Faking Effects on the Factor Structure of a Quasi-Ipsative Forced-Choice Personality Inventory

Abstract

Research has shown that faking behavior affects the factor structure of single-stimulus (SS) personality measures. However, no published research has analyzed the effects of this phenomenon on the factor structure of forced-choice (FC) personality inventories. This study examines the effects of faking, induced in a laboratory setting, on the construct validity of a quasi-ipsative FC personality inventory based on the Five-Factor Model. It also examines the moderator effect of the type of experimental design (between-subject and within-subject design) on factor analyses. The results showed that (a) data fit to a structure of five-factors in the two conditions (honest and faking) in both experimental designs; (b) model fit indices are also good or excellent in all cases; and (c) Burt-Tucker's congruence coefficients between convergent factors of conditions analyzed are very high. These findings provide evidence that the quasi-ipsative FC format is a robust instrument that controls the effects of faking on factor structure. Finally, we discuss theoretical and practical implications of these findings for personnel selection and assessment.

A Meta-Analysis of the Faking Resistance of Forced-Choice Personality Inventories

“…The findings have implications for the theory of faking at work. The psychometric theory of faking effects suggests that factor structure, construct validity, criterion validity, and incremental validity of personality inventories are negatively affected by faking (Salgado, 2016 ; Otero et al, 2020 ; Martínez et al, 2021a , b ). In connection with this, the findings of this meta-analytic study imply that other characteristics being equal, FC personality inventories would be more robust against the reduction of the validity levels than their counterpart SS personality inventories would be.…”

Effects of Faking on the Predictive Validity of a Quasi-Ipsative Forced-Choice Personality Inventory: Implications for Sustainable Personnel Selection

“…Otero et al [24] also reported evidence of the convergent-discriminant validity of the QI5F_tri using an SS personality inventory. Exploratory factor analyses confirmed the five-factor structure of the QI5F [25].…”

“…Hence, alternatives to SS personality inventories should be considered to reduce the impact of faking on the selection processes. In this sense, forced-choice (FC) inventories are instruments for assessing personality that better control the effects of faking [25,[53][54][55][56].…”

“…These effects would, in turn, lead to a decrease in reliability and a decrease in predictive validity. Nevertheless, recent empirical findings have shown the robustness of the quasi-ipsative FC personality inventories (without algebraic dependency) to the effects of faking: (a) the faking effects were considerably less critical to the average score of the evaluated groups [59]; and (b) the quasi-ipsative FC inventories showed a high degree of measurement invariance (that is, construct validity), under both honest and faking response instructions [25]. Therefore, considering these results, the negative effects of faking on the predictive validity of quasiipsative FC inventories can be expected to be minor or non-existent.…”

“…An extensive number of studies have examined the capacity of personality measures to predict several occupational and academic outcomes since they are a widely used assessment procedure in organizational and educational settings [7][8][9][10][11][12][13][14][15][16][17][18][19][20][21][22]. However, most of this research has been carried out using traditional single-stimulus (SS) personality measures, which are more susceptible to the potential adverse effects of faking (response distortion) [23][24][25][26].…”