Detecting underreport in real-world assessment contexts: The utility of multiple-scale indicators

Resumen

Underreport of symptoms and personality characteristics is a relevant problem for psychological assessment. Nevertheless, most of the studies in this field use simulation designs. This study aims at comparing underreport prevalence in real-world samples of different contexts, using single-scale and multiple-scale underreport indicators from the Minnesota Multiphasic Personality Inventory–2 (MMPI-2) to identify the best one and assess its implications on the clinical scales scores. Using a Differential Prevalence Group design, 1438 participants were assessed with the MMPI-2, grouped in three samples from two non-clinical contexts (community and organizational) and one clinical context. The organizational sample had the highest prevalence of underreporting. Overall, multiple-scale indi-cator LKS ≥ T65 performed better at distinguishing these samples. Analy-sis of variance revealed that LKS ≥ T65 was also the only indicator in which participants doing underreport consistently varied from honest re-sponders in the clinical scales scores, while also having lower probability of producing both type I and II errors. The existence of underreport has clear implication on the clinical scales results. The multiple-scale indicator is the most robust and should be used in the detection of underreporting. This is a relevant implication for psychological assessment in different contexts, mainly in the organizational context.

Áreas y palabras clave

Cómo citar (APA 7)

Gonzalez, B. (2022). Detecting underreport in real-world assessment contexts: The utility of multiple-scale indicators. https://doi.org/10.6018/analesps.502511