Factorial surveys reveal social desirability bias over self-reported organic fruit consumption

Abstract

Purpose Surveys measuring consumers’ preferences for sustainable food might suffer from socially desirable responding. Social desirability stems in part from social norms about sustainable lifestyles, when respondents need approval from others and when privacy is not guaranteed during survey completion. While various studies showed this phenomenon through laboratory experiments and by comparing different modes of survey administration, no research adopted factorial survey experiments (FSEs) to measure which factors are perceived by consumers as critical for socially desirable answering. The purpose of this paper is to fill this gap, at least for young consumers in a case study with organic fruit. Design/methodology/approach In total, 143 under-30 respondents were involved in an FSE. Each respondent evaluated six hypothetical scenarios (n=858) describing a consumer surveyed about his/her preferences for organic fruit. Respondents indicated whether they believed participants would have answered honestly or not to the survey described in each scenario. Generalized linear mixed models were used to model how scenario attributes were perceived to influence honest answering. Findings Respondents believe that people are more prone to bias their answers the more they seek approval from others. Moreover, the presence of acquaintances during survey completion is another critical driver of survey misreporting. Originality/value This study, by using a novel robust quasi-experimental approach, confirms that social desirability could lead consumers to misreport their preferences when surveyed about an organic fruit. This confirms that well-designed surveys, adopting proper remedies for social desirability should be adopted even for those food products, like fruit, which are usually deemed to be less subjected to misreporting. It also introduces FSEs as a flexible tool for collecting insights from consumers about potential antecedents of their behavior.