ABSTRACT Personality tests are susceptible to faking, leading to increased use in forced choice (FC) measures. However, a major challenge is that FC tests are often ipsative, meaning scores are dependent upon one another, therefore distorting inter‐individual comparisons. Despite this, ipsativity is poorly understood as it occurs in real‐world settings, including its degree, causes, and impacts. Drawing on 49 independent samples with matched FC and single‐stimulus (SS) measures, we operationally defined and assessed ipsativity meta‐analytically. Comparing normative SS measures to FC scores, we found that FC intercorrelations are meaningfully lower, though ipsativity varies considerably, with most measures being partially ipsative. Item desirability balance within blocks emerged as a critical factor influencing ipsativity, with balanced blocks (i.e., equal statement desirability) more likely to produce ipsativity. Furthermore, summative scoring exhibited stronger ipsative effects than item‐response‐theory (IRT) scoring, yet IRT scoring did not fully alleviate ipsativity unless some item blocks were unbalanced in desirability. Importantly, ipsativity was tied to diminished construct validity but also more resistance to mean shifts in scores due to faking. These results demonstrate that more normative FC measures exhibit more favorable validity but may be susceptible to faking. Thus, thoughtful FC design is required to balance FC goals.
Speer et al. (Sun,) studied this question.