Search

The factors influencing trip departure time are taking more importance in urban planning practice since congestion is increasingly being addressed by travel demand management (TDM) strategies. In this paper we formulate and estimate a joint travel mode-departure time model for commuting trips using combined revealed preference (RP) and stated choice (SC) data. The RP data considered nine alternative modes and up to 11 time periods, and the level-of-service data were obtained at an unusual level of precision using GPS measurements. The travel time, cost and cost divided by the wage rate coefficients were fairly similar in both the RP and SC environments, suggesting equal error variances for both datasets. The only parameters that differed between each type of data were those associated with the schedule delay early (SDE) and late (SDL) variables required by Small's Scheduling Model. This may be due to the potentially different temporal perspectives between RP choices (longer term decisions) and SC decisions, arguably shorter term given the nature of the experiment and the context presented in it (implementation of a congestion charging policy and a flexible working-hours scheme). The models were used to forecast the impacts of a hypothetic congestion charging scheme in Santiago, showing that the schedule delay coefficients derived from the SC context produced a smoother and less-peaked temporal distribution of travel demand than the RP parameters. [ABSTRACT FROM AUTHOR]

Studying road pricing acceptability has become increasingly relevant. Even though the policy has been effective in reducing the growing levels of congestion in several cities by curbing the indiscriminate use of the car, it still faces an initial rejection among the population in general. We discuss this issue based on a comprehensive review and set out to define what could be an acceptable road pricing scheme for Santiago de Chile. For this purpose, we estimated a hybrid discrete choice model that allowed us to segment the population according to their attitudes toward the measure. The preferred road pricing scheme in Santiago would charge only for entrance to the smallest of three cordons, around the historic centre of the city, during either the morning (or morning and evening) rush hours, provided that revenues were used to enhance public transport in the city. Since this is different from the best technical solution, which would imply charging in the largest cordon area, we suggest testing the preferred scheme in a pilot trial, following international recommendations and successful previous cases. [ABSTRACT FROM AUTHOR]

The interest in individuals’ non-strictly rational behaviour has permeated into discrete choice models pushed by several psychological theories. We analysed Satisficing Theory, which is particularly useful when decision makers have to face the cognitive burden of complex decisions. Three principles of the theory are discussed noting that the third, partial pay-off functions, has not been addressed in the literature. We implement the three principles mathematically obtaining a discrete choice model in which the decision maker chooses the first satisfactory alternative. The model formulation is analytically derived, as well as its properties. The Stochastic Satisficing model allows variable or constant marginal rates of substitution and enables the explicit characterization of non-compensatory behaviours. The model can also explain attribute saturation and non-attendance of high order needs when basic needs are not fulfilled. We analysed the model performance on synthetic data, showing that it is likely to be unbiased and consistent for relatively common samples sizes. When tested on real data, the model proves its flexibility to also adapt to constant marginal rates of substitution. We conclude that the model is a good characterization of Satisficing behaviour for simple datasets. [ABSTRACT FROM AUTHOR]

When designing stated-choice experiments modellers may consider offering respondents an “indifference” alternative to avoid stochastic choices when utility differences between alternatives are perceived as too small. By doing this, the modeller avoids adding white noise to the data and may gain additional information. This paper proposes a framework to model discrete choices in the presence of indifference alternatives. The approach allows depicting the likelihood function, independent of the number of alternatives in the choice-set and in the subset of indifference alternatives, offering a new approach to existing methods that are only defined for binary choice situations. The method is tested with the help of simulated and real data observing that the proposed framework allows recovering the parameters used in the generation of the synthetic datasets without major difficulties in most cases. Alternative approaches, such as considering the indifference option as an opt-out alternative or ignoring the indifference choices are clearly outperformed by the proposed framework and appear not capable of recovering parameters in the simulated set. [ABSTRACT FROM AUTHOR]

Fare evasion is a problem in many public transport systems around the world. Policies to reduce this problem are generally aimed at improving control systems and increasing fines for offenders. In this paper, we attempt to identify the joint impact of different variables explaining fare evasion using an econometric study. The variables found to be statistically significant are the level of inspection, the proximity to a Metro or intermodal station, the bus occupancy level, the period of the day, the geographic location and number of passengers boarding and alighting at each bus stop, among others. We propose a novel approach to perform cost-benefit evaluation in order to help authorities increase the cost-effectiveness of ticket inspection strategies on a given time horizon. We obtain new evidence that indicates that inspection strategies can be cost-effective even when evaders are not given a fine. [ABSTRACT FROM AUTHOR]

Experience products are those the quality of which cannot be ascertained until after consumption, forcing consumers to base their purchase decision on an expectation of the product's quality. This expected quality is based on cues available before purchase, among which price is noteworthy, as consumers tend to believe that higher prices imply higher quality. But price also stresses the consumers' budget restriction, inducing a double -and conflicting- global effect on purchase probability. Using the traditional formulation of Random Utility Models for experience goods (i.e. introducing all attributes directly in the utility function) can lead to an endogeneity problem due to the omission of expected quality , introducing bias on the results. Using a stated wine choice experiment conducted in China as a case study, we correct for endogeneity by modelling each alternative's expected quality as a latent variable, explained by all available quality cues, including price . Then we explain choice as a trade-off between price and expected quality . This allows us to separate both effects of price and correct for at least one source of endogeneity while being consistent with behavioural theory; this has either been ignored or not treated correctly in previous literature. Moreover, as the model requires only a single quality indicator for each alternative to achieve identification, the respondents’ burden increases marginally. Our results show that the use of latent variables reduces endogeneity and effectively allows to measure both effects of price separately, obtaining higher significance and correct signs for its parameters. [ABSTRACT FROM AUTHOR]

Latent class (LC) models have been used for decades. In some cases, models of this kind have exhibited difficulties in identifying distinct classes. Identifiability is key to determining the presence or absence of the different population cohorts represented by the latent classes. Theoretical identifiability addresses this issue in general, but no empirical identifiability analysis of this kind of model has been performed previously. Here, we analyse the theoretical properties of LC models to establish necessary conditions on the classes to be identifiable jointly. We then, establish a measure of behavioural difference and relate it to empirical identifiability; this measure highlights factors that are crucial for identifiability. We show how these factors affect identifiability through simulation experiments in which classes are known, and test elements such as the proportion of individuals belonging to each latent class, different correlation structures and sample sizes. In our experiments, each latent class corresponds to a different choice heuristic. We present a graphical diagnostic that supports the measure of behavioural difference that promotes identifiability and provide examples of model non-identifiability, partial identifiability, and strong identifiability. We conclude by discussing how non-identifiability can be detected and understood in ways that will inform survey design and analysis. • We analyse the theoretical properties of latent class models to establish necessary conditions for class identifiability. • We establish a measure of behavioural difference and relate it to empirical identifiability, highlighting its crucial factors. • We provide a graphical diagnostic for identifiability with examples of model non-identifiability, and weak and strong identifiability. [ABSTRACT FROM AUTHOR]

Endogeneity is a common problem in econometric modelling that may lead to estimating inconsistent parameters. In the scientific literature, the Multiple Indicator Solutions (MIS) method is used to correct for endogeneity. This approach uses indicators that, in practice, tend to be collected as discrete using Likert scales; however, theoretically, the MIS method is derived considering continuous indicators. To close this research gap, this paper focuses on characterizing the impact of discrete indicators when correcting for endogeneity using the MIS method in the case of discrete choice models (DCM). Our findings show that (i) under some conditions, using discrete indicators instead of continuous ones seems not to be a problem, however, (ii) there is also evidence that indicates that the correction could fail under not unusual circumstances. • Endogeneity is an anomaly that may yield inconsistent model parameters. • We designed a Monte Carlo experiment and applied a SP survey, to examine the impact of using discrete indicators. • Our findings show that the algorithm used to produce discrete indicators affects the endogeneity correction. • Using real data, we show that the correction with continuous indicators worked better than with discrete indicators. [ABSTRACT FROM AUTHOR]

Abstract: Stated choice studies have been applied regularly to the valuation of time savings and other attributes of travelling as perceived by individuals. In such experiments, respondents often provide reference levels for the attributes and the hypothetical choices presented to them are pivoted around actual behaviour. However, most individuals are not able to provide reference levels for the number of casualties on the road they travel. Thus, if valuation of this important element is attempted, it is the researcher who must provide casualty risk reference levels to the respondents. Some studies have applied route choice experiments including a safety attribute but the majority has been limited to only one particular road section with a common baseline risk for all respondents. This study discusses the setting up and results of a more generalized route choice experiment including a safety attribute. Respondents provided, at an initial stage, their travel times and costs related to a recent trip by car. Then, expected numbers of casualties for different trip lengths were calculated based on travel distances and traffic densities. So, the calculated number of severe injuries and fatalities (casualties) per year, on the road section the respondent had travelled, entered as a third attribute in the choices, together with the reported travel times and costs. Route choice was analysed using multinomial logit and mixed logit models. From the latter models we obtained point estimates for the value of the statistical life ranging from € 7.3 million to € 19.1 million. [Copyright &y& Elsevier]

Abstract: The still controversial Transantiago public transport system, in Santiago de Chile, has a topological structure that often requires its users to make one or more transfers to reach their destinations. This was rarely necessary in the previous non-integrated system and users reacted with unexpected displeasure when it started even though fare integration in the new system means that transferring involved no extra costs. This study investigates users'' subjective valuations of the transfer experience and its associated elements (walking and waiting times), analysing how these vary for different types of transfer combinations. In particular, we determine the relative preferences the following transfer combinations: metro–metro, metro–bus, bus–metro and bus–bus, with emphasis on the importance of various physical characteristics such as information availability, the existence of station escalators and the ability to board the first available bus or train. We also estimate the relative valuation of the different time component values (walking, waiting and in-vehicle) of trips including a transfer, and also derived the penalties users assign to trips that require transferring at intermodal stations during the morning peak hour. The trip time components most heavily penalised were the walking time involved in transferring and the final walking time to the user''s destination. [Copyright &y& Elsevier]

Abstract: In recent years there has been a re-focus on the valuation of a statistical life from the ex post or human capital method to an ex ante willingness to pay (WTP) approach. This is in part a recognition that we need to focus on establishing the amount, ex ante, that individuals are willing to pay to reduce the risk of exposure to circumstances that might lead to death or degree of injury in the road environment. This study sets out a framework in which to identify the degree of preference heterogeneity in WTP of pedestrians to avoid being killed or injured. A stated choice experiment approach is developed. The empirical setting is a choice of walking route for a particular trip that a sample of pedestrians periodically undertakes in Australia. Mixed logit models are estimated to obtain the marginal (dis)utilities associated with each influence on the choice amongst the attribute packages offered in the stated choice scenarios. These conditional estimates are used to obtain the WTP distributions for fatality and classes of injury avoidance, which are then aggregated to obtain estimates for pedestrians of the value of risk reduction (VRR). [Copyright &y& Elsevier]

Abstract: Recent advances in the specification of the utility function of mixed logit models allow the analyst, in principle, to consider a vast variety of individual heterogeneity. Nevertheless, when estimating the model it is common practice to experience severe difficulties in discriminating between different specifications to infer the “true” data generating process. We investigate possible sources for this difficulty focusing on the confounding effects inherent in two basic assumptions of discrete choice model utilities: linearity in the parameters and added error terms. We analyse the role of these assumptions in giving rise to confounding effects and why this increases the difficulty of discriminating among different structures. Finally, we investigate how these problems may affect benefit appraisal using these models. Empirical evidence is provided for two different environmental contexts and a more typical transport context using various kinds of data. [Copyright &y& Elsevier]

Abstract: Discrete-choice-models have come a long way since the field erupted in the seventies with the seminal work of McFadden and colleagues at Berkeley. For more than 25years the multinomial logit (MNL) model and its close relative, the hierarchical or nested logit (NL) model (a generalization that allowed treating correlated alternatives grouped in nests, the simultaneous brain child of , and ) reigned supreme, using mainly revealed preference data. At the end of the 90s the modeller’s tool kit expanded considerably with the installation of both stated preference data, as a strong ally to help treating the problems posed by new alternatives and latent or secondary variables, and the most flexible member of the family, the mixed logit (ML) model, that allows considering two remaining deficiencies of the old guard, heteroskedasticity and taste variations. This paper provides a glimpse of the field and speculates on the potential use of these methods to complement sensory work in the effort to better understand customer preferences. [Copyright &y& Elsevier]

Abstract: Passenger demand and, in particular, mode choice between the islands of Gran Canaria and Tenerife has experienced important changes in the last decade. In 2005 the jetfoil, which had been the dominant mode for many years, was replaced by a slower but cheaper fast ferry service. This induced important changes in the market shares of all competing modes (airplane, slow ferry and another fast ferry with a shorter in-sea time, but needing a bus connection in land). We estimated several discrete choice models, with data collected two years before, with the aim to test their forecasting performance in relation with observed behaviour. Interestingly, we found that an easy to interpret multinomial logit model allowing for systematic taste variations performed best in forecasting. We also discuss some model assumptions related to forecasting that allow replicating the effects of introducing a new mode more accurately. We finally show how the model can be used to examine the social benefits of a related infrastructure improvement project in the island of Gran Canaria. [Copyright &y& Elsevier]

Abstract: A stated preference experiment is developed looking at airplane and a fictitious high-speed train trips between Santiago and Concepción in Chile. Four variables are used – travel time, fare, comfort, and service delay; a factorial fractional design allowed the experimental size to be reduced to nine pairs of options. Trip characteristics and socio-economic information were also gathered for each respondent. With the purpose of overcoming the potential imperfections that pure SP models are known to have, we incorporated revealed preference data, obtained in a previous study, including bus, train and airplane travellers. Mixed stated and revealed preference models were estimated and compared with those obtained from the stated preference data alone. Finally, a second group of stated preference data is added. Different specifications were tested with mixed stated and revealed preference data. Moreover, by segmenting the sample we were able to derive subjective values of time for different kinds of individuals. [Copyright &y& Elsevier]

Abstract: We treat the problem of fitting alternative specific constants (ASC) in models estimated with a mixture of revealed preference (RP) and stated preference (SP) data to forecast the market shares of new alternatives. This important problem can have non-trivial solutions, particularly when some of the SP alternatives are completely revamped versions of existing ones (i.e., an advanced passenger train replacing a normal railway service). As there is no explicit treatment of this problem in the literature we examined it in depth and illustrated it empirically using data especially collected to analyse mode choice in a corridor to the West of Cagliari. We propose a hopefully useful guide to this art (as no practical recipes seem to serve all purposes). Careful specification of the systematic component of utility functions in RP and SP, including the ASC, serves to illuminate the true nature of the underlying error structure in the different data sets, yielding superior forecasting models. [Copyright &y& Elsevier]

Abstract: Random utility models are typically based on the assumptions that individuals exhibit compensatory behaviour and that their choice sets are pre-specified. These assumptions may be unrealistic in many practical cases; in particular, it has been argued that thresholds may be part of choice–rejection mechanisms (i.e. if the threshold of any attribute for a given alternative is surpassed the alternative is rejected) and, therefore, act as explicit criteria to determine the set of feasible alternatives. We formulate a semi-compensatory two-stage discrete choice model incorporating randomly distributed thresholds for attribute acceptance. The formulation allows us to estimate the parameters of the threshold’s probability distribution; these parameters can be expressed as a function of the socio-economics characteristics of the individual and the conditions under which the choice process takes place. Applying the model to real and simulated data allowed us to conclude that if thresholds exist in the population, the use of compensatory models such as Multinomial Logit or even Mixed Logit, can lead to serious errors in model estimation and, therefore, in the computation of marginal rates of substitution such as the subjective value of time, as well as in prediction. Although the Mixed Logit model can deal with random parameters allowing to treat taste differences, we found it could not be used to accommodate non-compensatory behaviour or to represent the existence of thresholds. [Copyright &y& Elsevier]

Abstract: This paper explores the complexity and cognitive burden associated to stated choice experiments. Complexity is analysed in terms of design dimensions such as the number of available alternatives, the number of attributes used to define these alternatives, the number of levels for those attributes, the range of attribute levels and the number of choice situations presented to each respondent. These design dimensions were systematically varied according to an experimental design in a first design hierarchy; the second hierarchy comprised the attributes of each alternative (travel times and travel cost components). To study the complexity of the experiment we specified a heteroskedastic logit model with the scale parameter specified as a function of the design dimensions. This allowed us to separate the effects of choice complexity from the marginal utility estimates. Our results show that all five design dimensions affect the choice variance, or put it another way, they affect choice consistency. However, we did not find systematic effects on willingness-to-pay estimates. [Copyright &y& Elsevier]

Abstract: Although noise levels in urban areas frequently surpass the designated norms the consequences on health are only beginning to be examined. A stated preference experiment is used to estimate the willingness-to-pay for reducing noise levels in a group-based residential location context. The experiment considers variations in travel time to work, monthly house rent, sun orientation of the dwelling and subjective noise level inside it; objective noise levels are also measured after the experiment. Multinomial and mixed logit models are estimated based on a consistent microeconomic framework, including non-linear utility functions and allowing for various stratifications of the data. The more flexible models allow for the treatment of repeated observations problem common to stated preference data and provide a better fit to the data, although willingness-to-pay results remain almost invariant. Subjective values of time derived from these models are consistent with previous values obtained in the country, giving support to the experimental design and quality of the sample. [Copyright &y& Elsevier]

In Chile, as in most less-developed nations, if life savings are valued at all the human capital approach is used in a rather non-consistent fashion. As part of a 5-year research project on the value of transport externalities, a stated preference (SP) experiment was carried out in order to assess the value of a statistical life for Chilean interurban motorways. Interviewees had to choose among different routes for a hypothetical trip, based on the following attributes: travel time, toll charge and level of risk. The results of our experiment show that people were sensitive to the risk variable, thus “stating” a preference for safer routes. Several models were estimated with linear and non-linear utility specifications, and also incorporating the effects of socio-economic variables in a novel and interesting fashion. We were able to estimate subjective values of time consistent with previous values obtained in the country and reasonably looking values (in comparison to Chilean prices and foreign experience) of a statistical life. The paper discusses the experimental design, data collection and analysis, with emphasis on the role of lexicographic individuals that are a feature of SP studies that has not been carefully explored in the literature. We also present our modelling results and compare our derived values (of time and of a statistical life) with values found previously and/or elsewhere. [Copyright &y& Elsevier]

A stated preference ranking experiment is designed to estimate the willingness-to-pay (WTP) for reducing the amount of atmospheric pollution in a group-based residential location context. Important issues are the proper definition of the context and the variable metric for the environmental attribute. Sample members were asked to rank 10 options arising from variations in the attributes travel time to work and to study, rent of the house and an environmental attribute (the number of days of Alert, in terms of concentration of PM10, at a dwelling’s location). Multinomial logit models based on a consistent microeconomic framework were estimated for various stratifications of the data (income, pollution sensitivity, and type of dwelling currently inhabited). From these subjective values of time and WTP were derived for reductions in the number of days of alert and hence the amount of pollutant concentration at a given location. The WTP came out at about 1% of the family income for reducing one contingence day per year; this is approximately 60% higher than an estimate reported for the city of Edmonton, Canada, but the average PM10 concentration in Santiago is about six times higher. [Copyright &y& Elsevier]

• Analysis of potential greywater reuse uptake using existing consumer preferences research. • Investigate the impact of current and potential policy schemes. • Forecast the volume of water that could be recovered. • Quantify potential impact on interventions and behavioural change. • Show that addressing scepticism in the population can avoid need for monetary incentives. The residential reuse of greywater has attracted interest in recent years as a strategy to face water security problems. Nowadays, some cities such as Santiago de Chile are seeking to promote new laws that allow residential greywater reuse and make the incorporation of the necessary infrastructure (machinery and a parallel pipe system) mandatory for new buildings. The success of any such schemes, in terms of the amount of mains water that can be saved, is clearly influenced by the decision that individual consumers make on whether or not to use the parallel system, as they will also be the ones to face the potential externalities produced by the system (e.g., odours, noise from technology). Understanding and anticipating the behaviour of individuals is not an easy task, especially in the context of systems not yet widely implemented, but the groundwork has been laid with the application of approaches that allow analysts to determine the heterogeneity in consumer preferences based on the qualities of the product or service. However, there has been a lack of focus on making predictions that quantify the impact of acceptability on the volume of water recovered, driven in part by methods that been applied. This paper presents a way of predicting policy effectiveness and potential greywater reclaim benefits based on individuals' preferences. For this, we use two existing models that allow us to make predictions of greywater reuse for different domestic purposes. In a case study application to the city of Santiago de Chile, we carry out scenario tests to predict the potential uptake under potential future policy settings and show how allowing for an additional permitted use of greywater could save several hundred litres of water per month per household. [ABSTRACT FROM AUTHOR]

The huge increase in higher education coverage in many developing countries has gone hand-in-hand with an additional supply of private colleges and with the enrolment of low to middle-class students, previously excluded from a historically elitist education segment. The larger diversity of both "suppliers and consumers", unseen a few years ago, calls for methodological approaches that recognize heterogenous tastes and eventually, to classify individuals into mutually exclusive groups, something that can improve the design of public policy. The importance of college choice in educational systems using voucher schemes, makes it relevant to know what are the main variables determining such choice and whether they differ among different groups. Chile, one of the countries with the most extensive voucher system in education, experienced a significant increase in higher education enrolment (over 250% over the last 15 years), and has faced fierce political controversy due to the high heterogeneity in college quality. We use a latent class choice model to capture differences between groups, an approach that performs significantly better than simpler models previously used in this area. We found that groups of individuals with different selection patterns exist, which could not be defined ex ante , and that those groups are not only differentiated by their income level but mainly, by how they performed at high school. From the different sensitivities to college characteristics such as cost, quality, and location, identifying these groups allows us to derive different policy prescriptions. [ABSTRACT FROM AUTHOR]

We aim to understand the effect of different information types on risk perception and examine the relationship between perceived risk and travel behaviour during a pandemic outbreak. A hybrid choice model structure, incorporating a multiple discrete-continuous extreme value model, was formulated and estimated to explore travellers' mode choice and usage changes. We used a risk perception map to visually explain which risk elements felt unfamiliar and uncontrollable to travellers. Virus variation , Potential sequelae , and Long-term coexistence of coronavirus with humans were perceived as the most unfamiliar and uncontrollable risk elements. The model results indicate that increased perceived risk tends to reduce travellers' use of public transport and increase the use of shared bikes and private cars. Reducing passengers' perceived risk is critical to encourage the re-uptake of public transport in the post-pandemic era. As travellers also show significant heterogeneity, governments should aim to design targeted intervention strategies to encourage different travellers to return to public transport when considering risk communication. • This paper examines the effect of different information on perceived risk and the influence of perceived risk on travel behaviour changes during the Covid-19 pandemic. • A risk perception map is used to find out risk elements that feel unfamiliar and uncontrollable to travellers. • A hybrid choice structure, incorporating a multiple discrete-continuous extreme value (MDCEV) model is formulated and estimated to understand mode choice and changes in the frequency of usage of these modes. • As Travellers show significant heterogeneity, targeted intervention strategies for different types of travellers are required to encourage re-uptake of public transport. [ABSTRACT FROM AUTHOR]

• Evading behaviour decreases when dissatisfaction with other users' evasion behaviour increases. • Evading behaviour increases when satisfaction with reliability decreases. • Reliability is clearly the most relevant attribute related to overall satisfaction. • A behavioural fare-evader profile for Santiago is built and policy recommendations are provided. • Behavioural contagion is a plausible theoretical justification for our results. Fare evasion, in public transport (PT), is a major problem in certain countries and threatens the PT administrators' funds directly. Hence, identifying the possible causes for this behaviour can be highly relevant. Governments and PT administrators aim to deliver first-class services to satisfy customers, who will then tend to reuse the services and recommend them to friends and family, attracting new users. Customer surveys help PT administrators determining which items are more relevant for users. We hypothesise that a direct link exists between overall and attribute-specific satisfaction and fare evading behaviour; we also test the notion that the evading behaviour of other users may directly impact the satisfaction of a given user. As such, we propose a general PT satisfaction-evasion behavioural model which considers (dis)satisfaction with other users' fare evading behaviour, attribute-specific satisfaction, overall satisfaction, reuse intention, and self-reported fare evading behaviour. We base our analysis on a single intercept survey (n = 2,002), designed on the basis of focus groups conducted on users of two PT modes, Metro and bus, in Santiago de Chile. Using structural equation models (SEM) we are able to capture heterogeneous perceptions and behaviours among PT users. In particular, we found that evading behaviour increases when satisfaction with (i.e. attitude towards) other users ' fare evading behaviour increases (contagion effect), and when satisfaction with reliability decreases. We develop a behavioural fare evader profile, that could be useful in policy making. For overall satisfaction, we test a variety of models confirming that reliability is the most crucial satisfaction construct in our case. We provide policy recommendations for deterring fare evasion in Santiago based on our results, but our framework is clearly generalisable to other PT settings. [ABSTRACT FROM AUTHOR]

Public transport (PT) administrators require mechanisms to prioritise investments for improving their services, not only to maintain existing customers but also to attract new users. Customer satisfaction surveys allow detecting the level of global satisfaction with the PT system in addition to more specific satisfaction with its various attributes. Statistical analyses of these data allow determining which attributes impact more strongly on overall satisfaction. Although several econometric frameworks have been used for this task, we found only three studies loosely based on psychological theories to justify the models obtained. To address this critical gap in the literature, we postulate the existence of a Maslow's hierarchy of transit needs , with three types of attributes: functional (utilitarian), security (protection) and hedonic. To test our hypothesis, we estimate structural equation models (SEM), SEM-Multigroup, and SEMM (finite mixture) models, and assess their differences in four different cities with Bus Rapid Transit (BRT)-type systems in Latin America. Our results confirm the existence of a hierarchy in the different BRT-type system contexts, allowing to derive more generalisable conclusions. Finally, we provide direct policy recommendations by constructing a set of priorities for our case studies, concerning reliability, safety, customer services and comfort; which is generalisable to any PT system setting. • We postulate a Maslow's hierarchy of transit needs : functional, security, hedonic. • We use a benchmark survey applied to four Latin American cities with BRT-type systems. • We test our hypothesis with SEM, SEM ordinal, SEM-MGA and SEMM (mixture) models. • We confirm the hierarchy's existence with six SEM-MGA cases and two latent class models. • We offer direct policy recommendations for PT administrators. [ABSTRACT FROM AUTHOR]

We conducted a stated choice experiment with visitors to the Teide National Park (TNP), where a hypothetical park shuttle bus connecting its main points of interest was simulated. Using these data, we estimated a model focused on capturing systematic and random heterogeneity in the park visitors' preferences. We obtained visitors' willingness-to-pay (WTP) for saving time while searching for a parking space (when using cars), and also visitors' WTP for reducing the waiting time required to start the visit in the shuttle bus. Moreover, regarding the environmental impact of the visit, we obtained individuals' WTP for reducing CO 2 emissions. Our results suggest that visitors would be willing to pay nearly 11€ for reducing the time spent finding a parking space and 9€ for reducing the waiting time to start the visit, in one hour; further, they would be willing to pay 3€ for reducing in 20 g the CO 2 emissions per occupant. These values are higher for females, for German visitors and for those who are regular bike riders at their home location. Moreover, we found that visitors aged between 55 and 60 had the highest WTP. These results may be used to design transport management policies for relieving visitors' dependence on personal vehicles, helping to increase the visit quality and reduce the negative externalities associated with current mobility patterns in the park. [ABSTRACT FROM AUTHOR]

The estimation of indicators of willingness-to-pay (WTP) and the computation of associated measures of uncertainty have attracted much interest in the general area of choice modelling, but have also been characterised by a substantial amount of confusion and misguided discussions. We examine this problem in depth, both theoretically and empirically, in an effort to illustrate the most appropriate approaches to solving this problem, which has important repercussions for practice in our field. Our findings should be useful to analysts advising on transport policy and infrastructure project evaluation, in particular when requiring estimates of the value of travel time. [ABSTRACT FROM AUTHOR]

• We provide a general framework to estimate willingness-to-pay for risk reductions, allowing for respondents' use of multiple-choice heuristics. • Not allowing for the use of multiple-choice heuristics by respondents can lead to important bias in willingness-to-pay results. • Survey engagement may impact the results of non-market good valuations, challenging the classical utility maximisation approach. • Having low confidence in government may increase status-quo choices in stated choice experiments designed to estimate willingness-to pay. Estimating the value of non-market goods, such as reductions in mortality risks due to traffic accidents or air pollution, is typically done using stated choice (SC) data. However, issues with potential estimation biases due to the hypothetical nature of SC experiments arise, as protest choices are common and survey engagement is not constant across respondents. Further, if respondents choose to use different choice mechanisms and this is not considered, the results may also be biased. We designed an SC experiment to estimate the willingness to pay (WTP) for mortality risk reductions, that allowed us to simultaneously estimate the WTP to reduce the risk of traffic accident deaths and cardiorespiratory deaths due to air pollution. We formulated and estimated a multiple heuristic latent class model that also considered two latent constructs: Institutional Belief , to consider protest responses, and survey Engagement as a class membership covariate. We found, first, that individuals with lower institutional belief gave a higher probability of choice to the status-quo alternative, shying away from programs involving governmental action. Second, that not identifying respondents who do not appropriately engage in the experiment, biased the WTP estimators. In our case WTP decreased up to 26% when two different choice heuristics were allowed for in the model. [ABSTRACT FROM AUTHOR]

Service provider administrators need to identify which perceived service quality (PSQ) elements are more relevant for users. By doing this, specific tactical and operational policies can be implemented to retain and attract new customers. In the public transport (PT) arena, few PSQ studies account for both service encounter and global satisfaction. Further, although some studies consider customer heterogeneity, we believe it has not been adequately captured. Regarding the problem of modelling PSQ from a PT service provider, we present a case study from Santiago, Chile. We analyse the PSQ derived from an extensive ( n  =  25,094 ) urban bus system satisfaction survey using structural equation models (SEM). Explicitly, we incorporate heterogeneity for both travel characteristics and sociodemographic attributes utilising a Multiple Indicator Multiple Cause (MIMIC) approach. We model two simultaneous regression equations regarding satisfaction with the bus-line (service encounter) being used and with the system (global), correcting for heterogeneity in all the satisfaction constructs via the SEM-MIMIC approach. Our main result is that the most critical variable for service encounter satisfaction is frequency/waiting time. For global satisfaction, the most significant attribute is tangibles/image, which includes satisfaction with the allied Metro service, with other users’ behaviour, and with information availability. As the perceived waiting time affects all satisfaction constructs negatively, we consider it a critical policy variable to tackle. Our model can serve as a planning tool for any PT administrator. The framework applies to any service setting with independent service-specific and global satisfaction attributes. [ABSTRACT FROM AUTHOR]

Understanding individual choices over time and measuring day-to-day variability in travel behaviour is important to capture the full range of travel behaviour. Although not very common, to date several multi-day travel surveys have been conducted and panel data is available to model different transport choices. However, determining the length of a panel that allows revealing variability in travel behaviour remains an open question. Also, no final agreement has been reached about modelling the various dimensions of correlation over the repeated observations. In this paper, we use the six-week panel data from the Mobidrive survey to estimate a mode choice model that accounts for correlation across individual observations over two time periods: all days of a single week and different days of the week (e.g. all Mondays) in the wave. We first analyse these effects separately, estimating different models for each type of correlation; then we try to disentangle the relative effects of each type of correlation, estimating both types jointly. We found that both types of correlation appeared highly significant when estimated alone, while only the correlation across a given day over the six-week period remained significant, when both types were estimated jointly. This implies that for the Mobidrive panel there is much less variability in mode choice across weeks than across the days of each week. It also suggests that one week could be an appropriate length for a panel to estimate modal choice and to correctly reveal day-to-day variability. [ABSTRACT FROM AUTHOR]

This paper investigates the valuation of crowding in public transport trips and its implications in demand estimation and cost-benefit analysis. We use a choice-based stated preference survey where crowding levels are represented by means of specially designed pictures, and use these data to estimate flexible discrete choice models. We assume that the disutility associated with travelling under crowded conditions is proportional to travel time. Our results are consistent with and extend previous findings in the literature: passenger density has a significant effect on the utility of travelling by public transport; in fact, the marginal disutility of travel time in a crowded vehicle (6 standing-passengers/m 2 ) is 2.5 times higher than in a vehicle with available seats. We also compare the effects of different policies for improving bus operations, and the effect of adding crowding valuation in cost-benefit analysis. In doing that, we endogenise the crowding level as the result of the equilibrium between demand and supplied bus capacity. Our results indicate that important benefits may be accrued from policies designed to reduce crowding, and that ignoring crowding effects significantly overestimate the bus travel demand the benefits associated with pure travel time reductions. [ABSTRACT FROM AUTHOR]

Local and central authorities have long been interested in taking advantage of investments in mass transit to achieve more sustainable urban development. In the case of Santiago de Chile, a highly income-segregated metropolis that also suffers an increasing urban sprawl, its underground (Metro) and BRT corridor networks offer a unique chance to revert these malaises. In effect, Santiago's modern transit infrastructure is undercapitalized, there are many areas with little or no development in the vicinity of Metro and BRT stations that could be densified providing housing to the growing population, and hopefully even turned into sub centres with urban equipment, servicing the poorer areas, thus diminishing social and spatial segregation. We sought to understand and measure how different government incentive packages could attract private investment into such areas. We used a combination of stated choice and best-worst scaling data to examine the potential power of various government grants to encourage density development, including social integration, at these locations. Our results allowed us to identify a typology of urban areas that respond differently to the incentive packages. The differentiated analysis of the urban areas, their tendencies and perception of potential real estate developers is a significant first step to design ad hoc strategies to encourage sustainable development in the surroundings of Metro and BRT stations. [ABSTRACT FROM AUTHOR]

During the year 2020, the COVID-19 pandemic affected mobility around the world, significantly reducing the number of trips by public transport. In this paper, we study its impact in five South American capitals (i.e., Bogotá, Buenos Aires, Lima, Quito and Santiago). A decline in public transport patronage could be very bad news for these cities in the long term, particularly if users change to less sustainable modes, such as cars or motorbikes. Notwithstanding, it could be even beneficial if users selected more sustainable modes, such as active transport (e.g., bicycles and walking). To better understand this phenomenon in the short term, we conducted surveys in these five cities looking for the main explanation for changes from public transport to active and private modes in terms of user perceptions, activity patterns and sociodemographic information. To forecast people's mode shifts in each city, we integrated both objective and subjective information collected in this study using a SEM-MIMIC model. We found five latent variables (i.e., COVID-19 impact, Entities response , Health risk , Life related activities comfort and Subjective well-being), two COVID-19 related attributes (i.e. , new cases and deaths), two trip attributes (i.e., cost savings and time), and six socio-demographic attributes (i.e. , age, civil status, household characteristics, income level , occupation and gender) influencing the shift from public transport to other modes. Furthermore, both the number of cases and the number of deaths caused by COVID-19 increased the probability of moving from public transport to other modes but, in general, we found a smaller probability of moving to active modes than to private modes. The paper proposes a novel way for understanding geographical and contextual similarities in the pandemic scenario for these metropolises from a transportation perspective. [ABSTRACT FROM AUTHOR]

Binary stated choices between traveller’s current travel mode and a not-yet-existing mode might be used to build a forecasting model with all (current and future) travel alternatives. One challenge with this approach is the identification of the most appropriate inter-alternative error structure of the forecasting model. By critically assessing the practise of translating estimated group scale parameters into nest parameters, we illustrate the inherent limitations of such binary choice data. To overcome some of the problems, we use information from both stated and revealed choice data and propose a model with a cross-nested logit specification, which is estimated on the pooled data set. [ABSTRACT FROM AUTHOR]

Congestion charging is currently being considered as an important public policy in an increasing number of cities around the world, but evidence shows the importance of gaining public acceptability prior to its implementation. We analyse which factors should be considered to increase acceptability in the case of the Spanish city of Las Palmas de Gran Canaria. We applied a three-stage methodology: first a qualitative survey using focus groups, second Likert scales and exploratory factor analysis on a sample of 89 individuals, and finally, a stated choice (SC) experiment to a different sample of 206 respondents to value their preferences. The SC experiment was designed as a cordon-price scheme, including system features and considering three different uses for revenues: improving the current bus transport system, creating an underground line and increasing green areas in the city. Our qualitative analysis shows the previous resistance to accept any charging system, the lack of confidence on politicians and stresses the importance given to the use of revenues. On the other hand, values obtained from the SC experiment suggest that that public acceptability relies on the characteristics of the congestion charging scheme. In particular more than one third of the population would be willing to pay a daily fare of €2.22 if revenues from the system were used to increase the size of green areas instead of reinvesting this into the transport system. [ABSTRACT FROM AUTHOR]

Abstract: The formulation of hybrid discrete choice (HDC) models including both observable alternative attributes and latent variables associated with attitudes and perceptions has become a renewed topic of discussion in recent years. Even though there have been developments related to HDC model estimation and theoretical parameter identification, many practical and empirical issues related with HDC modelling have not been treated yet. In particular, it is known that as the HDC model estimates are not unique, it is necessary to impose some constraints on the model estimation process. In this paper we analyse the impact of different normalization approaches on parameter recovery in a simulated environment, identifying their advantages and disadvantages; we also analyse the impact of data variability on parameter recovery. We found serious problems when arbitrary values are used for normalization and when data variability is low, especially regarding the generation of the latent variables. The discrete choice model component appears to be more robust to these issues. Regarding parameter normalization, we recommend to normalize the variances associated with the HDC model’s structural equations instead of the parameters of its measurement equations, as it is done more often in practice. [Copyright &y& Elsevier]

Abstract: The coast line of the Canary Islands (Spain) has suffered from a plague of Diadema antillarum (sea urchin) over the last decades. This has attracted the attention of local authorities since it is becoming a serious environmental problem. We set out to analyse the potential market for sea urchin meat in this region, where this species is not subject to commercial exploitation and it is relatively unknown. The paper uses data from a sample of volunteers who were interviewed before and after tasting different dishes cooked with sea urchin. We applied discrete choice models considering that answers before and after experiencing this food belonged to different types of data; this is a mixed data approach in the state of practice. Our main conclusion is that there is indeed a market niche for having specialised restaurants serving sea urchin as a delicacy food. [Copyright &y& Elsevier]

Abstract: In recent years there has been a re-focus on the valuation of a statistical life from the ex post or human capital method to an ex ante willingness to pay (WTP) approach. This is in part a recognition that we may have been undervaluing the cost of fatalities and injuries to society associated with crashes, but also a strong belief in the need to focus on establishing the amount, ex ante, that individuals are willing to pay to reduce the risk of exposure to circumstances that might lead to death or degree of injury on the road network. This study has developed a framework in which to identify the degree of preference heterogeneity in willingness to pay by individuals who are drivers or passengers in cars to avoid being killed or injured. A stated choice experiment approach is developed. The empirical setting is a choice of route for a particular trip that a sample of individuals periodically undertakes in Australia. The particular trip is described in enough detail to provide the respondent with a familiar market environment, providing all the relevant background information required for making a decision. Mixed logit models are estimated to obtain the marginal (dis)utilities associated with each influence on the choice amongst the attribute packages offered in the stated choice scenarios. These estimates are used to obtain the WTP distributions for fatality and injury avoidance, which are then aggregated to obtain estimates of the value of risk reduction (VRR), of which the fatality class is also known as the value of a statistical life (VSL). [Copyright &y& Elsevier]

Abstract We analyse the relationship between transit passengers' satisfaction and loyalty. Understanding passengers' behavioural intentions after experiencing a service is an essential task for transit managers. We use structural equation models (SEM) to explore the relationship among various satisfaction latent constructs. In particular, we introduce Loyalty , which represents the intent to recommend the service. We also introduce the concept of Critical Incidents (CI), i.e. closure of a transit line in the last three months (planned) or a service anomaly in the past month (unplanned), and hypothesise that CI negatively affects any attribute-specific satisfaction construct. We additionally model the concept of Involvement , measured as the intent to participate in future public transport (PT) marketing studies, and hypothesise that both Overall Satisfaction and Loyalty may affect this variable. Finally, we conduct an SEM Multi-Group Analysis (SEM-MGA), with the objective to determine whether heterogeneity is present in passengers' satisfaction models, by incorporating users' travel and demographic characteristics (i.e. gender, age, nationality, time of day, travel frequency, and trip purpose). Our findings show that CI significantly impact all attribute-specific satisfaction constructs, specially the unplanned events during the last month. We also find that Loyalty is influenced by Overall Satisfaction and also by specific satisfaction constructs. By comparing various SEM models, we find that service satisfaction constructs, such as speed and waiting time at the platform, are the most relevant towards Overall Satisfaction. The SEM-MGA serves as a tool to test for heterogeneity in the satisfaction models within user groups. We find differences in time of day, age, and travel frequency. Finally, we consider that the Involvement relationship needs further research. Our framework allows for more detailed policy-making in PT systems regarding heterogeneous subpopulations and more concrete policy variables, such as Critical Incidents. Highlights • Research in PT loyalty is key for identifying policies to encourage transit patronage. • Overall Satisfaction and specific service quality attributes directly affect Loyalty. • Critical Incidents decrease satisfaction with all attribute-specific service elements. • Loyalty influences Involvement somewhat. Further research is warranted. • SEM-MGA verifies heterogeneity in satisfaction by time, age, and travel frequency. [ABSTRACT FROM AUTHOR]

Abstract Transportation Research (TR) was established in 1967 with the vision of promoting multi-disciplinary (economics, engineering, sociology, psychology) research on transport systems. The journal has continuously expanded its wings becoming a world-leading journal, now publishing research work through six parts, A to F, respectively addressing Policy and Practice , Methodological , Emerging Technologies , Transport and Environment , Logistics and Transportation Review , and Traffic Psychology and Behaviour. This study aims to celebrate the first half century of the journal through a bibliometric study of the publications on all six parts between 1967 and 2016. It uses the most reliable database for academic research, the Web of Science Core Collection , to identify the leading trends in all TR journals in terms of impact, topics, authors, universities, and countries. Moreover, it uses the Visualization of Similarities (VOS) viewer software to analyse bibliographic coupling, co-citation, citation, co-authorship, and co-occurrence of keywords. [ABSTRACT FROM AUTHOR]

Fare evasion is a problem in many public transport systems around the world and policies to reduce it are generally aimed at improving control and increasing fines. We use an econometric approach to attempt explaining the high levels of evasion in Santiago, Chile, and guide public policy formulation to reduce this problem. In particular, a negative binomial count regression model allowed us to find that fare evasion rates on buses increase as: (i) more people board (or alight) at a given bus door, (ii) more passengers board by a rear door, (iii) buses have higher occupancy levels (and more doors) and (iv) passengers experience longer headways. By controlling these variables ( ceteris paribus ), results indicate that evasion is greater during the afternoon and evening, but it is not clear that it is higher during peak hours. Regarding socioeconomic variables, we found that fare evasion at bus stops located in higher income areas (municipalities) is significantly lower than in more deprived areas. Finally, based on our results we identified five main methods to address evasion as alternatives to more dedicated fine enforcement or increased inspection; (i) increasing the bus fleet, (ii) improving the bus headway regularity, (iii) implementing off-board payment stations, (iv) changing the payment system on board and (v) changing the bus design (number of doors or capacity). Our model provides a powerful tool to predict the reduction of fare evasion due to the implementation of some of these five operational strategies, and can be applied to other bus public transport systems. [ABSTRACT FROM AUTHOR]

The estimation of discrete choice models requires measuring the attributes describing the alternatives within each individual’s choice set. Even though some attributes are intrinsically stochastic (e.g. travel times) or are subject to non-negligible measurement errors (e.g. waiting times), they are usually assumed fixed and deterministic. Indeed, even an accurate measurement can be biased as it might differ from the original (experienced) value perceived by the individual. Experimental evidence suggests that discrepancies between the values measured by the modeller and experienced by the individuals can lead to incorrect parameter estimates. On the other hand, there is an important trade-off between data quality and collection costs. This paper explores the inclusion of stochastic variables in discrete choice models through an econometric analysis that allows identifying the most suitable specifications. Various model specifications were experimentally tested using synthetic data; comparisons included tests for unbiased parameter estimation and computation of marginal rates of substitution. Model specifications were also tested using a real case databank featuring two travel time measurements, associated with different levels of accuracy. Results show that in most cases an error components model can effectively deal with stochastic variables. A random coefficients model can only effectively deal with stochastic variables when their randomness is directly proportional to the value of the attribute. Another interesting result is the presence of confounding effects that are very difficult, if not impossible, to isolate when more flexible models are used to capture stochastic variations. Due the presence of confounding effects when estimating flexible models, the estimated parameters should be carefully analysed to avoid misinterpretations. Also, as in previous misspecification tests reported in the literature, the Multinomial Logit model proves to be quite robust for estimating marginal rates of substitution, especially when models are estimated with large samples. [ABSTRACT FROM AUTHOR]

Approximately one car occupant per year is killed as a result of landslides or avalanches in Norway, compared to 150–200 fatalities due to accidents. Still, protection from landslides is a major transport safety issue, possibly due to the concern and dread felt by the car-driving population travelling through landslide-prone areas. A main challenge when valuing landslide risk economically lies in distinguishing it from the standard road accident risk of having a collision or running off the road. In this paper we present an approach to the valuation of landslide risk, using a stated choice experiment where internet survey respondents were asked to choose between route alternatives that differed in terms of landslide risk, casualty risk, time use and cost. Thus, landslide risk was explicitly valued as an attribute, besides travel time and casualty risk, conveying a valuation of a feature different than the risk of fatality or injury caused by ordinary road accidents. The stated route choice data were analysed using mixed logit models. We obtained point estimates for the value of landslide risk removal per kilometre driven ranging from about EUR 0.2 to EUR 0.3. These results indicate that landslide risk was perceived as something different from casualty risk related to collisions with other road users or driving off the road. More research is warranted for developing the procedures for valuation of landslide risk before such valuations can enter as input into cost-benefit analyses of landslide-reducing measures. [ABSTRACT FROM AUTHOR]

Recent research has proposed fitting responses from discrete choice experiments to asymmetric value functions consistent with prospect theory, taking into account respondents’ reference points in their valuation of choice attributes. Previous studies have mainly concentrated on travel time and cost attributes, while evidence regarding road safety attributes is very limited. This paper investigates the implicit utility of a road safety attribute, defined as the number of casualties per year in alternative car trip choices, when safety improves or deteriorates. Using appropriate statistical tests we are able to reject symmetric preferences for losses and gains in the level of safety and estimate a sigmoid value function that exhibits loss aversion and diminishing sensitivity. This adds an interesting psychological dimension to the preference of road safety. Possible implications of this finding for policy making are discussed. [ABSTRACT FROM AUTHOR]

We derive values of travel time savings (VOT) for the Madrid–Barcelona corridor, linking the two largest cities in Spain, based on the estimation of discrete choice models among the main public transport services in the corridor: air transport, high speed rail (HSR) and bus. The new HSR alternative (which started to operate in February 2008) competes directly with one of the densest airline domestic markets in the world, and its introduction produced substantial improvements in level of service, achieving reductions in travel time of more than 50% over the conventional train. A specifically designed revealed preference (RP) survey coupled with highly precise measurements of the level-of-service variables for all alternatives allowed us to estimate progressively more flexible models, the specification of which also considered different interactions of socioeconomic variables (and other important factors affecting mode choice) with the typical level-of-service attributes. We also examined income effect through the introduction of the expenditure rate multiplying all time components. This specification allowed us to analyse the income elasticity of the willingness to pay (WTP) in order to provide useful information to quantify users׳ benefits during the lifespan of a given project. We found, as expected, that HSR and air transport users exhibit substantially higher values for saving travel time than bus travellers. Also as expected, savings of waiting time are more valued than savings of access time, and the latter more valued than savings of in-vehicle travel time. We were also able to demonstrate the relevance of considering a differentiated analysis for travellers that did not pay for their travel expenses, providing a theoretically sound explanation for the large differences encountered in WTP with respect to other travellers. Finally, the obtaining of the elasticity values for the choice probabilities provided us valuable information to understand competition in this market. [ABSTRACT FROM AUTHOR]