"Psychologists and laypeople are interested in the idea that people who are similar to each other will have relatively good relationships.
Unfortunately, many examinations of the "relationship quality and personality similarity" hypothesis suffer from two limitations that compromise previous findings:
1) defining similarity between partners in a way that confounds similarity and normativeness (or social desirability),
2) defining relationship quality in terms of only one measure reported by one partner.
The current study addresses both limitations.
It addresses the first by differentiating partners' distinctive similarity (the degree to which partners' share unique patterns of non-normative personality attributes) from personality normativeness (Furr, 2008).
It addresses the second limitation by having both partners describe their relationship and their relationally-based affective states along numerous dimensions (e.g., overall closeness, knowledge of each other’s goals and overall time spent alone together engaged in various activities, etc). Each member of sixty opposite-sex pairs of acquaintances provided self-reports of personality, relational attributes, and immediate affective states (both pre- and post- an interaction with their partner).
Results revealed that distinctive similarity (but not simple, unadjusted similarity) is associated with a host of positive relationship attributes and experiences, though these associations are more robust from females' perspectives than from males' perspectives. In addition, normativeness is associated with one's immediate affective state with a partner.
These results unconfound previous research by showing that it is having unique configurations of personality traits in common with one’s partner that best predicts relationship satisfaction."