Characteristics of adult attachment and implications for relationship dissolution

The present study assessed the relationship between attachment style and reasons for relationship dissolution. It was hypothesized that reasons for dissolution would form distinct factors representative of the attachment styles. Individuals classified into one of three attachment groups were expected to endorse the factor hypothesized to be theoretically related to their style significantly more than the remaining groups. It was also hypothesized that there would be intrarater consistency in dissolution reasons across individual raters' relationships. Ancillary hypotheses were proposed regarding convergent validity of the two attachment measures, the influence of parents, attachment history, social desirability and differences in length and number of relationships. Subjects were students from Hahnemann University in Philadelphia. Demographic data were reported. Subjects classified by attachment style were administered parental and attachment history questions followed by the Relationship Dissolution Index (RDI) and an abbreviated Marlowe-Crowne Social Desirability Scale. A factor analysis was performed on the RDI. Seven factors were extracted including "Avoidance" and "Anxious/Ambivalence" factors. No "Secure" factor emerged. ANOVAS were performed and subjects classified as avoidant on the two attachment measures endorsed the "Avoidance" factor significantly more than those classified as secure. Anxiously/ambivalently attached subjects did not endorse the "Anxious/Ambivalence" factor significantly more than the remaining groups. However, they endorsed the "Future Issues," and "Partner Pathology" factors significantly more than secure subjects and the "Compatibility Dissatisfaction" factor significantly more than the avoidant and secure groups. Only avoidant individuals showed some intrarater consistency on the RDI across relationships. As anticipated, substantial convergent validity was found between the two attachment measures and avoidant individuals had the shortest relationships. However, anxious/ambivalent subjects had the longest, contrary to expectation. No significant differences were found regarding number of relationships, degree of parental influence, length of parental marriage, degree of maternal/paternal acceptance or paternal mixed messages, relative to subjects' attachment style. Secure subjects perceived their parents' marriage as significantly higher in quality than anxious or avoidant individuals, partially supporting the hypotheses. Avoidant individuals reported significantly more rejection and mixed messages by mothers and rejection by fathers than secure subjects. These results were substantially consistent with the hypotheses. Finally, the effect of social desirability on responding was non-significant as expected. A discussion of the results and implications for future research was included.