Attribution in Relationships

The first investigation to study connections between attributions and close relationships was conducted by Bruce Orvis, Harold Kelley, and Deborah Butler (1976). They asked college-age couples to list examples of behavior, for oneself and one's partner, for which each had a different explanation. Several categories of behavior yielded divergent attributions (e.g., "Actor criticizes or places demands upon the partner"). More generally, for behavior resulting in negative outcomes, respondents exonerated themselves and blamed their partners. Later work suggests that this egocentric bias in attribution by close relationship partners holds mainly for couples experiencing distress. For those who are less distressed, they attribute bad outcomes to the situation and good outcomes to their partner or to their collaboration with their partner (see below).

An important implication of the results of Orvis and colleagues' investigation is that attributions made directly to one's partner, or indirectly in public and available to one's partner, may represent an attempt to influence the partner about why problematic events are occurring. For example, a spouse may say, "Our problems have been caused mainly by his inability to break the controlling influence his parents have over what he does." Whether or not the spouse believes that this control factor is critical, she may be making the attribution in an attempt to influence the partner to sever the control his parents have in his life. Helen Newman (1981) elaborated on attribution as a form of persuasion and ongoing communication in close relationships.

This early work by Orvis and colleagues confirmed the value of Jones and Nisbett's (1972) divergent perspectives hypothesis, with the important qualification that attributions often reflect self-esteem motivation when couples are making attributions about their relationships. Another amplification of this hypothesis was revealed in a study by John Harvey, Gary Wells, and Marlene Alvarez (1978). They showed that relationship partners who are distressed not only diverge in their attributions about relationship problems, but also cannot readily predict one another's attributions about the sources of the problems.