TRANSIT MARKET RESEARCH USING ATTITUDINAL MARKET
SEGMENTATION, STRUCTURAL EQUATION MODELING AND MODE CHOICE
MODELING
Yoram Shiftan, Transportation Research Institute, Technion
Maren L. Outwater, P.E., Cambridge Systematics, Inc
Yu Shuang Zhou, Cambridge Systematics Inc.
1. INTRODUCTION
This paper presents a new approach to identify potential transit markets and develop strategies to
increase public transport ridership. The approach uses structural equation modeling (SEM) to
simultaneously identify the attitudes of travel behaviors and the causal relationships between
traveler’s socioeconomic profile, traveler attitudes and travel behavior. Travel attitudes are used
to identify distinct market segments in order to best serve the needs of each segment and to
develop plans to increase transit ridership.
Contrary to the most sophisticated market research used in the private sector, the market
segments traditionally used in transportation planning are most often based on socioeconomic
characteristics such as income, gender, or automobile ownership. To better understand the
reasons that different travelers have for choosing their mode for everyday travel, our approach
breaks away from these stereotypes and instead determines the attitudes that drive each
segment’s mode choice. These attitudes often cut across social and economic groupings.
The approach developed in this paper is based on a household survey including three types of
questions: socio-economic, travel behavior and attitudinal questions. These data measure a
person’s sensitivity to a broad range of travel experience that they may encounter during their
trip and characteristics of the different travel modes that they may consider when choosing to
travel. Using exploratory and confirmatory factor analysis, we group the various attitudinal
questions into a series of traveler factors.
We then use Structural Equation Modeling (SEM) to simultaneously estimate statistical models
between the travel behavior, demographic and socio-economic data, and the attitudinal data
collected in the survey. SEM enables to test a set of linear models to identify the structural
attitudes of travel behaviors and to quantify the causal relationships between travelers’
socioeconomic status or demographic profile and their attitudes. This approach provides a
highly detailed understanding of the characteristics, attitudes, and preferences of potential
customers. From the SEM, we can determine the characteristics of transit service to which each
market segment will be most attracted. We then perform exploratory and confirmatory cluster
analysis using the attitudinal factor scores (from the SEM) to group the entire sample of
households and riders into market segments that share the same set of attitudes towards their
travel experience.
© Association for European Transport 2004
We demonstrate this approach using two case studies developed by Cambridge Systematics.
The first case study was developed for the Utah Transit Authority to assist them in identifying
their potential markets and the best strategies to increase transit ridership. The second case study
was developed for the San Francisco Bay Area Water Transit Authority to assist them in
evaluating expanded ferry service.
In both cases, using the developed SEM models and the socio-economic data available from the
US Census at a detailed geographical level we obtained an understanding of the concentration of
segments throughout the study area. This spatial dimension separates public transit from the
design and marketing of almost every other service or product. Public transit must not only
match the preferences of its target customers (i.e., market segments), it must do so for each
Origin-Destination pair (i.e., travel markets). The preferences of each segment must be linked to
this geographical dimension in order to deploy the most competitive transit services in the
appropriate travel market. The UTA study had a special segment indicating likelihood to use
transit. In the WTA case, the market segments were used to estimate stated-preference mode
choice models for 14 alternative modes, which separated the traveler’s reaction to time savings
and other attributes by attitudinal market segment and recognized that modal choices are
different for market segments that differ in their attitude and therefore have different sensitivity
to travel stress or desire to help the environment.
2. IDENTIFYING TRAVELER ATTITUDES USING FACTOR ANALYSIS
Both household surveys included three main common parts. The first part asked respondents
about their travel habits for both work and non-work purposes. The last part included some
socio-demographic questions. The second and main part of the survey inquired about
various attitudinal variables (38 in the UTA case and 30 in the WTA case) that were used to
measure respondents’ sensitivity to a broad range of experiences that they may encounter during
their trip, as well as characteristics of the different travel modes that they may consider for their
travel. The WTA survey included also a stated preference part that was used to estimate mode
choice models. The various attitudinal statements used are shown in Table 1 for the UTA study
and in Table 2 for the WTA study. Each survey respondent was asked to provide his or her rating
level of agreement or disagreement with each statement, with a scale of 0 to 10, where 0 means
strong disagreement, 10 means strong agreement, and 5 means neutral with the statement.
A factor analysis step is performed to analyze the various attitudinal variables collected from the
household survey. Factor analysis involves a statistical procedure that transforms a number of
possibly correlated variables into a smaller group of uncorrelated variables called principal
components or factors. The objectives of performing factor analysis are to reduce the number of
variables (data reduction) and to classify variables, so as to detect the structural relationships
between variables (structure detection).
The following are two types of factor analysis we conducted for this market analysis:
Exploratory factor analysis (EFA). A process where the underlying data determine the
structure and content of the resulting factors. It is used to explore the survey data to determine
the nature of factors that account for the co-variation between variables without imposing any a
priori hypothesis about the number and structure of factors underlying the data.
© Association for European Transport 2004
Confirmatory factor analysis (CFA). A process where we apply judgment regarding the
structure and content of the factors and then estimate the statistical results of these established
factors. It is supported by the results from exploratory factor analysis, as well as theoretical
hypotheses as to which variables are correlated with which factors.
We started with exploratory factor analysis (EFA). The main purpose of the EFA is to guide us
in constructing the confirmatory factor analysis (CFA). Three sets of factors, using six, seven
and eight factors, were tested in the CFA to determine the right group of factors that explain all
the attitudinal questions. Based on this analysis it was decided to use a set of seven factors in the
UTA study (shown in Table 1) and six factors in the WTA study (shown in Table 2) that were
significant and all the loadings had the correct sign. However, in the UTA study there was one
variable (Q33 – anxious) that did not fit into any of the seven factors, and was left out of the
analysis.
CFA uses analysis of Goodness of Fit Index (GFI) to analyze whether the data confirms the
theoretical construct. The Goodness of Fit Index is a measure of the relative amount of variances
and co-variances jointly accounted for by the model. It can be thought of as roughly analogous
to the R2 in multivariate regression. The closer the GFI is to 1.00; the better is the fit of the
model to the data. The GFI for the UTA study is 0.864, indicating that 86.4 percent of the covariation in the data can be reproduced by the given model. The GFI for the WTA study is 0.896.
Thus, in both cases the confirmatory factor analysis supports the six and seven-dimension
attitudinal construct.
3. IDENTIFY TRAVELER ATTITUDES USING STRUCTURAL EQUATION
MODELING (SEM)
Attitude models provide a means to link traveler attitudes to existing socioeconomic and
demographic data that are readily available through the 2000 U.S. Census data and Public Use
Microdata Sample (PUMS) data. This makes it possible to relate the traveler attitudinal factors
that were used to create the market segments to the socioeconomic data in the Census and to
identify the spatial distribution of the segments in the population.
There were 522 survey records in the general household UTA survey and 852 in the WTA one,
with various demographic and socioeconomic data including age, gender, education level,
marriage and job status, household size and number of kids, number of workers, auto ownership
and household income. The demographic and socioeconomic data from the household survey
were recoded into categories that match the categories of the 2000 Census data and PUMS data.
Therefore, the structural equation model estimated from the household survey can be applied to
the region’s overall population using the same sets of socioeconomic variables.
SEM is a modeling technique that enables us to test a set of linear equations simultaneously. It is
used in this study to identify the structural relationship between attitudinal statements, and to
quantify the causal relationships between travelers’ socioeconomic status and demographic
profile and travel attitudes. The primary objective of this structural equation modeling is to
improve the statistical reliability of the relationship between the socioeconomic data and the
attitudinal factors.
© Association for European Transport 2004
The following are two types of variables used in the SEM:
Manifest variables are observed variables that are directly measured from household surveys.
In this study, there are two main groups of manifest variables: 1) the attitudinal variables, which
are the ratings that travelers indicated regarding their attitude toward various travel statements;
and 2) socioeconomic and demographic variables, such as household size, household income,
and vehicle ownership.
Latent variables are unobserved variable that are not directly measured, but are inferred by the
relationships or correlations among manifest variables in the analysis. There are also two groups
of latent variables in the SEM: 1) the attitudinal factors representing the most important
attitudinal dimensions for traveler behavior, and 2) error terms associated with each variable
involved in the SEM model.
Conceptually, every variable has an associated measurement error. So the SEM model includes
an error term for each variable.
The SEM is constructed in AMOS 4.0. AMOS uses path diagrams to represent relationships
among manifest and latent variables. Ovals or circles represent latent variables, while rectangles
or squares represent manifest variables (see Figure 1 as an example for the UTA case). Singleheaded arrows in the path diagram represent causal effects.
In the SEM structure, people’s socioeconomic and demographic characteristics are regarded
as exogenous variables, while the ratings of attitudinal statements are endogenous variables.
A SEM is used to capture the causal influences of the exogenous variables on the
endogenous variables through sets of underlying attitudinal factors. For this purpose, three
basic sets of simultaneous equations were estimated concurrently in the SEM:
•
Functions between attitudinal factors and socioeconomic, demographic variables;
•
Functions between ratings of the attitudinal statements and underlying attitudinal factors;
and
•
Functions between latent variables.
All the linear equations in SEM are estimated simultaneously. The results of the SEM process
are a final set of traveler factors that are estimated by two sets of equations simultaneously. One
set of equations relates the attitudinal factors to the socioeconomic/demographic data and the
other set of equations relates them to the ranking of the attitudinal statements. Using the
estimated SEM attitudinal factors scores can be calculated using the estimated coefficients for
functions between attitudinal factors and socioeconomic variables. The factor scores are then
used in market segmentation as inputs to the cluster analysis.
3.1 Functions between Attitudinal Factors and Attitudinal Variables
Each latent factor is associated with multiple attitudinal statements through a CFA structure.
The SEM is a confirmatory rather than exploratory modeling method, because the analyst
predetermines the model structure. If the analyst assumes no direct relationship between an
attitudinal factor and an attitudinal statement, the path coefficient in the diagram is set to zero.
© Association for European Transport 2004
For each attitudinal factor, there is one and only one path coefficient that is fixed to be one.
This is the anchor variable that is used to set the scale of measurement for the latent factor and
residuals. The SEM model estimates all other path coefficients. Table 1 and 2 present the
estimation results of the path coefficients for all the attitudinal variables for each attitudinal
factor in the SEM process for the UTA and WTA cases accordingly. Statistics on standard error
(Std Error), significance (t-value) and critical ratio (C.R.) for each variable are also presented in
these tables. Critical ratio is the estimate divided by its standard error. Assuming the survey is a
random sample with standard normal distributions, estimates with critical ratio more than 1.96
are significant at the 95 percent confidence level.
3.2 Functions between Attitudinal Factors and Socioeconomic Variables
The attitudinal latent variables are specified as linear equations of observed socioeconomic and
demographic variables, which act as the indicators of the underlying attitudinal structure toward
travel. The socioeconomic and demographic variables include respondents’ age, gender, marital
status, education level, employment status, household income, number of kids under age 18,
number of workers in the household, and number of vehicles in the household. Table 3 and 4
show the estimation results for the parameters of socioeconomic variables for one of the factor
that is common to both studies: Desire to help the environment. Table 3 shows the results for the
UTA case and Table 4 shows them for the WTA case.
3.3 Functions between Attitudinal Factors
The causal influences of latent variables upon one another are also represented as linear
equations in the SEM. There are three pairs of causal relationships being modeled in the UTA
model: 1) sensitivity to time as a function of desire for productivity and reliability, 2) sensitivity
to time as a function of the need for fixed schedules, and 3) preference for transit as a function of
sensitivity to safety and privacy. In the WTA model there are four pairs of causal relationships
between factors being modeled: 1) need for flexibility as a function of need for time savings, 2)
sensitivity to travel stress as a function of need for flexibility, 3) insensitivity to transport cost as
a function of need for time savings, and 4) desire to help the environment as a function of
sensitivity to personal travel experience.
4. MARKET SEGMENTATION MODELS
The core concept of market segmentation is to view a market as several segments rather than one
homogeneous group. Each market segment is unique in its characteristics and attitudes toward
travel behavior. Market segmentation results can provide the basis for strategic marketing plans
that involve applying different marketing strategies to different market segments. The objective
of cluster analysis is to identify unique travel groups for market profiling. It is useful to the
extent that people within the same cluster share similar attitudes toward travel behavior, while
people in different clusters hold different views.
© Association for European Transport 2004
The following are two types of cluster analysis that we conducted for this market analysis:
Exploratory cluster analysis. A process where the statistics of the data determine the structure
and content of the resulting clusters. It is used to explore the survey data to determine the nature
of clusters that account for the co-variation between factors without imposing any a priori
hypothesis about the number and structure of clusters underlying the data.
Confirmatory cluster analysis. A process where we apply judgment regarding the structure
and content of the clusters, and then statistically estimate them. It is based on the results from
exploratory cluster analysis, as well as theoretical hypotheses as to which factors are correlated
with which clusters.
Confirmatory cluster analysis is built upon the results from the exploratory cluster analysis. It
uses the same tools as exploratory cluster analysis except that the factors used to define the
various clusters and the number of clusters is predetermined. In our study, three of the eight
factors were used to assign the household sample into eight clusters or segments in each of the
studies. In the UTA study we used the factors: Sensitivity to time, Need for fixed schedules, and
Willingness to use transit. In the WTA study we used the factors: Desire to help the
Environment, Need for time saving, and Sensitivity to travel Stress. We used these factors,
because they have the highest statistical reliability based on the structural equation modeling
discussed in the previous section. The resulting structure of the confirmatory cluster analysis is
presented in Figure 2 for the UTA study and in Figure 3 for the WTA study. As can be seen
from figure 2, the sensitivity of time was used to divide the sample into two groups: one for
travelers who have a high sensitivity of time and one for those who have low sensitivity of time.
Next, each of these two groups was divided into two subgroups based on their need for fixed
schedule. Finally, the willingness to use transit factor was used to further divide the subgroups
into eight market segments. Each market segment is identified with a descriptive name that
invokes the primary drivers behind the traveler attitudes in that segment.
A similar process was conducted for the WTA study as shown in Figure 3. Although three
factors were used to create the eight market segments, each of the eight market segments is best
understood using the information (or the factor scores) of all eight factors. In addition, the
structural equation modeling provides us with significant demographic/socioeconomic variables
that can be used to further describe each segment including household size, number of kids and
workers in the households, auto ownership, employment status, education level, gender, marital
status, age and income.
These variables give us a vital link to the census data. With this link, we are able to determine
the concentration of each market segment in each census block group. Thus, not only do we
know the attitudinal preference of each market segment, but also where each market segment is
distributed geographically. We strongly caution the reader, however, that these socioeconomic
variables cannot be used individually or even two or three at a time to predict which market
segment an individual belongs in. Their strength in predicting segment membership comes from
using all ten variables in conjunction. These socioeconomic data are available from the census,
and are used to segment the population for each block group.
MARKET SEGMENTATION RESULTS
© Association for European Transport 2004
This section presents the main characteristics of the various market segments in each study.
One means of evaluating the specific traveler attitudes present in each of the market segments
was to calculate average factor scores for the original factors that are present within each
segment. These were compared to the overall mean total factor scores to identify whether each
market segment was higher or lower than the overall average. The following conclusions can be
made regarding these analyses.
For the UTA Study:
Anxious Ambers, Productive 9 to 5-ers and Cautious Flyers have low desire to help the
environment and Cautious 9 to 5-ers have the lowest desire to help the environment. Green
Riders, Routine Riders and Routine Flyers have high desire to improve air quality. Green Flyers
have the highest desire to help the environment.
Anxious Ambers, Productive 9 to 5-ers and Cautious 9 to 5-ers have low willingness to use
transit and Cautious Flyers have the lowest willingness to use transit. Routine Riders, Green
Flyers and Routine Flyers have high willingness to use transit and Green Riders have the highest
willingness to use transit
Anxious Ambers, Green Riders and Green Flyers have low desire for productivity and reliability.
Productive 9 to 5-ers, Routine Riders, Cautious Flyers and Routine Flyers have high desire for
productivity and reliability with Cautious 9 to 5-ers have the highest desire for productivity and
reliability.
Anxious Ambers and Green Riders have low sensitivity to time and flexible schedule.
Productive 9 to 5-ers and Routine Riders have low sensitivity to time and fix schedule. Cautious
Flyers and Green Flyers have high sensitivity to time and flexible schedule. Cautious 9 to 5-ers
have high sensitivity to time and fix schedule with Routine Flyers have the highest sensitivity to
time and fix schedule.
Anxious Ambers, Productive 9 to 5-ers and Cautious Flyers have high sensitive to safety and
privacy with Cautious 9 to 5-ers having the highest sensitivity to safety and privacy. Routine
Riders, Green Flyers and Routine Flyers have low sensitivity to safety and privacy, while Green
Riders is the least sensitive segment to safety and privacy
Anxious Ambers is the most sensitive segment to stress and comfort, Green Riders and Routine
Riders have high sensitive to stress and comfort. Productive 9 to 5-ers, Cautious Flyers, Green
Flyers, Cautious 9 to 5-ers and Routine Flyers have low sensitivity to stress and comfort
For the WTA Study:
Anxious Amblers, Calm Chargers, and Frazzled Flyers are the most sensitive to personal travel
experience, while Green Cruisers and Reserved Recyclers are the least sensitive;
Tense Trekkers and Reserved Recyclers are the most sensitive to cost (note that this factor
identifies insensitivity to cost and is, therefore, reversed in concept from the other factors),
while Joe Six Pack and Calm Chargers are the least sensitive to cost;
Tense Trekkers are more sensitive to stress than all other categories, while Joe Six Packs are the
least sensitive to stress;
© Association for European Transport 2004
Tense Trekkers and Relaxed Runabouts have the highest need for flexibility, while Green
Cruisers have the least need for flexibility;
Frazzled Flyers have the highest need for time savings, while Joe Six Packs have the least need
for time savings; and
Reserved Recyclers have the highest desire to help the environment, while Joe Six Packs have
the least desire to help the environment.
APPLICATION OF MARKET SEGMENTS TO UTA SERVICE AREA
The survey-based market segmentation model was then applied to the whole population in the
study area. Zonal-level socioeconomic and demographic data were used to calculate the score of
each attitudinal factor using the estimated parameters from Structural Equation Model. The
resulting scores of were then used to divide the population into eight segments.
The information on market segments can be very useful in designing public transport service.
For example, market segments with a high need for time savings and a high need for flexibility
(such as Relaxed Runabouts and Tense Trekkers in the WTA study) are more difficult to serve
with fixed-route transit systems. But market segments with a desire to help the environment and
sensitivity to stress (such as Reserved Recyclers in the WTA study) are more likely to be well
served by public transport service. The UTA study directly identified markets with high
willingness to use transit.
MODE CHOICE MODELS
For the WTA study, the household survey data was combined with preferences and attitudes of
travelers to estimate stated preference choice models. Multinomial logit models were estimated
for three travel purpose and the estimation results are presented in Table 5. These models use
the market segments developed above to better model mode choice by those segments and
understand the differences in mode choice behavior among these markets. 14 alternatives were
specified including two auto modes (drive alone and carpool), six bus/rail modes differentiated
based on access/egress modes, and six ferry modes differentiated based on access/egress
modes. All the models presented above are multinomial logit models. Various nesting
structures were also tested but did not improve the likelihood of the models and the nesting
coefficients (logsum values) were found to be not significantly different than one indicating that
the alternatives are not forming a significant nest. As shown in Table 5, market segment-related
LOS and submode-specific constants are estimated to better understand the implications of
various market segments on their mode choice behavior. Only one LOS variable, total travel
time, is estimated for market segments that are sensitive to travel time. The in-vehicle travel
time coefficient for these market segments is the sum of this coefficient and the in-vehicle travel
times of the specific mode. It is found that the market segments that are more sensitive to time
have a larger and more negative coefficient than the other market segment coefficients.
The sensitivity to travel costs is exactly the same across all the market segments in every model,
because no market segment-specific cost coefficients were estimated that were significant and
logical. As expected, the values of time for time sensitive market segments are higher than that
of other market segments. It is also found that, these market segments are slightly more sensitive
to time when executing shopping/other trips than when commuting to work.
© Association for European Transport 2004
Additional constants were estimated for various market segments to understand the influence of
various factors like travel stress and environmental friendliness towards mode choice behavior.
Overall, it was found that stress sensitive travelers are prone to prefer auto modes to transit
modes for making non-work trips. Environmental friendly commuters seem to be inclined to
ride transit modes for recreational trips. This constant was not significant and did not have the
correct sign in work and shopping/other trips models. These effects are displayed in Figure 4.
CONCLUSIONS
Transit must compete for market share in much the same way as other products and services
compete for customers. The first step most companies take in refining their strategy involves
understanding their market place according to the key attitudes their potential customers most
value when making their decision which product or service to buy. In this study we
demonstrated the use of Structural Equation approach as a powerful tool to improve our
understanding of travel behavior and to improve transit services. We described the various
factors that best describe the marketplace for local travel within the UTA and WTA service
areas. We used these factors to group all potential customers into eight market segments.
This approach can significantly increase our ability to answer important questions for better
transit planning such as:
•
What attitudes and preferences drive each market segment’s choice for local travel
options?
•
What strategies would be most effective for each market segment?
•
Where are the “easy-to-reach” (and “hard-to-reach”) markets?
•
What strategies are most likely to be effective at different locations?
The information on the various key attitudes and eight market segments is critical to designing
transit services that meet the needs of target market segments. This information is also very
useful in showing stakeholders and policy makers, which market segments, will be too
demanding to provide cost-effective transit services. For example, market segments with a high
value of time and a high need for safety and privacy (such as Cautious 9 to 5-ers in the UTA
study) are more difficult to serve with fixed-route transit systems. But market segments with a
low value of time and low need to privacy (Green Riders in the UTA study) are more likely
served by modest improvements to existing transit services. The WTA study also shows how
attitude and identification of such market segments can improve mode choice models.
© Association for European Transport 2004
Table 1. Attitudinal Factors and Attitudinal Variables in SEM, UTA Case
Variable Statements
Factor 1: Desire to help improve air quality
I would be willing to pay more when I travel if it would help improve air quality
People who drive alone should pay more to help improve air quality
I would switch to a different form of transportation if it would reduce air pollution
Use of public transportation can help improve air quality
Factor 2: Desire for Productivity and Reliability
I would like to make productive use of my time when I travel
I would much rather do something else with the time that I spend traveling
I prefer a travel option that has predictable travel time from day to day
If my travel option is delayed, I want to know the cause and length of the delay
When traveling, I like to keep as close as possible to my departure and arrival
schedules
Factor 3: Sensitivity to Time
I am usually in a hurry when I make a trip
I would change my form of travel if it would save me some time
I use the fastest form of transportation regardless of cost
Driving is usually the fastest way to get where I need to go
Factor 4: Sensitivity to Safety and Privacy
I don’t mind traveling with strangers
I feel safe using public transportation
I feel safe walking both near my home and near my destination
When traveling, I like to talk and visit with other people
I worry about getting in an accident when I travel
I avoid traveling through certain areas because they are unsafe
I prefer driving because I like to be alone while I travel
Having my privacy is important to me when I travel
Factor 5: Need for Fixed Schedules
I need to travel mostly during rush hour times
I need to make trips according to a fixed schedule
Factor 6: Sensitivity to Stress and Comfort
Having a stress-free trip is more important than reaching my destination quickly
I avoid traveling at certain times because it is too stressful
I don’t mind delays as long as I am comfortable
It is important to have comfortable seats when I travel
A clean vehicle is important to me
Factor 7: Willingness to Use Transit
I wouldn’t mind walking a few minutes to get to and from a bus or a TRAX stop
I would ride transit if services were available to my destination when I need to
travel
If I rode public transportation I wouldn’t mind changing between buses or between
bus and TRAX
I know how to reach my destination using public transportation
I would use public transportation more often if it was cheaper to ride
© Association for European Transport 2004
Coeff
icient
Std
Error
C.R.
1
0.89
0.86
0.28
0.10
0.09
0.06
8.88
9.28
4.93
1
0.82
0.73
0.68
0.68
0.13
0.11
0.11
0.11
6.26
6.36
6.08
6.43
1
0.93
0.85
0.52
0.14
0.14
0.08
6.48
6.33
6.18
1
0.85
0.55
0.26
-0.21
-0.51
-0.56
-0.61
0.08
0.07
0.07
0.07
0.09
0.08
0.08
10.95
7.74
3.64
-2.89
-5.76
-7.23
-7.98
1
0.89
0.13
7.02
1
0.73
0.64
0.26
0.18
0.12
0.10
0.07
0.06
6.18
6.36
3.84
3.20
1
0.88
0.09
9.67
0.78
0.09
8.72
0.64
0.36
0.10
0.08
6.58
4.34
Table 2.
Factor/Variable
Attitudinal Factors and Variables in SEM, WTA Case
Variable Statements
Coefficient
Std
Error
t-value
Factor One
Desire to help the environment
PAYENVIR
I would be willing to pay more when I
travel if it would help the environment.
1.000
MODENVIR
I would switch to a different form of
transportation if it would help the
environment.
0.949
0.028
33.447
TRNENVIR
Use of transit can help improve the
environment.
0.376
0.018
20.887
Factor Two
Need for timesavings
CHANGMOD
I would change my form of travel if it would
save me some time.
1.000
HURRY
I am usually in a hurry when I make a trip.
0.911
0.023
39.283
FASTEST
I always take the fastest route to my
destination even if I have a cheaper
alternative.
0.760
0.024
32.044
NOSTRESS
Having a stress-free trip is more important
than reaching my destination quickly.
-0.680
0.030
-22.978
CROWDSOK
I’ll put up with crowds if it means I’ll get to
my destination quickly.
0.657
0.020
32.082
COMFORT
I don’t mind delays as long as I am
comfortable.
-0.511
0.021
-23.848
DLDRIVE
I don’t like to drive, but it is usually the
fastest way to get where I need to go.
0.418
0.025
16.791
© Association for European Transport 2004
Table 2. Attitudinal Factors and Variables in SEM, WTA Case (continued)
Factor/Variable
Variable Statements
Coefficient
Std
Error
t-value
Factor Three
Need for flexibility
VARIETY
I need to make trips to a wide variety of
locations each week.
1.000
NEEDFLEX
I need to have the flexibility to make many
trips during the day if necessary.
0.841
0.031
27.555
REGULAR
Generally, I make the same types of trips at
the same times of the day.
-0.489
0.023
-21.654
Factor Four
Sensitivity to Travel Stress
ANXIOUS
I am usually anxious and unsettled by the
time I reach my destination.
1.000
NOSTRESS
Having a stress-free trip is more important
than reaching my destination quickly.
1.106
0.080
13.883
BRIDGES
Driving on the bridges across the bay is
stressful for me.
1.266
0.074
17.099
STRESSFL
I avoid making certain trips at certain times,
because it is too stressful to make the trip.
1.519
0.080
19.059
Factor Five
Insensitivity to transport cost
CONVENNT
I use the most convenient form of
transportation regardless of cost.
1.000
PAYENVIR
I would be willing to pay more when I
travel if it would help the environment.
0.248
0.032
7.797
FASTEST
I always take the fastest route to my
destination even if I have a cheaper
alternative.
0.680
0.070
9.716
Factor Six
PRFDRIVE
Sensitivity to personal travel experience
I would prefer to drive than to be driven.
TRANCOMF
The people who ride transit to work are like
me.
-1.181
0.081
-14.608
WALKING
I am comfortable walking near my
destination during the day.
-0.678
0.055
-12.289
FERNOBUS
I would ride a ferry, but I wouldn’t ride the
bus.
1.273
0.092
13.887
PRFALONE
I prefer to make trips alone, because I like
the time to myself.
0.861
0.072
12.025
DLDRIVE
I don’t like to drive, but it is usually the
fastest way to get where I need to go.
-0.352
0.065
-5.413
© Association for European Transport 2004
1.000
Table 3.
Structural Equation for the Desire to Help Improve Air
Quality, UTA
Stratification
Estimate
S.E.
C.R.
P-Value
Age 18-24
0.256
0.299
0.857
0.392
Age 25-34
0.098
0.241
0.407
0.684
Factor 1: Desire to help improve air quality
Age
Age 35-64
0.091
0.196
0.467
0.641
Education
Low education
-0.159
0.2
-0.792
0.428
Auto Ownership
0 Auto
-0.398
0.797
-0.499
0.618
1 auto
-0.065
0.249
-0.262
0.793
1-person household
0.273
0.297
0.92
0.357
2-persons household
-0.007
0.209
-0.032
0.974
Income < 35K
0.817
0.222
3.672
0
Income 35-50K
0.291
0.212
1.37
0.171
Income 50-100K
0.416
0.211
1.97
0.049
Number of Kids
0 kid
0.252
0.197
1.278
0.201
Gender
Male
0.011
0.196
0.055
0.956
Marital Status
Married
-0.08
0.211
-0.379
0.705
Employment Status
Employed
0.155
0.216
0.721
0.471
Not in labor force
-0.108
0.715
-0.151
0.88
0-worker household
0.451
0.275
1.636
0.102
1-worker household
0.287
0.216
1.333
0.183
Household Size
Household income
Level
Household Workers
Number
© Association for European Transport 2004
Table 4
Structural Equation for the Desire to Help the
Environment Factor, WTA Case
Name
Estimate Std Error
C.R.
P
3.400
0.001
0.118
4.939
0.000
0.093
12.269
0.000
0.935
0.093
10.093
0.000
AGE5564
0.664
0.115
5.749
0.000
65-74
AGE6574
0.782
0.189
4.138
0.000
College student
None
CSTUDENT
0.457
0.306
1.492
0.136
Household size
1 Person
HHSIZE_1
-0.358
0.402
-0.892
0.372
2 Persons
HHSIZE_2
0.319
0.088
3.631
0.000
3 Persons
HHSIZE_3
0.011
0.104
0.110
0.912
Households with
children (<18 years
old)
0 Kid
HHSU18_1
0.020
0.112
0.175
0.861
1 Kid
HHSU18_2
0.411
0.106
3.868
0.000
2 Kids
HHSU18_3
0.334
0.229
1.463
0.143
Household income
$25-50,000
INC2550K
0.601
0.153
3.937
0.000
<$25,000
INC25K
0.838
0.374
2.239
0.025
$50-75,000
INC5075K
0.375
0.114
3.281
0.001
Vehicles per
household
0 vehicle
VEHS0
-1.669
0.528
-3.162
0.002
1 vehicle
VEHS1
0.191
0.115
1.667
0.096
Workers per
household
0 worker
WORK0
-0.568
0.186
-3.055
0.002
1 worker
WORK1
-0.269
0.095
-2.830
0.005
Households with
more workers than
vehicles
None
WORKCARS
-0.214
0.143
-1.499
0.134
Sensitivity to
personal travel
experience factor
None
f6_mode
-0.658
0.059
-11.151
0.000
Variable*
Stratification
Age
18-24
AGE1824
0.838
0.247
25-34
AGE2534
0.583
35-44
AGE3544
1.145
45-54
AGE4554
55-64
© Association for European Transport 2004
Table 5
Mode Choice Model Estimation Results – SP
HBW - SP
HBSh/Oth - SP
t-stat
Coeff
HBRec - SP
Constants
Modes
Coeff
t-stat
Carpool
Auto
-0.2085101
-0.6
-0.3319733
-0.76
-1.447214
Coeff
t-stat
-2.98
BART
Transit
1.861703
3.37
1.3173711
1.91
2.174823
2.26
Other rail
Transit
0.5900163
0.99
0.6385499
0.88
1.514266
1.47
Bus
Transit
0.9602115
1.66
-0.1090708
-0.15
1.275635
1.31
Ferry
Transit
0.0184942
0.03
0.235847
0.29
-0.652326
-0.62
-1.6825113
-4.21
-1.4432361
-2.76
0.014167
0.02
-0.683682
-3.65
-0.2864258
-1.23
-0.548317
-1.64
-4.58
Drive access
Auto
Transit Access/Egress
Transit
Level of service
Submodes/market segments
Total cost
Rail/bus
-0.0038383
-5.4
-0.0026914
-2.96
-0.006327
Ferry
-0.0031572
-3.52
-0.0012804
-1.14
-0.002563
-1.7
Auto
-0.0012912
-4.95
-0.0006963
-2.42
-0.001495
-3.94
Auto
-0.0367257
-8.49
-0.0247593
-4.04
-0.04603
-5.52
Rail/bus
-0.0233347
-5.95
-0.0156354
-2.98
-0.038986
-5.08
Ferry
-0.0241803
-3.59
-0.0173758
-1.96
-0.029138
-2.75
Walk time
Transit
-0.0297759
-6.75
-0.0229923
-3.82
-0.027511
-4.22
Transit access/egress time
Transit
-0.0602804
-4.91
-0.0480998
-3.15
-0.062081
-3.04
Drive access time
Transit
-0.01077
-0.62
-0.0689959
-2.83
-0.061132
-1.95
Out-of-vehicle time
Auto
-0.0431927
-2.1
-0.0386627
-1.47
-0.025037
-0.95
Total travel time
Time-sensitive market segments*
-0.00776316
-2.14
-0.0094045
-2.12
-0.005607
-1.02
In-vehicle time
Socioeconomic data
Submodes
Household income
Drive alone
4.46E-06
1.69
-9.29E-07
-0.3
-6.90E-06
-2.24
Rail/bus drive access
7.06E-06
2.3
7.10E-06
1.68
-1.69E-05
-3.59
Ferry drive access
1.47E-05
3.86
-2.13E-06
-0.45
-1.19E-05
-2.05
-2.15E-07
-0.06
-5.27E-07
-0.12
-2.39E-05
-4.3
7.40E-06
1.83
-1.16E-05
-2.17
-1.49E-05
-2.57
0.85
Rail/bus walk/transit access
Ferry walk/transit access
Vehicles per household
Drive alone
0.1358595
1.7
0.41993736
3.69
0.070405
Rail/bus walk/transit access
-0.6047203
-4.05
-0.3465766
-1.76
0.017849
0.09
Ferry walk/transit access
-0.4965095
-3.01
-0.1281851
-0.58
-0.083009
-0.39
Rail/bus drive access
Ferry drive access
0.0257747
0.27
0.30569162
2.23
0.009961
0.06
-0.2805015
-2.17
0.41639055
2.86
-0.263244
-1.12
0.573667
Additional constants
Market segment
Auto modes
Stress-related market segments**
-0.00307112
-0.02
1.06684314
4.81
Ferry modes
Stress-related market segments**
0.12487234
0.54
0.75732604
2.45
Carpool, transit and ferry
modes
Summary statistics
Pro-environmental market
segments***
Log-likelihood at convergence
---
---
---
---
--0.720028
-1754.4966
-1115.5829
-780.3198
Rho-squared with respect to zero
0.3278
0.3926
0.4575
Rho-squared with respect to constants
0.1187
0.0928
0.1112
Auto out-of-vehicle time/in-vehicle time
1.18
1.56
0.54
Bus/rail out-of-vehicle time/in-vehicle time
1.28
1.47
0.71
Ferry out-of-vehicle time/in-vehicle time
1.23
1.32
0.94
$17.07
$21.34
$18.47
Bus/rail – value of time
$3.65
$3.49
$3.70
Ferry – value of time
$4.60
$8.14
$6.82
Other statistics
Auto – value of time
*
Time-sensitive market segments are Calm Charger, Frazzled Flyer, Relaxed Runabout, and Tense Trekker.
© Association for European Transport 2004
2.41
--3.37
**
Stress-related market segments are Anxious Ambler, Frazzled Flyer, Reserved Recycler, and Tense Trekker.
***
Pro-Environmental Market Segments are Green Cruiser, Reserved Recycler, Relaxed Runabout, and Tense Trekker.
© Association for European Transport 2004
© Association for European Transport 2004
Figure 1
SEM Model Structure
Attitudinal
Statements
(Endogenous)
Attitudinal Factors
(Latent)
Privacy &
Safety
Productivity &
Reliabiity
Willingness to
Use Transit
Privacy &
Safety
Fixed
Schedule
Stress
& Comfort
Sensitivity
to Time
© Association for European Transport 2004
Socioeconomic
Status
(Exogenous)
Figure 2
Market Segmentation for UTA
All Travelers
Factor 3:
Sensitivity
to Time
Factor 5:
Need for
Fixed
Schedule
Low Sensitivity ofTime
Flexible
Schedule
Factor 7:
No Transit
Willingness
to Use
Transit
Anxious
Market
Amblers
Segment
Attitudinal
Focus
None of
These
Factors
High Sensitivityof Time
Fixed
Schedule
Flexible
Schedule
Fixed
Schedule
Transit
No Transit
Transit
No Transit
Transit
No Transit
Transit
Green
Riders
Productive
9 to 5-ers
Routine
Riders
Cautious
Flyers
Green
Flyers
Cautious
9 to 5-ers
Routine
Flyers
Transit
Fixed Fixed Schedule Time
Schedule
and
Transit
© Association for European Transport 2004
Time
Time
Time, Fixed
and Transit and Fixed Scheduleand
Schedule
Transit
Figure 3 Market Segmentation for WTA
All Trans-Bay
Trippers
Factor
One
Factor
Two
Factor
Four
Market
Segment
Focus
Modest Environmental
Less Time
Savings
Strong Environmental
More Time
Savings
Less Time
Savings
More Time
Savings
Less
Stressed
More
Stressed
Less
Stressed
More
Stressed
Less
Stressed
More
Stressed
Less
Stressed
More
Stressed
Joe
Six-Pack
Anxious
Ambler
Calm
Charger
Frazzled
Flyer
Green
Cruiser
Reserved
Recycler
Relaxed
Runabout
Tense
Trekker
None of
These
Factors
Stress
Time
Time &
Stress
© Association for European Transport 2004
Environment Environment Environment Environment
& Stress
& Time
Time &
Stress