Comparing Continuous and Dichotomous Scoring of Social Desirability Scales: Effects of Different Scoring Methods on the Reliability and Validity of the Winkler-Kroh-Spiess BIDR Short Scale

Abstract

Survey researchers often include measures of social desirability in questionnaires. The Balanced Inventory of Desirable Responding (BIDR; Paulhus, 1991) is a widely used instrument that measures two components of socially desirable responding: self-deceptive enhancement (SDE) and impression management (IM). An open question is whether these scales should be scored dichotomously (counting only extreme values) or continuously (taking the mean of the answers). This paper compares the two methods with respect to test-retest reliability (stability) and internal consistency using a short German version of the BIDR (Winkler, Kroh, & Spiess, 2006). Tests of criterion validity are also presented. Data are taken from a post-stratified national probability sample of German family doctors (n = 166). All retest reliabilities exceed .70. No significant differences in test-retest reliability are found for the SDE subscale and the combined scale; the IM subscale attains significantly higher test-retest reliability when scored continuously. Internal consistency is significantly higher for the continuously scored scales in one of two Waves. Tests of criterion validity yield expected results. Overall, these results suggest that the short German scale is a valid measure of socially desirable responding and support the case for continuous rather than dichotomous scoring of BIDR scales.