Features

2007 Wallace Foundation Distinguished

Lecture
What Makes Education Research “Educational”?
by Deborah Loewenberg Ball and Francesca M. Forzani

Education research is plagued by skeptics who doubt its quality and More––and more diverse––people need to learn more––and
relevance. Inhabitants of schools of education have been among the more varied––things than ever before. An educated citizenry is
sharpest critics, and internal battles rage over method and rigor. Yet needed to tackle societal problems of health care, hunger, energy,
often lacking is research that explains causes or examines the inter-
poverty, and environmental sustainability. Moreover, as life
expectancies increase and as globalization and the digital revolu-
play at the heart of educational practice and policy. This article argues
tion alter basic notions of access and interaction, the kind of edu-
for a conception of research in education that deliberately presses cation required will change, and it will need to extend across the
into what is called here the instructional dynamic. Using a sample of life span.
studies that exemplify this quintessentially educational perspective, Yet delivering effective education remains a problem. Formal
the authors unpack key features of research that probes inside schooling and educational programs often fail. Students retain
education. They discuss how such research complements in essen-
misconceptions even after instruction, basic academic skills are
often undeveloped, and many youth leave school unprepared to
tial ways the other kinds of scholarship that examine and inform
participate competently in a democratic and diverse society. Most
education. troubling is that education is delivered unevenly and inequitably.
In the United States as elsewhere, the nature of the educational
opportunities available to students living in poverty or to those
Keywords: education research; education researchers; schools
who are members of underrepresented groups is on average infe-
of education; teaching practice rior to that available to their middle-class and White counter-
parts. These inequities produce significant disparities in academic
achievement and in employment opportunities. For individual
and societal reasons, it is imperative to address these problems.

S
chools of education are perennially under fire. Critics point To date, however, solutions have been elusive.
to the uselessness of education research, the low intellec- One impediment is that solving educational problems is not
tual demand of teacher education, and the weak academic thought to demand special expertise. Despite persistent problems
qualifications of education students and their professors. Disdain of quality, equity, and scale, many Americans seem to believe that
for education schools is not new. From Bestor (1953) to Lyon work in education requires common sense more than it does the
(2002), commentators have heaped criticism on the colleges, sort of disciplined knowledge and skill that enable work in other
schools, and departments whose central mission is “education” fields. Few people would think they could treat a cancer patient,
(Conant, 1963; Judge, 1982; Koerner, 1963; Levine, 2006). The design a safer automobile, or repair a bridge, for these obviously
very existence of these departments and schools is threatened; require special skill and expertise. Whether the challenge is recruit-
some leading universities have closed or downsized their schools ing teachers, motivating students to read, or improving the math
of education. We argue that schools of education have a special curriculum, however, many smart people think they know what
role in addressing problems of educational improvement. it takes. Because schooling is a common experience, familiarity
The basis of our argument—and the answer to the question masks its complexity. Powell (1980), for example, referred to edu-
in our title—is that what makes education research “educational” cation as a “fundamentally uncertain profession” about which the
is a particular orientation to scholarship that we call research in perception exists that ingenuity and art matter more than profes-
education. sional knowledge. Yet the fact that educational problems endure
The larger problem that animates this question is the broad despite repeated efforts to solve them suggests the fallacy of this
challenge of educational improvement and transformation. reliance on common sense. Disciplined research on problems and
Economic sustainability and quality of life depend on education. solutions could help in education, just as it does in other domains.

Educational Researcher, Vol. 36, No. 9, pp. 529–540

DOI: 10.3102/0013189X07312896
© 2007 AERA. http://er.aera.net DECEMBER 2007 529
 Less clear, however, is what would characterize disciplined
knowledge in education. In recent years, debates about method environments
and evidence have swamped the discourse on education research teachers
to the exclusion of the fundamental question of what constitutes
education research and what distinguishes it from other domains
of scholarship. The panorama of work represented at professional
students
education meetings or in publications is vast and not highly content
defined. What are the essential scope and content of education students
research? What are the central questions and problems that quin-
tessentially define the domain and its subdomains?
Ironically, the low status often assigned to education creates an
incentive for education faculty members to emulate work in the
other social science disciplines. This has meant that research that is
ostensibly “in education” frequently focuses not inside the dynam- FIGURE 1. The instructional triangle. From Cohen, Raudenbush,
ics of education but on phenomena related to education––racial & Ball, 2003, p. 124. Copyright 2003 by American Educational
identity, for example, young children’s conceptions of fairness, or Research Association. Adapted with permission.
the history of the rise of secondary schools. These topics and others
like them are important. Research that focuses on them, however,
often does not probe inside the educational process. Until educa-
tion researchers turn their attention to problems that exist primar- Education and the Instructional Dynamic
ily inside education and until they develop systematically a body of Education is the deliberate activity of helping learners to develop
specialized knowledge, other scholars who study questions that bear understanding and skills. Although it occurs in many settings and
on educational problems will propose solutions. Because such solu- through a wide range of mechanisms, it is typically associated
tions typically are not based on explanatory analyses of the dynam- with schools, where it consists of interactions among teachers,
ics of education, the education problems that confront society are students, and content, in the various environments of schools (see
likely to remain unsolved. For example, knowing that the number Figure 1; Cohen, Raudenbush, & Ball, 2003). By interactions, we
of books in a child’s home and the educational level of the child’s mean active processes of interpretation that constitute teaching
parents are major factors in predicting school success does not and learning. Teachers interpret and represent subject matter to
explain how these factors influence learning. Nor does such knowl- students, who interpret their teachers, the content, and their
edge help in the design of interventions for particular students. classmates and then respond and act. In turn, teachers interpret
In this article, we label an approach to research that focuses their students, all of this in overlapping contexts and over time
inside educational transactions. We call this research in education (Lampert, 2001). We consider these multiple interactions, which
and distinguish it, without implying superiority, from inquiry we call the instructional dynamic, to be the defining feature of
into phenomena related to education. We argue that this orien- education.
tation to education research is necessary for the production of the Other scholars have characterized the fundamental nature of
sort of disciplined knowledge that might contribute directly to education in related ways. Schwab (1978), for example, writing
solutions to pressing problems in education. In addition, we about the deliberate, eclectic, and multifocal activity of curricu-
argue that phenomena outside educational settings can be stud- lum design, identifies four “commonplaces” of education—
ied with a special educational perspective complementary to the subject matter, learners, milieux, and teachers––which, he argues,
theoretical perspectives offered by other disciplines. This educa- must be coordinated in developing curriculum. This deliberate
tional perspective offers a lens for presenting and studying par- coordination constitutes for Schwab the special demand of edu-
ticular sorts of phenomena as forms of teaching and learning. cational work. Hawkins (1967/1974) refers to “I, Thou, and It”
We begin by defining education and the instructional dynamic to emphasize that what distinguishes teacher-student relation-
that constitutes it. We discuss this instructional dynamic not just ships is their purposeful engagement in a content or subject to be
in relation to schools but as a metaphor for interactions that take learned—an external “It.” And McDonald (1992) identifies a
place in many other settings. We then examine several cases of “wild triangle of relations––among teacher, students, subject”
research in education and research with an educational perspec- (p. 1) to highlight the uncertainty that characterizes teaching. He
tive to illustrate the distinctive attention to the instructional argues that the points of this triangle, always shifting, are what
dynamic that these studies bring to research-based problem solv- teachers must manage as they work their sensitive craft. The key
ing. By using studies that vary in both content and method, we idea for Schwab is the need to coordinate the four commonplaces
aim to exemplify the breadth of our argument about research that in curriculum building; for Hawkins, it is how engagement in
is inside education. We do not intend to displace studies that use subject matter defines the special relationship of teachers and stu-
tools from fields such as philosophy, anthropology, or psychology dents; McDonald focuses on the improvisational and responsive
to inform educational questions in important ways; instead, our nature of teaching. Others have used similar triangular represen-
argument highlights work that probes the instructional dynamic tations, but with different elements (e.g., Bransford, 1979), to
inside educational situations and problems. We conclude by con- model key constituents of the educational process.
sidering the implications of this argument for the nature and The instructional triangle that we use here (Cohen et al.,
purposes of schools of education. 2003) is distinct from the models described above. Its focus is the
530 EDUCATIONAL RESEARCHER
dynamic that constitutes teaching and learning as teachers and identifies the interactions as teaching and learning or designs for
students interpret one another and their environments over time. them as such. For example, whether or not textbook authors and
It is a descriptive, not a normative, construct. curriculum designers craft materials in ways that anticipate learn-
This instructional dynamic occurs not only in schools, where ers’ likely dispositions and prior ideas, users’ interpretations of the
teachers help students learn academic subjects, but also in many material and their environments nonetheless shape their reading
other settings in which a teacher or leader helps others (learners) and use of them. The same can be said of policy. Enactors react to
acquire skills and understand new ideas. In businesses, for exam- policies on the basis of their interpretations of them and of the envi-
ple, executives must help employees learn new processes. To do ronments of their work, whether or not policy makers design for
so, they explain the processes, give examples, and guide the learn- their learning. Curriculum designers or policy makers can try to
ers. This clearly is instruction and is not simple. New equipment improve implementation by aligning their designs to anticipate and
or tools have no value to a business unless employees can be respond to enactors’ knowledge, beliefs, and interpretations. Our
taught to use them effectively. Advertising also entails instruc- argument is not that curriculum designers or policy makers or oth-
tional thinking; successful marketing takes into account what ers are necessarily successful at teaching enactors or users, or even
customers know and care about, seeks to attract their attention, that they conceive their work in such ways, but that these situations
and “teaches” them what is special about the product. can be usefully examined from an instructional perspective.
A similar instructional dynamic is also evident in the imple- The instructional dynamic that we describe, whether it occurs
mentation of many forms of social policies and programs. Health in schools or elsewhere, is a specific and complex phenomenon
care workers who effectively institute a public health program in that warrants specially focused investigation. It is this special
a rural area, for example, teach patients to employ sanitation tech- focus that we argue constitutes the unique province of education
niques or to engage in other preventive health measures. These research. It is, however, one that education researchers do not, in
workers attend to how the new techniques relate to patients’ cur- the main, seek to highlight and claim. Instead, education research
rent practices and beliefs, and they monitor compliance. When frequently focuses not on the interactions among teachers, learn-
patients do not understand the new practices or are skeptical of ers, and content—or among elements that can be viewed as
their effectiveness, effective health care workers try to find ways such—but on a particular corner of this dynamic triangle.
to help patients learn more and appreciate the validity of the Researchers investigate teachers’ perceptions of their job or their
treatment. Educational reformers seeking to implement a new workplace, for example, or the culture in a particular school or
curriculum in a school face the similar task of making sure that classroom. Many excellent studies focus on students and their
the teachers who will use the new materials understand the pro- attitudes toward school or their beliefs about a particular subject
gram’s goals and know how to make choices about when and how area. Scholars analyze the relationships between school funding
to use the materials provided. and student outcomes, investigate who enrolls in private schools,
We refer to the instructional dynamic, then, not only to rep- or conduct international comparisons of secondary school grad-
resent instruction as it occurs in schools but as a metaphor for the uation requirements. Such studies can produce insights and
interactions that constitute work in many other settings. Even if information about factors that influence and contribute to edu-
these interactions are not labeled as instructional or thus under- cation and its improvement, but they do not, on their own, pro-
stood, there is a teacher involved. The teacher shapes the interac- duce knowledge about the dynamic transactions central to the
tions of the learners with new ideas or materials. Effective process we call education. Knowing about and understanding
teachers, whether they are classroom instructors, health care teachers, learners, content, or environments––or even knowing
workers, business executives, policy makers, or other leaders, use and understanding all of these entities––is not a substitute for
their knowledge of their learners to help them develop new ideas knowing about and understanding the dynamic relationships
and skills. They represent and provide examples of new ideas in among them that constitute the core of the educational process.
multiple forms, highlight important features, and demonstrate Research that focuses on this dynamic is research that probes not
connections between new content and what learners already only the corners of the instructional triangle depicted in Figure 1
know. They design specific activities whereby learners can engage but also the arrows. These arrows represent the dynamic process
with the content, including opportunities to encounter and build of interpretation and mutual adjustment that shapes student
new ideas and skills and to analyze, question, apply, and practice learning, instructional practice, or policy implementation.
that new learning. As learners work, teachers provide feedback Consider a frequently cited study of class size in Tennessee ele-
and encouragement and strategically re-present content when mentary schools. Finn and Achilles (1990) investigated whether
necessary. They also offer incentives for learners to participate in smaller classes positively affected student achievement in com-
these activities, and they regulate the environment so that it is parison with larger classes. The authors investigated the ques-
conducive to learning. tion by comparing students’ performance on standardized and
The transactions that are part of this instructional dynamic curriculum-based tests in reading and mathematics. They found
occur not only when regular teachers and learners meet in person that students in the smaller classes outperformed their peers in
but also asynchronously as designers create policies, curricula, and larger classes in both subjects and that minority students partic-
other educational materials for use by others who must learn from ularly benefited from being in smaller classes. The results suggest
them. People function as teachers or learners in many situations that reducing class size affected the instructional dynamic in ways
that we do not usually associate with those roles. The content of that were productive of improved student learning. The study did
their interactions constitutes the material that is being taught and not, however, explain how this worked. Improvement might have
learned. These transactions occur whether or not anyone explicitly occurred because teachers were able to pay more attention to

DECEMBER 2007 531

individual students. Would the same have been true if the teach- (LD) elementary school students in which instruction was the
ers had not known the material adequately? Would reduced class central focus. They hypothesized that the extent of students’
size work better for students at some ages than at others, or bet- learning in reading would be a function of what students did in
ter in some subjects than in others? Creating effective policies for class and that features of the curriculum and of teachers’ behav-
class size requires insight into how smaller classes work to affect iors in relation to that curriculum would influence students’
students’ achievement. activities in class. Three primary research questions guided their
Questions about why class size matters to student learning are study: What is the nature of reading activities in LD classes?
squarely instructional. They concern what happens inside class- What types of student activities lead to the greatest improvement
rooms when teachers help a smaller number of students negoti- in reading test performance? What types of instructional situa-
ate new content. Researchers might consider, for example, the tions generate these student activities?
types of instructional activities that teachers are able to plan for To answer these questions, Leinhardt and her colleagues (1981)
and enact when working with only a relatively small number of used a pretest, a posttest, and observational measures of 14 variables
students or the extent to which smaller class size allows for more related to student and teacher behavior. The data collected
productive interactions among students. These fundamentally included information about students’ activities and teachers’
instructional questions are enabled by Finn and Achilles’ (1990) moves, timing, and pacing. Using multiple regression analysis, the
work but require the sort of educational lens that is absent from researchers found that (a) teacher behaviors influence student
this initial investigation of the effect of class size on achievement. behaviors in class, and student behaviors in turn influence learn-
Knowing that class size matters is valuable for educators and pol- ing; and (b) most of the variance in students’ posttest performance
icy makers, and Finn and Achilles’ conclusions constitute a sig- is due to teacher and student behavior. In particular, Leinhardt
nificant contribution to our knowledge about schools. Questions et al. found that increases in students’ silent reading time and in the
about why or how class size matters, however, are just as impor- amount of direct, supervised instruction led to improvement in
tant. Too often, they are ignored. students’ reading achievement.
Studies in the area of education often take their perspective This study of the factors that lead to reading achievement pro-
from a discipline outside education––one that is sociological, for vides a strong example of research in education because of its focus
example, or anthropological or psychological. Such work can and inside instruction. Rather than simply looking for correlations
often does contribute helpful knowledge about problems in edu- between students’ background characteristics and their achieve-
cation, but it is not sufficient for unpacking, understanding, and ment or looking at the relationship between enrollment in an LD
solving those problems. Education is inherently transactional; an classroom and learning, Leinhardt et al. attended to how instruc-
orientation to probing the processes inside instances of education tional activities mediate students’ learning. Their study was, in fact,
is what makes education research special. The instructional per- noteworthy not only for its substantive contributions but for the
spective also can serve as a useful lens for interactions in fields not precision of its observational measures and for the strategies that it
typically associated with instruction. We argue that the unique offered for sampling the instructional domain. It demonstrated
contribution that education research can make is attention to the both special appreciation of factors pivotal to student achievement
instructional dynamic. In the next section we elaborate on our and expert skill in measuring those factors and using the results to
conceptions of research in education and research with an edu- inform classroom practice. A study of reading achievement that
cational perspective, and we employ several cases to illustrate the lacked this instructional focus would miss the single most impor-
distinctive features of each. tant set of factors in explaining children’s learning in reading.
What distinguishes the approach taken by Leinhardt et al. is
Research in Education its explicit focus on how teachers and students interpret one
Research in education investigates questions about the instruc- another and instructional activities in classroom contexts and
tional dynamic at play in problems directly related to schooling how these together constitute teaching and learning. Other stud-
and the formal educational process. It treats instruction, at vari- ies in reading focus instead on students’ difficulties, on the chal-
ous grain sizes, as the key variable in educational problems. To lenges of particular texts, or on the developmental trajectories of
clarify what distinguishes research in education from research on struggling readers. Each of these offers valuable insights for tack-
a particular corner of the instructional triangle, we offer six cases ling problems of reading achievement; by themselves, however,
of studies that meet our definition. The first three investigate they do not test interventions or explain how interventions might
problems of classroom instruction, the fourth is a study of edu- work to improve students’ reading skills.
cational program implementation across classrooms, and the final
Case 2: Studies of Conceptual Change in Science
two consider problems in teacher education.
Our second case is from a program of work conducted in the 1980s
Case 1: Measuring Reading Instruction and Its Effects that focused on children’s scientific misconceptions and how cur-
Our first case is from a series of studies of reading instruction con- riculum and instruction could be designed to address them. This
ducted in the early 1980s. Noting that much of the literature on program of research examined instruction at a grain size still finer
children’s reading achievement at the time relied on inadequate than in the studies of Leinhardt and her colleagues (1981).
sampling of classroom events and insufficient models of relation- In an initial case study, Smith and Anderson (1984) examined
ships among classroom events and learning outcomes, Leinhardt, a teacher’s unsuccessful attempt to use a set of published cur-
Zigmond, and Cooley (1981) launched a study of reading riculum materials to help her fifth-grade students understand
achievement in self-contained classrooms for learning disabled photosynthesis. Following suggestions in the teacher’s guide, the

532 EDUCATIONAL RESEARCHER

teacher engaged students in an experiment in which they planted experimental text and employed reading strategies aimed at con-
grass in both light and dark environments. The experiment was ceptual change developed more accurate conceptions of photo-
designed to illustrate the fact that plants use light to make food synthesis than those who did not. The researchers argued that
(photosynthesis) and that plants deprived of light will die. effective instructional materials directly address both teachers’
Despite the successful enactment of the experiment (grass planted assumptions about students and how they learn and students’
in the light thrived; grass planted in the dark began to die), stu- beliefs about the topic at hand. The researchers also suggested
dents maintained their initial belief that plants get food by ingest- that teachers should present students with situations and tasks
ing materials from the soil and concluded that light simply makes that confront them with evidence that competes with their initial
plants “healthier.” misconceptions.
In investigating why the students failed to replace their origi- These studies of conceptual change in science meet our criteria
nal misconceptions with an accurate understanding of photosyn- for research in education because the research questions focus on
thesis, Smith and Anderson focused on several sets of key the interactions among teachers, students, and curriculum materi-
interactions at play in this distinctively instructional problem: als that take place during instruction. Smith and Anderson (1984)
They considered the teacher’s interpretation and use of the and Anderson and Roth (1989) understood that students’ persis-
teacher’s guide and the way that her presentation and orchestra- tent misconceptions were fundamentally an instructional problem.
tion of the instructional activities suggested in the guide inter- Thus they did not merely investigate students’ beliefs about how
acted with students’ prior beliefs about how plants get food. plants obtain food or teachers’ epistemological views, nor did they
Although Smith and Anderson paid close attention to individual focus on teachers’ beliefs about the nature of science. Instead, they
elements of the instructional triangle––in particular to children’s studied how those beliefs influenced how teachers and students
prior beliefs about how plants obtain food, to the teacher’s took up and used the curriculum materials during instruction.
knowledge of photosynthesis and beliefs about learning, and to They also explored how teachers’ guides could be designed to pro-
the characteristics of the curriculum materials––their central con- mote effective use by teachers and how written texts could address
cern was not with these individual elements but with how all of students’ likely initial beliefs and help them build a bridge from
them interacted during instruction. misconceptions to more accurate views. They were, in short, atten-
After multiple classroom observations and interviews with the tive to the dynamic nature of the educational process and to the
students and the teacher, Smith and Anderson concluded that the places in which instructional interactions could break down.
students failed to develop an understanding of photosynthesis Other studies in science education inform questions about the
because their teacher neither addressed their initial misconcep- teaching and learning of science but do not probe science instruc-
tions with sufficient directness nor used the curriculum materials tion directly. Examples include studies of 10-year-olds’ abilities
and activities to help them relate new observations to old ideas. to produce logical explanations, of teachers’ beliefs about science,
The teacher neglected the suggestion in the teacher’s guide that and of attitudes toward science among girls and minority stu-
she define photosynthesis, for example, and she seldom directly dents. Studies such as these address important questions but do
challenged students’ incorrect interpretations of the results of not explain the processes of teaching and learning that combine
their experiment. Smith and Anderson noted that the teacher to produce instruction in the environments of school classrooms.
believed that students’ empirical observations of the dying grass They offer resources for better hypotheses about these interac-
plants would lead the students to derive logically the ideas that tions but do not directly examine them.
she wanted them to learn. This belief seemed to have led her to
pay little attention to students’ original ideas and to misinterpret Case 3: The Cultural Modeling Project
or ignore several crucial suggestions in the teacher’s guide for Just as the above-mentioned researchers probed the instructional
framing the experiment and guiding students’ thinking about it. dynamic inside reading or science instruction, Lee (2007) has
In an experiment that built on Smith and Anderson’s conclu- focused on the educational transactions at the heart of urban
sions, Anderson and Roth (1989) designed an intervention meant African American students’ struggle to read and interpret litera-
to investigate whether a different set of curriculum materials ture. In an analysis of her own teaching of African American stu-
might help students learn more about photosynthesis. Nineteen dents in an urban Chicago high school, Lee hypothesized that
students with misconceptions about how plants obtain food were traditional high school English instruction did not capitalize suf-
randomly divided into three groups and assigned to read about ficiently on students’ extant literacy skills. She observed that
photosynthesis in either an experimental text or a typical text. although African American youth tended to perform poorly in
The experimental text included features designed to help students English classes and on measures of literacy achievement, many
shift from their initial misconceptions to an accurate under- also routinely engaged in sophisticated language play and used
standing of photosynthesis. It continually reminded students of effective strategies for interpreting metaphors, symbols, irony,
the definition of photosynthesis, for example, and prompted and satire in their daily speech. Academic literary analysis, Lee
them to explain clearly how plants and animals get their food. noted, requires the same strategies and appreciation for the uses
Anderson and Roth administered pretests and posttests of stu- and nuances of language. To help students make use of this
dents’ understanding of photosynthesis and used clinical inter- potentially powerful knowledge and skill inherent in their every-
views to probe students’ recall of the text, their reading strategies, day speech, Lee developed, implemented, and studied an instruc-
and their explanations of how plants get their food. tional approach that she called cultural modeling.
On the basis of the results of the interviews and posttests, the The central premise of cultural modeling is that African
researchers concluded that the students who worked from the American English, and the oral genres that African Americans

DECEMBER 2007 533

often use, can function as a bridge to sophisticated literary analy- school exit exams on dropout rates and studies of teachers’ beliefs
sis. Teachers who worked with Lee to implement the program about low-income and minority students. These are important
invested in learning about African American English and the cul- inquiries that inform our understanding of underachieving stu-
tural backgrounds of their students and worked on honing their dents, but they are not about educational transactions.
knowledge of literary analysis and English composition. During
instruction, they assisted students in identifying the problem- Case 4: The Implementation of K–12 Mathematics
solving strategies that they routinely employed to make sense of Reform in California
metaphors and other complex speech forms outside school and Identifying effective instructional regimes and school reform pro-
to connect those strategies with what they were expected to do in grams and the factors that lead to their success is a key problem
academic literary analysis. In teaching students to identify and in education and one in which the instructional dynamic is cen-
interpret metaphors in canonical texts, for example, teachers trally relevant. How teachers use curriculum materials and what
sometimes began by presenting students with the lyrics to a pop- they do with students matters for what students learn. Thus an
ular song and asking them to explain how they would decode effective evaluation of a particular program implemented across
metaphors in those lyrics. Once students were able to make multiple sites necessarily takes into account differences in the way
explicit the strategies that they employed with the familiar text, the program is used at the classroom level. One implication of
teachers helped them use the same strategies to approach the new this is that educational innovations are likely to meet with more
text and to offer careful oral and written interpretations of it. success when they are informed by knowledge about effective
Cultural modeling as Lee has developed it helps to mediate the instructional practices and when the teachers who implement
interactions among teachers, students, and content that consti- them have opportunities to learn to engage in those practices in
tute instruction. Individual corners of the instructional triangle the context of the innovation. One way to frame the problem of
play important roles in cultural modeling, but it is the transac- implementation, then, is as a problem of professional learning.
tions among the corners that matter most––the arrows of the The task for the designers of programs and policies is to teach
instructional triangle. The program depends heavily on teachers’ implementers to use the new ideas effectively.
knowledge of students’ cultural backgrounds, speech patterns, Cohen and Hill’s (2001) investigation of California’s attempt
and out-of-school activities in addition to their expertise in liter- during the 1990s to improve mathematics instruction in its ele-
ary analysis. Employed effectively, however, cultural modeling mentary schools illustrates this focus on the instructional
helps teachers use what they know to design instructional activi- dynamic that is at the core of effective program implementation.
ties specially tailored to their students and to create a classroom The goal of the California math reform was to engage all students
climate that supports ambitious academic instruction. in more intellectually demanding work on mathematics. The
In developing this alternative curriculum and studying its reform included unusually detailed guidance for teaching and
implementation and effectiveness, Lee does not limit her atten- learning, including curriculum frameworks, instructional mate-
tion to teachers or students or even to curriculum content. Her rials, assessments, and professional education for teachers.
study is not primarily about the African American students in the Thousands of workshops were offered around the state to allow
Chicago school where the investigation took place. It is also not teachers to learn about these new materials, to study students’
primarily about the teachers who worked with those students, work on the new state assessments, and to collaborate in solving
although it is attentive to the prior beliefs about African American the problems of delivering the new curriculum.
adolescents and about teaching English that those teachers In their study, Cohen and Hill conceived of the implementa-
brought to the work. Lee’s study is not, in short, a work of soci- tion problem in California as one of professional learning. Their
ology or anthropology, although it draws on knowledge and on analysis focused on the instructional dynamic at play in reformers’
ways of knowing from those fields. Lee peered inside the instruc- explanation of the new program to teachers and in teachers’
tional dynamic to understand the interactions among teachers, attempts to use the new curriculum. They considered, for example,
students, and content that happen during the teaching and learn- how teachers’ beliefs about mathematics teaching and learning
ing of literary analysis. She considered how teachers draw on their influenced their willingness to adopt the suggested instructional
knowledge of students and of academic content to design and use strategies and materials. They also examined the extent to which
curriculum materials and instructional strategies and how stu- teachers’ opportunities to participate in relevant professional edu-
dents take up new ideas as they participate in instruction in the cation was related to their reports of using instructional practices
context of what they already know and in the midst of the par- aligned with the reform.
ticular environment of school. Cohen and Hill concluded that although many teachers were
Lee’s focus on the instructional dynamic distinguishes her enthusiastic about the ideas associated with the reform, they were
analysis of cultural modeling from other studies that investigate also deeply reluctant to change their instruction and, when they
the problem of low academic achievement among African did adopt new practices, did so only in a piecemeal fashion.
American students by asking questions merely about students or Others adapted reformers’ ideas so that in practice they differed
teachers. Much of the research that pertains to the education of little from conventional forms of instruction. Cohen and Hill’s
underachieving students is inattentive to the instructional central finding was that the reform succeeded in changing teach-
dynamic. Examples of such work include studies of African ers’ practice to some degree but only when teachers had signifi-
American youth culture, of segregation among peer groups in cant opportunities to participate in professional development
middle schools, or of language use in rural Appalachian commu- that helped them understand the reform and learn new ways to
nities. They also include investigations of the effects of high teach. Only a small fraction of California elementary teachers had

534 EDUCATIONAL RESEARCHER

these opportunities, however, and the fragmented and incoher- compatible with professional views of good mathematics teach-
ent reform that resulted met with limited success. ing, she held other ideas about her own subject-matter knowledge
Cohen and Hill’s study suggests that educational innovation and about how she would learn to teach new topics that appeared
might be viewed in much the same way as classroom instruction: to inhibit her in learning what she needed to learn to provide a
The ideas associated with a reform do not act on their own to conceptually based explanation for the division-of-fractions algo-
change practice but are subject to various kinds of use. Reformers rithm. The researchers observed in interviews with her over the
function as instructors who represent new ideas to teachers, who course of the school year that she spoke increasingly about the
are the learners in that situation. The reformers design curriculum role of practice and of sources such as her students’ textbook in
materials that are more or less attentive to teachers’ prior beliefs and learning to teach and seemed to discount the value of her math-
practices, and they create or fail to create effective opportunities for ematics methods course at the university. Moreover, the
teachers to learn and practice with the new materials. Reform researchers found that the methods course itself seemed to com-
implementation can thus be conceived of as an educational process. pound this difficulty.
Studies of it can profitably investigate the multiple interactions that According to the observations of Borko et al., the instructor in
occur as reformers teach about a new program and learners learn the methods course moved too quickly through material,
about and adopt––or fail to adopt––the new ideas. assumed that student teachers in the course had already mastered
Cohen and Hill’s analysis of the California math reform the necessary mathematical knowledge, and failed to explain key
departs from conventional program evaluation in its attention to conceptual shifts that he made in explaining the division-of-
variability in implementation. Rather than rely on aggregate data fractions algorithm. He appeared unaware that students’ repeated
about the extent to which desired practices have been adopted requests for additional explanations of the invert-and-multiply
across sites, Cohen and Hill considered varying degrees of adap- rule might indicate confusion. The research team also found that
tation and reasons for the variation. The knowledge produced by the teacher education program as a whole may have contributed
an implementation study such as this one is potentially powerful to the student teacher’s difficulty. Borko et al. concluded that stu-
for reformers. It offers explanations for why reform succeeds or dent teachers’ reliance on independent practice and on “ideas that
fails and provides a basis for suggestions for more effective will work” was created by the multiple demands placed on them
implementation. by the design of the program, which left them with little time for
thoughtful engagement with course material or lesson planning.
Case 5: Preparing Mathematics Teachers The researchers also found that the program had allowed the stu-
The same attention to the instructional dynamic that the above- dent teacher in the case study to test out of a required course on
mentioned studies apply to classroom instruction and reform elementary mathematics concepts that might have helped her
implementation can also support research on teacher education. acquire mathematical knowledge fundamental for teaching. They
Questions about how teacher education affects K–12 pupil argue, in conclusion, that teacher education must challenge stu-
achievement and what qualities of student teachers matter are dent teachers’ fundamental beliefs about learning, teaching, and
fundamentally educational and require research designs that learning to teach.
attend to the instructional dynamic. We illustrate this point with As in the other studies that we have discussed here, this analy-
two additional cases, one that is small scale and qualitative and sis attends carefully to complex interactions among students,
the other, quite different, that is large scale and mixed method. instructors, and content in trying to explain how instruction
In the first, Borko and colleagues (1992) closely analyzed an works and, in this case, where it fails. Borko et al. demonstrate an
episode in which a student teacher was unable to provide an accu- appreciation for the reality that no single element of teacher edu-
rate, conceptually based justification for a division-of-fractions cation is responsible for the student teacher’s difficulty––not the
algorithm (the invert-and-multiply rule) in response to a student teacher’s extant knowledge of mathematics or beliefs
pupil’s question. This episode occurred in the context of the about teaching and not the content of methods courses, the skill
authors’ longitudinal study of the student teacher’s learning of teacher educators, or the structure and expectations of the
across courses and field experiences in her teacher education pro- teacher education program. Rather, they point out how these
gram. Surprised by the episode, which occurred late in their constituents of the instructional triangle interact with each other
study, the researchers looked for an explanation in the interac- in ways that either promote or inhibit learning. The researchers
tions among the student teacher, her instructors in the teacher target these interactions for close study, situating their work
education program, and the teacher education curriculum. They firmly in the domain of education.
were particularly attentive to how the student teacher’s knowl-
edge of mathematics and her beliefs about good teaching influ- Case 6: Investigating Pathways Into Teaching in
enced her willingness and ability to apply the instructional New York City Schools
strategies suggested in her teacher education program. A recently launched study of teacher education programs in New
Drawing on their extensive interviews, observations, and a York City by Boyd and colleagues (2006) exhibits a similarly
questionnaire, Borko and her colleagues analyzed the student instructional focus. This large-scale, mixed-method analysis
teacher’s knowledge of mathematics and beliefs about good examines features of the multiple pathways into teaching that
mathematics teaching, experience with university coursework, exist in New York City and the impact of those features on where
and classroom thinking and actions. They found that her teachers teach, how long they remain in the classroom, and stu-
knowledge of fractions was insufficient for teaching the content dent achievement in reading and math. Although Boyd and his
effectively and that, although she held some beliefs that were colleagues have not concluded their work in this study, their

DECEMBER 2007 535

research questions and methods demonstrate significant attention labor market, Boyd et al. establish their work within the domain
to the instructional features of teacher training and how those fea- of education research.
tures affect pupil achievement and the teacher labor market. In each of the six studies we have described here, researchers
A central premise of this Pathways study is that labels for indi- combine foundational knowledge of education––of its history
vidual teacher education programs, such as traditional and alter- and purposes, of the problems that plague it, and of the resources
native, may mask critical differences in program features and in that can contribute to its success––with an analytic perspective
the educational experiences that students have across programs. that foregrounds the multiple interactions that constitute the
The researchers argue that a complex set of interactions exists instructional dynamic in the situation. Their awareness of these
among teachers’ background characteristics, teachers’ opportuni- interactions and of the points at which they can be enhanced or
ties to learn in various programs, and features of the school con- disrupted and their ability to choose research questions that focus
texts in which prospective teachers complete field experiences and the investigation on the instructional dynamic highlight a dis-
ultimately choose to teach. They hypothesize that these interac- tinctive expertise.
tions, which vary from one program to the next, influence both
pupil outcomes and the teacher labor market. The researchers The Special Niche of Education Research
assert that such interactions need to be unearthed and closely Education is not the only research field that struggles to define its
examined if researchers and policy makers are to make well- special focus and to clarify the domains that researchers in it are,
informed decisions about whether––and how––different path- or should be, uniquely equipped to study. The field of organiza-
ways to teaching work. tional behavior, for example, suffers from similar problems.
To identify and understand these interactions and move beyond Heath and Sitkin (2001) found that most published work in the
blurry distinctions between types of programs, the research team field of organizational behavior emphasized the behavior of indi-
used interviews with program faculty, state and program docu- viduals and groups and was comparatively less attentive to the
ments, faculty surveys, and syllabi from methods courses to collect organizational aspects of organizational behavior. Most work in
data on more than a hundred teacher education programs located the field, they contended, addresses behavior that is common
in 18 colleges and universities. They also examined in detail spe- across many arenas—not particularly central to organizations or
cific characteristics of pathways into teaching in five areas, includ- organizing—and is studied by researchers in many other social
ing program structures; subject-specific preparation in reading and science fields. Heath and Sitkin argue that if most work ostensi-
math; preparation in learning and child development; preparation bly about organizational behavior is merely about interesting
to teach racially, ethnically, and linguistically diverse students; and behavior, then the field risks failing to address questions that are
the characteristics of field experiences. To understand students’ truly relevant to organizations. To examine whether and how a
experiences in these programs and to enable management of selec- study probes questions at the heart of organizational behavior,
tion bias, the researchers surveyed the 2003 cohort of entering pro- they deploy two tests. One appraises the degree to which a study
gram participants and the 2005 cohort of first-year teachers. They is central to understanding organizations; the second considers,
supplemented this information with administrative data, including for any given investigation, whether special expertise of organi-
demographic characteristics, salary, education, performance on cer- zational behavior is entailed.
tification exams, certification status, and career paths, from New The organizational centrality test asks, How much would we
York City’s comprehensive database. Finally, they collected back- understand about organizations if we understood everything
ground data and test scores for pupils in New York schools and used there was to know about (the topic proposed for study)?
them to construct value-added measures of student achievement. According to Heath and Sitkin (2001), if the answer is “not
Boyd et al. plan to use these data to peer into the black box of much,” then the topic is probably peripheral to the study of orga-
teacher education and weigh a range of inputs and outputs and nizational behavior. Education researchers could perform an
their interactions. In particular, they are attempting to distinguish analogous test. Researchers considering whether to study the cul-
among student teachers’ own characteristics, the features of indi- ture of African American peer groups in a middle school, for
vidual teacher education programs, and the characteristics of the example, might test the educational centrality of the study’s focus
schools in which student teachers work and to estimate the effects and discover that even if they learned everything there was to
of each of these variables on teachers’ career paths and on the edu- know about African American peer group culture, they would
cational outcomes of their students. These analyses are likely to still know little about the education of this particular group. It is
enable the researchers to consider questions such as which aspects not that the proposed study would be without merit; on the
of particular pathways into teaching affect the selection and contrary, it would be a topic of likely interest to psychologists,
recruitment of excellent teaching students, which aspects act inde- sociologists, and others interested in culture, social groups, and
pendently of students’ entering characteristics to improve their peer influences. The study’s results might have important impli-
ability to teach, and which are effective in preparing candidates for cations for education research. But the research itself would not
teaching in particular kinds of schools. Although at a much larger probe the special phenomenon that is education.
grain size than the study conducted by Borko et al. (1992), the Heath and Sitkin’s (2001) core competence test works similarly:
Pathways study is similarly attentive to the fact that teacher edu- Researchers may ask about a given study, Is this a topic about
cation consists of a set of interactions among instructors, learners, which organizational behavior researchers have unique insights
content, and environments. In choosing to train their analytic lens that are not likely to be shared by researchers in related social sci-
directly on those interactions as they search for explanations for ence disciplines such as psychology, sociology, political science,
differences in pupil achievement and fluctuations in the teacher or economics? If the answer is no, Heath and Sitkin argue, then

536 EDUCATIONAL RESEARCHER

perhaps the topic is not best studied by organizational behavior sought to uncover why some of these programs reduce recidivism
researchers. Scholars in education might ask themselves a similar and some do not. In doing so, she examined how program design
question to help focus their work on topics that are central to and administration interact with the needs of both prisoners and
education and that put their unique knowledge of and apprecia- prison staff and how these interactions can determine a program’s
tion for the importance of educational transactions at the core of success or failure. As Lin put it, program implementation consists
their investigations. If they do not do so, who will? of “the interaction of a program’s rules and resources with an
As we argue above, an instructional dynamic is present and of environment that filters, interprets, and makes use of those rules
central import not just in schools but in other organizations as and resources in different ways” (p. 35). Lin found that the most
well. Education research can, therefore, contribute a lens useful successful programs are those in which staff members encourage
to work in other domains; this perspective can offer leverage on prisoners’ participation, communicate openly with prisoners, and
complex dynamics with which many struggle. In the next section show a commitment to the programs and the quality of their
we discuss two studies, both from the field of political science, administration. Instructors in successful programs, like effective
that demonstrate the application of this perspective to problems instructors in more traditional educational situations, tailor their
outside education. teaching to individual prisoners’ needs and refuse to let struggling
participants drift.
Research That Applies an Educational Perspective Lipsky (1980) and Lin (2000) are but two examples of
to Problems Outside Education researchers who apply an educational perspective to problems
We claimed earlier that teachers make strategic decisions about outside education; although they are both in the field of political
the presentation of content and that the interactions among science, other examples exist across the social sciences. Some
teachers, learners, and ideas are of central import. This frame- social problems that could usefully be illuminated from an edu-
work applies not only to instances of explicit teaching and learn- cational perspective are instead investigated using other lenses.
ing but also to circumstances in which individuals in leadership The onus is on education researchers to define the central con-
or service-delivery positions interpret, represent, and administer tributions of their field and to promote greater awareness of how
programs, policies, and other content-rich materials. Researchers an educational perspective might prove useful to problem solving
from across the social sciences who find themselves addressing across multiple domains. We turn next to examine the essential
problems with an instructional dynamic might usefully apply role of schools of education in articulating the quintessential
such a framework to their analyses. Lipsky’s (1980) study of nature of research in education and in cultivating the expertise
human service organizations, Street-Level Bureaucracy, offers a demanded for such research.
helpful illustration.
Lipsky (1980) examines how “street-level bureaucrats”–– Implications for the Nature and Purposes
teachers, social workers, public interest lawyers, and police officers–– of Schools of Education
influence the implementation of public policy. His is an educational Schools of education should acknowledge a special responsibility
lens, although he does not label it as such. Viewing street-level to produce disciplined knowledge in education. They must be
bureaucrats as mediators and interpreters of policy, Lipsky investi- places that both house and prepare researchers with special
gates the fine-grained decisions they make about how to administer knowledge of the instructional dynamic and with the skills
policies and programs. In particular, he studies how these profes- required to study it. At the center of every school of education
sionals socialize citizens to certain expectations of government ser- must be scholars with the expertise and commitment necessary to
vices, determine eligibility for government benefits and sanctions, study educational transactions, whether they occur in schools or
and oversee the treatment that citizens receive in and through gov- elsewhere. These scholars must ensure that problems in education
ernment programs. Lipsky explains that “in a sense street-level benefit from sufficient scholarly attention and that younger
bureaucrats implicitly mediate aspects of the constitutional rela- scholars receive the training they need to develop core compe-
tionship of citizens to the state” (p. 4); he argues in conclusion that tence in education research. If schools and departments
these workers, rather than legislatures or high-ranking administra- of education do not make the development of knowledge in
tors, actually make policy. education––and thus the housing and preparation of education
Key to this study is Lipsky’s attention to how midlevel bureau- researchers––their primary reason for being, and if they do not
crats mediate policy. Just as classroom instructors or educational work actively to disseminate that knowledge among policy mak-
reformers strategically represent content to learners, the workers ers and members of the public, then educational problem solving
in his study decided how to make resources and opportunities will be left to researchers and professionals without the requisite
accessible to clients on the basis of their understanding of both expertise. And the core technology of education will remain inad-
program aims and client needs. Although the interactions among equately studied.
bureaucrats, programs, clients, and settings are not consciously One implication of this argument is that schools of education
instructional, Lipsky’s approach to studying the dynamic can be need to embrace unapologetically the worlds of both practice and
represented by the instructional triangle. scholarship. Educational practice is the core concern of education
Attention to a similar dynamic is evident in Lin’s 2000 study research, and those who study it must know it well. But educa-
of social policy in prisons, Reform in the Making. In this study, tion researchers must also arm themselves with the special ana-
Lin examined the implementation of rehabilitation programs in lytic skills that will allow them to usefully bridge the alleged
prisons, including literacy programs, high-school equivalency divide between theory and practice. It is along this divide that
classes, vocational training, and drug-treatment programs. She education researchers have special expertise.

DECEMBER 2007 537

 Ironically, analyses of the challenges and troubles so often must help beginning education researchers to (a) understand the
faced by schools of education during the past century frequently environments and problems that are the focus of education
point to this borderland as the source of the troubles. Critics research and the theoretical perspectives that enable productive
explain that the perception that education is an applied field and study of those environments and problems, (b) develop questions
that knowledge produced in and about it is “soft” and lacking in for study that are central to education and that capitalize on the
rigor has led schools of education to emulate the social sciences core competence of education researchers, and (c) learn to design
more than to define their own focus and niche. Clifford and and choose appropriate methods for education research. In this
Guthrie (1988), for example, argued that schools of education, scenario, prospective researchers who choose schools of education
particularly those located in research universities, have become for their professional training would do so because their primary
“ensnared improvidently” in the academic and political cultures interest is in acquiring training for studying educational prob-
of their universities and have tried to develop an “instant science lems. Those who are interested in educational questions but find
of education” that they hope will please their colleagues in disci- themselves drawn more clearly to the disciplinary perspective of
plinary departments but that often focuses on issues peripheral to a different social science field would choose training in another
instruction and to the complexity of life in schools. In the process, unit of the university.
Clifford and Guthrie assert, schools of education have failed to It is not the case, however, that only researchers who study
satisfy the needs and expectations of either the university or the instruction should call schools of education home. Disciplinary
schools where research might be used to improve classroom learn- scholars who work from perspectives outside education can make
ing. Labaree (2004) concurred; he contended that faculty mem- substantial contributions to educational problem solving and to
bers in education are torn between commitments to the public the preparation of education researchers. As we have tried to
schools and to the academy and are isolated from both, and that show, research in education is often enabled by studies that do
as a result they have continued to draw “unrelenting fire” from not investigate educational transactions but rather contribute
both sides. These analyses seem accurate. However, where some fundamentally important insights about elements of the instruc-
have found the position of education research at the intersection tional triangle. Many scholars whose work circles the instruc-
of theory and practice to be a near-fatal problem, we argue that tional dynamic but does not probe deeply inside it nonetheless
education researchers must claim and take advantage of the bor- maintain a focus on problems related to education, and the field
derland. It is precisely through work along this boundary that we has much to gain by housing them and supporting their work. In
can begin to address problems in education––and thus justify our addition, the school of education should serve as a resource to
occupation of university real estate. scholars in disciplinary departments across campus who choose
The production of education research and the cultivation of to investigate problems related to education.
education researchers are, in fact, the only purposes to which What we envision, then, is a porous and multidisciplinary
schools of education in research universities can lay exclusive school of education, one not unlike many that currently exist on
claim. Others have argued that the special function of a school of American campuses. We argue, however, that research focusing
education is to prepare teachers. Teacher education is indeed a on educational transactions should be the principal concern of
fundamentally important part of the mission of the school of edu- schools of education. It seems reasonable to assume that every
cation, and schools and departments of education in colleges and school of education should have a central core of researchers who
universities can make special contributions to the training of are unlikely to find a home in any other unit of the university and
teachers (see Ball & Forzani, 2007). A close inspection of the who produce significant work that would not be expected to
landscape of teacher preparation in the United States, however, appear elsewhere on campus.
makes clear that schools of education in research universities have Three elements thus constitute the central argument of this
no unique stake in teacher education. One reason is scale: article: (a) understanding the dynamic interactions among teach-
Teachers make up the single largest occupational group in the ers, students, content, and environments; (b) viewing phenom-
United States, and schools of education lack the capacity to pre- ena not usually considered education-related as instances of
pare sufficient numbers of teachers for American schools. Most teaching and learning and applying parallel questions and
students preparing to become teachers receive professional train- approaches in studies of those phenomena; and (c) viewing the
ing in a comprehensive university, in a liberal arts college, or, transactions that constitute education as uniquely distinguishing
increasingly, in a nonprofit organization such as Teach for research in education from other useful inquiries about or related
America. Furthermore, schools of education do not “own” to education. We began by pointing to the proliferation of
teacher preparation, even on their own campuses; prospective problems in education that confront not only the United States
teachers complete a significant portion of their study in other aca- but societies around the world; we conclude with the admonition
demic disciplines, making teacher preparation an all-university that if education researchers fail to articulate the nature and value
endeavor. Schools of education are, however, uniquely able to of specialized knowledge in education, then these problems will
produce disciplined knowledge in education and to maintain a be left to common sense or to professionals with expertise not pre-
robust agenda of research that might yield solutions to important cisely suited to the work.
problems. Several potential pitfalls and misunderstandings attend our
To claim and define the problem space that is uniquely theirs, argument. One is that we may be misinterpreted as suggesting
faculty members in schools of education must develop profes- that research with an educational perspective is only about class-
sional training deliberately designed to prepare the next genera- room instruction. Our use of the term instructional is deliberate,
tion of scholars to do high-quality work in the field. Such training for with it we signal the essential frame of the interactional

538 EDUCATIONAL RESEARCHER

dynamic represented by the instructional triangle. However, we and environments in student persistence; and examining critically
emphasize that we use this frame not only to represent classroom whether and how outreach activities are worthwhile. We argue that
instruction but also as a metaphor for the interactions at the core schools of education have a special role in addressing central prob-
of phenomena other than classroom teaching. And although we lems of educational improvement in higher education, in precol-
contend that instruction in schools is a central problem to which lege schooling, and in contexts other than formal institutional ones.
education research must attend, the usefulness of the instruc- The distinctive expertise that education schools can bring to the
tional perspective for studying interactions in other domains is a institutional mission is an opportunity to seize and develop. Doing
central part of our argument. so will require, however, a disciplined articulation of expertise in
A second potential pitfall of our argument is that it may be education and the courage to define it as such. It will also require
interpreted as a judgment that some studies are valuable and oth- focused development and extension, for the claim to expertise is
ers are not. We seek to name and highlight the importance of without value unless we can increasingly demonstrate what the
research inside educational processes because we argue that it is a instructional perspective offers to studies of teaching and learning
special contribution unique to education research. We do not as well as to other phenomena.
intend to diminish the value of learning more about individual
NOTE
elements of the instructional triangle, such as teachers’ beliefs,
school funding, or various curricula, nor do we wish to deny the An earlier version of this article was presented by the first author as
key contributions that such work makes to research in education. the invited Wallace Foundation Distinguished Lecture at the 2007
We hope that our efforts to bring the instructional perspective to annual meeting of the American Educational Research Association in
Chicago. We gratefully acknowledge Anne-Lise Halvorsen for com-
light will not lead some to conclude that we view other goals for
ments on an earlier draft of this article. Our ideas have also been deeply
research related to education as less worthy. influenced by the thinking of David K. Cohen, who has written exten-
Third, our argument is not intended to specify restrictively sively about policy making and implementation as instances of teaching
who belongs in a school of education. Educational phenomena and learning and with whom the first author has worked on ideas about
are usefully studied using tools and perspectives from other dis- instruction.
ciplines, and the interdisciplinary culture needed to support
inquiry into education depends on intellectual diversity among REFERENCES
the faculty. Too rare, however, are scholars steeped in the instruc- Anderson, C. W., & Roth, K. J. (1989). Teaching for meaningful and
tional perspective or whose specialization is instruction. Our self-regulated learning of science. In J. Brophy (Ed.), Advances in
argument would therefore support a closer look at the composi- research on teaching (Vol. 1, pp. 265–309). Greenwich, CT: JAI.
tion of education faculty to ensure that it reflects significant Ball, D. L., & Forzani, F. M. (2007). Teacher education in research uni-
versities: Staking out the territory. Manuscript in preparation.
expertise in the field’s unique area of competence.
Bestor, A. (1953). Educational wastelands. Urbana, IL: University of
Our argument also suggests questions for further investigation Chicago Press.
and elaboration. One centers on specialization in our field. What Borko, H., Eisenhart, M., Brown, C., Underhill, R., Jones, D., & Agard, P.
distinguishes educational psychology, for example, from other (1992). Learning to teach hard mathematics: Do novice teachers and their
areas of specialization in psychology? What characterizes this spe- instructors give up too easily? Journal for Research in Mathematics
cialization (see Calfee & Berliner, 1996)? What makes “educa- Education, 23, 194–222.
tional” sociologists, anthropologists, and historians different from Boyd, D. J., Grossman, P., Lankford, H., Loeb, S., Michelli, N. M., &
their disciplinary colleagues who do not study problems related Wyckoff, J. (2006). Complex by design: Investigating pathways into
to education? teaching in New York City schools. Journal of Teacher Education, 57,
A second important question is whether there is a set of core 155–166.
problems on which research in education especially focuses. Most Bransford, J. D. (1979). Human cognition: Learning, understanding, and
remembering. Belmont, CA: Wadsorth.
other fields can be mapped by a consistent set of questions and
Calfee, R. C., & Berliner, D. C. (1996). Introduction to a dynamic and
problems, but a look at the literature in education research or at relevant educational psychology. In D. C. Berliner & R. C. Calfee
the myriad subspecializations with which education scholars affil- (Eds.), Handbook of educational psychology (pp. 1–11). New York:
iate (“special interest groups”) suggests that we do not currently Macmillan Reference Books.
have such a set of core problems. Clifford, J. J., & Guthrie, J. (1988). Ed school: A brief for professional
A third important question is how our argument about education. Chicago: University of Chicago Press.
research relates to the professional education for which schools of Cohen, D. K., & Hill, H. C. (2001). Learning policy: When state educa-
education are responsible. Studies of professional education could tion reform works. New Haven, CT: Yale University Press.
constitute a key activity of faculties of education. Given this, Cohen, D. K., Raudenbush, S. W., & Ball, D. L. (2003). Resources,
schools might design programs of professional education not only instruction, and research. Educational Evaluation and Policy Analysis,
to prepare their own students for practice but also to serve as lab- 25, 119–142.
Conant, J. B. (1963). The education of American teachers. New York:
oratories for the study of teaching practice and teacher education.
McGraw-Hill.
Schools of education are rich with potential to explore and Finn, J. D., & Achilles, C. M. (1990). Answers and questions about class
develop this special competence in education. The larger institu- size: A statewide experiment. American Educational Research Journal,
tions that they inhabit are themselves educational institutions. As 27, 557–577.
such, schools of education house the core expertise for the entire Hawkins, D. (1974). I, thou, and it. In D. Hawkins, The informed vision:
enterprise––understanding how particular programs and interven- Essays on learning and human nature (pp. 48–62). New York:
tions work; studying the interplay among recruitment, support, Agathon. (Original work published 1967)

DECEMBER 2007 539

Heath, C., & Sitkin, S. B. (2001). Big-B versus big-O: What is organi- McDonald, J. P. (1992). Teaching: Making sense of an uncertain craft.
zational about organizational behavior? Journal of Organizational New York: Teachers College Press.
Behavior, 22, 43–58. Powell, A. G. (1980). The uncertain profession: Harvard and the search for
Judge, H. (1982). American graduate schools of education: A view from educational authority. Cambridge, MA: Harvard University Press.
abroad. New York: Ford Foundation. Schwab, J. J. (1978). The practical: Translation into curriculum. In I.
Koerner, J. D. (1963). The miseducation of American teachers. Baltimore: Westbury & N. J. Wilkof (Eds.), Science, curriculum, and liberal edu-
Penguin. cation: Selected essays (pp. 365–383). Chicago: University of Chicago
Labaree, D. F. (2004). The trouble with ed schools. New Haven, CT: Yale Press.
University Press. Smith, E. L. & Anderson, C. W. (1984). Plants as producers: A case
Lampert, M. (2001). Teaching problems and the problems of teaching. study of elementary science teaching. Journal of Research in Science
New Haven, CT: Yale University Press. Teaching, 21, 685–698.
Lee, C. D. (2007). Culture, literacy, and learning: Taking bloom in the
midst of the whirlwind. New York: Teachers College Press. AUTHORS
Leinhardt, G., Zigmond, N., & Cooley, W. W. (1981). Reading
instruction and its effects. American Educational Research Journal, 18, DEBORAH LOEWENBERG BALL is William H. Payne Collegiate
343–361. Professor in Education at the University of Michigan, 610 East
Levine, A. (2006). Educating school teachers. Washington, DC: University Avenue, 1110 School of Education Building, Ann Arbor, MI
Education Schools Project. 48109–1259; dball@umich.edu. Her areas of specialization include the
Lin, A. C. (2000). Reform in the making: The implementation of social pol- study of efforts to improve teaching through policy, reform initiatives,
icy in prison. Princeton, NJ: Princeton University Press. teacher education, and mathematical knowledge for teaching.
Lipsky, M. (1980). Street-level bureaucracy: Dilemmas of the individual
in public services. New York: Russell Sage Foundation. FRANCESCA M. FORZANI is a doctoral student in education at the
Lyon, R. (2002, November). The key to progress in education? Lessons University of Michigan, 610 East University Avenue, 1228 School of
from medicine, welfare, and other fields. Paper presented at the policy Education Building, Ann Arbor, MI 48109–1259; fforzani@umich.edu.
forum “Rigorous Evidence: The Key to Progress in Education,” Her areas of specialization include the design, implementation, and eval-
Washington, DC. uation of reform initiatives, particularly in the area of teacher education.

540 EDUCATIONAL RESEARCHER