ABSTRACT
This chapter discusses models for analyzing relationship research in dyads for cross-sectional data, for longitudinal data, and models for analyzing data from group and family research, and statistical power considerations. One of the biggest challenges when doing relationship research is that observations of two or more members are inherently not independent from one another. Nonindependence occurs when the scores of group members are statistically related. For the family context, nonindependence is handled by estimating the variance due to the families. The Actor—Partner Interdependence Model is perhaps the most widely used model in dyadic research, given the wide array of issues that can be addressed using this model. Another important model in dyadic research is the Common Fate Model. The dyadic growth curve model of Raudenbush et al. has been designed to analyze change at the level of the dyad members.
