Sense, nonsense, and the use of personality measures

Criticize measurement-based personality research by recounting a well-known limitation: indices of relationship are often inflated because of the similarity of content across measures of presumably different constructs. In this paper we endorse the major contention of Nicholls et al. that measurement-based research may be conducted thoughtlessly, and we extend their thinking by pointing out an important underlying issue—the problem of defining what constitutes a “syndrome.” In contrast to them, however, we argue that the problems in measurement they cite do not entail the consequences they foresee and, in addition, have limited application.