Construct Validity of Dichotomous and Polychotomous Choice Contingent Valuation Questions

In this note we conduct construct validity tests for dichotomous choice (DC) and polychotomous choice (PC) contingent valuation questions. Contrary to previous results, we find that DC and PC estimates of willingness to pay are theoretically valid, convergent valid, and similar in terms of statistical precision. Similar to previous results, PC respondents are less sensitive to information than DC respondents. We conclude that DC and PC valuation questions are construct valid for this study. Sequential PC valuation questions could be used in studies where obtaining information about the certainty or intensity of respondent preferences would be useful.     

The impact of offering two versus three alternatives in choice modelling experiments

Researchers designing choice modelling experiments have some latitude over the number of choice alternatives that can be offered in each choice set. There is some evidence that design dimensions, including the number of alternatives available in each choice set, can influence model outcomes. A key issue is whether referendum formats with binary options are preferable to choice sets with multiple alternatives.A choice modelling experiment was performed where questionnaires delivered to two split samples differed only according to whether two or three alternatives were offered to respondents in each choice set. The results indicate that more robust models could be constructed from the three-alternative split compared to the two-alternative split. One reason for the difference is that respondents tended to display serial non-participation in the two-alternative format, choosing an alternative consistently without regard for changes in the attributes. For practitioners of the CM technique, the results suggest that it may be preferable to offer more than two alternatives in a choice set that includes a status quo option.     

Should Reference Alternatives in Pivot Design SC Surveys be Treated Differently?

Analysts are increasingly making use of pivot style Stated Choice (SC) data in the estimation of choice models. These datasets often contain a reference alternative whose attributes remain invariant across replications for the same respondent. This paper presents evidence to suggest that respondents react differently to the attributes of these reference alternatives and those of purely hypothetical alternatives. While some such evidence has been reported in the existing literature, this paper goes further and details a number of different departures from a common treatment of the two types of alternatives, relating both to the observed part of utility and the error term.     

Revealing Differences in Willingness to Pay due to the Dimensionality of Stated Choice Designs: An Initial Assessment

Stated choice (SC) methods are now a widely accepted data paradigm in the study of the choice responses of agents. Their popularity has spawned an industry of applications in fields as diverse as transportation, environmental science, health economics and policy, marketing, political science and econometrics. With rare exception, empirical studies have used a single SC design, in which the numbers of attributes, alternatives, choice sets, attribute levels and ranges have been fixed across the entire design. As a consequence the opportunity to investigate the influence of design dimensionality on behavioural response has been denied. Accumulated wisdom has promoted a large number of positions on what design features are specifically challenging for respondents; and although a number of studies have assessed the influence of subsets of design dimensions, there exists no single study (that we are aware of) that has systematically varied all of the main dimensions of SC experiments. This paper reports some initial findings on what influences, in aggregate, specific design configurations have on the mean willingness to pay for specific attributes using a Design of Designs (DoD) SC experiment in which the ‘attributes’ of the design are the design dimensions themselves. The design dimensions that are varied are the number of choice sets presented, the number of alternatives in each choice set, the number of attributes per alternative, the number of levels of each attribute and the range of attribute levels. The empirical evidence, using a sample of respondents in Sydney choosing amongst trip attribute bundles for their car commuting trip, suggests that, within the boundaries of design dimensionality investigated, mean estimates of WTP for travel time savings in the aggregate cover a range that is appropriate for reporting a global mean and a set of meaningful values for sensitivity testing in project appraisal and demand prediction. When these aggregated mean estimates are conditioned on all design dimensions we do not find any systematic differences due to specific design dimensions; however when each design dimension is assessed without controlling for the other dimensions we find evidence to support differences in aggregate mean WTP attributable to the number of attributes per alternative and the number of alternatives in a choice set. Research funded under the Australian Research Council Large Grants Scheme, Grant A00103962.     

Risk–risk versus standard gamble procedures for measuring health state utilities

Contingent valuation (CV) has been widely used to measure the potential benefits derived from different policy decisions. However, doubt now exists about the validity of the CV method and alternative approaches to benefit valuation have been proposed. The paper reports on the results of a study which was designed to test the viability of two of the most prominent of the alternatives: the risk–risk (RR) and standard gamble (SG) approaches. If individual preferences are consistent with the axioms of von Neumann–Morgenstern expected utility theory (EUT) then the two methods should generate the same interval scales for any given set of health states. However, the results show that SG utilities are substantially higher than RR ones, thus casting doubt on these axioms. The paper discusses alternatives to EUT which might better expalin the discrepancies found. It also considers whether the results might be explained in terms of status-quo bias and/or by the relative difficulty of RR questions. The results presented may have important implications for other areas of applied research in which there exists uncertainty about outcomes.     

Attending to the Reasons for Attribute Non-attendance in Choice Experiments

This paper focuses on behavioural reasons underlying stated attribute non-attendance in choice experiments. In order to identify and incorporate procedures for dealing with heterogeneous attribute processing strategies, we ask respondents follow-up questions regarding their reasons for ignoring attributes. Based on these statements, we conclude that the standard way of assigning a zero impact of ignored attributes on the likelihood is inappropriate. We find that some respondents act in accordance with the passive bounded rationality assumption since they ignore an attribute simply because it does not affect their utility. Excluding these genuine zero preferences, as the standard approach essentially does, might bias results. Other respondents claim to have ignored attributes to simplify choices. However, we find that these respondents have actually not completely ignored attributes. We argue along the rationally adaptive behavioural model that valid preference information may indeed be elicited in these cases, and we illustrate how recoding of non-attendance statements conditional on stated reasons may be a more appropriate solution than the current standard way of taking stated non-attendance into account.     

Conservative dichotomous choice responses in the active policy setting: DC rejections below WTP

An important feature of a Contingent Value (CV) study is that researchers design a survey that guides respondents to answer dichotomous choice (DC) questions as if they represent once-and-for-all choices. Researchers frequently construct hypothetical markets to satisfy this condition; yet detractors assert that ‘hypotheticality’ leads inevitably to inflated DC responses. For active policy questions, however, some respondents may suspect that a CV informs an actual policy issue; so to reject a DC might induce the policy-maker to reintroduce the policy with a price reduction or a program improvement. With potential incentives to deflate a DC response when policies are active, we locate two types of respondents that represent two different incentives. One class is expected to be able to risk permanent rejection of a waiver from one automobile emissions inspection. This class more frequently rejects a DC value known to improve existing conditions. Another respondent class is expected to be risk averse to defeat of the program or to excessive delay. Predictably, these respondents more frequently accept a DC value that represents a known gain. Conservative DC responses have implications for the use of CV in active policy contexts, opening a role for theory to assist practitioners in these circumstances.      

Quantifying public preferences for agri-environmental policy in Scotland: A comparison of methods

This paper compares two methods for determining policy priorities for reform of Scottish agricultural support. Multifunctional agriculture attempts to establish a new balance between traditional commodity support and payment for the production of non-market goods and services that are increasingly demanded by the public. Supplying non-market goods presents particular problems for optimal policy design, not least the elicitation of consumer demand for those goods. From public focus groups, a range of attributes was derived as central to the Scottish public’s preferences for future agri-environmental reform. This information was then combined in two separate survey methods using the Analytical Hierarchy Process (AHP) and choice experiments (CE). Both applications suggest that the public has defined preferences and a willingness to pay (using general income taxation) to affect changes beyond the status quo, and that policy payments should be targeted towards both environmental and social benefits. The divergent preference orderings derived from the alternative methods can be considered in the light of previous methodological debates on question framing, bounded rationality and respondent uncertainty. We speculate about the validity of alternative methodologies for informing particular policy questions.     

Attribute Causality in Environmental Choice Modelling

When selecting attributes in environmentalChoice Modelling studies, preference should begiven to those attributes that aredemand-relevant, policy-relevant, andmeasurable. The use of these criteria willoften result in a short list of environmentalattributes of which some are causally related. The inclusion of attributes that have a``cause-effect' relationship may stimulate somerespondents to seek to understand the causalrelations among attributes in order to assigngreater meaning to the alternatives, andpotentially, simplify the decision makingprocess. This may have implications for theweights they assign to each of the attributeswhen identifying the preferred alternatives,and subsequently for the implicit prices and/orwelfare estimates. A test of the impact ofincluding an attribute that causesimpacts on ecosystem health as well as anattribute relating to ecosystem health effectson parameter estimates, implicitprices and welfare estimates is conducted. Twoquestionnaires are developed, one with the`causal' attribute included and one without. Acomparison of results indicates that when the`causal' attribute is included in the vector ofchoice attributes, the implicit value of asingle endangered species falls by 34 per centwhilst no significant difference is detected inthe parameter estimates. Importantly, however,estimates of compensating surplus for a givenpolicy package do not differsignificantly across the two treatments. Thisimplies that to the extent that the inclusionof a `causal' attribute reduces the implicitprices for one or more of the `effect'attributes, the associated loss in utility isapproximately offset by the utility nowassociated with the new attribute.     

What are the consequences of ignoring attributes in choice experiments? Implications for ecosystem service valuation

This paper investigates the sensitivity of choice experiment values 3AL for ecosystem services to ‘attribute non-attendance’. We consider three cases of attendance, namely that people may always, sometimes, or never pay attention to a given attribute in making their choices. This allows a series of models to be estimated which addresses the following questions: To what extent do respondents ignore attributes in choice experiments? What is the impact of alternative strategies for dealing with attribute non-attendance? Can respondents reliably self-report non-attendance? Do respondents partially attend to attributes, and what are the implications of this? Our results show that allowing for the instance of ‘sometimes attending’ to attributes in making choices offers advantages over methods employed thus far in the literature.     

Preferences with and without prices - does the price attribute affect behavior in stated preference surveys?

The experimental as well as the non-market valuation literature include several examples of how an introduced price can affect behavior in otherwise not expected ways. It has become standard to include a price vector as an attribute in choice experiments, something that enables us to estimate a marginal willingness to pay for other attributes. We test the impact on preferences by an inclusion of a price in a choice experiment. Preferences are affected, as might be expected. However, also the relative ranking of individual attributes is affected.      

An Analytical Framework for Joint vs Separate Decisions by Couples in Choice Experiments: The Case of Coastal Water Quality in Tobago

This paper offers an analytical framework for analyzing joint and separate decisions by couples in the context of choice experiments for nonmarket valuation. It reports results from an attribute-based stated preference study in which members of couples are asked to conduct a choice-experiment first individually and then jointly. The choice context was the selection of which beach to visit while on vacation in Tobago. Available alternatives differed in attributes related to coastal water and beach quality such as level of coastal development and fish abundance. Tests of preference equality are reported and structured so as to identify the intra-couple decision-making patterns under taste heterogeneity with both finite and continuous mixed logit. Results from the latter suggest that women’s preferences are found to be predominant in the joint choice-experiment. Results suggest caution in using individual choice rather than joint couple choice when valuing quality changes impacting on couple activities, such as water and beach quality in Tobago, and call for further research on the topic.     

A Joint Latent-Class Model: Combining Likert-Scale Preference Statements With Choice Data to Harvest Preference Heterogeneity

In addition to choice questions (revealed and stated choices), preference surveys typically include other questions that provide information about preferences. Preference-statement data include questions on the importance of different attributes of a good or the extent of agreement with a particular statement. The intent of this paper is to model and jointly estimate preference heterogeneity using stated-preference choice data and preference-statement data. The starting point for this analysis is the belief that the individual has preferences, and both his/her choices and preference statements are manifestations of those preferences. Our modeling contribution is linking the choice data and preference-statement data in a latent-class framework. Estimation is straightforward using the E-M algorithm, even though our model has hundreds of preference parameters. Our estimates demonstrate that: (1) within a preference class, the importance anglers associate with different Green Bay site characteristics is in accordance with their responses to the preference statements; (2) estimated across-class utility parameters for fishing Green Bay are affected by the preference-statement data; (3) estimated across-class preference-statement response probabilities are affected by the inclusion of the choice data; and (4) both data sets influence the number of classes and the probability of belonging to a class as a function of the individual’s type.     

The internalization of externalities in the production of electricity: Willingness to pay for the attributes of a policy for renewable energy

This paper investigates the willingness to pay of a sample of residents of Bath, England, for a hypothetical program that promotes the production of renewable energy. Using choice experiments, we assess the preferences of respondents for a policy for the promotion of renewable energy that: (i) contributes to the internalization of the external costs caused by fossil fuel technologies; (ii) affects the short-term security of energy supply; (iii) has an impact on the employment in the energy sector; and (iv) leads to an increase in the electricity bill. Responses to the choice questions show that our respondents are in favour of a policy for renewable energy and that they attach a high value to a policy that brings private and public benefits in terms of climate change and energy security benefits. Our results therefore suggest that consumers are willing to pay a higher price for electricity in order to internalize the external costs in terms of energy security, climate change and air pollution caused by the production of electricity.     

Modelling preference heterogeneity for theatre tickets: a discrete choice modelling approach on Royal Danish Theatre booking data

This article analyses the behavioural choice for theatre tickets using a rich data set for 2010–2013 from the sale system of the Royal Danish National Theatre. A consumer who decides to attend a theatre production faces multiple sources of price variation that involves a choice by the consumer among different ticket alternatives. Three modelling approaches are proposed in order to model ticket purchases: conditional logit with socio-demographic characteristics, nested logit and latent class. These models allow us explicitly to take into account consumers’ preference heterogeneity with respect to the attributes associated with each ticket alternative (quality of the seat and day of the performance). In addition, the willingness to pay of choice attributes is estimated. Final results suggest that customers’ characteristics in terms of age and frequency of theatre attendance characterize different patterns of behaviour in the choice of theatre ticket.     

Distal order effects in stated preference surveys

Stated preference researchers have previously demonstrated that a good's placement among a sequence of goods in a set of valuation questions (i.e. proximal order effects) can have a substantial impact on people's valuations of these different goods. However, the economic consequences of potential order effects stemming from other questions in a survey, prior to the valuation tasks, have received surprisingly little attention. Using an online climate change survey, we identify order effects created by prior attitude-elicitation questions, and we assess the potential impact of these distal order effects on willingness to pay (WTP) estimates for stylized climate change policies. We find that the order used in prior questions may change people's opinions toward various attributes of the good to be valued, and thereby change WTP by a substantial amount. This paper emphasizes the significance of order effects stemming from preliminary survey questions, and supports a call for diligence in the random ordering of all potentially influential preliminary information in stated preference surveys to minimize inadvertent effects from any single arbitrary ordering.     

Does attribute order influence attribute-information processing in discrete choice experiments?

The existing empirical evidence shows that both contingent valuation and discrete choice experiment (DCE) methods are susceptible to various ordering effects. However, very few studies have analysed attribute-ordering effects in DCEs, and no study has investigated their potential influence on information-processing strategies, such as attribute non-attendance (ANA). This paper tests for attribute-ordering effects and examines whether the order of attributes describing the alternatives affects respondents’ propensity to attend to or ignore an attribute. A split-sample approach is used, where one sample received a DCE version in which the positions of the first and last non-monetary attributes are switched across the sequence of choice tasks compared with the other sample. The results show that attribute order does not affect welfare estimates in a significant way under the standard assumption of full attribute attendance, thus rejecting the notion of procedural bias. However, the welfare estimates for the attributes whose order was reversed and the share of respondents who ignored them differ significantly between the two attribute-ordering treatments once ANA behaviour is accounted for in the estimated choice models. These results highlight the important role of information-processing strategies in the design and evaluation of DCEs.     

Cost thresholds,cut-offs and sensitivities in stated choice analysis: Identification and implications

Within the discrete choice literature, there is growing recognition that some respondents do not process all attributes when evaluating their choice outcomes. Worryingly, the cost attribute is often among those attributes that are likely to be ignored by respondents. We use probabilistic decision process models (similar in form to latent class models, but where we define the classes to describe specific heuristics) to facilitate situations where respondents adopt cost thresholds and cut-offs. We further develop this model to address the potential confounding between preference heterogeneity and processing heterogeneity by simultaneously allowing for a segmentation of respondents based on their sensitivities to cost. Results, based on an empirical dataset on the existence value of rare fish species in Ireland, provide further confirmation that a share of respondents did not attended to cost. Importantly, however, when heterogeneity to cost levels is accounted for the inferred incidence of complete non-attendance is markedly lower, to the extent that when cost thresholds and cut-offs are also accommodated it almost disappears. This modelling approach leads to significant gains in model fit and has important implications for welfare analysis.     

Provided and perceived status quo in choice experiments: Implications for valuing the outputs of multifunctional rural areas

This article presents a choice experiment analyzing multifunctional rural development policies targeting conservation and recovery of environmental, social and cultural assets. In choice experiments a base alternative is usually included in order to estimate the welfare change associated to policy proposals. This study is concerned with the much neglected issue of the impact on policy analysis of the definition of a ‘status quo’ alternative either as an objective assessment by experts, or as a self-reported perception by respondents. Convergent validity analysis and prospective policy scenarios show a significant impact of different status quo specifications on individuals' preferences and related welfare measures when complex and unfamiliar biophysical systems are involved in policy analysis and evaluation.