Response bias and the Personality Inventory for DSM–5: Contrasting self- and informant-report.

Abstract

Previous research has raised concerns that scores derived from the Personality Inventory for DSM-5 (PID-5; American Psychiatric Association, 2013; Krueger, Derringer, Markon, Watson, & Skodol, 2012) may be compromised by response styles such as underreporting or overreporting. The informant-report form of the PID-5 (PID-5-IRF; Markon, Quilty, Bagby, & Krueger, 2013) has been recommended for use when response bias is an assessment concern. The purpose of the current investigation was to evaluate PID-5 and PID-5-IRF scale score elevations across participants exhibiting signs of overreporting or underreporting. A total of 245 adults completed the PID-5 and the Revised NEO Personality Inventory (NEO PI-R; Costa & McCrae, 1992). A family member or friend of at least 1 year's acquaintance completed the PID-5-IRF for 216 of these. A total of 211 target-informant pairs were available for analysis. Participants were categorized as overreporting and underreporting according to NEO PI-R validity scale cutoffs. The majority of PID-5 scale scores were elevated in those identified as overreporting; more than half of the PID-5-IRF scale scores were similarly elevated. The majority of PID-5 scale scores were lower in those scoring above underreporting cut-offs; however, PID-5-IRF scales were not as consistently or strongly impacted. PID-5 scales were strongly impacted by response bias, whereas PID-5-IRF scores were less strongly impacted overall, and more so by overreporting bias. Caution when using these instruments in the assessment of personality disorders prone to over- or underreporting may be warranted. (PsycINFO Database Record

Sex differences in associations of socioemotional dispositions measured in childhood and adolescence with brain white matter microstructure 12 years later

“…Finally, we found a sex-moderated association between prosociality and FA in the SFOF, Although we observed significant associations between FA and CADS dispositions rated by one informant, these did not replicate using the other raters' scoring. This is consistent with previous studies reporting only low to moderate agreement between informant ratings of temperament and personality traits across youth self-reports and adult raters (Boson, Brandstrom, & Sigvardsson, 2018;Capaldi & Rothbart, 1992;Quilty, Cosentino, & Bagby, 2018;Tackett, 2011). Similarly, CADS dimensions rated by parents and youth are only modestly correlated (Class et al, 2019).…”

Validity of informant report interpretations: Role of examinee performance and symptom invalidity.

“…Many of the same incentives that drive volitional examinee exaggeration may similarly entice an informant, potentially invalidating interpretations of informant-reported examinee functioning. Extant research in this area has primarily focused on personality (Quilty et al, 2018) and adult ADHD assessment (Marshall et al, 2016; Nelson & Lovett, 2019) but has not been conducted in the context of adult neuropsychological evaluations of neurocognitive disorders. Given that performance and symptom invalidity have been repeatedly shown to be strongly associated with worse performance on examinee-completed cognitive tests and higher symptom reports on examinee-completed self-report measures, respectively (Larrabee, 2012), we expected that examinee response invalidity would relate to greater informant-reported dysfunction in the examinee.…”

“…Some data from the personality assessment literature suggests that examinee symptom invalidity is associated with overreporting on informant report measures (Quilty et al, 2018). There is also evidence that examinee response invalidity in adult ADHD evaluations is associated with attenuated inter-rater reliability (Nelson & Lovett, 2019).…”

A Brief but Comprehensive Review of Research on the Alternative DSM-5 Model for Personality Disorders

“…Other examples are studies showing that individual differences in trait facets are relatively stable across periods of 2 weeks [57, 235], 4 months [285], and more than 1 year [286]. Further issues that have been addressed include measurement invariance or item bias due to age [287,288], gender [217,289], and clinical status [290]; response styles in PID-5 self-reports [154,155,291,292]; heritability and familial aggregation of maladaptive traits [39,289,[293][294][295]; and perceived likability, impairment, functionality, as well as desire and ability for change of maladaptive traits [68,158,[296][297][298].…”