The Impact Message Inventory–Circumplex (IMI–C): A Replication Study of Its Circumplex Structure in a Dutch Sample

Abstract

The 56-items Impact Message Inventory–Circumplex (IMI–C) has been developed to describe the interpersonal styles of target persons in terms of the 2-dimensional interpersonal circle, divided in 8 subscales (octants). The IMI–C is based on the assumption that a target person's interpersonal styles can validly be measured by assessing the internal (cognitive and emotional) reactions of the IMI–C respondent (e.g., therapists), interacting with the target person (e.g., psychotherapy patients). This replication study tested the circumplexity of a computer-delivered Dutch IMI–C. The computer-delivered Dutch IMI–C approached circumplexity in a large clinical sample (N = 812) of psychotherapy and psychiatric patients, after removing 3 items from the Submissive octant scale. These 3 items clearly disturbed the IMI–C circumplex structure. It is concluded that, even with the inevitable problem of nested data, circumplexity was approached sufficiently to justify the interpretation of the IMI–C at the octant level, provided that the 3 troublesome Submissive items are removed from the calculations of octant scores. Our replication test of the Dutch IMI–C circumplexity might be an impetus for researchers in the United States to invest in the further validation of the original American version.