PSYCHOMETRIC PROPERTIES OF THE HUNGARIAN ADULT ATTACHMENT SCALE

Background and purpose - The revised Adult Attachment Scale (AAS) developed by N. L. Collins is a widely used questionnaire to measure adult attachment. However, its psychometric properties have not been investigated in Hungary. We aimed to confirm the key psychometric properties of the Hungarian version of the AAS focusing on reliability indices on a population that consisted of depressed and non-depressed young adults. Methods - The AAS is a self-report questionnaire, in which two different dimensional evaluating systems are possible: the original (close, depend, and anxiety) and the alternative scoring system (anxiety, avoidance). Our study population consisted of young adults with a history of major depression ( n = 264, median age = 25.7 years) and their never-depressed biological siblings (n = 244, median age = 24.0). Results - The internal consistency of close, anxiety, and avoidance scales were satisfactory (Cronbach-a >0.7). The consistency of the depend scale was slightly lower than expected (Cronbach-a = 0.62). Test-retest reliability was good for all of the scales, it ranged from 0.73 to 0.78 after 14 months of follow-up period. The scale showed good discrimination as tested by the differences of close and anxiety attachment dimensions between the groups (p<0.01). Moreover, we were able to differentiate the currently depressed subjects based on these attachment dimensions. Exploratory and confirmatory factor analyses were conducted, and a bifactor solution proved optimal model fit. Conclusion - The three dimensions of the AAS has not been confirmed. However, the close and anxiety scales of AAS were found to be adequate. Our results also indicate that attachment features correlate with major depressive episodes in adulthood.