THE CREATION OF A MODEL TO ESTIMATE THE CHANGE IN NON-MOTORIZED
TRANSPORTATION USE AFTER A LAND USE CHANGE
By
Mark Jones

A THESIS

Submitted to
Michigan State University
In partial fulfillment of the requirements
For the degree of
MASTER OF URBAN AND REGIONAL PLANNING
Urban and Regional Planning
2012

ABSTRACT
THE CREATION OF A MODEL TO ESTIMATE THE CHANGE IN NON-MOTORIZED
TRANSPORTATION USE AFTER A LAND USE CHANGE
By
Mark Jones
Studies have shown that an increase in non-motorized transportation (NMT) has been
linked to healthy benefits such as an increase in physical activity, lower body mass index, and a
decreasing chance of developing a chronic disease. New theories of economic development, such
as Placemaking, have also emphasized increasing NMT. However, there is no tool to quantify
the changes in NMT ability before and after a development. This project develops such a
modeling tool and applies it to the City Center II development in East Lansing, MI. A literature
review identifies built environment variables that explain the variance in American Community
Survey walking, biking, and riding the bus to work data. After conducting correlations with the
means to work and built environmental data, the final variables are used in a linear regression,
along with socio-economic controls. A service area is created utilizing ArcGIS’s Network
Analyst tool to define an area ½ mile around the development. Derived regression coefficients
are applied to the service area in order to estimate the pre-development amount of people who
took NMT to work. The anticipated final redevelopment project is then reproduced in ArcGIS
and the service area is recreated. The same regression coefficient formula is applied to the new
service area and the result is the estimated number of people who will take NMT to work postdevelopment. The results show that the development will increase the amount of people who
take NMT to work from 28.58% to 30.68%. Finally, the benefits and policy implications of using
a model to assess the change in NMT use before and after a land use is discussed.

ACKNOWLEDGEMENTS

This project would not have been possible without the support from many people. First of
all, I would like to thank my parents for not asking too many questions about what my project
was about, as it went through many variations. Also, I would like to thank my girlfriend,
Amanda Vig, for the support and encouragement. Finally, I would like to thank the Tri-County
Regional Planning Commission, particularly Chief Transportation Planner Paul Hamilton, for
offering me an internship. I was able to conduct most of my thesis research while concurrently
creating a model that was applied to their TAZs. This helped me stay on track to finish the thesis
and it gave me an outlet to ask interpretation and method questions.

iii

TABLE OF CONTENTS

LIST OF TABLES...........................................................................................................v
LIST OF FIGURES........................................................................................................vi
LIST OF ABBREVIATIONS........................................................................................vii
CHAPER 1 INTRODUCTION..............................................................................….....1
1.1 Background of Transportation Modeling………………………….………..1
1.2 Benefits of Non-Motorized Transportation………...……………….………3
1.3 Non-Motorized Transportation Variables…………...…………….………..4
1.4 Travel Distances………………………………………………….…………8
1.5 Non-Motorized Transportation Variable Selection……………….………...9
CHAPTER 2 APPLICATION…………………………………………………..……...12
CHAPTER 3 METHODS……………………………………………………..………..14
CHAPTER 4 RESULTS………………………………………………………...………17
CHAPTER 5 DISCUSSION…………………………………………………...……….21
5.1 Model Results……………………………………………………..………...21
5.2 Policy Implications……………………………………………..…………...22
CHAPTER 6 LIMITATIONS AND FUTURE RESEARCH……………..…….……...25
6.1 Critique of Model………………………………………………..….……….25
6.2 Future Use for the Model………………………………………..….……….27
APPENDIX A……………………………………………………………….…...…...…31
APPENDIX B……………………………………………………………….…...……...33
APPENDIX C………………………………………………………………...….……...34
REFERENCES……………………………………………………………..…….……..36

iv

LIST OF TABLES

Table 1 Non-Motorized Transportation Variables....................................................6
Table 2 Non-Motorized Transportation Variables...................................................11
Table 3 Regression Outputs....................................................................................18
Table 4 Variable Measurements..............................................................................19
Table 5 Changes in Estimates..................................................................................20
Table 6 Correlations……………………………………………...……….……….31
Table 7 Model without SES Controls………………………………………..……33
Table 8 Full Model…………………………………………………….…………..34

v

LIST OF FIGURES
Figure 1 Pre-Development........................................................................................15
Figure 2 Post-Development......................................................................................15

vi

LIST OF ABBREVIATIONS
ACS – American Community Survey
CIAA - Corridor Improvement Authority Act
GIS – Geographic Information System
MACIA - Michigan Avenue Corridor Improvement Authority
MSA – Metropolitan Statistical Area
NMT – Non-Motorized Transportation
PLT – Path, Lanes and Trails
SES – Socio-Economic Status
TAZ – Transportation Analysis Zone
TCRPC – Tri-County Regional Planning Commission
VTPI – Victoria Transportation Policy Institute

vii

CHAPTER 1
INTRODUCTION

1.1 Background of Transportation Modeling
Today, decision makers rely on a spectrum of modeling tools in order to make
transportation decisions. For example, every federally designated metropolitan planning
organization employs some sort of transportation forecasting model (Hardy, 2011). The problem
with these modeling systems is that very few look at the transportation and land use system in an
integrated way and non-motorized transportation (NMT) is rarely included in these models
(Hardy, 2011; Rodriguez & Joo, 2004).
The relationship between land use and transportation has been studied for over 50 years,
but little understanding exists about how the built environment influences travel behavior (Frank
et al., 2008). Many planners understand that conceptually and intuitively, a transportation/land
use connection exists and should be addressed at all levels of transportation and land-use
planning; however, they lack information on exactly how that connection works and how it can
be measured (Moore & Thorsnes, 2007).
Research has shown that there is a connection between the built environment and which
mode of transportation an individual will take (Abdul et al., 2007; Chen et al., 2008). This
influence has also shown to extend to an individual’s choice of taking NMT (Winters et al.,
2010; Killingsworth et al., 2003; Belanger, 2010; Handy et al., 2006; Soltani et al., 2006).
However, there has also been a growing amount of research that suggests people will self-select
areas that encourage NMT to live in, so the built environment will not influence walking or
biking more, but it will attract those who like to take those modes of transportation (Handy et al.,

1

2006). In any case, the built environment has a role in being able to predict which mode of
transportation an individual will take.
Issues including, but not limited to, lack of reliable data, computational power, or
knowledge of non-motorized travel behavior have hampered modeling NMT for metropolitan
areas (Iacono et al., 2010). Despite the lack of understanding, the federal government still spent
$3.17 billion on 10,012 bicycle- and pedestrian-related projects for states and counties, between
1992 and 2004 (Cradock et al., 2009).
In 2009, researchers looked at the 10,012 projects from 1992 and 2004 and came up with
three policy recommendations for improving public health with federal grants, one of which was
to show an explicit link between transportation and public health (Cradock et al., 2009). This can
be accomplished through quantifying the change in non-motorized transportation after land use
changes. However, the transportation modeling world has been slow to include these variables in
their forecasting. This paper aims to fill the gap in understanding and predicting the connection
between land use change and its effects on NMT.
This paper develops a model that is used to estimate the number of people who currently
take NMT to work and thenestimates the amount of people who take NMT to work after a land
use change. The example land use change is a proposed mixed-use development in downtown
East Lansing, Michigan – City Center II. The creation of such a model has other benefits than
acquiring federal grant money, including:
1. Local officials can use this model as a marketing tool to show citizens how a
project will increase the NMT ability of an area.
2. Developers are able to calculate the improvements in NMT a development will
create in order to help obtain more public funding.

2

3. Citizens are able to identify how walkable/bikable an area is post-development
prior to project completion, which would enable them to reserve a unit sooner.

1.2 Benefits of Non-motorized Transportation
There is evidence that non-motorized transportation is associated with economic
development. From a local government standpoint, walkability increases property values; it is
cheaper to build and maintain sidewalks and bicycle parking facilities than it is to build and
maintain roads and parking lots/structures; and pedestrian infrastructure projects have a higher
employment multiplier than other transportation projects (Pivo & Fisher, 2011; Litman, 2009;
Garrett-Peltier, 2011).
There are also personal economic benefits for users of NMT. Costs for such things as
parking, vehicle maintenance, and fuel can be reduced through taking NMT to work (Litman,
2009). In addition, other overlooked or undervalued benefits arise when taking NMT. For
instance, it is less expensive to ride a bike for exercise than to join a health club, and one is less
likely to obtain a traffic violation, among other benefits (Litman, 2011).
Researchers have also found that non-motorized transportation is also linked to such
health benefits as a lower body mass index (Konikov-Titievsky, 2010; Lovasi et al., 2009; Berry
et al., 2010) and a lower risk of developing a chronic disease (Stone & Gostin, 2003). In
addition, the mental health and social benefits of the reduced isolation from living in a walkable
community can help raise the quality of life and the social capital of an individual (Leyden,
2003; Wood et al., 2010; Rogers et al., 2011).
Non-motorized transportation has differing benefits for differing cohort groups.
Specifically, walking is a very valuable health tool for senior citizens. For seniors, walking is a

3

way to lower blood sugar, reduce body fat, lower blood pressure, improve bone density, and
keep good mental health (Witt, 2010). Research has shown that in addition to personal, social,
and organizational related factors, the built environment plays an important role in inhibiting
senior participation in physical activity (Li et al., 2005; Joseph, 2006; King, 2006). The senior
citizen population – those over 65 years old – increased by 15.1% in the United States between
2000 and 2010, and it is estimated to more than double between 2010 and 2050 (Administration
on Aging, 2010). Therefore, it is more important than ever for communities to make sure that the
built environment encourages walking and does not impede it.
From a theoretical standpoint, non-motorized transportation is also a key asset in
Placemaking. The process and result of making a place with unique features that people want to
use, be in, enjoy and/or remember, because it was interesting or pleasurable, is called
Placemaking (About 21c3). According to the Center for 21st Century Communities, there are
eight key assets which are critical to making a vibrant community, two of which are physical
design &walkability and transit. Placemaking is key to job creating and retaining knowledgeable
workers in the New Economy (Wyckoff, 2010). Therefore, non-motorized transportation is also
an important element in future economic development.

1.3 Non-motorized Transportation Variables
Non-motorized transportation typically includes walking, biking, and variants such as
small-wheeled transport and wheelchair travel (VTPI, 2010). However, a growing body of
research shows that taking public transit will have the same health, environmental and
economically beneficial uses as walking and biking (Maghelal, 2011; Litman, 2012). This has

4

led some researchers to expand their definition of NMT to include public transit (Appleyard,
2010; Neiman, 2010).
Researchers have derived many ways to calculate the influence of the built environment
on people’s ability to walk, ride a bike, or take public transportation. This section will look into
the variables most commonly included and the methods used.
Densities, diversity of land uses, and the design of streets and transit systems are
commonly assembled to create a walkability index (Lovasi et al., 2011). Perhaps the first well
known study to attempt to quantify walkability was the 1000 Friends of Oregon 1993 report,
Making the Land Use, Transportation, Air Quality Connection. The variables chosen to analyze
walkability were ease of street crossings, sidewalk continuity, local street characteristics (grid vs.
cul-de-sac), and topography (1000 Friends of Oregon, 1993).
Since the 1993 Friends of Oregon report, there have been many variations of the elements
that go into a NMT index or predictive models. The walkability index consisting of net
residential density, retail floor area ratio, intersection density, and land-use mix created by
Lawrence Frank from the University of British Columbia. has become the most common
calculation for walkability (Frank et al., 2006; Frank et al., 2007; duToit et al., 2007; Leslie, et
al., 2007; Owen et al., 2007; Frank et al., 2008; Gebel, 2009; Marshall et al., 2009; Sallis et al.,
2009; Frank et al., 2010). Other researchers have used this model but did not include the retail
floor area ratio because of a lack of resources (Sundquist et al., 2011; de Sa, 2011).
There are indices or models whose variables are chosen because of the availability of
information/data, and Frank et al. (2010) acknowledges that using the four elements previously
mentioned is sometimes out of necessity and not proven by research:

5

“Although many other variables have been hypothesized and in some cases shown to
be related to active transport such as sidewalks, traffic calming and intersection
characteristics, those variables are not widely available in existing datasets, so the
present four-component index represents a practical starting point and improved
versions can be developed as conceptualization and data availability improve.”

There are examples of NMT models and indexes that use other variables. Table 1
provides an overview of these variables. Each of these variables has a multitude of ways to
calculate them.

Table 1 Non-Motorized Transportation Variables
Built Environment
Variable
Public Transportation
Employee Density
Mixed-land use

Park Access
Population Density
Presence of Sidewalks

Street Connectivity

Reference
Lovasi et al., 2009; Konikov-Titievsky, 2010; Seeley,
2005; Forsyth et al., 2008;
Maghelal, 2011; Forsyth et al., 2008; Kim, 2006
de Sa, 2011; Forsyth et al., 2008; Kim, 2005;
Konikov-Titievsky, 2010; Lovasi, 2006;
Lovasi et al., 2009; Ozbil et al., 2008; Song et
al., 2007
Lovasi, 2006; Forsyth et al., 2008;
Konikov-Titievsky, 2010; Lovasi et al., 2009; Kim,
2005;
Harkey et al., 1998; Lovasi, 2006; Maghelal, 2011;
Seeley, 2005; Lovasi et al., 2009; Forsyth et
al., 2008
du Toit et al., 2007; Forsyth et al., 2008; de Sa, 2011;
Maghelal, 2011; Ozbil et al., 2008; Song et
al., 2008

According to the Victoria Transport Policy Institute, connectivity “refers to the density of
connections in path or road network and the directness of links…Connectivity can apply both

6

internally (streets within that area) and externally (connections with arterials and other
neighborhoods).” With the advent of geographic information systems (GIS) in the 1990’s, street
pattern has been more commonly used as a measure of walkability (Forsyth et al., 2008). There
are many different ways to calculate connectivity, including block length, block size, block
density, intersection density, street density, connected node ratio, link-node ratio, grid pattern
pedestrian route directness, and effective walking area (Dill, 2004; Tresidder, 2005). Nearly all
NMT models include at least one component of street connectivity.
Mixed-land use is defined as the level of integration within a given area of different types
of uses for physical use (Konikov-Titievsky, 2010). Land use mix, especially the close proximity
of shopping, employment and food to residential areas is shown to be related to greater walking
and cycling among residents (Saelens et al., 2003; Mouden et al., 2006).
Similar to street connectivity, researchers have found many ways to calculate land-use
mix. For example, Frank et al., (2010) created an entropy score, which consisted of five land
uses: residential, retail, (excluding “big box” stores), entertainment (including restaurants),
office, and institutional (including schools and community institutions). The amount of land
cover per census tract was calculated and then normalized between 0 and 1, with 0 being a single
use and 1 indicating a complete even distribution of floor area between the five uses.
Other examples of calculations for mixed-land use include taking the percentage of total
parcel area in different land uses and applying those unique ratios to the model (Forsyth et al.,
2008). Resources such as Walkscore.com use destination based analysis in order to calculate
land-use mix. Walk score uses an algorithm to create a single score based on the distance to such
destinations as restaurants, parks, post offices, hospitals, shopping centers, bars, grocery stores,
etc. (Walk Score, 2011).

7

1.4 Travel Distances
Researchers have created predictive models and indexes for a variety of study areas.
The size of the study area varies greatly from large places such as TAZs, neighborhoods, census
block groups, zip codes to smaller areas like ½ and ¼ mile buffers and 1 km buffers. However,
spatial units, such as census block groups, can be too large to capture the variations in urban
form that occur at a much smaller scale (Frank et al., 2007). A 1-kilometer (0.6 mi) buffer
created along a road network is the distance that can be covered in a ten-minute walk (Frank et
al., 2007). Therefore, a smaller buffer, one that is in non-motorized transportation traveling
distance to the study area is the best choice for analysis.
The average distance traveled by destination walking and bicycling varies slightly. In a
study conducted in Montreal, Canada the median walking distance was 650 meters (0.4 miles)
and the average distance traveled by bicycle was two kilometers (1.2 miles) with a high degree
of variation (Larsen et al., 2010). In 1999, the Wisconsin Survey research Laboratory conducted
a Bicycle and Pedestrian Survey. They found that the most common distance traveled by walking
was between ½ mile and a mile; the most common distance traveled by bicycle was between ¼
and ½ mile each way.
The average distance walking to transportation is very similar to that of walking and
biking to any other destination. In a case study conducted at the Mountain View, California
public transportation station area, people walked 0.43 miles from the north and 0.65 miles from
the south to the transit stop (Park, 2008). In an Atlanta, GA study the median distance traveled to
a transit station was 0.47 miles (Ozbil, 2010). O’Sullivan (1996) found that the average walking
distances to suburban bus stations was 0.4 miles and the average distance walked to central
business district bus stops was 0.2 miles, in Calgary.

8

The selection of a network or circular buffer has a considerable influence on the results of
a project (Oliver at al., 2007). A circular buffer, commonly known as ‘straight-line’ or ‘as-thecrow-flies’, can include disconnected areas in the calculation of a service area. A network buffer
is created along pathways that are accessible from the study location; therefore, network buffers
are more of a reflection of the NMT area that a person can cover. One method for calculating a
network buffer is to use the Network Analyst Extension in ArcView, where the size of the buffer
varies based on the connectivity of the road network – more intersections allow a greater area to
be covered on the ground (Frank et al., 2007).

1.5 Non-motorized Transportation Variable Selection
Several aspects of social-economic status (SES), including educational attainment and
income have been shown to contribute to an individuals’ amount of physical activity (Cerin et
al., 2009). Physical characteristics of the neighborhood environment, access to recreational
activities, perceived safety and security are all potential reasons for the discrepancy in resident’s
SES and their neighborhood’s walkability (Cerin & Leslie, 2008; Edwards & Tsouros, 2006).
Other researchers have also discovered a link between incomes and built environmental variables
that influence walking. For example, higher income neighborhoods experience greater park use
which results in higher income individuals walking more – this is because high-SES residents
have access to parks with higher levels of safety, maintenance and attractiveness (Leslie et al.,
2010).
Researchers have commonly used controls in their predictive models in order to isolate
the influence that the built environment has on walkability/bikability. Some of the more

9

common examples of controlled variables include: income, sex, age, educational attainment,
ethnicity, and car availability (Owen et al., 2007; Maghelal, 2011; Kim, 2005).
In regard to commuting to work, higher salary income and more expensive housing is
associated with a greater propensity to working from home, but a lower propensity of walking or
biking to work. College education in most cases is associated with taking NMT to work. Lastly,
other variables such as car ownership, race, and gender are shown to influence the likelihood of
taking NMT to work, but the results have sharp differences across sub-regions within
metropolitan areas (Plaut, 2005).
A shortcoming of current methods to quantify NMT is that they utilize all variables at
their disposal without checking if those variables contribute to increasing NMT or if they have
multicollinearity. For example, Lovasi et al, (2009) included population density and bus access
in their walkability index. Although, it would seem likely that bus stops are placed in pockets of
high population density. In addition, most of the projects focus on one aspect of NMT, although,
all three have the same positive benefits.
The above research studies and the variables identified were used to select the current
study potential variable list. Including the destination based variables: bus stop density,
employee density, population density, commercial property density, park/school density, and
household density, three connectivity variables (connected node ratio, intersection density, and
effective walking area) and two land use variables (path, lane, and trail density, and sidewalk
presence) were selected to be included in the predictive model. Table 2 has a brief description of
the measure, definition and data source used to calculate the variables.

10

Table 2 Non-Motorized Transportation Variables

Connectivity

Measure
Connected node
ratio

Definition
Ratio of real nodes (intersections
with three or more legs) to total
nodes

Data source
Node shapefile created
by Mark Jones using all
roads file from TCRPC*

Ratio of parcels per acre

TCRPC

Employees per acre

TCRPC

Bus stop density

Bus stops per acre

TCRPC shapefile
obtained from Capital
Area Transportation
Authority

Population density

Population per acre

2005-2009 ACS**

Household density

Ratio of households per acre

2005-2009 ACS

Commercial
property density

Acres that are zoned commercial
divided by total acres

TCRPC generalized
zoning file

Park and school
density

Ratio of land area for parks,
elementary schools, middle
schools, high schools, and MSU’s
main campus (acres) to the land
area (acres).

TCRPC

Sidewalk presence

Destination

Node shapefile created
by Mark Jones using all
roads file from TCRPC

Effective walking
area
Employee density

Land Use

Intersection density Ratio of real nodes per acre

Ratio of sidewalk length (ft.) to
road length (ft.)

Tri-County Regional
Planning Commission
(TCRPC) all roads and
sidewalks file

Path, lane, and trail
density

The average length of bike trails,
widened shoulders, walking trails,
and bike lanes (ft.) per acre

TCRPC

* Tri-County Regional Planning Commission
** American Community Survey

11

CHAPTER 2
APPLICATION

The City Center II proposed project is at the intersection of Michigan Avenue and West
Grand River Avenue, in East Lansing, MI. The $105 million, 5.25-acre project includes a tenstory, mixed-use building, which will consist of residential units, retail, restaurant and office
space and a 400-seat performing arts theater. One block to the north of the 10 story structure will
be a five-story, mixed-use building with retail and a restaurant on the first floor and residential
units on the upper floors. Also, the project includes a parking garage, a four-story residential
building, and nine (9) townhouses (City of East Lansing, 2012).
The City Center II project is just one of many developments and rehabilitations occurring
along the Michigan/Grand River corridor since the creation of the Michigan Avenue Corridor
Improvement Authority (MACIA). The MACIA was created by the 2005 Corridor Improvement
Authority Act (CIAA), P.A. 280 of 2005, and amended in 2007. The CIAA provides for the
creation of a public economic development corporation that will work to “correct and prevent
deterioration in business districts, encourage historic preservation, promote economic growth”
(Michigan Ave Corridor, 2011).
The Act allows multiple municipalities to collaborate and establish one unified authority
with jurisdiction over the entire corridor. The authorities can create broad initiatives, but they
must include three things: allow for mixed use and high density residential, expedite permitting,
and support non-motorized transportation. The partners in MACIA include the City of Lansing,
City of East Lansing, Lansing Township, and Michigan State University. According to the
Michigan Avenue Corridor Improvement Authority website:

12

“The ultimate goal of the project is to revitalize and beautify the Michigan Avenue
corridor to make it more walkable, attract a larger base of customers and create a
pleasant living environment for surrounding neighbors.”
With the City Center II project being constructed under this setting, it would imply that
the project will increase the amount of people who take non-motorized transportation. However,
there is no tool at the disposal of the Authority which can estimate the change in the amount of
people who take non-motorized transportation after a development. This study develops such as
tool and applies it to the City Center project to assess the effects of development on NMT.

13

CHAPTER 3
METHODS

The 2005-2009 American Community Survey (ACS) means to work data – walking,
biking and public transportation - for all 117 census tracts within the Lansing tri-county
metropolitan area (Ingham, Eaton & Clinton Counties) is used as the dependant variable in the
NMT model. The eleven (11) variables in Table 3 are selected from the literature review and the
values are calculated per census tract, and are the independent variables.
Utilizing ArcGIS, the NMT variables are calculated per census tract in order to run
correlations. The sidewalk presence, path, lane, and trail density, commercial property density,
and park and school density variables are calculated using the Intersect tool. The connected node
ratio, bus stop density, employee density, and intersection density are calculated using the Spatial
Join tool. Finally, the population density and household density values are calculated using data
gathered from the 2005-2009 ACS website.
A one-half mile service area is created around the current structures in the development
region. The service area is created using ArcGIS’s network analysis tool and by placing facilities
around the perimeter of the project area. The resulting service areas are merged into one
shapefile resulting in the pre-development NMT service area.
The plans used for site plan review for the City Center II proposal were obtained from the
City of East Lansing website, and are used for the post development measurements for the
current NMT model. Using these plans, the physical environment changes are duplicated in
ArcGIS (Figure 1). A new service area is created using the new built environment characteristics
and a final shapefile is created called the post-developed NMT service area (Figure 2).

14

Figure 1 Pre-Development

Figure 2 Post-Development

For interpretation of the references to color in this and all other figures, the reader is referred to the
electronic version of this thesis.
15

The same ArcGIS tools are utilized to calculate the eight built environment values in the
pre- and post-development service areas. However, the population density is more difficult to
determine. The service areas are over laid with 2010 TIGER/Line shapefile block data. The
blocks that the service areas covered are included in the population density calculation. The final
pre-development value of 12.75 people/acre is the sum of the blocks that the service area covered
(580.77 acres) divided by the 2010 population of those blocks (7,403).
Currently, where the proposed development will occur, there are three vacant buildings
owned by the City of East Lansing and six residences with 59 occupants (Mullins, 2012). The
City Center II development will contain three (3) one-bedroom units, one-hundred and three
(103) two-bedroom units, and sixteen (16) three-bedroom units. Assuming that there will be full
occupancy, the development will increase the population by sixty-three (63).
Finally, the derived regression coefficients in the full model are applied to the pre- and
post-development service areas and their built environment variable values. The results are then
compared in order to estimate the change in the amount of people who take NMT to work after
the construction of City Center II.

16

CHAPTER 4
RESULTS

All variables are texted to identify in any multicollinearity exits. Any variable that
demonstrates multicollinearity is eliminated from consideration. Household density, effective
walking area and bus stop density are subsequently eliminated because of multicollinearity,
leaving eight variables to be included in the model(s) (See Appendix A for correlations). The
control variables selected were based on data availability from the ACS. Per-capita income1,
median age, ethnicity2, gender3, and educational attainment4 data for all Lansing tri-county
census tracts are correlated. Per-capita income and median age are found to have a correlated
value of 0.7352; therefore, median age was eliminated. Ethnicity and education were found not
to be significant when inserted into the predictive models, so those control variables were also
left out. The final control variables left in the model were per-capita income and gender. The
final model, with control variables and without, is displayed in Table 3.

1
2
3
4

Per-capita income in the past 12 months (in 2009 inflation-adjusted dollars)
Percent of “white only” population
Percent male
Percent of population with an Associate’s, Bachelor’s, Master’s, or Professional Degree
17

Table 3 Regression Outputs
(See Appendix B & C for full regression outputs)
Model w/o SES
Controls

Full Model

Income

-3.992E-6***
(-2.003)

Gender

0.367*
(1.742)

Intercept

-0.237**
(-2.568)

-0.271**
(-2.003)

Population density

0.009**
(2.277)

0.003
(0.730)

Path, lane, and trail
density

0.003***
(3.759)

0.003***
(4.669)

Sidewalk ratio

0.056***
(2.734)

0.046**
(2.341)

Commercial property 0.179
density
(1.588)

0.153
(1.444)

Real node ratio

0.297**
(2.555)

0.251**
(2.294)

Park/school density

0.200**
(2.195)

0.237***
(2.754)

Employee density

0.005***
(4.586)

0.004***
(3.783)

Intersection density

-0.799***
(-4.886)

-0.597***
(-3.703)

R2

0.703

0.744

0.681

0.720

Adjusted R

2

F-statistic
31.951***
30.874***
n
117
117
Note: Numbers in parentheses are t-statistics, */**/*** indicates
significance at the 0.10/0.05/0.01 percent confidence levels

18

The final result is an estimate that 28.58% of the workers sixteen and over within the City
Center II pre-development NMT service area take NMT to work. The percentage will increase to
30.68% post-development. If these estimated percentages are applied to the calculated
population totals, the City Center II project will increase the amount of people who take nonmotorized transportation to work from 2,115 to 2,291, within its NMT service area.
Table 4 shows the final calculations for all nine variables for the pre- and postdevelopment non-motorized transportation service areas. Refer to table 3 for units of
measurement for each variable.

Table 4 Variable Measurements

Population Density
Path, lane, and trail density
Sidewalk ratio
Commercial property density
Real node ratio
Park/school density
Employee density
Intersection density

Pre-Development
12.75
80.06
1.9001
0.1565
0.9047
0.2736
14.794
0.3084

Post-Development
12.86
81.35
1.9211
0.1572
0.9274
0.2748
17.648
0.3111

All of the variables increased in density, although, employee density had the largest increase.
Employee density increased from approximately 15 employees per acre to 18 – resulting in a
19.3% gain.
Table 5 explores where the greatest gains will occur post-development, according to the
model. The top line of percentages shows the estimated influence that each variable had in the
overall amount of people who took NMT to work pre-development. The second line of
percentages has the estimated change post-development. Those two lines added up equal the

19

third line, the final estimated percentage of people who will take NMT to work postdevelopment.

20

Table 5 Changes in Estimates
Pre-Development
NMT Estimates
Path, Lane, and Trail Density
Connected Node Ratio
Sidewalk Presence
Park/School Density
Employee Density
Population Density
Commercial Property Density
Intersection Density

5

24.018%
22.708%
8.740%
6.486%
5.918%
3.825%
2.394%
-18.413%

Changes in
Estimates After
Development
0.387%
0.57%
0.097%
0.027%
1.142%
0.033%
0.011%
-0.162%

Total

5

24.405%
23.278%
8.837%
6.513%
7.059%
3.858%
2.405%
-18.575%

The TOTAL line adds up to 57.78% because the constant value of (-27.1%) needs to be added.
21

CHAPTER 5
DISCUSSION

The purpose of this project was to develop a model that estimates the change in NMT
after a land use change. The example of a redevelopment in East Lansing was chosen because of
the ease of access to the development details. Therefore, this example of the application of the
model is reflective of the land use characteristics of the Lansing MSA. This area includes the
State of Michigan Capitol, Michigan State University, and other unique land uses.
While the results of this model are particular to the Lansing MSA, the methods can be
universally applied. This can result in substantial changes in the way that non-motorized
transportation grants are applied for and awarded. In addition, if a method to quantify the
outcomes of NMT projects can be created, the disproportioned rate of traditional motorized
transportation projects compared to non-motorized transportation may start to decrease.

5.1 Model Results
This model was created using demographics and built environment data from the Lansing
tri-county region, which has a large student population. As previously mentioned in the literature
review, income has an influence on a person’s likelihood of taking NMT to work, and students
are predominantly lower-income. Students also live in close proximity to college campuses,
where many of them work. Michigan State University, which has enrollment of over 47,000, has
an extensive system of sidewalks, walking trails, and bike lanes. Therefore, this model has the
regional influence of a high correlation between paths, lanes, and trails and taking NMT to work.

22

This is not a negative on the model; it has simply adapted to the unique characteristics of the area
it was designed for.
Urban designers emphasize the importance of micro-scale features that add up to
influence an individual’s use and experience of the neighborhood environment (Alfonzo et al.,
2008). This project adds to that literature. The eight built environment characteristic changes
accumulated to increase the amount of people who took non-motorized transportation to work,
but it was through small increases to each variable. Although the total percentage of the amount
of people who took NMT to work increased by 2.1%, it was through incremental increases in all
variables.
In regard to the City Center II development, according to the City of East Lansing City
Center II project website, the “project will make East Lansing more pedestrian and bicycle
friendly by bringing residents closer together to where they work, shop and partake in leisure
activities.” This analysis suggests that the statement is statistically factual. However, it remains
uncertain that that the results indicate a functional difference in NMT from a planning,
community, or placemaking perspective. The development will not significantly increase the
population density, but it will have a positive effect on the employee density. The development
will bring more jobs to the area, but the overall increase in NMT use – from 28.6% to 30.7% - is
negligible.

5.2 Policy Implications
In 2005, Public Law 109-59 was passed. Section 1807 established a Non-Motorized
Transportation Pilot Program in which four communities would receive $6,250,000 per fiscal
year, for four years, in order “to demonstrate the extent to which bicycling and walking can carry

23

a significant part of the transportation load, and represent a major portion of the transportation
solution, within selected communities." Over the course of four years the four communities Columbia, MO; Marian County CA; Minneapolis Area, MN; Sheboygan County, WI - were
required to “construct... a network of nonmotorized transportation infrastructure facilities,
including sidewalks, bicycle lanes, and pedestrian and bicycle trails, that connect directly with
transit stations, schools, residences, businesses, recreation areas, and other community activity
centers."
The final report for the pilot programs was published in April, 2012 and it went into
detail regarding the insights and lessons learned. One of the lessons learned was that it took
longer time than the program offered to identify, plan, implement projects, and collect data for
program evaluation (Federal Highway Administration, 2012). This process could be streamlined
by forecasting demand changes using the NMT model presented in this study, similar to how
other transportation models operate. The ability to predict changes in usage prior to the actually
construction of infrastructure and changes in land use would be an advantage when apply for
federal grants and reporting on their progress.
On August 1, 2010 the State of Michigan was the first state in the union to pass Complete
Streets legislation (P.A. 135, of 2010). The City of Lansing was the first municipality in
Michigan to pass a Complete Streets Ordinance in 2009 and Lansing Township was the first
township to pass the ordinance in 2011. The State of Michigan Law and local ordinances require
that transportation designing and planning should consider all road users, including pedestrians
and bikers, during all phases of a transportation project. The decisions to include such things as
bike lanes and sidewalks will ultimately be sensitive to budget, time, and feasibility, among other
factors. This leaves a gap in decision making when prioritizing complete streets projects.

24

Without the ability to predict the benefits of a complete street, non-motorized transportation
cannot be judged or valued relative to other transportation mode projects.
Future improvements to regional non-motorized transportation models will help place
bicycles and pedestrians on a “level playing field” with modernized modes in transportation
planning (Porter et al., 1999). States spend only 1.6% of their federal expenditures on bicycle
and walking, or approximately $2.17 per capita (Alliance for Biking & Walking, 2012). As the
amount of non-motorized transportation benefits expands, the ability to spend the small amount
of appropriated money wisely becomes a priority.

25

CHAPTER 6
LIMITATIONS AND FUTURE RESEARCH

6.1 Critique of Model
This model is development specifically for the Lansing-East Lansing MSA. Therefore,
the coefficient values may not accurately reflect other regions. Some variables, such as
topography and subway access, are omitted because they did not apply to this region. While
others, such as retail floor area ratio, sidewalk cover, road speed, and traffic counts, are left out
because of a lack of information/data. Despite these data limitations, the final model did account
for a large amount of the variance (r-square value of 0.744).
Further research needs to be conducted in order to understand the relationship between
the variables found not to be significant and NMT. Out of the eight variables included in the
model two were not significant at the 0.10, 0.05, or 0.01 levels – population density and
commercial property density. Commercial property density had a significance level of 0.153, so
the variable was close to being significant. However, population density had a significance level
of 0.467. The research on population density is mixed when it comes to increasing NMT. Some
researchers have found that it has a positive effect while others disagree. For example, Newman
(1992) found that in Australian cities when walking and cycling to work is graphed against
population density the result is an S curve.
Newman postulates that the decrease in walking and biking to work between 40 and 90
persons/ha is because there is more access to public transportation. After 90 persons/ha there
appears to be a sharp increase in walking/biking to work. This could be because amenities are
close enough to not need public transportation, or the public transportation is getting overused.

26

This theory was accounted for by including taking public transportation to work in the model,
but possibly there is more to the story.
The 2005-2009 ACS data set presents the smallest geographic area (census tracts) that
includes data related to non-motorized transportation. However, unlike information that planners
typically rely upon in the decennial Census long forms – such as means to work – the ACS uses
a moving average rather than point-in time data. While the creation of the ACS will give
planners a new set of refreshed data every year, there are a number of problems with using the
ACS dataset. For instance, the sampling error associated with the decennial census long form
was much lower than that of the ACS (Blodgett, 2009).
The problems with using ACS data occur when the study area goes through changes.
However, for the current study area, the population has not comparably grown very much since
2000. According to the census, the Lansing tri-county population grew 3.64% between 2000 and
2010, near one-third of the national average of 9.71%.
In general, The ACS is a good resource for socio-economic data in larger study areas,
such as national, states, larger cities, counties, and metro areas; however, the reliability decreases
as the study area gets smaller. Therefore, if this model is going to be adapted to a rapidly
changing region, more accurate survey data is needed.
There is a difference between walking for transport and walking for recreation. This
project aimed to estimate the change in resident’s behavior towards taking NMT as a means for
transportation to work. Although, some researchers have found that the built environment
characteristics that influence transport to work also influence recreational travel (Cerin et al.,
2006; Alfonzo et al., 2008). In contrast, other researchers have found no relationship between the
built environment and recreational walking (Learnihan et al., 2011). Still other researchers found

27

that different built environment attributes were associated with recreational walking versus
transportation walking (Lee &Moudon, 2006). The only survey data available related to NMT in
the Lansing region was American Community Survey means to work. So, these differences were
not able to be explored.
Other researchers have found that people are willing to walk further if the area is highly
walkable (Park, 2008). Therefore, because there was a relatively large amount of people taking
NMT to work pre-development (28.58%) – implying a highly walkable area – the effects of
increasing the walkability of the area may have been felt further out then the half-mile service
area created for this project. This concern could be solved by utilizing better survey data, which
was not at the disposal of this researcher.
Finally, in the process of creating this model other questions have risen. For example, do
the same variables influence the likelihood that people will take NMT from different socioeconomic classes? For example, is the access to close employment the most significant variables
when it comes to predicting if high-income individuals take NMT to work? Conversely, is access
to sidewalks, paths, trails, and bike lanes the best predictor for low- to medium-income
individuals?

6.2 Future Use for the Model
This model was applied to a relatively small proposed redevelop in a dense, mixed-use
downtown area. However, the methods used could also be applied to larger land use changes.
Expand and give other thematic potential applications, nothing too specific in this section. We
are trying to make your last points on your soap box, not open up lots of questions.

28

In addition to quantifying the effects of proposed land use changes, this model would be a
valuable tool for master plans, neighborhood plans, non-motorized transportation plans, and any
other plans that are created to guide the land use decisions of a community. Many community
plans involve aesthetically pleasing long-term goals/strategies for underutilized areas. For
example, the City of East Lansing published an East Village Master Plan in 2006. This 36 acre
mixed-use redevelopment project will replace the current student housing in the area. However,
this project has yet to get off the ground. The Master Plan published has lots of beautiful
renderings of the proposed redevelopment, but there are no quantified benefits within the Plan.
The ability to show how much an area will benefit from reinvestment may go a long way in
gaining momentum for the project.
Transportation models can help foster economic development by assessing the need,
quantity, and benefit of infrastructure improvements. The economic benefits of non-motorized
transportation have already been discussed, and if communities are subscribing to the theory of
placemaking, then the ability to quantify changes will be a great tool to possess. This paper went
into the benefits, methods, and possible uses of such a model. The need for better research data
has been exhausted in this paper and in countless others, but that is the last step in making this
model possible and realizing the benefits.
This project used the only non-motorized transportation survey data available for the Lansing
area. However, the question this project asked and the method itself were unique and they add to
the growing amount of research related to improving non-motorized transportation in America.
Once a model is created distinctive to that community, calculating how a new development
changes the ability for the residents to take NMT will be easy. These simple methods can have a

29

place within a community’s site plan review, or at least as a marketing tool to attract future
residents of a development.

30

APPENDICES

31

APPENDIX A
Table 6 Correlations

Means to
Work
Means to Work
Population Density
PLT Density
Bus Stop Density
Sidewalk Presence
Comm. Property
Connected Node
Ratio
Park / School Density
Residential Density
Effective Walking
Area
Employee Density
Intersection Density

Population PLT
Density
Density

Bus Stop
Density

1
.569**
.569**
.592**
.597**
.313**

.569**
1
.531**
.711**
.720**
.365**

.569**
.531**
1
.417**
.440**
0.81

.592**
.711**
.417**
1
.745**
.531**

.597**
.720**
.440**
.745**
1
.276**

Commercial
Property
Density
.313**
.365**
.081
.531**
.276**
1

.333**

.349**

.077

.458**

.455**

.175

1

.550**
.324**

.434**
.858**

.481**
.318**

.357**
.721**

.491**
.672**

-.016
.510**

.174
.384**

.110

.656**

.249**

.591**

.564**

.353**

.401**

.608**
.224**

.329**
.667**

.287**
.259**

.613**
.661**

.424**
.629**

.583**
.443**

.263**
.447**

** Correlation is significant at the 0.01 level (2-tailed)
* Correlation is significant at the 0.05 level (2-tailed)

32

Sidewalk
Presence

Connected
Node Ratio
.333**
.349**
.077
.458**
.455**
.175

Table 6 (cont’d)
Park / School
Density
Means to Work
Population Density
PLT Density
Bus Stop Density
Sidewalk Presence
Comm. Property
Connected Node
Ratio
Park / School Density
Residential Density
Effective Walking
Area
Employee Density
Intersection Density

Residential
Density

Employee
Density

.324**
.858**
.318**
.721**
.672**
.510**

Effective
Walking
Area
.110
.656**
.249**
.591**
.564**
.353**

.550**
.434**
.481**
.357**
.491**
-.016

.608**
.329**
.287**
.613**
.424**
.583**

.224*
.667**
.259**
.661**
.629**
.443**

.174

.384**

.401**

.263**

.447**

1
.138

.138
1

.071
.812**

.263**
.223*

.186*
.780**

.071

.812**

1

.141

.894**

.263**
.186*

.223*
.780**

.141
.894**

1
.307**

.307**
1

** Correlation is significant at the 0.01 level (2-tailed)
* Correlation is significant at the 0.05 level (2-tailed)

33

Intersection Density

APPENDIX B
Table 7 Model without SES Controls

Regression Statistics
Multiple R
0.838
R Square
0.703
Adjusted R Square
0.681
Standard Error
0.08848
Observations
117
ANOVA

Regression
Residual
Total

df
8
108
116

Coefficients
Intercept
-0.237
Pop_Density
0.009
PLT Density
0.003
Sidewalk Presence
0.056
Commercial_Density
0.179
Connected Node
Ratio
0.297
Parks & Schools
Density
0.200
Intersection Density
-0.799
Employee Density
0.005

SS
2.001
0.846
2.947

MS
0.250
0.008

F
31.951

Standard
Error
0.092
0.004
0.001
0.020
0.113

t Stat
-2.568
2.277
3.759
2.734
1.588

P-value
0.012
0.025
0.000
0.007
0.115

0.116

2.555

0.024

0.091
0.164
0.001

2.195
-4.886
4.586

0.030
0.000
0.000

34

Significance
F
.000

APPENDIX C
Table 8 Full Model
Regression Statistics
Multiple R
0.8628
R Square
0.74442
Adjusted R Square
0.72031
Standard Error
0.08285
Observations
117
ANOVA
df
Regression
Residual
Total

Significance
SS
MS
F
F
10 2.1192561 0.21193 30.8741756 4.94287E-27
106 0.7276021 0.00686
116 2.8468582

Coefficients
Intercept
-0.271
Pop_Density
0.003
PLT Density
0.003
Sidewalk Presence
0.046
Commercial_Density
0.153
Connected Node
Ratio
0.251
Parks & Schools
Density
0.237
Intersection Density
-0.597
Employee Density
0.004
Income
0.000
Gender
0.367

Standard
Error
0.135
0.004
0.001
0.019
0.106

t Stat
-2.003
0.730
4.669
2.341
1.444

P-value
0.048
0.467
0.000
0.021
0.152

0.110

2.294

0.024

0.086
0.161
0.001
0.000
0.211

2.754
-3.703
3.783
-3.659
1.742

0.007
0.000
0.000
0.000
0.084

35

REFERENCES

36

REFERENCES
1000 Friends of Oregon. (1993). Making the Land Use, Transportation, Air Quality Connection.
Retrieved from: http://www.friends.org/resources/reports
Abdul, R. P., Pendyala, R. M., Bhat, C. R., & Waddell, P. A. (2007). Modeling residential sorting
effects to understand the impact of the built environment on commute mode choice.
Transportation, 34(5), 557-573.
About 21c3. (n.d.). Retrieved January 3, 2011, from: http://www.mml.org/resources/21c3/about.html
Administration on Aging. (2010). Projected Future Growth of the Older Population. Retrieved from:
http://www.aoa.gov/AoARoot/Aging_Statistics/future_growth/future_growth.aspx
Alfonzo, M., Boarnet, M. G., Day, K., Mcmillan, T., & Anderson, C. L. (2008). The relationship of
neighbourhood built environment features and adult parents walking. Journal of Urban
Design, 13(1), 29.
Alliance for Biking & Walking (2012). Bicycling and walking in the U.S.: 2012 Benchmarking
Report. Retrieved from:
http://www.peoplepoweredmovement.org/site/index.php/site/memberservices/2012_benchma
rking_report/
Appleyard, B. S. (2010). New methods to measure urban environments for consumer behavior
research: Individual access corridor analysis of environmentally sustainable travel to rapid
transit. University of California, Berkeley.
Belanger, K. M. (2010). An analysis of walking and bicycling behavior in suburban multifamily
housing: A case study in eugene, oregon. University of Oregon.
Berry, T. R., Spence, J. C., Blanchard, C. M., Cutumisu, N., Edwards, J., & Selfridge, G. (2010).A
longitudinal and cross-sectional examination of the relationship between reasons for
choosing a neighbourhood, physical activity and body mass index. International Journal of
Behavioral Nutrition and Physical Activity, 7, 57-57.
Blodgett, J. (2009). The american community survey vs. the decennial census long form. Retrieved
from: http://mcdc2.missouri.edu/pub/data/acs/acsVScensus.shtml
Cerin, E., Leslie, E., & Owen, N. (2009). Explaining socio-economic status differences in walking
for transport: An ecological analysis of individual, social and environmental factors. Social
Science & Medicine, 68(6), 1013.
Cerin, Ester & Leslie, Eva. (2008). How socio-economic status contributes to participation in
leisure-time physical activity." Social Science & Medicine 66 (12), 2596-609.
Chen, C., Gong, H., & Paaswell, R. (2008). Role of the built environment on mode choice decisions:
Additional evidence on the impact of density. Transportation, 35(3), 285-299.

37

City of East Lansing. (2012). City Center II. Retrieved from:
http://www.cityofeastlansing.com/CityCenterII/
Cradock, A. L., Troped, P. J., Fields, B., Melly, S. J., Simms, S. V., Gimmler, F., et al. (2009).
Factors associated with federal transportation funding for local pedestrian and bicycle
programming and facilities. Journal of Public Health Policy, 30, S38-72.
de Sa, E. (2011). Associations between leisure-time and transport-related physical activity with
objective measures of the built environment. York University (Canada)
Dill J. (2004) Measuring network connectivity for bicycling and walking. Transport Research Board
2004 Annual Meeting Transportation Research Board 2004, CD-ROM.
duToit, L., Cerin, E., Leslie, E., &Owen, N. (2007). Does walking in the neighbourhood enhance
local sociability? Urban Studies, 44(9), 1677-1695
Edwards, P. &Tsouros, A. (2006) Promoting physical activity and active living in urban
environments. World Health Organization.
Federal Highway Administration (2012). Report to the U.S. Congress on the Outcomes of the
Nonmotorized Transportation Pilot Program SAFETEA-LU Section 1807. Retrieved from:
http://www.fhwa.dot.gov/environment/bicycle_pedestrian/ntpp/2012_report/final_report_apri
l_2012.pdf
Forsyth, A., Hearst, M., Oakes, J. M., & Schmitz, K. H. (2008). Design and destinations: Factors
influencing walking and total physical activity. Urban Studies, 45(9), 1973-1996.
Frank, L. D., Sallis, J. F., Conway, T. L., Chapman, J. E., Saelens, B. E., & Bachman, W. (2006).
Many pathways from land use to health: Associations between neighborhood walkability and
active transportation, body mass index, and air quality.Journal of the American Planning
Association, 72(1), 75-87.
Frank, L. D., Saelens, B. E., Powell, K. E., & Chapman, J. E. (2007). Stepping towards causation:
Do built environments or neighborhood and travel preferences explain physical activity,
driving, and obesity? Social Science and Medicine, 65(9), 1898-1914.
Frank, L., Bradley, M., Kavage, S., Chapman, J., & Lawton, T. K. (2008). Urban form, travel time,
and cost relationships with tour complexity and mode choice. Transportation, 35(1), 37-54.
Frank, L. D., Sallis, J. F., Saelens, B. E., Leary, L., Cain, K., Conway, T. L., et al. (2010). The
development of a walkability index: Application to the neighborhood quality of life study.
British Journal of Sports Medicine, 44(13), 924.
Garrett-Peltier, Heidi (2011). Pedestrian and bicycle infrastructure: a national study of employment
impacts. Political Economy Research Institute.
Gebel, K., Bauman, A., & Owen, N. (2009). Correlates of non-concordance between perceived and
objective measures of walkability. Annals of Behavioral Medicine, 37(2), 228-38.

38

Handy, S., Cao, X., & Mokhtarian, P. L. (2006).Self-selection in the relationship between the built
environment and walking. American Planning Association. Journal of the American Planning
Association, 72(1), 55-74.
Hardy, M. H. (2011). Simplified integrated transportation and land use modeling to support
metropolitan planning decisions: An application and assessment. George Mason University.
Harkey, D., Reinfurt, D., &Knuiman, M. (1998). Development of the bicycle compatibility index.
Transportation Research Record 1636, Paper No. 98-1073.
Iacono, M., Krizek, K. J., & El-Geneidy, A. (2010). Measuring non-motorized accessibility: Issues,
alternatives, and execution. Journal of Transport Geography, 18(1), 133-140.
Joseph, A. (2006). Where older people walk: Assessing the relationship between physical
environmental factors and walking behavior of older adults. Georgia Institute of Technology.
ProQuest Dissertations and Theses, p. 416.
Killingsworth, R. E., De Nazelle, A., & Bell, R. H. (2003). Building a new paradigm: Improving
public health through transportation. Institute of Transportation Engineers.ITE Journal,
73(6), 28-28.
Kim, Nam Seok. "Trip Generation Model for Pedestrians Based on NHTS 2001." University of
Maryland, College Park, 2005. United States – Maryland
King, D. K. (2006). Individual and neighborhood effects on active lifestyles and social isolation in a
sample of community-dwelling elderly: A socio-ecological study. Dissertation Abstracts
International: Section B: The Sciences and Engineering, p.2820.
Konikov-Titievsky, L. (2010). Neighborhood walkability and body mass index in New York
City. Ph.D. dissertation, Columbia University, United States -- New York.
Larsen, J., El-Geneidy, A., &Yasmin, F. (2010). Beyond the quarter mile: Re-examining travel
distances by active transportation. Canadian Journal of Urban Research, 19(1), 70-88.
Learnihan, V., VAN NIEL, ,KP, Giles-Corti, B., & Knuiman, M. Effect of scale on the links
between walking and urban design. Geographical Research. 49(2), 183-191.
Lee, C., &Moudon, A. V. (2006). Correlates of walking for transportation or recreation purposes.
Journal of Physical Activity & Health, 3, S77-S98.
Leslie, E., Coffee, N., Frank, L., Owen, N., Bauman, A., & Hugo, G. (2007). Walkability of local
communities: Using geographic information systems to objectively assess relevant
environmental attributes. Health and Place, 13(1), 111-122.
Leyden, K. M. (2003). Social capital and the built environment: The importance of walkable
neighborhoods. American Journal of Public Health, 93(9), 1546-51.

39

Li, F., Fisher, K., Brownson, R., Bosworth, M.. (2005). Multilevel modeling of built environment
characteristics related to neighborhood walking activity in older adults. Journal of
Epidemiology and Community Health. 59(7), 558-564.
Litman, T.A. (2009). The economic value of walking. Victoria Transport Policy Institute.
Litman, T.A. (2011). Evaluating non-motorized transportation benefits and costs. Victoria Transport
Policy Institute.
Litman, T.A. (2012). Evaluating public transit benefits and costs. Victoria Transport Policy Institute.
Lovasi, G. S. (2006). Neighborhood walkability, physical activity and cardiovascular risk.University
of Washington.
Lovasi, G., Neckerman, K., Quinn, J., Weiss, C., & Rundle, A.. (2009). Effect of individual or
neighborhood disadvantage on the association between neighborhood walkability and body
mass index. American Journal of Public Health, 99(2), 279-84.
Maghelal, P. K., P.H.D. (2011). Walking to transit: Influence of built environment at varying
distances. Institute of Transportation Engineers.ITE Journal, 81(2), 38-43.
Marshall, J. D., Brauer, M., & Frank, L. D. (2009). Healthy neighborhoods: Walkability and air
pollution.Environmental Health Perspectives, 117(11), 1752-9.
Michigan Avenue Corridor. (2011). Welcome to the Michigan Avenue Corridor! Retrieved from:
http://www.michiganavecorridor.com/
Moore, T., &Thorsnes, P. (2007). The Transportation/Land use connection: 2007 edition. Planning
Advisory Service Report, (546), I-12,14-17,19-49,51-78,80-243,245-283,285-335,337357,359-367,369-0_2.
Mouden, A., Lee, C., Cheadle, A., Garvin, C., Johnson, D., Schmid, T., Weathers, R., Lin, L. (2006).
Operational definitions of walkable neighborhood: theoretical and empirical insights. Journal
of Physical Activity and Health.3 (1). S99-S117.
Moudon, A., Lee, C., Cheadle, A., Collier, C., Johnson, D., Schmid, T., Weather, R. (2005). Cycling
and the built environment, a US perspective. Transportation Research Part D: Transport
and Environment, 10(3), 245-261.
Neiman, A. (2010). Urbanization and health: Examining associations among neighborhood design,
safety and quality of life. Dissertation Abstracts International: Section B: The Sciences and
Engineering, p.6145.
Newman, P. (1992). The compact city: An australian perspective. Built Environment, 18(4), 285300.
O’Sullivan, S. (1996). Walking distances to and from light rail transit stations in calgary. University
of Calgary (Canada)

40

Oliver, L. N., Schuurman, N., & Hall, A. W. (2007). Comparing circular and network buffers to
examine the influence of land use on walking for leisure and errands. International Journal of
Health Geographics, 6(1476-072), 41-41.
Owen, N., Cerin, E., Leslie, E., duToit, L., Coffee, N., Frank, L. D., et al. (2007). Neighborhood
walkability and the walking behavior of australian adults. American Journal of Preventive
Medicine, 33(5), 387-395.
Özbil, A., Peponis, J. and Stone, B. (2008). Modeling street connectivity, pedestrian movement and
land- use according to standard GIS street network representations: a comparative study’,
unpublished paper presented to the 4th Joint Conference of the Association of European
Schools of Planning, Chicago, USA.
Ozbil, A. N. (2010). Walking to the station: The effects of street connectivity on walkability and
access to transit. Georgia Institute of Technology).
Park, S. (2008). Defining, measuring, and evaluating path walkability, and testing its impacts on
transit users mode choice and walking distance to the station. University of California,
Berkeley)
Pivo, G., & Fisher, J. (2011). The walkability premium in commercial real estate investments.
United States, Richmond: American Real Estate and Urban Economic Association.
Plaut, P. O. (2005). Non-motorized commuting in the US. Transportation Research: Part D:
Transport and Environment, 10(5), 347-356.
Porter, C., J. Suhrbier, and W. L. Schwartz. Forecasting Bicycle and Pedestrian Travel: State
of the Practice and Research Needs. In Transportation Research Record: Journal of the
Transportation Research Board, No. 1674, Transportation Research Board of the National
Academies, Washington, D.C., 1999, pp. 94-101.
Rodriguez, D. A., & Joo, J. (2004). The relationship between non-motorized mode choice and the
local physical environment. Transportation Research: Part D: Transport and Environment,
9(2), 151-173.
Rogers, S. H., Halstead, J. M., Gardner, K. H., & Carlson, C. H. (2011). "Examining walkability and
social capital as indicators of quality of life at the municipal and neighborhood scales":
Erratum. Applied Research in Quality of Life, 6(2), 215-216.
Saelens, B., Sallis, J., Frank, L.. (2003). Environmental correlations of walking and cycling: findings
from the transportation, urban design, ad planning literatures. Environment and Physical
Activity. 25 (2) p.80-91
Sallis, J. F., Saelens, B. E., Frank, L. D., Conway, T. L., Slymen, D. J., Cain, K. L., et al. (2009).
Neighborhood built environment and income: Examining multiple health outcomes. Social
Science & Medicine, 68(7), 1285-1293.

41

Seeley, E. (2010). Comparing walkability of ethnically diverse, low-income neighborhoods of
sacramento, california. University of California, Davis.
Soltani, A., Primerano, F., Allan, A., & Somenahalli, S. (2006). Design for movement: Linking nonwork travel and activity level to local urban design dimensions. Urban Design International,
11(3-4), 173-186.
Song, Y., &Quercia, R. G. (2008). How are neighbourhood design features valued across different
neighbourhood types? Journal of Housing and the Built Environment, 23(4), 297-316.
State of Michigan Public Act 135. (2010). Michigan Complete Streets.
Stone, L. A., &Gostin, L. O. (2003). The built environment and its relationship to the publics health:
The legal framework. American Journal of Public Health, 93(9), 1390-4.
Sundquist, K., Eriksson, U., Kawakami, N., Skog, L., Ohlsson, H., &Arvidsson, D. (2011).
Neighborhood walkability, physical activity, and walking behavior: The swedish
neighborhood and physical activity (SNAP) study. Social Science & Medicine, 72(8), 12661273.
Tresidder, M. (2005). Using GIS to measure connectivity: An exploration of issues. Unpublished
Field Area Paper. Portland University: School of Urban Studies and Planning.
U.S. Census Bureau, American Fact Finder. (2010). Means to Work. Retrieved from:
http://factfinder2.census.gov
U.S. Census Bureau, American Fact Finder. (2010). Total Population. Retrieved from:
http://factfinder2.census.gov
VTPI. (2010). TDM Encyclopedia. Retrieved from: http://www.vtpi.org/tdm/tdm25.htm
Walk Score. (2011). Walk Score methodology. Retrieved from:
http://www2.walkscore.com/pdf/WalkScoreMethodology.pdf
Winters, M., Brauer, M., Setton, E., Teschke, K. (2010). Built environment influences on healthy
transportation choices: bicycling versus driving. Journal of Urban Health. 87(6), 969-993
Wisconsin Survey Research Laboratory (1999).Bicycle and Pedestrian Transportation Survey.
Retrieved from:
http://www.dot.wisconsin.gov/library/publications/format/stats/bikepedsurvey.pdf
Witt, C. (2010). The benefits of walking for seniors. Impowerage. Retrieved from:
http://impowerage.com/fitness/activities/the-benefits-of-walking-for-seniors
Wood, L., Frank, L. D., & Giles-Corti, B. (2010). Sense of community and its relationship with
walking and neighborhood design. Social Science & Medicine, 70(9), 1381.

42

Wyckoff, M. (2010). Placemaking, sense of place and place-based initiatives: Key parts of regional
and local economic development strategies. Planning & Zoning News. 29 (1), 6-12.

43