Dealing with social desirability bias: an application to charitable giving

Purpose - This paper aims to address the issue of survey distortion caused by one of the most common and pervasive sources of bias, namely social desirability bias (SDB). Despite 50 years of research, there are still many unanswered questions about its conceptualisation and operationalisation. The authors argue that traditional measures of SDB are inadequate and that the context in which the research is being conducted should be reflected in the measures employed. Hence, the authors develop and validate a multi-dimensional scale that may be used to measure the degree of SDB present in responses to giving surveys. Design/methodology/approach - Following initial scale development procedures a convenience sample of 820 donors to a national charity was employed to refine the resultant scale items. Exploratory factor analysis and reliability tests were conducted to establish the dimensionality of the new scale and its reliability. Using a separate sample of 1,500 active donors, the scale was then subject to confirmatory procedures to test its predictive validity. Findings - The findings support the assertion that SDB is a multi-dimensional construct consisting of six dimensions. However, in the context of postal surveys it is found that self-deception and the degree of intrinsic benefit accruing to a donor are the primary determinants of the level of SDB an individual will exhibit. The authors also highlight the significance of the SDB issue since in the survey reported here, 65 per cent of respondents were found to over-report their giving. Originality/value - This is one of the first published studies that has been able to explore the predictive validity of a SDB scale. The work has expanded our understanding of the determinants of SDB and provided an instrument that may now be employed to reduce a significant proportion of this error in giving surveys.