MITIGATING HYPOTHETICAL BIAS IN STATED PREFERENCE DATA: EVIDENCE FROM SPORTS TOURISM

One of the major criticisms of stated preference data is hypothetical bias. Using a unique dataset of both stated and actual behavior, we test for hypothetical bias of stated preference survey responses. We consider whether respondents tend to overstate their participatory sporting event behavior ex ante when compared to their actual behavior at different registration fees. We find that stated behavior accurately predicts actual behavior at a middle level of respondent certainty, overpredicts actual behavior at a lower level of certainty, and underpredicts behavior at a higher level of certainty. This result suggests that respondent uncertainty corrections can be used to mitigate hypothetical bias and stated preference data can be used to better understand actual behavior in situations where no data exist. (JEL L83, Q51, Z2)     

RELIABILITY OF STATED PREFERENCES FOR CHOLERA AND TYPHOID VACCINES WITH TIME TO THINK IN HUE,VIETNAM

We examine the effect of giving respondents time to think about their stated choices (SC) in a survey of cholera and typhoid vaccine preferences in Hue, Vietnam. Because neither vaccine is widely available in Vietnam, we used the SC approach (a stated preference technique) and gave half of our respondents overnight to think about their choices to make the hypothetical valuation scenario as similar to a real‐life choice situation as possible. Respondents who were given extra time made fewer choices that violated internal validity tests of utility theory, and had lower average willingness to pay (WTP), confirming a result found in similar studies in the contingent valuation literature. (JEL D12, I18, C25)     

DUBIOUS AND DUBIOUSER: CONTINGENT VALUATION AND THE TIME OF DAY

We collect contingent valuation data from 524 student survey respondents over a 3‐day, 72‐hour period. Data analysis of a hypothetical campus referendum focuses on time‐of‐day effects on willingness to pay for a renewable energy project. We find that subjects responding to the survey during the night‐time hours (i.e., between 12 a.m. and 6 a.m.) do not display the law of demand, offering theoretically invalid responses to questions with important policy implications. Results from this research may have serious implications for the contingent valuation method (CVM). In short, just like your father said, nothing good happens after midnight when using the CVM. (JEL Q51)     

The effects of response rate changes on the index of consumer sentiment

From 1979 to 1996, the Survey of Consumer Attitudes response rate remained roughly 70 percent. But number of calls to complete an interview and proportion of interviews requiring refusal conversion doubled. Using call-record histories, we explore what the consequences of lower response rates would have been if these additional efforts had not been undertaken. Both number of calls and initially cooperating (vs. initially refusing) are related to the Index of Consumer Sentiment (ICS), but only number of calls survives a control for demographic characteristics. We assess the impact of excluding respondents who required refusal conversion (which reduces the response rate 5-10 percentage points), respondents who required more than five calls to complete the interview (reducing the response rate about 25 percentage points), and those who required more than two calls (a reduction of about 50 percentage points). We found no effect of excluding any of these respondent groups on cross-sectional estimates of the ICS using monthly samples of hundreds of cases. For yearly estimates, based on thousands of cases, the exclusion of respondents who required more calls (though not of initial refusers) had an effect, but a very small one. One of the exclusions generally affected estimates of change over time in the ICS, irrespective of sample size.     

ISSUE INVOLVEMENT AND RESPONSE EFFECTS IN PUBLIC OPINION SURVEYS

This research tests the widespread assumption that responseeffects due to variations in question form, wording, or contextwill be greatest among respondents who are least involved withan issue. A meta-analysis of results from 15 split-ballot experimentsconducted over a five-year period indicates that the responseeffects of using counterarguments or middle alternatives insurvey questions are significantly larger, as would be expected,among respondents who are less involved with a given issue thanamong those who are highly involved with it. But the effectsof question order and response order appear to be largely unrelatedto how involved a respondent is with a particular issue. Issueinvolvement, then, appears to specify some response effects,but not others.     

What respondents really expect from researchers

This article addresses the issue of falling response rates in telephone surveys. To better understand and maintain respondent goodwill, concepts of psychological contract and respondent expectations are introduced and explored. Results of the qualitative study show that respondent expectations are not only socially contingent but also ego-expressive, utilitarian, pleasurable, and epistemic by nature. Although results are reassuring in terms of commercialization of the psychological contract, they indicate some radical changes that are needed for the respondents to accept its continuation. The article discusses several practical and theoretical implications of such changes and suggests a series of corresponding propositions aimed at facilitating and inspiring future developments in this field.     

The Relationship Between Distance from Gambling Venues and Gambling Participation and Problem Gambling Among U.S. Adults

In this article we examine the relationship between extent of gambling for U.S. adults and the distance from their residence to the nearest casino or track. We employ data from a telephone survey of U.S. adults conducted in 2011–2013. The chances that the respondents gambled in the past year, were frequent gamblers, or were problem gamblers were greater if they lived close to a casino. The chances that the respondents gambled in the past year or were frequent gamblers were greater if they lived close to a horse or dog track. The effects of closeness to a casino on the likelihood of past-year gambling, frequent gambling, and problem gambling, as well as the effect of closeness to a track on past-year gambling, extended to about 30 miles from the respondent’s home. In addition, the concentration of casinos within 30 miles of the respondent’s home was positively related to the respondents’ chance of being a frequent or problem gambler. If a respondent had no casinos within 30 miles, he or she had a 2.7 % chance of being a problem gambler; if one casino, a 3.9 % chance; if six or more, a 6.2 % chance. The authors estimate that at least part of this effect is causal.     

Certainty and overconfidence in future preferences for food

We examine consumer certainty of future preferences and overconfidence in predicting future preferences. We explore how preference certainty and overconfidence impact the option value to revise today’s decisions in the future. We design a laboratory experiment that creates a controlled choice environment, in which a subject’s choice set (over food snacks) is known and constant over time, and the time frame is short – subjects make choices for themselves today, and for one to two weeks ahead. Our results suggest that even for such a seemingly straightforward choice task, only 45% of subjects can predict future choices accurately, while stated certainty of future preferences (one and two weeks ahead) is around 80%. We define overconfidence in predicting future preferences as: the difference between actual accuracy at predicting future choices and stated certainty of future preferences. Our results suggest strong evidence of overconfidence. We find that overconfidence increases with the level of stated certainty of future preferences. Finally, we observe that the option value people attach to future choice flexibility decreases with overconfidence. Overconfidence in future preferences affects economic welfare because it says people have too much incentive to lock themselves into future suboptimal decisions.     

Scale direction effects in agree/disagree and item-specific questions: a comparison of question formats

The effect of the response scale direction on response behavior is a well-known phenomenon in survey research. While there are several approaches to explaining how such response order effects occur, the literature reports mixed evidence. Furthermore, different question formats seem to vary in their susceptibility to these effects. We therefore investigate the occurrence of response order effects in Agree/Disagree (A/D) and Item-Specific (IS) questions. We conducted an experiment among n = 930 students in which we varied the scale direction (decremental vs. incremental) within A/D and IS questions and asked respondents to evaluate the questionnaires. The results reveal response order effects within the A/D but not within the IS question format. Furthermore, respondent’s evaluations suggest that completion of the IS questionnaires requires more consideration than the completion of the A/D questionnaires. Altogether, our findings indicate that IS questions are more robust against response order effects than A/D questions.     

Cross‐community comparability of attitude questions: An application of item response theory

The use of attitude questions is very common in comparative surveys as it allows researchers to gauge the perspectives of respondents toward social issues and explain cross‐country differentials in attitudes. Comparative studies implicitly assume that equivalently worded items are measuring the same construct in different settings. However, the results of these studies might be questionable if the measurement invariant assumption is violated and different groups of respondents do not have a shared understanding of the attitude items. This paper uses item response theory to compare the measurement of items and to test whether equivalently worded attitude questions about family dissolution are understood in the same way across various communities in India, Malaysia, Pakistan and the Philippines. The paper also examines the interaction between the respondent’s gender, the sex of the spouse leaving the marriage and the responses to the attitude questions about acceptability of family dissolution.     

Race-of-interviewer effects and survey questions about police violence

ABSTRACT

This research uses binary logistic regression to test for a connection between the race of interviewer and race of respondent on five questions in the General Social Survey about the use of physical force by the police. Results indicate two instances of race-of-interviewer effect: (1) black respondents were more likely to voice disapproval about whether the police can strike a citizen trying to escape when speaking to a black interviewer, and (2) white respondents were less likely to voice approval of police striking an adult male citizen in the presence of a black interviewer. Secondary findings indicate that education is consistently significant regardless of race of respondent and the survey question, while social class, sex, age, and region are significant in only limited scenarios.     

Preference for realistic art predicts support for Brexit

Following the UK’s EU referendum in June 2016, there has been considerable interest from scholars in understanding the characteristics that differentiate Leave supporters from Remain supporters. Since Leave supporters score higher on conscientiousness but lower on neuroticism and openness, and given their general proclivity toward conservatism, we hypothesized that preference for realistic art would predict support for Brexit. Data on a large nationally representative sample of the British population were obtained, and preference for realistic art was measured using a four‐item binary choice test. Controlling for a range of personal characteristics, we found that respondents who preferred all four realistic paintings were 15–20 percentage points more likely to support Leave than those who preferred zero or one realistic paintings. This effect was comparable to the difference in support between those with a degree and those with no education, and was robust to controlling for the respondent’s party identity.     

On the numerical forms of contingent valuation responses

Responses elicited by means of contingent valuation are often expressed in numerical forms termed “prominent numbers”, characterised by inexactness. This might arise because subjects find exactness in response unnecessary or valuation tasks intrinsically difficult. It has been suggested that the provision of numerical cues could influence the ease of completion of the valuation task, and thus its outcome. We describe an experiment undertaken to test three hypotheses. These are, first, that the perceived level of difficulty of the valuation task falls as the exactness of numerical cues provided to subjects increases. Second, it is hypothesised that subjects choose values characterised by higher degrees of exactness if they have been cued by numbers characterised by higher degrees of exactness. Third, the size of the chosen value is expected to be lower if the perceived task difficulty is higher. All three hypotheses are supported. Potential explanations for, and implications of, these findings are discussed.     

The Measurement of a Middle Position in Attitude Surveys

Five split-ballot experiments, plus replications, were carriedout in several national surveys to compare the effects of offeringor omitting a middle alternative in forced-choice attitude questions.Explicitly offering a middle position significantly increasesthe size of that category, but tends not to otherwise affectunivariate distributions. The relation of intensity to the middleposition is somewhat greater on Offered forms than on Omittedforms (less intense respondents being more affected by questionform than those who feel more strongly), but in general formdoes not alter the relationship between an item and a numberof other respondent characteristics. Finally, in one instancethere is evidence that form can change the conclusion aboutwhether two attitude items are related, but the results areof uncertain reliability.     

The older the bolder: Does relative age among peers influence children’s preference for competition?

Having been relatively old at school tends to have a long lasting positive effect on professional achievement, in particular in competitive environments (sports, politics). We investigate the roots of this pattern and study whether the relative age position of children at school plays a role in shaping their preference for competition. We run a controlled experiment in high schools across two states in Australia. Our participants are students who are the very oldest or very youngest in their classroom. We elicit their preference for competition using the Niederle and Vesterlund (2007) task and find evidence of an effect of relative age. This effect is concentrated on male students. We find no differences in self-confidence, and risk attitudes between relatively old and young students. These results suggest that the observed pattern may come from pure preference for competition.     

The meaning of “fundamental preferences”

We present the definition and meaning of “fundamental preferences” that are interpersonally comparable, ordinal and endemonistic. We also dispel a number of misunderstandings concerning them. In the article “A cause of preference is not on object of preference” (Soc Choice Welfare (1993) 10: 57–68), Professor Broome misinterprets the notion of “fundamental preferences” in confusing an observer's device for a psychological transformation of the observed (as if an economist studying wealth meant that he becomes wealthy, or if physicians had to be sick — this is well shown in his interpretation of a sentence of mine in p 65 where the crucial switch to the first person is his own). Considering a new set of variables that includes both structural parameters and former variables, hence variables of different kinds, assumes nothing new concerning the observed object; namely, it says neither that the consumption of bread becomes “a cause” of the taste for jam, nor that the individual likes (or dislikes) his own tastes, or anything like this (the accusation of “fantasy”). We shall suggest that certain other views receive a similar treatment in this paper. This misunderstanding is regrettable, since the consideration of fundamental preferences is unavoidable in social ethics, both when one has to compare all-encompassing individual situations, and for the preferences of the hypothetical identical individuals in an Original Position device where they evaluate at once what they might have and what they might be I wish to thank Professor Broome for comments on an earlier version of this note. . Therefore, perhaps the full argument must be stated again (see the works in references). To begin with, we should face the issue relevant for social ethics directly, rather than dealing with it in devious ways. The question arises if: (1) distributive justice is a question (he who says it is not wants to impose his own view of it), (2) individual happiness has any relevance for the quality of society (imagine a society of despaired people). Then, one can show that the relevant issue turns out to be: can one say that a person is happier than another? These persons are in specific situations.     

THE IMPORTANCE OF PERSISTENCE IN RESPONDENT SELECTION FOR PREELECTION SURVEYS

Different techniques for respondent selection can affect dataquality. These differences can result in variations in the distributionof partisans in preelection surveys, which in turn can havean effect on the distribution of candidate preference. Persistencein trying to interview designated respondents in telephone householdsincreased the number of Republicans in a 1984 sample, and thereforeReagan's margin over Mondale. Such differences in interviewingtechniques might account for some of the variations in nationalpreelection estimates of the outcome of the presidential election,and they suggest that caution be used in comparing marginalsfor party identification from different surveys employing differentrespondent selection techniques.     

Too Hard to Shout Over the Loudest Frame: Effects of Competing Frames in the Context of the Crystallized Media Coverage on Offshore Outsourcing

This study investigates the effects of competing frames in newspaper coverage of offshore outsourcing, an issue that is characterized by a predominantly negative, unemployment-focused media framing. The findings of a randomized, controlled experiment (N = 152) demonstrate that conventional framing effects do hold for this issue and for this media context by moving recipients’ attitudes in the direction consistent with the valence of the frame. However, they also show the backfire effect of the positively valenced frame among recipients with greater interest in political and economic news, who become less supportive of outsourcing if they read a story framing outsourcing from a consumer-oriented perspective. Our results contribute to the ongoing debate about the limits of framing effects on forming opinion about contentious policy issues and demonstrate the challenges for nondominant perspectives to make their way to news-savvy audiences even when the nature of the issue in question necessitates considering them.     

The influence of interviewer characteristics on support for democracy and political engagement in Sub-Saharan Africa*

This study investigates how an interviewer’s characteristics affect how respondents answer survey questions about democracy and political engagement. I analyze data from the 2008 Afrobarometer surveys, in which 810 interviewers surveyed 27,713 respondents across 20 countries in sub-Saharan Africa. Using these data, I study how interviewer education, age, and gender affect two outcomes: (1) response distributions to attitudinal and behavioral survey questions and (2) the likelihood of respondents saying ‘don’t know’ to a survey question. The analysis also investigates how the respondent’s perception of who sponsored the survey (NGO, private sector, government) affects attitudes. The results show that these interviewer characteristics affect the quality of survey data on political attitudes and behaviors. In the discussion, I consider the implications for research based on public opinion data about democracy and political engagement.     

Playing the Race Card in the Post-Willie Horton Era: The Impact of Racialized Code Words on Support for Punitive Crime Policy

To date, little is known about the precise impact of raciallycoded words and phrases. Instead, most of what we know aboutracialized messages comes from studies that focus on pictorialracial cues (for example, the infamous "Willie Horton" ad) oron messages with an extensive textual narrative that is lacedwith implicit racial cues. Because in a "post-Horton" era strategicuse of racially coded words will often be far more subtle thanthose explored in past studies, we investigate the power ofa single phrase believed by many to carry strong racial connotations:"inner city." We do so by embedding an experiment in a nationalsurvey of whites, where a random half of respondents was askedwhether they support spending money for prisons (versus antipovertyprograms) to lock up "violent criminals," while the other halfwas asked about "violent inner city criminals." Consistent withthe literature on issue framing, we find that whites’racial attitudes (for example, racial stereotypes) were muchmore important in shaping preferences for punitive policieswhen they receive the racially coded, "inner city" question.Our results demonstrate how easy it is to continue "playingthe race card" in the post–Willie Horton era, as wellas some of the limits of such framing effects among whites withmore positive racial attitudes.