Stated choice: a study in predictive validity using an aggregate truth set

Stated Choice models expand the ability of transportation planners to forecast future trends. The Stated Choice approach can forecast demand for new services or policies. However, Stated Choice models are subject to a range of experimental error not found within Revealed Preference (RP) designs. Primary among the concerns facing researchers is the ability of respondents to understand and operate upon hypothetical choice scenarios in a manner that will reproduce choices made under actual situations. These concerns are specified in the magnitude of a scaling factor. Efforts to estimate the scaling factor has proceeded by linking real decisions taken from a revealed preference survey with comparable decisions made under hypothetical conditions. However, where the alternative is new, actual decision data is not available. This study examines the level of error incorporated in a study where no RP data is available. The test of predictive validity focuses on the switching behavior of commuters at a single employment site. The actual data used to test the forecast is limited to company wide or aggregate ridership levels on the public transit mode taken two years after estimation of the SC model. The Fowkes and Preston hypothesis is examined and shown to bound the future actual value between forecasts derived from probabilistic and deterministic methods. The results show that with the passage of time, the probabilistic method approaches the reported ridership levels within 15 percent error.     

Investigating preference heterogeneity in Value of Time (VOT) and Value of Reliability (VOR) estimation for managed lanes

This paper presents an empirical study in investigating user heterogeneity of Value of Time (VOT) and Value of Reliability (VOR). Combined Revealed Preference (RP) and Stated Preference (SP) data were used to understand traveler choice behavior regarding the usage of managed lanes (MLs). The data were obtained from the South Florida Expressway Stated Preference Survey, which focused on automobile drivers who had traveled on the I-75, I-95, or SR 826 corridors in South Florida. Mixed logit modeling was applied and indicated an average value of $13.55 per hour for VOT and $16.13 per hour for VOR. Potential sources of heterogeneity in user sensitivities to time, reliability, and cost were identified and quantified by adding interaction effects of the variables in the mixed logit model. The findings indicated that various socioeconomic demographic characteristics and trip attributes contributed to the variations in VOT and VOR at different magnitudes. The results of this study contribute to a better understanding on what attributes lead to higher or lower VOT and VOR and to what extent. These findings can be incorporated into the demand forecasting process and lead to better estimates and enhanced analytical capabilities in various applications, such as toll feasibility studies, pricing strategy and policy evaluations, and impact analysis.     

Passengers’ valuations of train seating layout,position and occupancy

The layout of seating within train carriages, of which there are numerous possibilities, and also the occupancy of that seating can be expected to impact on passengers’ experiences of a train journey. However, there is very little evidence on how rail passengers value different seating experiences. On the back of exploratory research, and including attitudinal evidence, this paper provides significant and original insights into rail passengers’ preferences in this area. The primary evidence base is a Stated Preference experiment, complemented by a novel Revealed Preference exercise that uses CCTV footage to observe where rail passengers prefer to sit.The valuations, expressed as travel time multipliers, obtained from the Stated Preference exercise are generally plausible and exhibit a wide range according to the precise seating configuration, the occupancy level and the seating position within a layout. The innovative Revealed Preference analysis provides an encouraging degree of support to the Stated Preference results which, reassuringly, can themselves be reconciled with related valuations widely used in the railway industry in Britain and also the findings of the exploratory and attitudinal research. Whilst we are interested here solely in obtaining valuations, we demonstrate how they can be used in demand forecasting.     

Investigating the market for inter-modal freight technologies

Freight carried by rail has traditionally been mainly low value bulk commodities. The market for transport of such commodities appears at best static and is forming a smaller proportion of the total demand for freight transport. There is thus an urgent need for rail operators to develop practical and cost effective inter-modal systems, which offer high quality services to consignors of consumer goods whose premises are not usually connected to the rail network. Much long-haul traffic of this type is international. In continental Europe, a number of inter-modal technologies—including swapbodies and piggyback—have long been in use. Development of similar technologies for use within the more constrained loading gauge of Great Britain, has received a great boost from the impending opening of the Channel Tunnel. The alternative technologies are discussed, before turning to ways of stimating the market for them. A large part of the paper is devoted to reporting on a computerised survey using our LASP (Leeds Adaptive Stated Preference) technique. The reason for using hypothetical Stated Preference data is the inadequate nature and extent of data on actual choice decisions, particularly in circumstances in which confidential freight rates are individually negotiated and little general merchandise goes by rail. By bringing together the results of this survey with information on costs and quality of service, the likely future market for inter-modal freight technologies is assessed. It is seen that the potential for inter-modal services within Britain is very limited, although there should be a good opportunity on the major corridor from London to Scotland through the West Midlands and the North West. When the Channel Tunnel is opened, however, the potential for services between Britain and continental Europe will be enormous, provided that an adequate quality of service can be offered.     

An investigation on the performances of mode shift models in transit ridership forecasting

This paper aims at investigating the over-prediction of public transit ridership by traditional mode choice models estimated using revealed preference data. Five different types of models are estimated and analysed, namely a traditional Revealed Preference (RP) data-based mode choice model, a hybrid mode choice model with a latent variable, a Stated Preference (SP) data-based mode switching model, a joint RP/SP mode switching model, and a hybrid mode switching model with a latent variable. A comparison of the RP data-based mode choice model with the mode choice models including a latent variable showed that the inclusion of behavioural factors (especially habit formation) significantly improved the models. The SP data-based mode switching models elucidated the reasons why traditional models tend to over-predict transit ridership by revealing the role played by different transit level-of-service attributes and their relative importance to mode switching decisions. The results showed that traditional attributes (e.g. travel cost and time) are of lower importance to mode switching behaviour than behavioural factors (e.g. habit formation towards car driving) and other transit service design attributes (e.g. crowding level, number of transfers, and schedule delays). The findings of this study provide general guidelines for developing a variety of transit ridership forecasting models depending on the availability of data and the experience of the planner.     

Interactive stated choice surveys: a study of air travel behaviour

Stated preference (SP) experiments are becoming an increasingly popular survey methodology for investigating travel behaviour. Nevertheless, some evidence suggests that SP experiments do not mirror decisions in real markets. With an increasing number of real world decisions made using the internet, an opportunity exists to improve the realism of the SP counterparts of such choices by aligning the choice environment with such online portals. In this paper, we illustrate the benefits of such an approach in the context of air travel surveys. Our survey is modelled on the interface and functionality of an online travel agent (OTA). As with a real OTA, many ticket options are presented. Sort tools allow the options to be reordered, search tools allow options to be removed from consideration, and a further tool allows attributes to be hidden and shown. Extensive use of these tools is made by the 462 respondents. A traditional SP component was also completed by the respondents. Our exploratory analysis as well as random utility model estimation results confirm not only that respondents seem to engage more actively with the interactive survey, but also that the resulting data allows for better performance in model estimation compared to a more conventional SP experiment. These results have implications for the study of other complex travel choices where interactive surveys may similarly be preferable to standard approaches.     

Factors influencing the propensity to cycle to work

This paper describes the development of a mode choice model for the journey to work with special emphasis on the propensity to cycle. The model combines Revealed Preference (RP) and Stated Preference (SP) data to form a very large and comprehensive model. RP data from the National Travel Survey was combined with a specially commissioned RP survey. A number of SP surveys were also undertaken to examine the effects of different types of en-route and trip end cycle facilities and financial measures to encourage cycling.The development of the model is described in detail. The model was used to forecast trends in urban commuting shares over time and to predict the impacts of different measures to encourage cycling. Of the en-route cycle facilities, a completely segregated cycleway was forecast to have the greatest impact, but even the unfeasible scenario of universal provision of such facilities would only result in a 55% increase in cycling and a slight reduction in car commuting. Payments for cycling to work were found to be highly effective with a £2 daily payment almost doubling the level of cycling. The most effective policy would combine improvements in en-route facilities, a daily payment to cycle to work and comprehensive trip end facilities and this would also have a significant impact on car commuting.     

Inconsistent choices in Stated Choice data;Use of the logit scaling approach to handle resulting variance increases

The scaling approach is a statistical estimation method that allows for differences in the amount of unexplained variation in different types of data, which can then be used together in the analysis. This approach has been mostly used in context of combining Stated Preference and Revealed Preference data, but has also been used as a method of identifying systematic differences in the variance of choices within a single Stated Preference data set, e.g. for investigation of learning and fatigue effects. This paper investigates whether a scaling approach is suitable for handling inconsistencies in Stated Choice data. Both the number of inconsistent choices, based on a test of violations of the transitivity axiom, and education are used as scaling variables. Scaling effects appear to exist due to inconsistent choices, and the amount of unexplained variance is shown to increase as the number of inconsistent choice increase. Scaling due to inconsistencies significantly improves the models and reduces the valuations of travel time. In addition, the scaling approach makes the valuations of travel time from the Stated Choice data more consistent with the valuations from Contingent Valuation data included in the same study. In spite of the fact that education is the only significant explanatory variable for the number of inconsistent choices, scaling due to education gives no significant improvement in the model.     

A Stated Preference Experiment for Measuring Service Quality in Public Transport

Abstract

This paper develops a Stated Preference (SP) experiment that provides a way to measure service quality in public transport. The paper introduces an empirical procedure for optimising the SP experiment. This procedure permits the identification of the choice alternatives defining the experiment by simulating the choices of a user sample. By using the data collected from an experimental survey, a Multinomial Logit model was calibrated. This model is a way of identifying the importance of service quality attributes on global customer satisfaction and calculating a Service Quality Index, which provides an operationally appealing measure of current or potential service effectiveness.     

Commute responses to employment decentralization: Anticipated versus actual mode choice behaviors of new town employees in Kunming,China

This study examines workers’ mode-choice responses to a typical job decentralization policy implemented in China’s urban development – government job relocation (GJR) to new towns in the urban periphery. Broadly, the literature suggests that job decentralization tends to increase car commuting; however, little is known about the effects of China’s GJR initiatives on individuals’ commuting mode choices. Using Kunming as a case study, this study examines how workers’ commuting mode choices have shifted in response to the GJR policy. Our study analyzes two travel survey datasets that span the job relocation process: (1) stated preference (SP) data on workers’ anticipated mode choices after a move of workplace to a planned new town; and (2) revealed preference (RP) data on workers’ actual choices of commuting mode after their jobs were moved. The findings suggest that after job relocation, workers’ actual commuting modes shift from more sustainable modes towards cars. The determinants of workers’ mode choices differ substantially between the hypothetical and actual setting of job relocation. The anticipated mode choices are largely determined by socio-demographic characteristics whereas the actual mode choices are strongly influenced by travel time and housing locations. The evidence from this study offers two important implications for future planning practice of job decentralization. First, planners and policy-makers should be skeptical about the transportation benefits of job decentralization. Second, while SP surveys can assist planners to predict individuals’ mode-choice responses, the robustness of SP results should be carefully assessed before translating into the evidence base for informing job decentralization policy-makings.     

Valuation of improved railway rolling stock: A review of the literature and new evidence

The paper provides a comprehensive review of a large amount of previously unpublished British evidence about the valuation of new railway rolling stock and its effects on demand. An important conclusion is that incentives to bias Stated Preference responses and unfamiliarity with the rolling stock being considered have led to inflated values. This has broader implications for the use of Stated Preference techniques. Also provided is fresh empirical evidence for a range of different types of rolling stock and specific rolling stock attributes. A novel aspect of the research was the use of disaggregate Revealed Preference choice data. The estimated rolling stock values are much lower than those obtained from previous studies.     

Economical welfare maximisation analysis: assessing the use of existing Park-and-Ride services

This paper assesses travellers’ responses to the use of existing Park-and-Ride (P&R) services based on an economical welfare maximisation approach. Specifically, the paper presents a modelling framework to estimate consumer surplus and producer surplus (business profits) on the basis of modal choice probabilities. The paper draws on evidence from Stated Preference surveys conducted around two P&R sites in Sapporo, Japan, where P&R services occupy a modest market space. Overall, the results suggest that business profit increases when economical welfare is maximised, as a consequence of increased demand. It is also shown that P&R choice is not only influenced by parking fees, but also by the fares and other attributes of alternative transportation modes. Accordingly, the interactions of P&R with alternative transportation modes should be taken into consideration in any strategic transportation policies oriented towards motivating sustainable transport mode choices.     

Accounting for inertia in modal choices: some new evidence using a RP/SP dataset

Inertia is related with effect that experiences in previous periods may have on the current choice. In particular, it has to do with the tendency to stick with the past choice even when another alternative becomes more appealing. As new situations force individuals to rethink about their choices new preferences may be formed. Thus a learning process begins that relaxes the effect of inertia in the current choice. In this paper we use a mixed dataset of revealed preference (RP)-stated preference (SP) to study the effect of inertia between RP and SP observations and to study if the inertia effect is stable along the SP experiments. Inertia has been studied more extensively with panel datasets, but few investigations have used RP/SP datasets. In this paper we extend previous work in several ways. We test and compare several ways of measuring inertia, including measures that have been proposed for both short and long RP panel datasets. We also explore new measures of inertia to test for the effect of “learning” (in the sense of acquiring experience or getting more familiar with) along the SP experiment and we disentangle this effect from the pure inertia effect. A mixed logit model is used that allows us to account for both systematic and random taste variations in the inertia effect and for correlations among RP and SP observations. Finally we explore the relation between the utility specification (especially in the SP dataset) and the role of inertia in explaining current choices.     

Selfishness and altruism in the distribution of travel time and income

Most economic models assume that individuals act out their preferences based on self-interest alone. However, there have also been other paradigms in economics that aim to capture aspects of behavior that include fairness, reciprocity, and altruism. In this study we empirically examine preferences of travel time and income distributions with and without the respondent knowing their own position in each distribution. The data comes from a Stated Preference experiment where subjects were presented paired alternative distributions of travel time and income. The alternatives require a tradeoff between distributional concerns and the respondent’s own position. Choices also do not penalize or reward any particular choice. Overall, choices show individuals are willing forgo alternatives where they would be individually well off in the interest of distributional concerns in both the travel time and income cases. Exclusively self-interested choices are seen more in the income questions, where nearly 25 % of respondents express such preferences, than in the travel time case, where only 5 % of respondents make such choices. The results also suggest that respondents prioritize their own position differently relative to regional distributions of travel time and income. Estimated choice models show that when it comes to travel time, individuals are more concerned with societal average travel time followed by the standard deviation in the region and finally their own travel time, while in the case of income they are more concerned with their own income, followed by a desire for more variability, and finally increasing the minimum income in their region. When individuals do not know their fate after a policy change that affects regional travel time, their choices appear to be mainly motivated by risk averse behavior and aim to reduce variability in outcomes. On the other hand, in the income context, the expected value appears to drive choices. In all cases, population-wide tastes are also estimated and reported.     

Stated choice valuation of urban traffic air pollution and noise

In this study environmental problems related to urban traffic are valued by Stated Choice. Stated Choice is a Stated Preference method in which non-market goods are assessed relatively to each other and not in absolute amounts as with the more common Contingent Valuation Method. The use of Stated Choice to estimate people’s willingness to pay to reduce environmental problems caused by road traffic can be seen as an extension of the established use of Stated Choice in transport research. Despite the fact that the results from this study have been used by the Norwegian Public Roads Administration in their cost benefit analyses since 1995, this paper points to the fact that considerable uncertainty remains with regard to (1) the area of validity of the results, (2) the impact on the valuations of interaction effects and other general methodological problems with Stated Preference methods, and (3) problems related to the complex choice situation of Stated Choice.     

High speed rail market projection: Survey design and analysis

This paper summarises work done to assess the market potential and likely financial performance of a planned high-speed rail link connecting Sydney, Canberra and Melbourne, under a variety of scenarios concerning the services offered and the possible market responses of the competing transport modes in the corridor. In the Australian context, such a link has the characteristics of an essentially new transport mode since existing rail services are extremely poorly developed. The expectation was that generated demand could be a substantial part of the overall ridership. A programme of market research was commissioned by the private consortium considering the project, designed to support forecasting models capable of predicting both diverted and generated travel on the new service. A major survey of current travellers was conducted in 1988, followed by an extensive collection of stated preference evidence about the factors affecting the travel decisions of both existing travellers and those who had not recently made any journeys in the corridor. The paper focuses mainly on the design and organisation of the surveys, on the analysis approach, and on the methods used to generate forecasts for simulated populations and scenarios.     

Hypothetical bias in Stated Choice Experiments: Is it a problem? And if so,how do we deal with it?

The extent to which Stated Choice (SC) experiments suffer from hypothetical bias continues to be a controversial topic in the literature. This research provides further evidence in this debate by examining the existence of hypothetical bias in a transport-related SC experiment. Data for this research were sourced from a University of Sydney study exploring the effect of exposure-based charging on motorist behaviour. The sample included 148 Sydney motorists who were recruited to take part in the 10-week GPS driving field study (Revealed Preference/RP data). In addition, participants were also required to complete an SC survey which was designed to mimic the RP decision context in order to capture what participants indicated they would do as opposed to what participants actually did in reaction to the charging regime.The current state of practice for measuring hypothetical bias in the literature is to compare aggregate differences in model outcomes using SC and RP data sources. Aggregate analysis is limited in its scope and does not allow for the calculation of the prevalence of hypothetical bias (i.e., how many participants are affected by hypothetical bias). This research is uniquely structured to allow for individual categorisation of hypothetical bias by comparing SC and RP data from the same sample for the direct purpose of investigating the prevalence of hypothetical bias. Furthermore, the extent to which mitigation techniques (cheap talk and certainty scales) influence hypothetical bias is also explored. The findings from this research show that the SC model estimates are prone to hypothetical bias and that the mitigation techniques have potential to compensate for this inherent bias.     

Stated preference methods and travel demand forecasting: An examination of the scale factor problem

Stated Preference choice models may have incorrect scale properties, because their coefficients incorporate the influence of factors which do not affect actual behaviour, and this paper examines the consequences of this scale factor problem for demand forecasting. Empirical evidence is presented which suggests that the scale factor problem is not serious and the results are generalised to consider the consequences of the scale factor problem in situations other than those prevailing in the empirical example. The paper also examines the predictive performance of three procedures for forecasting changes to an existing demand in the light of the scale factor problem. This indicates that reliance should not be placed on the incremental logit model and that a simple procedure which amends a standard logit model forecast in the light of its performance in explaining the observed initial demand is generally preferred to unadjusted logit model forecasts.     

On the development of time period and mode choice models for use in large scale modelling forecasting systems

A substantial amount of research is presently being carried out to understand the complexities involved in modelling the choice of departure time and mode of travel. Many of these models tend to be far too complex and far too data intensive to be of use for application in large scale model forecasting systems, where socio-economic detail is limited and detailed scheduling information is rarely available in the model implementation structure. Therefore, these models generally work on the basis of a set of mutually exclusive time periods, rather than making use of continuous departure time information. Two important questions need to be addressed in the use of such models, namely the specification used for the time periods (in terms of length), and the ordering of the levels of nesting, representing the difference in the sensitivities to shifts in departure time and changes in the mode of travel. This paper aims to provide some answers to these two questions on the basis of an extensive analysis making use of three separate Stated Preference (SP) datasets, collected in the United Kingdom and in the Netherlands. In the analysis, it has proved possible to develop models which allow reasonably sound predictions to be made of these choices. With a few exceptions, the results show higher substitution between alternative time periods than between alternative modes. Furthermore, the results show that the degree of substitution between time periods is reduced when making use of a more coarse specification of the time periods. These results are intended for use by practitioners, and form an important part of the evidence base supporting the UK Department for Transport’s advice for practical UK studies in the WebTAG system.¹