Polychotomous Choice Research Articles

Full-Information Selection Bias Correction for Discrete Choice Models with Observation-Conditional Regressors

We examine self-selection in polychotomous choice models that construct attribute values for each alternative conditioned on observed choices. Using observations made only when the alternative was chosen ignores private information which was a basis for the decision, biasing resulting estimates. We suggest a full-information maximum likelihood procedure that performs well at the extremes of the choice set in our sample, and use an “identification at infinity” weighting to identify levels. We apply the model to understanding fishing location choice in the economically significant Bering Sea pollock fishery, where expected catches at each location are constructed from harvests observed when that location is chosen.

Are All U.S. Credit Unions Alike? A Generalized Model of Heterogeneous Technologies with Endogenous Switching and Fixed Effects

This paper formally models heterogeneity in credit unions’ production technologies as evidenced by (endogenously selected) differing output mixes. We show that failure to account for this observed heterogeneity is likely to lead to biased, inconsistent estimates of credit unions’ technology and potentially misleading results about the industry structure. The estimates are also likely to be biased if one overlooks unobserved credit union specific effects, as customarily done in the literature. To address these concerns, we develop a generalized model of endogenous switching with polychotomous choice that is able to accommodate fixed effects in both the technology selection and the outcome equations. We use this model to estimate returns to scale for the U.S. retail credit unions from 1994 to 2011. Unlike recent studies, we find that not all credit unions enjoy increasing returns to scale. A non/negligible number of large institutions operate at decreasing returns to scale, indicating that they should either cut back in size or switch to a different technology by adjusting the output mix.

Preference uncertainty in stated preference studies: facts and artefacts

The ordinal scale and polychotomous choice methods are two widely used techniques for estimating preference uncertainty in stated preference studies. This article presents the results of two experiments that apply these estimation techniques. The first experiment was designed to compare and contrast the scores of the ordinal scale and polychotomous choice method. The second experiment was conducted to test a scale that combines verbal expressions with numerical and graphical interpretations: a composite scale. The results of the study can be summarized in three key findings. First, the polychotomous choice method generates a higher proportion of ‘yes’ responses than the conventional dichotomous choice elicitation format. Second, the composite scale generates a significantly higher proportion of certain responses. Finally, the ordinal scale performs poorly on the ground of construct validity.

Respondent uncertainty in contingent valuation of preventing beach erosion: An analysis with a polychotomous choice question

Respondent uncertainty is often considered as one of the main limitations of stated preference methods, which are nowadays being widely used for valuing environmental goods and services. This article examines the effect of respondent uncertainty on welfare estimates by applying the contingent valuation method. This is done in the context of beach protection against erosion. Respondent certainty levels are elicited using a five-category polychotomous choice question. Two different uncertainty calibration techniques are tested, namely one that treats uncertain responses as missing and another in which uncertain ‘yes’ responses are recoded as ‘no’ responses. We found no evidence that the former technique offers any gains over the conventional model assuming certainty. The latter calibration technique systematically reduces welfare estimates. The reduction is statistically significant only when the most certain ‘yes’ responses are recoded as ‘no’ responses. The article further identifies determinants of respondent uncertainty. Finally, it explores how real market experience affects respondent uncertainty.

Nonparametric Identification and Estimation in a Roy Model With Common Nonpecuniary Returns

We consider identification and estimation of a Roy model that includes a common nonpecuniary utility component associated with each choice alternative. This augmented Roy model has broader applications to many polychotomous choice problems in addition to occupational sorting. We develop a pair of nonparametric estimators for this model, derive asymptotics, and illustrate small-sample properties with a series of Monte Carlo experiments. We apply one of these models to migration behavior and analyze the effect of Roy sorting on observed returns to college education. Correcting for Roy sorting bias, the returns to a college degree are cut in half. This article has supplementary material online.

Seafood Substitution and Mislabeling: WTP for a Locally Caught Grouper Labeling Program in Florida

Abstract Recent media reports have documented the fraudulent mislabeling of grouper at several Florida restaurants whereby customers were sold lower-valued species instead of grouper. This article uses data from 400 seafood eating Floridians to determine awareness of reported substitution, effects on consumption, and willingness to pay (WTP) for a product integrity labeling program. Using a novel frequency-based polychotomous choice (PC) contingent valuation format, designed for products that are purchased often, WTP is estimated based on level of commitment to the product integrity labeling program. Sixty-two percent of respondents were aware of the reported substitution, 57% of those changed their seafood consumption, and respondents were willing to pay an average premium of $0.83 to $3.18 for grouper entrees at restaurants with a product integrity label. The factors that helped to explain the change in consumption and WTP were identified and can be used to help improve current labeling and marketing ef...

An investigation of preference uncertainty in the contingent valuation study

This study investigates the respondents’ preference uncertainty in the dichotomous choice (DC) contingent valuation study for gathering information on willingness to pay (WTP). To this end, we use polychotomous choice (PC) question and DC question with certainty level for two split samples. We find that incorporating preference uncertainty has a significant effect on the WTP estimates and, more importantly, the DC question with certainty level produces more efficient WTP estimate than the PC question.

Mobility and the Return to Education: Testing a Roy Model with Multiple Markets

Self–selected migration presents one potential explanation for why observed returns to a college education in local labor markets vary widely even though U.S. workers are highly mobile. To assess the impact of self–selection on estimated returns, this paper first develops a Roy model of mobility and earnings where workers choose in which of the 50 states (plus the District of Columbia) to live and work. Available estimation methods are either infeasible for a selection model with so many alternatives or place potentially severe restrictions on earnings and the selection process. This paper develops an alternative econometric methodology that combines Lee's (1983) parametric maximum order statistic approach to reduce the dimensionality of the error terms with more recent work on semiparametric estimation of selection models (e.g., Ahn and Powell (1993)). The resulting semiparametric correction is easy to implement and can be adapted to a variety of other polychotomous choice problems. The empirical work, which uses 1990 U.S. Census data, confirms the role of comparative advantage in mobility decisions. The results suggest that self–selection of higher educated individuals to states with higher returns to education generally leads to upward biases in OLS estimates of the returns to education in state–specific labor markets. While the estimated returns to a college education are significantly biased, correcting for the bias does not narrow the range of returns across states. Consistent with the finding that the corrected return to a college education differs across the U.S., the relative state–to–state migration flows of college– versus high school–educated individuals respond strongly to differences in the return to education and amenities across states.

Optimal decision rules for fixed-size committees in polychotomous choice situations

This paper derives optimal decision rules for fixed-size committees in polychotomous choice situations. Earlier studies focus on the dichotomous choice model and thus the present extension broadens the scope of applications of the theory of collective decision making.

Semiparametric identification and heterogeneity in discrete choice dynamic programming models

Empirical discrete choice dynamic programming models have become important empirical tools. A question that arises in estimation and interpretation of the results from these specifications is which combination of data and assumptions are needed to overcome problems of heterogeneity, selection, and omited variables bias. This paper addresses this question by considering nonparametric identification of a version of the model that allows for quite general forms of unobservable and information structures. I show that the model can be identified under conditions similar to a static polychotomous choice model. Using a stochastic version of an ‘identification of infinity’ argument, utility can be identified up to a monotonic transformation of the observables under strong support conditions and two types of exclusion restriction. The first type is similar to a standard static exclusion restriction: a variable that influences the first period decision, but does not enter the second period decision directly. The second type requires a variable that does not affect the utility of the first option directly, but is known during the first period, and has predictive power on the choice during the second. I also provide two specifications under which the full error structure can be identified. This requires the additional assumption of stochastic innovations in the observables. I then use the model to estimate schooling decisions in which students deciding whether to drop out of high school account for the option value of attending college.

Yea-Saying in Contingent Valuation Surveys

A number of recent papers have provided estimates of willingness to pay derived in contingent valuation surveys exceeding those revealed in experimental or real-life markets. One possible explanation for the overestimation of values is the presence of yea-saying. With the objective of reducing the occurrence of yea-saying, a new elicitation format, referred to as the dissonance-minimizing (D M) format, is proposed. An empirical comparison of the DM format with the conventional dichotomous-choice (DC) format and the ambivalence-reducing polychotomous choice (PC) format suggested by Ready, Whitehead, and Blomquist (1995) is made. Results are encouraging for the DM.

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.

Biodiversity values for alternative management regimes in remote UK coniferous forests: an iterative bidding polychotomous choice approach

The overwhelming majority of commercial forestry in the UK comprises blanket non-native coniferous species with low biodiversity levels. The UK government is committed under the Rio Principles and Helsinki Guidelines to conserve and enhance biodiversity in British forests and woodlands. This paper assesses the passive use value of biodiversity in remote non-visited coniferous forests, and the costs of diversification and restructuring. Focus groups were employed to aid the design and framing of the study. An iterative bidding polychotomous choice format was used to elicit willingness-to-pay values for different biodiversity standards and management options. Whilst, passive use benefits exceeded costs for biodiversity conservation, considerable ambivalence in choice occurred between different levels of biodiversity. The policy relevance of the results are discussed, and compared to the practical implementation measures for biodiversity being adopted by the Forestry Commission.

Alcohol and Earnings: Does Drinking Yield a Wage Premium?

This paper examines the relationship between alcohol consumption and earnings for prime-age males. Wage differentials for non-, moderate, and heavy drinkers are estimated using a polychotomous choice model, which accounts for the endogenous relationship be- tween drinking and earnings. We find that moderate alcohol consumption leads to increased earnings relative to abstention. However, heavy drinking leads to reduced earnings relative to moderate drinking. Heavy drinkers possess flatter age-earnings profiles and attain lower returns for higher education than non- and moderate drinkers. These results are in contrast to previous research on substance abuse, which finds no earnings drop-off for heavy users.

Majority vote of even and odd experts in a polychotomous choice situation

The performance of majority vote by an odd number of voters in a dichotomous situation is a much-studied subject. In this paper, we study the performance of group consensus in a situation of polychotomous choice. The differences in majority vote behaviour between even and odd numbers of expert panellists are examined, and the effects of adding new members are derived. Unlike the dichotomous model, optimality in the present context may not be uniquely defined, so the probabilities of the consensus being “correct” or “erroneous” are both considered as the number of experts is increased.

Contingent Valuation When Respondents Are Ambivalent

Respondents to contingent valuation surveys may have difficulty resolving ambivalence over trade-offs between money and changes in levels of environmental amenities. In two separate studies, respondents were given the opportunity to express intensity of preferences through polychotomous choice valuation questions. This elicitation method produced slightly higher rates of usable responses than dichotomous choice, but with wide ambivalence regions. Dichotomous choice responses appear to reflect a conservatism strategy, in that ambivalent respondents tend to reject any move away from the baseline.

The Computation of Opportunity Costs in Polychotomous Choice Models with Selectivity

This article provides general formulas for the computation of opportunity costs (or foregone earnings) of chosen alternatives in sample selection models with polychotomous choices. With observed choice probabilities and outcomes of alternatives chosen by some individuals, the opportunity costs for individuals who do not choose those alternatives can be evaluated. For the probability indexed sample selection models with logit choice probabilities, which include the polychotomous choice model of Lung-fei Lee (1983), the formulas are simple and reveal restrictions implicitly imposed on those models. These formulas can be useful for empirical studies. Copyright 1995 by MIT Press.

Semiparametric maximum likelihood estimation of polychotomous and sequential choice models

This article considers semiparametric estimation of discrete choice models. The estimation method is a semiparametric maximum likelihood method which generalizes Klein and Spady (1993) to the estimation of polychotomous choice and sequential choice models. Special emphasis is on the estimation of models with indices of which the density function is not bounded away from zero, issues on trimming indices, the correction of asymptotic bias, and issues of negative kernel density estimation. A semiparametric efficiency bound is derived for the polychotomous choice model with index restrictions. Our estimator is shown to be √ n-consistent, asymptotically normal, and almost asymptotically efficient. Monte Carlo experiments are provided to evaluate the finite-sample performance of the estimator.

Nonparametric identification and estimation of polychotomous choice models

In this paper we provide conditions guaranteeing the identification of nonparametric polychotomous choice models. In these models, neither the subutility function of observable attributes nor the distribution of the unobservable random terms is specified parametrically. Sets of nonparametric functions that possess properties that are often implied by economic theory and satisfy the restrictions required to identify the models are described. We use the identification results to develop nonparametric strongly-consistent estimators for the subutility function of observable attributes. The results concern models in which the distribution of the unobservable random terms both depend and do not depend on the observable characteristics.

Imputing a continuous income variable from a bracketed income variable with special attention to missing observations

Abstract This paper proposes a method, based on polychotomous discrete choice methods, to impute a continuous measure of income when only a bracketed measure of income is available and for only a subset of the obsevations. The method is shown to perform well with CP5 data.