Journal of Choice Modelling最新文献

Stated Preference (SP) valuation methods are often challenged by the existence of Hypothetical Bias (HB), often as individuals overstating their willingness to pay for a good or service in a hypothetical elicitation. A relatively new method shown to effectively reduce this upward bias is priming. However, these existing priming methods rely on relatively lengthy word or sentence tasks in order to prime respondents. Such tasks are costly in terms of survey time and participant effort, resulting in cognitive overload with benefits limited only to the elicitation. We propose a “real payment priming” method, which takes advantage of a real valuation, where actual payment would occur, prior to a hypothetical valuation. Results show that priming through real payment on one good effectively reduces potential HB in the subsequent hypothetical valuation on another good. Our method enables a wider scope of applications particularly when researchers have multiple valuation tasks, obviating the need for an extra priming task, or that the two goods are identical or similar.

Consideration set formation (CSF) is a two-stage decision process in which people first select a subset of products to consider and then evaluate and choose from the selected subset of products. CSF models typically use stated consideration or infer it from choice data probabilistically. This study explores CSF by means of eye-tracking and evaluates how measures of visual consideration compare to stated consideration. We develop a model of CSF behavior, where stated and visual consideration are embedded in the specification of the utility function. We propose three different measures of visual consideration and show that one third of respondents (∼34%) use CSF behavior and that stated consideration diverges substantially from visual consideration. Surprisingly, many product types stated as not considered receive more visual attention, not less. Our findings suggest that stated consideration may be in part a measure of preferences rather than of consideration, implying concerns with endogeneity when including stated consideration data in choice models. Accounting for CSF in discrete choice analysis increases our understanding of the decision process, and can target concerns with biased estimates when analyzing data from two-stage decision processes.

We investigate preferences for COVID-19 vaccines using data from a stated choice survey conducted in the US in March 2021. To analyse the data, we embed the Choquet integral, a flexible aggregation operator for capturing attribute interactions under monotonicity constraints, into a mixed logit model. We find that effectiveness is the most important vaccine attribute, followed by risk of severe side effects and protection period. The attribute interactions reveal that non-pecuniary vaccine attributes are synergistic. Out-of-pocket costs are independent of effectiveness, incubation period, and mild side effects but exhibit moderate synergistic interactions with other attributes. Vaccine adoption is significantly more likely among individuals who identify as male, have obtained a bachelor’s degree or a higher level of education, have a high household income, support the democratic party, had COVID-19, got vaccinated against the flu in winter 2020/21, and have an underlying health condition.

We examine identifiability and distinguishability in Decision Field Theory (DFT) models and highlight pitfalls and how to avoid them. In the past literature, the models’ parameters have been put forward as being able to capture the psychological processes in a decision maker’s mind during deliberation. DFT models have been widely used to analyse human decision making behaviour, and many empirical applications in the choice modelling domain rely solely on data concerning the observed final choice. This raises the question if such data are rich enough to allow for the identification of the model’s parameters. Insight into identifiability and distinguishability is crucial as it allows the researcher to determine which behavioural and psychological conclusions can or cannot be drawn from the estimated DFT model and how a DFT model can be specified in such a way that resulting parameters have meaningful interpretations. In this paper, we address this issue. To do this, we first show which specifications of DFT are equivalent to conventional probit models. Then, building on this equivalence result, we apply established analytical methods to highlight and explain the identification and distinguishability issues that arise when estimating DFT models on conventional choice data. We find evidence that some of the DFT models’ special cases suffer from identifiability issues. Our results warrant caution when DFT models are used to infer psychological processes and human behaviour from conventional choice data, and they help researchers choose the correct specification of DFT models.

According to the behavioral decision theory, time discounting is often used to explain preference reversals. However, the discounting theory fails to explain some types of preference reversals. Furthermore, preference reversals are limited to those along the time axis (i.e., temporal distance). To extend our knowledge of preference reversals in various choice contexts, this study constructs an analytical framework that combines the time discounting notion of behavioral decision theory and construal level theory developed in social psychology. We put forward three propositions for discounting: magnitude effect (the higher the construal level, the smaller the discounting rate), sign effect (the discounting rate is smaller for losses than for gains), and generalization of distance (discounting applies not only to temporal distance but also to psychological distances such as social distance). These propositions were validated in two studies. In Study 1, we conducted a series of three experiments on a lottery choice task using two samples of respondents (i.e., students and a web panel). In Study 2, we estimated the discounting rates of the higher and lower construal levels by employing multiple intertemporal choice tasks. While many choices involve trade-offs among attributes, the effects of changes in psychological distances are not clear. However, by identifying whether these attributes evoke high or low construal levels and whether the aspects are related to gains or losses, our approach greatly facilitates the analysis of how evaluation and preference are affected by psychological distance, and consequently, that of preference reversal behavior.

In stated preference surveys, data regarding the considered alternatives is sometimes collected prior to the preferred alternative. When the chosen alternative is not in the stated consideration set, the consideration data is inconsistent with the choice data. Several modeling approaches have been used in such situations. Some researchers ignore the consideration data and assume all alternatives are considered. Others only use the consistent choice data and delete the inconsistent observations. The most intricate methods use a latent consideration set formation approach in modeling the choice process. We extend the latent consideration set formation model to incorporate the stated consideration data but allow for inconsistencies in consideration and choice data, and allow for individual-level heterogeneity in the consideration and the choice process. We compare the recovery of the mean population preference parameters of our model with the existing approaches through simulation. The results show that if there is a similar effect of the attributes in both the consideration phase and the choice phase, the mixed logit model is not outperformed by the two-stage models. In contrast, when there is a sufficiently different effect of attributes in the consideration and the choice phase, two-stage models can recover the mean population preference parameters better than the mixed logit model. Furthermore, we can conclude that having stated consideration data barely improves the recovery of the mean preference parameters compared to a latent consideration set choice model that only uses choice data. Finally, we illustrate the models using empirical data about preferences for mobile phones.

Workforce participation decisions involve multiple stages: search, screening and offer evaluation. Standard econometric practice focusses on these stages in isolation. We conceptualize the focal behaviours as separate sequential decision stages, and provide a stated preference measurement framework for online job search and choice with a behaviourally consistent modelling approach. We demonstrate this approach in an empirical application of 275 dentists who completed an online survey including two Discrete Choice Experiments: the first mimicked an online job search site in which dentists decided which jobs they would apply to and the second presented dentists with a job offer which they accepted or rejected. Modelling these tasks requires a two-stage econometric model that incorporates the likelihood of application (first stage) into the job offer choice (second stage). The model detects differences in preferences (hence behaviours) across stages, facilitating the differentiation of policy aimed at search and job choice behaviours. Job screening occurs during search and the marginal propensity to apply for a job-type differs from the offer stage. We suggest that the approach presented provides a valuable way to investigate how dentists particularly, and perhaps the health workforce more generally, respond at different stages of workforce participation decisions and discuss practical implications.

There is an increasing interest in the use of stated preference methods to understand individuals' preferences for health and healthcare. There is also a growing interest in understanding heterogeneity in individuals' preferences. Consequently, stated preference studies frequently consider models that capture either or both observed and unobserved preference heterogeneity. A popular preliminary investigation into heterogeneity involves split-sample analysis to compare subgroups' preferences e.g., comparing patients with clinicians, or older patients with younger. In fixed-effects models, the constant variables (the individuals’ characteristics) remain stable across choice sets and therefore only enter the choice model when interacted with various attributes and/or levels. However, subgroups of respondents may differ on multiple variables that may not easily be implemented with interaction terms because of complexity and a lack of power thus only one, or a few, variables are typically taken into account in each subgroup model. This paper presents an overview of methods for matching and balancing samples to weight individuals with different characteristics in subgroup analysis and an example of how unweighted comparisons may produce erroneous conclusions regarding the degree of heterogeneity in preferences. We illustrate the issue with synthetic and empirical datasets to explore methods for matching subgroups before and within simple choice models. Our results show that entropy balancing and propensity score matching could be more appropriate than analyses using unmatched preference data when heterogeneity is driven by multiple factors. The paper concludes with a discussion of when matching and weighting may and may not be useful in healthcare decision-making.

Sustainable intensification seeks to increase outputs from existing farmland in ways that have a lower environmental impact. An extensive literature has examined the determinants of farmers' adoption of the different agro-ecological cropping systems needed to achieve these goals. However, the farmers' preferences for the attributes of these systems and the decision processes for choosing between available systems is still poorly understood. To fill this gap, this paper proposes a methodology that relies on a discrete choice experiment to analyse farmers’ preferences for cropping systems and estimate the heterogeneity of decision processes among farmers. We modelled three major types of decision processes potentially used by farmers to evaluate the systems that are not consistent with the standard utility maximization framework. These findings offer insights into the behavioural patterns of respondents and should help crop system promoters and developers to better understand how their proposed systems are likely to be evaluated by different types of farmers.

This work proposes a discrete choice model that jointly accounts for heterogeneity in preferences and in decision making procedures adopted by respondents, as well as for non-linearities in the utility function, allowing for the potential effect of salient attributes in choice experiments. We present an innovative application in the context of preferences towards nuclear energy, with data obtained from a nationwide online survey conducted in Italy. Results show that most of the variation in the choice data is indeed due to heterogeneity in the decision process, where the saliency heuristic plays an important role. Furthermore, the proposed model provides more conservative monetary valuations as opposed to standard models, potentially leading to substantial differences in cost-benefit analysis. Implications for choice modeling practitioners are discussed, emphasizing the need to account for saliency effects when modeling the choice data.