Don’t Use Abbreviated Personality Measures

New research by Crede, Harms, Niehorster and Gaye-Valentine1 in the Journal of Personality and Social Psychology investigates the impact of using abbreviated personality measures.  Short answer: don’t do it.

In their study, the researchers surveyed 437 employed people (collected via StudyResponse) and 395 undergraduates.  Personality was assessed with common 1-item, 2-item, 4-item, 8-item, 6-item, and 8-item measures, along with a variety of outcomes, including job performance, GPA, stress, and health behaviors.

Almost universally, longer measures resulted in higher correlations with outcomes.  For example, Conscientiousness predicted Job Task Performance .6 with the 8-item measure (Saucier, in this case), with correlations as low as .2 with 1-item measures.

This is most critical for personality researchers who often rely on incremental variance found over the Big Five to provide evidence of a new, unique personality trait.  For example, they use a single-item measure and their new measure in a survey study, find new variance in job performance or life satisfaction using their new measure over the Big Five, and declare that they have found a new, useful personality trait.  This study suggests that evidence for many such traits may be fallacious; instead, the incremental variance found is better explained by the lack of a reasonable multi-item personality trait for comparison.

The authors also find that the use of abbreviated personality measures increases both Type I and Type II statistical conclusion errors.  Single-item measures are especially risky. Although the shortened length of the scales is attractive from a practical perspective, this is completely empirically unjustified – the loss of validity is simply not worth it.

All of this together suggests a very simple decision rule: don’t use abbreviated personality measures.

1. Credé, M., Harms, P., Niehorster, S., & Gaye-Valentine, A. (2012). An evaluation of the consequences of using short measures of the Big Five personality traits. Journal of Personality and Social Psychology, 102 (4), 874-888 DOI: 10.1037/a0027403 [↩]