Psychometric Properties of Turkish Compass of Shame Scale

The Compass of Shame Scale (CoSS) was developed to measure the ways individuals deal with shame. The scale consists of 48 items and four factors (withdrawal, attack other, attack-self, avoidance). This study aimed to investigate the psychometric properties of the Compass of Shame Scale (CoSS) in a Turkish sample. The sample consisted of 496 adults (137 male, 359 female) aged between 18 and 66 years. The data for test-retest analysis were collected from 76 participants two weeks after the first measurement. Participants were asked to complete the Compass of Shame Scale, Trait Shame and Guilt Scale, Trait Anger and Anger Expressions Inventory, Difficulties in Emotion Regulation Scale, Beck Depression Inventory, and a Demographic Form. In order to test the construct validity of CoSS an exploratory factor analysis (EFA) with varimax rotation was conducted and its correlations with theoretically similar constructs were tested. The results of the factor analysis were compatible with the fourfactor structure of the original scale. Regarding correlations, withdrawal, attack self, and attack other showed significant correlations with trait shame, anger, anger expressions, difficulties in emotion regulation, and depressive symptoms. Avoidance, however, had small but significant correlations with pride, anger, and internalized anger. These findings support the construct validity of the scale. The Cronbach’s alpha values for the internal consistency reliabilities of the factors were estimated and ranged between .71 and .89. The values calculated for testretest reliability ranged from .71 to .80. In addition, significant gender differences were found for the withdrawal and attack self factors such that women scored significantly higher than men in these factors. These findings provided strong evidence for the reliability and validity of the Turkish CoSS.