Inferring attribute non-attendance from stated choice data: implications for willingness to pay estimates and a warning for stated choice experiment design

There is a growing interest in traveller behaviour research to explore alternative information processing strategies (often referred to as heuristics or rules) adopted by individuals when assessing packages of attributes describing alternatives in a choice set, and making a choice. One popular attribute processing rule relates to attributes not being considered (i.e., being ignored), for all manner of reasons, referred to in the small but growing literature as attribute non-attendance or non-preservation. Researchers have used a mixture of methods to study the role of attribute non-attendance, including supplementary questions on whether each attribute is ignored or not, and methods in which the functional form of the utility expressions defining an alternative can recognise the possibility, up to a probability, of an attribute being ignored. Although supplementary questions are worthy of further consideration, despite the controversy as to the reliability of the response, recent interest has focused on ways to establish the incidence of attribute non-attendance without recourse to such evidence. In this paper we use an existing data set of choice amongst four attributes describing alternative car non-commuting trips, to illustrate the proposed method, and to compare values of travel time savings under each possible combination of non-attendance attributes relative to a model in which all attributes are assumed to be fully attended to. The paper reveals a major concern with the way that attribute levels and ranges are selected in the design of choice experiments, which can induce non-attendance situations where willingness to pay estimates cannot be obtained.     

Making use of respondent reported processing information to understand attribute importance: a latent variable scaling approach

In recent years we have seen an explosion of research seeking to understand the role that rules and heuristics might play in improving the predictive capability of discrete choice models, as well as delivering willingness to pay estimates for specific attributes that may (and often do) differ significantly from estimates based on a model specification that assumes all attributes are relevant. This paper adds to that literature in one important way—it explicitly recognises the endogeneity issues raised by typical attribute non-attendance treatments and conditions attribute parameters on underlying unobserved attribute importance ratings. We develop a hybrid model system involving attribute processing and outcome choice models in which latent variables are introduced as explanatory variables in both parts of the model, explaining the answers to attribute processing questions and explaining heterogeneity in marginal sensitivities in the choice model. The resulting empirical model explains how lower latent attribute importance leads to a higher probability of indicating that an attribute was ignored or that it was ranked as less important, as well as increasing the probability of a reduced value for the associated marginal utility coefficient in the choice model. The model does so by treating the answers to information processing questions as dependent rather than explanatory variables, hence avoiding potential risk of endogeneity bias and measurement error.     

Consumer valuation of new car attributes: An econometric analysis of the demand for domestic and Japanese/Western European imports

Although previous studies of new vehicle choice decisions have shown that consumers evaluate the attributes of domestic vehicles differently from those of imported vehicles, the assumption underlying these results is overly restrictive. In particular, there is no reason to assume, a priori, that the attributes of all imported vehicles have similar effects upon consumer choice behavior. A competing hypothesis, for example, is that consumers of U.S. and European automobiles have similar attribute valuations that are differentiated from those of Japanese consumers. In the present paper, a multinomial logit model of vehicle type choice is developed which more completely differentiates imports by country of origin. Using a 1985 national sample of new car buyers to estimate the model, willingness to pay measures were calculated and nested hypothesis tests on attribute valuation performed. The results supported the hypothesis that consumers value vehicle attributes differently depending upon its country of origin. Moreover, consumers do not have similar valuations on the attributes of all imports implying that an arbitrary distinction between domestic and imported vehicles may produce misleading results.     

Identifying a behaviourally relevant choice set from stated choice data

Stated choice experiments are designed optimally in a statistical sense but not necessarily in a behavioural choice making sense. Statistical designs, and consequently model estimation, assume that the set of alternatives offered in the experiment are processed by respondents with a specific processing strategy. Much has been studied about attribute processing using discrete choice methods in travel choice studies, but this paper focuses more broadly on processing of alternatives in the choice set offered in the experiment. This paper is motivated by the primary idea that the distribution of predicted choice probabilities associated with a set of alternatives defining a given choice set might provide strong evidence on the strategies that agents appear to use when choosing a preferred alternative. In an empirical setting of a choice set of size three, four model specifications are considered including a model for the selection of the best alternative in the full choice set and three variants of a best–worst regime. Using state choice data on road pricing reform, the empirical analysis examines which model specification delivers the most accurate prediction of the chosen alternative. The results suggest which alternatives really matter in choice making and hence the alternatives that might be included in a choice set for model specification.     

Parameter transfer of common-metric attributes in choice analysis: implications for willingness to pay

There is a growing literature that promotes the presence of a mix of compensatory and semi-compensatory processing strategies in the way that individuals evaluate packages of attributes in real or hypothetical markets, and make choices. This paper proposes a specification for the utility form in a choice model to test if, given a pair of attributes with a common-metric (e.g., components of travel time or cost), the attribute with the dominating level defines the marginal (dis)utility that is assigned to both attributes. We refer to this processing strategy as a parameter transfer rule. We use a stated choice data set, in the context of car driving individuals choosing between tolled and non-tolled routes, to estimate a mixed logit model which incorporates the presence of the parameter transfer rule and the conventional fully compensatory rule, both existing up to a probability. We find that if this parameter transfer heuristic is part of the mix, the WTP is more than 30% higher, on average, than when only a fully compensatory rule is imposed. We also contrast the parameter transfer rule with other semi-compensatory heuristics which have been investigated in other papers, and show that the finding adds further support to the accumulating evidence that a semi-compensatory attribute processing rules tend to result in higher mean WTP estimates compared to the fully compensatory attribute processing rule.     

Experience as a conditioning effect on choice: Does it matter whether it is exogenous or endogenous?

Previous choice studies have proposed a way to condition the utility of each alternative in a choice set on experience with the alternatives accumulated over previous periods, defined either as a mode used or not in a most recent trip, or the mode chosen in their most recent trip and the number of similar one-way trips made during the last week. The paper found that the overall statistical performance of the mixed logit model improved significantly, suggesting that this conditioning idea has merit. Experience was treated as an exogenous influence linked to the scale of the random component, and to that extent it captures some amount of the heterogeneity in unobserved effects, purging them of potential endogeneity. The current paper continues to investigate the matter of endogeneity versus exogeneity. The proposed approach implements the control function method through the experience conditioning feature in a choice model. We develop two choice models, both using stated preference data. The paper extends the received contribution in that we allow for the endogenous variable to have an impact on the attributes through a two stage method, called the Multiple Indicator Solution, originally implemented in a different context and for a single (quality) attribute, in which stage two is the popular control function method. In the first stage, the entire utility expression associated with all observed attributes is conditioned on the prior experience with an alternative. Hence, we are capturing possible correlates associated with each and every attribute and not just one selected attribute. We find evidence of potential endogeneity. The purging exercise however, results in both statistical similarities and differences in time and cost choice elasticities and mean estimates of the value of travel time savings. We are able to identify a very practical method to correct for possible endogeneity under experience conditioning that will encourage researchers and practitioners to use such an approach in more advanced non-linear discrete choice models as a matter of routine.

     

Aggregation of common-metric attributes in preference revelation and implications for willingness to pay

There is a growing interest in process heterogeneity in the way that individuals evaluate packages of attributes in real or hypothetical markets and make choices. We consider the role of the relative magnitude of pairs of attributes that are defined on a common metric (e.g., minutes or dollars), to look at the extent to which attributes might be added in preference revelation, in contrast to the commonly adopted single rule of compensatory behaviour. The focus is on a choice model specification that allows for different treatments of pairs of attributes across a sample, in contrast to studies that impose a single rule on all observations, and that does not require supplementary information on whether specific individuals claimed to have added up attributes; rather we structure a non-linear utility function that permits a probabilistic aggregation of each attribute. We translate this into a willingness to pay for travel time-savings for car commuters, in the context of tolling roads in Sydney, and contrast it with the results from the additive model, and a model where self-stated attribute processing information is taken into account. The empirical evidence suggests that mean willingness to pay increases when the addition rule is accounted for. This is a potentially important message for environmental applications where two or more attributes have a common metric.     

Towards a practical method to establish comparable values of travel time savings from stated choice experiments with differing design dimensions

Individual’s process the information in stated choice (SC) experiments in many different ways. In order to accommodate decisions rules that are used in processing information, there is good sense in conditioning the parameterisation of stated choice design attributes on these rules. In particular, rules might be invoked to cope with the dimensionality of the SC design. In this paper, we investigate the impact of rules such as attribute aggregation and reference dependency on preference profiles for specific design attributes, as well as the design specification, as we vary the dimensionality of an SC design. The heteroscedastic extreme value logit model is estimated to identify the role of design dimensionality and attribute processing rules, after accounting for scale differences across sixteen pooled data designs The empirical evidence, drawn from a study in Sydney of car commuter route choice, suggests that accounting for the way that stated choice designs are processed, given their dimensionality, does make a statistically significant difference on measures of willingness to pay, as does accounting for scale differences between pooled data designs. This evidence has practical value in guiding the design of SC experiments and in adjusting results from different SC designs when comparing the evidence. We propose a simple adjustment formula to use in adjusting VTTS from different studies so that they are comparable. From a practical policy perspective, the evidence sends a warning about the risk of undervaluing mean VTTS if the attribute processing rules are not accounted for.     

Extending stated choice analysis to recognise agent-specific attribute endogeneity in bilateral group negotiation and choice: a think piece

This paper is a think piece on variations in the structure of stated preference studies when modelling the joint preferences of interacting agents who have the power to influence the attribute levels on offer. The approach proposed is an extension of standard stated choice methods, known as ‘stated endogenous attribute level’ (SEAL) analysis. It allows for interactive agents to adjust attribute levels off a base stated choice specification that are within their control, in an effort to reach agreement in an experimental setting. This accomplishes three goals: (1) the ability to place respondents in an environment that more closely matches interactive settings in which some attribute levels are endogenous to a specific agent, should the modeller wish to capture such behaviour; (2) the improved ability of the modeller to capture the behaviour in such settings, including a greater wealth of information on the related interaction processes, rather than simply outcomes; and (3) the expansion of the set of situations that the modeller can investigate using experimental data.

Users of a regional telephone-based traveler information system – a study of TravInfo™ users in the San Francisco Bay Area

This paper reports the results of a 1998 survey carried out among a small sample of people who had called the TravInfo Travel Advisory Telephone System (TATS) in April 1997 looking for current information about traffic conditions in the San Francisco Bay Area. The survey employed a series of tradeoff questions designed to identify the specific attributes of the information that the respondents felt to be most important, and to estimate the relative values that these users had for various possible information improvements. Despite the small size of the sample, it proved feasible to investigate how the implicit valuations placed on each attribute of the service (update frequency, extent of road coverage, and level of customization) varied among different user segments. For the most part, respondents appeared to value basic enhancements – those that establish an initial quality differential above the baseline set by free broadcast traffic information – more than they value further subsequent improvements. Overall, more frequent information updates are the highest priority among the range of possible enhancements explored in the survey, followed by an extension of coverage to include major arterials in addition to freeways. However, the evidence suggests relatively little value in door-to-door coverage – that is, coverage of streets beyond the freeway and major arterial networks. This research produced some estimates of the average monetary values ("willingness to pay") that the sample of current users attached to information improvements of various kinds. While these values varied among subgroups in ways that fit a priori expectations, it is not clear whether they provide an unbiased picture of the users' absolute levels of willingness to pay, such that they could be used reliably to inform (for example) information pricing decisions. The value of this research lies more clearly in the light it sheds on the relative structure of user preferences, and on the variations in preferences between information attributes and market segments. Overall, the authors conclude that the prospects for self-sustaining ATIS services are unclear. In response to a direct question, a majority of users (perhaps influenced by a strategic bias) indicated that they were unwilling to pay for ATIS services; some of these same people, however, indicated later that they might indeed be willing to buy a particular enhanced ATIS package, in direct competition with free broadcast services.     

A practical concern about the relevance of alternative-specific constants for new alternatives in simple logit models

When a “new” alternative is introduced, post-estimation, into a logit model, analysis initially exclude an ASC or occasionally assume a correspondence with an existing alternative. Using a recent data set for mode and route choice, the paper highlights the forecasting implications of ignoring the ASC in the utility expression of a “new” alternative, and how sensitive the market shares are to the inclusion/ exclusion of the ASC.     

Design and development of a stated choice experiment for interdependent agents: accounting for interactions between buyers and sellers of urban freight services

Stated choice experiments have proven to be a powerful tool in eliciting preferences across a broad range of choice settings. This paper outlines the elements of a group-based experiment designed for interdependent urban freight stakeholders, along with the procedure to administer the questionnaire sequentially. The focus is on the design of a computer-assisted personal survey instrument and the value in disseminating the details of a new approach to design and collect stated choice data for interacting agents. The paper also discusses how to specify a reference alternative, and then how to recruit appropriate real-market or representative decision-making group members to participate in a subsequent phase of the survey, which incorporates the reference alternative and contextual information from an initial phase. The empirical strategy, set out in some detail, provides a new framework within which to understand more fully the role that specific attributes, such as variable user charges, influencing freight distribution chains might play, and who in the supply chain is affected by specific attributes in terms of willingness to pay for the gains in distribution efficiency.

Embedding risk attitude and decision weights in non-linear logit to accommodate time variability in the value of expected travel time savings

In recent years we have seen important extensions of logit models in behavioural research such as incorporation of preference and scale heterogeneity, attribute processing heuristics, and estimation of willingness to pay (WTP) in WTP space. With rare exception, however, a non-linear treatment of the parameter set to allow for behavioural reality, such as embedded risk attitude and perceptual conditioning of occurrence probabilities attached to specific attributes, is absent. This is especially relevant to the recent focus in travel behaviour research on identifying the willingness to pay for reduced travel time variability, which is the source of estimates of the value of trip reliability that has been shown to take on an increasingly important role in project appraisal. This paper incorporates, in a generalised non-linear (in parameters) logit model, alternative functional forms for perceptual conditioning (known as probability weighting) and risk attitude in the utility function to account for travel time variability, and then derives an empirical estimate of the willingness to pay for trip time variability-embedded travel time savings as an alternative to separate estimates of time savings and trip time reliability. We illustrate the richness of the approach using a stated choice data set for commuter choice between unlabelled attribute packages. Statistically significant risk attitude parameters and parameters underlying decision weights are estimated for multinomial logit and mixed multinomial logit models, along with values of expected travel time savings.     

Bayesian imputation of non‐chosen attribute values in revealed preference surveys

Obtaining attribute values of non‐chosen alternatives in a revealed preference context is challenging because non‐chosen alternative attributes are unobserved by choosers, chooser perceptions of attribute values may not reflect reality, existing methods for imputing these values suffer from shortcomings, and obtaining non‐chosen attribute values is resource intensive. This paper presents a unique Bayesian (multiple) Imputation Multinomial Logit model that imputes unobserved travel times and distances of non‐chosen travel modes based on random draws from the conditional posterior distribution of missing values. The calibrated Bayesian (multiple) Imputation Multinomial Logit model imputes non‐chosen time and distance values that convincingly replicate observed choice behavior. Although network skims were used for calibration, more realistic data such as supplemental geographically referenced surveys or stated preference data may be preferred. The model is ideally suited for imputing variation in intrazonal non‐chosen mode attributes and for assessing the marginal impacts of travel policies, programs, or prices within traffic analysis zones. Copyright © 2012 John Wiley & Sons, Ltd.     

Random utility models with implicit availability/perception of choice alternatives for the simulation of travel demand

Random utility models are undoubtedly the most used models for the simulation of transport demand. These models simulate the choice of a decision-maker among a set of feasible alternatives and their operational use requires that the analyst is able to correctly specify this choice-set for each individual.Some early applications basically ignored this problem by assuming that all decision-makers chose from the same pre-specified choice-set. This assumption may be unrealistic in many practical cases and cause significant misspecification problems (P. Stopher, Transportation Journal of ASCE 106 (1980) 427; H. Williams, J. Ortuzar, Transportation Research B 16 (1982) 167).The problem of choice-set simulation has been dealt within the literature following two basically different approaches:

• •simulating the perception/availability of an alternative implicitly in the choice model,
• •simulating the choice-set generation explicitly in a separate model.

The implicit approach is more convenient from an operational point of view, while the explicit one is more appealing from a theoretical point of view.In this paper, a different approach to the modeling of availability/perception of alternatives in the context of random utility model is proposed. This approach is based on the concept of intermediate degrees of availability/perception of each alternative simulated through a model (or “inclusion function”) which in turn is introduced in the systematic utility of standard random utility models.This model, named implicit availability/perception (IAP), may be differently specified depending on assumptions made on the joint distribution of random residuals and the way in which the average degree of availability/perception is modeled.In this paper, a possible specification of the IAP model, based on the assumption of random residual distributed as i.i. Gumbel and with the average degree of availability/perception modeled as a binomial logit, is proposed.The paper also proposes ML estimation models in two cases: in the first, only information on alternatives choices is available, while in the second, this information is complemented with others on variables related to a latent (i.e., non-observable) alternatives availability/perception degree (e.g., information on car availability of decision-maker i used as an indirect measurement of the unknown and non-observable availability/perception degree of alternative car for decision-maker i in a modal split).The proposed specification is tested on mode choice data; the calibration results are compared with those of a similar logit specification with encouraging results in terms of goodness of fit.