Componential Analysis of Interpersonal Perception Data

We examine the advantages and disadvantages of 2 types of analyses used in interpersonal perception studies: componential and noncomponential. Componential analysis of interpersonal perception data (Kenny, 1994) partitions a judgment into components and then estimates the variances of and the correlations between these components. A noncomponential analysis uses raw scores to analyze interpersonal perception data. Three different research areas are investigated: consensus of perceptions across social contexts, reciprocity of attraction, and individual differences in self-enhancement. Finally, we consider criticisms of componential analysis. We conclude that interpersonal perception data necessarily have components (e.g., perceiver, target, measure, and their interactions), and that the researcher needs to develop a model that best captures the researcher's questions.