Discrete Choice Methods Research Articles

An econometric analysis of willingness-to-pay for sustainable development: A case study of the Volčji Potok landscape area

When the market for a certain good is competitive enough, economic activities can be studied by the market pricing mechanism. Because this is usually not feasible in case of environmental goods with embodied natural and cultural heritage, particular methods for economic valuation of such goods have to be applied. The present article represents the economic valuation of the Landscape Development and Protection Area of Volčji Potok, which is an important Slovenian cultural landscape area with internationally recognized qualities. For this purpose we combined classical contingent valuation with a closed-version of discrete choice method, where the protest responses have been removed. By using econometric analysis we obtained the value of willingness-to-pay and established its determinants. We also made an attempt to control for different biases that arise in such analyses. At last, we used the adjusted average individual value of willingness-to-pay to calculate the aggregate willingness-to-pay. The present analysis represents one of the very few applications of the method to Central and Eastern European countries.

Semi-parametric discrete choice measures of willingness to pay

A semi-parametric discrete choice method is proposed to recover welfare measures from individual choice data. The proposed method is compared with the traditional binary choice models in an application to measure benefits of recreation trips.

A hazard model of US airline passengers’ refund and exchange behavior

This study explores the use of discrete choice methods for airline passenger cancellation behavior. A discrete time proportional odds model with a prospective time scale is estimated based on the occurrence of cancellations (defined as refund and exchange events) in a sample of tickets provided by the Airline Reporting Corporation. Empirical results based on 2004 data from eight domestic US markets indicate that the intensity of the cancellation process is strongly influenced by both the time from ticket purchase and the time before flight departure. Higher cancellations rates are generally observed for recently purchased tickets, and for tickets whose associated flight departure dates are near. Cancellations rates are influenced by several other covariates, including departure day of week, market, and group size.

High Desert Wine

This paper uses data collected via a consumer intercept survey at the annual Harvest Wine Festival in Southern New Mexico to determine the impact wine attributes on consumer preferences. The hypothesis that price, reputation, and origin significantly affect the demand for New Mexico wines is evaluated using discrete choice methods. Results indicate that consumers' preferences and resulting wine choices are affected by wine attributes including wine price, winery reputation, and the origin of the wine.

Modelling Deforestation and Land‐Use Change: Sparse Data Environments

AbstractLand‐use change in developing countries is of great interest to policy‐makers and researchers with diverse interests. Concerns about consequences of deforestation for global climate change and biodiversity have received the most publicity, but loss of wetlands, declining land productivity and watershed management are also problems facing developing countries. Analyses of these problems are especially constrained by lack of data. This article reviews modelling approaches for data‐constrained environments that involve discrete choice methods including neural nets and dynamic programming, and research results that link individual household survey data with satellite images using geographic positioning systems.

Making sense of patient priorities: applying discrete choice methods in primary care using 'think aloud' technique

Delivering effective health care within limited budgets requires an understanding of patient priorities. Discrete choice experiments (DCEs) provide patients with choices, where each choice differs in terms of certain attributes (such as waiting times, quality of care). Although this technique has significant potential in examining priorities, its use raises practical and conceptual issues. This paper describes the development of a DCE evaluating patient priorities in primary care. Twenty patients completed a DCE using a 'think aloud' protocol, where they verbalized their thinking while making choices. The analysis examined their decision-making processes. There was evidence that patients reinterpreted some attributes, and related some to others outside the task. The cost attribute was interpreted in a variety of ways, dominating some patients' decision-making, being seen as irrelevant by others and being interpreted appropriately by some. The deree to which patients exhibited trading in line with theoretical assumptions also varied. Some choices in the hypothetical task were restricted by their previous experience, but more frequently patients tested the boundaries of the task in ways which directly reflected the primary care context. Patient interpretation of the discrete choice task was varied and some went beyond the formal boundaries of the task to make their choices. This highlights the importance of piloting attributes, providing clear instructions about the task and developing models of patient decision-making so that responses can be interpreted correctly.

Principe, intérêts et limites de la méthode des choix discrets pour la révélation des préférences en santé

Patients play an increasingly active role in the medical decision making process. Hence, methods for eliciting public preferences have become essential tools for enlightening the choices of health authorities. Discrete choice modelling is the most recently applied technique in the field of health. This article deals with the principle, the strengths and the weaknesses of this method. Often used in English-speaking countries, the discrete choice method has several advantages over other elicitation methods since it allows estimation of marginal utilities. Compliance rates are also higher and more information is provided. Considering these qualities, further research should be undertaken on discrete choice modelling in France.

Eliciting Stated Preferences for Health-Technology Adoption Criteria Using Paired Comparisons and Recommendation Judgments

Objectives The principal aim of this study was to illustrate applying discrete-choice methods for eliciting preferences for technology adoption criteria, including threshold values for cost-effectiveness ratios. A secondary objective was to compare the criteria weights of a sample of industry stakeholders with the results obtained from a study of National Institute for Health and Clinical Excellence (NICE) recommendations. Methods We administered a two-stage stated-preference (SP) question format to a sample of respondents who have expertise in applying economic analysis to technology adoption decisions. We elicited paired comparisons and recommendation judgments for similar criteria from a sample of International Society for Pharmacoeconomics and Outcomes Research members. Respondents evaluated nine pairs of hypothetical drugs, first indicating which drug was “better,” then indicating what they would recommend to a reimbursement authority such as the National Health Service. Stated-choice studies often obtain only paired-comparison judgments. Results Parameter estimates from the initial paired-comparison question indicate only incremental cost-effectiveness ratio and number of affected patients influence evaluations. These two factors have identical weights in the second recommendation question, but all four factors were considered in determining the recommendation threshold. Our sample was more willing than NICE to accept trade-offs between cost-effectiveness and other drug features and was less concerned about cost-effectiveness ratios and information uncertainty. Nevertheless, our sample was in agreement with NICE about the importance of the number of patients who would benefit as a criterion for influencing adoption recommendations. Conclusion This study demonstrates that including a question allowing respondents the opportunity to reject both alternatives in a forced-choice paired comparison can be important for obtaining accurate preference estimates. Our results demonstrate that SP methods offer a feasible means of quantifying a broad set of valuations that incorporate the preferences of patient, citizen, payer, and other stakeholder groups.

The influence of wine attributes on region of origin equity: An analysis of the moderating effect of consumer's perceived expertise

AbstractThis research addresses the interaction between the components of brand. More specifically, the authors evaluate how the region of origin as a component of a wine brand adds value to a wine purchaser. Previous research suggests that the region of origin equity is significantly moderated by the other wine attributes with which it is combined on the wine label: commercial brand, level of price, type of bottler, grape variety. The authors test whether the moderating effects depend on self‐perceived consumer expertise. They use a discrete choice method to survey 1,162 European wine purchasers, about equally selected from France, Austria, Germany, and the UK. Results show that the region of origin equity is indeed significantly moderated by the other wine attributes. They also show these moderating effects are more important for consumers who perceive themselves as “more expert” than for consumers who feel they are “novices.” [EconLit citation: M310]. © 2006 Wiley Periodicals, Inc. Agribusiness 22: 323–341, 2006.

Empirical implications of response acquiescence in discrete‐choice contingent valuation

The use of discrete-choice contingent valuation (CV) to elicit individuals' preference, expressed as maximum willingness-to-pay (WTP), although primarily developed in environmental economics, has been popular in the economic evaluation of health and healthcare. However, a concern with this method is the potential for 'over-estimating' WTP values due to the presence of response acquiescence, or 'yea-saying' bias. Based on a CV survey conducted to estimate physicians' valuation of clinic computerization, the extent of such bias was estimated from a within-sample open-ended valuation question following the respondents' discrete choice response. Analysis of this data suggests that not only was response acquiescence an issue, but also that the parametric estimation of mean and median WTP, the most common approach to estimating WTP from discrete-choice data, would potentially magnify such bias (to various degrees depending on the distributional assumptions applied). The possible extent of CV design versus analysis in discrete-choice methods therefore warrants further exploration.

Preferences over dental restorative materials among young patients and dental professionals

Knowledge of the patient's view is necessary, as patients' rights are an issue of increasing importance in dental practice. This study aimed to examine how dentists (n = 42), dental assistants (n = 65), and young dental patients (n = 306) in Norway and Denmark valued three attributes of dental restorations, namely, expected longevity, the appearance, and the risk of an adverse reaction. According to the discrete choice method, participants were presented with several scenarios, describing different levels of the attributes, for comparison of two restorations (A and B). The differences among patients, dentists, and dental assistants were examined by logistic regression analyses. The likelihood of a dentist favoring one restoration over another was very sensitive to the difference in the expected longevity of the restorations, and much less to differences in visibility. The patients had considerable sensitivity to differences in visibility and much less to differences in duration. Danish teenagers attach less importance to visibility of the restoration and greater importance to the risk of an adverse reaction than Norwegian teenagers. All differences were statistically significant. Preferences associated with properties of dental restorations differ between teenage patients and dental professionals and this should be addressed in clinical decision-making.

Applicability of discrete-choice methods to economic evaluations of complementary and alternative medicine

This review discusses the potential contributions of discrete-choice methods of stated preference assessment to future economic evaluations of complementary and alternative medicine. The manner in which some complementary and alternative medicine practitioners and researchers have framed the concept of therapeutic benefits including process-related benefits is first discussed. The mechanics of a discrete-choice exercise including assessment of health state utilities and willingness to pay are then described. Potential limitations of the discrete-choice method are discussed. This review concludes with a discussion of future research opportunities. The author suggests that discrete-choice methods may be integrated into economic evaluations through the benefit assessment process, particularly within the complementary and alternative medicine context.

Combining Top-Down and Bottom-Up Approaches To Energy-Economy Modeling Using Discrete Choice Methods

Recently, hybrid models of the energy-economy have been developed with the objective of combining the strengths of the traditional top-down and bottom-up approaches by simulating consumer and firm behavior at the technological level. We explore here the application of discrete choice research and modeling to the empirical estimation of key behavioral parameters representing technology choice in hybrid models. We estimate a discrete choice model of the industrial steam generation technology decision from a survey of 259 industrial firms in Canada. The results provide behavioral parameters for the CIMS energy-economy model. We then conduct a policy analysis and show the relative effects of an information program, technology subsidy, and carbon dioxide tax on the uptake of alternative industrial steam generation technologies, including boilers and cogeneration systems. We also show how empirically derived estimates of parameter uncertainty can be propagated through the model to provide uncertainty estimates for major model outputs.

Preferences of patients for emergency services available during usual GP surgery hours: a discrete choice experiment

A study was undertaken to investigate patients' strength of preferences for attributes or characteristics associated with delivery of emergency primary care services available during usual GP surgery hours and to investigate the trade-offs between attributes. A discrete choice experiment was used to quantify patients' strength of preferences for several key attributes of usual-hours emergency primary care. The attributes were chosen to reflect the findings of previous research, current policy initiatives and discussions with local key stakeholders. A self-complete questionnaire was administered to NHS Direct callers and adult attenders at Accident and Emergency, GP services and the NHS Walk-in Centre in the locality. Regression analysis was used to estimate the relative importance to patients of the different attributes. An overall response of 71% (n=432) was achieved. All but one of the attributes was a statistically significant predictor of preference. The attribute 'being kept informed about waiting time' was the most important. This was followed by 'quality of the consultation', 'having a consultation with a nurse', 'having a consultation with a doctor' and 'contacting the service in person'. Respondents were prepared to wait an extra 68 min to have a consultation with a doctor, but an extra 2 h 9 min for information about expected waiting time. There were no measurable preference differences between patients surveyed at different NHS entry points. Respondents younger than 45 years held strong preferences with respect to how they wanted to make contact with the system, whereas older respondents appeared not to hold strong preferences, seemingly indifferent between the alternatives. There was weak evidence which showed the younger group more strongly preferred accessing services via an integrated telephone system than making contact in person. This study showed that local solutions for reforming emergency primary care during hours when the GP surgery is open should take account of the strength of patient preferences. The discrete choice method was acceptable, and the results directly informed the development of a local service framework for such care.

A model for multiple brand choice

This paper considers a simultaneous choice model that describes consumer selection of multiple brands. The proposed model considerably extends research possibilities in the demand area by relaxing the restrictive assumption of mutually exclusive alternatives, which characterizes multinomial discrete choice methods. An essential contribution lies in the explicit treatment of stochastic dependencies among latent brand utilities. Two empirical applications illustrate how the model determines brand demand functions and yields quantitative insights into the competitive structure of the examined markets. An alternative specification of the multivariate, simultaneous probit model assuming independent covariance structure is also addressed.

Determining clinician satisfaction with telemedicine

We compared two methods of assessing clinician satisfaction with telemedicine. Clinicians completed a self-administered questionnaire with a preference ranking scale and a discrete choice method. Sixty-three clinicians completed the questionnaire (a 78% response rate). The preference ranking method showed that clinicians ranked the level of skill of the other clinician as the most important (55%) attribute of a teleconsultation and the completeness of the history as the second most important attribute (42%). The discrete choice method showed that clinicians ranked completeness of the patient's history as the most important attribute and risk of receiving wrong advice as the second most important. The use of discrete choice preferences provides an alternative and more objective method for collecting data about preferences in telemedicine. However, its use is not simple and requires the participants to be engaged in person rather than sent a questionnaire by post.

Politicians’ and hospital managers’ trade-offs in the choice of reimbursement scheme: a discrete choice experiment

Trade-offs in the choice of hospital reimbursement schemes are widely discussed in the health economics literature but no one has previously, to this authors’ knowledge, made an attempt at quantifying how purchasers and providers trade-off anticipated outcomes of reimbursement schemes. The purpose of this study is to elicit Danish county council politicians’ and hospital managers’ preferences for the anticipated incentives and consequences embedded in reimbursement schemes using the discrete choice method. Results indicate that politicians and hospital managers agree that increasing the number of patients treated is the most important objective for the choice of reimbursement scheme. The second most important objective to the politicians is to provide budget safety whereas inducing increased quality of treatment is third. Hospital managers rank inducement of increased quality of treatment as the second most important objective and have the county's budget safety as their third most important objective.

THE MIXED LOGIT MODEL: THE STATE OF PRACTICE

The mixed logit model is considered to be the most promising state of the art discrete choice model currently available. Increasingly researchers and practitioners are estimating mixed logit models of various degrees of sophistication with mixtures of revealed preference and stated choice data. It is timely to review progress in model estimation since the learning curve is steep and the unwary are likely to fall into a chasm if not careful. These chasms are very deep indeed given the complexity of the mixed logit model. Although the theory is relatively clear, estimation and data issues are far from clear. Indeed there is a great deal of potential mis-inference consequent on trying to extract increased behavioural realism from data that are often not able to comply with the demands of mixed logit models. Possibly for the first time we now have an estimation method that requires extremely high quality data if the analyst wishes to take advantage of the extended behavioural capabilities of such models. This paper focuses on the new opportunities offered by mixed logit models and some issues to be aware of to avoid misuse of such advanced discrete choice methods by the practitioner.

Asymmetric price effects in the telecommunications services markets

Asymmetric price effects have been observed and reported on in the product literature, but not in the service literature. This paper investigates the asymmetric effects for a number of telecommunications services. Discrete choice methods are used to collect the data. Asymmetries in the price effects are observed between a national and local brand in the case of local, interstate long distance, and cellular markets, while the price effects are reasonably symmetric in the case of paging and Internet services. We also find that the national brands have more symmetric effects between each other, and that the asymmetries are much more pronounced between the national and local brands. In contrast to findings from the product literature, higher quality brands do not dominate in attracting nonbuyers into the market.

Symptom-based outcome measures for asthma: the use of discrete choice methods to assess patient preferences

This paper reports on an application of discrete choice modelling to the measurement of patient preferences over asthma symptoms. A sample of patients with moderate to severe asthma was asked to choose between a series of pairs of scenarios characterised by different combinations of asthma symptoms. Their responses were analysed using a random effects ordered probit model. The results implied that patients weighted some symptoms more highly than others. Discrete choice modelling proved to be a useful approach for developing preference based outcome measures, although the results show how, in contexts where preferences over health care outcomes based on symptoms or some measure of health status are involved, a conventional linear additive model may not always be suitable.