Journal of Choice Modelling最新文献

In travel choice models, variables describing alternative attributes such as travel time may have to be specified as stochastic because the analyst may not have accurate measurements of the attribute values considered by the decision-maker. Such stochasticity in alternative attributes is different from unobserved heterogeneity in the coefficients representing travellers’ response to those attributes. Specifying only one of these as random while keeping the other fixed can potentially result in biased parameter estimates, inferior goodness-of-fit, and distorted information for policy analysis. Therefore, in this study, we propose an integrated choice and stochastic variable modelling framework with random coefficients (i.e., an ICSV-RC framework) that allows the analyst to accommodate stochasticity in alternative attributes and random coefficients on such attributes. In addition, we show that ignoring either source of stochasticity – stochasticity in alternative attributes or unobserved heterogeneity in response to the attributes – results in models with inferior goodness-of-fit and a systematic bias in all parameter estimates. We demonstrate this using simulation experiments for two different travel choice settings, one involving labelled mode choice alternatives and the other involving unlabelled route choice alternatives. In addition, we present an empirical analysis in the context of truck route choice to highlight the importance of accommodating both sources of variability – stochasticity in travel times and random heterogeneity in response to travel times.

In this paper, we assess the role of emotions in choices. We elicited emotions using an innovative facial expression analysis approach, comparing the results to those from a traditional hedonic liking scale. To this end, we conduct an experiment combing surveys and actual wine tasting. The results show a positive and significant relationship between positive emotions (joy), positive experiences (valence), and wine choices, especially in the case of wines with credence attributes (organic and selected vintage organic wines). Moreover, we conclude that facial recognition mechanisms can be useful to elicit evoked emotions and provide complementary information to the traditional liking scales to better understand decision making.

Discrete choice experiments (DCEs) are increasingly used in several scientific domains. Since their results may be used to inform governmental decision-making, it is important that the validity of the method is continuously scrutinized. An often-studied design artefact is the impact of the presentation order of alternatives, attributes, and choice sets on the results of a DCE. No systematic review of the literature on ordering effects existed until now, and many applied studies using a DCE do not explicitly consider the role of ordering effects. I conducted a systematic review of the literature on ordering effects in this study. Using a three-step snowball sampling strategy, 85 studies were identified across various scientific domains. The majority of included studies documented statistically significant ordering effects. Alternative and attribute ordering effects are primarily caused by lexicographic behaviours, while choice set ordering effects seem to be caused by respondent learning, fatigue, or anchoring. Although ordering effects may not always occur, the majority of studies that did find statistically significant effects warrants the use of mitigation methods. An overview of potential mitigation methods for the applied DCE literature is presented, including randomization of presentation orders, advance disclosure of DCE core elements, and inclusion of alternative-specific constants (ASCs), attribute level overlap, and an instructional choice set (ICS). Finally, several directions for future methodological research on this topic are provided, particularly regarding heterogeneity in ordering effects by study design traits and respondent characteristics, and interactions between ordering effects. Insights in these aspects would further our understanding of respondents’ processing of DCEs.

Pooling data from different subgroups offers advantages of shrinking standard errors and simplifying characterization of the data structure. The ability to pool data also facilitates meta-analysis to evaluate consensus among multiple studies and to inform benefit transfer to new choice settings. Testing for poolability requires accounting for differences in response variance or scale among subgroups. This is commonly done by assuming a single scale factor within each subgroup of interest. This assumption may not hold for many subgroups, especially when task non-attendance is present. We use data from a prior DCE study to show that task non-attendance, and by extension the assumption of a single scale factor across subgroups, can lead to inaccurate conclusions when determining poolability. To address this concern, we propose a latent-class/random-parameters Logit (LCRP) model specification that accommodates task non-attendance or other causes of scale differences within subgroups and directly tests for poolability.

Departure time choice is a key component of travel behavior that directly influences the spatial and temporal distribution of travel demand. This paper tries to develop a modeling framework for choosing the departure time that minimizes travel costs. In this regard, a modeling framework for generating departure time recommendations is proposed and applied to real commuting trips. The methodology is an extension of the departure time choice model with unreliable travel time. Two cases are considered. The first calculates the optimal time of departure when the mean of the travel time varies by time of day but the variance is constant. An exact solution to the departure time choice problem is provided for this case. In the second case, both the mean and variance vary with the time of day. A numerical solution is proposed; it is proved that the sequence of the numerical solution is contractive with a unique fixed point obtainable for any initial guess. We apply both to the departure time planning problem for a transportation operator that offers repetitive mandatory trips on a dense network. The case study offers two insights into departure time choice analyses. First, the assumption that the travel time variance at peak hours is constant induces biases for the optimal departure time. However, this assumption provides plausible results for the off-peak period. Second, travelers relying on personal judgment may have significantly different costs of travel than passengers making their decisions based on the system's recommendations.

With the increased affordability of eye-tracking technology, its applications in discrete choice experiments (DCEs) are rapidly increasing. It is critical to understand the current state of research, challenges, and potential value of this technology for future studies. This article provides an interdisciplinary perspective on three main themes of this literature – (i) utilizing visual attention measures to identify the effect of top-down and bottom-up processing on information search and preferences, (ii) modelling advancements to incorporate visual attention measures into the discrete choice models, (iii) examining the effect of the DCE design on the consumer's information search processes. Then, we highlight four areas of improvement in these themes. First, visual attention measures alone might not be sufficient proxies for representing information processing. We lay out a research agenda to precisely measure information processing by integrating eye-tracking and electroencephalogram (EEG) data. Second, traditional static behaviour models do not effectively leverage the dynamic nature of eye-tracking data. We propose to adapt dynamic behavioural models from cognitive psychology where the mathematical representation of the decision-making process is consistent with the eye-tracking data. Third, existing studies provide descriptive (instead of prescriptive) insights about the effect of DCE design on information search. Thus, instead of DCE design, eye-tracking data can be used ex-post to select behavioural models aligned with observed search patterns. Fourth, convenience sampling protocols in eye-tracking studies raise questions about the internal validity of findings. Future DCEs with eye-tracking should adopt protocols used in randomised control trial studies.

This study investigates the process of stage transition in traveler's adoption of carsharing. The carsharing adoption behaviors are classified into five stages using the transtheoretical model: precontemplation (PC), contemplation (C), preparation (PA), action (A), and maintenance (M). In Beijing, a comprehensive survey combining stated preference and revealed preference methods is conducted to collect data on travelers' adoption attitudes and related attributes, such as socioeconomic characteristics, travel features, and carsharing usage environments. Specifically, ten latent attitudinal variables are derived from the technology acceptance model and theory of planned behavior. An integrated choice and latent variable model is developed to examine the impact of these explanatory factors on the transition processes between the five adoption stages. The findings reveal that the effects of explanatory factors on stage transitions in carsharing adoption behaviors differ across stages. Factors such as education level, income, having children, car ownership, familiarity with carsharing, personal condition, and social pressures play a crucial role in the early stages of understanding before carsharing adoption, namely the PC-C and C-PA stage transitions. Conversely, in the later stages (PA-A and A-M), travelers show a greater emphasis on carsharing infrastructure, including station reachability, vehicle availability, and perception toward rental stations and shared vehicles. Additionally, factors related to travel habits and expectations are influential throughout all stages of the carsharing adoption process. Lastly, three categories of measures, namely soft, neutral, and hard measures, are derived from the results to guide policymakers and carsharing operators in formulating strategies for the operation and management of carsharing systems.

In response to the Covid-19 pandemic, many countries have adopted measures to contain the spread of the virus, including mandatory quarantine for inbound travellers. This research investigates the preferences of residents of New South Wales, Australia, towards the mandatory quarantine protocol adopted in the state. Heterogeneity in individual preferences is explored by advancing the Logit Mixed Logit (LML) model defined by Train (2016). Two approaches are suggested to decompose individual heterogeneity in this framework and are applied to data collected via a stated preference experiment. The empirical findings demonstrate that on average, the community prefers returned travellers be quarantined in dedicated quarantine facilities rather than be quarantined at home or using hotels, but are mostly indifferent to how long travellers are quarantined for, and how many travellers are allowed to return to Australia. The sample do however have a preference, on average for travellers having to pay less to quarantine, meaning they wish to see greater government subsidies. However, the modelling approach demonstrates that the common use of averages potentially masks diverse preferences, and is not representative of community wants and desires, thus possibly leading to incorrect inferences about policy impacts.

Community support for climate change adaptation projects markedly benefits effective protection. A relevant driver of community support is the perceived attribution of responsibility to individuals. If individuals attribute responsibility for adaptation to others, e.g. public authorities, this reduces the adaptation efforts of the individual, might induce preference uncertainty, and can lead to maladaptation. We study individuals' perceptions of personal responsibility and preferences for coastal protection in a setting in which individuals have little formal responsibility. To do so, we collect data from the Maldives, a small island development state with significant risks of seaborne hazards where responsibility for coastal protection formally rests with the central government without significant involvement of local communities. Using survey measures and a Discrete Choice Experiment (DCE), we investigate respondents' sense of personal responsibility and their preferences for climate change adaptation distinguishing between preferences for hard, man-made structures and soft, working-with-nature protection approaches. The results show that responsibility perception plays an important role for stated willingness to support protective measures. However, they further show a mismatch between formally assigned and perceived responsibility for protection with a majority of respondents having a strong sense of personal responsibility for protection. In addition, the DCE results indicate a misalignment of people's preferences and the measures implemented by the government. While the latter belong to the group of hard protection measures, the majority of respondents show a clear preference for soft protection. We discuss the implications of these findings and highlight the importance of a better understanding of drivers of responsibility perceptions.

Using rank-ordered logit regression, researchers typically analyze consumer preference data collected with Best-Worst Scaling (BWS) surveys. We propose a generalized rank-ordered logit (GROL) model that allows flexibility in modeling preference heterogeneity. The GROL and mixed rank-ordered logit model (MROL) accommodate preference heterogeneity. However, the GROL also allows one to model heterogeneity as a function of demographic or environmental variables. A Monte Carlo experiment compares the estimates of accuracy and precision of the proposed GROL estimation with the MROL specification. Simulation results suggest that the GROL model performs comparatively well when the GROL or the MROL is the true data-generating process (dgp). Coefficient and willingness-to-pay estimates of the GROL are more precise and accurate compared to the MROL when the MROL is the true dgp. We surmise that the increased precision of the GROL estimator arises from the added flexibility for modeling different sources of heterogeneity. An empirical application analyzes a BWS survey on consumer preferences for single-use eating-ware (SUEW) products made from biobased materials. Findings suggest that consumers value most product degradability and using non-plastic materials to fabricate SUEW. Consumers also valued the rapidity of product degradability and using non-plastic materials to make SUEW plates. Respondent attentiveness also affected willingness-to-pay (WTP) estimates across attributes. Results suggest attentive respondents were about $3.00 more WTP for biodegradable SUEW than inattentive respondents.