A method for the analysis of data about dyadic relationships.

Psychological research in many subfields focuses on dyads and dyadic relationships because they influence individual cognitions, emotions, and behavior. However, the measurement of relationship characteristics has been elusive. Although direct observational methods have been hailed as the gold standard, they have some disadvantages, such as their high cost and external validity. We propose a factor-analytic method, specifically the bifactor model, using self-report data from each individual of a dyad. This method has been applied to many contexts but not yet to dyadic data. We show that the information that can be gleaned from these data is relatively low-cost but rich in terms of the insight it yields about the nature of the dyadic relationship. We demonstrate two different applications of the bifactor model to data from married couples.•The bifactor model can be used to identify the individual-specific and dyadic aspects of data reported by the members of a dyad.•The estimates of the bifactor model of dyadic data can yield information about hierarchical aspects of a relationship, which are not directly assessed.