Characteristics of Premium Transit Services that Affect Choice of Mode (2014)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

D-1 A p p e n d i x d Transit Service Attribute Models Contents D-1 Overview D-2 Model Results of Unscaled Attributes D-15 Model Results of Scaled Attributes D-24 Details of the Transit Attribute Models D-33 Summary of Key Findings on Transit Attributes Overview The objective of the current research effort is to evaluate the relative importance of a selected set of transit service attributes (see TABLE D-3) and to quantify their effect in the context of mode choice decision. To achieve this, we employ MaxDiff analysis on preference data collected in the surveys. The estimation of utility function in this analysis is typically performed using multinomial logit (MNL) estimation. Other algorithms that could have been used in this estimation process, included maximum likelihood, neural networks, and the Hierarchical Bayes model. In the current study, the data from the MaxDiff experiments were used to estimate two MNL choice models for each study area: (1) a MNL model for commute trips, and (2) a MNL model for non-commute trips. All survey respondents participated in the MaxDiff experiments, except those respondents who indicated that they were “very uninformed” about the public transit system in the study area. (In Chicago, 7.2% of respondents indicated that they were “very uninformed” about the city’s transit system. For Charlotte, this share is 8.1%.) Here, each respondent was asked to choose his/her “most preferred” and “least preferred” transit option from a set of three transit options, each of which included information on in-vehicle travel time, fare, and three transit attributes. This exercise was repeated eight times for each respondent. For each model, the results provide a coefficient value (or utility) for each transit attribute evaluated in the MaxDiff experiments. Then, the attributes can be ordered by their utilities to identify the relative importance of each attribute. Two types of MaxDiff analyses were completed from the data collected in Chicago and Charlotte: 1. Estimates of unscaled coefficients, which capture the behavior of transit attributes relative to each other. These coefficients were used to estimate the marginal rates of substitution and relative importance of attributes (more on this in the next section).

D-2 Characteristics of premium Transit Services that Affect Choice of Mode and cost trade-offs relative to these additional service attributes (more on this in the next section). As such, the marginal rates of substitution, expressed as equivalent minutes of in- vehicle travel time, used in any planning or modeling context, should be derived only from the scaled coefficients. Model Results of Unscaled Attributes Two separate models were specified for commute and non-commute trips for each study area, as indicated in the previous section. Each estimated model considered travel time, travel cost, and interaction effects of the transit attributes with a number of individual demographics (such as gender, age, student status, employment status, indication of any mobility-related problem, and indicator of the length of residency in the study area), household socio- demographics (such as presence of children, family income, and number of vehicles in the household), trip characteristics (indicator of group travel and trip distance), and attitudinal variables (such as willingness to walk to access the transit system, pro-transit attitude, environment and productivity/time savings, pro-car attitude, transit averse, and low transit comfort level). The final model specification was obtained based on a systematic process of eliminating variables found to be statistically insignificant and unintuitive. Salt Lake City FIGURE D-1 presents marginal rates of substitution from the MaxDiff portion of the Salt Lake City survey, which evaluated the relative importance of unbundled non-traditional attributes. Results are presented as equivalent minutes of time for work and non-work trips. The marginal rates of substitution capture how much individuals are willing to pay (in terms of extra travel time) for improvements in the transit attributes. Thus, the first entry in FIGURE D-1 suggests that, on average, commuters the in Salt Lake City area would be willing to add about 15 minutes and non-commuters would be willing to add about 14 minutes to their in-vehicle travel time to always have seats available on their transit vehicle. The values for each attribute time coefficient obtained from the MaxDiff model. 2. Estimates of scaled coefficients of bundled (and not bundled) attributes. The scaled coefficients were estimated so that the effect of transit attributes on individuals’ mode choice decision may be incorporated. The purpose of scaling the coefficients is to incorporate the importance of these attributes relative to other mode choice factors (such as time and cost). The unscaled coefficients are not sufficient to use directly because the data collection method does not adequately capture the time are computed (in minutes) by dividing the overall (unscaled) coefficient by the transit travel

Transit Service Attribute Models D-3 FIGURE D-1. Values of unscaled marginal rates of substitution (min. of IVT) for Salt Lake City. Marginal Rates of Substuon (mins) 2520151050 Easy to board; doors are level with platform or curb Station/stop is well maintained and clean Benches at station/stop are clean and comfortable Transit vehicle is very new and clean Seats are comfortable and a good size Transit runs from 4 AMuntil 11 PM WiFi available on transit vehicle Real me info available Station/stop close to coffee shop, dry cleaners, grocery, restrooms Station/stop is well lit with police presence Enhanced station/stop security Parkingwithin 10 mins. walk from station 1 in 10 trips experience a delay of 5 mins. or more Easy to immediately idenfy route on transit vehicle Station/stop effectively protects you from bad weather Transit vehicle has effective air conditioning and heating Station/stop within 10 mins. walk of your home or work Fare machines are fast and easy to use Convenient transfer (short walk or on same platform) Frequency: every 10 mins in peak; every 20 mins in off peak Always seats available on transit vehicle minutes Non Work Trips Work Trips

D-4 Characteristics of premium Transit Services that Affect Choice of Mode are comfortable and a good size, and easy to board; doors are level with platform or curb) are also among the least important attributes for both work and non-work travelers. FIGURE D-1 also presents average marginal rates of substitution for premium bundles of on-board amenities and modernized station/stops. The bundle values vary in importance from 7 minutes to 11 minutes depending on the bundle and the trip purpose, whereas the unscaled and unbundled values above often exceed the coefficients for the bundles. These results demonstrate that, on average, on-board amenities are more important to work travelers and modernized station/stops are more important to non-work travelers. Chicago FIGURE D-2 and FIGURE D-3 present the MaxDiff model results (i.e., unscaled coefficients), the marginal rates of substitution, and the relative importance of transit attributes for commute and non-commute trips for the Chicago area (the detailed model results for commute and non-commute trips are presented later in this section). In addition, FIGURE D-2. and FIGURE D-3 provide a visual representation of the (unscaled) marginal rates of substitution and relative importance of transit attributes for the Chicago area, respectively. The scaled results corresponding to this study area are discussed in the next section of this appendix. The coefficients presented in FIGURE D-2 and FIGURE D-3 are the overall (unscaled) coefficients and not just the base coefficients. The overall coefficient for each transit attribute may be calculated by adding the contribution of each interacting variable with the corresponding base coefficient. For example, the overall coefficient for real-time information (info) for commute trips the in Chicago area was calculated as follows: 0.833 - 0.478 × 0.114 = 0.779, where the base coefficient is 0.833 (see TABLE D-1), the coefficient corresponding to full-time student status is -0.478, and 11.4% of individuals in the dataset are full-time students. Ease of use (easy to immediately identify route on transit vehicle, convenient transfers, and fast, easy-to-use fare machines) attributes are more important to non-work travelers than work travelers. This may be because non-work travelers are less familiar with the system and more likely to use different routes at different times of day. Cleanliness (station/stop is well maintained and clean, benches at station/stop are clean and comfortable, and transit vehicle is very new and clean) are among the least important attributes for both work and non-work travelers. This may reflect a tolerance among transit riders for cleanliness of public property and/or a state of good repair for the Utah Transit Authority (UTA), thus making this attribute less important than in other systems where cleanliness is not as good. Some aspects of comfort (seats

FIGURE D-2. Unscaled marginal rates of substitution (in minutes) for Chicago. -40-35-30-25-20-15-10-50 Real-Time Info (Station/stop design features bundle) Station/Stop Security (Station/stop design features bundle) Station/Stop Lighting/Safety (Station/stop design features bundle) Station/Stop Shelter (Station/stop design features bundle) Proximity to Services (Station/stop design features bundle) Cleanliness of Station/Stop (Station/stop design features bundle) Station/Stop Benches (Station/stop design features bundle) On-Board Seating Availability (On-board features bundle) On-Board Seating Comfort (On-board features bundle ) On-Board Temperature (On-board features bundle ) Cleanliness of Transit Vehicle (On-board features bundle ) Productivity features (On-board features bundle) Route Name/Num Identification (Not part of a bundle) Reliability (Not part of a bundle) Schedule Span (Not part of a bundle) Transit Frequency (Not part of a bundle) Transfer Distance (Not part of a bundle) Station/Stop Distance (Not part of a bundle) Parking Distance (Not part of a bundle) Ease of Boarding (Not part of a bundle) Fare Machines (Not part of a bundle) Minutes Commute trip Non-Commute trip

FIGURE D-3. Relative importance of transit attributes (in %) for Chicago. 0 1 2 3 4 5 6 7 8 Real-Time Info (Station/stop design features bundle) Station/Stop Security (Station/stop design features bundle) Station/Stop Lighting/Safety (Station/stop design features bundle) Station/Stop Shelter (Station/stop design features bundle) Proximity to Services (Station/stop design features bundle) Cleanliness of Station/Stop (Station/stop design features bundle) Station/Stop Benches (Station/stop design features bundle) On-Board Seating Availability (On-board features bundle) On-Board Seating Comfort (On-board features bundle ) On-Board Temperature (On-board features bundle ) Cleanliness of Transit Vehicle (On-board features bundle ) Productivity features (On-board features bundle) Route Name/Num Identification (Not part of a bundle) Reliability (Not part of a bundle) Schedule Span (Not part of a bundle) Transit Frequency (Not part of a bundle) Transfer Distance (Not part of a bundle) Station/Stop Distance (Not part of a bundle) Parking Distance (Not part of a bundle) Ease of Boarding (Not part of a bundle) Fare Machines (Not part of a bundle) Percentage Commute trip Non-Commute trip

Transit Service Attribute Models D-7 Several interesting observations may be made from ( FIGURE D-2 and FIGURE D-3). Overall, station/stop distance and on-board seating availability are the two most important transit attributes for both commuters and non-commuters in the Chicago area. From the viewpoint of relative rank of attributes, station/stop distance may be identified as the most important transit attribute for commuters, followed by on-board seating availability. A close inspection indicates that these two attributes have almost identical marginal rates of substitution, suggesting that station/stop distance and on-board seating availability are equally important to Chicago commuters. The non-commuters in this area, however, consider on-board seating availability slightly more important than station/stop distance feature. In addition, station/stop distance and on-board seating availability are the most important “unbundled” and “on-board feature bundle” attributes for commuters and non-commuters. In the context of on-board features, the importance of seating availability may be due to the fact that Chicago commuters and non-commuters tend to undertake long-distance trips (the average trip distances for Chicago commuters and non- commuters are 13 miles and 12 miles, respectively). Among the attributes in the “station/stop design feature bundle,” cleanliness of station/stop is the most important attribute for commuters as well as for non-commuters. In this regard, an interesting observation is that cleanliness of transit vehicle is rated as one of the less important transit attributes (cleanliness of transit vehicle is rated as the 20th and 17th most important attribute by commuters and non-commuters, respectively). This may be because unlike most stations/stops in Chicago area, the transit vehicle fleets are kept clean and well maintained (Chicago Transit Authority 2011). Ease of boarding and on-board seating comfort attributes also have low ratings, which may suggest that both commuters and non-commuters are fairly satisfied with the current conditions of these two attributes. Though, in general, there are similarities between commuters’ and non-commuters’ preference levels, a number of attributes are considered more important by commuters than non- commuters. These attributes include availability of good station/stop shelter, clean and comfortable station/stop benches, productivity features, transit frequency, and parking distance. On the other hand, attributes such as station/stop security, lighting/safety, proximity to services, and longer schedule span are considered more important by non-commuters (relative to commuters). A point to note here is that the relative importance of transit attributes is not fixed but changes with travel distance, as shown in FIGURE D-4, and for commute and non-commute trips, respectively. For commuters, some transit attributes such as route name/number, station/stop lighting/safety features, cleanliness of station/stop, and transit frequency become less important while transfer distance becomes noticeably more important as commute distance increases. This may be because a longer commute distance is likely to involve one or more transfers. A convenient transfer (e.g., on the same platform or only a short walk) would reduce a commuter’s total out-of-vehicle travel time as well as overall travel time. In contrast, transfer distance becomes less important for non-commute trips because the need to be at work on time is not applicable to this group of travelers (see FIGURE D-5).

Relative importance of attributes at mean distance FIGURE D-4. Relative importance of transit attributes by distance for Chicago (commute trip). 1 3 5 7 9 11 13 15 17 19 21 1 3 5 7.5 10 13.1 15 17.5 20 22.5 25 27.5 30 32.5 35 37.5 40 42.5 45 47.5 50 s e t u b i r t t a f o e c n a t r o p m i e v i t a l e R ( 1 = m o s t i m p o r t a n t , 2 1 = l e a s t i m p o r t a n t ) Distance (in miles) Real-Time Info Station/stop design features Station/Stop Security Station/Stop Shelter Station/Stop Lighting/Safety Proximity to Services Cleanliness of Station/Stop Station/Stop Benches On-Board Seating Availability On-Board Seating Comfort On-Board Temperature Cleanliness of Transit Vehicle Productivity features Route Name/Num Identification Reliability Schedule Span Transit Frequency Transfer Distance Station/Stop Distance Parking Distance Ease of Boarding Fare Machines

Relative importance of attributes at mean distance FIGURE D-5. Relative importance of transit attributes by distance for Chicago (non-commute trip). 1 3 5 7 9 11 13 15 17 19 21 1 3 5 7.5 10 12.1 15 17.5 20 22.5 25 27.5 30 32.5 35 37.5 40 42.5 45 47.5 50 s e t u b i r t t a f o e c n a t r o p m i e v i t a l e R ( 1 = m o s t i m p o r t a n t , 2 1 = l e a s t i m p o r t a n t ) Distance (in miles) Real-Time Info Station/stop design features Station/Stop Security Station/Stop Shelter Station/Stop Lighting/Safety Proximity to Services Cleanliness of Station/Stop Station/Stop Benches On-Board Seating Availability On-Board Seating Comfort On-Board Temperature Cleanliness of Transit Vehicle Productivity features Route Name/Num Identification Reliability Schedule Span Transit Frequency Transfer Distance Station/Stop Distance Parking Distance Ease of Boarding Fare Machines

D-10 Characteristics of premium Transit Services that Affect Choice of Mode An intuitive and interesting observation is that the level of importance of transit attributes is not the same across all strata of commuters in Chicago. For example, parking distance is regarded as the most important transit attribute by full-time employed commuters relative to self-employed/part-time employed commuters with the same characteristics. (In the interest of brevity, these results are only presented in Appendix B.) Though station/stop cleanliness has an overall ranking of five, this is the most important transit attribute for respondent commuters who are full-time students, female, long-time Chicago residents, or are traveling in a group. For non- commuting female travelers, long-time Chicago residents, or group travelers, transit frequency is the most important attribute (see Appendix E). Transit frequency is also ranked highest among other non-commuting strata, such as full-time workers, retired individuals, individuals aged 35 to 55, individuals from households with 1 or 2 vehicles, and individuals from households with at least $75,000 income. From a transit strategy formulation standpoint, these results may be useful. For example, findings from this study may be used by transit providers to formulate strategies that would provide improved service to specific demographic segments. Charlotte Figure D-6 shows the MaxDiff model results for Charlotte and the marginal rates of substitution. The relative importance of transit attributes is presented in FIGURE D-7 and FIGURE D-8 for commute and non-commute trips for Charlotte (see Appendix E for the model results). The scaled results corresponding to this study area are also discussed in the next section. Commuters and non-commuters in the Charlotte area behave in a similar manner to commuters and non-commuters in Chicago, with a few exceptions (though, in general, unscaled marginal rates of substitution are higher for Charlotte non-commute trips relative to similar trips for Chicago). For example, station/stop distance and cleanliness of station/stop are the two (almost equally) important attributes for commuters in Charlotte, while on-board seating availability and cleanliness of station/stop are the most important transit attributes for non- commuters. Both commuters and non-commuters in Charlotte almost always consider station/stop security and lighting/safety to be more important than their counterpart travelers in Chicago. Relative to non-commuters, Charlotte commuters are more sensitive to attributes such as reliability and parking distance that have a direct impact on travel time, while Charlotte non- commuters are more sensitive to attributes such as schedule span and frequency that have a direct impact on trip scheduling. An interesting contrast between Chicago and Charlotte is that the relative importance of attributes does not change with travel distance for commuters and non-commuters in Charlotte (the only exception is the station/stop benches attribute for Charlotte commuters—see FIGURE D-7). To conserve space, the figure showing relative importance of transit attributes by distance for Charlotte non-commuters has been omitted. However, like Chicago, the level of importance of attributes does vary between different segments of commuters and non-commuters (see Appendix E).

FIGURE D-6. Unscaled marginal rates of substitution (in minutes) for Charlotte. -50-45-40-35-30-25-20-15-10-50 Real-Time Info (Station/stop design features) Station/Stop Security (Station/stop design features) Station/Stop Lighting/Safety (Station/stop design features) Station/Stop Shelter (Station/stop design features) Proximity to Services (Station/stop design features) Cleanliness of Station/Stop (Station/stop design features) Station/Stop Benches (Station/stop design features) On-Board Seating Availability (On-board features ) On-Board Seating Comfort (On-board features ) On-Board Temperature (On-board features ) Cleanliness of Transit Vehicle (On-board features ) Productivity features (On-board features ) Route Name/Num Identification (Not in bundle) Reliability (Not in bundle) Schedule Span (Not in bundle) Transit Frequency (Not in bundle) Transfer Distance (Not in bundle) Station/Stop Distance (Not in bundle) Parking Distance (Not in bundle) Ease of Boarding (Not in bundle) Fare Machines (Not in bundle) Minutes Commute trip Non-Commute trip

FIGURE D-7. Relative importance of transit attributes (in %) for Charlotte. 0 1 2 3 4 5 6 7 8 Real-Time Info (Station/stop design features) Station/Stop Security (Station/stop design features) Station/Stop Lighting/Safety (Station/stop design features) Station/Stop Shelter (Station/stop design features) Proximity to Services (Station/stop design features) Cleanliness of Station/Stop (Station/stop design features) Station/Stop Benches (Station/stop design features) On-Board Seating Availability (On-board features ) On-Board Seating Comfort (On-board features ) On-Board Temperature (On-board features ) Cleanliness of Transit Vehicle (On-board features ) Productivity features (On-board features ) Route Name/Num Identification (Not in bundle) Reliability (Not in bundle) Schedule Span (Not in bundle) Transit Frequency (Not in bundle) Transfer Distance (Not in bundle) Station/Stop Distance (Not in bundle) Parking Distance (Not in bundle) Ease of Boarding (Not in bundle) Fare Machines (Not in bundle) Percentage Commute trip Non-Commute trip

Relative importance of attributes at mean distance FIGURE D-8. Relative importance of transit attributes by distance for Charlotte (commute trip). 1 3 5 7 9 11 13 15 17 19 21 1 3 5 7.5 10 11.8 15 17.5 20 22.5 25 27.5 30 32.5 35 37.5 40 42.5 45 47.5 50 s e t u b i r t t a f o e c n a t r o p m i e v i t a l e R ( 1 = m o s t i m p o r t a n t , 2 1 = l e a s t i m p o r t a n t ) Distance (in miles) Real-Time Info Station/stop design features Station/Stop Security Station/Stop Shelter Station/Stop Lighting/Safety Proximity to Services Cleanliness of Station/Stop Station/Stop Benches On-Board Seating Availability On-Board Seating Comfort On-Board Temperature Cleanliness of Transit Vehicle Productivity features Route Name/Num Identification Reliability Schedule Span Transit Frequency Transfer Distance Station/Stop Distance Parking Distance Ease of Boarding Fare Machines

TABLE D-1. Relative importance of attributes by trip purpose—comparison of Chicago, Charlotte, and Salt Lake City. Aribute Aribute Bundle Commute Non Commute Chicago Charloe Salt Lake Chicago Charloe Salt Lake Real-time info* Station/stop design 14 15 12 15 14 9 Station/stop security Station/stop design 10 7 9 5 6 12 Station/stop lighting/safety Station/stop design 8 4 10 6 3 13 Station/stop shelter Station/stop design 9 6 5 10 5 4 Proximity to services Station/stop design 18 16 11 14 15 1 Cleanliness of station/stop Station/stop design 5 2 16 3 2 16 Station/stop benches Station/stop design 15 19 15 20 19 15 On-board seating availability On-board features 2 3 14 1 1 14 On-board seating comfort On-board features 19 20 20 18 20 20 On-board temperature On-board features 12 8 17 12 10 17 Cleanliness of transit vehicle On-board features 20 18 21 17 18 21 Productivity features** On-board features 17 17 19 19 16 19 Route name/number identification Not part of a bundle 11 11 6 11 13 3 Reliability Not part of a bundle 16 14 7 16 17 11 Schedule span Not part of a bundle 7 12 13 4 8 7 Transit frequency Not part of a bundle 4 10 1 8 7 5 Transfer distance Not part of a bundle 13 13 2 13 12 6 Station/stop distance Not part of a bundle 1 1 4 2 4 2 Parking distance Not part of a bundle 3 5 8 9 9 10 Ease of boarding Not part of a bundle 21 21 18 21 21 18 Fare machines Not part of a bundle 6 9 3 7 11 8 *The attribute was not part of station/stop design features bundle in the survey for Salt Lake City. **The attribute was referred to simply as “Wi-Fi” in the survey for Salt Lake City.

Transit Service Attribute Models D-15 D-15 Finally, there are more similarities between the residents the in Chicago and Charlotte areas (in terms of their preferences regarding the transit attributes) than between the residents in Chicago and Salt Lake City or between the residents in Charlotte and Salt Lake City (see TABLE D-1). In particular, the residents in Salt Lake City rate attributes such as proximity to services, cleanliness of station/stop, on-board seating availability, reliability, and transfer distance quite differently than do the Chicago and Charlotte residents. One possible reason for these differences may be that the survey design and administration was undertaken in a slightly different manner for Salt Lake City, compared to the the Chicago and Charlotte surveys. For instance, all respondents participated in the surveys for Chicago and Charlotte areas (except for respondents who indicated that they were “very uninformed” about the local transit system), whereas only respondents who were transit users participated in the Salt Lake City survey. Also, in the Salt Lake City survey, the station/stop design feature bundle did not contain the real-time information attribute and the “productivity features” attribute was included simply as “Wi-Fi” in the on-board features bundle. Model Results of Scaled Attributes The MaxDiff model results need to be scaled before the coefficients can be incorporated in planning analysis. This scaling is completed to associate the bundled coefficients in the MaxDiff model with the bundled coefficients estimated in the mode choice models. This ensures that there the individual characteristics represented in the MaxDiff model are consistent with the trade-offs identified in the mode choice model estimation process (described in Appendix E). For this exercise, depending on the coefficients that are available from the mode choice models, transit attributes are divided into two groups: 1. Bundled transit attributes include transit attributes that are part of station/stop design features and on-board features bundles. In general, for each study area/trip purpose combination, the mode choice model provides coefficients for the overall bundle. These coefficients are then used as control variables to adjust/scale the magnitude of the MaxDiff coefficients of the corresponding bundle. Specifically, assume that Ci is the overall coefficient for bundle i (estimated by the mode choice model) and Mij is a MaxDiff model coefficient that corresponds to attribute j in bundle i. To obtain the adjusted coefficient for attribute j, Sij, the following formula is applied: where J is the total number of attributes in bundle i. 2. Unbundled transit attributes includes all other transit attributes. For scaling purpose, the unbundled attributes are divided into two categories as follows: – Category A: This includes attributes for which coefficients are available from the mode choice models. Reliability and schedule span attribute fall under this category. The unscaled coefficients of these two variables can be replaced by the mode choice model results in a straightforward manner. iJ j ij ij ij C M M S 1

D-16 Characteristics of premium Transit Services that Affect Choice of Mode – Category B: All other unbundled attributes belong to this category. The overall coefficients for the station/stop design and on-board features bundles available from the mode choice and MaxDiff models are used to scale the coefficients under this category. In particular, assume that the overall coefficients for the station/stop design and on-board features bundles from the mode choice model are a1 and a2, and from the MaxDiff model are A1 and A2, respectively. Also, assume that the overall coefficient for Category B from the MaxDiff model is A3. The overall mode choice coefficient for Category B, a3, is predicted as follows: where ., 11 21 21 Amae AA aa m Then, using the formula presented above, the value of a3 may be distributed among its component attributes. To illustrate, for Chicago commuters (transit mode = bus), a1 = 0.124, a2 = 0.146, A1 = 6.275, A2 = 3.570, and A3 = 6.897. Using these values, we obtain m = -0.008, e = 0.175, and a3 = 0.119. Now, using these values, the adjusted coefficient, for example, for route name/number identification attribute .016.0119.0 897.6 908.0 The mode choice coefficients used in scaling the MaxDiff model coefficients are provided in TABLE D-2 for reference. These are described more fully in Appendix E. 33 Amea Salt Lake City TABLE D-3 and TABLE D-4 present the coefficients for the stop design and on-board amenity bundles from mode choice models, respectively, for work and non-work travel from Salt Lake City data. The tables also show the corresponding scaled coefficients and marginal rates of substitution for the attributes in the two bundles. These results demonstrate that there is a wide variation in stop design attributes for work and non-work travel, and that there is much less variation in on-board attributes for work and non-work travel.

Transit Service Attribute Models D-17 TABLE D-2. Coefficients from the mode choice model used in the MaxDiff models. Variable Commute Non-Commute City Bus Train Bus Train In vehicle travel me (minutes) Salt Lake City -0.034 -0.034 -0.049 -0.049 Chicago -0.025 -0.025 -0.019 -0.019 Charlotte* -0.022 -0.022 -0.008 -0.008 Premium stop design amenies Salt Lake City 0.166 0.166 0.061 0.061 Chicago 0.124 0.124 0.084 0.084 Charlotte 0.103 0.060 0.172 0.172 Premium on board amenies Salt Lake City 0.127 0.127 0.148 0.148 Chicago 0.146 0.146 0.205 0.205 Charlotte 0.101 0.101 0.180 0.180 Reliability** Salt Lake City -0.024 -0.024 -0.025 -0.025 Chicago 0.147 0.135 0.088 0.088 Charlotte 0.101 0.101 0 0 *The In-Vehicle Travel Time Coefficient for non-commute trips in Charlotte is lower than what is considered reasonable and so, for the purposes of the MaxDiff model scaling, the Chicago non-commute in-vehicle travel time coefficient was used instead. **The definition of the reliability coefficient and the estimation for this variable changed for Chicago and Charlotte and the coefficients are not directly comparable, because the stated preference questions were refined to reflect values closer to transit agency definitions (x% of trips delayed by y minutes or more) rather than the original (x in 10 trips experience delay of y minutes or more).

D-18 Characteristics of premium Transit Services that Affect Choice of Mode TABLE D-3. Scaled coefficients of bundled attributes for work travel: Salt Lake City. TABLE D-4. Scaled coefficients of bundled attributes for non-work travel: Salt Lake City. TABLE D-5 and TABLE D-6 present adjusted MaxDiff model results and the corresponding marginal rates of substitution for the Chicago and Charlotte areas, respectively. A comparison indicates that, at aggregate level, Chicago commuters value unbundled attributes most and station/stop design features least (though these features still add value), regardless of the mode of transit, while Chicago non-commuters value on-board amenities the most, with the unbundled attributes a close second. A similar trend can also be observed among Charlotte commuters, but Charlotte non-commuters valued the on-board, station, and unbundled attributes almost equally. In the mode choice models, the aggregate-level bundle variables (and their coefficients) may be replaced by the corresponding component attributes (and adjusted coefficients) in a straightforward manner. For the unbundled attributes, the alternative specific constants for bus and train modes should be adjusted before the scaled coefficients are included in the mode choice models. One way to adjust the mode-specific constants is to run the mode choice models with and without the “anchor” variables. The differences in the magnitude of the mode-specific constants between the two models may then be used to make suitable adjustments to the mode-specific constants of the final mode choice models that include all unbundled attributes. Aribute Mode Coefficient Std. Err t stat Value Notes Stop design Bundle (0 = standard, 1 =modern) Bus,train 0.061 0.075 0.809 1.55 mes IVTT Station/Stop Lighng/Safety 0.008 N/A N/A 0.196 mes IVTT Station/Stop Shelter 0.014 N/A N/A 0.371 mes IVTT Proximity to Services 0.019 N/A N/A 0.474 mes IVTT Cleanliness of Station/Stop 0.006 N/A N/A 0.145 mes IVTT Station/Stop Benches 0.006 N/A N/A 0.159 mes IVTT Station/Stop Security 0.009 N/A N/A 0.219 mes IVTT On board amenities (0 = standard, 1 =modern) Bus,train 0.148 0.075 1.978 3.754 mes IVTT WiFi 0.029 N/A N/A 0.735 mes IVTT On Board Seating Availability 0.055 N/A N/A 1.411 mes IVTT On Board Seating Comfort 0.016 N/A N/A 0.409 mes IVTT On Board Temperature 0.033 N/A N/A 0.853 mes IVTT Cleanliness of Transit Vehicle 0.015 N/A N/A 0.388 mes IVTT Attribute Mode Coefficient Std. Err t stat Value Notes Stop design (0 = standard, 1 =modern) Bus,train 0.166 0.043 3.843 4.628 mes IVTT Staon/Stop Lighng/Safety 0.032 N/A N/A 0.878 times IVTT Staon/Stop Shelter 0.040 N/A N/A 1.099 mes IVTT Proximity to Services 0.030 N/A N/A 0.842 mes IVTT Cleanliness of Station/Stop 0.015 N/A N/A 0.419 mes IVTT Staon/Stop Benches 0.018 N/A N/A 0.489 times IVTT Staon/Stop Security 0.032 N/A N/A 0.884 mes IVTT On board amenities (0 = standard, 1 =modern) Bus,train 0.127 0.051 2.478 3.552 mes IVTT WiFi * 0.019 N/A N/A 0.540 mes IVTT On Board Seating Availability 0.044 N/A N/A 1.230 times IVTT On Board Seating Comfort 0.018 N/A N/A 0.507 mes IVTT On Board Temperature 0.029 N/A N/A 0.810 mes IVTT Cleanliness of Transit Vehicle 0.016 N/A N/A 0.442 mes IVTT

Transit Service Attribute Models D-19 Once the transit attributes are incorporated, the mode choice models may be used to analyze future scenarios such as testing the effects of improving one or a set of attributes on transit mode share. TABLE D-5 and TABLE D-6 also provide information on how much additional time the commuters and non-commuters in the study areas may be willing to add to their in-vehicle travel time to include the corresponding attributes on the route (see the “Scaled Marginal Rates of Substitution” column). For example, Chicago commuters would be willing to increase their transit travel time by about 5 minutes to have all the station/stop design-related attributes on their routes. The results in the tables indicate that time values are very similar for station/stop design and on-board attributes for commuters in Chicago and Charlotte (except that the train mode for Charlotte commuters is valued less). The non-commuters are, in general, willing to pay more than commuters to have these attributes on their routes. A “-” entry in the tables indicates that either the coefficient is not statistically significant or the coefficient is constrained to zero to avoid a negative value. Chicago and Charlotte TABLE D-5 and TABLE D-6 present adjusted MaxDiff model results and the corresponding marginal rates of substitution for the Chicago and Charlotte areas, respectively. FIGURE D-9 and FIGURE D-10 present the same information cumulatively for Chicago and Charlotte, respectively. A visual comparison indicates that, at an aggregate level, both Chicago commuters value unbundled attributes most and station/stop design features least (though, these features still add value), regardless of the mode of transit, while Chicago non-commuters value on-board amenities the most, with the unbundled attributes a close second. A similar trend can also be observed among Charlotte commuters, but Charlotte non-commuters valued the on- board, station and unbundled attributes almost equally. In the mode choice models, the aggregate-level bundle variables (and their coefficients) may be replaced by the corresponding component attributes (and adjusted coefficients) in a straightforward manner. For the unbundled attributes, the alternative specific constants for bus and train modes should be adjusted before the scaled coefficients are included in the mode choice models. Once the transit attributes are incorporated, the mode choice models may be used to analyze future scenarios, such as testing the effects of improving one or a set of attributes on transit mode share. TABLE D-5 and TABLE D-6 also provide information on how much additional time the commuters and non-commuters in the study areas may be willing to add to their in-vehicle travel time to include the corresponding attributes on the route (see the “Scaled Marginal Rates of Substitution” column). For example, Chicago commuters would be willing to increase their transit travel time by about 5 minutes to have all the station/stop design-related attributes on their routes. The results in the tables indicate that time values are very similar for station/stop design and on-board attributes for commuters in Chicago and Charlotte (except that the train mode for Charlotte commuters is valued less). The non-commuters are, in general, willing to pay more than commuters to have these attributes on their routes. A “-” entry in the tables indicates that either the coefficient is not statistically significant or the coefficient is constrained to zero to avoid a negative value.

TABLE D-5. Scaled MaxDiff model results and marginal rates of substitution for Chicago. Aribute Scaled Coefficient (t stat) Scaled Marginal Rates of Substuon(with respect to IVTT, in minutes) Commute trips Non commute trips Commute trips Non commute trips Bus Train Bus Train Bus Train Bus Train Staon/stop design features bundle 0.124 (2.5) 0.124 (2.5) 0.084 (2.3) 0.084 (2.3) 4.96 4.96 4.42 4.42 Real me info 0.015 0.015 0.008 0.008 0.62 0.62 0.44 0.44 Staon/stop security 0.021 0.021 0.016 0.016 0.85 0.85 0.84 0.84 Staon/stop lighng/safety 0.022 0.022 0.016 0.016 0.86 0.86 0.82 0.82 Staon/stop shelter 0.021 0.021 0.013 0.013 0.86 0.86 0.69 0.69 Proximity to services 0.010 0.010 0.009 0.009 0.40 0.40 0.50 0.50 Cleanliness of staon/stop 0.022 0.022 0.016 0.016 0.90 0.90 0.86 0.86 Staon/Stop Benches 0.012 0.012 0.005 0.005 0.48 0.48 0.27 0.27 On board features bundle 0.146 (2.5) 0.146 (2.5) 0.205 (3.5) 0.205 (3.5) 5.88 5.88 10.79 10.79 On board seang availability 0.054 0.054 0.078 0.078 2.15 2.15 4.09 4.09 On board seang comfort 0.019 0.019 0.026 0.026 0.77 0.77 1.39 1.39 On board temperature 0.035 0.035 0.046 0.046 1.41 1.41 2.41 2.41 Cleanliness of transit vehicle 0.016 0.016 0.030 0.030 0.64 0.64 1.56 1.56 Producvity features 0.022 0.022 0.025 0.025 0.87 0.87 1.34 1.34 Unbundled features 0.285 0.273 0.181 0.190 11.41 10.93 9.55 10.00 Route name/number Idenficaon 0.016 0.016 0.011 0.012 0.63 0.63 0.58 0.63 Reliability 0.147 (2.1) 0.135 (2.1) 0.088 (1.6) 5.88 5.40 4.63 Schedule span* 0.019 0.019 0.015 0.016 0.77 0.77 0.78 0.86 Transit frequency 0.021 0.021 0.013 0.014 0.82 0.82 0.68 0.74 Transfer distance 0.014 0.014 0.009 0.010 0.56 0.56 0.46 0.50 Staon/stop distance 0.023 0.023 0.015 0.017 0.92 0.92 0.80 0.88 Parking distance 0.021 0.021 0.013 0.014 0.84 0.84 0.68 0.74 Ease of boarding 0.005 0.005 0.005 0.005 0.21 0.21 0.24 0.26 Fare machines 0.019 0.019 0.013 0.014 0.78 0.78 0.69 0.75 *Note: Schedule span was included as a separate variable in the Max Diff experiments, but the coefficients estimated separately for this variable were unreasonably high, so these were estimated along with the remaining unbundled features, relative to the reliability measure, which was also included as a separate variable in the Max Diff experiments.

TABLE D-6. Scaled MaxDiff model results and marginal rates of substitution for Charlotte. Aribute Scaled Coefficient (t stat) Scaled Marginal Rates of Substuon (with respect to IVTT, in minutes) Commute Trips Non Commute Trips Commute Trips Non Commute Trips** Bus Train Bus Train Bus Train Bus Train Staon/stop design features bundle 0.103 (2.4) 0.060 (1.6) 0.172 (2.1) 0.172 (2.1) 4.68 2.73 9.05 9.05 Real me info 0.011 0.006 0.020 0.020 0.51 0.30 1.06 1.06 Staon/stop security 0.017 0.010 0.030 0.030 0.76 0.44 1.56 1.56 Staon/stop lighng/safety 0.018 0.011 0.031 0.031 0.83 0.48 1.62 1.62 Staon/stop shelter 0.018 0.010 0.030 0.030 0.81 0.47 1.57 1.57 Proximity to services 0.011 0.006 0.017 0.017 0.51 0.29 0.89 0.89 Cleanliness of staon/stop 0.020 0.012 0.033 0.033 0.92 0.53 1.74 1.74 Staon/Stop Benches 0.008 0.005 0.012 0.012 0.35 0.21 0.62 0.62 On board features bundle 0.101 (2.1) 0.101 (2.1) 0.180 (3.1) 0.180 (3.1) 4.59 4.59 9.47 9.47 On board seang availability 0.032 0.032 0.063 0.063 1.46 1.46 3.32 3.32 On board seang comfort 0.012 0.012 0.019 0.019 0.56 0.56 1.02 1.02 On board temperature 0.026 0.026 0.046 0.046 1.20 1.20 2.42 2.42 Cleanliness of transit vehicle 0.013 0.013 0.024 0.024 0.60 0.60 1.26 1.26 Producvity features 0.017 0.017 0.027 0.027 0.76 0.76 1.45 1.45 Unbundled features 0.548 0.410 0.741 0.864 9.89 8.01 10.61 9.50 Route name/number Idenficaon 0.015 0.010 0.023 0.040 0.69 0.45 1.23 1.10 Reliability 0.101 (2.3) 0.101 (2.3) 4.59 4.59 Schedule span* 0.014 0.009 0.028 0.025 0.63 0.41 1.47 1.31 Transit frequency 0.016 0.010 0.028 0.048 0.73 0.47 1.49 1.34 Transfer distance 0.012 0.008 0.025 0.042 0.56 0.36 1.29 1.15 Staon/stop distance 0.021 0.014 0.033 0.057 0.98 0.63 1.76 1.58 Parking distance 0.019 0.012 0.027 0.047 0.88 0.57 1.44 1.28 Ease of boarding 0.002 0.001 0.010 0.017 0.09 0.06 0.52 0.47 Fare machines 0.016 0.010 0.027 0.046 0.74 0.47 1.40 1.26 *Note: Schedule span was included as a separate variable in the Max Diff experiments, but the coefficients estimated separately for this variable were unreasonably high, so these were estimated along with the remaining unbundled features, relative to the reliability measure, which was also included as a separate variable in the Max Diff experiments ** These non-commute trips have been scaled to an in-vehicle travel time coefficient of -0.0019 from Chicago because the Charlotte in-vehicle travel time coefficient was too low to be reasonable.

D-22 Characteristics of Premium Transit Services that Affect Choice of Mode FIGURE D-9. Cumulative scaled equivalent minutes of in-vehicle travel time for Chicago.

Transit Service Attribute Models D-23 FIGURE D-10. Cumulative scaled equivalent minutes of in-vehicle travel time for Charlotte.

D-24 Characteristics of premium Transit Services that Affect Choice of Mode Details of the Transit Attribute Models The following tables provide more detailed model results for the Maximum Difference Scaling models presented in Chapter 3 of TCRP Report 166: TABLE D-7. MaxDiff model estimation results for commute trips (t-stat) – Chicago TABLE D-8. MaxDiff model estimation results for non-commute trips (t-stat) – Chicago TABLE D-9. Relative rank of attributes by individual demographics, household demographics, trip characteristics, and attitudinal variables (relative importance in %) – Chicago (commute trips) TABLE D-10. Relative rank of attributes by individual demographics, household demographics, trip characteristics, and attitudinal variables (relative importance in %) – Chicago (non-commute trips) TABLE D-11. MaxDiff model estimation results for commute trips (t-stat) – Charlotte TABLE D-12. MaxDiff model estimation results for non-commute trips (t-stat) – Charlotte TABLE D-13. Relative rank of attributes by individual demographics, household demographics, trip characteristics, and attitudinal variables (relative importance in %) – Charlotte (commute trips) TABLE D-14. Relative rank of attributes by individual demographics, household demographics, trip characteristics, and attitudinal variables (relative importance in %) – Charlotte (non-commute trips)

TABLE D-7. MaxDiff model estimation results for commute trips (t-stat)—Chicago. LOS/Attribute Attribute Bundle Coefficient (t-stat) Individual Demographics Variables Household Demographics Variables Trip Characteristics Attitude Full-time student (base: part-time student/not student) Employment status (base: self-employed/part- time employed/ not employed) Female (base: male) Long- time resident (> 5 years, base: short-time resident) Has mobility problem (base: no mobility problem) Age (base: age < 35) Number of vehicles in the HH (base: no vehicles) Family income (base: income < 35k) Presence of kid(s) in the HH (base: no kids in the HH) Group travel (Actual distance /mean distance) Pro-Transit Attitude Environment, Productivity, and Time Savings Pro-Car Attitude Transit Averse Low Transit Comfort Level Willingness to walk Full-time employed Retired 35 ≤ Age < 55 Age ≥ 55 Number of vehs in the HH = 1 Number of vehs in the HH = 2 Number of vehs in the HH = 3+ 35k ≤ Income < 50k 50k ≤ Income < 75k 75k ≤ Income < 100k Income ≥ 100k Willing to walk ≤ 2 min. Willing to walk ≥ 10 min. IVTT - -0.051 (-10.2) Fare - -1.17 (-16.87) 0.107 (3) Real-Time Info S t a t i o n / s t o p d e s i g n f e a t u r e s 0.833 (11.22) -0.478 (-2.27) #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A Station/Stop Security 0.51 (3.66) #N/A #N/A #N/A 0.616 (4.08) #N/A #N/A 0.447 (2.99) #N/A #N/A #N/A 0.445 (2.35) #N/A #N/A #N/A #N/A -0.411 (-2.59) #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A Station/Stop Lighting/Safety 1.16 (10.46) #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A 0.551 (2.64) #N/A #N/A #N/A -0.148 (-1.89) #N/A #N/A #N/A 0.259 (2.44) -0.284 (-2.45) #N/A #N/A Station/Stop Shelter 0.691 (5.37) #N/A #N/A #N/A #N/A #N/A #N/A 0.431 (2.63) 0.523 (2.56) #N/A #N/A #N/A #N/A 0.348 (1.98) #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A Proximity to Services 0.516 (7.3) #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A -0.213 (-1.95) #N/A #N/A Cleanliness of Station/Stop 1.31 (11.93) #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A -0.164 (-2.13) -0.17 (-1.91) #N/A #N/A #N/A #N/A #N/A #N/A Station/Stop Benches 0.427 (4.81) 0.654 (2.75) #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A 0.309 (2.03) #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A On-Board Seating Availability O n - b o a r d f e a t u r e s 1.02 (9.93) #N/A #N/A #N/A #N/A #N/A #N/A 0.425 (2.89) #N/A #N/A #N/A 0.49 (2.48) #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A On-Board Seating Comfort 0.572 (6.73) #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A -0.31 (-2.09) #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A On-Board Temperature 0.7 (6.04) #N/A #N/A #N/A #N/A #N/A #N/A #N/A 0.408 (2.21) #N/A 0.408 (2.43) 0.491 (2.53) #N/A #N/A #N/A #N/A -0.472 (-2.97) #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A Cleanliness of Transit Vehicle 0.711 (5.47) #N/A -0.361 (-2.37) #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A 0.261 (2.97) #N/A 0.216 (1.99) -0.641 (-3) #N/A Productivity features 0.651 (8.17) #N/A #N/A #N/A #N/A #N/A #N/A #N/A -0.339 (-2.05) #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A -0.45 (-2.02) #N/A Route Name/Num Identification N o t p a r t o f a b u n d l e 0.92 (6.31) #N/A #N/A #N/A 0.326 (2.21) #N/A #N/A #N/A #N/A #N/A #N/A 0.401 (2.16) #N/A #N/A #N/A #N/A #N/A #N/A -0.226 (-3.03) #N/A #N/A #N/A #N/A #N/A -0.761 (-3.47) #N/A Reliability 0.552 (7.82) #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A Schedule Span 0.778 (5.31) #N/A #N/A #N/A #N/A #N/A #N/A 0.328 (1.97) 0.411 (1.98) -0.296 (-1.93) #N/A #N/A #N/A 0.399 (2.15) 0.613 (2.8) #N/A #N/A #N/A #N/A #N/A #N/A -0.223 (- 2.38) #N/A #N/A #N/A #N/A Transit Frequency 0.794 (5.04) #N/A #N/A #N/A 0.482 (3.17) #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A 0.681 (3.2) 0.49 (2.83) #N/A #N/A -0.167 (-2.15) #N/A #N/A -0.19 (-2.04) #N/A #N/A #N/A #N/A Transfer Distance 0.669 (6.49) #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A 0.147 (1.83) #N/A #N/A #N/A #N/A #N/A #N/A #N/A Station/Stop Distance 0.77 (5.27) #N/A 0.301 (1.91) #N/A #N/A #N/A #N/A #N/A #N/A #N/A 0.347 (2.17) #N/A #N/A 0.597 (3.22) 0.512 (2.44) #N/A #N/A #N/A #N/A #N/A #N/A -0.228 (-2.43) #N/A #N/A #N/A #N/A Parking Distance 1.1 (5.67) #N/A 0.358 (2.24) #N/A #N/A #N/A #N/A #N/A #N/A #N/A 0.292 (1.72) 0.687 (3.31) #N/A #N/A #N/A #N/A -0.412 (-2.54) #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A -0.373 (-2.4) Ease of Boarding 0.378 (4.88) #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A -0.37 (-2.1) #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A Fare Machines #N/A #N/A #N/A #N/A #N/A #N/A #N/A 0.582 (3.91) 0.83 (4.21) 0.764 (5.09) 0.685 (4.47) 0.92 (4.88) #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A

TABLE D-8. MaxDiff model estimation results for non-commute trips (t-stat)—Chicago. LOS/Attribute Attribute Bundle Coefficient (t-stat) Individual Demographics Variables Household Demographics Variables Trip Characteristics Attitude Full- time studen t (base: part-time student/ not student) Employment status (base: self-employed/part- time employed/ not employed) Femal e (base: male) Long- time resident (> 5 years, base: short-time resident) Has mobility problem (base: no mobility problem) Age (base: age < 35) Number of vehicles in the HH (base: no vehicles) Family income (base: income < 35k) Presence of kid(s) in the HH (base: no kids in the HH) Group travel (Actual distance /mean distance) Pro- Transit Attitud e Environment, Productivity, and Time Savings Pro-Car Attitude Transit Averse Low Transit Comfort Level Willingness to walk Full-time employed Retired 35 ≤ Age < 55 Age ≥ 55 Number of vehs in the HH = 1 Number of vehs in the HH = 2 Number of vehs in the HH = 3+ 35k ≤ Income < 50k 50k ≤ Income < 75k 75k ≤ Income < 100k Income ≥ 100k Willing to walk ≤ 2 min. Willing to walk ≥ 10 min. IVTT - -0.056 (-5.6) 0.011 (3.27) Fare - -1.46 (-21.09) 0.42 (4.82) 0.156 (6.93) Real-Time Info S t a t i o n / s t o p d e s i g n f e a t u r e s 0.748 (9.47) Station/Stop Security 0.791 (4.75) 0.498 (2.73) 0.751 (4.01) 0.595 (2.3) -0.427 (-1.9) Station/Stop Lighting/Safety 1.04 (8.31) 0.413 (1.81) 0.693 (2.79) 0.331 (1.95) Station/Stop Shelter 1.16 (13.82) -0.18 (-1.86) Proximity to Services 0.744 (8.2) 0.447 (2.18) Cleanliness of Station/Stop 0.963 (5.84) 0.485 (2.65) 0.443 (2.42) Station/Stop Benches 0.458 (5.71) On-Board Seating Availability O n - b o a r d f e a t u r e s 1.32 (7.62) 0.337 (1.79) 0.752 (2.54) -0.429 (-2.4) 0.499 (2.19) 0.412 (3.67) On-Board Seating Comfort 0.783 (5.44) -0.347 (-2) On-Board Temperature 0.946 (11.25) Cleanliness of Transit Vehicle 0.544 (6.16) 0.456 (1.97) Productivity features 0.646 (4.67) -0.359 (-2.16) 0.649 (2.95) 0.256 (2.55) Route Name/Num Identification N o t p a r t o f a b u n d l e 0.462 (3) 0.564 (3.28) 0.42 (2.43) 0.263 (2.48) 0.262 (2.13) 0.575 (2.04) Reliability 0.673 (8.38) Schedule Span 1.31 (11.73) 0.357 (1.97) -0.238 (-2.28) Transit Frequency 1.66 (12.2) -0.6 (- 2.71) -0.5 (-2.71) -0.169 (-2.3) -0.326 (-3.22) Transfer Distance 0.456 (2.61) 0.424 (2.44) 0.408 (2.35) 0.52 (2.85) -0.18 (-2.51) Station/Stop Distance 1.08 (6.56) 0.88 (2.49) 0.384 (2.08) 0.33 (1.84) -0.412 (-3.32) 0.395 (2.82) Parking Distance 0 (0) 1.41 (9.64) 1.15 (7.63) 1.58 (6.79) 0.463 (1.96) Ease of Boarding 0 (0) 0.508 (4.89) 0.876 (3.59) Fare Machines 0 (0) 0.589 (4.09) 0.791 (3.39) 0.961 (4.28) 0.695 (2.89) 0.671 (3.34) 0.424 (2.78)

TABLE D-9. Relative rank of attributes by individual demographics, household demographics, trip characteristics, and attitudinal variables (relative importance in %)—Chicago (commute trips). Attribute Attribute Bundle Variables Individual Demographics Household Demographics Trip Characteristics Attitude Willingness to Walk Full-time student (base: part-time student/ not student) Employment status (base: self- employed/part-time employed/ not employed) Female (base: male) Long- time resident (> 5 years, base: short- time resident) Has mobility problem (base: no mobility problem) Age (base: age < 35) Number of vehicles in the HH (base: no vehicles) Family income (base: income < 35k) Presenc e of kid(s) in the HH (base: no kids in the HH) Group travel (Actual distance /mean distance) Pro- Transit Attitude Environ- ment, Produc- tivity, and Time Savings Pro-Car Attitude Transit Averse Low Transit Comfort Level Willing to walk ≤ 2 min. Willing to walk ≥ 10 min. Full-time employed Retired 35 ≤ Age < 55 Age ≥ 55 Number of vehs in the HH = 1 Number of vehs in the HH = 2 Number of vehs in the HH = 3+ 35k ≤ Income < 50k 50k ≤ Income < 75k 75k ≤ Income < 100k Income ≥ 100k Real-time info S t a t i o n / s t o p d e s i g n f e a t u r e s 20 (2.33%) 7 (5.42%) 6 (5.53%) 8 (5.05%) 6 (5.53%) 6 (5.53%) 9 (4.82%) 9 (4.93%) 6 (5.36%) 8 (4.87%) 9 (4.60%) 6 (5.53%) 9 (5.08%) 9 (4.78%) 7 (5.36%) 5 (6.19%) 6 (5.53%) 5 (5.74%) 6 (5.59%) 6 (5.53%) 7 (5.67%) 6 (5.44%) 7 (5.64%) 5 (6.31%) 5 (5.67%) Station/stop security 18 (3.35%) 17 (3.32%) 18 (3.39%) 5 (6.83%) 18 (3.39%) 18 (3.39%) 7 (5.54%) 18 (3.02%) 18 (3.28%) 20 (2.98%) 7 (5.27%) 18 (3.39%) 18 (3.11%) 18 (2.93%) 18 (3.28%) 20 (0.74%) 18 (3.39%) 18 (3.52%) 18 (3.42%) 18 (3.39%) 18 (3.47%) 18 (3.33%) 17 (3.45%) 15 (3.86%) 18 (3.47%) Station/stop lighting/safety 2 (7.61%) 3 (7.55%) 2 (7.70%) 4 (7.04%) 2 (7.70%) 2 (7.70%) 3 (6.71%) 4 (6.87%) 2 (7.47%) 3 (6.78%) 6 (6.40%) 2 (7.70%) 4 (7.07%) 1 (9.82%) 3 (7.46%) 2 (8.62%) 2 (7.70%) 4 (6.98%) 1 (7.79%) 2 (7.70%) 2 (7.90%) 1 (9.26%) 6 (5.93%) 2 (8.78%) 2 (7.90%) Station/stop shelter 12 (4.53%) 11 (4.50%) 12 (4.59%) 13 (4.19%) 12 (4.59%) 12 (4.59%) 4 (6.49%) 2 (7.19%) 12 (4.45%) 13 (4.04%) 14 (3.81%) 12 (4.59%) 6 (6.33%) 12 (3.97%) 12 (4.44%) 10 (5.13%) 12 (4.59%) 12 (4.76%) 12 (4.64%) 12 (4.59%) 9 (4.71%) 12 (4.51%) 12 (4.67%) 10 (5.23%) 12 (4.70%) Proximity to services 17 (3.39%) 16 (3.36%) 17 (3.43%) 18 (3.13%) 17 (3.43%) 17 (3.43%) 19 (2.99%) 17 (3.05%) 17 (3.32%) 19 (3.02%) 19 (2.85%) 17 (3.43%) 17 (3.15%) 17 (2.96%) 17 (3.32%) 15 (3.83%) 17 (3.43%) 17 (3.56%) 17 (3.46%) 17 (3.43%) 17 (3.51%) 17 (3.37%) 20 (2.05%) 14 (3.91%) 17 (3.51%) Cleanliness of station/stop 1 (8.60%) 2 (8.53%) 1 (8.70%) 1 (7.95%) 1 (8.70%) 1 (8.70%) 2 (7.58%) 1 (7.75%) 1 (8.44%) 2 (7.66%) 4 (7.23%) 1 (8.70%) 2 (7.98%) 4 (7.52%) 1 (8.42%) 1 (9.73%) 1 (8.70%) 1 (7.90%) 2 (7.66%) 1 (8.70%) 1 (8.92%) 2 (8.55%) 1 (8.86%) 1 (9.92%) 1 (8.92%) Station/stop benches 4 (7.09%) 18 (2.78%) 19 (2.83%) 19 (2.59%) 19 (2.83%) 19 (2.83%) 20 (2.47%) 19 (2.53%) 20 (2.75%) 11 (4.30%) 20 (2.36%) 19 (2.83%) 19 (2.60%) 19 (2.45%) 19 (2.75%) 16 (3.17%) 19 (2.83%) 19 (2.94%) 19 (2.87%) 19 (2.83%) 19 (2.91%) 19 (2.79%) 18 (2.89%) 16 (3.23%) 19 (2.91%) On-board seating availability O n - b o a r d f e a t u r e s 5 (6.69%) 5 (6.64%) 4 (6.77%) 7 (6.19%) 4 (6.77%) 4 (6.77%) 1 (8.36%) 7 (6.04%) 4 (6.57%) 6 (5.96%) 2 (8.33%) 4 (6.77%) 7 (6.22%) 7 (5.86%) 5 (6.56%) 3 (7.58%) 4 (6.77%) 3 (7.03%) 4 (6.85%) 4 (6.77%) 4 (6.95%) 4 (6.66%) 3 (6.90%) 4 (7.72%) 3 (6.94%) On-board seating comfort 15 (3.75%) 14 (3.72%) 15 (3.80%) 16 (3.47%) 15 (3.80%) 15 (3.80%) 17 (3.31%) 15 (3.39%) 15 (3.68%) 17 (3.34%) 17 (3.16%) 15 (3.80%) 15 (3.49%) 15 (3.28%) 15 (3.68%) 18 (1.95%) 15 (3.80%) 15 (3.94%) 15 (3.84%) 15 (3.80%) 13 (3.90%) 15 (3.73%) 15 (3.87%) 12 (4.33%) 15 (3.89%) On-board temperature 11 (4.59%) 10 (4.56%) 11 (4.65%) 12 (4.25%) 11 (4.65%) 11 (4.65%) 13 (4.05%) 5 (6.56%) 11 (4.51%) 5 (6.48%) 5 (6.57%) 11 (4.65%) 12 (4.27%) 11 (4.02%) 11 (4.50%) 19 (1.69%) 11 (4.65%) 10 (4.83%) 11 (4.70%) 11 (4.65%) 8 (4.77%) 11 (4.57%) 11 (4.74%) 9 (5.30%) 11 (4.77%) Cleanliness of transit vehicle 10 (4.67%) 20 (2.28%) 10 (4.72%) 11 (4.31%) 10 (4.72%) 10 (4.72%) 12 (4.12%) 13 (4.21%) 10 (4.58%) 12 (4.16%) 13 (3.92%) 10 (4.72%) 11 (4.33%) 10 (4.08%) 10 (4.57%) 9 (5.28%) 10 (4.72%) 9 (4.90%) 10 (4.77%) 10 (4.72%) 5 (6.62%) 10 (4.64%) 4 (6.27%) 20 (0.53%) 10 (4.84%) Productivity features 14 (4.27%) 13 (4.24%) 14 (4.32%) 15 (3.95%) 14 (4.32%) 14 (4.32%) 15 (3.77%) 21 (1.85%) 14 (4.19%) 16 (3.81%) 16 (3.59%) 14 (4.32%) 14 (3.97%) 14 (3.74%) 14 (4.19%) 13 (4.84%) 14 (4.32%) 13 (4.49%) 14 (4.37%) 14 (4.32%) 11 (4.43%) 14 (4.25%) 14 (4.40%) 18 (1.52%) 14 (4.43%) Route name/num. identification N o t p a r t o f a b u n d l e 6 (6.04%) 6 (5.99%) 5 (6.11%) 3 (7.56%) 5 (6.11%) 5 (6.11%) 8 (5.33%) 8 (5.45%) 5 (5.92%) 7 (5.38%) 3 (7.29%) 5 (6.11%) 8 (5.61%) 8 (5.28%) 6 (5.92%) 4 (6.84%) 5 (6.11%) 11 (4.78%) 5 (6.18%) 5 (6.11%) 6 (6.27%) 5 (6.00%) 5 (6.22%) 19 (1.20%) 4 (6.26%) Reliability 16 (3.62%) 15 (3.59%) 16 (3.66%) 17 (3.35%) 16 (3.66%) 16 (3.66%) 18 (3.20%) 16 (3.27%) 16 (3.55%) 18 (3.23%) 18 (3.05%) 16 (3.66%) 16 (3.36%) 16 (3.17%) 16 (3.55%) 14 (4.10%) 16 (3.66%) 16 (3.81%) 16 (3.71%) 16 (3.66%) 15 (3.76%) 16 (3.60%) 16 (3.73%) 13 (4.18%) 16 (3.76%) Schedule span 8 (5.11%) 9 (5.07%) 8 (5.17%) 9 (4.72%) 8 (5.17%) 8 (5.17%) 5 (6.40%) 3 (7.04%) 19 (3.10%) 10 (4.55%) 11 (4.29%) 8 (5.17%) 3 (7.17%) 3 (7.99%) 8 (5.00%) 7 (5.78%) 8 (5.17%) 7 (5.36%) 8 (5.22%) 8 (5.17%) 14 (3.78%) 8 (5.08%) 9 (5.26%) 7 (5.89%) 7 (5.30%) Transit frequency 7 (5.21%) 8 (5.17%) 7 (5.27%) 2 (7.74%) 7 (5.27%) 7 (5.27%) 10 (4.60%) 11 (4.70%) 7 (5.11%) 9 (4.64%) 10 (4.38%) 7 (5.27%) 10 (4.84%) 2 (8.47%) 2 (8.26%) 6 (5.90%) 7 (5.27%) 14 (4.32%) 7 (5.33%) 7 (5.27%) 12 (4.11%) 7 (5.18%) 8 (5.37%) 6 (6.01%) 6 (5.41%) Transfer distance 13 (4.39%) 12 (4.36%) 13 (4.44%) 14 (4.06%) 13 (4.44%) 13 (4.44%) 14 (3.87%) 14 (3.96%) 13 (4.31%) 15 (3.91%) 15 (3.69%) 13 (4.44%) 13 (4.08%) 13 (3.84%) 13 (4.30%) 12 (4.97%) 13 (4.44%) 6 (5.63%) 13 (4.49%) 13 (4.44%) 10 (4.56%) 13 (4.37%) 13 (4.53%) 11 (5.06%) 13 (4.55%) Station/stop distance 9 (5.05%) 4 (6.97%) 9 (5.11%) 10 (4.67%) 9 (5.11%) 9 (5.11%) 11 (4.46%) 12 (4.56%) 8 (4.96%) 4 (6.53%) 12 (4.25%) 9 (5.11%) 1 (8.33%) 5 (7.36%) 9 (4.95%) 8 (5.72%) 9 (5.11%) 8 (5.31%) 9 (5.17%) 9 (5.11%) 16 (3.69%) 9 (5.03%) 10 (5.21%) 8 (5.83%) 8 (5.24%) Parking distance 3 (7.22%) 1 (9.49%) 3 (7.30%) 6 (6.67%) 3 (7.30%) 3 (7.30%) 6 (6.37%) 6 (6.51%) 3 (7.08%) 1 (8.14%) 1 (9.86%) 3 (7.30%) 5 (6.70%) 6 (6.31%) 4 (7.07%) 11 (5.11%) 3 (7.30%) 2 (7.58%) 3 (7.39%) 3 (7.30%) 3 (7.49%) 3 (7.18%) 2 (7.44%) 3 (8.33%) 9 (4.95%) Ease of boarding 19 (2.48%) 19 (2.46%) 20 (2.51%) 20 (2.29%) 20 (2.51%) 20 (2.51%) 21 (2.19%) 20 (2.24%) 21 (2.43%) 21 (2.21%) 21 (0.04%) 20 (2.51%) 20 (2.30%) 20 (2.17%) 20 (2.43%) 17 (2.81%) 20 (2.51%) 20 (2.61%) 20 (2.54%) 20 (2.51%) 20 (2.57%) 20 (2.47%) 19 (2.56%) 17 (2.86%) 20 (2.57%) Fare machines - - - - - - 16 (3.37%) 10 (4.91%) 9 (4.92%) 14 (4.01%) 8 (5.08%) - - - - - - - - - - - - - -

TABLE D-10. Relative rank of attributes by individual demographics, household demographics, trip characteristics, and attitudinal variables (relative importance in %)—Chicago (non-commute trips). Attribute Attribute Bundle Variables Individual Demographics Household Demographics Trip Characteristics Attitude Willingness to Walk Full-time student (base: part-time student/n ot student) Employment status (base: self- employed/part-time employed/ not employed) Female (base: male) Long- time resident (> 5 years, base: short- time resident) Has mobility problem (base: no mobility problem) Age (base: age < 35) Number of vehicles in the HH (base: no vehicles) Family income (base: income < 35k) Presence of kid(s) in the HH (base: no kids in the HH) Group travel (Actual distance/ mean distance) Pro- Transit Attitude Environ- ment, Produc- tivity, and Time Savings Pro-Car Attitude Transit Averse Low Transit Comfort Level Willing to walk ≤ 2 min. Willing to walk ≥ 10 min. Full-time employed Retired 35 ≤ Age < 55 Age ≥ 55 Number of vehs in the HH = 1 Number of vehs in the HH = 2 Number of vehs in the HH = 3+ 35k ≤ Income < 50k 50k ≤ Income < 75k 75k ≤ Income < 100k Income ≥ 100k Real-time info S t a t i o n / s t o p d e s i g n f e a t u r e s 11 (4.49%) 12 (4.51%) 11 (4.62%) 13 (4.01%) 11 (4.60%) 12 (4.30%) 11 (4.74%) 12 (4.21%) 12 (4.46%) 12 (4.42%) 13 (3.91%) 11 (4.81%) 13 (4.23%) 12 (4.25%) 11 (4.55%) 12 (4.73%) 11 (4.64%) 11 (4.85%) 11 (4.74%) 12 (4.48%) 11 (4.97%) 10 (4.78%) 11 (4.62%) 12 (4.57%) 10 (4.72%) Station/stop security 9 (4.75%) 10 (4.77%) 9 (4.88%) 5 (6.91%) 9 (4.86%) 10 (4.54%) 9 (5.01%) 3 (8.68%) 10 (4.72%) 10 (4.67%) 4 (7.25%) 19 (2.34%) 11 (4.47%) 10 (4.50%) 9 (4.81%) 10 (5.01%) 9 (4.91%) 9 (5.12%) 9 (5.01%) 10 (4.73%) 9 (5.26%) 8 (5.06%) 9 (4.89%) 10 (4.84%) 9 (4.99%) Station/stop Lighting/safety 6 (6.24%) 6 (6.27%) 2 (8.97%) 8 (5.57%) 6 (6.39%) 6 (5.97%) 6 (6.59%) 8 (5.86%) 6 (6.20%) 7 (6.14%) 1 (9.07%) 6 (6.69%) 7 (5.88%) 7 (5.91%) 6 (6.32%) 6 (6.58%) 2 (8.51%) 6 (6.74%) 6 (6.59%) 6 (6.22%) 5 (6.91%) 5 (6.65%) 6 (6.43%) 6 (6.36%) 6 (6.56%) Station/stop shelter 5 (6.96%) 5 (6.99%) 5 (7.16%) 6 (6.21%) 5 (7.13%) 4 (6.66%) 4 (7.35%) 6 (6.53%) 4 (6.92%) 4 (6.85%) 7 (6.07%) 3 (7.46%) 5 (6.56%) 5 (6.59%) 4 (7.05%) 3 (7.34%) 5 (7.20%) 4 (7.52%) 4 (7.35%) 4 (6.94%) 6 (6.52%) 4 (7.42%) 5 (7.17%) 4 (7.09%) 4 (7.31%) Proximity to services 12 (4.46%) 13 (4.48%) 12 (4.59%) 14 (3.99%) 12 (4.57%) 13 (4.27%) 12 (4.71%) 13 (4.19%) 13 (4.44%) 13 (4.39%) 14 (3.89%) 12 (4.79%) 4 (6.73%) 13 (4.23%) 12 (4.52%) 13 (4.71%) 12 (4.62%) 12 (4.82%) 12 (4.71%) 13 (4.45%) 12 (4.95%) 11 (4.76%) 12 (4.60%) 13 (4.55%) 11 (4.69%) Cleanliness of station/stop 7 (5.78%) 7 (5.80%) 7 (5.95%) 3 (7.76%) 7 (5.92%) 7 (5.53%) 7 (6.10%) 4 (7.92%) 7 (5.74%) 8 (5.69%) 10 (5.04%) 7 (6.19%) 8 (5.44%) 8 (5.47%) 7 (5.85%) 8 (6.09%) 7 (5.98%) 7 (6.24%) 7 (6.10%) 7 (5.76%) 7 (6.40%) 6 (6.16%) 7 (5.95%) 8 (5.89%) 7 (6.07%) Station/stop benches 17 (2.75%) 18 (2.76%) 17 (2.83%) 19 (2.45%) 17 (2.82%) 18 (2.63%) 17 (2.90%) 17 (2.58%) 18 (2.73%) 18 (2.70%) 18 (2.40%) 17 (2.95%) 18 (2.59%) 19 (2.60%) 18 (2.78%) 18 (2.90%) 17 (2.84%) 17 (2.97%) 17 (2.90%) 17 (2.74%) 17 (3.05%) 17 (2.93%) 17 (2.83%) 17 (2.80%) 16 (2.89%) On-board seating availability O n - b o a r d f e a t u r e s 3 (7.92%) 3 (7.96%) 3 (8.15%) 2 (8.88%) 3 (8.12%) 1 (11.90 %) 2 (8.36%) 5 (7.43%) 9 (5.31%) 2 (7.79%) 5 (6.91%) 1 (8.49%) 1 (10.28 %) 2 (7.50%) 2 (8.02%) 1 (8.35%) 3 (8.19%) 2 (8.55%) 2 (8.36%) 1 (10.36% ) 2 (8.78%) 2 (8.44%) 3 (8.16%) 2 (8.07%) 2 (8.32%) On-board seating comfort 10 (4.70%) 11 (4.72%) 10 (4.83%) 12 (4.20%) 10 (4.81%) 11 (4.50%) 10 (4.96%) 11 (4.41%) 11 (4.67%) 11 (4.62%) 12 (4.10%) 10 (5.04%) 12 (4.43%) 11 (4.45%) 10 (4.76%) 11 (4.95%) 10 (4.86%) 10 (5.07%) 10 (4.96%) 11 (4.68%) 10 (5.21%) 9 (5.01%) 10 (4.84%) 11 (4.79%) 18 (2.75%) On-board temperature 8 (5.68%) 8 (5.70%) 8 (5.84%) 10 (5.07%) 8 (5.82%) 8 (5.43%) 8 (5.99%) 9 (5.33%) 8 (5.64%) 9 (5.59%) 11 (4.95%) 8 (6.08%) 10 (5.35%) 9 (5.38%) 8 (5.75%) 9 (5.99%) 8 (5.87%) 8 (6.13%) 8 (5.99%) 8 (5.66%) 8 (6.29%) 7 (6.05%) 8 (5.85%) 9 (5.78%) 8 (5.96%) Cleanliness of transit vehicle 15 (3.26%) 16 (3.28%) 15 (3.36%) 18 (2.91%) 14 (3.35%) 16 (3.12%) 15 (3.45%) 16 (3.06%) 16 (3.24%) 16 (3.21%) 9 (5.23%) 15 (3.50%) 16 (3.08%) 16 (3.09%) 16 (3.31%) 16 (3.44%) 15 (3.38%) 15 (3.52%) 15 (3.45%) 16 (3.25%) 15 (3.62%) 16 (3.48%) 15 (3.36%) 16 (3.33%) 14 (3.43%) Productivity features 14 (3.88%) 15 (3.89%) 14 (3.99%) 16 (3.46%) 19 (1.76%) 15 (3.71%) 14 (4.09%) 15 (3.64%) 15 (3.85%) 15 (3.81%) 16 (3.38%) 14 (4.15%) 15 (3.65%) 4 (7.36%) 15 (3.93%) 15 (4.09%) 14 (4.01%) 14 (4.19%) 14 (4.09%) 9 (5.40%) 14 (4.30%) 15 (4.13%) 14 (3.99%) 15 (3.95%) 13 (4.07%) Route name/num. identification N o t p a r t o f a b u n d l e 16 (2.77%) 17 (2.78%) 16 (2.85%) 9 (5.50%) 16 (2.84%) 17 (2.65%) 16 (2.93%) 10 (4.97%) 17 (2.76%) 17 (2.73%) 17 (2.42%) 16 (2.97%) 17 (2.61%) 18 (2.63%) 17 (2.81%) 17 (2.92%) 16 (2.87%) 16 (2.99%) 16 (2.93%) 14 (4.34%) 16 (3.07%) 12 (4.63%) 16 (2.86%) 7 (6.34%) 15 (2.91%) Reliability 13 (4.04%) 14 (4.06%) 13 (4.16%) 15 (3.61%) 13 (4.14%) 14 (3.87%) 13 (4.26%) 14 (3.79%) 14 (4.01%) 14 (3.97%) 15 (3.52%) 13 (4.33%) 14 (3.80%) 15 (3.83%) 13 (4.09%) 14 (4.26%) 13 (4.18%) 13 (4.36%) 13 (4.26%) 15 (4.03%) 13 (4.47%) 13 (4.30%) 13 (4.16%) 14 (4.11%) 12 (4.24%) Schedule span 4 (7.86%) 4 (7.90%) 4 (8.09%) 4 (7.02%) 4 (8.06%) 3 (7.52%) 3 (8.30%) 1 (9.39%) 3 (7.81%) 3 (7.74%) 6 (6.86%) 2 (8.43%) 3 (7.40%) 3 (7.45%) 3 (7.96%) 2 (8.29%) 4 (8.13%) 3 (8.49%) 3 (8.30%) 3 (7.84%) 4 (7.13%) 3 (8.38%) 4 (8.10%) 3 (8.01%) 3 (8.26%) Transit frequency 2 (9.96%) 1 (10.00 %) 1 (10.25 %) 1 (8.89%) 1 (10.21 %) 2 (9.53%) 1 (10.52 %) 2 (9.35%) 1 (9.90%) 1 (9.80%) 2 (8.69%) 5 (6.82%) 2 (9.38%) 1 (9.44%) 1 (10.09 %) 3 (7.34%) 1 (10.30 %) 1 (9.66%) 1 (10.52 %) 2 (9.93%) 1 (8.87%) 1 (10.62 %) 1 (10.26 %) 1 (10.15 %) 1 (10.47 %) Transfer distance 18 (2.74%) 9 (5.30%) 18 (2.82%) 11 (4.63%) 18 (2.80%) 19 (2.62%) 18 (2.89%) 18 (2.57%) 19 (2.72%) 19 (2.69%) 19 (2.39%) 18 (2.93%) 19 (2.58%) 20 (2.59%) 19 (2.77%) 7 (6.18%) 18 (2.83%) 18 (1.79%) 18 (2.89%) 18 (2.73%) 18 (3.03%) 18 (2.92%) 18 (2.82%) 18 (2.79%) 17 (2.87%) Station/stop distance 1 (11.76 %) 2 (8.82%) 6 (6.67%) 7 (5.79%) 2 (8.67%) 5 (6.20%) 5 (6.84%) 7 (6.08%) 5 (6.44%) 6 (6.38%) 8 (5.65%) 4 (6.95%) 6 (6.10%) 6 (6.14%) 5 (6.56%) 5 (6.83%) 6 (6.70%) 5 (7.00%) 5 (6.84%) 5 (6.46%) 3 (7.18%) 14 (4.27%) 2 (9.12%) 5 (6.60%) 5 (6.81%) Parking distance - - - - - - - - 2 (8.41%) 5 (6.79%) 3 (8.27%) - - 17 (2.63%) - - - - - - - - - - - Ease of boarding - - - - 15 (3.12%) 9 (5.03%) - - - - - - - - - - - - - - - - - - - Fare machines - - - 17 (3.16%) - - - - - - - 9 (5.09%) 9 (5.43%) 14 (3.95%) 14 (4.08%) - - - - - - - - - 19 (2.67%)

TABLE D-11. MaxDiff model estimation results for commute trips (t-stat)—Charlotte. LOS/Attribute Attribute Bundle Coefficient (t-stat) Individual Demographics Variables Household Demographics Variables Trip Characteristics Attitude Full-time student (base: part- time student/ not student) Employment status (base: self- employed/part-time employed/not employed) Female (base: male) Long- time resident (> 5 years, base: short-time resident) Has mobility problem (base: no mobility problem) Age (base: age < 35) Number of vehicles in the HH (base: no vehicles) Family income (base: income < 35k) Presence of kid(s) in the HH (base: no kids in the HH) Group travel (Actual distance /mean distance) Pro- Transit Attitude Environment, Productivity, and Time Savings Pro-Car Attitude Transit Averse Low Transit Comfort Level Willingness to Walk Full-time employed Retired 35 ≤ Age < 55 Age ≥ 55 Number of vehs in the HH = 1 Number of vehs in the HH = 2 Number of vehs in the HH = 3+ 35k ≤ Income < 50k 50k ≤ Income < 75k 75k ≤ Income < 100k Income ≥ 100k Willing to walk ≤ 2 min. Willing to walk ≥ 10 min. IVTT - -0.082 (-17.44) Fare - -0.991 (-9.15) 0.154 (1.74) Real-time info S t a t i o n / s t o p d e s i g n f e a t u r e s 0.959 (14.71) #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A Station/stop security 1.34 (17.33) #N/A #N/A #N/A #N/A #N/A #N/A #N/A 0.547 (2.88) #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A 0.186 (2.17) #N/A #N/A 0.206 (2.2) #N/A #N/A Station/stop Lighting/safety 0.891 (6.8) #N/A #N/A #N/A 0.452 (3.09) 0.564 (3.84) #N/A #N/A 0.394 (2.05) #N/A #N/A #N/A 0.512 (2.35) #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A 0.264 (2.24) 0.204 (2.1) #N/A #N/A Station/stop shelter #N/A #N/A 0.443 (2.64) #N/A 0.33 (2.28) #N/A #N/A 0.27 (1.9) #N/A 0.873 (3.93) 0.859 (4.19) 0.739 (3.24) #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A Proximity to services 0.858 (8.02) #N/A #N/A #N/A 0.291 (2.13) #N/A #N/A #N/A #N/A #N/A #N/A #N/A -0.582 (- 3.04) #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A 0.203 (2.45) #N/A #N/A #N/A #N/A Cleanliness of station/stop 0.934 (4.95) #N/A #N/A #N/A #N/A #N/A #N/A #N/A 0.539 (2.77) #N/A #N/A #N/A 0.64 (2.33) 1.16 (4.7) 0.781 (3.14) 0.644 (2.91) #N/A #N/A #N/A #N/A #N/A #N/A 0.321 (2.8) #N/A #N/A #N/A Station/stop benches 0.982 (8.28) #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A -0.314 (- 3.19) #N/A #N/A #N/A #N/A #N/A #N/A #N/A On-board seating availability O n - b o a r d f e a t u r e s 1.44 (17.08) #N/A #N/A #N/A #N/A #N/A #N/A #N/A 0.537 (2.8) #N/A #N/A #N/A #N/A 0.465 (2.66) #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A On-board seating comfort 2.36 (3.07) -0.587 (-2.01) #N/A #N/A #N/A 0.39 (2.89) #N/A #N/A #N/A -1.82 (-2.36) -1.81 (-2.34) -2.02 (-2.59) #N/A #N/A #N/A #N/A #N/A -0.467 (-2.22) #N/A #N/A 0.193 (2.36) #N/A #N/A 0.278 (3.11) #N/A #N/A On-board temperature 0.937 (7.05) #N/A #N/A #N/A #N/A #N/A #N/A 0.415 (2.73) 0.476 (2.33) #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A 0.27 (1.95) Cleanliness of transit vehicle 0.66 (10.14) #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A 0.206 (2.37) #N/A #N/A Productivity features 0.752 (5.77) #N/A #N/A #N/A -0.257 (-1.92) #N/A #N/A 0.302 (2.25) #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A 0.354 (2.45) #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A Route name/num. identification N o t p a r t o f a b u n d l e 0.671 (4.52) #N/A #N/A #N/A 0.276 (2.04) #N/A #N/A 0.306 (2.11) 0.579 (3) #N/A #N/A #N/A 0.44 (2.16) 0.397 (2.43) #N/A #N/A #N/A #N/A #N/A #N/A 0.196 (2.45) #N/A #N/A #N/A #N/A #N/A Reliability 0.798 (7.45) #N/A #N/A #N/A 0.28 (2.06) #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A -0.238 (-2.86) #N/A #N/A #N/A #N/A Schedule span 0.856 (7.76) #N/A #N/A #N/A 0.307 (2.22) #N/A #N/A #N/A 0.432 (2.32) #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A -0.172 (- 2.05) -0.217 (-1.97) #N/A #N/A #N/A Transit frequency #N/A #N/A #N/A #N/A 0.313 (2.39) 0.284 (2.06) #N/A #N/A #N/A 0.508 (3.33) #N/A #N/A 0.838 (3.67) 1.07 (5.8) 0.708 (4) 1.02 (6.64) #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A Transfer distance 1.32 (12.6) #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A -0.292 (- 2.06) -0.576 (- 2.62) #N/A #N/A #N/A #N/A #N/A 0.172 (1.92) #N/A -0.384 (- 2.87) Station/stop distance 1.24 (8.98) #N/A #N/A #N/A 0.381 (2.62) #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A 0.532 (2.86) #N/A 0.414 (2.6) #N/A #N/A #N/A #N/A 0.216 (2.43) -0.265 (- 3.01) -0.247 (-2.17) #N/A #N/A #N/A Parking distance 1.06 (5.77) #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A 0.638 (2.33) 0.686 (2.87) 0.568 (2.32) 0.492 (2.26) #N/A #N/A #N/A #N/A 0.233 (2.71) #N/A #N/A #N/A #N/A #N/A Ease of boarding #N/A #N/A #N/A #N/A #N/A #N/A 1.13 (3.11) #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A 0.166 (1.89) 0.682 (4.06) #N/A Fare machines 0.366 (2.16) #N/A #N/A #N/A 0.292 (2.06) #N/A #N/A #N/A 0.415 (2.28) #N/A 0.652 (4.07) 0.671 (3.44) 0.458 (2.04) 0.372 (2.09) 0.413 (2.12) #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A #N/A

TABLE D-12. MaxDiff model estimation results for non-commute trips (t-stat)—Charlotte. LOS/Attribute Attribute Bundle Coefficient (t-stat) Individual Demographics variables Household Demographics variables Trip Characteristics Attitude Full- time student (base: part-time student/n ot student) Employment status (base: self- employed/part-time employed/not employed) Female (base: male) Long- time resident (> 5 years, base: short-time resident) Has mobility problem (base: no mobility problem) Age (base: age < 35) Number of vehicles in the HH (base: no vehicles) Family income (base: income < 35k) Presen ce of kid(s) in the HH (base: no kids in the HH) Group travel (Actual distance /mean distance) Pro- Transit Attitude Environ- ment, Produc- tivity, and Time Savings Pro-Car Attitude Tran sit Aver se Low Transit Comfort Level Willingness to Walk Full-time employed Retired 35 ≤ Age < 55 Age ≥ 55 Numb er of vehs in the HH = 1 Number of vehs in the HH = 2 Numb er of vehs in the HH = 3+ 35k ≤ Income < 50k 50k ≤ Income < 75k 75k ≤ Income < 100k Income ≥ 100k Willing to walk ≤ 2 min. Willing to walk ≥ 10 min. IVTT - -0.038 (-4.97) Fare - -1.4 (-9.37) 0.456 (1.97) 0.307 (2.92) Real-time info S t a t i o n / s t o p d e s i g n f e a t u r e s 1.13 (11.31) -0.322 (- 2.68) Station/stop security 1.24 (6.58) 0.661 (1.97) 0.512 (2.27) Station/stop lighting/safety 1.18 (4.15) 0.475 (1.97) 0.798 (3.03) 1.22 (4.03) 1.05 (3.04) 0.551 (1.91) -0.394 (-2.69) 0.5 (2.85) -1.13 (-3.09) -0.561 (-2.16) Station/stop shelter 1.47 (9.86) 0.446 (2.05) Proximity to services 0.927 (8.96) 0.278 (2.43) Cleanliness of station/stop 1.84 (15.93) 0.269 (2.04) Station/stop benches 0.462 (3.45) 0.467 (2.28) On-board seating availability O n - b o a r d f e a t u r e s 1.52 (8.42) 0.824 (3.18) -0.523 (-2.08) 0.486 (2.21) On-board seating comfort 0.84 (5.36) -0.455 (-2.26) On-board temperature 1.36 (12.58) Cleanliness of transit vehicle 0.709 (6.98) Productivity features 1.34 (7.24) 1.19 (2.46) -0.627 (-2.97) -0.677 (-2.95) Route name/num. identification N o t p a r t o f a b u n d l e 1.09 (9.82) 0.665 (2.16) Reliability 0 (0) 0.586 (3.02) 0.665 (3.66) 0.828 (3.8) Schedule span 1.3 (10.15) 0.627 (2.5) -0.497 (-3.72) -0.581 (-3.42) 0.393 (2.54) Transit frequency 1.11 (5.94) -0.606 (-1.94) 0.456 (1.86) 0.834 (2.8) Transfer distance 1.47 (8.73) 0.625 (2.08) -0.55 (-2.62) Station/stop distance 1.74 (9.38) 0.539 (2.45) -0.433 (-1.99) Parking distance 0.839 (4.56) 0.826 (3.39) 1.1 (3.85) -0.265 (-1.91) -0.297 (-2.18) 0.295 (1.91) Ease of boarding 0 (0) 0.63 (4.74) 0.595 (2.08) Fare machines 0 (0) 0.727 (3.59) 0.995 (2.96) 0.745 (4.43) 0.445 (2.34) 0.389 (2.09)

TABLE D-13. Relative rank of attributes by individual demographics, household demographics, trip characteristics, and attitudinal variables (relative importance in %)—Charlotte (commute trips). Attribute Attribute Bundle Variables Individual Demographics Household Demographics Trip Characteristics Attitude Willingness to Walk Full-time student (base: part- time student/not student) Employment status (base: self- employed/part-time employed/ not employed) Female (base: male) Long- time resident (> 5 years, base: short- time resident) Has mobility problem (base: no mobility problem) Age (base: age < 35) Number of vehicles in the HH (base: no vehicles) Family income (base: income < 35k) Presenc e of kid(s) in the HH (base: no kids in the HH) Group travel (Actual distance/ mean distance) Pro- Transit Attitude Environment, Productivity, and Time Savings Pro-Car Attitude Transit Averse Low Transit Comfort Level Willing to walk ≤ 2 min. Willing to walk ≥ 10 min. Full-time employed Retired 35 ≤ Age < 55 Age ≥ 55 Number of vehs in the HH = 1 Number of vehs in the HH = 2 Number of vehs in the HH = 3+ 35k ≤ Income < 50k 50k ≤ Income < 75k 75k ≤ Income < 100k Income ≥ 100k Real-time info S t a t i o n / s t o p d e s i g n f e a t u r e s 8 (5.38%) 8 (5.08%) 8 (5.21%) 12 (4.55%) 9 (4.88%) 9 (4.90%) 11 (4.86%) 13 (4.29%) 7 (5.33%) 8 (5.29%) 8 (5.38%) 11 (4.49%) 11 (4.15%) 9 (4.59%) 10 (4.57%) 9 (5.19%) 7 (5.52%) 7 (5.30%) 8 (5.21%) 8 (4.93%) 9 (5.34%) 10 (5.17%) 9 (4.88%) 8 (5.02%) 8 (5.24%) Station/stop security 3 (7.51%) 3 (7.10%) 3 (7.27%) 5 (6.35%) 4 (6.82%) 3 (6.85%) 4 (6.80%) 3 (8.45%) 2 (7.45%) 2 (7.39%) 2 (7.52%) 6 (6.27%) 6 (5.80%) 5 (6.41%) 6 (6.38%) 3 (7.25%) 3 (7.71%) 3 (7.40%) 3 (7.27%) 2 (7.85%) 3 (7.46%) 3 (7.23%) 2 (7.87%) 3 (7.01%) 3 (7.32%) Station/stop lighting/safety 11 (5.00%) 11 (4.72%) 11 (4.84%) 4 (6.37%) 2 (7.40%) 12 (4.56%) 13 (4.52%) 8 (5.75%) 10 (4.95%) 11 (4.92%) 11 (5.00%) 5 (6.57%) 13 (3.86%) 11 (4.26%) 12 (4.24%) 12 (4.82%) 10 (5.13%) 10 (4.92%) 11 (4.84%) 11 (4.58%) 12 (4.96%) 6 (6.23%) 6 (5.57%) 11 (4.66%) 11 (4.87%) Station/stop shelter - 18 (2.35%) - 19 (1.56%) - - 19 (1.37%) - 11 (4.85%) 12 (4.74%) 16 (4.15%) - - - - - - - - - - - - - - Proximity to services 12 (4.81%) 12 (4.55%) 12 (4.66%) 8 (5.45%) 12 (4.36%) 13 (4.39%) 14 (4.35%) 14 (3.84%) 12 (4.77%) 13 (4.73%) 12 (4.82%) 19 (1.29%) 14 (3.71%) 12 (4.11%) 13 (4.09%) 13 (4.64%) 11 (4.94%) 11 (4.74%) 12 (4.66%) 13 (4.41%) 5 (5.91%) 12 (4.63%) 13 (4.37%) 12 (4.49%) 12 (4.69%) Cleanliness of station/stop 10 (5.24%) 10 (4.95%) 10 (5.07%) 15 (4.43%) 11 (4.75%) 11 (4.78%) 12 (4.74%) 4 (6.59%) 9 (5.19%) 10 (5.15%) 10 (5.24%) 3 (7.37%) 2 (9.06%) 2 (8.21%) 3 (7.52%) 11 (5.05%) 9 (5.37%) 9 (5.16%) 10 (5.07%) 10 (4.80%) 11 (5.20%) 5 (6.77%) 11 (4.75%) 10 (4.89%) 10 (5.10%) Station/stop benches 7 (5.51%) 7 (5.20%) 7 (5.33%) 11 (4.66%) 8 (4.99%) 8 (5.02%) 9 (4.98%) 12 (4.40%) 6 (5.46%) 7 (5.42%) 7 (5.51%) 10 (4.60%) 10 (4.25%) 8 (4.70%) 9 (4.68%) 8 (5.31%) 6 (5.65%) 16 (3.69%) 7 (5.33%) 7 (5.05%) 7 (5.47%) 9 (5.30%) 8 (5.00%) 7 (5.14%) 7 (5.36%) On-board aeating availability O n - b o a r d f e a t u r e s 2 (8.07%) 2 (7.63%) 2 (7.82%) 3 (6.83%) 3 (7.32%) 2 (7.36%) 2 (7.30%) 2 (8.85%) 1 (8.01%) 1 (7.94%) 1 (8.08%) 4 (6.74%) 3 (8.24%) 4 (6.89%) 5 (6.86%) 2 (7.79%) 2 (8.28%) 2 (7.95%) 2 (7.82%) 4 (7.40%) 2 (8.02%) 2 (7.76%) 4 (7.33%) 2 (7.54%) 2 (7.86%) On-board seating comfort 1 (9.94%) 1 (12.51%) 1 (12.81% ) 1 (11.19% ) 1 (13.99%) 1 (12.07%) 1 (11.97% ) 1 (10.56% ) 18 (3.00%) 19 (3.03%) 19 (1.91%) 1 (11.04% ) 1 (10.21% ) 1 (11.30% ) 1 (11.24% ) 1 (12.77%) 1 (10.89% ) 1 (13.03%) 1 (12.81% ) 1 (13.13%) 1 (13.15% ) 1 (12.73% ) 1 (13.42% ) 1 (12.35% ) 1 (12.89% ) On-board temperature 9 (5.25%) 9 (4.97%) 9 (5.09%) 14 (4.44%) 10 (4.77%) 10 (4.79%) 3 (6.86%) 5 (6.32%) 8 (5.21%) 9 (5.17%) 9 (5.26%) 12 (4.39%) 12 (4.06%) 10 (4.48%) 11 (4.46%) 10 (5.07%) 8 (5.39%) 8 (5.17%) 9 (5.09%) 9 (4.82%) 10 (5.22%) 11 (5.05%) 10 (4.77%) 9 (4.90%) 5 (6.59%) Cleanliness of transit vehicle 17 (3.70%) 17 (3.50%) 17 (3.58%) 16 (3.13%) 17 (3.36%) 18 (3.38%) 17 (3.35%) 18 (2.95%) 17 (3.67%) 18 (3.64%) 18 (3.70%) 18 (3.09%) 19 (2.86%) 19 (3.16%) 18 (3.14%) 17 (3.57%) 17 (3.80%) 17 (3.64%) 17 (3.58%) 17 (3.39%) 16 (3.68%) 16 (3.56%) 12 (4.41%) 18 (3.45%) 17 (3.60%) Productivity features 15 (4.22%) 15 (3.99%) 15 (4.08%) 18 (2.35%) 15 (3.82%) 16 (3.85%) 8 (5.35%) 17 (3.37%) 15 (4.18%) 16 (4.15%) 15 (4.22%) 17 (3.52%) 17 (3.25%) 16 (3.60%) 16 (3.58%) 5 (5.98%) 14 (4.33%) 14 (4.15%) 15 (4.08%) 16 (3.87%) 13 (4.19%) 14 (4.06%) 16 (3.83%) 15 (3.94%) 15 (4.11%) Route name/num. identification N o t p a r t o f a b u n d l e 16 (3.76%) 16 (3.56%) 16 (3.64%) 13 (4.49%) 16 (3.41%) 17 (3.43%) 10 (4.96%) 9 (5.59%) 16 (3.73%) 17 (3.70%) 17 (3.77%) 9 (5.20%) 9 (4.62%) 18 (3.21%) 17 (3.20%) 16 (3.63%) 16 (3.86%) 15 (3.71%) 16 (3.64%) 12 (4.46%) 15 (3.74%) 15 (3.62%) 17 (3.41%) 17 (3.51%) 16 (3.66%) Reliability 14 (4.47%) 14 (4.23%) 14 (4.33%) 9 (5.11%) 14 (4.06%) 15 (4.08%) 16 (4.05%) 15 (3.57%) 14 (4.44%) 15 (4.40%) 14 (4.48%) 16 (3.73%) 16 (3.45%) 14 (3.82%) 15 (3.80%) 15 (4.32%) 13 (4.59%) 13 (4.41%) 14 (4.33%) 15 (4.10%) 17 (3.12%) 13 (4.30%) 15 (4.06%) 14 (4.18%) 14 (4.36%) Schedule span 13 (4.80%) 13 (4.54%) 13 (4.65%) 7 (5.51%) 13 (4.35%) 14 (4.38%) 15 (4.34%) 7 (5.76%) 13 (4.76%) 14 (4.72%) 13 (4.81%) 13 (4.01%) 15 (3.70%) 13 (4.10%) 14 (4.08%) 14 (4.63%) 12 (4.92%) 12 (4.73%) 13 (4.65%) 14 (4.40%) 14 (3.81%) 17 (3.45%) 14 (4.35%) 13 (4.48%) 13 (4.68%) Transit frequency - - - 20 (1.48%) 19 (1.44%) - - - 19 (2.82%) - - 14 (3.92%) 8 (4.63%) 17 (3.39%) 8 (4.86%) - - - - - - - - - - Transfer distance 4 (7.40%) 4 (7.00%) 4 (7.16%) 6 (6.26%) 5 (6.71%) 4 (6.75%) 5 (6.69%) 6 (5.91%) 3 (7.34%) 3 (7.28%) 3 (7.41%) 7 (6.18%) 7 (5.71%) 6 (6.32%) 7 (6.29%) 7 (5.56%) 15 (4.28%) 4 (7.29%) 4 (7.16%) 5 (6.79%) 4 (7.35%) 4 (7.12%) 3 (7.59%) 4 (6.91%) 9 (5.11%) Station/stop distance 5 (6.95%) 5 (6.57%) 5 (6.73%) 2 (7.69%) 6 (6.31%) 5 (6.34%) 6 (6.29%) 10 (5.55%) 4 (6.89%) 4 (6.84%) 4 (6.96%) 8 (5.80%) 4 (7.67%) 7 (5.93%) 2 (7.88%) 4 (6.71%) 4 (7.13%) 5 (6.85%) 5 (6.73%) 3 (7.49%) 8 (5.43%) 8 (5.35%) 5 (6.31%) 5 (6.49%) 4 (6.77%) Parking distance 6 (5.94%) 6 (5.62%) 6 (5.75%) 10 (5.03%) 7 (5.39%) 7 (5.42%) 7 (5.38%) 11 (4.74%) 5 (5.89%) 5 (5.85%) 5 (5.95%) 2 (7.95%) 5 (7.56%) 3 (7.79%) 4 (7.39%) 6 (5.73%) 5 (6.10%) 6 (5.85%) 6 (5.75%) 6 (6.65%) 6 (5.90%) 7 (5.72%) 7 (5.39%) 6 (5.55%) 6 (5.79%) Ease of boarding - - - - - 6 (5.78%) - - - - - - - - - - - - - - - - 19 (0.84%) 16 (3.57%) - Fare machines 18 (2.05%) 19 (1.94%) 18 (1.99%) 17 (3.12%) 18 (1.86%) 19 (1.87%) 18 (1.86%) 16 (3.50%) 20 (2.04%) 6 (5.62%) 6 (5.82%) 15 (3.86%) 18 (3.19%) 15 (3.73%) 19 (1.74%) 18 (1.98%) 18 (2.11%) 18 (2.02%) 18 (1.99%) 18 (1.88%) 18 (2.04%) 18 (1.97%) 18 (1.86%) 19 (1.92%) 18 (2.00%)

TABLE D-14. Relative rank of attributes by individual demographics, household demographics, trip characteristics, and attitudinal variables (relative importance in %)—Charlotte (non-commute trips). Attribute Attribute Bundle Variables Individual Demographics Household Demographics Trip Characteristics Attitude Willingness to Walk Full-time student (base: part- time student/not student) Employment status (base: self- employed/part-time employed/not employed) Female (base: male) Long- time resident (> 5 years, base: short- time resident) Has mobility problem (base: no mobility problem) Age (base: age < 35) Number of vehicles in the HH (base: no vehicles) Family income (base: income < 35k) Presence of kid(s) in the HH (base: no kids in the HH) Group travel (Actual distance/ mean distance) Pro- Transit Attitude Environment, Productivity, and Time Savings Pro- Car Attitude Transit Averse Low Transit Comfort Level Willing to walk ≤ 2 min. Willing to walk ≥ 10 min. Full-time employed Retired 35 ≤ Age < 55 Age ≥ 55 Number of vehs in the HH = 1 Number of vehs in the HH = 2 Number of vehs in the HH = 3+ 35k ≤ Income < 50k 50k ≤ Income < 75k 75k ≤ Income < 100k Income ≥ 100k Real-time info S t a t i o n / s t o p d e s i g n f e a t u r e s 11 (4.97%) 11 (4.84%) 11 (4.87%) 10 (4.95%) 11 (5.03%) 11 (5.24%) 13 (4.56%) 12 (4.29%) 11 (5.24%) 11 (5.11%) 11 (5.24%) 11 (5.00%) 10 (5.37%) 12 (4.94%) 11 (5.11%) 11 (5.24%) 11 (5.04%) 11 (5.24%) 10 (5.40%) 12 (5.11%) 14 (3.95%) 10 (5.26%) 12 (5.08%) 10 (5.53%) 9 (5.77%) Station/stop security 9 (5.45%) 10 (5.31%) 1 (8.19%) 2 (7.68%) 9 (5.52%) 9 (5.75%) 12 (5.00%) 11 (4.71%) 9 (5.75%) 9 (5.61%) 9 (5.75%) 10 (5.48%) 8 (5.89%) 10 (5.43%) 10 (5.61%) 9 (5.75%) 9 (5.53%) 9 (5.75%) 9 (5.93%) 9 (5.61%) 8 (6.06%) 9 (5.77%) 9 (5.57%) 9 (6.07%) 8 (6.34%) Station/stop lighting/safety 10 (5.19%) 3 (7.09%) 10 (5.08%) 9 (5.17%) 10 (5.25%) 10 (5.47%) 1 (7.98%) 1 (9.12%) 10 (5.47%) 10 (5.34%) 10 (5.47%) 1 (9.86%) 9 (5.61%) 11 (5.16%) 3 (7.83%) 10 (5.47%) 10 (5.26%) 10 (5.47%) 15 (3.76%) 11 (5.34%) 9 (5.77%) 3 (7.82%) 10 (5.30%) 18 (0.24%) 16 (3.16%) Station/stop shelter 5 (6.46%) 5 (6.29%) 5 (6.33%) 5 (6.44%) 1 (8.53%) 4 (6.82%) 7 (5.93%) 8 (5.58%) 4 (6.82%) 4 (6.65%) 4 (6.82%) 5 (6.50%) 3 (6.99%) 6 (6.43%) 5 (6.65%) 4 (6.82%) 4 (6.55%) 4 (6.82%) 4 (7.03%) 4 (6.65%) 4 (7.19%) 5 (6.84%) 5 (6.61%) 4 (7.19%) 3 (7.51%) Proximity to services 14 (4.07%) 14 (3.97%) 15 (3.99%) 13 (4.06%) 14 (4.13%) 13 (4.30%) 16 (3.74%) 14 (3.52%) 14 (4.30%) 14 (4.19%) 14 (4.30%) 14 (4.10%) 14 (4.41%) 14 (4.06%) 14 (4.19%) 14 (4.30%) 14 (4.13%) 14 (4.30%) 13 (4.43%) 10 (5.45%) 12 (4.53%) 13 (4.31%) 15 (4.17%) 13 (4.54%) 12 (4.74%) Cleanliness of station/stop 2 (8.09%) 1 (7.88%) 2 (7.92%) 1 (8.06%) 2 (8.19%) 1 (8.54%) 2 (7.43%) 6 (6.99%) 1 (8.53%) 2 (8.32%) 1 (8.53%) 2 (8.14%) 1 (8.74%) 2 (8.05%) 1 (8.32%) 1 (8.53%) 2 (8.20%) 1 (8.53%) 1 (8.80%) 1 (9.54%) 1 (9.00%) 1 (8.56%) 1 (8.27%) 1 (9.00%) 1 (9.40%) Station/stop benches 18 (2.03%) 20 (1.98%) 19 (1.99%) 20 (2.02%) 18 (2.06%) 19 (2.14%) 15 (3.75%) 19 (1.76%) 18 (2.14%) 18 (2.09%) 18 (2.14%) 18 (2.04%) 18 (2.20%) 18 (2.02%) 18 (2.09%) 18 (2.14%) 18 (2.06%) 18 (2.14%) 18 (2.21%) 18 (2.09%) 18 (2.26%) 18 (2.15%) 18 (2.08%) 17 (2.26%) 17 (2.36%) On-board seating availability O n - b o a r d f e a t u r e s 4 (6.68%) 4 (6.51%) 4 (6.55%) 4 (6.66%) 4 (6.77%) 3 (7.05%) 6 (6.13%) 2 (8.90%) 3 (7.05%) 3 (6.88%) 3 (7.05%) 4 (6.72%) 13 (4.74%) 5 (6.65%) 4 (6.87%) 3 (7.05%) 1 (8.94%) 3 (7.05%) 3 (7.27%) 3 (6.87%) 3 (7.43%) 4 (7.07%) 4 (6.83%) 3 (7.44%) 2 (7.77%) On-board seating comfort 15 (3.69%) 15 (3.60%) 16 (3.62%) 14 (3.68%) 15 (3.74%) 14 (3.90%) 17 (3.39%) 15 (3.19%) 15 (3.89%) 15 (3.80%) 15 (3.89%) 15 (3.71%) 15 (3.99%) 15 (3.68%) 15 (3.80%) 15 (3.89%) 15 (3.74%) 15 (3.89%) 14 (4.02%) 15 (3.80%) 13 (4.11%) 14 (3.91%) 16 (3.77%) 14 (4.11%) 18 (1.97%) On-board temperature 7 (5.98%) 7 (5.82%) 7 (5.86%) 7 (5.96%) 6 (6.06%) 6 (6.31%) 9 (5.49%) 10 (5.17%) 6 (6.31%) 6 (6.15%) 6 (6.31%) 7 (6.01%) 5 (6.46%) 7 (5.95%) 7 (6.15%) 6 (6.31%) 6 (6.06%) 6 (6.31%) 6 (6.50%) 6 (6.15%) 6 (6.65%) 7 (6.33%) 7 (6.11%) 6 (6.65%) 4 (6.95%) Cleanliness of transit vehicle 17 (3.12%) 18 (3.04%) 18 (3.05%) 18 (3.11%) 17 (3.16%) 16 (3.29%) 18 (2.86%) 17 (2.69%) 17 (3.29%) 17 (3.21%) 17 (3.29%) 17 (3.13%) 17 (3.37%) 17 (3.10%) 17 (3.21%) 17 (3.29%) 17 (3.16%) 17 (3.29%) 16 (3.39%) 17 (3.21%) 16 (3.47%) 17 (3.30%) 17 (3.19%) 16 (3.47%) 15 (3.62%) Productivity features 1 (11.12%) 8 (5.74%) 8 (5.77%) 17 (3.12%) 7 (5.97%) 7 (6.22%) 10 (5.41%) 18 (2.52%) 7 (6.21%) 7 (6.06%) 7 (6.21%) 8 (5.92%) 6 (6.37%) 8 (5.86%) 8 (6.06%) 7 (6.21%) 7 (5.97%) 7 (6.21%) 7 (6.41%) 7 (6.06%) 7 (6.55%) 8 (6.24%) 8 (6.02%) 7 (6.56%) 5 (6.85%) Route name/num. identification N o t p a r t o f a b u n d l e 13 (4.79%) 13 (4.67%) 13 (4.69%) 12 (4.78%) 13 (4.85%) 12 (5.06%) 14 (4.40%) 13 (4.14%) 13 (5.05%) 13 (4.93%) 13 (5.05%) 13 (4.82%) 12 (5.18%) 3 (7.68%) 13 (4.93%) 13 (5.05%) 13 (4.86%) 13 (5.05%) 12 (5.21%) 14 (4.93%) 11 (5.33%) 12 (5.07%) 14 (4.90%) 12 (5.33%) 11 (5.57%) Reliability - 19 (2.51%) - - - - 19 (2.68%) 16 (3.15%) - - - - - - - - - - - - - - - - - Schedule span 8 (5.71%) 9 (5.57%) 9 (5.60%) 8 (5.70%) 8 (5.79%) 8 (6.03%) 11 (5.25%) 5 (7.32%) 8 (6.03%) 8 (5.88%) 8 (6.03%) 9 (5.75%) 7 (6.18%) 9 (5.69%) 9 (5.88%) 8 (6.03%) 8 (5.79%) 8 (6.03%) 8 (6.22%) 8 (5.88%) 15 (3.93%) 16 (3.35%) 3 (7.61%) 8 (6.36%) 7 (6.64%) Transit frequency 12 (4.88%) 12 (4.75%) 12 (4.78%) 11 (4.86%) 12 (4.94%) 18 (2.34%) 5 (6.32%) 3 (7.39%) 12 (5.15%) 12 (5.02%) 12 (5.15%) 12 (4.91%) 11 (5.27%) 13 (4.86%) 12 (5.02%) 12 (5.15%) 12 (4.95%) 12 (5.15%) 11 (5.31%) 13 (5.02%) 10 (5.43%) 11 (5.17%) 13 (4.99%) 11 (5.43%) 10 (5.67%) Transfer distance 5 (6.46%) 5 (6.29%) 5 (6.33%) 5 (6.44%) 5 (6.55%) 4 (6.82%) 7 (5.93%) 8 (5.58%) 4 (6.82%) 4 (6.65%) 4 (6.82%) 5 (6.50%) 3 (6.99%) 1 (9.17%) 5 (6.65%) 4 (6.82%) 4 (6.55%) 4 (6.82%) 4 (7.03%) 4 (6.65%) 4 (7.19%) 5 (6.84%) 5 (6.61%) 4 (7.19%) 13 (4.70%) Station/stop distance 3 (7.65%) 2 (7.45%) 3 (7.49%) 3 (7.62%) 3 (7.75%) 2 (8.07%) 3 (7.02%) 7 (6.61%) 2 (8.07%) 1 (10.31%) 2 (8.07%) 3 (7.69%) 2 (8.27%) 4 (7.61%) 2 (7.87%) 2 (8.07%) 3 (7.75%) 2 (8.07%) 2 (8.32%) 2 (7.87%) 2 (8.51%) 2 (8.10%) 2 (7.82%) 2 (8.51%) 6 (6.68%) Parking distance 16 (3.69%) 16 (3.59%) 17 (3.61%) 15 (3.68%) 16 (3.74%) 15 (3.89%) 4 (6.72%) 4 (7.37%) 16 (3.89%) 16 (3.80%) 16 (3.89%) 16 (3.71%) 16 (3.99%) 16 (3.67%) 16 (3.79%) 16 (3.89%) 16 (3.74%) 16 (3.89%) 17 (2.75%) 16 (3.79%) 17 (2.65%) 15 (3.90%) 11 (5.10%) 15 (4.11%) 14 (4.29%) Ease of boarding - - - 19 (2.76%) - 17 (2.76%) - - - - - - - - - - - - - - - - - - - Fare machines - 17 (3.11%) 14 (4.28%) 16 (3.26%) 19 (1.98%) - - - - - - - - - - - 19 (1.73%) - - - - - - - -

Transit Service Attribute Models D-33 Summary of Key Findings on Transit Attributes TABLE D-15 shows a comparison of scaled values of premium transit attributes across all three cities where surveys were conducted. The values are in equivalent minutes of in-vehicle travel time and coefficients from which these values have been derived were estimated for bus and train modes combined. For Chicago and Charlotte, the values shown are averages of bus and train values. TABLE D-15. Scaled equivalent minutes of in-vehicle travel time for premium service attributes. Attribute CommuteTrips Non CommuteTrips Charloe Salt Lake City Chicago Charloe Salt Lake City Chicago Station/stop design features bundle 3.71 4.61 4.97 -9.05 1.57 4.42 Real-time info 0.40 * 0.62 -1.06 * 0.44 Station/stop security 0.60 0.88 0.85 -1.56 0.22 0.84 Station/stop lighting/safety 0.66 0.88 0.86 -1.62 0.20 0.82 Station/stop shelter 0.64 1.10 0.86 -1.57 0.37 0.69 Proximity to services 0.40 0.84 0.40 -0.89 0.47 0.50 Cleanliness of station/stop 0.73 0.42 0.90 -1.74 0.15 0.86 Station/stop benches 0.28 0.49 0.48 -0.62 0.16 0.27 On-board features bundle 4.58 3.53 5.84 -9.47 3.8 10.79 On-board seating availability 1.46 1.23 2.15 -3.32 1.41 4.09 On-board seating comfort 0.56 0.51 0.77 -1.02 0.41 1.39 On-board temperature 1.20 0.81 1.41 -2.42 0.85 2.41 Cleanliness of transit vehicle 0.60 0.44 0.64 -1.26 0.39 1.56 Productivity features 0.76 0.54** 0.87 -1.45 0.74** 1.34 Unbundled features 8.94 *** 11.17 -10.61 *** 9.77 Route name/number Identification 0.57 0.63 -1.23 0.61 Reliability 4.59 5.64 - 4.63 Schedule span* 0.52 0.77 -1.47 0.82 Transit frequency 0.60 0.82 -1.49 0.71 Transfer distance 0.46 0.56 -1.29 0.48 Station/stop distance 0.80 0.92 -1.76 0.84 Parking distance 0.72 0.84 -1.44 0.71 Ease of boarding 0.08 0.21 -0.52 0.25 Fare machines 0.60 0.78 -1.40 0.72 All premium service features 17.23 8.14 21.98 29.13 5.37 24.98 *The attribute was not part of station/stop design features bundle in the survey for Salt Lake City. ** The attribute was referred simply as “Wi-Fi” in the survey for Salt Lake City. ***The scaling process was not applied for unbundled features in Salt Lake City during Phase 1. have been presented separately for both commute and non-commute trips. In the case of Salt Lake City, the

D-34 Characteristics of premium Transit Services that Affect Choice of Mode There are key findings derived from this comparison that are useful for planning purposes: Reliability and on-board seating availability have relatively high values irrespective of trip purpose or geographic location. In terms of size, Chicago and Charlotte can be considered as ends of the spectrum. Salt Lake City may fall somewhere in the middle. The size of a city does not necessarily influence the importance of premium service characteristics consistently. There appears to be not much variation in the values of premium transit attributes for commute trips. At an aggregate level, both station/stop design features and on-board features bundles have similar values for commuters across all three cities which fall in a relatively narrow range between 4 and 6 minutes. On the other hand, for non-commute trips, there seems to be some differences in the values attributes for Chicago and Charlotte are more closely aligned (10 to 11 minutes combined); Salt Lake City did not have these attributes scaled in the analysis. Overall, the premium service characteristics account for 17 to 29 minutes of in-vehicle travel time in Chicago and Charlotte. Of this amount, 8 to 18 minutes are accounted for with station and on-board amenities. In Salt Lake City, 5 to 8 minutes (depending on trip purpose) of in-vehicle time are associated with station and on-board amenities. The set of attributes having relatively high scaled values was compared to the corresponding set with unscaled values (see TABLE D-1). It was observed that the relative importance of transit attributes could change after the scaling process. Specifically, reliability and on-board temperature which were not relatively important according to unscaled values were found to be quite important in the scaled version. This shows the importance of scaling of the attribute values in which they are appropriately traded-off against travel times and costs. The scaling process makes the attribute values more realistic in the broader perspective so that they can be used in mode choice modeling and planning. of transit attributes based on city size. The values for station amenities range from 2 to 9 minutes and the values for on-board amenities range from 4 to 11 minutes. The unbundled