Attribution and denial in socially desirable responding

Paulhus's (1984) Balanced Inventory of Desirable Responding (BIDR) contains scales designed to assess the two major components of socially desirable responding. The Self-Deception Scale (SDS) assesses the tendency to give favorably biased but honestly-held self-descriptions; the Impression Management Scale (IMS) assesses the tendency to give deliberately favorable self-descriptions. Research by Millham (1974) and Roth, Snyder and Pace (1986) has distinguished two tactics of desirable responding: (a) attribution: the claiming of positive attributes, and (b) denial: the rejection of negative attributes. This thesis presents three studies designed to evaluate the relative importance of these two distinctions in the BIDR. The first study, a factor analysis of 130 cases, demonstrated that both the content (self-deception vs. impression-management) and tactic (attribution vs. denial) were important in determining responses to the BIDR. The IMS items, including both attribution and denial, formed one factor. The attribution SDS items fell on a second factor. Surprisingly, the denial SDS items fell closer to the IMS factor. Rosenberg's Self-Esteem scale was most highly correlated with the attribution SDS items. Study 2 was a similar factor analysis of the data from a much larger dataset (N = 670). The factor pattern was identical to that in Study 1. Moreover, the SDS attribution items again predicted adjustment, including high self-esteem, low social anxiety and low empathic distress. Study 3 (N = 137) was designed to determine whether the critical difference between the attribution and denial items depends on: (a) whether the item refers to positive or negative attributes, or (b) whether the statement as a whole is favorable or unfavorable. To test these competing hypotheses, 20 negations were written, one for each of the 20 original assertions on the SDS. Results showed that items referring to positive characteristics (I am a saint; I am not a saint) formed a distinct factor from items referring to negative characteristics (I am a sinner; I am not a sinner). Simple negations (I am not a sinner) fell on the same factor as their corresponding assertions (I am a sinner) but at the opposite pole. Finally, the correlations with various personality measures were consistent with Studies 1 and 2. These results clarify the distinction between attribution and denial components. The distinction is not simply one of keying direction, that is, whether the statement as a whole is desirable or undesirable. Rather, the critical factor is whether the item content refers to a positive or negative characteristic. This distinction is critical in measuring self-deception, but not impression management.Arts, Faculty ofPsychology, Department ofGraduat