LAND USE TRANSPORTATION INTERACTION:

AN EXAMINATION OF THE 1995 NPTS DATA

By

Catherine L. Ross, Ph.D.

Professor of City Planning

and

Anne E. Dunning
Graduate Research Assistant

Georgia Institute of Technology

Graduate City Planning Program
College of Architecture
Atlanta, Georgia 30332-0155
USA

Prepared for:

U.S. Department of Transportation

Federal Highway Administration

October 1997
 Land Use and Transportation Interaction:
An Examination of the 1995 NPTS Data

Table of Contents

EXECUTIVE SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
INTRODUCTION AND OVERVIEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
Summary of Literature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
Key Terms and Definitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Edge City, Second City, and Area Type 9
Transit Availability 10
Urban Sprawl 10
TRENDS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Comparison to Historical NPTS Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
New Variables Available for Land Use Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
CONTRIBUTING ELEMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Measures for People . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Population Density 13
Median Household Income/Poverty 17
Race and Hispanic Origin 20
Age 22
Education 25
Measures for Places . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
Area Type 25
Residential Density 30
Age of Housing 36
Housing Tenure 37
Measures for Employment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
Employment Density 39
Retail Employment 42

1
FINDINGS AND CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
Measures for People . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
Measures for Places . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
Area Type 45
Residential Density 46
Age of Housing 47
Housing Tenure 47
Measures for Employment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
OTHER RESEARCH . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49

2
 Index of Tables
TABLE 1: MILES DRIVEN LAST YEAR BY POPULATION DENSITY AND GENDER . . . . . . . . . . . . . . . . 13
TABLE 2: DRIVERS PER ADULT BY POPULATION DENSITY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
TABLE 3: VEHICLES PER ADULT BY POPULATION DENSITY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
TABLE 4: ONE-WAY WORK TRIP BY POPULATION DENSITY AND GENDER . . . . . . . . . . . . . . . . . . . . . 15
TABLE 5: TRANSIT AVAILABILITY BY POPULATION DENSITY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
TABLE 6: DISTANCE TO TRANSIT FROM THE HOUSEHOLD BY POPULATION DENSITY . . . . . . . . . . 16
TABLE 7: MODE OF TRANSPORTATION BY POPULATION DENSITY . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
TABLE 8: ANNUALIZED INDIVIDUAL TRAVEL BEHAVIOR BY POPULATION DENSITY . . . . . . . . . . . 17
TABLE 9: BLOCK GROUP MEDIAN HOUSEHOLD INCOME BY AREA TYPE . . . . . . . . . . . . . . . . . . . . . . 17
TABLE 10: TRANSIT AVAILABILITY BY BLOCK GROUP MEDIAN HOUSEHOLD INCOME . . . . . . . . . 19
TABLE 11: DISTANCE TO TRANSIT FROM HOUSEHOLD BY POVERTY STATUS . . . . . . . . . . . . . . . . . . 19
TABLE 12: ANNUALIZED INDIVIDUAL TRAVEL BEHAVIOR BY HOUSEHOLD INCOME . . . . . . . . . . . 20
TABLE 13: RACE BY AREA TYPE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
TABLE 14: TRANSIT AVAILABILITY BY RACE OR HISPANIC ORIGIN . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
TABLE 15: MODE OF TRANSPORTATION BY RACE OR HISPANIC ORIGIN . . . . . . . . . . . . . . . . . . . . . . . 21
TABLE 16: AGE BY AREA TYPE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
TABLE 17: FAMILY LIFE CYCLE BY AREA TYPE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
TABLE 18: ONE-WAY WORK TRIP BY AGE AND GENDER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
TABLE 19: TRANSIT AVAILABILITY BY FAMILY LIFE CYCLE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
TABLE 20: DISTANCE TO TRANSIT BY FAMILY LIFE CYCLE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
TABLE 21: TRANSIT AVAILABILITY BY EDUCATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
TABLE 22: DRIVERS PER ADULT BY AREA TYPE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
TABLE 23: VEHICLES PER ADULT BY AREA TYPE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
TABLE 24: WORK LOCATION BY AREA TYPE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
TABLE 25: ONE-WAY WORK TRIP BY AREA TYPE AND GENDER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
TABLE 26: TRANSIT AVAILABILITY BY AREA TYPE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
TABLE 27: DISTANCE TO TRANSIT FROM THE HOUSEHOLD BY AREA TYPE . . . . . . . . . . . . . . . . . . . . 28
TABLE 28: AUTOMOBILE COMMUTING BY AREA TYPE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
TABLE 29: ANNUALIZED INDIVIDUAL TRAVEL BEHAVIOR BY AREA TYPE . . . . . . . . . . . . . . . . . . . . 30
TABLE 30: BLOCK GROUP RESIDENTIAL DENSITY BY AREA TYPE . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
TABLE 31: MILES DRIVEN LAST YEAR BY RESIDENTIAL DENSITY AND GENDER . . . . . . . . . . . . . . . 31
TABLE 32: ONE-WAY WORK TRIP BY RESIDENTIAL DENSITY AND GENDER . . . . . . . . . . . . . . . . . . . 32
TABLE 33: TRANSIT AVAILABILITY BY RESIDENTIAL DENSITY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
TABLE 34: DISTANCE TO TRANSIT FROM THE HOUSEHOLD BY RESIDENTIAL DENSITY . . . . . . . . . 33
TABLE 35: MODE OF TRANSPORTATION BY RESIDENTIAL DENSITY . . . . . . . . . . . . . . . . . . . . . . . . . . 33

3
TABLE 36: ANNUALIZED INDIVIDUAL TRAVEL BEHAVIOR BY RESIDENTIAL DENSITY . . . . . . . . . 34
TABLE 37: WORK LOCATION BY RESIDENTIAL DENSITY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
TABLE 38: EMPLOYMENT DENSITY BY RESIDENTIAL DENSITY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
TABLE 39: AGE OF HOUSING BY AREA TYPE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
TABLE 40: BUS AVAILABILITY FOR RECENT BUILDS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
TABLE 41: PERCENTAGE OF RENTER-OCCUPIED HOUSING BY AREA TYPE . . . . . . . . . . . . . . . . . . . . 37
TABLE 42: TRANSIT AVAILABILITY BY HOUSING TENURE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
TABLE 43: DISTANCE TO TRANSIT BY HOUSING TENURE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
TABLE 44: WORK TRACT EMPLOYMENT DENSITY BY HOME BLOCK GROUP AREA TYPE . . . . . . . 39
TABLE 45: MILES DRIVEN LAST YEAR BY EMPLOYMENT DENSITY AND GENDER . . . . . . . . . . . . . . 41
TABLE 46: ONE-WAY WORK TRIP BY EMPLOYMENT DENSITY AND GENDER . . . . . . . . . . . . . . . . . . 41
TABLE 47: RETAIL TRADE BY AREA TYPE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
TABLE 48: MILES DRIVEN LAST YEAR BY RETAIL EMPLOYMENT AND GENDER . . . . . . . . . . . . . . . . 43
TABLE 49: ONE-WAY WORK TRIP BY RETAIL EMPLOYMENT AND GENDER . . . . . . . . . . . . . . . . . . . . 43

Index of Figures
FIGURE 1: PERSON TRIPS BY POPULATION DENSITY FOR 1990 AND 1995 . . . . . . . . . . . . . . . . . . . . . . 11
FIGURE 2: MODE OF TRANSPORTATION BY POPULATION DENSITY . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
FIGURE 3: MEDIAN HOUSEHOLD INCOME BY AREA TYPE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
FIGURE 4: DISTANCE TO TRANSIT BY AREA TYPE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
FIGURE 5: ANNUALIZED INDIVIDUAL TRAVEL BEHAVIOR BY RESIDENTIAL DENSITY . . . . . . . . . . 34
FIGURE 6: WORK TRACT EMPLOYMENT DENSITY BY HOME BLOCK GROUP AREA TYPE . . . . . . . . 40

4
EXECUTIVE SUMMARY
There is currently a great deal of discussion about the interaction between land use and
transportation. The 1995 NPTS provides some ability to investigate this question through the
inclusion of variables that measure the interaction of land use and travel behavior. Population
density is the primary quantifiable land use descriptor variable. Population density has been
further manipulated to isolate area types (urban, second city, suburban, town and rural). Other
variables that attempt to quantify land use include residential density and work tract employment
density. Characteristics of the population or built environment such as race, age, income, and
retail employment further identify land use impacts across different population groups.

Greater population density is associated with decreasing annual miles driven, greater bus
availability, decreased dependency on single occupancy vehicles and increased use of transit.
The private automobile is still the dominant mode of travel although African Americans, Asians
and Hispanics are slightly more likely to use other modes of transportation.

Increasing population density is associated with fewer person trips, fewer person miles traveled,
and fewer person miles per trip. Residents of densely populated areas report the fewest vehicle
trips, vehicle miles traveled, and vehicle miles per trip. Less densely populated areas tend to
have more drivers per adult and more vehicles per adult.

Second cities tend to follow national averages with regard to several transportation parameters,
for example, drivers per adult, vehicles per adult, percent of persons working from home, and
auto-dependency. Approximately 20% of second city residents go to work by a mode other than
the private automobile. Residents of second cities report the highest number of person trips of
any area type. Persons in suburban areas make the next highest number of person trips. A
surprisingly high number of low-income residents live in second cities, which have limited
transit availability.

Results of the 1995 NPTS identify the locational preferences of specific segments of the
population. High-income households generally tended to locate in suburban areas while
middle-income households are most often found in rural areas. Low-income households are
generally found in urban or rural areas.

Distance to work and travel time to work decrease as the percentage of retail trade in an area
increases. Urban areas have the smallest percentage of residents working in census tracts with
over 25% participation in retail trade. Second cities have the highest percentage with 28.8% of
residents working where more than 25% of jobs are in retail trade. Retail employment and
employment density at the work census tract have some measurable correlations to travel
behavior.

At the home block group, increasing housing density is associated with greater transit availability
and closer proximity to transit. Bicycle and walk trips increase as residential density increases.
Increasing residential density is also associated with increasing employment density. At
residential densities between 100 and 1,499 housing units per square mile, people are less likely

5
 to work at jobs with no fixed workplace. Low residential density areas have the largest
percentage of people working at home.

Residential density, retail employment, income, area type, and population density all provide
important descriptors for transportation behavior and policy implementation. This National
Personal Transportation Special Report carefully examines these and other aspects of people,
places and employment that may link land use to transportation choices and behavior. Questions
underlaying this analysis of that link include:

· What is the relationship between vehicle availability and urban sprawl?

· How do people travel in edge cities?
· How do population density, employment, access to goods and services, and
transit availability affect household travel behavior?
· What land use characteristics at the residence and/or workplace end seem to be
the best predictors of travel behavior?
· What impact does urban sprawl or dispersion have on travel behavior and
transportation investment costs?
· What is the impact of edge cities on travel behavior?
· Have urban areas developed in ways that require us to travel in private vehicles
and necessitate long vehicle trips (and vehicle emissions)?
· Do higher residential densities offer some chance of reducing vehicle trips and
emissions?
· Does transit accessibility change people's travel behavior for all trips or only the
work trip (peak period transit service vs. off-peak service)?

INTRODUCTION AND OVERVIEW

Transportation professionals increasingly look to land use as a possible explanatory factor of
transportation behavior. The Federal Highway Administration (FHWA) designed the 1995
National Personal Transportation Survey to include several variables representing land use. The
resulting data provides a basis to quantitatively explore land use and transportation interaction.

Summary of Literature
The following is a survey of current literature concerning the effect of land use on transportation.
Studies that explore this relationship can help further our understanding of travel patterns and
travel behavior now and in the future.

Pushkarev and Zupan's (1977) study on optimum density for transit types found that both high
residential density and the high density and relative size of the trip-end destination (workplace)

6
are major determinants of public transportation use. The study also concluded that clustering
nonresidential floor-space in central business districts and placing moderate to high density
residences (7 to 15 dwellings per acre) close to those clusters was the most effective in
promoting transit use.

The reality of development through the latter part of this century is quite the opposite of that
pattern. Low density and a doughnut hole of population and employment density in city centers
increasingly characterize modern cities. Policies such as the Federal Highway Acts and the
Standard Zoning Enabling Acts have drastically affected land use, expanding housing and
employment into suburban areas. Instead of the Central Business District (CBD) containing the
vast majority of a region’s office floor-space, many new clusters of office buildings have sprung
up in suburban areas (Pivo, 1990). Instead of dense clusters of buildings, as were found in the
street grid of the traditional downtown, these suburban office complexes are spaced far apart with
vast expanses of parking acreage in between. Often, the new complexes offer more real space for
cars than for the people who drive them, and mass transit is atypical in these areas (Leinberger
and Lockwood, 1986). This transit and pedestrian unfriendly environment, coupled with the fact
that these complexes were designed as single-use centers, means shopping, dining, and other
day-to-day activities tend to be accessible by auto.

In addition, recent years have brought an increasing awareness of the trend of American cities to
form nodes of urban activity in the midst of suburbs surrounding central cities. These nodes have
transferred travel activity from radial activity focused on the concentrated central core of a city to
tangential movement between the outer nodes. Joel Garreau’s definitive book, Edge City: Life on
the New Frontier, characterizes these nodes as edge cities and explains this growing
phenomenon.

According to research conducted in the 1980’s, the migration of white-collar office and service
job centers to the suburbs resulted in an increase rather than a decrease in travel time and
distance to work. Robert Cervero (1989) contends this is an outgrowth of "jobs-housing spatial
imbalance" brought on by factors beyond the simple lack of land-use planning. Possible causes
include fiscal and exclusionary zoning, two wage-earner households tending to locate close to
one workplace and not the other, and the fast pace of job-turnover coupled with an unwillingness
to relocate close to a new job (Cervero, 1989).

In contrast, Gordon and Richardson emphasized in the 1990 NPTS Special Report "Geographic
Factors Explaining Worktrip Length Changes" that average work trip duration either fell slightly
or grew by much smaller percentages than distances. The suburbanization of jobs and residences
has allowed people to live away from activity centers and use roads with less congestion than city
streets. With longer distances but less congestion, travel time has not suffered from sprawl1 .

Another study of five communities in the San Francisco Bay area did not focus explicitly on trip
length but looked instead at the number of trips by mode. A primary finding was that land use
characteristics of the neighborhoods (where person trips were generated) were not associated
with number of person trips made, but were associated with transit and non-motorized trips
(Kitamura et al, 1997). High density was found to be associated with lower fractions of auto

7
trips, and higher percentages of non-motorized trips. a community was found to be statistically
correlated with an increase in non-motorized trips. Eight attitudinal factors were entered into the
analysis. The factors included pro-environment, pro-transit, automotive mobility, time pressure,
and urban form and added increasing explanatory power to the models used to predict travel
mobility. This led researchers to conclude that "attitudes are at least more strongly, and perhaps
more directly associated with travel than are land use characteristics." (Kitamura et al, p. 154).

Many studies have shown similar findings with regard to density and its correlation to transit
usage versus auto usage and also identified other elements which contribute to transportation
mode choice. In addition to low densities and ample free parking, suburban business areas are
characterized by a single dominant land use: office space. It is believed that mixed-used
developments, combining offices, shops, restaurants, banks and other activities may be important
to relieving automobile congestion by reducing the number of trips. In pedestrian-friendly
mixed-use suburban activity centers, it is hypothesized that walking can take the place of noon-
or peak-hour auto trips to conduct errands.

In her review of density/travel pattern literature, Ruth Steiner identifies assumptions that underlie
the views of the proponents of high-density, mixed-used land use patterns (Steiner, 1994). These
assumptions include:

· People are willing to move into high density developments

· Travel patterns will change once people locate in a high density development
· People in high density developments will make fewer and shorter auto trips
· People in high density developments will walk and use transit more frequently

Another study attempting to account for both density and socioeconomic makeup came to the
conclusion that "population density, employment density, and land-use mix are related to mode
choice [even] when non-urban-form [socioeconomic] factors are controlled" (Frank and Pivo,
1994). The study went on to test the hypothesis that the relationship of population density,
employment density and mode choice is non-linear, enabling the identification of thresholds of
density where shifts from one mode (auto) to others (transit or walking) occur. Significant shifts
from auto use to walking or transit occur at certain employment density levels (20-75 employees
per acre, and at > 125 employees per acre). For shopping trips, population densities need to
exceed 13 persons per acre before a significant shift from auto use to walking or transit occurs
(Frank and Pivo, 1994).

Key Terms and Definitions

Several conventions were developed to facilitate research with the 1995 NPTS data. These
conventions include the following definitions and explanations.

8
Edge City, Second City, and Area Type
Joel Garreau defined five factors that determine an edge city:

· "Has five million square feet or more of leasable office space--the workplace of
the Information Age,
· "Has 600,000 square feet or more of leasable retail space,
· "Has more jobs than bedrooms,
· "Is perceived by the population as one place,
· "Was nothing like 'city' as recently as thirty years ago2."

People often know where these edge cities exist in their own states, but quantitatively defining an
edge city for the purposes of the NPTS poses a challenge. NPTS variables deal primarily with
people, rather than spaces; hence, population, household, and employment densities can be used
to explain these urban phenomena, rather than floor space and community perceptions.

David R. Miller and Ken Hodges of Claritas, Inc. established a standard for defining urbanization
categories using relational population densities3. Under this system, Claritas defines a grid
system across the United States based on 1/30th of a degree latitude and longitude, which
amounts to roughly 900,000 cells of about four square miles each. The total population of a
given cell and its eight surrounding cells (a 3x3 grid) divided by the total area of all nine cells
determines the given cell's grid density. Claritas then ranks all of the grid cell densities for the
nation into one hundred equal groups (a scale of 0 to 99).

The highest grid cell density in a 5-mile radius (5x5 grid, excluding the corners) determines the
local density maximum in an area. Population centers emerge where grid cell densities only
decrease moving away from a local maximum and no other local maximum with a greater
density appears in closer proximity.

Area type classifications depend on the calculated grid cell densities and population center
densities. Simple grid cell densities define rural areas (grid cell densities less than or equal to 19)
and small towns (grid cell densities greater than or equal to 20 and less than or equal to 39). This
classification results in groupings similar to the groups created by the Urbanized Area definition
of 1,000 persons per square mile minimum. Claritas associates population center densities
greater than 79 with urban areas; second cities comprise remaining population center densities.
Areas around second city and urban areas form suburban areas. Lines of different slopes
distinguish suburban areas around the population centers of second cities and urban areas.

9
 Area Type Determination Calculations

Area Type Determination Calculation

Rural Area GCD # 19
Town 20 # GCD # 39
Urban Area PCD $ 79 (urban population center)
and
GCD $ 40 (not town or rural)
and
GCD $ 0.80 @ PCD +9.8
Second City PCD < 79 (not an urban population center)
and
GCD$ 40 (not town or rural)
and
GCD $ 1.7368 @ PCD - 64.208
Suburban Area GCD$ 40
and
Area û Urban Area
and
Areaû Second City

GCD = Grid Cell Density Source: "A Population Density Approach to

PCD = Population Center Density Incorporating an Urban-Rural Dimension into Small
Area Lifestyle Clusters" by Miller and Hodges

Second cities differ from Garreau’s edge cities in that second cities can be quantitatively defined
and rely entirely on contextual population densities; whereas, edge cities receive their
classifications from community perceptions and measurements of space. Existing political
definitions of local borders do not affect the NPTS area type classification system.

Transit Availability
Transit availability is defined as bus availability. The 1995 NPTS assumes that the bus is the
basic form of transit. Streetcar, subway, and commuter rail are assumed to exist only where a
bus system has been established.

Urban Sprawl
Urban sprawl describes the tendency for people who associate themselves with an urban center to
live farther and farther away from that urban center. Sprawl is difficult to define quantitatively.
The area type coding provided by Claritas offers a good proxy for sprawl. Population density
defines the edge of an urban area’s impact on population, as opposed to political boundaries
which may not indicate the true form of population dynamics. Suburban areas, as defined by
Claritas, are assumed to be associated with an urban area or a second city. Suburbs can,
therefore, be classified as the sprawled outer edges of the urban area. In this analysis, travel
behavior found in suburbs and second cities represent the effects of urban sprawl.

10
TRENDS
The 1995 Nationwide Personal Transportation Survey provided groundbreaking precedent to
provide new ways of exploring the effects of land use on travel. The new land use survey
questions, combined with improvements such as travel diaries, establishes a standard for future
studies.

Comparison to Historical NPTS Data

Previous NPTS surveys have provided data on characteristics such as area densities, populations,
and differences between central cities and areas outside central cities. Population density
provides the greatest comparison between past surveys and the 1995 NPTS

FIGURE 1: PERSON TRIPS BY POPULATION DENSITY FOR 1990 AND 1995

5.00

4.42 4.42 4.35

Trips per Day per Person

4.50
4.15
4.04
4.00

3.50 3.37
3.14 3.27 3.17
3.00 2.82

2.50

2.00
0 to 249 250 to 999 1,000 to 4,000 to 10,000 & up
3,999 9,999
1995 Population Density (People/Square Mile)
1990

In the 1990 NPTS special report "Travel by Households without Vehicles4," Charles Lave and
Richard Crepeau found that the number of person trips per day for the total NPTS sample peaked
at population densities between 250 and 999 people per square mile. Data values between 1990
and 1995 show an overall increase in the number of trips people took across all population
densities. The numbers are difficult to compare, but comparison of person trips to population
density remains remarkably similar. In 1995, people tended to make more person trips per day in
medium-density areas.

11
 FIGURE 2: MODE OF TRANSPORTATION BY POPULATION DENSITY

100%
90%

80%
70%
60%
Percent People

50%

40%
30%
20%

10%
0%
0 to 249 250 to 999 1,000 to 4,000 to 10,000 & All
3,999 9,999 up
Population Density (People/Square Mile)

Private Vehicle Public Transit Taxi Bicycle/Walk

Mode choice trends have also remained consistent in the 1995 NPTS. In the 1990 Special Report
“Recent Nationwide Declines in Carpooling5,” Erik Ferguson found trends of decreasing private
vehicle use as population density increases. In addition, transit use increased as population
density increased. The data in Figure 2 indicate that these trends remain constant in the 1995
NPTS.

New Variables Available for Land Use Study

Beyond population density, the 1995 NPTS began exploring more aspects of the developed
environment than previous surveys. Several census categories can be applied to the NPTS data
to offer more information on social characteristics. This report focuses on the following land use
and population characteristics:

12
 Measures for People Measures for Places Measures for Employment
Population density Area Type Employment density
Income Residential density Retail employment
Poverty status Age of Housing
Race/ethnicity mix Housing tenure
Hispanic origin
Age
Educational attainment
Retail employment

The 1995 NPTS also includes self-reports of transit accessibility, household vehicle availability,
and customer evaluations of highway and public transportation.

CONTRIBUTING ELEMENTS
The issues, terms, methodologies, and trends discussed to this point all contribute to the analysis
of the 1995 Nationwide Personal Transportation Survey data. The previous section identified
historical trends in NPTS data. The literature review has established the current background of
intellectual debate regarding land use and transportation. Using these contexts and the concepts
of the key terms defined earlier, this report will now employ the new variables available for land
use study to analyze the interaction of land use and transportation. This section divides these
analyses into categories of measures for people, places, and employment.

Measures for People

Population Density
Traditionally, analysts have used population density and MSA size to measure the effects of land
use on different aspects of transportation. Population density provides a good indicator, for
instance, of annual miles driven.

TABLE 1: MILES DRIVEN LAST YEAR BY POPULATION DENSITY AND GENDER

Annual Miles Driven

Male Female
People per Mile2 Mean Median Mean Median
0 to 249 17,991 14,000 10,607 9,000
250 to 999 17,670 15,000 10,288 9,000
1,000 to 3,999 15,415 12,000 8,976 8,000
4,000 to 9,999 14,316 12,000 8,307 6,500
10,000 & up 11,479 9,000 7,276 5,000

The first table shows that high population densities are associated with driving fewer miles
annually. Males typically drive 1.5 to nearly 2 times as many miles as females do, but the
correlation between density and annual miles driven holds true for both genders at all population

13
 densities. Presumably, low population density is associated with increased distance between
destinations and greater miles driven each year.

TABLE 2: DRIVERS PER ADULT BY POPULATION DENSITY

People per Mile2

Drivers per Adult 0 to 249 250 to 999 1,000 to 3,999 4,000 to 9,999 10,000 & up Total
Less than One 10.90% 10.20% 12.60% 15.70% 36.80% 15.80%
One Driver 82.80% 84.80% 82.40% 81.10% 62.20% 79.90%
More than One 6.20% 5.00% 5.00% 3.20% 0.90% 4.30%
Total 100.00% 100.00% 100.00% 100.00% 100.00% 100.00%

Areas of high population density do not follow the same trends in drivers per adult as block
groups with lower densities. The most densely populated areas have the highest percentage of
residents with less than one driver per adult. In areas with population densities above 10,000
people per square mile, approximately 36.8% of residents have less than one driver per adult.
This ratio differs greatly from the average of 15.8% across all density categories. For density
levels between 4,000 and 9,999 people per square mile, 15.7% of the people have less than one
driver per adult. In contrast, 82.8% of the people have one driver per adult in the 0 to 249 people
per square mile density level while only 62.2% have one driver per adult at population densities
above 10,000 people per square mile.

TABLE 3: VEHICLES PER ADULT BY POPULATION DENSITY

People per Mile2

Vehicles per Adult 0 to 249 250 to 999 1,000 to 3,999 4,000 to 9,999 10,000 & up Total
Less than One 17.10% 17.60% 20.00% 27.10% 53.60% 25.10%
One Vehicle 57.50% 63.30% 64.80% 61.20% 41.50% 59.10%
More than One 25.50% 19.00% 15.20% 11.70% 4.90% 15.80%
Total 100.00% 100.00% 100.00% 100.00% 100.00% 100.00%

The number of vehicles per adult follows trends similar in most respects to those patterns set by
the number of drivers per adult. Population densities of over 10,000 people per square mile have
the highest percentage (53.6%) of adults with less than one vehicle. At density levels under 250
people per square mile, only 17.1% of adults have less than one vehicle. Conversely, 25.5% of
adults in the lowest-density areas have more than one vehicle, but only 4.9% of adults living in
population densities above 10,000 people per square mile own multiple vehicles. Across all
density levels, an average of 15.8% of all adults have more than one vehicle.

14
 TABLE 4: ONE-WAY WORK TRIP BY POPULATION DENSITY AND GENDER

Distance to Work (Miles) Time to Work (Minutes)

People per Male Female Male Female
Mile2 Mea Media Mea Media Mea Media Mea Media
n n n n n n n n
0 to 249 17 12 13 10 24 20 20 15
250 to 999 17 10 12 8 24 20 20 15
1,000 to 3,999 14 9 11 7 22 17 19 15
4,000 to 9,999 12 8 9 6 23 20 20 15
10,000 & up 11 7 9 5 26 20 26 20

Table 4 shows that people living in low-density areas generally travel longer distances to work,
and their commute times are longer than the commute times of their higher population density
counterparts. As population density increases, commute times and distances decrease slightly
where population densities are less than 10,000 people per square. At densities greater than
10,000 people per square mile, distances continue to decrease, but trip times suddenly increase.
This increase likely indicates that short distances cannot alleviate long commute times in densely
populated and congested areas. An alternative explanation is that this increase reflects the
additional travel time associated with transit use.

At all density levels, women have shorter commute distances and times, indicating that
households are located closer to where women work than to where men work. It is not clear
whether households locate closer to where women work or if women find jobs closer to home.

TABLE 5: TRANSIT AVAILABILITY BY POPULATION DENSITY

Transit Availability
2
People per Mile Bus Service No Bus All
Available
0 to 249 20.1% 79.9% 100.0%
250 to 999 41.0% 59.0% 100.0%
1,000 to 3,999 69.4% 30.6% 100.0%
4,000 to 9,999 88.8% 11.2% 100.0%
10,000 & up 98.0% 2.0% 100.0%
Total 63.4% 36.6% 100.0%

As shown in Table 5, transit (bus) availability increases with increased population densities. In
the least densely populated areas, bus service is available to only 20.1% of the population. This
percentage increases over fourfold to 98.0% in areas with population densities of 10,000 people
per square mile and greater.

15
 TABLE 6: DISTANCE TO TRANSIT FROM THE HOUSEHOLD BY POPULATION DENSITY

Distance to People per Mile2

Transit 0 to 249 250 to 999 1,000 to 3,999 4,000 to 9,999 10,000 & up All
Less than .1 mile 18.5% 20.1% 26.0% 38.4% 57.9% 36.0%
.1 to .24 mile 2.4% 5.6% 13.0% 17.4% 18.3% 14.3%
.25 to .49 mile 3.0% 6.5% 10.4% 13.3% 11.2% 10.8%
.5 to .99 mile 18.7% 29.6% 35.1% 25.2% 11.3% 25.1%
1 mile & up 57.4% 38.2% 15.5% 5.7% 1.3% 13.8%
Total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0%

As shown in Table 6, the most densely populated areas have transit located most closely to the
household. For areas with population densities of 4,000 people per square mile and greater, the
largest share of transit is located within .1 mile of the household. As population density
decreases, the distance from transit to the residence increases; this is true except for transit
located less than .1 mile from the household. People living in the least densely populated areas
live farthest from transit, with over half of transit located at least .5 mile away from the
household.

TABLE 7: MODE OF TRANSPORTATION BY POPULATION DENSITY

People per Mile2

Mode 0 to 249 250 to 999 1,000 to 3,999 4,000 to 9,999 10,000 & up All
Private Vehicle 93.1% 93.3% 92.0% 89.6% 69.4% 89.3%
Public Transit 3.5% 2.9% 3.1% 3.0% 11.0% 4.0%
Taxi 0.0% 0.1% 0.1% 0.2% 1.0% 0.2%
Bicycle/Walk 3.3% 3.8% 4.8% 7.2% 18.5% 6.5%
Total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0%

Beyond describing trip characteristics, population density also affects mode choice preferences.
As shown in Table 7, the private vehicle dominates as the preferred mode of transportation.
Between 4,000 and 9,999 people per square mile, people use private vehicles 89.6% of the time.
Above 10,000 people per square mile, private vehicle utilization drops dramatically to 69.4%.
Areas with population densities less than 250 people per square mile possess the highest share of
private vehicle usage, which may be attributable to the few mode choice options available in
low-density areas. In contrast, high population density reduces the private vehicle’s popularity.
Usage of alternative modes of transportation drastically increases for population densities over
10,000 people per square mile, while private vehicle utilization drops by roughly 25% to 69.4%.
Notably, bicycling and walking (18.5%) outperforms public transit (11.0%) at the highest
density. This demonstrated preference merits further exploration of investments for urban
pedestrian environments and bicycle right-of-way.

16
 TABLE 8: ANNUALIZED INDIVIDUAL TRAVEL BEHAVIOR BY POPULATION DENSITY

Annualized Individual Travel Behavior

Population Person Person Miles Person Miles Vehicle Vehicle Miles Vehicle Miles
Density Trips Traveled (PMT) per Trip Trips Traveled (VMT) per Trip
0 to 249 1,515 16,900 11 958 10,560 11
250 to 999 1,614 15,345 10 1,025 9,762 10
1,000 to 3,999 1,615 14,414 9 1,020 8,458 8
4,000 to 9,999 1,586 12,837 8 968 7,827 8
10,000 & up 1,476 9,029 6 668 4,880 7
Overall 1,568 14,064 9 951 8,523 9

Table 8 summarizes data relating population density, trips and miles traveled. The data reveal a
tendency toward fewer person trips in areas with the highest and lowest densities, with some
variation in between. The person miles traveled (PMT), however, declines as population density
increases, suggesting fewer miles associated with each trip at higher densities.

Vehicle trips decrease steadily as population density increases. The vehicle miles traveled
(VMT) associated with these trips also decreases. The exception occurs in areas where the
population density is 10,000 or higher in which the average number of miles per trip increases
slightly to 5.4.

Median Household Income/Poverty

TABLE 9: BLOCK GROUP MEDIAN HOUSEHOLD INCOME BY AREA TYPE

Household Income Area Type

Second City Rural Suburban Town Urban All
$0 to $24,999 32.6% 28.2% 0.9% 16.6% 28.2% 19.7%
$25,000 to $34,999 23.8% 43.2% 16.5% 24.7% 27.0% 26.5%
$35,000 to $44,999 16.9% 22.4% 21.8% 21.1% 19.7% 20.6%
$45,000 to $54,999 12.1% 5.6% 22.9% 18.2% 13.1% 15.0%
$55,000 & up 14.6% 0.6% 37.9% 19.4% 12.0% 18.3%
Total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0%

17
 FIGURE 3: MEDIAN HOUSEHOLD INCOME BY AREA TYPE

50%
Percent of Area Population

40%

30%

20%

10%

0%
$0 to $25,000 $35,000 $45,000 $55,000 &
$24,999 to to to up
$34,999 $44,999 $54,999
Block Group Median Household Income

Second City Rural Suburban Town Urban All

As summarized in Table 9, wealthy households dominate in the suburbs, middle income

households prevail in rural areas, and households in the lowest income category are most
common in second cities and in urban and rural areas. Suburban areas have the highest
percentage of households with a median income of $55,000 and higher; these households
comprise 37.9% of all households in suburban areas, twice the overall percentage for this income
category. Rural areas have the lowest percentage share of households in the two highest income
categories; households with incomes of $45,000 and higher comprise only 6.1% of all
households. The middle income categories prevail in rural areas where households with median
incomes of $25,000 to $44,999 comprise 65.6% of all households.

In second cities, the percentage of low-income residents (32.6%) is greater than the percentage of
low-income residents in both rural areas (28.2%) and urban areas (28.2%). This indicates a
growing trend for the poor who have traditionally resided in inner cities to follow the waves of
people leaving central cities for outlying areas. This movement of low-income groups will create
significant challenges for meeting transportation needs: second cities must plan for an influx of
low-income residents who cannot afford private vehicles and must depend on public
transportation for mobility. Because they have been recently developed, second cities do not
have the public transportation infrastructure which urban areas have developed over decades.
Transit accessibility will become increasingly important.

18
 TABLE 10: TRANSIT AVAILABILITY BY BLOCK GROUP MEDIAN HOUSEHOLD INCOME

Transit Availability
Household Bus Service No Bus Total
Income Available
$0 to $24,999 59.4% 40.6% 100.0%
$25,000 to $34,999 56.3% 43.7% 100.0%
$35,000 to $44,999 64.1% 35.9% 100.0%
$45,000 to $54,999 70.6% 29.4% 100.0%
$55,000 & up 72.8% 27.2% 100.0%
Total 63.4% 36.6% 100.0%

As shown in Table 10, transit availability is positively related to median household income: as
household income increases, transit availability also increases. This finding merits attention
because transit usage is typically associated with the lowest income categories; however, these
data indicate that only 60% of households with incomes of $0 to $24,999 have access to bus
service. Because low income households are more commonly dependent on transit for mobility,
the lack of available public transportation has social and economic implications.

TABLE 11: DISTANCE TO TRANSIT FROM HOUSEHOLD BY POVERTY STATUS

Percent of Block Group Living in Poverty

Distance to Transit Less than 4% 4% to 6% 7 to 12% 13% & up All
Less than .1 mile 28.6% 31.6% 36.5% 47.6% 36.0%
.1 to .24 mile 12.6% 13.5% 14.4% 17.0% 14.3%
.25 to .49 mile 10.2% 11.4% 11.0% 10.9% 10.8%
.5 to .99 mile 32.0% 27.2% 24.9% 15.7% 25.1%
1 Mile & up 16.7% 16.3% 13.2% 8.9% 13.8%
Total 100.0% 100.0% 100.0% 100.0% 100.0%

Table 11 reveals a tendency for transit accessibility to be greatest for those areas in which the
percent of the population living in poverty is the greatest. In those areas where more than 13% of
block groups live in poverty, 47.6% live less than .1 mile from transit. As distance from transit
increases the block groups with more than 13% of its residents living in poverty decreases.
However, results also indicate that there are areas having significant numbers living below
poverty that are located from .5 to over a mile from transit. For example, 24.9% of block groups
that have 7 to 12% living in poverty are located from .5 to .99 of a mile from transit.

19
 TABLE 12: ANNUALIZED INDIVIDUAL TRAVEL BEHAVIOR BY HOUSEHOLD INCOME

Annualized Individual Travel Behavior

Block Group Median Person Person Miles Person Miles Vehicle Vehicle Miles Vehicle Miles
Household Income Trips Traveled (PMT) per Trip Trips Traveled (VMT) per Trip
$0 to $24,999 1,482 12,173 8 821 7,026 9
$25,000 to $34,999 1,585 13,594 9 968 8,526 9
$35,000 to $44,999 1,567 14,761 9 984 9,161 9
$45,000 to $54,999 1,592 15,040 9 988 8,896 9
$55,000 & up 1,619 15,199 9 1,001 9,109 9
Overall 1,568 14,064 9 951 8,523 9

As shown in Table 12, both trips and miles of travel are positively associated with income.
Person trips and PMT generally increase as household income increases. The average number of
miles associated with each trip also increases. Vehicle trips generally increase as household
income increases. The VMT associated with these trips also increases.

Race and Hispanic Origin

TABLE 13: RACE BY AREA TYPE

Area Type
Race Second City Rural Suburban Town Urban All
White 73.5% 88.9% 80.4% 86.5% 53.7% 78.2%
African- 16.2% 6.5% 9.9% 7.1% 28.3% 12.5%
American
Asian 1.9% 0.4% 3.6% 1.1% 4.5% 2.2%
Other 8.4% 4.2% 6.1% 5.3% 13.5% 7.1%
Total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0%

Racial mix varies significantly in different area types. Whites form the majority in all types of
areas, but they dominate most in rural areas (88.9%), where all other racial groups combined
account for only 11.1% of the population. African-Americans have the most significant presence
in urban areas, with one African-American person for every two white persons.
African-Americans also have a significant, albeit greatly diminished, presence in second cities.
Although second cities have certain population characteristics similar to urban areas, second
cities have far less diversity in terms of racial mix when compared to urban areas.

20
 TABLE 14: TRANSIT AVAILABILITY BY RACE OR HISPANIC ORIGIN

Transit Availability
Race/Hispanic Origin Bus Service No Bus Total
Available
White 59.3% 40.7% 100.0%
African-American 80.0% 20.0% 100.0%
Asian 86.5% 13.5% 100.0%
Other 75.8% 24.2% 100.0%
All 63.3% 36.7% 100.0%

Hispanic 76.8% 23.2% 100.0%

Non-Hispanic 62.2% 37.8% 100.0%
All 63.4% 36.6% 100.0%

Table 14 shows that both African-Americans and Asians have higher than average transit
availability, while the availability of transit for whites is below average. Transit is also available
to a greater than average percentage of Hispanics.

TABLE 15: MODE OF TRANSPORTATION BY RACE OR HISPANIC ORIGIN

Race of Household Reference Person Reference Hispanic Status

Mode White African- Asian Other All Hispanic Non- All
American Hispanic
Private Vehicle 91.3% 79.0% 86.1% 84.8% 89.4% 84.8% 89.8% 89.3%
Public Transit 3.0% 10.1% 4.5% 5.2% 4.0% 5.6% 3.9% 4.0%
Taxi 0.1% 0.5% 0.1% 0.3% 0.2% 0.2% 0.2% 0.2%
Bicycle/Walk 5.5% 10.4% 9.2% 9.7% 6.4% 9.4% 6.1% 6.5%
Total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0%

Table 15 shows that the private vehicle is the dominant mode of transportation for all groups.
Whites rely on private vehicles more than any other group and less on public transit than these
groups. African-Americans depend on private vehicles less than all other groups and more on
public transit and bicycling and walking. Hispanics use private vehicles less than Non-Hispanics
and less than the average. As with African-Americans, they are more likely to use public transit
and bicycling and walking.

21
Age

TABLE 16: AGE BY AREA TYPE

Area Type
Age Group Second City Rural Suburban Town Urban All
5 to 15 16.0% 19.2% 17.9% 19.3% 16.0% 17.8%
16 to 19 5.6% 6.4% 5.8% 6.0% 5.3% 5.8%
20 to 29 18.0% 11.4% 14.5% 12.9% 17.5% 14.6%
30 to 39 17.4% 18.1% 19.7% 19.4% 21.0% 19.1%
40 to 49 14.8% 15.5% 16.6% 16.3% 14.1% 15.6%
50 to 59 9.0% 11.1% 10.2% 9.9% 9.0% 9.9%
60 to 69 8.9% 9.0% 8.0% 8.2% 8.3% 8.0%
70 & up 10.4% 9.3% 7.2% 8.1% 8.7% 8.6%
Total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0%

TABLE 17: FAMILY LIFE CYCLE BY AREA TYPE

Area Type
Family Life Cycle Second Rural Suburban Town Urban All
City
Single Adult, No Children 20.1% 12.5% 16.2% 13.6% 24.6% 17.0%
Two or More Adults, No Children 23.4% 23.4% 24.6% 23.2% 23.3% 23.6%
Single Adult, Youngest Child 0-5 1.8% 1.0% 1.0% 1.8% 2.8% 1.6%
Two or More Adults, Youngest Child 0-5 13.1% 14.0% 16.3% 16.5% 13.4% 14.8%
Single Adult, Youngest Child 6-15 2.8% 2.5% 2.5% 2.1% 3.2% 2.6%
Two or More Adults, Youngest Child 6-15 12.1% 17.6% 15.5% 17.0% 10.6% 14.8%
Single Adult, Youngest Child 16-21 1.2% 0.8% 1.1% 1.2% 1.1% 1.1%
Two or More Adults, Youngest Child 16-21 3.4% 5.3% 5.0% 4.9% 2.9% 4.4%
Single Adult Retired 9.9% 8.9% 6.1% 7.1% 8.2% 7.9%
Two or More Adults Retired 12.2% 13.8% 11.7% 12.6% 9.9% 12.1%
Total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0%

The 1995 NPTS contains information on age and life cycle patterns, see Tables 16 and 17. As
shown in Table 17, cities attract young adults, both single and married, who have no children.
These findings indicate a preference by these groups to locate in more densely populated urban
settings. Rural areas and towns, in contrast, have lower than average percentages of single,
childless adults. Towns and rural and suburban areas are more likely than average to be
populated by households with two or more adults and school-age children, indicating a possible
educational component in choice of residential location. Rural areas attract a lower percentage of
young adults (20-29 years old) than urban areas and second cities.

22
 TABLE 18: ONE-WAY WORK TRIP BY AGE AND GENDER

Distance to Work (Miles) Time to Work (Minutes)

Male Female Male Female
Age Group Mea Media Mea Media Mea Media Mea Media
n n n n n n n n
16 to 19 7 5 7 4 14 10 15 10
20 to 29 13 8 12 8 22 15 21 15
30 to 39 15 10 12 8 25 20 22 18
40 to 49 16 10 11 8 25 20 21 15
50 to 59 15 10 10 7 25 20 20 15
60 to 69 14 8 7 5 25 15 16 15
70 & up 8 5 7 4 20 15 16 13

Table 18 shows a tendency for younger workers to have shorter work trip distances and trip
times. This is true regardless of gender. The trip distance of males increases until the age of 49
when it begins to decrease. However, work trip times for males reaches a peak at age 30 and
levels off through age 69, indicating a stable trip time independent of distance. Females have
shorter work trip distances and travel times across all age groups. Mean trip distance peaks
between the ages of 20 and 40 and begins to decline thereafter. However, work trip times exhibit
a slight peak in the 30 to 39 age group category. Not surprisingly, work trip distance and trip
times decline significantly for workers in the 70 and up age group category.

TABLE 19: TRANSIT AVAILABILITY BY FAMILY LIFE CYCLE

Transit Availability
Family Life Cycle Bus Service No Bus Total
Available
Single Adult, No Children 71.7% 28.3% 100.0%
Two or More Adults, No Children 62.5% 37.5% 100.0%
Single Adult, Youngest Child 0-5 67.7% 32.3% 100.0%
Two or More Adults, Youngest Child 0-5 64.9% 35.1% 100.0%
Single Adult, Youngest Child 6-15 66.0% 34.0% 100.0%
Two or More Adults, Youngest Child 6-15 58.4% 41.6% 100.0%
Single Adult, Youngest Child 16-21 69.4% 30.6% 100.0%
Two or More Adults, Youngest Child 16-21 56.8% 43.2% 100.0%
Single Adult Retired 64.2% 35.8% 100.0%
Two or More Adults Retired 57.5% 42.5% 100.0%
All 63.3% 36.7% 100.0%

Table 19 shows the relationship between family life cycle and availability of transit at the
residence. For all life cycle categories, transit service is available to over 55% of households,

23
compared to the overall average of 63.3%. The data indicate that households with a single adult
are more likely to live where transit is available. This tendency is greatest for single adults with
no children (71.7%) and holds true for all stages of life. In contrast, households with two or
more adults are less likely to live where transit is available, indicating less need or preference to
use transit. Households with two or more adults and young children are an exception to this
general tendency, with 64.9% reporting transit availability at the residence. According to these
data, transit availability at the residence is closely associated with family life cycle.

TABLE 20: DISTANCE TO TRANSIT BY FAMILY LIFE CYCLE

Distance to Transit from Household (Miles)

Family Life Cycle Less .1 to .25 to .5 to 1 Mile All
than .1 .24 .49 .99 & up
Single Adult, No Children 24.4% 19.9% 18.6% 16.1% 12.2% 19.4%
Two or More Adults, No Children 21.6% 22.6% 23.3% 25.5% 23.7% 23.2%
Single Adult, Youngest Child 0-5 2.6% 1.4% 1.3% 1.1% 1.3% 1.8%
Two or More Adults, Youngest Child 0-5 13.3% 13.5% 12.8% 16.7% 20.1% 15.1%
Single Adult, Youngest Child 6-15 3.4% 2.9% 2.4% 2.6% 1.6% 2.8%
Two or More Adults, Youngest Child 6-15 11.5% 12.7% 11.7% 15.4% 19.7% 13.8%
Single Adult, Youngest Child 16-21 1.3% 1.0% 1.1% 1.2% 1.2% 1.2%
Two or More Adults, Youngest Child 16-21 3.2% 3.7% 3.6% 4.7% 4.9% 3.9%
Single Adult Retired 9.3% 9.8% 10.3% 5.6% 4.6% 7.9%
Two or More Adults Retired 9.5% 12.6% 14.8% 11.1% 10.6% 11.1%
Total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0%

As shown in Table 20, transit is most closely located near households with no children and a
working-age adult (24.4%). Households with two or more adults and no children are evenly
distributed across all transit access categories, with the highest percentage occurring between 5
and .99 miles (25.5%). Families with two or more adults and children under 16 are more likely
to live one-half mile from transit or more, indicating less dependence on transit than other family
types. Single retirees are more likely than average to live within .5 mile from transit, while
families with two or more retired adults are more likely to live from .1 to .99 miles from transit,
also indicating less dependence on transit than their single counterparts.

24
Education
TABLE 21: TRANSIT AVAILABILITY BY EDUCATION

Transit Availability
Education of Household Bus Service No Bus Total
Reference Person Available
Less Than HS Graduate 54.0% 46.0% 100.0%
High School Graduate 58.4% 41.6% 100.0%
Some College, No Degree 66.1% 33.9% 100.0%
Associate Degree 62.7% 37.3% 100.0%
Bachelors Degree 70.0% 30.0% 100.0%
Some Grad/Prof School 68.5% 31.5% 100.0%
Grad/Prof School Degree 71.2% 28.8% 100.0%
All 63.1% 36.9% 100.0%

As shown in Table 21, transit availability generally increases as education increases, indicating a
positive relationship between the two. On average, transit is available to 63.1% of households.
This compares to 54.0% for households in which the reference person has less than a high school
education and 58.4% for households in which the reference person has graduated from high
school. The percentage for households in which the reference person has attended college
exceeds the average, with the exception of the associate degree category. The positive
relationship between transit availability and education does not hold true for the category of
persons who have some graduate or professional school. The percentage for this category is
68.5%, a decrease of 1.5% compared to households in which the reference person has a bachelors
degree (70.0%).

Measures for Places

Area Type
The 1995 NPTS bases urban/rural coding on population densities at a location and in relation to
neighboring locations (see Key Terms and Definitions for an explanation) .

TABLE 22: DRIVERS PER ADULT BY AREA TYPE

Area Type
Drivers per Adult Second City Rural Suburban Town Urban Overall
Less than One 16.7% 11.9% 10.7% 11.5% 32.2% 15.8%
One Driver 80.0% 82.0% 84.6% 83.2% 66.2% 79.9%
More than One 3.3% 6.1% 4.7% 5.3% 1.6% 4.3%
Total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0%

25
The majority of all adults in America drive. This ratio of drivers to adults varies, however, with
area type. In towns, second cities, suburban, and rural areas, over 80% of the population has a
ratio of one driver for every adult.

Rural areas have the largest percentage of ratios above one driver per adult. The NPTS defines
adults as persons eighteen years of age or older; whereas, many states allow people to earn
driver’s licenses at sixteen years of age. Sixteen and seventeen year olds account for ratios above
one driver per adult. Rural areas, therefore, have the largest percentage of their young people
driving. Rural residents require private transportation for much of daily living, and young people
need to attain driving privileges for mobility.

Urban areas have the lowest ratio of drivers per adult. The high percentage (32.2%) of urban
populations having less than one driver per adult indicates less dependence on private vehicles.
Urban areas offer more options for public transit, and many destinations can be accessed by
walking or bicycling.

TABLE 23: VEHICLES PER ADULT BY AREA TYPE

Area Type
Vehicles per Adult Second City Rural Suburban Town Urban Total
Less than One 27.1% 18.4% 20.1% 18.3% 47.0% 25.1%
One Vehicle 61.6% 56.1% 65.6% 62.4% 46.1% 59.1%
More than One 11.3% 25.4% 14.3% 19.4% 6.9% 15.8%
Total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0%

Vehicle ownership per adults follows patterns similar to the patterns of drivers per adult. In rural
areas, over one quarter of the adults have more than one vehicle, but in urban areas, nearly half of
the residents have less than one vehicle for each adult. Over one quarter of the residents of
second cities also have less than one vehicle for each adult. In second cities, towns, and
suburban areas, over 60% of the population have exactly one vehicle per adult.

The land use of an area can affect the number of vehicles per adult: close access to destinations
and plentiful transportation facilities may induce less vehicle ownership in urban areas. The
number of vehicles in an area can also affect land use. High levels of vehicle ownership require
parking structures, lots, and facilities to accommodate the vehicles. Rural areas have the space
necessary to support high vehicle ownership; whereas, land values in urban areas make vehicle
ownership expensive.

26
 TABLE 24: WORK LOCATION BY AREA TYPE

Area Type
Work Location Second City Rural Suburban Town Urban All
Work from Home 5.2% 8.2% 5.3% 5.8% 5.0% 5.9%
No Fixed Work Place 2.0% 2.6% 2.0% 1.9% 2.6% 2.2%
Work at Work Location 92.8% 89.2% 92.7% 92.3% 92.4% 91.9%
Total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0%

As shown in Table 24, there is little variation in work location across area type. The percentage
of people who work at the work location vastly exceeds the percentages of people who work at
home and those who have no fixed work place. The distribution is very similar within the second
city, suburban, town and urban categories. The percentage distribution within these categories
hovers near the overall percentages for each category. Rural areas vary from this pattern, with
more people working from home than average and fewer people working at the work location
than average.

TABLE 25: ONE-WAY WORK TRIP BY AREA TYPE AND GENDER

Distance to Work (Miles) Time to Work (Minutes)

Male Female Male Female
Mea Media Mea Media Mea Media Mea Media
n n n n n n n n
Second City 12 6 9 5 21 15 18 15
Rural 18 12 13 10 24 20 20 15
Suburban 14 10 11 8 24 20 21 17
Town 16 10 12 7 24 20 19 15
Urban 11 7 9 6 26 20 25 20

The impact of area type on the distance to work by gender is reported in Table 25. Across all
area types males generally travel greater distances to work and the mean travel time to work for
males is also greater. The travel time to work for males and females is roughly equivalent in
urban areas and their distance to work in urban areas is different by only two seconds. The
distance to work for rural males is 18 miles and for females it is 13 miles. This is the greatest
difference in distance to work across all area types.

27
 TABLE 26: TRANSIT AVAILABILITY BY AREA TYPE

Transit Availability
Area Type Bus Service No Bus Total
Available
Second City 81.9% 18.1% 100.0%
Rural 14.3% 85.7% 100.0%
Suburban 87.4% 12.6% 100.0%
Town 37.6% 62.4% 100.0%
Urban 98.3% 1.7% 100.0%
All 63.4% 36.6% 100.0%

Table 26 summarizes the relationship between area type and transit availability. Urban areas
have the highest percentage of available bus service (98.3%), exceeding the overall average of
63.4% by nearly one-third. Suburban areas (87.4%) and second city areas (81.9%) also exceed
the average by a large margin, indicating a tendency for these areas to have transit service
available. Rural areas and towns both fall well-below the average with 14.3% and 37.6% service
respectively.

TABLE 27: DISTANCE TO TRANSIT FROM THE HOUSEHOLD BY AREA TYPE

Area Type
Distance to Transit Second City Rural Suburban Town Urban All
from Household
Less than .1 mile 37.9% 21.4% 28.2% 22.1% 52.5% 36.0%
.1 to .24 mile 16.0% 1.6% 13.4% 6.3% 19.6% 14.3%
.25 to .49 mile 12.0% 4.9% 11.6% 5.7% 12.0% 10.8%
.5 to .99 mile 24.3% 18.3% 34.4% 27.5% 14.3% 25.1%
1 mile & up 9.7% 53.8% 12.3% 38.4% 1.6% 13.8%
Total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0%

28
 Percent of Households within FIGURE 4: DISTANCE TO TRANSIT BY AREA TYPE

60%

50%

40%
Area Type

30%

20%

10%

0%
Less than .1 .1 to .24 mile .25 to .49 .5 to .99 mile 1 mile & up
mile mile
Distance to the Bus

Second City Rural Suburban

Town Urban All

Approximately 52.5% of persons living in urban areas are less than .1 mile form transit while this
is true for only 21.45 of those living in rural areas. When the distance to transit increases to
between .1 to .24 mile, 19.6% of urban area residents enjoy this high level of accessibility, but
this is true for only 1.6% of those living in rural areas. Residents living in urban areas and in
second cities enjoy greater accessibility to transit. Approximately 53.8% of residents of rural
areas live at least one mile or further from transit as do 38.4% of persons living in towns. Only
1.6% of urban dwellers are one mile or more from transit. It is clear that residents of rural areas
and towns are transit constrained.

TABLE 28: AUTOMOBILE COMMUTING BY AREA TYPE

Area Type
Commute Auto Usage Second City Rural Suburban Town Urban All
Go to Work by Auto 80.1% 67.3% 83.0% 75.5% 69.4% 75.9%
Do Not Go to Work by Auto 19.9% 32.7% 17.0% 24.5% 30.6% 24.1%
Total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0%

Table 28 shows that 75.9% of the population generally travels to work by auto. Second cities
and suburban areas exceed this average with 80.1% and 83.0% of workers, respectively, using
autos to travel to work. Rural and urban areas both fall below this average with 67.3% and

29
 69.4% respectively. Work-related auto in towns falls at the average. These data indicate a
greater dependence in second cities and suburban areas on auto use.

TABLE 29: ANNUALIZED INDIVIDUAL TRAVEL BEHAVIOR BY AREA TYPE

Annualized Individual Travel Behavior

Area Type Person Person Miles Person Miles Vehicle Vehicle Miles Vehicle Miles
Trips Traveled (PMT) per Trip Trips Traveled (VMT) per Trip
Second City 1,609 13,445 8 988 7,982 8
Rural 1,549 16,833 11 961 10,432 11
Suburban 1,595 13,790 9 1,009 8,431 8
Town 1,579 15,350 10 1,002 9,563 10
Urban 1,488 9,820 7 731 5,359 7
Overall 1,568 14,064 9 951 8,523 9

With 1,549 person trips, rural residents have the second lowest number of overall trips, and they
make the third highest number of vehicle trips at 961. Rural residents are tied much more to
their personal vehicles than residents of other areas. They also cover the most distance at 16,833
person miles annually. Townsfolk cover the next highest distance at 15,350 person miles and
9,563 vehicle miles. Urban residents make the fewest number of trips and cover the shortest
distance by far with 731 vehicle trips. Part of the reason why the number of trips remains so low
for urban residents may have to do with issues of data collection: trips of less than one block or
equal to one half mile may be undercounted.

Residential Density
Area types provide a broad look at the geographic landscape. Residential density allows a closer
look at land uses where people live and also a link to measures of people. Residential and
population densities reflect similar parameters: both indicate the extent of concentration where
people live. Population density measures the number of people per square mile; residential
density measures the number of living units per square mile. A proportional increase in
residential density may correspond with a proportional increase in population density. The two
measures diverge in instances where more or fewer people live in a household, compared to the
average. Variables such as race or age may impact residential density. Some cultures, for
instance, typically live in large households with extended families, while other cultures value
independence from family. Similarly, large numbers of single-person households may appear
where high concentrations of young adults live.

30
 TABLE 30: BLOCK GROUP RESIDENTIAL DENSITY BY AREA TYPE

Area Type
Block Group Housing Second City Rural Suburban Town Urban All
Units per Mile2
0 to 99 2.7% 81.3% 1.4% 26.8% 0.4% 23.0%
100 to 499 12.1% 12.9% 13.6% 39.5% 0.4% 16.9%
500 to 1,499 32.7% 4.3% 36.2% 22.6% 6.1% 21.6%
1,500 to 2,999 34.3% 1.3% 34.4% 9.8% 23.8% 20.7%
3,000 & up 18.1% 0.2% 14.4% 1.4% 69.2% 17.9%
Total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0%

Table 30 compares block group residential density to area type and reveals that urban and rural
areas are at opposite ends of the spectrum with regard to residential density. Urban areas have
the greatest percentage of dense residential areas and the lowest percentage of areas with sparse
dwellings. Over 80% of rural areas have residential densities under 100 housing units per square
mile. Second cities and suburban areas are comparable to each other with over 65% of block
groups having between 500 and 3,000 housing units per square mile.

TABLE 31: MILES DRIVEN LAST YEAR BY RESIDENTIAL DENSITY AND GENDER

Annual Miles Driven

Block Group Housing Male Female
Units per Mile2 Mean Median Mean Median
0 to 99 17,956 14,000 10,637 9,500
100 to 499 17,523 14,000 10,088 9,000
500 to 1,499 15,382 12,000 8,987 8,000
1,500 to 2,999 14,351 12,000 8,485 7,000
3,000 & up 12,360 10,000 7,387 5,000

The number of miles an individual drives annually consistently decreases as residential density
increases. This trend appears in both mean and median measures of central tendency and holds
true across gender. The genders diverge, however, in actual numbers of miles driven. In block
groups with over 3,000 housing units per square mile, females drive 7,387 miles annually, which
is 4,973 miles fewer per year than males do on average, a 60% difference. In areas with
residential densities lower than 100 housing units per square mile, the large difference between
the genders increases up to 7,134 miles annually on average, which is also a 60% difference.

31
 TABLE 32: ONE-WAY WORK TRIP BY RESIDENTIAL DENSITY AND GENDER

Distance to Work (Miles) Time to Work (Minutes)

Block Group Housing Male Female Male Female
Units per Mile2 Mea Media Mea Media Mea Media Mea Media
n n n n n n n n
0 to 99 17 12 13 10 24 20 20 15
100 to 499 17 10 12 8 25 20 20 15
500 to 1,499 14 9 11 7 22 18 20 15
1,500 to 2,999 13 8 10 6 23 18 20 15
3,000 & up 11 7 9 6 24 20 24 20

For those block groups with 0-99 units per mile, men drive 17 miles while females drive an
average of 13 miles one-way for the work trip (Table 32). At the very highest density 3,000 and
up males drive 11 miles while females drive 9 miles. The statistics for the time to work
corresponds to the pattern observed for distance traveled with males generally traveling greater
distances and having correspondingly longer travel times. Males drive for approximately 24
minutes and females for 20 minutes at the lowest density block group and 24 minutes and 24
minutes respectively in the densest block group levels of 3,000 or more housing units. The
distance to work decreases for both males and females as housing unit density increases. This is
not true for travel time where males living in block groups with 0 to 99 units travel 24 minutes
and males living in block groups with more than 3,000 units per mile also travel 24 minutes on
average. For females, travel time to work is 20 minutes in low residential density areas and
reverses the trend and increases to 24 minutes as density increases. As housing density increases
distance to work decreases for males and females. However, as density increases we do not see a
decrease in travel time for males or females. Travel time to work for females is constant at 20
minutes except for an increase in travel time for women living in the most densely populated
block groups. There are many possible explanations including congestion associated with
densely populated areas as well as the mode of travel or the time of day when the trip occurs.

TABLE 33: TRANSIT AVAILABILITY BY RESIDENTIAL DENSITY

Block Group Transit Availability

Residential Density
(Housing Units/ Mile2) Bus Service No Bus Total
Available
0 to 99 20.3% 79.7% 100.0%
100 to 499 44.3% 55.7% 100.0%
500 to 1,499 70.1% 29.9% 100.0%
1,500 to 2,999 85.8% 14.2% 100.0%
3,000 & up 96.4% 3.6% 100.0%
All 63.4% 36.6% 100.0%

32
As housing density increases the availability of bus increases from 20.3% for block groups with 0
to 99 units to 96.4% for block groups with 3,000 or more housing units. The general availability
of bus is approximately 63.4% while 36.6% of residents do not have bus service available across
all housing density levels (Table 33).

TABLE 34: DISTANCE TO TRANSIT FROM THE HOUSEHOLD BY RESIDENTIAL DENSITY

Block Group Residential Density (Housing Units/ Mile2)

Distance to Transit 0 to 99 100 to 500 to 1,500 to 3,000 All
from Household 499 1,499 2,999 & up
Less than .1 mile 17.8% 19.4% 25.7% 35.1% 54.3% 36.0%
.1 to .24 mile 2.6% 6.0% 13.0% 18.3% 17.2% 14.3%
.25 to .49 mile 3.3% 7.2% 10.5% 13.5% 11.5% 10.8%
.5 to .99 mile 19.1% 31.2% 36.0% 26.9% 14.6% 25.1%
1 Mile & up 57.2% 36.2% 14.8% 6.2% 2.4% 13.8%
Total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0%

For those persons having transit available the distance to transit decreases for a larger number of
households in the densest block groups (Table 34). At the 0 to 99 level approximately 17.8% of
households are less than .1 mile. When residential density increases up to 3,000 units 54.3% of
households are less than .1 mile an increase of more than 300%. At the lowest residential density
there is a drop at the .1 mile to .49 mile range with a total of 5.9% of households located between
those distances. These numbers change to 19.1 % for households that are located beyond .5 mile
of transit. At the 3,000 and up density level only 2.4% of households are located at a distance of
one mile or greater from transit. While approximately 57.2% of households are located more
than a mile from transit in the lowest density level. The lack of accessible transit service (within
one-quarter mile) in low density residential areas means that persons living in rural areas that are
transit dependent have limited or no transit alternative. The distance from the transit station or
bus stop is critically important to the decision whether or not to use transit at all.

TABLE 35: MODE OF TRANSPORTATION BY RESIDENTIAL DENSITY

Block Group Residential Density (Housing Units/ Mile2)

Mode of 0 to 99 100 to 500 to 1,500 to 3,000 All
Transportation 499 1,499 2,999 & up
Private Vehicle 93.2% 92.8% 92.2% 90.3% 76.0% 89.3%
Public Transit 3.5% 3.1% 3.0% 2.8% 8.4% 4.0%
Taxi 0.0% 0.1% 0.1% 0.1% 0.7% 0.2%
Bicycle/Walk 3.3% 4.1% 4.7% 6.8% 14.9% 6.5%
Total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0%

33
Table 35 reports the results of the primary mode used by respondents. They were asked to
identify the mode used for the longest portion of the trip taken. The pre-dominance of the private
vehicle is evident across all residential densities. The mode of transportation used by most
households in the lowest density areas is the private vehicle used by 93.2% of households. Where
residential density is the greatest approximately 76.0% of households use the private vehicle.
The densest residential areas display significantly lower dependence on the private vehicle. The
availability of transit and other modes explains some of this as well as the existence of large
numbers of urban poor that do not own automobiles. Public transit is used by 8.4% of
households in the densest residential areas and only 3.5% use it in the lowest density areas. The
largest number of bicycle and walk trips are made by households in the densest residential areas
and this is probably influenced by the proximity of trip destinations. Overall 89.3 use the private
vehicle and 4% use transit.

TABLE 36: ANNUALIZED INDIVIDUAL TRAVEL BEHAVIOR BY RESIDENTIAL DENSITY

Block Group Annualized Individual Travel Behavior

Residential Density
(Housing Units/ Mile2) Person Person Miles Person Vehicle Vehicle Miles Vehicle
Trips Traveled Miles per Trips Traveled Miles per
(PMT) Trip (VMT) Trip
0 to 99 1,521 16,973 11 959 10,562 11
100 to 499 1,604 15,092 9 1,011 9,590 9
500 to 1,499 1,601 14,366 9 1,010 8,283 8
1,500 to 2,999 1,588 12,923 8 989 8,020 8
3,000 & up 1,532 10,304 7 771 5,764 7
Overall 1,568 14,064 9 951 8,523 9

FIGURE 5: ANNUALIZED INDIVIDUAL TRAVEL BEHAVIOR BY RESIDENTIAL DENSITY

1,800 18,000

1,600 16,000

1,400 14,000
Annualized Trips

12,000
Annual Miles

1,200

1,000 10,000

800 8,000

600 6,000

400 4,000 Person Miles Traveled (PMT)

200 Person Trips 2,000

Vehicle Miles Traveled (VMT)
Vehicle Trips
0 0
0 to 99 100 to 499 500 to 1,499 1,500 to 3,000 & up Overall 0 to 99 100 to 499 500 to 1,500 to 3,000 & up Overall
2,999 1,499 2,999
Residential Density (Housing Units/Square Mile) Residential Density (Housing Units/Square Mile)

34
The impact of residential density on Vehicle Miles Traveled (VMT), Vehicle Trips, and Person
Miles Traveled (PMT) is illustrated in Table 36. As residential density increases there is a
corresponding decrease in person miles traveled, vehicle trips, and vehicle miles traveled. From
the lowest density residential areas to the densest, the number of person miles traveled decreased
by approximately 60.7%, vehicle trips decreased by 58.35% and vehicle miles traveled decreased
by 54.31%. So, as residential density increased travel in all three categories experienced a
sizable decrease. However, this decrease in travel as density increased was not true for person
trips which increased although by only .007%. So residents made slightly more person trips in
the densest residential areas but traveled fewer personal miles, made fewer vehicle trips, and
reduced the total number of vehicle miles traveled. Increased residential density results in the
sizable reduction in specific categories of travel.

TABLE 37: WORK LOCATION BY RESIDENTIAL DENSITY

Block Group Residential Density (Housing Units/ Mile2)

Place of Work 0 to 99 100 to 499 500 to 1,499 1,500 to 2,999 3,000 & up All
Work from Home 7.9% 6.0% 5.1% 5.4% 4.8% 5.9%
No Fixed Work Place 2.5% 1.7% 1.7% 2.5% 2.4% 2.2%
Work at Work Location 89.6% 92.3% 93.2% 92.0% 92.8% 91.9%
Total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0%

Table 37 reports the impact of residential density on workplace choice. As residential density
increases the percentage of persons working from the home decreases from 7.9% to 4.8% while
the number of persons working at a work location increases from 89.6% to 92.8%. This increase
in employment at a work location may be attributed to a variety of sources for example, greater
and more diverse employment opportunities, the availability of transit, wages, and more walk and
bicycle trips.

TABLE 38: EMPLOYMENT DENSITY BY RESIDENTIAL DENSITY

Work Tract Block Group Residential Density (Housing Units/ Mile2)

Employment Density
(Employees per Mile2) 0 to 99 100 to 499 500 to 1,499 1,500 to 2,999 3,000 & up Overal
l
0 to 174 96.0% 42.2% 6.4% 1.4% 0.5% 36.6%
175 to 799 1.9% 44.8% 40.0% 23.2% 7.9% 21.0%
800 to 1,999 0.6% 7.8% 32.0% 38.1% 24.5% 18.9%
2,000 to 6,499 0.7% 3.6% 17.4% 30.3% 42.3% 16.8%
6,500 & up 0.7% 1.6% 4.2% 6.9% 24.7% 6.7%
Total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0%

35
For those people who work at a fixed work location, patterns of employment density follow
patterns of residential density. The data in table 38, which come from census data provided by
Claritas (as opposed to NPTS data provided by the Federal Highway Administration), show
trends where increasing employment density corresponds with increasing residential density. By
far, areas with fewer than 100 housing units per square mile have the highest percentage of
people working in areas with fewer than 175 jobs per square mile (96.0%). Similarly, areas with
over 3,000 housing units per square mile have the highest percentage of people who work in
census tracts with over 6,500 employees per square mile (24.7%). Overall, however, the largest
percentage of all people work in tracts with fewer than 175 jobs per square mile (36.6%), which
could represent a turnaround trend from the days when cities as commercial centers were seen as
primary employment centers.

Age of Housing
The age of housing provides another important indicator for residential area land use. New
housing in an area implies population growth in that area, and transportation infrastructure must
meet the needs of the population where it exists.

TABLE 39: AGE OF HOUSING BY AREA TYPE

% of Block Group Area Type

Housing Units Built in
the Last Ten Years Second City Rural Suburban Town Urban All
0-20% 80.6% 87.1% 75.2% 70.7% 93.5% 80.6%
21-40% 11.0% 11.5% 13.0% 19.2% 4.8% 12.3%
41-60% 4.8% 1.3% 6.9% 6.2% 1.5% 4.4%
61-80% 2.4% 0.0% 3.3% 2.2% 0.2% 1.7%
81-100% 1.2% 0.1% 1.6% 1.6% 0.1% 1.0%
Total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0%

New development has occurred in the last ten years primarily outside of urban and rural areas.
Joel Garreau classifies edge cities as areas that were "nothing like ’city’ as recently as thirty years
ago6." These development configurations started appearing in America much later than
traditional urban areas. The similar concept of second city approximates edge city for the NPTS
data. Presumably, housing for second cities would be remarkably young; however, the NPTS
data do not indicate substantial youth for second cities compared to town and suburban areas.
Town and suburban areas have been developed more recently, with only 29.3% and 24.8% of
their block groups respectively containing over 20% housing built in the last ten years. This is
19.4% in second cities. Suburban areas, where 11.8% of the block groups constructed 40% of
their housing units in the last ten years, have the highest percentage of very young communities.
Existing housing structure in urban and rural areas has maintained a substantially dominant
presence with 93.5% and 87.1% of block groups containing less than 20% housing units built in
the last ten years (Table 39).

36
 TABLE 40: BUS AVAILABILITY FOR RECENT BUILDS

Bus % of Block Group Housing Units Built in Last 10 Years

Availability 0-20% 21-40% 41-60% 61-80% 81-100% All
Bus Service 64.4% 54.8% 65.9% 72.9% 67.1% 63.4%
Available
No Bus 35.6% 45.2% 34.1% 27.1% 32.9% 36.6%
Total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0%

Public transit infrastructure develops over a multi-year process. Established urban areas
implemented sophisticated public transit decades ago. Newer towns experiencing high growth
may find themselves facing heavy need for transit before infrastructure can be developed. Urban
and rural areas have the highest percentages of block groups with less than 20% housing built in
the last ten years. Urban areas have well-established transit systems with 98.3% of the
population served by transit; with just 14.3% of rural residents claiming transit availability, rural
areas have little transit infrastructure.

Transit infrastructure appears to meet new demand. All block groups with over 40% housing
units built in the last ten years surpass the overall average of 63.4% of the people served by
transit. Areas with between 61 and 80% new housing units achieve the highest level of transit
availability with 72.9% of the population of these areas served (Table 40).

Housing Tenure
Age of housing represents a physical depiction of an area’s growth; housing tenure indicates a
population characteristic integral to an area’s residential land use. Knowing whether residents
rent or own their homes may provide insight into the stability of the area or the likelihood of
residents to own a personal vehicle or utilize public transit.

TABLE 41: PERCENTAGE OF RENTER-OCCUPIED HOUSING BY AREA TYPE

Renter-Occupied Housing Area Type

in the Block Group Second City Rural Suburban Town Urban All
0-9% 9.7% 4.9% 25.4% 15.9% 4.9% 13.1%
10-19% 15.1% 45.1% 22.0% 29.4% 9.3% 24.7%
20-29% 15.3% 30.5% 15.8% 21.1% 9.7% 18.7%
30-49% 25.3% 17.4% 18.6% 23.0% 18.6% 20.6%
50-100% 34.6% 2.1% 18.2% 10.5% 57.5% 23.0%
Total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0%

37
As shown in Table 41, area type appears to influence the amount of renter-occupied housing.
Urban areas have high rental capacity. Second cities have more rental housing than less densely
populated areas but less rental housing than urban areas provide. The largest concentration of
rental units in rural areas, 45.1%, occurs in block groups containing between 10 and 19% of
rental units. Rural areas match urban areas with only 4.9% of block groups containing less than
10% rental housing. More than 75% of all rural block groups contain between 10 and 29% rental
units. Towns also show a peak at 10 to 19%, but towns have a more even spread that includes
33.5% of block groups with 30 to 100% renter-occupied housing.

TABLE 42: TRANSIT AVAILABILITY BY HOUSING TENURE

Transit Availability
Housing Tenure Bus Service No Bus Total
Available
Owned 57.9% 42.1% 100.0%
Rented 77.2% 22.8% 100.0%
Provided By Job or Military 50.7% 49.3% 100.0%
Other 64.7% 35.3% 100.0%
All 63.3% 36.7% 100.0%

As shown in Table 42, public transit serves 50% or more of all housing types. Transit is most
closely associated with rental communities, where 77.2% of renters have access to bus service.
Rental communities typically include more people without personally owned vehicles compared
to communities where home ownership is more common. Approximately 58% of communities
have available bus service in areas where home ownership prevails.

TABLE 43: DISTANCE TO TRANSIT BY HOUSING TENURE

Housing Tenure
Distance to Owned Rented Provided Other All
Transit from By Job or
Household Military
Less than .1 mile 29.0% 49.2% 30.3% 65.4% 36.0%
.1 to .24 mile 14.2% 14.5% 6.7% 12.0% 14.3%
.25 to .49 mile 11.6% 9.4% 1.8% 0.0% 10.8%
.5 to .99 mile 28.0% 19.6% 48.9% 17.2% 25.1%
1 Mile & up 17.2% 7.3% 12.4% 5.4% 13.7%
Total 100.0% 100.0% 100.0% 100.0% 100.0%

38
Renters live closer to transit than owners. Of the 77.2% of renters reporting bus service
available to them, 49.2% live within one-tenth of a mile of bus service. Only 29% of served
owners report that proximity to transit, but even 82.8% of owners with bus service available to
them live within one mile of transit. The typically dense nature of rental housing may explain
why renters receive better service from public transit: one bus route can easily provide
transportation for a large number of renters located in a small area (Table 43).

With only half of the community living in homes provided by employers or the military receiving
transit service, 48.9% of this segment of the community lives between one half and one mile
away from bus service. Another 30.3% of this community lives less than one-tenth of a mile
away from transit. These figures resemble transit availabilities for owner more than for renters.
Military personnel may live on bases which provide self-contained communities without need for
extensive transportation to non-military locations.

Measures for Employment

Employment Density

TABLE 44: WORK TRACT EMPLOYMENT DENSITY BY HOME BLOCK GROUP AREA TYPE

Work Tract Area Type

Employment Density
(Workers per Mile2) Second City Rural Suburban Town Urban All
0 to 174 15.4% 60.7% 6.0% 32.3% 1.7% 22.5%
175 to 799 20.8% 16.4% 17.9% 25.9% 6.5% 18.3%
800 to 1,999 23.8% 10.8% 23.4% 16.0% 13.9% 18.2%
2,000 to 6,499 23.1% 8.2% 29.5% 16.5% 32.3% 22.2%
6,500 & up 16.9% 3.9% 23.1% 9.3% 45.6% 18.9%
Total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0%

39
 FIGURE 6: WORK TRACT EMPLOYMENT DENSITY BY HOME BLOCK GROUP AREA TYPE

70%

60%

50%
Percent of Workers

40%

30%

20%

10%

0%
0 to 174 175 to 799 800 to 1,999 2,000 to 6,500 & up
6,499
Work Tract Employment Density (Workers/Square Mile)
Second City Rural Suburban
Town Urban All

The NPTS data indicate that employment density patterns and population density patterns typically
mirror one another. People living in rural areas tend to work in areas of low employment density.
People living in urban areas tend to work in areas of high employment density. A graph of work tract
employment densities shows striking difference in trend lines according to area type. The trend line
for urban areas sweeps up as work tract employment density increases, while the trend line for rural
areas turns downward. The trend line for town residents decreases with a downward slope not as
steep as the slope for the rural trend line.
.
Trend lines for residents of suburbia and second cities display employment location decisions
similar to each other. These trend lines include maximum values. Percentages of second city
residents working in tracts with low employment densities increase as employment densities
increase, up to a maximum value with 23.8% of second city residents working in census tracts
with employment densities from 800 to 1,999 jobs per square mile. A lower percentage of
second city residents work in tracts with higher employment densities. Employment density for
suburban residents reaches a maximum value with 29.5% of suburban residents working in tracts
with employment densities from 2,000 to 6,499 jobs per square mile; fewer suburban residents
work in tracts with over 6,500 jobs per square mile.

40
The graphed trend lines translate easily into indications of density preference. Urban residents
live in areas of high population density and, presumably, live near an area of high employment
density; therefore, 45.6% of urban residents work in tracts with over 6,500 jobs per square mile.
Rural residents live in sparsely developed areas where job densities remain low. Approximately
60.7% of rural residents work in tracts with less than 175 jobs per square mile (Table 44).

TABLE 45: MILES DRIVEN LAST YEAR BY EMPLOYMENT DENSITY AND GENDER

Work Tract Annual Miles Driven

Employment Density
(Employees per Mile2) Male Female
Mean Median Mean Median
0 to 174 19,367 15,000 11,277 10,000
175 to 799 18,399 15,000 11,144 10,000
800 to 1,999 17,466 15,000 10,956 10,000
2,000 to 6,499 16,537 15,000 10,895 10,000
6,500 & up 14,543 12,000 10,353 10,000

Females do not change their travel behavior as drastically as males do as employment density at
the work tract changes. Males working in low employment densities from 0 to 174 employees
per square mile annually drive an average of 19,367 miles while females drive 11,277 miles
(Table 45).

TABLE 46: ONE-WAY WORK TRIP BY EMPLOYMENT DENSITY AND GENDER

Work Tract Distance to Work (Miles) Time to Work (Minutes)

Employment Density Male Female Male Female
(Employees per Mile2) Mean Median Mean Median Mean Median Mean Median
0 to 174 12 7 9 6 18 15 15 11
175 to 799 12 8 9 6 20 15 17 15
800 to 1,999 13 8 10 7 22 15 19 15
2,000 to 6,499 13 9 10 7 23 20 21 15
6,500 & up 15 10 13 9 30 25 28 24

People live farther away from work when the workplace is located in a tract with high
employment density than when the workplace is located in a tract with low employment density.
The distance from home to the workplace does not vary by much; however, distance consistently
increases for both men and women as employment density increases. Men working in work
tracts with over 6,500 jobs per square mile commute 15 miles to work on average, which is 3
miles more than men who work in work tracts with under 175 jobs per square mile. Females

41
 working in high density work tracts commute 4 miles more on an average one-way trip to work
than females working in low-employment density work tracts.

Average commute times for both men and women almost double, going from 18 minutes in a
low-employment density area to 30 minutes in a high-employment density area for males and
going from 15 minutes in a low-employment density area to 28 minutes in a high-employment
density area for females. This strong increase in commute time may be attributable to the
increased traffic encountered in areas of high-employment density during peak hours (Table 46).

Retail Employment
Employment density gives indications for travel behavior for all workplaces. Dissecting
employment by Standard Industrial Classification (SIC) codes can offer a more refined view of
how certain industries affect an area’s transportation. This refined view becomes increasingly
important when one or two industries dominate an area’s employment.

TABLE 47: RETAIL TRADE BY AREA TYPE

Percent of the Work Home Block Group Area Type

Tract’s 16+ Population
Working in Retail Trade Second City Rural Suburban Town Urban All
0 to 9% 28.3% 23.5% 31.4% 26.7% 38.8% 29.5%
10 to 14% 16.8% 23.4% 18.7% 18.5% 19.1% 19.2%
15 to 24% 26.2% 32.1% 25.4% 29.6% 23.8% 27.4%
25% & up 28.8% 21.0% 24.5% 25.2% 18.2% 23.9%
Total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0%

Economists historically started defining American cities with a monocentric model in which all
development formed around a central business district. Retail employment has traditionally been
associated with urban areas. The metamorphosis of American land use has resulted in malls and
retail areas on the outer edges of metropolitan areas and outside cities. The 1995 NPTS shows
the plurality of workers living in urban block groups (38.8%) work in census tracts with less than
10% of the population working in retail trade. Urban areas also have the smallest percentage of
workers (18.2%) who are employed in tracts with over 25% retail trade participation. Second
cities have taken the lead in retail trade where 28.8% of residential block groups have residents
who work in tracts with over 25% of their employees working in retail. Rural areas follow
closely with where 32.1% of the workforce of rural block groups working where 15 to 24% of
jobs are in retail (Table 47).

42
 TABLE 48: MILES DRIVEN LAST YEAR BY RETAIL EMPLOYMENT AND GENDER

Percent of the Work Annual Miles Driven

Tract’s 16+ Population Male Female
Working in Retail Trade Mean Median Mean Median
0 to 9% 17,184 14,000 10,788 10,000
10 to 14% 18,164 15,000 11,225 10,000
15 to 24% 17,184 15,000 10,797 10,000
25% & up 17,077 15,000 11,044 10,000

The percent of retail employment in one’s work census tract does not appear to affect an
individual’s annual mileage driven. The median annual mileage for females remains constant at
10,000 miles, regardless of work tract retail employment. For males, median miles decrease
from 15,000 miles annually to 14,000 miles annually when census tract participation in retail
trade falls below 10%. This slight variation does not appear significant since mean mileage is
fairly consistent. The results suggest retail employment in the work tract does not affect annual
driving distances (Table 48).

TABLE 49: ONE-WAY WORK TRIP BY RETAIL EMPLOYMENT AND GENDER

Percent of the Work Distance to Work (Miles) Time to Work (Minutes)

Tract’s 16+ Population Male Female Male Female
Working in Retail Trade Mea Media Mea Media Mea Media Mea Media
n n n n n n n n
0 to 9% 14 10 11 8 25 20 23 20
10 to 14% 13 9 10 7 23 18 21 15
15 to 24% 13 8 10 6 21 15 19 15
25% & up 12 7 10 6 19 15 18 15

Retail employment may, however, provide some effect on work trips. High participation in the
retail workforce seems to indicate shortened commutes in terms of both distance and time. Males
working in census tracts with over 25% of the jobs in retail trade commute two miles fewer than
males working in census tracts with under 10% retail employment. This two-mile reduction in
commute distance results in a six-minute reduction in commute time. Females correspondingly
reduce their commutes by one mile or five minutes. Retail industries typically begin their
workdays after the morning peak period and often end their workdays after evening peak periods.
People commuting in areas with a high percentage of retail trade will find the population’s
commutes spread over a longer period of time than the typical peak period and commuters will
encounter fewer delays due to congestion.

43
FINDINGS AND CONCLUSIONS

Measures for People

Greater population density is associated with a number of travel outcomes including decreasing
annual miles driven for both genders for all population densities. This is most certainly affected
by the fact that bus availability increases with population density offering other travel choices for
urban residents. The existence of transit in more dense populations is associated with fewer
miles driven and transit is located most closely to households in the most densely populated
areas. Clearly increased density is highly correlated with decreased dependency on the single
occupancy vehicle resulting in fewer annual miles driven. People in lower density areas travel
longer distances to work and have longer commute times, however recently the most densely
populated areas are showing an increase in commute times. To some extent this is attributable to
increasing congestion levels in urban areas.

Less densely populated areas tend to have more drivers per adult and more vehicles per adult.
The lack of available travel options (modes) explains some of this. The private automobile
dominates as the most preferred mode of travel but the use of transit increases as population
density increases. Therefore we see greater reliance on the use of the automobile where travelers
have few or no alternatives for traveling. Whites are more heavily dependent on the single
occupancy vehicle than are other races. African Americans and Hispanics are slightly more
likely to use other forms of transportation. Transit availability is below average for whites and
above average for African Americans, Asians and Hispanics. Transit availability is generally
associated with increases in educational attainment. As educational levels increase these persons
tend to locate where they have more access to transit.

Wealthier households are most prevalent in suburban areas, middle income households are most
common in rural areas, and households with the lowest incomes are most common in second
city, urban and rural areas. Areas with the greatest percent of block groups living in poverty
tend to have the greatest accessibility (least distance) to transit.

Households with no children and a working-age adult to locate most closely to transit, while
families with children under 16 generally live farther from transit. Younger workers tend to have
shorter work trip distances and trip times, regardless of gender and females have shorter work
trip distances than males across all age groups. Cities attract young adults, both single and
married , who have no children; families with children tend to live in suburban areas and towns
while the elderly have their greatest presence in second cities.

Increasing density is associated with fewer person trips, person miles traveled (PMT) and person
miles per trip. Vehicle trips, vehicle miles traveled (VMT) and vehicle miles per trip are all
lowest in the most densely populated areas. Increased densification is one way to reduce miles
traveled although the critical question is what level of density would be necessary to cause a
significant reduction in miles traveled or a substantial increase in transit usage.

44
Measures for Places

Area Type
Controlling for area type revealed several correlations between land use and transportation.
These new NPTS data which quantify travel characteristics by area type for the first time help
identify and verify important trends in the interaction of urban form and travel behavior. Rural,
urban and second city areas show noteworthy travel characteristics.

Rural area residents depend heavily on private transportation. They make fewer person trips than
almost any other area residents in the country, yet their annual person miles traveled surpass the
residents of other areas. Vehicle trips show a similar pattern. Rural areas lead with the highest
ratio of young drivers to the adult population, and over one quarter of the residents own more
than one vehicle per adult. Rural areas have a lower than average percentage of workers who go
to work by private auto, but these areas also have the highest percentage of people who work
from home, which indicates either preference or a lack of mobility. Only 14.3% of rural
residents have access to transit service, compared to a national average of 63.4%. For over half
of those people who do have transit in their rural areas, bus service does not reach closer than one
mile from home.

Urban areas can be described at the opposite end of the spectrum. These areas lead with almost a
third of the population having less than one driver per adult and less than two percent of the
population having more than one driver per adult. Almost half of urban residents share private
vehicles or live without them. Urban workers commute across the shortest distances of anyone,
but the time length of their work trips slightly exceeds the duration commute trips in other areas.

Urbanites enjoy an array of transportation options, including transit, which is available for 98.3%
of the urban population. Over half of these residents have access to transit within one tenth of a
mile of their homes. Even with transit options available, 69.4% of urban workers commute to
work in an auto. This number represents less auto commuting than the national average of
75.9%, but it also expresses a strong preference for commuters to take an auto to work.

Urban dwellers reported the lowest number of person trips of anyone. Urban residents reported
driving just over half as many vehicle miles annually as rural residents. Urbanites cover less
distance in vehicles than people from any other area type.

Second cities, here defined as areas of concentrated population density with population centers
less dense than the core found in urban areas (see Key Terms and Definitions), have aroused a
great deal of interest in recent years. While edge city population and employment clustering has
been happening over decades, categorizing this new development phenomena has proven
challenging. The 1995 NPTS shows why: second cities display characteristics of several of the
other area types.

45
In some contexts, second cities follow national averages. For 80% of the population of second
cities, there is one driver for every adult, and there is one vehicle for each adult for 61.6% of the
population, which is slightly higher than the national average. Second cities follow the national
average for the percentage of workers working from home and the percentage of workers
working at a fixed workplace. Unlike urban and rural areas, where less than 70% of the workers
commute to work by auto, second city residents resemble the people of suburban areas in terms
of auto dependency for their commutes. Only 20% of second city residents go to work by a mode
other than a private automobile.

In other respects, second cities resemble traditional urban areas. Second city workers find
employment at distances from work equivalent to the distances that urban workers traverse, but
the commute for second city residents takes between five and seven minutes less time (between
one fifth and one fourth of the travel time).

Second city residents reported the highest number of person trips of any area type. The next
highest number of person trips came from suburban areas. Person miles traveled and vehicle
miles traveled by second city residents fell short of the national average. Second city residents
make frequent short trips.

In many ways, second cities represent a middle ground between urban and rural areas.
Approximately 82% of second city residents have bus service available to them, which is a far
higher percentage than for people in towns and rural areas but a lower percentage than for urban
and suburban residents. Those second city residents who have transit service available to them
typically live close to transit: 37.9% live within one tenth of a mile of transit, which is a far
higher percentage than in rural, suburban, and town areas.

Second cities will continue to provide transportation challenges. People living in second cities
enjoy the benefits of agglomeration, but they also prefer easy access to open spaces. The
conveniences of a thriving small city atmosphere quickly grows into a transportation challenge
when attempting to meet the needs of diverse residents. For instance, NPTS data indicate a high
percentage of low-income residents already live in second cities. Without the extensive transit
coverage that already exists in urban areas, the growing numbers of people who cannot afford
private vehicles will have a difficult time surviving in automobile-dependent second city
communities. Traffic congestion in these areas provides another challenge. Finding
transportation alternatives for these communities to meet their needs while maintaining the
in-between character of second cities poses a creative challenge to all transportation
professionals.

Residential Density
As might be expected, high residential densities typically occur in urban areas and low residential
densities typically occur in rural areas. About two thirds of second city and suburban people live
where residential densities range between 500 and 2,999 housing units.

Travel distances, including annual miles driven, commute trip length, person miles, and vehicle

46
miles, decrease as residential density increases. Interestingly, the distribution of commute time
for males is U-shaped, indicating longer trips at the lowest and highest residential densities.
Likewise, person trips and vehicle trips for both genders increase to a maximum value and then
decrease as residential density increases. People in areas of medium residential density make the
highest number of trips and have the longest commutes. Males in these areas have the longest
commute times.

Transit availability and residential density share a positive correlation. Increasing housing
density is associated with greater transit availability and closer proximity to transit. The
availability of alternative transportation facilities reflects itself in mode choice. People living in
higher residential densities rely less on private vehicles for trip-making than their counterparts
living in lower residential densities do. Bicycle and walk trips increase as residential density
increases.

Residential density correlates to some degree with employment, as well. Increasing employment
density is associated with increasing residential density. Residential density does not correspond
greatly with place of work decisions, but some slight variation exists. At residential densities
between 100 and 1,499 housing units per square mile, people are less likely to work at an unfixed
workplace. Low residential density areas have the greatest percentage of people who work at
home.

Age of Housing
The age of housing provides an indicator for the growth or decline in an area. New housing in an
area implies population growth and increased transportation demand. Second cities, which are
relatively new phenomena, should be expected to have a high proportion of recent builds. In
fact, second cities, towns, and suburban areas have the greatest proportion of housing built in the
last ten years. Urban and rural communities established the base of their housing infrastructure
prior to the last decade.

Implementing transportation strategies for new communities poses a challenge to keep pace with
growth. Data show that new builds are receiving transit service. Approximately, 72.9% of the
people who live in block groups comprised of 61-80% housing units built in the last ten years
have bus service. The least amount of bus service available (54.8%) occurs in block groups with
21-40% new builds. Transportation planners are generally meeting the demands of growth where
development is concentrated.

Housing Tenure
Housing tenure offers an indication of the likelihood of community residents to use transit.
Public transit serves over 50% of all housing types, but transit is most closely associated with
rental communities, where 77.2% of the residents have transit availability. Rental units, which
are typically densely oriented, are easy for transit to serve. Urban areas have the highest
percentages of renter-occupied housing compared to other area types. Non-rental units are
typically located in suburban areas.

47
Measures for Employment
Urban residents live in areas with high employment density and rural and town residents work in
areas with lower employment densities. Second city and suburban residents work in areas with
moderate employment density. However, the annual miles driven decreases for both genders as
work tract employment density increases. In addition, distance to work and time to work
increase for both men and women as employment density increases. The increased availability of
other modes of travel in densely populated areas, including the walk mode, would suggest
decreasing travel and commute times but the role of congestion must be considered.

Distance to work and travel time to work decrease as the percentage of retail trade in an area
increases. Urban areas have the smallest percentage (18.2%) of block groups with over 25%
retail trade , second cities have taken the lead in retail trade with 28.8% of block groups in
second cities having over 25% of their population in retail. Rural areas have 32.1% of rural
block groups with 15 to 24% of their populations working in retail.

OTHER RESEARCH
The land use special report for the 1995 NPTS provides a starting point for research in several
directions. Further work with the 1995 NPTS may include a closer look at the integrated effect of
land use and population variables. For instance, do particular races in urban areas have to pay to
park more than others? Does a parking fee affect mode choice for urban African Americans?
For urban Hispanics?

Other aspects of these data can be broadened. Now that employment density has been
established as a standard for land use studies, this variable can be further integrated into
descriptions of areas. This report explored the effects of employment density and retail
employment on travel behavior. Other areas of employment should also receive attention to
determine how to meet the needs of communities based upon their employment centers. The
changing Standard Industrial Classification (SIC) system will provide some interesting data in
this regard.

Future studies of land use and transportation should refine and expand independent variables.
Zoning, for example, will provide another interesting dimension to this exploration. Self-
reporting of land use characteristics should provide a good indication. Also, integrating NPTS
data with geographic information systems (GIS) in the future will open new areas of exploration.

The issue of how land use interacts with transportation opens more questions than one report can
possibly answer. This report endeavored to explore some initial areas of interest and lay a
foundation for future research in this area.

48
REFERENCES
Brindle, Ray, "Lies, Damned Lies and ’Automobile Dependence,’" Australian Transport
Research Forum, Volume 19, pp. 117-131.

California Air Resources Board; The Land Use and Transportation Linkage; Office of Strategic
Planning; Sacramento, CA.

Cervero, Robert, “Land Use Mixing and Suburban Mobility,” Transportation Quarterly, 1988.

Cervero, Robert, “Jobs-Housing Balancing and Regional Mobility,” JAPA, Spring, 1989, pp.
136-150.

Ferguson, Erik, “Recent Nationwide Declines in Carpooling,” 1990 Nationwide Personal Transportation Survey: Travel
Mode Special Reports, US Department of Transportation Federal Highway Administration, December 1994.

Frank, Lawrence D. & Pivo, Gary, “Impacts of Mixed Use and Density on Utilization of Three
Modes of Travel: Single-Occupant Vehicle, Transit, and Walking,” Transportation
Research Record, No. 1466, 44, 1994.

Garreau, Joel, Edge City: Life on the New Frontier, Doubleday, New York, 1998.

Lave, Charles and Crepeau, Richard, “Travel by Households Without Vehicles,” 1990 Nationwide Personal
Transportation Survey: Travel Mode Special Reports, US department of Transportation Federal Highway
Administration, December 1994.

Newman, Peter W. G. and Kenworthy, Jeffrey R., "Gasoline Consumption and Cities: A
Comparison of US Cities with a Global Survey,” APA Journal, Winter, 1989, pp. 24-37.

USDOT, "Edge City and ISTEA: Examining the Implications of Suburban Development
Patterns," 1992.

49
1. Gordon, Peter and Richardson, Harry W., Geographic Factors Explaining Work Trip Length
Changes,” 1990 Nationwide Personal Transportation Survey: Special Reports on Trip and
Vehicle Attributes, US Department of Transportation Federal Highway Administration, February
1995, Chapter 2

2. Garreau, Joel, Edge City: Life on the New Frontier, Doubleday, New York, 1998, pp. 6-7.

3. Miller, David R. and Hodges, Ken, "A Population Density Approach to Incorporating an
Urban-Rural Dimension into Small Area Lifestyle Clusters," paper presented at the Annual
Meeting of the Population Association of America, Miami, Florida, May 4-7, 1994.

4. Lave, Charles and Crepeau, Richard, "Travel by Households without Vehicles," 1990
Nationwide Personal Transportation Survey: Travel Mode Special Reports, US Department of
Transportation Federal Highway Administration, December 1994, Chapter 1, p. 24.

5. Ferguson, Erik, "Recent Nationwide Declines in Carpooling," 1990 Nationwide Personal

Transportation Survey: Travel Mode Special Reports, US Department of Transportation Federal
Highway Administration, December 1994, Chapter 2, p. 18.

6. Garreau, ibid.

50