Measures of Desirability Beliefs and Their Validity as Indicators for Socially Desirable Responding

Abstract

Social desirability (SD) bias is a serious threat for the quality of survey data, and SD beliefs predict the strength and direction of this bias. The validity of the most frequently used indicator for desirability beliefs, the one-point measure, depends on the neutrality and monotonic assumptions to be empirically valid. In the current study, the authors examine these completely untested assumptions by drawing on data about topics that are often the subject of survey research. These are prejudice against the elderly, charitable donations, voting in political elections, and environmental consciousness. The authors find that the neutrality assumption does not hold for substantial parts of the sample for all topics and the monotonic assumption is not valid for two of them. They introduce two new, more elaborated measures for desirability beliefs that allow for the valid prediction of SD bias when either one or both of the otherwise necessary assumptions do not hold.