Validating a new measure of self-complexity.

The construct of self-complexity has been recognized as important for understanding psychological adjustment. However, research in this area has been impeded by measurement problems. In particular, the most commonly used measure, Linville's (1985, 1987) H statistic, cannot assess this construct appropriately. In this article, we present a new measure designed to tap two components of self-complexity separately: the number of self-aspects and the average distinction among them. We conducted a validation study with 131 Chinese college students. The results indicate that (a) this new measure is not confounded with conscious social desirability bias, (b) it has incremental validity beyond existing measures, and (c) the two components as assessed by this new measure are relatively independent and have different psychological functions. We discuss the implications of these findings for future studies of self-complexity and clinical practice.