Analysis of Network and Non-Network Factors on Traveler Choice Toward Improving Modeling Accuracy for Better Transportation Decisionmaking

CHAPTER 2. TRAVELER BEHAVIOR OVERVIEW

This chapter presents a state-of-the-art overview of previous studies addressing traveler decisionmaking. It organizes travel behavior knowledge by decision horizon. At the within-day and day-to-day levels, route and departure time choices were the primary focus areas. Travelers’ experiences from day to day influence their future decisions, and the line between these daily choices and behavioral patterns quickly becomes blurred. For example, a traveler may stop using public transportation after a bad experience even if the utility is otherwise perceived as quite high. Since mode choice is subject to available modes, this tends to be modeled and studied as a behavioral pattern in the time span of weeks or months. Finally, lifestyle and mobility choices reflect the self-imposed or otherwise imposed constraints to which travelers are subjected over longer time frames. Much research has been conducted within each area to understand how various factors influence these choices, but there is less understanding of the mechanisms that operate to define these travel habits, patterns, and long-term constraints. These mechanisms and how they relate to different levels of traveler decisionmaking are also discussed.

LONG-TERM LIFESTYLE AND MOBILITY DECISIONS

This section examines urban form variables and the self-selection phenomenon to understand travelers’ lifestyle choices. The influence of added network capacity on travel behavior is briefly discussed as it relates to traveler characteristics.

The effects of price and traveler characteristics on utility are relatively well understood, but existing knowledge on attitudes about mobility and lifestyle and how these attitudes are manifested in behavior is limited. Besides income, psycho-social attributes and their influence on car ownership have been examined.⁽¹⁾ Hiscock et al. and Cullinane found psycho-social benefits in car use, especially for young males.^(2,3) In these studies, car owners felt car use improves prestige, protection, autonomy, and self image.

For decades, the supply-oriented approach to transportation planning revealed that network equilibrium often results in increased travel in response to increased capacity, such that adding capacity may only alleviate congestion in the short term. Furthermore, adding freeway capacity is thought to induce additional travel. Fujii and Kitamura explored the relationship between individuals’ activities and the travel environment to determine whether this is the case for commuters between the time they leave work and the time they go to sleep.⁽⁴⁾ The authors used structural equations to conduct impact analysis of hypothetical freeway lanes in the Osaka-Kobe metropolitan area on residents’ time use and travel. The model examined the number of trips during this period, the total out-of-home activity and travel durations, the number of home-based trip chains, and the total amount of time spent at home after arriving for the first time until going to sleep. Their model of travel preferences suggests that older married individuals tend to have a lower preference toward in-home and out-of-home activities, meaning they have lower preferences toward all activity types. People with higher incomes have large preference indicators for both in-home and out-of home activities but more so for out-of-home activities. Time use and travel variables are treated as endogenous in this study; therefore, the impacts of supply changes cannot be thoroughly addressed. However, the results suggest that additional freeway lanes induce little traffic, indicated by only slight increases in number and duration of out-of-home activities. Much of the time savings from added capacity is allocated to in-home activities.

Effect of Transit-Oriented Development (TOD) and Urban Density on Behavior Patterns and Long-Term Choices

Much research on travel behavior and land use interactions consists of aggregate analyses. This focus on the relationship between urban form and aggregated travel patterns provides little insight into the underlying factors and mechanisms by which urban form influences individual choices.⁽⁵⁾ Disaggregate analysis of household and individual-level behaviors suggests that behavior differences are greater among neighborhoods than among individuals within neighborhoods, and attitudes play an important role in decisionmaking. It is necessary to understand how urban form shapes choice sets since discrete choice theory is only able to illustrate how the factors influence choices within a given choice set.⁽⁵⁾

Holtzclaw et al. attempted to determine which factors influence home location selection and associated transit use the most.⁽⁶⁾ Using odometer readings from emissions systems inspections in San Francisco, CA; Chicago, IL; and Los Angeles, CA, the authors predicted a household’s vehicle miles traveled (VMT) as a function of home-zone density, proximity to jobs, transit service and access to jobs by transit, availability of local shopping, and pedestrian and bicycle “friendliness” (i.e., the attractiveness of these options as compared to driving). The elasticities for vehicle ownership with respect to density for the three cities were -0.33, -0.32, and -0.35, respectively. Elasticities for VMT (per capita) with respect to density were -0.35, -0.40, and -0.43, respectively. Since residents in these cities have above average access to transit, while also noting that the model did not control for parking costs, income, and other relevant variables, applying this model across more cities may not yield such results. For example, the model does not control for attitudes toward driving and public transit, differences in living, vehicle ownership cost, or the cost and quality of transit. These variables differ significantly in most major U.S. cities, and attitudes typically exert a strong influence on travel patterns. However, the magnitudes are surprisingly similar for three urban areas that differ significantly in terrain and climate. One should note that density often acts as a proxy for other urban characteristics.

Equally important to the understanding of how these factors may reduce VMT is an understanding of what factors individuals most prefer in “neo-traditional” developments. In Lund’s survey where California residents were asked to identify their top three reasons for choosing to live in a TOD, only 33.9 percent cited transit accessibility as a top reason.⁽⁷⁾ More often, residents preferred type or quality of housing (60.5 percent), cost of housing (54 percent), or quality of neighborhood (51.7 percent). Lund also found that residents who listed transit as one of their top three reasons were 13 to 40 times more likely to use transit than those who did not, suggesting significant effects of self-selection in such developments.⁽⁷⁾ This endogeneity is discussed further in the next section.

Residential Self-Selection and Vehicle Ownership

Researchers have sought to disentangle the impact of travel preferences and self-selection in home location choice and how this ultimately impacts differences in observed travel patterns across distinct neighborhood designs. Cao et al. suggested that attitudes and socio-demographics are confounding influences in such studies.⁽⁸⁾ While definitive conclusions have not emerged, general neighborhood design distinctions (e.g., walk-oriented versus auto-oriented, existence of bicycle lanes, distance to work, and non-work trip purposes) appear responsible for at least half of the observed VMT differences.^(8–10)

For example, surveys by Frank et al. in Atlanta reveal that despite driving preferences, residents living in a walkable neighborhood tend to drive far less than those living in auto-oriented neighborhoods.⁽¹¹⁾ The least walkable neighborhoods generated roughly 45.5 mi of travel per worker per day, while the most walkable neighborhoods generated only 28.3 mi. Furthermore, those who prefer an auto-oriented neighborhood but live in a walkable neighborhood tend to drive significantly less (25.7 mi per worker per day) than their counterparts in auto-oriented neighborhoods (42 mi) despite their stated preference. Of those who prefer walkable neighborhoods, VMT per worker per day average 25.8 and 36.6 mi, respectively, for residents of walkable versus auto-centric. Thus, while someone may prefer to live in a different neighborhood, it appears that he/she will still conform to the travel opportunities of the home neighborhood. It also merits mention that households residing in suburban settings (versus more traditional neighborhoods) tend to be older and have more members. As expected (by VMT patterns), they also own more vehicles per household member.⁽¹²⁾ The neighborhoods in Frank et al.’s study had similar densities, though they differed in household size and income.⁽¹¹⁾

More recently, Aditjandra et al. applied dynamic (quasi-longitudinal) structural equation models to understand the residential self-selection phenomenon in the United Kingdom.⁽¹³⁾ A total of 219 participants who had moved to their current residence in the last 8 years were asked how they drive now compared to before they moved using a five-point scale from “a lot less” to “a lot more.” Results suggest that socio-demographic characteristics are the main influence on changes in car ownership, but changes in neighborhood characteristics, in particular, safety factors and shopping accessibility, had an important influence. This method was demonstrated in the United States by Cao et al.⁽¹⁴⁾ The findings from the UK study corroborate with Cao et al.’s study, suggesting that controlling for residential self-selection, neighborhood design impacts on travel behavior may be similar in different geographical settings despite different planning contexts.^(13,14) In the United States, car ownership is associated with yard size and availability of off-street parking, whereas in the United Kingdom, shopping/facility accessibility and safety of residential neighborhoods most influences vehicle ownership. Again, one should note that such variables can often proxy for other characteristics. For example, yard size could indicate home lot size or that the residence is a single family dwelling unit.

These proxy issues indicate the need to better understand human interactions and the mechanisms that drive behavior. After all, if a family moves, their friends may still live in the old neighborhood and exhibit the former travel behavior. As many studies have shown, geography is one of the best indicators of frequency and duration of social contact.

MEDIUM-TERM BEHAVIORAL PATTERNS

Socio-Demographics and Household Composition

The impact of socio-demographic variables on travel behavioral patterns is a well-studied topic. Several studies found significant relationships between travel and variables such as age, gender, household composition, and income. For example, Newbold et al. used the General Social Survey dataset in Canada to determine the travel pattern differences of older (65+) and younger people.⁽¹⁵⁾ The data are available for different time periods and can therefore control for generational differences. The study found significant differences in trip duration and frequency across generations. Employment level and health status were also significant predictors of trip duration and frequency.

Gender differences in trip duration, frequency, and mode choice are significant in many studies, which attest women to be more likely to change their behavior toward more sustainable travel modes.^(16,17) Moriarty and Honnery and Best and Lanzendorf found no significant differences between men and women in total number of trips and distance traveled but found differences in activity types.^(18,19) Whereas men make more work trips, women make more trips for maintenance activities. Researchers consistently found that household composition influences trip type, duration, and frequency. Key stages in households include the gain or loss of employment, having children, and retirement.⁽²⁰⁾ Students, unemployed, and part-time employed households with no children are more likely to use non-motorized forms of transportation, and high-income or retiree households are less likely to use non-motorized transportation. Car ownership, also endogenous to some model systems, is found in many studies to be significant with a tendency for people to use cars versus public transit. This trend is significant with high-income groups.⁽²¹⁾ Giuliano, Giuliano and Narayan, and Giuliano and Dargay studied differences in travel behavior between different socio-demographic groups in the UK and the United States.^(22–24) According to these studies, Americans make 4.4 trips per day with a length of 43 mi compared to 3 trips per day and 16 mi in the UK. In both countries, travelers over age 65 travel roughly half the distance of younger participants. The difference between countries is explained by the lower income and significantly higher transportation costs in the UK compared to the United States.

Bomberg and Kockelman surveyed over 500 commuters in Austin, TX, to gather information on their driving behavior during and after an abrupt increase in fuel prices.⁽²⁵⁾ For most of summer 2005, price increases were comparable to previous years; however, between August and September 2005, prices increased 36 percent from $2.16 to $2.93/gal. Ordered-probit models to classify the travel behavior change suggest that travelers are most likely to respond by reducing overall driving; this reduction is achieved through increased use of other modes or trip chaining. A traveler’s built environment characteristics were more influential in behavior change than even income, education, and average driving. Some drivers even adapted their driving style, suggesting some drivers adopt a series of strategies to cope with system changes. Respondents were surveyed again in 2006 to gather information about response to transportation policy measures. Though there was substantial support for alternative modes and reduced fuel dependency, respondents willingness to pay for driving increased ($1.45/gal as distance from the central business district (CBD) increased by one standard deviation from the mean (3.74 mi).

It is worth noting that some urban form variables were evaluated in addition to the traveler characteristics in these studies. Residents of less dense urban areas tend to travel further. As a result, density influences the price of travel and therefore the travel behavior.⁽²⁶⁾ In the United States, urban form is thought to reinforce car use and dependency.⁽²³⁾

Effect of Travel Demand Management Measures and Parking Pricing on Mode Choice

A Transit Cooperative Research Program (TCRP) report found that eliminating minimum parking space requirements and charging market rates for residential parking spaces could reduce vehicle ownership per household enough to reduce household VMT by 30 percent.⁽²⁷⁾ Additionally, charging employees for parking at work was linked to a 10 to 30 percent decrease in single-occupancy vehicle (SOV) mode share depending on the quality of transit alternatives.⁽²⁷⁾ In Portland, OR, establishing maximum parking ratios and a parking limit maximum appeared to reduce the downtown parking ratio by half from roughly 3.4 long-term spaces per 1,000 ft² of commercial space in 1973 to 1.5 spaces per 1,000 ft² in 1990.⁽²⁷⁾ These parking policies, alongside some transportation demand management (TDM) measures and transit enhancements, are credited with increasing Portland’s downtown transit share from 20 to 25 percent in the early 1970s to a downtown commuter transit share of 30 to 35 percent in the 1980s and 1990s.⁽²⁸⁾ As expected, many urban design variables influence mode share (e.g., cities with few parking spaces per employee tend to have higher transit mode share) since limits on parking are implicitly reflected in the shadow price associated with parking.⁽²⁸⁾

Using the 6-week Mobidrive study, Schlich and Axhausen explored repetitious travel behavior.⁽²⁹⁾ Because people rarely evaluate all their options at each new opportunity and because constraints are relatively similar from day to day, habits are formed but mediated by each day’s changing needs. Schlich and Axhausen found that behavior is more variable on weekend days than working days.⁽²⁹⁾ For each individual in the study, variability was sharply reduced and then constant after 2 weeks (i.e., the respondent looked similar over 3 weeks and over 5 weeks). The authors recommended observing participants over a 2-week period.

Learning, Experience, and Inertia

Inertia, or a traveler’s propensity to continue making the same choices based on past experience, is not yet well understood. In 2011, Cherchi and Manca demonstrated that the significance of inertial effect varies substantially with model specification, and this effect is not stable during a stated-preference experiment.⁽³⁰⁾ Depending on a participant’s past experience and exposure to options, the inertial affect also varies, pointing to a need for well-designed and controlled experiments.

Using an agent-based model employing Bayesian perception updating, Chorus et al. determined a perceived value of acquiring travel time information as the difference between expected regret induced by a choice before and after acquiring information.⁽³¹⁾ Simulations revealed that this value of information, even for drivers who considered transit as an alternative to driving, is influenced by three factors: information irrelevance, information unreliability, and preference for driving options. These same factors also limit the effect of received information on mode choice when the information is highly favorable toward transit. The authors suggest that only transit information that is freely provided and easily accessible has the potential to be used by drivers. This information should also be reliable and include aspects of comfort, dynamic conditions, convenience, and perhaps even environmental friendliness. Given the difficulty in meeting these conditions of low-cost, high-quality information, Chorus et al. suggested that it may be more efficient to demonstrate the car’s limited attractiveness in certain conditions, such as inclement weather or road accidents.⁽³¹⁾

The role of inertia in influencing users’ responses to real-time information had previously been captured by Srinivasan and Mahmassani in the context of route switching decisions, as discussed in the next section.⁽³²⁾

PRE-TRIP, EN-ROUTE, AND DAY-TO-DAY BEHAVIOR

Jan et al. found that travelers habitually follow the same route for the same trip, but route variations increase with longer travel distances.⁽³³⁾ The dominant factors for route choice are travel time and distance.^(33–35) Considerable research effort has focused on the effects of route choice behavior under traffic information systems, the dynamic aspects of the route choice behavior, and the relationships among route choice, departure time, and trip-chaining decisions. (See references 32, 36–39.)

Traveler information substantially influences route choice. Abdel-Aty et al. studied route changes in Los Angeles, CA.⁽³⁵⁾ Only a small percentage of the respondents (15 percent) reported using more than one route on their commute. Of that 15 percent, 34 percent said they changed routes after seeing traffic conditions. Drivers with higher incomes and education levels predicted more route changes, perhaps reflecting schedule flexibility and arrival time expectations.

Mahmassani and Herman performed a survey of commuters in Austin, TX, and yielded a binary logit model that relates route switching propensity to four types of factors: geographic and network condition variables, workplace characteristics, individual attributes, and use of information (radio traffic reports).⁽⁴⁰⁾ They found that variables describing the characteristics of the commute itself had a dominant effect relative to workplace rules or individual characteristics. Information in the form of radio traffic reports also appeared to have a strong impact. Regular listeners to traffic are more likely to switch routes. The only socio-demographic attribute significant in the model was age.⁽⁴¹⁾

In a similar experiment, Avineri and Prashker examined the impact of information on traveler learning, differentiated by travelers’ risk aversion.^(42,43) The results suggest that when information about travel times is provided, travelers do not always choose the route with the least expected travel time. Giving static information to users serves to increase traveler heterogeneity. In this case, individuals learned more quickly to prefer either routes with less travel time or routes with less variability in travel time. When examined at an aggregate level, this combination could be seen as a “non-learning effect,” or no change. Furthermore, higher variation in travel times is associated with lower sensitivity to travel time differences. Avineri and Prashker found in some cases that increasing travel time variability of a less attractive route could increase its perceived attractiveness.⁽⁴²⁾ This underlines the need for better models of learning and reinforced habits as an alternative to utility maximization.

Beyond these dimensions, only a few studies have addressed destination adjustment in response to real-time information for discretionary (shopping) travel.⁽⁴⁴⁾ The remainder of this section discusses the effects of other network and non-network factors.

Effect of Tolling and Other Costs on Mobility Decisions

Travel cost as part of demand management is a powerful tool to influence travel behavior. Hensher and King examined the influence of parking costs in the CBD, a park-and-ride facility with public transit access, and the related mode choice as well as destination choice (including a “forgo the trip” alternative) in Sydney, Australia.⁽⁴⁵⁾ Each participant was required to consider six alternatives in a stated preference questionnaire. In 97 percent of the responses, cost was the most significant factor in determining location choice and mode choice. Similar results were found by Handy et al. in a study of whether Americans drive by choice or through necessity.⁽⁴⁶⁾

Congestion pricing of roadways presents a valuable opportunity to rationalize road networks by helping ensure that travelers pay for the delay costs they impose on others. A study of Seattle, WA, travelers with Global Positioning System (GPS) vehicle units estimated that variable network pricing (to reflect the congestion impacts of different demand levels over space and time) would reduce regional VMT by 12 percent and total travel time by 7 percent with a 6-to-1 benefit-cost ratio.⁽⁴⁷⁾ Using GPS tolling meters, the study followed participants to establish a baseline tolling routine. Participants were then given a monetary travel budget sufficient to cover the cost of their routine for the duration of the study period, creating an incentive to reduce certain forms of travel to save/make money. This policy approach is very similar to Kockelman and students’ credit-based congestion pricing policy proposal, though VMT results differ in their network simulations of the Austin and Dallas-Ft. Worth regions of Texas, where marginal social cost pricing of freeways for all links by time of day is consistently estimated to result in VMT savings of under 10 percent.^(48,49)

Saleh and Farrell investigated the influence of congestion pricing on the peak spreading of departure time choice.⁽⁵⁰⁾ Taking into account the scheduling flexibility of respondents, results suggest that non-work activities and work schedule flexibility impact departure time choice for the trip to work. Furthermore, respondents were less willing to pay tolls to depart earlier than usual.

Similarly, a TCRP report discusses a number of elements that influence travelers’ decisions to use a high-occupancy vehicle (HOV) lane.⁽⁵¹⁾ The report concludes that so many urban, facility, and vehicle characteristics interact with one another that it is difficult to delineate the effect of HOV lanes on travelers. However, the success of HOV lanes, both in terms of drivers served and benefits to the road network, is attributed to combinations of the following characteristics:

• Urbanized population of 1.5 million or more.
  
  
• An orientation, preferably radial, to a city center, focusing on major employment centers with preferably more than 100,000 jobs.
  
  
• Geographic barriers.
  
  
• Congestion in general purpose lanes.
  
  
• Realistic potential for 25 to 30 buses per hour.
  
  
• Peak hour travel time savings of preferably 1 min/mi or more or at least 5 min of total travel time.

Effect of Walk Quality on Day-to-Day Travel Behavior and Patterns

Beyond information and pricing, the quality of the urban environment can influence route and activity timing decisions. Cervero and Kockelman examined many features of urban form that may reduce auto dependence.⁽⁵²⁾ Their gravity-based accessibility measure for access to commercial jobs was found to have an elasticity of -0.27, suggesting neighborhood retail shops and pedestrian-oriented design are more significant than residential densities in mode choice selection. Integrating aspects of pedestrian-oriented design, such as four-way intersections and vertical mixing of land uses, may result in significant VMT reductions. For example, a 10 percent increase in the number of four-way intersections in a neighborhood was associated with an average reduction of 5.19 percent of person miles traveled for non-work trips. A doubling of land use mix or variety is associated with a roughly 11 percent increase in modes other than SOV for non-work travel. These effects are discussed in more detail in chapter 4.

Besides urban density, mixed land use and high-quality pedestrian-oriented urban design increase the use of public transit and non-motorized transport modes.⁽⁵³⁾ Naess and Naess and Jensen found that, in general, car use increases with increasing distance from the city center.^(54,55) This could also be an indicator of self-selection or endogeneity. Similarly, Cervero studied the impact of compact, mixed use, and pedestrian-friendly design on mode choice.⁽⁵⁶⁾ The study quantified density and diversity and estimated the influence of each on mode choice. The influences were significant but modest. Surprisingly, the most important influence factor for mode choice was the sidewalk ratio. In well-developed pedestrian areas, commuters were more likely to use public transit and join carpooling initiatives.

Information, pricing, and urban form influence day-to-day and within-day behaviors, but they are understood and applied over time such that they also influence travel patterns. These and other influences are discussed in the following subsections.

Behavioral Mechanisms

Besides all the influencing factors and characteristics that help explain travel behavior changes, it is important to understand the underlying process of the perception and manifestation of these characteristics, which then lead to behavior adjustments. That is, how do patterns become lifestyle choices? Even though there are day-to-day travel variations as discussed in the previous sections, it has also been noted that travel patterns repeat themselves, which suggests that parts of travel behavior are habitual and influenced by inertial effects.⁽⁵⁷⁾ Furthermore, the effect of information, as discussed in previous sections, depends on whether travelers comply with the prescribed information. Inertia, information compliance, travel experience, and learning determine the system outcomes that feed back into supply and demand models.

Behavior adjustment implies that behavior is an outcome of experience or new information of the current conditions. This can be seen as a learning process, which leads to an adjustment of the behavior. Mahmassani and Chang studied an adjustment and experience-based model of perceived travel time for departure time choice.⁽⁵⁸⁾ Under the myopic adjustment rule, the perceived travel time is only a function of the latest day’s outcome. In laboratory experiments conducted to study the effectiveness of different information strategies on user responses to information, Srinivasan and Mahmassani found that route switching model specifications, which predict whether a user will switch paths in a given time interval, consistently outperformed models that view the process as a new choice at every opportunity.⁽³²⁾ These mechanisms are neither mutually exclusive nor collectively exhaustive. As a result, they can operate simultaneously and in conjunction with other mechanisms. The authors designed an experiment whereby virtual commuters were given trip times at three facilities (decision locations), real-time information about congestion on the facilities, a message alerting when they were stuck in a queue, and post-trip feedback consisting of departure time, arrival time, and trip time on the chosen path. Their empirical findings suggest that an individual’s negative experience with ATIS information has mixed effects on inertia, but congestion and information quality tend to reduce inertia. Drivers who experience lower switching costs and increased trip time savings tend to comply with information. In the sequential treatment, past negative experience relative to preferred arrival time seemed to increase likelihood of compliance. Inaccurate information decreased drivers’ compliance propensity.

Bayarma et al. examined multiday travel behavior as a stochastic process using 6-week travel diary data to explore how travel patterns vary and persist among heterogeneous individuals.⁽⁵⁹⁾ The authors classify weekday travel patterns into five representative patterns: public transport commuting; extensive car use involving three or four visits to a location; three to four shopping, leisure, and social trips; high fraction of trips which serve to transport another person; and mostly work visits and time spent on work-related activities. The authors found that transitions from a pattern to itself were frequent, especially for non-workers, but transitions from pattern to pattern varied substantially across individuals. Individuals with a driver’s license tended to have a higher level of day-to-day variability in their travel patterns. Residential location type also influenced variability in daily travel, with individuals living in a central area regularly pursuing more shopping and leisure activities. Gender, marital status, and number of household vehicles were insignificant in this study; age, household type, and employment status explained much of the variation.

In a seminal work on attitude-behavior theory, one study examined the interrelationships between attitudes and behavior from multiple modeling perspectives: multi-attribute models, hierarchical models, market segmentation models, and, to a lesser extent, structural equation models.⁽⁶⁰⁾ Simple models provide empirical support for behavioral feedback mechanisms, and attitudes and behavior are found to simultaneously influence one another. This concept of simultaneous influence has been explored in greater depth since the study, and market segmentation and structural equations models are still used to explore psycho-social influences on travel behavior. Beyond attitudes, perceptions and intentions also have a substantial influence on behavior. While attitudes and perception have been explored in great depth, less attention has been given to traveler intention until recently. Bamberg found that forming an implementation intention (when, where, and how to perform an action) increases the probability that a goal intention is manifested in behavior.⁽⁶¹⁾ In the study of 90 university students, forming an intent to ride a new bus route was the best predictor of whether the student did in fact ride the new bus route, even more so than current bus and auto use habits. While habit exerted a strong negative effect on whether one would test the route for the control group, this habit strength did not influence the experimental group. Thus, Bamberg points out that influencing behavior involves not only influencing the decisionmaking process but also the formation of implementation intention.⁽⁶¹⁾

The studies reviewed in this chapter reflect a diverse field of inquiry conducted by researchers in various disciplines. While considerable understanding exists on several aspects of both short- and long-term behaviors of travelers, the level and completeness of that knowledge is highly variable. More importantly, the ability to make operational use through models for the prediction of user responses to contemplated policies and interventions is limited, especially with regard to measures that entail capturing the dynamic aspects of user decisions in transportation systems. The remainder of the report illustrates how different choice dimensions are affected by given policies and programs and how these can be modeled effectively to support design and evaluation of these policy and programs.