1989–90 (base-year interview, first term of postsecondary enrollment).
SOURCE: U.S. Department of Education, National Center for Education Statistics, 1990 Beginning Postsecondary Students Longitudinal Study (BPS:90/94), Data Analysis System.

E D U C AT I O N S TAT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , F A L L 2 0 0 1 123
Methodology

include a bachelor’s degree, and what is the transfer rate Conclusion

for community college students? This contextual informa- This examination of alternative ways of defining potential
tion is not intended to constitute a statistical comparison transfer was undertaken in part to inform research at the
across studies. In general, however, BPS estimates of the design stage. Which definition (or definitions) is (are) most
percentage of beginning community college students appropriate for addressing a specific research question? The
whose expectations included a bachelor’s degree or higher, decision is not straightforward. This report illustrates the
as well as the percentage of students who transferred to a trade-off between restricting the pool of potential transfer
4-year institution, are higher than estimates based on other students and excluding substantial portions of the initial
data sets. Dougherty (1987, 1992), for example, reviewed cohort. For example, including only students who have an
several studies and concluded that 30 to 40 percent of all academic major and are taking courses leading toward a
community college entrants aspire to a bachelor’s degree, bachelor’s degree results in a high transfer rate (52 percent),
while the present study found that 71 percent of commu- but no more than about 1 in 10 community college students
nity college students in BPS expect to complete a bachelor’s meets this definition, and it excludes 4 out of 5 transfer
degree or higher. Similarly, while the overall transfer rate students. Restricting the pool to the 71 percent of students
found in this study is comparable to the average estimate of who expect to earn a bachelor’s degree or higher yields a
22 percent found by the Transfer Assembly project (Cohen transfer rate of 36 percent, but fully 95 percent of all
and Sanchez 1997), both the numerator and denominator of transfers have this expectation.
the Transfer Assembly project are more restrictive. Although
BPS data do not facilitate use of such a restrictive definition, Just as any statistic depends on the specific variables used
an approximation of it using BPS data yielded a transfer rate to indicate the underlying concepts, the transfer rate for
of 33 percent, somewhat higher than the rate resulting from community college students is sensitive to the specification
the Transfer Assembly project. of potential transfer. The most complete picture is provided
by using multiple indicators, but this approach is not
However, any comparisons between the estimates presented always practical. If data collection costs or other constraints
in this report and those from other studies must be consid- only permit one definition, one strategy is to define the
ered in light of differences in how the overall population of group of potential transfer students broadly enough that it
community college students is defined across studies. For still reflects community college students somewhat gener-
example, BPS excluded students taking courses only for ally, while not so broadly as to include students who never
remedial or avocational purposes without receiving credit, harbor plans to transfer to a 4-year college. The results of
while other estimates may include these students. As a this study present several alternatives with different advan-
result, the typical amount of remediation for students in tages and disadvantages.
BPS may underestimate, or otherwise differ from, the
amount of remediation found among community college Overall, the results provide national estimates of commu-
students in general—a factor that might be associated with nity college students’ academic expectations and transfer
transfer to a 4-year institution. Furthermore, this report activity. These estimates refine and update our understand-
focuses only on students enrolled in public 2-year colleges; ing of students’ intentions and paths to transfer. Selecting
including other less-than-4-year colleges, particularly less- an approach to defining potential transfer is a necessary first
than-2-year institutions, may lower the estimates. In step in any effort to analyze the impact of institution type
addition, BPS data are restricted to first-time beginning on persistence and attainment. This report has taken a step
postsecondary students; colleges conducting their own back and analyzed the definition itself by examining several
studies of transfer may include entering students who are alternative approaches using a complete nationally repre-
not first-time beginners as defined in the BPS study. Finally, sentative sample. While the results do not demonstrate the
it could also be that student aspirations change appreciably superiority of any single definition, they sharpen one’s appre-
from one cohort to the next and that estimates therefore ciation for the consequences of measurement decisions and
could depend in part on when the survey was administered. build a firmer foundation for future work on this population.

124 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
 Community College Transfer Rates to 4-Year Institutions Using Alternative Definitions of Transfer

References Grubb, W.N. (1991). The Decline of Community College Transfer

Berkner, L.K., Cuccaro-Alamin, S., and McCormick, A.C. (1996). Rates: Evidence From National Longitudinal Surveys. Journal of
Descriptive Summary of 1989–90 Beginning Postsecondary Higher Education, 62(2): 194–222.
Students: 5 Years Later With an Essay on Postsecondary Persistence Spicer, S.L., and Armstrong, W.B. (1996). Transfer: The Elusive
and Attainment (NCES 96–155). U.S. Department of Education, Denominator. New Directions for Community Colleges, 24(4):
National Center for Education Statistics. Washington, DC: U.S. 45–54.
Government Printing Office.
Brint, S., and Karabel, J. (1989). The Diverted Dream: Community
Colleges and the Promise of Educational Opportunity in America, Data source: The 1990 Beginning Postsecondary Students
1900–1985. New York: Oxford University Press. Longitudinal Study (BPS:90/94).
Cohen, A.M., and Sanchez, J.R. (1997). The Transfer Rate: A Model For technical information, see the complete report:
of Consistency. Los Angeles, CA: Center for the Study of Bradburn, E.M., and Hurst, D.G. (2001). Community College Transfer
Community Colleges. (ERIC ED409952) Rates to 4-Year Institutions Using Alternative Definitions of
Transfer (NCES 2001–197).
Dougherty, K. (1987). The Effects of Community Colleges: Aid or
Author affiliations: E.M. Bradburn, NCES; D.G. Hurst, Education
Hindrance to Socioeconomic Attainment? Sociology of Educa- Statistics Services Institute (ESSI).
tion, 60(2): 86–103.
For questions about content, contact Annette Allen
Dougherty, K. (1992). Community Colleges and Baccalaureate (annette.allen@ed.gov).
Attainment. Journal of Higher Education, 63(2): 188–214. To obtain the complete report (NCES 2001–197), call the toll-free
ED Pubs number (877–433–7827) or visit the NCES Web Site
Grubb, W.N. (1988). Vocationalizing Higher Education: The Causes
(http://nces.ed.gov).
of Enrollment and Completion in Public Two-Year Colleges,
1970–1980. Economics of Education Review, 7(3): 301–319.

E D U C AT I O N S TAT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , F A L L 2 0 0 1 125
Methodology

Classifying 2-Year Institutions

A Classification System for 2-Year Postsecondary Institutions
——————————————————————————————————
Ronald A. Phipps, Jessica M. Shedd, and Jamie P. Merisotis
This article was originally published as the Executive Summary of the Technical Report of the same name. The universe data are from the NCES
Integrated Postsecondary Education Data System (IPEDS).

Background needs, these unique American inventions are in fact a

A well-known classification system for higher education disparate group of institutions. The purpose of this report is
in the United States was developed by the Carnegie Com- to describe a classification system for 2-year institutions
mission on Higher Education. Originally published in that can contribute to research and provide a framework for
1973—and refined and updated in 1976, 1987, 1994, policy discussions.
and 2000—the Carnegie classification system divides
colleges and universities into categories based on their Methodology
degree-granting activity (doctoral, master’s, baccalaureate, This study used the IPEDS database. The universe of
and associate’s). In addition, the Carnegie classification institutions consists of 2,068 Title IV participating 2-year
system identifies myriad “specialized institutions,” such as postsecondary institutions that met the data criteria for this
theological seminaries, schools of law, teachers colleges, and study.2 K-means cluster analysis—in combination with
medical schools. various other procedures—was the primary method used to
classify these institutions.
A limitation of the Carnegie classification system is how
2-year institutions are categorized. While there are six Cluster analysis is a multivariate statistical procedure that
distinct subcategories for 4-year institutions, the 2000 attempts to mathematically form “clusters,” or groups of
Carnegie classification system places the 1,669 institutions relatively homogenous entities, based on measures of
that offer associate of arts degrees or certificates and—with similarity with respect to specific variables, while maximiz-
few exceptions—offer no baccalaureate degrees into a single ing the differences between groups. A focus group of
grouping, Associate’s Colleges (The Carnegie Foundation experts in the field—researchers, association leaders, and
for the Advancement of Teaching 2000). This single policy analysts—along with preliminary data analysis,
grouping of institutions accounts for more institutions than determined which variables were both policy relevant and
Doctoral/Research Universities, Master’s Colleges and appropriate to be included in the cluster analysis procedure.
Universities, and Baccalaureate Colleges combined (1,478). The results of the cluster analysis, in conjunction with
Furthermore, the Carnegie classification includes only subsequent analyses, revealed which variables created the
accredited and degree-granting institutions, excluding over most distinguishable categories of institutions.
700 of the 2,427 2-year institutions with participation
agreements for Title IV student aid funds, as found in the The Classification System
1997–98 “Institutional Characteristics Survey” of the The universe of institutions was first separated into three
Integrated Postsecondary Education Data System (IPEDS) sectors by institutional control—public, private not-for-
database. profit, and private for-profit. Further analyses were con-
ducted within each sector, resulting in the seven-category
About half of all students in postsecondary education are classification system described below (figure A).
enrolled in 2-year institutions.1 Placing all 2-year institu-
tions into one category, thereby masking their differences, Public institutions
has limited the understanding of this crucial segment of Size of institutional enrollment is the most distinguishing
higher education. While community colleges and other characteristic of public 2-year institutions. Below is a brief
2-year institutions often share a commitment to open description of the three categories within this sector:
access, comprehensiveness, and/or responsiveness to local
2
About 350 institutions were not included in this study because of inconsistencies in
1
This enrollment information is based on the NCES 1995–96 National Postsecondary their data reporting in the different IPEDS surveys.
Student Aid Study (NPSAS:96), Data Analysis System.

126 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
 A Classification System for 2-Year Postsecondary Institutions

Figure A.—Classification system of 2-year institutions

Universe of 2-year institutions

N=2,068

Public Institutions Private Not-For-Profit Private For-Profit

Institutions Institutions
N=1,029 N=309 N=730

Community Community Community Allied Health Connector Career Connector Certificate

Development Connector Mega-Connector
and Career
N=258 N=505 N=251 N=165 N=128 N=367 N=333

NOTE: The sum of the number of institutions in each category does not add to the total number of institutions due to missing data in the variables chosen for
categorization. In the universe of 2,068 institutions analyzed in this report, 61 institutions could not be placed in a final category: 15 public 2-year institutions, 16 private
not-for-profit institutions, and 30 private for-profit institutions.
SOURCE: U.S. Department of Education, National Center for Education Statistics, Integrated Postsecondary Education Data System (IPEDS), 1997 Full Collection Year.

Community Development and Career Institutions are institu- Allied Health Institutions are institutions that grant 100
tions with an unduplicated headcount of less than 2,000 percent of their awards in allied health programs. These
students. These institutions tend to confer awards and institutions tend to be small in enrollment and to have an
degrees primarily in job and career skills development, and exclusive focus on allied health training.
to focus on overall workforce development for the commu-
nities they serve. Connector Institutions are institutions that grant less than
100 percent of their awards in allied health programs. These
Community Connector Institutions are institutions with an institutions tend to confer awards and degrees that target
unduplicated headcount of 2,000 to 9,999 students. These job and career skills development, and to offer academic
institutions tend to confer awards and degrees that target programs with some component of general education that
job and career skills development, and to offer academic can facilitate transfer to 4-year institutions.
programs with some component of general education that
can facilitate transfer to 4-year institutions. Private for-profit institutions
A distinguishing characteristic of private for-profit 2-year
Community Mega-Connector Institutions are institutions with institutions is the percentage of total awards granted that
an unduplicated headcount of at least 10,000 students. are certificates. Below is a description of the two categories
These institutions tend to be in urban locations, to confer within this sector:
awards and degrees that target job and career skills develop-
ment, and to offer academic programs with some compo- Career Connector Institutions are institutions that grant less
nent of general education that can facilitate transfer to than 100 percent of their awards as certificates. They are
4-year institutions. degree-granting institutions—although many also offer
certificates—that target job and career skills development.
Private not-for-profit institutions Many of these institutions offer academic programs with
The percentage of total awards granted that are in allied some component of general education that can facilitate
health programs is the distinguishing characteristic of transfer to 4-year institutions.
private not-for-profit 2-year institutions. Two categories
were created within this sector:

E D U C AT I O N S TAT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , F A L L 2 0 0 1 127
Methodology

Certificate Institutions are institutions that grant 100 percent Data source: The NCES 1997–98 Integrated Postsecondary Education
of their awards as certificates. These institutions provide Data System (IPEDS).
specialized training, usually in a single job category or area. For technical information, see the complete report:
Phipps, R.A., Shedd, J.M., and Merisotis, J.P. (2001). A Classification
Reference System for 2-Year Postsecondary Institutions (NCES 2001–167).
Author affiliations: R.A. Phipps, J.M. Shedd, and J.P. Merisotis, The
The Carnegie Foundation for the Advancement of Teaching. Institute for Higher Education Policy.
(2000). The Carnegie Classification of Institutions of Higher
For questions about content, contact Aurora D’Amico
Education. Available: http://www.carnegiefoundation.org/Classifi- (aurora.d’amico@ed.gov).
cation/CIHE2000/Tables.htm
To obtain the complete report (NCES 2001–167), call the toll-free
ED Pubs number (877–433–7827), visit the NCES Web Site
(http://nces.ed.gov), or contact GPO (202–512–1800).

128 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
O T H E R P U B L I C AT I O N S AND F U N D I N G O P P O RT U N I T I E S

OTHER PUBLICATIONS
Programs and Plans of the National Center for Education Statistics: 2001
Celestine Davis (editor) ....................................................................................... 129
The Nation’s Report Card: Mathematics Highlights 2000
Shari L. Santapau ......................................................................................... 129
The Nation’s Report Card: State Mathematics 2000 Reports
Charlotte Solomon, Laura Jerry, and Anthony Lutkus ......................................... 130

FUNDING OPPORTUNITIES
The AERA Grants Program .............................................................................. 130
The NAEP Secondary Analysis Grant Program ............................................... 131

Other Publications The Nation’s Report Card: Mathematics

Highlights 2000
Programs and Plans of the National Center for
Shari L. Santapau
Education Statistics: 2001
Celestine Davis (editor) The National Assessment of Educational Progress
(NAEP), known as “The Nation’s Report Card,” is
This publication summarizes current NCES statistical
authorized by Congress, administered by NCES, and
programs, including major publications and plans for
overseen by the National Assessment Governing Board
future work. It includes descriptions, timelines, and
(NAGB). For over 30 years, NAEP has been the only
plans for all NCES data collections, such as the Com-
ongoing national indicator of what American students
mon Core of Data, Integrated Postsecondary Education
know and can do in major academic subjects. In 2000,
Data System, National Assessment of Educational
NAEP administered a mathematics assessment to a
Progress, Early Childhood Longitudinal Study, Third
national sample representative of all students at grades
International Mathematics and Science Study–Repeat,
4, 8, and 12. In addition, state-level results were
and National Postsecondary Student Aid Study. Also
collected at grades 4 and 8. The findings from the
described in this publication are NCES centerwide
NAEP 2000 Mathematics Assessment provide a picture
programs and services, such as statistical standards,
of U.S. students’ mathematics knowledge, skills, and
training, technology, and customer service.
achievement.
Editor affiliation: C. Davis, NCES.
This 16-page publication uses a full-color tabloid
For questions about content, contact Celestine Davis
(celestine.davis@ed.gov). format to present highlights from the 2000 mathemat-
To obtain this publication (NCES 2001–038), call the toll-free ics assessment. It describes the assessment content,
ED Pubs number (877–433–7827) or visit the NCES Web Site presents major findings, and provides information
(http://nces.ed.gov).

E D U C AT I O N S T AT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , F A L L 2 0 0 1 129
Other Publications and Funding Opportunities

about practices at home and in school that are related Funding Opportunities
to mathematics achievement. Results in 2000 are
The AERA Grants Program
compared to results in 1990, 1992, and 1996. The
publication also includes sample test questions and Jointly funded by the National Science Foundation
examples of student responses. (NSF), NCES, and the Office of Educational Research
and Improvement (OERI), this training and research
Author affiliation: S.L. Santapau, Educational Testing Service. program is administered by the American Educational
For questions about content, contact Arnold Goldstein Research Association (AERA). The program has four
(arnold.goldstein@ed.gov). major elements: a research grants program, a disserta-
To obtain this publication (NCES 2001–518), call the toll-free
ED Pubs number (877–433–7827) or visit the NCES Web Site
tion grants program, a fellows program, and a training
(http://nces.ed.gov). institute. The program is intended to enhance the
capability of the U.S. research community to use
large-scale data sets, specifically those of the NSF
The Nation’s Report Card: State Mathematics and NCES, to conduct studies that are relevant to
2000 Reports educational policy and practice, and to strengthen
Charlotte Solomon, Laura Jerry, and Anthony Lutkus communications between the educational research
community and government staff.
National Assessment of Educational Progress (NAEP)
assessments are administered at both the state and Applications for this program may be submitted at any
national levels. The NAEP 2000 Mathematics Assess- time. The application review board meets three times
ment collected state-level results for fourth- and eighth- per year. The following are examples of grants recently
graders who attended public schools in states and other awarded under the program:
jurisdictions that volunteered to participate. NAEP has
collected state-level mathematics results since 1992 at Research Grants
grade 4 and since 1990 at grade 8. ■ Mark Beasley, St. John’s University—Effects of
Educational Opportunity on the Intraschool
This series of reports provides each participating Distribution of Eighth-Grade Mathematics
jurisdiction with an overview of its results from the Achievement in the U.S. and Korea: Multilevel
2000 mathematics assessment as well as previous Analyses of TIMSS
assessments. Each jurisdiction receives its own custom- ■ Douglas Downey, Ohio State University—When
ized report, which presents results for public school the Time Is Right: Delayed Entry to Kindergarten
students in that jurisdiction, along with national and and Its Consequences for Stratification
regional results for comparison. Each report also
■ Ithel Jones, Florida State University—Social and
presents a second set of results that includes the
Academic Effects of Varying Types of Preschool
performance of special-needs students who were
Experiences
permitted accommodations in the test administration.
■ Lois Joy, Smith College—Gender Differences in
In addition, the report includes information about the
assessment content, the sample of students assessed, the Transition From College to Work: Salaries,
and the way results are reported. Occupations, and Job Changes in the Skilled Job
Market
Author affiliations: C. Solomon, L. Jerry, and A. Lutkus, Educational
Testing Service.
For questions about content, contact Arnold Goldstein
(arnold.goldstein@ed.gov).
To obtain a state report (NCES 2001–519), visit the NCES Web Site
(http://nces.ed.gov).

130 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S
 ■ Seongeun Kim, University of California, Los The NAEP Secondary Analysis Grant Program
Angeles—Prevention of School Violence and The NAEP Secondary Analysis Grant Program was
Crime: Investigation of Cross-Level Interaction developed to encourage education researchers to
Effects of Risk and Protective Factors on Adoles- conduct secondary analysis studies using data from the
cent Violence and Crime Using Multilevel Longi- National Assessment of Educational Progress (NAEP)
tudinal Methods and the NAEP High School Transcript Studies. This
■ John Warren, University of Washington—Trends program is open to all public or private organizations
in the Selectivity and Consequences of Adoles- and consortia of organizations. The program is typically
cent Employment, 1966–1997 announced annually, in the late fall, in the Federal
Register. Grants awarded under this program run from
Dissertation Grants 12 to 18 months and awards range from $15,000 to
■ Percy Abram, Stanford University—Does $100,000. The following grants were awarded for fiscal
Language Matter? The Effects of Language on the year 2001:
Development of Social Capital Among Latino
Students ■ David Grissmer, Rand Corporation—Analyzing
■ Nora Gordon, Harvard University—Tracking State NAEP Data to Address Educational Policy
Title I: From Revenues to Inputs to Outcomes Issues in K–12 Education
■ Lawrence Rudner, LMP Associates, Inc.—Scoring
■ Jenifer Hamil-Luker, University of North Caro-
lina, Chapel Hill—Differential Participation in Content Essays Using Bayesian Networks
and Returns to Education Over the Life Course ■ Robert Lissitz, University of Maryland—Science
■ Amy Lutz, SUNY-Albany—Bilingualism in the Achievement in Social Contexts: An Alternative
USA: Language Outcomes and Consequences for Method for Analysis of Data From NAEP
Hispanic Youth ■ Richard Niemi, University of Rochester—
■ David Most, University of California, Los Components of Knowledge in the NAEP 1998
Angeles—Funding and Finishing the Ph.D.: The Civics Main and Trend Assessments
Role of Various Patterns of Graduate Support ■ Daniel Sherman, American Institutes for
Mechanisms Research—Application of Small Area Estimation
■ Gregory Palardy, University of California, Santa Methods to NAEP
Barbara—An Improved Model for the Equitable ■ Claudia Gentile, Educational Testing Service—
Evaluation of School Effectiveness: An Applica- Evaluating the “Creative” in Creative Writing
tion of Multilevel Latent Variable Growth ■ Matthew Schultz, ACT, Inc.—Describing
Modeling Achievement Levels With Multiple Domain
Scores
For more information, contact Edith McArthur
(edith.mcarthur@ed.gov) or visit the AERA Grants
For more information, contact Alex Sedlacek (alex.sedlacek@ed.gov).
Program Web Site (http://www.aera.net/grantsprogram).

E D U C AT I O N S T AT I S T I C S Q U A R T E R LY — V O L U M E 3 , I S S U E 3 , F A L L 2 0 0 1 131
132 N AT I O N A L C E N T E R F O R E D U C AT I O N S TAT I S T I C S