Pitfalls of personality theory

Abstract

This paper considers whether personality theory should develop breadth by exploring more and more narrow personality traits, or depth by deepening our understanding of known, higher-order traits. It demonstrates that narrow personality traits are frequently statistical artefacts. I argue that they are not based on individual differences in behaviour at all; as such they may not represent causal influences, or indeed any real characteristic of individuals. Despite their careful psychometric development, they will often be nothing more than social constructions. If one must work with narrow scales, I argue that it is imperative to examine the items within each scale to ensure that they differ in meaning. I also demonstrate that correlating such scales together can produce nonsensical results because popular scales which supposedly measure different traits sometimes contain items which are identical in meaning, and so will be answered similarly. I show that this is a problem with several widely-used scales, and note that studies which are based on such misleading analyses are sometimes widely cited. I argue that developing theories which explore the origins of higher-order personality traits and their relationship to other variables is likely to be more useful than focusing on narrow traits.