Ole Bonnichsen

Using an Ex-ante Entreaty to Reduce Protest Zero Bias in Stated Preference Surveys - A Health Economic Case

When prompting respondents to act on the hypothetical markets that are presented to them in stated preference surveys, it is often observed that a proportion of the sample state a zero demand for the good in question even though their genuine demand is positive. Though the literature comes up with some means to calibrate the estimated preferences ex-post, this is far from being unproblematic and uncontroversial. In this paper we test whether a short ex-ante entreaty, presented to respondents in a health economic stated preference survey, can effectively reduce the number of protest zero bids. We first of all find that the protest reduction entreaty significantly reduces the number of protest zero bids in the survey. Furthermore, the results indicate that marginal demand for the individual attributes as well as general preferences do not seem to be influenced by the protest reduction entreaty. Compared to the traditional ex-post calibration of protest zero bids in stated preference surveys, these results suggest that an ex-ante reduction in protest zero bids by using a simple and short entreaty is possible without introducing other types of biases in the survey.

Elicitation of Ostomy Pouch Preferences: A Discrete-Choice Experiment

AbstractBackground: Previous studies about patients who have undergone ostomy surgery commonly address the issues of the surgery, complications, preoperative counseling, quality of life, and psychosocial changes following surgery. Only a limited number of studies deal with how technical improvements in stoma care would affect patients and, to the author’s knowledge, the present study is the first to elicit preferences for potential improvements in ostomy pouches in the form of monetary values. Objective: This article examines and measures Swedish patients’ preferences for potential improvements in ostomy pouch attributes. The theory, study design, elicitation procedure, and resulting preference structure of the sample is described. Methods: A discrete-choice experiment (DCE) was used to elicit preferences. Respondents were asked to choose between alternatives in choice sets, in which each alternative comprised a number of attributes relating to the adhesive, filter, and flexibility of ostomy pouches. The choices between alternatives made by the respondent imply a trade-off between the attributes and allow for the estimation of individuals’ willingness to pay (WTP) for the attributes of ostomy pouches when cost is included as an attribute. A total of 254 patients responded to the survey and preferences were estimated using a random parameter logit econometric specification. Results: Respondents had significantly positive WTP for all potential attribute improvements presented in the survey, with the strongest preference for reducing the number of leakages; flexibility of the system weighted second, and filter lifetime was the least important. Conclusion: The results confirm that the potential attribute improvements presented to the respondents in this study significantly and positively affect the utility that patients would obtain from a potential improvement in their ostomy pouch. This provides information as to how treatment options in terms of stoma management can be structured so as to maximize the benefits for patients.

Correcting for non-response bias in contingent valuation surveys concerning environmental non-market goods: an empirical investigation using an online panel

Data collection for economic valuation by using Internet surveys and pre-recruited Internet panels can be associated with severe disadvantages. Problems concerning sample coverage and sample representativeness can be expected. Representation errors may occur since people can choose whether to be part of an Internet panel and subsequently whether they wish to participate in the survey, thereby introducing two elements of potential self-selection. These elements may be correlated with preferences, thus making the respondents a non-random and non-representative sample, ultimately biasing results. This paper analyses a sample used for an Internet contingent valuation method survey eliciting preferences for improvements in water quality of a river. We find that some variables that affect the survey participation decision also affect willingness-to-pay, consequently biasing our welfare estimates. We show how adjusting willingness-to-pay for this bias can be accomplished by using a grouped data model incorporating a correlation parameter to account for selection.

The influence of time of day on decision fatigue in online food choice experiments

Purpose Fatigue effects related to answering a sequence of choice tasks have received much scrutiny in the stated choice experiments (SCE) literature. However, decision fatigue related to the time of day when respondents answer questionnaires has been largely overlooked in this literature even though time of day related fatigue effects are well known in the psychology literature. The purpose of this paper is to hypothesize that variations in the time of day when respondents answer an online food choice experiment will translate into observable fatigue effects in the food choices. Design/methodology/approach An empirical SCE concerning food choices is conducted using a web-based questionnaire for interviews in a pre-recruited online panel of consumers. Timestamps collected during the online interviews provide knowledge about the time of day at which each respondent has answered the survey. This information is linked with knowledge from a food sociology survey on typical meal times as well as biophysical research linking food intake to blood sugar and mental energy in order to generate a proxy variable for each respondent’s level of mental energy when answering the food choice tasks in the questionnaire. Findings Results show evidence of a time of day effect on error variance in the stated food choices as well as the subsequently estimated market share predictions. Specifically, respondents provide less consistent answers during the afternoon than at other times of the day. Originality/value The results indicate that time of day can affect responses to an online survey through increased fatigue and correspondingly less choice consistency. Thus, especially online surveys might account for this in data analysis or even restrict accessibility to the online survey for certain times of day.