Participants in this study were drawn from a project investigating the effect of living in a non-conventional family on a child's development. The family systems considered non-conventional included single mothers, unmarried cohabitees, communal groups and a control group of conventional two-parent families was also examined. Infants had originally been examined in the Strange Situation and no differences were found in the distribution of attachment classifications amongst the family lifestyle groups.

Participants were young adults (age range 17-19) who, as infants, had been classified as avoidant (N=12), resistant (N=6) and secure (N=12). This subsample represented all family types originally examined but it was noted that the participants who were contacted and available from the original conventional group, tended to have had insecure infant attachment styles.

Participants were examined using the Adult Attachment Interview and various means were used to assess negative life events to date.

Results were as follows:

• Attachment classification stability was 63% (p < 0.01).
• Type of family system was not significantly related to adult attachment style, negative life events, or attachment stability.
• Negative life events were associated with stability of attachment classification, mainly in terms of maintaining an insecure attachment style.

Overall, Hamilton concludes that the study has supported the idea that insecure infant attachment is related to participants being on a life course that involves a higher incidence of negative life events. It was also noted that acrimonious divorces, especially where children were young, appeared to be related to insecurity as an adult.

This is one of five articles on longitudinal attachment stability to early adulthood published together in the June 2000 issue of Child Development. For an overview of all five articles, refer to the general introduction (2000).