Citation

Bond SB, Bond M. J. Nerv. Ment. Dis. 2004; 192(12): 857-863.

Affiliation

School of Social Work, McGill University, Montreal, Quebec, Canada.

Copyright

(Copyright © 2004, Lippincott Williams and Wilkins)

DOI

unavailable

PMID

15583508

Abstract

This study examined whether an individual's attachment style and/or a couple's combination of attachment styles predicted violence within the marriage and explored whether other variables moderated the risk of violence. Measures of attachment style were administered to 41 discordant couples who presented to four different clinics. The couples' presenting complaints were not violence, and those who did report violence on questioning did not manifest severe violence, i.e., requiring shelters or legal intervention. Self-report measures of violence and marital satisfaction, including problem-solving communication, were also given. Using analysis of covariance and logistic regression, the relative contributions to strength of predicting being a victim of conjugal violence were calculated. An anxious attachment style was a significant predictor of females being victims of violence and of men not being victims. A dismissive style in men was predictive of men being victims when entered into the model with problem solving communication. The combination of anxiously attached females and dismissive males was a potent predictor of violence, and longer duration of marriage and poor problem-solving communication added power to the prediction. Marital interaction, which is influenced by couples' attachment styles and problem-solving communication, is a significant factor in marital partners experiencing physical violence. For couples with milder levels of violence, a more nuanced approach (compared with the legally based approach used for severe violence) seems indicated.