Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter.
Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 18
18
CHAPTER FOUR
DEVELOPING QUESTIONNAIRES
Questionnaires are at the heart of the survey effort, with survey Examples of introductions are:
questions the means by which the desired information is
obtained from respondents. However, perhaps less obviously, · "The Big Blue Bus [Santa Monica, California] needs
questionnaires also convey the purpose of the survey, its impor- your help to provide improved bus service. Please com-
tance, and the attitude of the sponsor toward respondents. plete this survey and return it to the surveyor."
· "The Santa Clara Valley Transportation Authority
In developing on-board and intercept surveys, transit (VTA) wants YOUR help to improve transit services by
agency staff must choose the questions to be asked that will completing this questionnaire and placing it in the
cover the topics identified in the survey planning and return box at the rear exit before leaving the vehicle."
accordingly design the questionnaire. Staff must write the · "Dear Customer: Please take a few minutes to complete
questions and answer choices and the text that introduces the this survey. Your answers will be used in evaluating
survey to riders. [Greater Cleveland] RTA services. When you leave the
vehicle, return the survey to the person collecting them
The wording of introductions, questions, and answer or drop it in any mailbox. No stamp is needed. Thank
choices and the formatting of the questionnaire affect the you for taking time to complete this survey, and enjoy
quality of data collected. A well-written introduction and your ride."
well-designed layout encourages transit users to partici- · "Dear Bus Rider: The Citizens Area Transit (CAT)
pate in the survey, thus minimizing nonresponse error. is conducting a survey to improve bus services in
Well-written questions, appropriate answer choices, and Las Vegas. Please complete this form and drop it in
an easy-to-follow questionnaire design help respondents to the envelope by the bus door. Thank you for your
understand the questions, provide accurate responses (thus cooperation."
minimizing measurement error), and answer all questions · "CTA would like to know more about your travel needs,
(minimizing item nonresponse). in order to serve you better. Please fill out this brief sur-
vey, and return it to the person who gave it to you."
Transit agency practices and relevant findings in the liter- · "Dear Bi-State Rider: Thank you for using Bi-State
ature are discussed for each aspect of questionnaire drafting Transit [St. Louis] services. In order to improve our ser-
and design. vices, we are conducting a short survey of Metrolink
and bus riders. Please take the time to complete this
questionnaire, and when you are done, simply follow
QUESTIONNAIRE INTRODUCTIONS the folding instructions and place it in the mail. We will
pay for postage. Your input will help us to serve you
On-board and intercept transit surveys generally include a better. Thank you for your help. Tom Irwin, Executive
short introduction that explains the purpose and use of the Director."
survey and requests the recipient to complete the ques-
tionnaire. Introductory scripts used in personal interviews In sum, introductions are short and focused on motivating
are similar to written introductions in self-administered response. Information relating to specific questions is placed
questionnaires. with the question to which it applies rather than in the
introduction.
Introductions typically request cooperation, convey that
the survey will help to improve transit service, and provide
instructions on where to return the questionnaire. Stating the TOPICS AND QUESTION WORDING
purpose of the survey and how the survey supports "group
values" (e.g., improving bus and rail service) serves to moti- On-board and intercept transit surveys are conducted for a
vate participation (Dillman 2000). By content and tone, range of purposes and the results are used in a wide variety
introductions also convey a respectful attitude toward of ways within transit agencies. Questions on the survey and
respondents, requesting their cooperation and assuring them response choices naturally need to serve the goals and objec-
that their answers will be taken seriously. tives of the survey project. Therefore, it is useful to begin by
OCR for page 19
19
considering how survey objectives translate into specific terms of gender, age, income, etc., common topics for survey
survey questions. questions. In other cases, considerable thought needs to be
given to how to translate research questions into survey ques-
Figure 1 schematically diagrams the relationships tions. An example is the question of how ridership can be
between agency needs, research questions, and survey ques- increased. As discussed here, in practice a variety of survey
tions. Agency needs relate to the goals and objectives dis- questions are used to address this issue.
cussed earlier. Each agency need results in one or more
research questions. For example, agency marketing efforts This section is organized around the six research ques-
can generate a variety of research questions ranging from tions:
who uses transit to how ridership can be increased.
1. Where and when do customers use transit?
In some cases, translating research questions into survey 2. Who uses transit?
questions is straightforward. Transit users can be profiled in 3. How satisfied are the customers?
Agency needs Research Survey Questions
Questions
Origin and
Where and when
destination
do customers use
Travel modeling
transit?
Trip purpose, fare
payment, other trip
characteristics
Rider
Who uses transit? demographics
Long-range and
areawide
planning
Frequency of
transit use; how
long use transit
How satisfied are
customers?
Route planning and Customer
scheduling satisfaction
Reasons to
use transit
Why do customers
use transit?
Marketing
Vehicles available
to HH; for trip
How can ridership Alternative mode
be increased?
Customer
communications
Needed
improvements
How effective are
agency Awareness, web
use, etc.
communications?
FIGURE 1 Translating agency needs into research questions and survey questions.
OCR for page 20
20
4. Why do customers use transit? TABLE 9
5. How can the agency attract increased ridership? ACTIVITY-BASED ORIGIN AND DESTINATION SURVEY
QUESTIONS FROM TRANSIT PERFORMANCE MONITORING
6. How effective are agency communications? SYSTEM
Activity at origin 1) Where did you come from before you got on
Question wording for each topic is discussed in the follow- this bus/train?
Work/work-related
ing sections. Home
College/other school
Shopping
Medical services
Where and When Do Customers Use Transit Social, religious worship, personal business
Other________________
Many on-board and intercept transit surveys are primarily
designed to obtain detailed information about each respon- Access mode 2) How did you get to this bus/train?
Walked
dent's current trip. Survey results form a profile of where and Drove my car
when customers use transit services. The substantial body of Dropped off by someone
O&D surveys reported by transit agencies reflects the use- Rode my bicycle
Rode a bus/train
fulness of on-board and intercept surveys to capturing travel Rode with someone who parked
behavior information. On-board and intercept surveys are
well suited to this purpose. Because the surveys are con- Activity at 3) Where are you going now?
ducted during customers' actual trips, on-board and intercept destination Work/work-related
Home
surveys are able to cover the entire universe of riders. By
College/other school
inquiring about the current trip, surveys are able to minimize Shopping
errors that arise from recalling past trips (as can occur in tele- Medical services
phone, Internet, or mail surveys). Social, religious worship, personal business
Other________________
The most common travel behavior questions in question- Egress mode 4) When you get off this vehicle, how will you get
naires provided by transit agencies for this study concern to your final destination?
specific aspects of "this trip." Walk
Drive my car
Get picked up by someone
Ride a bus/train
· Origin, Ride my bicycle
· Destination, Ride with someone who parked
· Purpose (work, shopping, return home, etc.), Source: McCollom Management Consulting 2004.
· Access mode (e.g., walk, auto, bus, train),
· Egress mode (e.g., walk, auto, bus, train),
· Duration of access/egress trips, O&D Questions
· Waiting time for bus or train on this trip,
· Other routes used on this trip today, and The most comprehensive O&D surveys obtain four locations
· Method of fare payment. for each trip: origin, boarding, alighting, and destination
(OBAD). Each location is geocoded for further analysis and
modeling. For riders who transfer between routes or modes,
Trip Purpose and Access and Egress Modes the surveys usually ask place of boarding the first bus or train
and where the rider will alight from the final bus or train, as
The wording of trip purpose, access, and egress questions well as the route numbers for each segment of the trip.
was standardized through the Transit Performance Monitor- Although some O&D surveys use the full set of OBAD loca-
ing System (TPMS). TPMS was designed to collect data on tions, others use boarding and alighting locations only.
transit customers through the use of on-board surveys using
standardized questions. The program was funded through a O&D questions are a challenging type of question to for-
cooperative agreement between FTA and APTA. From 1996 mulate and present on the questionnaire owing to the detailed
to 2003, the program collected survey results from approxi- nature of the information and the nonintuitive character of
mately three dozen transit agencies (McCollom Management the concepts employed in the questions; in particular, the
Consulting 2002, 2004). concept of a one-way trip.
Table 9 shows TPMS questions and response categories For modeling purposes, transit planners precisely define
for trip purpose, access, and egress. Trip purposes were what they mean by a trip and each segment of the trip, start-
asked in terms of O&D activities. Trips from home to work, ing from when travelers leave one activity site to when they
for example, were classified as work trips, as are trips from arrive at the next activity site. Transit users are generally not
work to home. familiar with this concept and may not think in these terms.
OCR for page 21
21
Thus, respondents may not understand the distinctions 3. "What are the nearest cross streets or intersection of
between origin and boarding, or between alighting and des- that place?" Boxes are provided for two street names.
tination. For some trips, of course, the origin and boarding 4. "What is the exact street address of that place?" Boxes
locations are the same and it may seem redundant to the are provided for street number, street name, city, and
respondent to be asked about both locations. In other cases, zip code.
respondents view the trip to begin at the boarding location; 5. "Where did you board this bus?" Boxes are provided
so again, the origin appears to be redundant even though, by for cross streets and city.
the intended definition, it is not. 6. "What time did you board this bus?"
7. "How did you get from the beginning location to the
Customers also may not understand how to report trips bus stop where you boarded this bus?" Response cate-
involving a transfer or stops along the way (linked trips, in gories are provided for "Walk," "Shuttle/vanpool,"
the planner's parlance). Sacramento Regional Transit "Rode with someone who parked," etc.
District reports that customers transferring from light rail fre-
quently report the transfer point as the end of the trip. Comparable information is asked for alighting location,
destination location, and activity. Surveys for TriMet, CTA,
Respondents may not even correctly identify the start and and the Southern California Regional Rail Authority
end of the trip; some respondents provide round-trip (known as Metrolink) follow similar question sequences
information--what appears to be a home-to-home trip, for (see Appendix C).
example--rather than a one-way home-to-shopping trip.
The concept behind this sequence of questions is that rid-
Another difficulty is that respondents simply may not ers will most readily be able to report their trip by starting
know the address or intersection information. Transit staff with their last activity and continuing sequentially. The sur-
report that riders who do not drive, in particular, may not vey communicates the idea of the start of the trip by asking,
know the street names at intersections, and therefore can- simply, "Where were you before riding this bus?" The survey
not report the intersection where they boarded the bus. communicates the definition of a one-way trip without using
Even when they know the exact address, riders may be the word "trip" and without needing to explain what is meant
reluctant to report the specific address, particularly for their by a one-way trip.
place of residence.
Most O&D questionnaires provide several ways to report
Even when precisely reported, location information can each location: as a landmark, cross streets, or address (with
pose data processing challenges, as when street names repeat city and zip requested to avoid problems with duplicate street
in multiple cities within an agency's service area (i.e., two or names). In the experience of transit agency staff, providing
more "Main Streets"). all three ways to present location information produces more
usable locations than providing only one or two ways.
These problems are reported uniformly across O&D sur-
veys. Although survey design, instructions, use of examples, The formatting of answer spaces for this information
and other measures may reduce the incidence of problems varies. Some surveys provide separate spaces for landmark,
with the data, no agency reported having completely sur- intersection, and address, as shown in the TARC survey in
mounted these challenges. Figure 2. In practice, respondents typically complete only
portions of the landmark/intersection/address section--hope-
Transit agencies seek to overcome these problems by fully reporting enough information for geocoding purposes.
asking riders to narrate their trip, usually from beginning to
end. This approach takes advantage of the linear nature of Other surveys consolidate spaces for all three types of
transit trips. location information. For example, a TriMet survey provided
blanks for the street name, nearest cross street, city, and zip
A typical sequence of OBAD questions is shown in code. (The full survey is in Appendix C.) TriMet staff
Appendix C from a survey conducted for TARC. (Note that believes that a simpler format is less intimidating to riders
the survey was printed on 17 in. by 11 in. paper so that the than the more extensive formats they used in earlier O&D
two pages appear opposite each other in the version used in surveys.
fieldwork.) The question sequence is:
Even more simply, a SMART survey asked for the corner
1. "Where were you before riding this bus (beginning of two streets; instructions were under the lines specifying,
location)?" Answer choices are "My home," "My "Address, street name, or landmark" and "Street name,"
work," "Visiting friend or relative," etc. respectively (Figure 2).
2. "What is the name of this place/building where you
began this trip?" Boxes are provided to fill in the name Some agencies have experienced difficulty with the
of a place; an example of a bank is provided. "street 1" and "street 2" terminology. CTA reports that these
OCR for page 22
22
FIGURE 2 TARC (top) and SMART origin questions.
words elicited O&D streets from some respondents in its Overall, it appears that avoiding the use of trip and one-
West Side survey (see Appendix D). The CTA Douglas Line way trip terminology is more effective than attempting to
survey asked for intersection, address, and landmark in one define or explain the meaning of these words.
space, as shown in Appendix C. CTA staff believes that this
approach is easier on field staff (because the questionnaire is Multisegment trips involving transfers also create confu-
simpler for respondents), although more labor intensive for sion for respondents. Agencies marked by high transfer rates
coding staff. In the view of CTA staff, a more complex ques- have used various approaches to improve the transfer infor-
tionnaire would need more highly trained field workers. mation provided to them. For example, a Lane Transit Dis-
trict survey first asked whether the rider was using one bus
Regardless of these wording and formatting choices, or more than one bus on this trip, and if the latter, to specify
transit agencies uniformly report at least some difficulties the route number(s) (see Appendix C). Lane Transit District
with obtaining accurate and logical descriptions of riders' staff believes that overall this approach worked well,
trips. Some agencies have experimented with reordering the although there was some confusion about multisegment trips.
questions. The CTA West Side Customer Travel Survey
begins by asking, "What bus route are you riding now?" Surveys sometimes include examples of answers. The
That question is followed by questions on alighting loca- experience appears to be that providing example answers (as
tion, transfers, and origin activity (home, work, school, in the TARC survey) is effective, whereas examples of the
etc.) and origin location. The concept is to begin with a very meaning of instructions can be counterproductive. The CTA
well-defined question (current bus route) and work back- West Side survey that illustrated the meaning of one-way
wards to boarding and origin locations. CTA staff believes trips with a home to work example was taken to mean by
that there is merit to this approach, although respondents some respondents that only home to work trips should be
continue to have difficulty with the "Where are you coming reported (see Appendix D).
from?" language.
Fare Payment
Metrolink also asked for the boarding station as the first
question for a survey of commuter rail riders, but found little An important aspect of each trip is how the fare is paid.
difference in data quality from sequences that started with the Agencies typically list fare payment options and ask respon-
"where came from" question. dents to check the one that applies to this trip. The list of
OCR for page 23
23
payment options can be quite extensive for agencies that trips per day and not simply the number of days transit is
have various passes and different rates for seniors, persons used. Capturing trips per day is most relevant for transit
with disabilities, and students. One approach is to present the systems that are often used for more than one round trip per
options in a matrix format, as illustrated on the TriMet ques- day. However, some surveys have found that an inordinate
tionnaire in Appendix C. Although there is reason to avoid number of riders report using transit for five trips per week.
matrix formats (as will be discussed later), the format Many of those responding in this way are presumably mean-
appears to be workable for fare payment. ing 5 round-trips per week, or 10 one-way trips. The days per
week wording sidesteps the problem of respondents not
understanding the concept of trips.
Who Uses Transit?
For questions about the duration of transit ridership, ques-
On-board and intercept transit surveys are widely used to
tion wording may or may not specify a threshold frequency of
profile characteristics of transit users. Transit agencies
use. Thus, Pace Suburban Bus asks simply, "When did you
reported that information on customer characteristics is
begin riding Pace?" TPMS wording specifies "How long have
highly valuable for planning and marketing purposes and of
you been a regular transit rider" and defines regular as at least
keen interest to upper management. Rider profiles help pro-
once each week. CTTransit (Hartford, Connecticut) defines
vide a picture of who is using bus and rail services, without
regular as at least once each month. Defining frequency works
which it is difficult to think concretely about how to provide
to ensure a uniform interpretation of the question.
or market the services.
The number of vehicles available to a household shows
Profile information can be categorized as travel behavior
whether the rider may have an automobile available for the
(going beyond describing the current trip) and demographic
current trip. Table 10 shows wording used by several transit
information.
agencies and in the 2001 National Household Travel Survey
and 2000 U.S. Census Bureau long-form questionnaire.
General Travel Questions Transit surveys often use trip-specific questions in addition
to, or instead of, the household vehicle question, as discussed
General travel questions used on questionnaires provided by later in this chapter.
transit agencies include:
A standardized TPMS question, widely adopted, concerns
· Frequency of using transit, the alternative mode for the current trip. If transit service
· How long the respondent has been using transit, were not available, would the respondent make the trip by
· Vehicles available to the household, and car, walking, riding with a friend, taxi, bicycle, or not make
· Alternative modes. the trip at all? The number that would not make the trip
shows the degree to which transit provides basic mobility for
Frequency of transit use shows the degree to which riders riders. The number that would use an automobile documents
are regular versus occasional users. Particularly when ana- transit's role in reducing traffic congestion, although the
lyzed by trip purpose and time of day of travel, the frequency "ride with a friend" response does not indicate whether the
question can show what portion of riders' overall travel is friend is already making the trip.
served by transit, and suggest areas for potentially attracting
current riders to use transit more often.
Demographic Questions
Duration of transit use is valuable information to show the
turnover rate among riders. Some agencies find relatively Demographic questions can provide insight into travel
high turnover among riders. TPMS found that nearly one- behavior and customer attitudes. Demographic data can also
half of riders in small and large systems have been using tran- be useful in assessing which markets transit is tapping and
sit for less than one year. However, the results are not possible untapped or underdeveloped markets.
uniform: a survey in Denver found that only 14% of RTD bus
riders were new riders (less than one year). It would be Demographic questions used on questionnaires provided
important to investigate the factors causing rapid turnover, by transit agencies cover the following topics:
where present, which could well lead to prescriptions to
retaining riders and thus growing overall ridership. · Gender,
· Age,
Wording of the frequency and turnover questions varies, · Race and ethnicity,
as shown in Table 10. Frequency of transit use can be asked · Have driver's license,
on a days per week (or month) basis, or trips within a defined · Household income,
time period (usually a week or month). The latter approach · Household size,
has the advantage of reflecting differences in the number of · Employment status,
OCR for page 24
24
TABLE 10
SURVEY QUESTIONS RELATED TO WHO USES TRANSIT
Frequency TPMS National Household Travel Survey (NHTS) Pace (suburban Chicago) LYNX (Orlando)
of transit How often do you use In the past two months, about how often have
How many days per week do How often do you ride the
use transit? you used public transportation such as buses,
you ride Pace? bus?
7 days a week subways, streetcars, or commuter trains?
Less than 1 day 57 days a week
6 days a week Two or more days a week [11+ times]
1 day 24 days a week
5 days a week About once a week [510 times]
2 days Once a week
4 days a week Once or twice a month [24 times]
3 days Once or twice a month
3 days a week Less than once a month [one time]
4 days Less than once a month
2 days a week Never
5 days
1 day a week
67 days
Twice a month
Once a month
First time riding
TriMet (Portland, OR) RTD (Denver)
How many trips have you How many one-way trips did you take on any
taken on a TriMet RTD bus last week? (A round trip counts as two
bus/MAX/streetcar in the trips.) Please do not include trips you may have
last month? (Count each taken on the 16th Street Mall Shuttle or to events
direction as one trip.) like the Rockies games. (Put "0" if none.)
0 or 1 One-way bus trips last week__________
2 to 6
7 to 12
13 to 29
30 or more
How long a TPMS Pace (suburban Chicago) CTTransit (Hartford, CT)
transit How long have you been
When did you begin How long have you
user a regular transit rider--
riding Pace? regularly--at least once a
at least once a week?
Less than 1 year ago month--been riding
Less than a month transit?
1 to 2 years ago
16 months
2 to 3 years ago Less than a month
712 months
3 to 4 years ago 16 months
12 years
4 to 5 years ago 712 months
34 years
5 to 7 years ago 12 years
More than 4 years
7 to 10 years ago 24 years
More than 10 years ago More than 4 years
OCR for page 25
TABLE 10 (Continued)
Vehicles NHTS 2000 Census TARC (Louisville) Sun Tran (Tucson) LYNX (Orlando)
available How many vehicles are How many automobiles, How many vehicles are in How many working How many cars or trucks are
to owned, leased, or vans, and trucks of one- running condition and vehicles (autos, in your household?
household available for regular use ton capacity or less are available for use by your trucks, motorcycles) ___
by the people who kept at home for use by household? are available in the
currently live in your members of your None household where you
household? Please be sure household? live or where you stay
1
to include motorcycles, ____ in the Tucson area?
2
mopeds, and RVs. 0
3
___ 1
4+
2
3
4 or more
Choice of TPMS TARC (Louisville)
mode If transit service were not
If bus service was not
were available how would you
available, how would you
current make this kind of trip?
make this trip?
mode not Use a car
Drive
available Walk
Ride with someone
Ride with a friend
Taxi
Use a taxi
Walk
Bicycle
Bicycle
I would not make this
trip Would not make this trip
25
OCR for page 26
26
· Home ZIP code, and without making distinctions between litter on the floor, over-
· Internet access. flow of trash cans, or brightness of the lighting.
Question wording is relatively straightforward for these It is possible to develop a concise attribute list that captures
questions, and often standardized. Table 11 shows TPMS the different aspects of the user experience and thus reduces
wording for demographic questions included in the TPMS survey length. Studies that use qualitative methodologies
project, and wording used by agencies for selected questions. such as focus groups or advanced statistical methodologies
such as factor analysis and structural equations have found
that the user experience can be distilled into 7 to 10 service
How Satisfied Are Customers? attributes (Strategic Marketing & Research 1997; Stuart I.
Brown Associates 1997; Stuart et al. 2000; Weinstein 2000;
Customer satisfaction and attitudinal sections of on-board
Miller et al. 2002; Zhou et al. 2004) Although the wording of
and intercept surveys address the basic question, "How are
the attributes varies between transit systems, they generally
we doing?" These questions reveal riders' level of satisfac-
cover the following areas:
tion and experience with bus and rail services. However, the
questions do not necessarily show why riders use transit or
· Timely service (frequency of service, predictability of
where agencies should focus their resources.
bus or train arrivals),
· Speed of service,
Questionnaires provided by transit agencies typically ask
· Cleanliness (on board and in station),
for a rating of overall service and ratings for various attri-
· Safety/police presence,
butes. The surveys use a variety of scales for these ratings.
· Comfort,
An alternative approach is to query riders' direct experience
· Driver courtesy and friendliness,
with transit rather than rating service attributes.
· Crowding,
· Cost/value, and
Attribute Ratings · Information availability and ease of use.
The most common approach to measuring customer satis- Interestingly, the studies identified crowding as an issue
faction is to ask respondents to rate overall service and rate a among larger agencies but not smaller agencies, whereas
series of attributes. The overall rating may be obtained as one driver courtesy and friendliness was highlighted only by the
item on the list of attribute ratings or as a separate question. smaller agencies.
The number of service attributes presented--and the level LYNX and TriMet surveys illustrate attribute lists that
of detail--varies widely. In questionnaires provided by tran- reflect these basic aspects of service (Table 12).
sit agencies, the number of attributes ranged from 5 (Sun
Tran, Tucson, Arizona) to 24 (CATS). Studies of customer
satisfaction in transit have included as many as 48 attributes Ratings Scales
(Morpace International 1999).
Ratings of attributes may employ either a verbal or numeric
Extensive attribute lists can include very specific aspects scale. Commonly used verbal scales include 5-point scales
of service, such as posted signs on bus stops (CATS) and ranging from "Very satisfied" to "Very dissatisfied" or from
cleanliness of train interior (Metrolink) (Table 12). These "Strongly agree" to "Strongly disagree." These scales have
attributes are of direct relevance and interest to operational the advantage that each point on the scale is clearly positive
divisions of the respective agencies and thus provide specific or negative, with a neutral point in the middle.
and timely feedback to operating personnel.
Another common verbal scale uses the short, easily under-
Lengthy attribute lists increase the length of the survey. stood words, "Excellent," "Good," "Fair," and "Poor"; how-
Whether the amount of information collected is greater than ever, whether "Fair" is a positive or negative rating can be
for a shorter list is open to question, given that ratings for open to interpretation.
attributes touching on similar aspects of service are often
highly correlated. The high level of correlation could reflect Numeric scales are also commonly used. Such scales
a similar level of performance across different attributes-- typically range from 1 to 5, 1 to 7, or 1 to 10. Numeric scales
for example, agencies that keep the trash emptied may keep are easily fit on the page. Scores can be averaged and the aver-
stations and railcars clean as well. On the other hand, strong age can be easily tracked over time. On the other hand, scores
correlations between attributes may reflect a limit on how can be difficult to interpret; is a "5" a satisfactory score?
many different aspects of the transit user experience riders
perceive. In a subway station environment, for example, Agencies sometimes combine numeric and verbal scales.
users may perceive the station to be "clean and well lit" For example, questionnaires may use a 1 to 5 scale but
OCR for page 27
27
TABLE 11
DEMOGRAPHIC SURVEY QUESTIONS
Gender TPMS
I am...
Male
Female
Age TPMS
My age is:
Under 15
15 to 18
19 to 24
25 to 34
35 to 49
50 to 64
65 or more
Income TPMS RTD (Denver) Sun Tran (Tucson) TARC (Louisville)
What is your total
Which one of the What do you What was your
household income?
following categories estimate was the estimated total
Under $20,000 best describes the combined total household income (in
$20,000$29,999 total annual income, annual income 2003) before taxes?
$30,000$39,999 before taxes, for all (before taxes) in 2003 <$20,000
$40,000$49,999 persons in your for everyone who
$20,000$39,999
$50,000$59,999 household? lives in that
$40,000$59,999
$60,000$79,999 Under $15,000 household?
$60,000$79,999
$80,000 or greater $15,000$24,999 Below $5,000
$80,000$99,999
$25,000$34,999 $5,000$9,999
$100,000+
$35,000$49,999 $10,000$19,999
$50,000$74,999 $20,000$29,999
$75,000$99,999 $30,000$39,999
$100,000 or more $40,000$49,999
$50,000$59,999
$60,000$74,999
$75,000 or more
Household TPMS NHTS Sun Tran (Tucson)
size Including yourself,
Including yourself, how many people Including yourself, how many
how many people live
live in your household? Please do not people live in the household
in your household ?
include anyone who usually lives where you live or where you
____
somewhere else or is just visiting, stay in the Tucson area?
such as a college student away at 1
school.
2
___ 3
4 or more
Internet NHTS CTTransit TARC (Louisville) Pace (suburban
access During the last 6 (Hartford, CT) Do you have access to Chicago)
months, did you have Do you have the Internet at any of Do you have access to
access to the Internet Internet access? the following locations? the Internet at home?
or world-wide web? Yes (Mark all that apply.) Yes
Yes No At home No
No At work at work?
At school
Yes
At the local library
No
I don't have access to
the Internet
Other places (specify)
OCR for page 28
28
TABLE 12
SURVEY QUESTIONS RELATED TO CUSTOMER SATISFACTION
TriMet (Portland, OR) Sun Tran (Tucson)
Please read the following statements and Do you agree or disagree with the following statements?
answer using the 5-point rating scale. Transit services operate on time.
Cleanliness inside bus I feel safe when riding the bus.
Safety while on-board Drivers are helpful and friendly.
Reliability of service Route/schedule information is easy to use.
Frequency of service Buses are clean and well-maintained.
Overcrowding Scale: Strongly agree, Agree, No opinion, Disagree, Strongly
Overall service disagree
Scale: 1 (poor) to 5 (excellent)
LYNX (Orlando) Pace (suburban Chicago)
Evaluate LYNX services: Please indicate your level of satisfaction with Pace service on a
Routing scale from 1 to 5, where 5 is Very Satisfied and 1 is Very
Dissatisfied.
On time
Safety Overall satisfaction with Pace
Cleanliness How driver obeys and enforces rules
Operator courtesy Accuracy of route information
Fare Availability of route information
Frequency Driver courtesy
Hours of operation Posted signs at bus stop
Overall service Personal safety on bus
Scale: Excellent, Good, Fair, Poor, Very poor, Ease of fare payment
Don't know Cleanliness inside buses
Total travel time
Service connections
Personal safety at bus stops
Responses of telephone representatives
Service when and where desired
Reliability of equipment
Buses running on time
Drivers safe driving
Value of service for fare paid
Availability of bus shelters
Notification of service changes
Scale: 1 (very satisfied) to 5 (very dissatisfied)
Metrolink (Los Angeles area) CATS (Charlotte, NC)
Please rate each feature associated with traveling Rate how well CATS performs in each area.
on Metrolink trains. Buses are on time
Travel time on Metrolink vs. car Bus passes are sold at convenient locations
On-time arrivals Travel time on the bus is reasonable
Connecting transit buses at station Buses are clean inside
Availability of free unreserved parking at station The Transit Center is clean and well maintained
Availability of paid and reserved parking at station It's easy to get bus information on the phone
Availability of seating on the train The transit system serves all parts of the city
Cleanliness of train interior Buses begin running early enough in the morning
Safe operation of trains Buses continue to run late enough at night
Personal security on the train There is frequent bus service on weekdays
Personal security at the station There is frequent bus service on Saturdays
Trains free of defects (heat, doors, etc.) There is frequent bus service on Sundays
Ease of purchasing tickets Bus fares are reasonable
Courtesy of Metrolink conductors You feel safe from crime at the Charlotte Transit Center
Cost of a Metrolink ride You feel safe from crime on the bus
OCR for page 29
29
TABLE 12 (Continued)
Communication of schedule changes Taking the bus is relaxing
Communication of delays The bus ride is comfortable
Schedule convenience Bus drivers are courteous
Ease of getting information at 800-371-LINK Bus drivers are knowledgeable
Scale: 1 (dissatisfied) to 5 (satisfied). Also asked for Bus drivers are safe drivers
importance on 1 to 5 scale Bus drivers are sensitive to the needs of passengers
Bus drivers greet you
Complaints/suggestions get a quick response
The system provides a valuable service to the community
Scale: 1 (very poor) to 5 (excellent). Also asked for
importance on 1 to 5 scale
anchor the "1" and "5" as "Very satisfied" and "Very dissat- trip are viewed as "choice" riders, who are using transit
isfied," respectively. because of the quality of service or other factors. Individuals
who do not own a car or do not have a car available for this
Simple "Yes" and "No" scales may be also be used where trip are classified as "transit dependent" or "captive." They
appropriate, as when asking whether the respondent would are assumed to use transit for lack of an alternative.
recommend transit to others.
Whether riders that do not have an automobile available
See Appendixes C and D for examples of ratings scales in are truly captive is open to question. National surveys show
sample questionnaires. that people living in zero-vehicle households still make far
more trips by private automobile (34%) than by transit (19%)
Direct Experience with Transit (Pucher and Renne 2003). Survey results from 18 transit
agencies, reported through the TPMS program, show that
Another approach to relating the customer experience to one-quarter of riders surveyed would "ride with someone
agency performance is to ask riders specific questions about else" if transit were not available for the trip, whereas 20%
their user experiences, either for the current trip or overall would walk or bike and 11% would take a cab (McCollom
experience. A good example is presented in a CTTransit on- Management Consulting 2004). Thus, even riders who, based
board survey. The survey asked respondents whether buses on lack of automobile ownership appear to be captive appear
arrive within 5 min of the scheduled time "Always," "Some- to have several means of transportation available to them.
times," "Most times," or "Not very often." Similar questions
pertained to the courteousness of bus operators, bus cleanli- Some surveys take a more nuanced approach to the auto-
ness, availability of timetables and notices, and helpfulness mobile availability question. TriMet, for example, has four
of telephone center representatives. (See Appendix D for the categories (results from 2000 survey are in parentheses):
full questionnaire.)
· I do have a car but prefer to use TriMet (43%).
Similarly, a GCRTA survey asked riders to "grade RTA's · I don't have a car because I prefer to use TriMet (14%).
services for this trip." Attributes included GCRTA arriving · I don't have a car available for me to use (28%).
at the stop as scheduled, GCRTA driver being courteous, · I don't drive or don't know how to drive (15%).
GCRTA seating comfort, and if the GCRTA shelter is clean.
These categories acknowledge that the availability and
Why Do Customers Use Transit? quality of transit service may encourage some riders to
forego buying a car (or a second car) even though they could
A key objective of many on-board and intercept surveys is to afford to do so. In a sense, they choose to be captive riders.
understand why riders use transit. What are the key drivers
that prompt members of the traveling public to choose the Other on-board and intercept transit surveys look not only
bus, subway, light rail, or commuter rail over alternatives at automobile availability but also to other factors that affect
that range from automobiles to walking? mode choice. Table 13 shows examples from several surveys
that ask why respondents are using transit for this trip, or why
Questionnaires provided by transit agencies show a wide they use transit generally. Variation in the wording of answer
variety of approaches to surveying on this topic. One choices reflects different local conditions, but answer
approach is to focus primarily on the availability of an auto- choices can be summarized as:
mobile as the primary alternative mode, with some surveys
simply asking whether the respondent had an automobile · Do not drive,
available for this trip. People who have a car available for the · No car available (or allows someone else to use the car),
OCR for page 30
30
TABLE 13
SURVEY QUESTIONS RELATED TO WHY CUSTOMERS USE TRANSIT
Reasons VTA (San Jose, CA) Pace (Chicago area) Bi-State Development Agency
for using (St. Louis)
What is your main reason for using Why did you use Pace
transit What is the main reason you
transit in Santa Clara County? today instead of another
use Bi-State transit services?
Have no other way way of traveling?
Cheaper than driving
Better use of time Don't drive
Faster than driving
Costs less No car available
I don't drive
Faster than driving Reduces pollution
No car available
Allows someone else to use the car Prefer transit
Traffic is too bad
Fewer problems than using car Can read or relax
Parking is too expensive
Car in shop Unavailable parking
More relaxing than a car
Good for the environment No license
Friends use transit
Other (specify) Cheaper than driving
Employer helps pay fare
Expensive parking
Better for the environment
Other
Other (specify)
TriMet (Portland, OR) Big Blue Bus (Santa Monica, CA)
What are the most important reasons you
What is the major reason you are using the bus for
are riding the bus today?
this one-way trip? (Check one best answer.)
I do not own a car
I do have a car but prefer to use TriMet. Someone else has the car
I don't have a car because I prefer to use TriMet. The bus is faster than walking
I don't have a car available for me to use. Parking is expensive
I don't drive or don't know how to drive. I do not like to drive in traffic
The bus is faster than driving
To help the environment
I do not have a driver's license
My employer pays for bus fare
Metrolink (Los Angeles)
Please think about when you first started riding Metrolink. What was the single most important
reason that made you take Metrolink? (Choose one.)
I had moved to a new residence and needed a new way to commute
I got a new job or job location and needed a new way to commute
My employer gave me a free Metrolink ticket
My employer paid for part of my Metrolink pass
A family member, friend, or co-worker told me about Metrolink
I have seen advertising for Metrolink and was curious
The MTA strike forced me to find alternate transportation
I took the train to a special event
I was traveling in a group
My car was being repaired
My car was being used by another family member
I could not drive my car for medical reasons
The high cost of gas
I served on jury duty and received a free Metrolink ticket
I received an offer in the mail
Other (please specify)
Availability TPMS SANDAG DART VTA (San Jose, CA)
of auto Do you have a car or
Did you have a car Do you have a car Was an auto available to
for this other personal vehicle
trip that you could have available to you you for this trip?
that you could have
used today instead to make this trip? Yes
used to make this
of the bus? Yes
trip? Yes, but with
Yes No inconvenience to others
Yes
No No
No
OCR for page 31
31
· Faster than driving, sit for most work trips; for example, to avoid parking costs or
· Avoid driving in traffic, avoid the stress of traffic, but take the automobile on days
· Better use of time (e.g., read and relax), when they will come home later in the evening. In this exam-
· Cost of driving and/or parking, and ple, the reasons for using transit (parking and traffic) are dif-
· Environmental considerations (reduces pollution). ferent from actions that would attract increased usage, which
might be more frequent evening service. As another exam-
This list is consistent with transit research findings that mode ple, some riders use transit for work but not shopping trips,
choice is largely driven by the cost and availability of park- owing to buses not conveniently serving trips to the mall.
ing, travel times (especially out of vehicle time), price, com-
fort and convenience (Charles River Associates 1997; Transit questionnaires tend to take a straightforward
Dueker et al. 1998; Schaller 1999; Miller et al. 2002). approach to assessing ways to increase ridership. Question-
naires provided by transit agencies ask what the agency
"could do to improve bus service" (Broward County Tran-
How Can an Agency Attract Increased Ridership? sit), what "would motivate you the most to continue riding or
ride more often?" (Pace Suburban Bus), or use similar word-
Closely allied to the issue of why people use transit is the ing. The surveys then provided a list of possible service
question of how to attract additional ridership. Although improvements. Respondents are instructed to check the one
related, the two issues are quite distinct. Some riders use tran- to three most important service improvements (Table 14).
TABLE 14
SURVEY QUESTIONS RELATED TO IMPROVING TRANSIT AND ATTRACTING INCREASED RIDERSHIP
Needed Broward County (FL) Transit RTD (Denver)
improvements Please tell us three things that we could What is the single most important area, if any, in
do to improve bus service. Please check which RTD should make improvements to its' bus
only up to three: service?
More bus routes Cost
Fewer transfers Comfort
Park & ride lots Convenience
More information Customer information
More frequent bus service Travel time
Express buses Park-n-ride
More evening and weekend service Bus driver performance
More comfortable buses Telephone information center (TIC)
Better on-time performance Security/safety
Bus stop shelters/benches None
Other (specify)
Pace (suburban Chicago) LYNX (Orlando)
What is the ONE item listed below that would What bus service improvements
motivate you the MOST to continue riding or to ride are most needed? (Check top three.)
more often? Frequency
Reduce fares New routes
Provide stops closer to my home Night and weekend service
Run buses more often Shelters
Change the schedule Pre-paid fare cards
Reduce travel time Free transfers
Run express service more often Additional transfer locations
Serve more destinations Other___________________
Improve on-time performance
Improve safety while driving
Improve safety while waiting
Make transit information more accessible
Provide a more convenient connection to final
destination
Run service from free parking lots to busy Metra stations
Other (please specify)
Nothing--will not consider riding or riding more often
Does not apply--I ride as often as I can
