Validation of the Moral Disengagement Scale among Some Iranian University Students

Abstract

Background: Moral disengagement is a variable in the social cognitive theory of morality and includes eight cognitive, psychosocial mechanisms by which moral self-sanctions are selectively disengaged from inhumane conduct. Objectives: The aim of the present study was to validate a university student replica of the moral disengagement scale among some Iranian university students. Methods: This validation study was based on the confirmatory factor analysis (CFA) method. The statistical population consisted of 346 undergraduate students at the Salman Farsi University of Kazerun. Also, 44 other undergraduate students were participated to examine the test-retest reliability of the scale. Both samples were selected by convenient sampling. The main sample completed the 32-item Moral Disengagement scale and the Marlowe-Crowne Social Desirability Scale. CFA (by AMOS 24), stability coefficients, Cronbach's alpha, and multiple analysis of variance (ANOVA) (all by SPSS 16) were used to study gender differences. Results: The indices of CFA for the 32-item scale were not satisfactory, then an item in all subscales with the lesser beta was dropped, and the scale included only 24-items. The indices of CFA of the 24-item scale were satisfactory. The internal consistency for the whole scale was desirable (α: .817) and for the subscales were adequate. Test-retest correlations were not desirable for the whole scale (r: .693) and for the subscales. The total score and the scores of some subscales were negatively correlated with social desirability. The total score and the scores of some subscales also were greater in males. Conclusions: The satisfactory indices of CFA and also the higher scores of males in the 24-item scale confirmed its construct validity. However, correlations between the scale and social desirability did not confirm the ideal divergent validity. Thus, assessing the social desirability beside the scale can clarify interpreting the scores. The obtained test-retest reliability suggests that this scale cannot assess a stable variable, and according beside the social cognitive theory, it is better to consider moral disengagement as a changeable and inconstant variable.