| Abstract: | Proposes a method for analyzing data from dyadic social interactions across 3 dichotomous variables. By accounting for the likely interdependency between 2 scores from a dyad, problems incurred with traditional analyses are circumvented. The analysis is demonstrated by an examination of data from 4 groups of 41 parent–child dyads: mother–daughter, mother–son, father–daughter, and father–son. The effects of sex-of-S, sex-of-partner, familial relationship between S and partner, and the interaction of these factors were tested. Results are compared to results derived from a repeated-measures ANOVA. It is shown that the latter analysis may mask important features of the data. (1 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved) |
