| Summary: | Giving behaviour is often studied through self-reports and questionnaires. Despite their obvious advantages, questionnaires are vulnerable to socially desirable responding leading to over-reporting of socially desirable items and vice versa. Yet, this phenomenon has received little to no attention from nonprofit researchers. To address this shortfall, the thesis first explores donors' motivation to give socially desirable responses by developing a conceptual model based on a multi-disciplinary review of social desirability bias and giving behaviour literature. A much better understanding of socially desirable responding via this' conceptual model will help to operationalise social desirability bias concept in the nonprofit context. This study has also identified several social desirability scales that have been used in the past to measure the same response bias. However, further review shows that these scales should be improved psychometrically if they are to be applied in a specific context, as addressed in this study. The author conducted a series of three research stages. The conceptual model was explored and tested using both qualitative and quantitative research approaches, mainly in-depth interviews with seven experts as well as postal surveys with 3000 donors from two established animal welfare charities in the UK. This resulted in an improved social desirability scale comprising of 26 scale items. In this study, the effects of socially desirable responding on self-reports of giving behaviour were modelled by matching the self-reported against the actual amount of donations recorded in the charity's database. Structural equation models were presented, linking the proposed eight antecedents of social desirability bias to the discrepancies in the amount of donations. The findings revealed social desirability bias to be a complex multidimensional phenomenon and showed that 'positive self deception' to be a key motive for donors to give socially desirable responses in the realm of postal questionnaires. The study concluded that although the effects of socially desirable responding are relatively small in the realm of postal questionnaire, it remains a mystery whether donors' memory failure may play a bigger role in the discrepancies in the amount of donations reported. Implications of these findings on the development of social desirability bias construct are also explored.
|
|---|
