Intercity Passenger Rail in the Context of Dynamic Travel Markets (2016)

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.

59 The NCRRP 03-02 amplified work plan called for the creation of two discrete choice models for the analysis of intercity travel. First, the data from the stated choice exercises were used to create a mode share model based on transportation level-of-service factors such as times, costs, and other traditional inputs such as demographics and party size. Then, additional models were run using attitudinal data as well as level-of-service factors from the same survey to estimate an ICLV model (Figure 42). C H A P T E R 5 Merging Economic Modeling Theory with Analysis of Attitudes and Preferences Z = demographics that predict the latent variable x = mode characteristics w = trip characteristics z = demographic characteristics U = mode utility, predicted by x, w, and z Figure 42. The ICLV Model merges basic values and trip-specific variables.

60 Intercity Passenger Rail in the Context of Dynamic Travel Markets 5.1 A Mode Choice Model with No Attitudes or Preferences 5.1.1 Model Design Stated preference (SP) and stated choice exercises incorporate a type of tradeoff analysis tool also used in choice-based conjoint analyses, but the latter method is more commonly used in market research, whereas the former is more common in the transportation literature. These types of exercises capture respondents’ preferences when making choices and allow for the esti- mation of discrete choice models and systematic analysis of respondents’ decisions. The research team developed an advanced experimental design that was customized for each respondent (as discussed in Chapter 3 and later in this chapter). Experimental designs dictate how the experiments produce the different levels of the different attributes in the SP experi- ments. The team used Bayesian D-efficient designs, which are considered the best experimental design for this type of project. Then, the research team produced different designs for each trip purpose and major OD pair. It was a major undertaking to optimize the SP scenarios and is a state-of-the-art methodology. Most studies generate just one experimental design; for this project, 108 designs were generated. The goal of this experimental design was to ensure designs that were customized to the different sensitivities faced by different respondent types (based on trip purpose) and for different OD pairs (e.g., long versus short trips). Data coding and model building were undertaken by the research team in Ox, a software package used for advanced modeling. Once the designs were created, the actual SP exercises were developed. The SP experiments were created to emulate as clearly as possible the current choices for a trip that respondents had taken in the past, while at the same time minimizing the complexity of the choice. As part of the NCRRP survey, the research team used a stated choice exercise during which each respondent was presented with eight consecutive travel scenarios similar to the one that can be seen in Figure 43. Several aspects of the scenarios—such as the access time, travel duration, and costs for the differ- ent modes—changed from scenario to scenario and respondents had to choose a mode for each of them. The research team tried to simplify the exercises as much as possible to make the choices as simple as possible, while still collecting the necessary data to accomplish the project’s goals. There- fore, trip characteristics like egress mode and egress travel time were either omitted or simplified, as they were not necessary for this project’s purposes. Respondents’ choices in the exercises were the basis for the later estimation of the two discrete choice models. 5.1.2 Results: Times and Costs First, a model was constructed using a traditional base of times and costs. The data from the stated choice exercise was used to develop multinomial logit models of intercity mode choice. These models were specified in a form that results in the coefficients having a monetary scale so that each value represents a “willingness to pay” for changes in the corresponding service vari- able. In addition, the effects of sociodemographic variables were represented as differences from a base group, arbitrarily selected as males under 35 years old. This results in coefficient values that are more easily interpreted than more conventional logit model utility coefficients. Table 13 shows the results of two alternative models that were developed in this form. The two model forms shown here are different only in the assumed form of the utility function; one is the conventional additive utility function in which the factors are added together and the other is multiplicative in which the factors are multiplied together. Looking at the first column that represents the multiplicative model, the first value, -74.56, indicates that on average male

Merging Economic Modeling Theory with Analysis of Attitudes and Preferences 61 Figure 43. Example of a travel scenario from the stated choice exercise presented to a respondent from Washington, DC. travelers under the age of 35 would pay almost $75 more to travel by car than by bus. This value is reduced by $39.91, to about $35, for females under 35 years. The values increase successively for older travelers to a maximum of over $200 (-74.56 + -130.36) for males between the ages of 55 and 64. The pattern for air versus car is similar, but with younger males preferring air over car (evidenced by a positive willingness to pay and assuming all other service variables are equal) and females even more so, but with the preference shifting to car with older travelers. The sociodemographic patterns illustrated by the multiplicative model are less significant for rail versus car but show a pronounced pattern of younger travelers favoring rail over car compared to the rail versus car preference for older travelers. The values of access, egress, and in-vehicle times are comparable to those obtained in previous intercity travel mode choice models and the income elasticity of 0.41 in the multiplicative models indicates that these values increase at rates approximately proportional to the square root (which would be an elasticity of 0.5) of the travelers’ incomes. Either the additive or multiplicative forms could be used directly as intercity mode choice models but of course the primary reason for developing these models here was to determine appropriate specifications for the more robust ICLV models. Such models incorporate both the SP and socio- demographic data used here as well as the attitudinal data that were also collected in the survey. Importantly, these choice data in combination with the attitudinal and demographic data that were collected in the survey also allowed the research team to estimate more advanced and

62 Intercity Passenger Rail in the Context of Dynamic Travel Markets sophisticated models, including the ICLV models that are summarized in Section 5.2 and pre- sented in full detail in NCRRP Web-Only Document 2, Technical Appendix: ICLV and Hybrid Model Development. 5.2 Integrated Choice and Latent Variable Modeling 5.2.1 Background: Activities in the Development of the ICLV Model As described previously, the research team undertook an ambitious survey effort designed to estimate ICLV models (also known as hybrid models) to help capture the attitudes and values that drive intercity mode choice, while accounting for level-of-service variables (e.g., time, cost, frequency, comfort) that also drive intercity mode choice. As summarized by Vij and Walker (2015), “Integrated Choice and Latent Variable (ICLV) models . . . [allow] for the incorporation of latent behavioral constructs within the framework employed by traditional models of disaggregate decision making. ICLV models were first proposed two-and-a-half decades ago by McFadden (1986) and expanded on by Ben-Akiva [Walker] et al. (2002). Rapid strides in optimization techniques and computational power and the ready availability of estimation software. . . . have since contributed to a veritable explosion in the number of studies estimating ICLV models. In the context of transportation and logistics, ICLV models have been applied to the study of travel mode choice (Paulssen et al. 2014), route choice (Prato et al. 2012), car ownership (Daziano and Bolduc 2013), . . . etc.” Mulplicave Addive log likelihood 33,337.60 33,901.00 Significance: p value 0.3411 0.33 Willingness to Pay Est. Significant Est. Significant Bus vs car base ($) 74.56 *** 42.10 * female 39.91 *** 45.72 *** age 35–44 44.39 ** 113.25 *** age 45–54 107.89 *** 186.66 *** age 55–64 130.36 *** 203.82 *** age 65+ 119.19 *** 178.76 *** one addional service per day 1.07 *** 1.10 *** Air vs car base ($) 44.72 21.43 female 37.82 *** 41.22 *** age 35–44 93.00 *** 117.99 *** age 45–54 83.74 *** 111.11 *** age 55–64 127.75 *** 155.83 *** age 65+ 155.87 *** 193.74 *** one addional service per day 5.67 *** 4.03 *** Rail vs car base ($) 0.37 35.91 female 5.12 2.75 age 35–44 13.97 33.95 age 45–54 8.48 41.84 * age 55–64 3.29 45.23 * age 65+ 13.29 57.80 ** one addional service per day 0.94 *** 0 To reduce access me ($/h) 95.82 *** 94.71 *** To reduce egress me ($/h) 46.80 *** 57.46 *** To reduce in vehicle me ($/h) 38.63 *** 57.38 *** Income elascity 0.41 *** 0.06 *p < 0.1 **p < 0.05 ***p < 0.01 Table 13. Results from preliminary models, based on times and costs only.

Merging Economic Modeling Theory with Analysis of Attitudes and Preferences 63 The gap between discrete choice models including only level-of-service variables and behav- ioral theory has encouraged different developments that attempt to enrich behavioral realism by explicitly modeling one or more components of the respondents’ decision-making process (e.g., accounting for attitudes and perceptions). The most general framework proposed is the ICLV methodology (Ben-Akiva, Walker et al. 1999; Ashok et al. 2002; Ben-Akiva, McFadden et al. 2002; Bolduc et al. 2005), with some examples of recent applications given in Abou-Zeid et al. (2010), Glerum et al. (2014), and Hess et al. (2013). This hybrid modeling (aka ICLV) approach integrates latent variable and latent class models with discrete choice methods to model the influence of latent variables and classes on the choice process. Latent variable models capture the formation and measurement of latent psychological factors, such as attitudes and perceptions, which explain unobserved individual heterogeneity. In other words, hybrid choice models allow the joining of models that can analyze both “hard” concepts like travel times, costs, comfort, frequency, etc. with “softer” concepts like how attitudes and values influence choice making. Most of the work to date has been academic with relatively small samples and little effort on understanding important policy implications. 5.2.2 Furthering the State of the Practice in ICLV Modeling In this NCRRP project, the research team estimated the effects of attitudes and values (beyond times and costs) on mode choice using data from the NEC and the Cascade Corridor. For this modeling work, the research team obtained over 6,000 respondents—a very large data set for such an effort—for use in the development of hybrid choice models to better understand the demand for these two major US intercity rail corridors. As far as is known from the literature, this is the largest scale study of its kind using hybrid choice techniques. This is important due to the immense needs of the NEC, in particular, and the necessary investment that needs to be made. The total sample size is roughly 5,112 respondents from the NEC recruited through online sample and a previous study of auto users in the NEC, with 513 respondents obtained from an online sample for the Cascade sample. Not only is this sample size much larger than in most typical studies using ICLV models, but also extensive work was done (as documented in NCRRP Web-Only Document 2, Technical Appendix: ICLV and Hybrid Model Development) on disentangling pure random heterogeneity from that which can be linked to underlying latent attitudes. 5.2.3 Model Specification Separate models were estimated for four different trip-purpose segments: • Work, composed of business travelers and conference attendees, with a total of 1,043 respondents • Vacation, with a total of 2,062 individuals • Visiting friends and relatives (VFR), with a total of 2,724 individuals • Other purposes, with a total of 735 individuals. A common specification was used as the starting point for all segments, and this was refined by excluding attributes that did not show a significant and meaningful influence in a given segment. The following subsections describe the individual components of the overall model structure, look- ing in turn at the role of explanatory variables, latent attitudes, attitudinal indicators, and modal constants. Key Explanatory Variables The components of utility that are related to explanatory variables are travel time (i.e., access time, in-vehicle time, and egress time) and travel cost. For air, bus, and rail, travel cost was defined

64 Intercity Passenger Rail in the Context of Dynamic Travel Markets as the per-person cost; for car, the research team recognized that the driver often pays a larger share and thus multiplied the total cost by a factor to deal with party size. For car, access time and egress time were set to zero. To capture random heterogeneity in sensitivities across respondents, the research team defined the individual coefficients to follow a log-uniform distribution, i.e., allowing for different time and cost sensitivities across respondents. This distribution has a similar shape to a lognormal distribution (being the exponential of a uniform rather than a normal distribution) but with a less extreme tail and initial tests showed it to obtain not only more meaningful results but also a slightly better fit. Separate coefficients were used for travel time on different modes and also for access and egress time, allowing the capture of differences in the perceived onerousness of different time components. For cost, the research team used a similar approach but additionally captured interactions with income, with separate effects for non-reporters. A complication arises, as a share of respondents did not report income. Rather than making an arbitrary assumption that these respondents had an average income, the research team used a separate mean effect for the random cost coefficient for these respondents but kept the level of the underlying heterogeneity for the uniform distribution (i.e., the log of the negative of the coefficient) the same. Latent Attitudes The latent attitude specification used in these models follows on from earlier factor analysis work carried out on the same data. In particular, the research team defined four latent variables: • Attitude toward cars • Attitude toward ICT • Attitude toward urbanism/sociability • Attitude toward privacy Each of these four factors was based on the use of observed variables, which are summarized in Table 14. Car Atude “Rather than owning a car, I would prefer to borrow, share, or rent a car just for when I need it.” “I love the freedom and independence I get from owning one or more cars.” “I feel I am less dependent on cars than my parents are/were.” ICT Atude “It would be important to me to receive e mail or text message updates about my bus or train trip.” “Being able to freely perform tasks, including using a laptop, tablet, or smartphone, is important to me.” Respondent owns smart technology (at least one smartphone, tablet, GPS device, or laptop). Urbanism Atude “I enjoy being out and about and observing people.” “I like to live in a neighborhood where I can walk to a commercial or village center.” “If everyone works together, we could improve the environment and future for the Earth.” Privacy Atude “The idea of being on a train or a bus with people I do not know is uncomfortable.” “I don't mind traveling with people I do not know.” “The thought of sharing a car with others for such a trip seems unpleasant to me.” Table 14. Attitudinal indicators used in the ICLV model.

Merging Economic Modeling Theory with Analysis of Attitudes and Preferences 65 Each of these latent attitudes is defined to have a deterministic and a random component, with latent attitude l for person n being: , ,zl n l n l nα = γ + ξ where the estimates of gl capture the impact of a range of sociodemographic characteristics of person n (zn) on the latent attitude, and where xl is a standard Normal variate (mean of 0, stan- dard deviation of 1), distributed across respondents, capturing the random element of the latent attitude. The sociodemographic terms tested for effects on the latent attitudes were as follows: • Gender (female dummy) • Age (using five categories of which 35–45 served as base) • Education: dummy for respondents without a graduate degree • Employment: dummy for respondents who are not employed Modal Constants The mode-specific constants for respondent n were specified to follow a Normal distribution across respondents, allowing for differences across individual travelers in their baseline prefer- ences for different modes. For identification purposes, the mean and standard deviation were set to zero for bus, on the basis of tests showing that the level of random heterogeneity was lowest for bus. The demographic characteristics for interactions with constants included: • Gender (female dummy): tested for the non-car modes • Age (using five categories of which 35–45 served as base): interacted with the non-car modes • Education: dummy for respondents without a graduate degree, interacted with the constant for non-car modes • Employment: dummy for respondents who are not employed, interacted with the constant for non-car modes (employed as base) • Households with fewer cars than adults: dummy interacted with the car constant (one or fewer cars per adult) • Households with more cars than licenses: dummy interacted with the car constant (one or fewer cars per license) While considerations for service quality were as follows: • Frequency: daily service frequency, interacted with the constant for non-car modes (Note: Fre- quency was included here as opposed to being listed as an explanatory variable because it was not included as a variable in the survey, i.e., it was not explicitly shown to respondents. The aim behind including this here is to test whether respondents in corridors with higher frequency of service for a given mode are more likely to choose that mode also in the hypothetical scenarios.) • West Coast dummy: interacted with the constant for non-car modes (East Coast as base) • Party size: terms for one other person and two plus other people, interacted with the constant for non-car modes (single person as base) • Trip length: terms of overnight, two nights, and three plus nights; interacted with the constant for non-car modes (day trip as base) • Distance effects for air: dummy terms for trips under 200 miles and trips over 400 miles, interacted with the air constant (200–400 miles as base) The combined utility specification now includes the following: • The impacts of the explanatory variables, with randomly distributed time and cost coefficients, where the latter is also interacted with income

66 Intercity Passenger Rail in the Context of Dynamic Travel Markets • The mode-specific constants, which include a deterministic component as well as a random part • An impact on the modal constants by the latent attitudes, which again include a deterministic and random component. Other Elements of Model Composition Two observations are offered to help understand the composition of the model: • Firstly, the sociodemographic terms included in the modal constants explained above relate to person as well as trip characteristics, while those sociodemographic terms mentioned earlier for the latent attitudes related only to person characteristics. This reflects the assumption that attitudes are stable for each person across different trips. • Secondly, all respondent characteristics included in the deterministic component of the latent attitude have also been included in the modal constant as well, thus avoiding a situation where a sociodemographic effect is erroneously interpreted as relating to attitudes when it may just relate to underlying modal preferences, or vice versa. As an example, it may well be the case that younger respondents travel less by car for reasons unrelated to their attitude toward cars. If age was included as a covariate only on the latent attitude toward cars but not separately on the modal constants, this inherent modal preference may erroneously be captured as an attitudinal difference. 5.3 Model Results The modeling effort undertaken for this work was substantial, with a total of 160 different param- eters used across the different models. The results are in turn very detailed and are documented in a number of different tables in NCRRP Web-Only Document 2, Technical Appendix: ICLV and Hybrid Model Development. The following sections will look at selected parts of the results in turn. 5.3.1 Model Fit Statistics and Value of Travel Time Measures The overall model fit statistics and headline value of travel time (VTT) are presented in Table 15; the VTT measures are presented for a mean income of $125,000 per year (this is a Work Vacaon Visit Friends,Relaves Other Respondents 1,043 2,062 2,724 735 Log likelihood (total) 15,827.00 30,867.10 48,659.70 20,354.90 Log likelihood (choice) 5,065.95 9,803.58 12,041.90 2,879.41 2 for choice model only 0.56 0.57 0.60 0.65 Value of Travel Time Mean Std.Dev. Mean Std. Dev. Mean Std. Dev. Mean Std. Dev. Access ($/h) 50.10 39.37 38.71 50.73 49.19 102.06 41.60 84.31 Egress ($/h) 58.37 98.27 25.86 28.70 14.75 47.36 18.02 45.89 Car in vehicle ($/h) 44.15 34.61 23.48 28.75 27.39 32.15 27.93 32.19 Bus in vehicle ($/h) 61.33 61.81 48.28 54.84 40.24 47.94 34.02 37.03 Air in vehicle ($/h) 49.35 130.69 20.72 58.81 29.01 34.30 36.65 38.72 Rail in vehicle ($/h) 46.08 43.81 22.96 21.96 23.96 23.58 26.31 29.20 Table 15. Model fit statistics and value of travel time measures at income of $125,000 per year: Impact of explanatory variables and mode constants.

Merging Economic Modeling Theory with Analysis of Attitudes and Preferences 67 high mean, but is not surprising due to the fact these are intercity travelers in two major US corridors making discretionary or business trips) and the mean and standard deviation of the distribution resulting from the ratio between the negative log-uniform distributions for the time and cost coefficients are reported. The overall model fits are not directly comparable across purposes, not only given the differ- ences in sample size, but also due to the use of different numbers of indicators across segments in the measurement models (details on this are given later). After estimation, it is possible to factor out the component of the overall log-likelihood relating to the stated choices alone, and the calculation of the r2 measures shows very similar performance across segments, where the high values show the relative ease of explaining mode choices for intercity travel, especially after accounting for random heterogeneity and the role of attitudes. The VTT measures, with the exception of access time (where VFR is very similar to work), are higher for work trips than for all other purposes. Even when remembering that these are calculated for a given income of $125,000, this is still not surprising giving the different time pressures faced on work trips. Egress time is valued substantially lower than access time for all purposes except work trips, where the opposite applies, potentially as a result of these trips being presented as outbound and respondents being more sensitive to the part of their journey relating to getting to their work location after exiting the main-mode station. Major differences exist across modes in the way travel time is valued, and the orderings differ across purposes. While for work trips, car in-vehicle time (IVT) is valued the least highest, this is not the case for the three remaining purposes, where the least onerous type of IVT applies to air for vacation and to rail for VFR and other purposes. The highest VTT generally applies to bus, which is clearly a comfort factor, except for purposes other than work, vacation, and VFR, where the VTT for air is marginally higher than that for bus. Overall, rail IVT is valued the least high. Finally, the standard deviations show that there exists extensive variation across individual travelers in how they value travel time. The actual values of time are calculated from the main estimates, which show that, with a few exceptions (access time for work, egress time, car travel time for work and air travel time for work), the estimates for the variance parameters are statistically significant, showing substantial heterogeneity, especially for the travel cost sensitivity. Also, across all four purposes, the sensitiv- ity to travel cost is higher for income non-reporters than for a respondent at the mean income of $125,000. The impact of income on cost sensitivity is strongest for work travel and lowest for vacation and VFR travel. 5.3.2 Mode Constants The mean values of the modal constants provide information on the underlying preference for different modes, all else being equal (i.e., same time, cost, etc.). The following estimates relate to a respondent in the base sociodemographic group (aged between 35 and 44, with a degree and employed, traveling alone on an East Coast day trip between 200 and 400 miles): • Male respondents on work trips prefer rail ahead of air, car, and then bus (the base), where, for female respondents, the difference between car and bus becomes negligible. • Male respondents on vacation trips prefer rail, ahead of bus, air, and car, where, for female respondents, air is ranked second. • Both male and female respondents on VFR trips prefer rail, ahead of air, car, and bus. • Both male and female respondents on trips for other purposes prefer rail, ahead of car, bus, and air.

68 Intercity Passenger Rail in the Context of Dynamic Travel Markets Numerous shifts in the modal constants are also observed, as follows: • Age has impacts for bus and air, where – On vacation trips, the likelihood of traveling by bus is higher for respondents under 35 and – The likelihood of traveling by air decreases for VFR purposes in the two highest age groups but increases in the highest age group for other purposes. • Having more vehicles than licenses in a household increases the probability of traveling by car for work trips. • Compared to respondents with a graduate degree, those without are – More likely to use bus on VFR and other trips and – More likely to use air on VFR trips. • Compared to respondents in firm employment, those not are – More likely to travel by bus for work reasons and – Less likely to travel by air for other reasons. • Compared to respondents on day trips, those who stay – Overnight are less likely to use bus or rail for VFR trips, – One or more nights away are less likely to use bus for other trips, and – Three or more nights away are more likely to use air for vacation trips and less likely to use rail for other trips. • Compared to respondents traveling alone, those traveling with – One other person are less likely to use bus, air, or rail for vacation trips and less likely to use rail or air for VFR trips or air for work trips and – Two or more people are more likely to use bus for work trips, less likely to use air or rail for vacation and VFR trips, and less likely to use air for other trips. • Increases in service frequency increase the likelihood of choosing bus for vacation and VFR trips and air for VFR and other trips. • Compared to the East Coast, travelers on the West Coast are less likely to travel by bus for vacation, air for other purposes, or rail for work and vacation. • Across all purposes, respondents on trips below 200 miles are less likely to use air than those traveling between 200 and 400 miles, where, above 400 miles, they are even more likely to travel by air. 5.3.3 Latent Attitudes In the creation of the ICLV, the role of the four basic values developed in this project were incorporated. Attitude Toward Privacy The latent attitude toward privacy has an effect on mode choice behavior in all four trip- purpose segments. The signs of the attitudinal parameters show those respondents with a more positive latent attitude are more likely to agree with the statement “I don’t mind traveling with people I do not know” and less likely to agree with the statements “The idea of being on a train or a bus with people I do not know is uncomfortable” and “The thought of sharing a car with others for such a trip seems unpleasant to me.” This thus shows that respondents with a more positive value for this latent attitude are less concerned about privacy, in the operation of the ICLV model. The demographic parameters show these respondents are more likely to be female for “other” purposes and are more likely to be older and less likely to be less educated (for work, vacation, and VFR) or to be unemployed (for all purposes other than work). Looking finally at the impacts of this latent attitude on the mode choice behavior, the param- eters show that respondents who are less concerned about privacy are less likely to choose car

Merging Economic Modeling Theory with Analysis of Attitudes and Preferences 69 or air (for all purposes other than work), while, for work trips, they are more likely to choose rail. The negative coefficient of the latent attitude on the probability of choosing rail for “other” purposes needs to be put into the context that the impact is even more negative on car and air, showing simply that these respondents are more likely to choose bus than others, which is a reasonable result. Attitude Toward Auto Orientation The signs of the attitudinal parameters show those respondents with a more positive latent atti- tude toward cars are more likely to agree with the statements “Rather than owning a car, I would prefer to borrow, share, or rent a car just for when I need it” and “I feel I am less dependent on cars than my parents are/were” and to disagree with the statement “I love the freedom and independence I get from owning one or more cars.” This thus identifies this latent attitude as an anti-car attitude, or at the very least as a reduced car lover attitude, in the operation of the ICLV model. The demographic parameters show that these anti-car respondents are less likely to be female for non-work trips (perhaps due to personal security issues), are more likely to be young (as expected given the changes in car attitudes across generations), are more likely to be less edu- cated (other than for vacation trips where there is no effect), and are less likely to be employed (for VFR trips only), where this is likely to be an income effect too, which the main income effect on cost fails to capture completely. Looking finally at the impacts of this latent attitude on mode choice behavior, the mode- oriented parameters show that respondents with a less positive car attitude (i.e., a more positive value for this latent attitude) are less likely to choose car (not surprisingly), but there is also a reduced probability of choosing air for work and VFR trips, possibly due to environmental con- siderations, though again possibly also due to some confounding with income effects. There is also a reduced probability of choosing rail for VFR trips (with bus being the base), reinforcing the earlier point that for VFR, there are possibly confounding effects with income for this latent attitude. Attitude Toward Urbanism The latent attitude toward urbanism was found to have an effect only for the “other” purposes segment. The signs of the attitude parameters show that those respondents with a more positive latent attitude are more likely to agree with the statements “I enjoy being out and about and observing people,” “I like to live in a neighborhood where I can walk to a commercial or village center,” and “If everyone works together, we could improve the environment and future for the earth,” identifying them as more sociable respondents. The demographic parameters show that these respondents are more likely to be female and to have a graduate degree. Looking finally at the impacts of this latent attitude on the mode choice behavior, the relevant parameters show that respondents with a more positive social latent atti- tude are more likely to choose air and rail than those with a less positive attitude, compared to car and bus. Attitude Toward Information Communications Technology The latent attitude toward information communications technology was only included for VFR and “other” purposes after no impact on mode choice behavior was found for work and vacation. This may be seen as surprising for work especially but could be the result of a relatively homogeneous group of work travelers, who all have a heightened use of ICT, making it hard to find an impact on mode choice. The signs of the attitude parameters show that those respondents with a more positive latent attitude are more likely to agree with the statements “It would be important to me to receive

70 Intercity Passenger Rail in the Context of Dynamic Travel Markets e-mail or text message updates about my bus or train trip” and “Being able to freely perform tasks, including using a laptop, tablet, or smartphone is important to me.” This thus identifies this latent attitude as a pro-ICT attitude. The demographic parameters show that, for VFR trips, these respondents are more likely to be female, highly educated, or employed and are more likely to be younger (for both VFR and “other” purposes). Looking finally at the impacts of this latent attitude on the mode choice behavior, the mode-based parameters show us that respondents with a more positive attitude toward ICT are less likely to choose car for VFR and “other” purposes, while, for “other” trips, they are also more likely to choose air. While the former effect is as expected, the latter is somewhat surprising as the use of ICT is less easy during air travel than for bus or rail, although that is starting to change with more abundant in-plane Wi-Fi, the ability to keep devices on during take-off and landing, etc. 5.3.4 Summary and Conclusions When seen in the context of the amount of data presented by the ICLV model in NCRRP Web-Only Document 2, Technical Appendix: ICLV and Hybrid Model Development, it is clear that a great deal of “raw material” is created in the thorough incorporation of a wide variety of complex subject matter in the model’s operation. Importantly, the research team has designed this ICLV model to create output material specifically designed to support the scenario testing tool to be introduced in Chapter 6. The combination of Chapters 5 and 6 will demonstrate that the results obtained in this study have the ability to help formulate real policy implications using advanced (and formerly only academic) modeling techniques in a real-world setting. The results have been designed to be applied in a way that is easy for policy makers to interpret and use. This indicates that studies that can obtain significant sample size and which can then estimate and apply complex models in rea- sonable ways can provide results that are meaningful and relevant to policy makers. As shown in the model development process, measuring and estimating both hard and soft attributes means that a relatively complex model is necessary to describe these complex behaviors. Yet, Chapter 6 will show that this complexity can be exploited to generate good clear policy implications that are useful for transportation practitioners.