The correlational analysis of dyad-level data in the distinguishable case

Many theories of interpersonal relationships distinguish between individual -level processes and dyadic or group-level processes. This suggests that tw o-person relationships should be studied at the level of the dyad as well a s at the level of the individual. We discuss correlational methods for dyad s when each dyad contains two different types of individuals (e.g., a husba nd and wife, a mother and child, or an expert and a novice). In such dyadic interaction designs, the dyad members are said to be distinguishable. We p resent a method for computing the overall correlation for distinguishable d yads, and we discuss a model for separating the dyad-level and individual-l evel components of such a correlation. The computational techniques and the ir interpretation are described using data from 98 heterosexual couples.