Construction of item-level scales from the Personality Assessment Inventory to assess levels of personality functioning.

The fifth edition of the Diagnostic and Statistical Manual for Mental Disorders (DSM-5; American Psychiatric Association, 2013) introduced the clinician-rated Levels of Personality Functioning Scale (LPFS) as an indicator of general personality functioning based on four elements: Identity, Self-Direction, Empathy, and Intimacy. Construct validation strategies were employed to select and evaluate items from the Personality Assessment Inventory (PAI; Morey, 2007) to measure the four elements of the LPFS. In Study 1, conceptual ratings of PAI items produced lists of candidate items for the four elements. In Study 2, a sample of student respondents (n = 312) was used to select the final items for the PAI-Levels of Personality Functioning (PAI-LPF). In Study 3, a large sample of adults (n = 505) gathered using Amazon's Mechanical Turk was used to cross-validate the psychometric properties of the PAI-LPF element scales. Means, standard deviations, and coefficient alpha values are reported for the PAI-LPF total score and element scales using the PAI community adult and clinical patient normative samples. The PAI-LPF offers clinicians and researchers the ability to include the LPFS as part of a comprehensive assessment of personality and psychopathology offered by the PAI. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

Prediction of psychotherapy process and outcome with the Personality Assessment Inventory.

The Personality Assessment Inventory (PAI; Morey, 1991) contains scales that were designed to make predictions about how an individual might respond to treatment, thereby allowing clinicians to attune treatment plans to a client's specific needs. The present study utilized two features of the PAI as predictors of treatment process and outcome in a sample of 47 outpatient veterans: the Treatment Rejection (RXR) scale and the Treatment Process Index (TPI). Data were collected for three treatment process and outcome measures: treatment utilization (ratio of appointments attended to appointments scheduled), therapist-rated therapeutic alliance, and symptom change over time. Results indicated that RXR significantly predicted utilization over and above the TPI. The TPI significantly predicted the rate of distress symptom decline over time, but RXR did not. Lastly, neither RXR nor the TPI were significant predictors of therapist-rated alliance.

The validity of a regression-based procedure for detecting concealed psychopathology in structured personality assessment.

Positive Impression Management (PIM) predicted scoring of the Personality Assessment Inventory (PAI; Morey, 2007) is a regression-based procedure designed to inform diagnostic hypotheses when respondents engage in positive response distortion (PRD). Observed scores that deviate above the PIM-predicted score are proposed to indicate areas of true psychopathology that are being concealed by the respondent. To test this proposition, the PAI was administered twice to 334 undergraduate students, using standard instructions and job applicant role-play instructions. For most PAI scales and subscales, the PIM-predicted deviation scores from the role-play condition were significantly correlated with scores on the corresponding scale obtained from the standard administration condition. Exceptions to the general findings are noted for some scales, and recommendations are offered for further research and clinical application of PIM-predicted scores.

Personality assessment screener in a primary care sample of low-income urban women.

The diagnostic efficiency of the Personality Assessment Screener (PAS; Morey, 1997) total score was evaluated using selected scales from the Patient Health Questionnaire (Spitzer, Kroenke, & Williams, 1999), the fourth edition of the Personality Diagnostic Questionnaire (Hyler, 1994), and the Alcohol Use Disorders Identification Test (Saunders, Aasland, Babor, DeLaFuente, & Grant, 1993) as reference standards. Complete data were collected from 110 women seeking treatment at an urban family medicine training clinic. Total PAS scores were effective in identifying patients with mood disorders, cluster B personality disorders, and alcohol use disorders, but the optimum cut scores were higher than the cut score of 19 recommended by Morey (1997). The 10 PAS element scores showed good convergent and discriminant correlations with the reference measures. These findings support the utility of the PAS to screen for major forms of psychopathology in an urban primary care setting.

Introduction to the special issue on the personality assessment inventory.

This special issue of the Journal of Personality Assessment brings together 13 new research studies on the Personality Assessment Inventory (PAI; Morey, 1991) that should inform users and stimulate future empirical activity with this measure. In 4 articles, authors evaluate the validity scales and indexes of the PAI using both analog and criterion designs and samples from a variety of clinical and forensic settings. In a 5th article, the authors describe a novel approach to profile interpretation using two PAI negative distortion measures. The authors present applications of the PAI to new populations and problems including a German translation of the PAI and profile information for male batterers and victims of head injury. The authors of 2 studies extend research on the validity of the PAI for the assessment of borderline personality disorder. In the final 3 studies, the authors evaluate the validity of PAI measures of violence and aggression to predict subsequent aggressive behavior and institutional misconduct. Finally, the authors offer several suggestions for future research with the PAI.

Another look at paradoxical severity effects in head injury with the Personality Assessment Inventory.

Previous investigations of psychiatric symptomatology after head injury using the MMPI-2 (Butcher, Dahlstrom, Graham, Tellegen, & Kaemmer, 1989) have consistently revealed greater Basic scale elevations in mild injuries versus more severe injuries. In this study, we tested this pattern of paradoxical severity effects using the Personality Assessment Inventory (PAI; Morey, 1991). We gathered PAI and MMPI-2 data from 34 patients with moderate-to-severe head injuries and from 52 patients with mild head injuries. MMPI-2 Basic scale profiles were consistent with the paradoxical severity effect; mild injury patients had significantly more elevated scores on four Basic scales (Scales 1, 2, 3, and 7). PAI Clinical scale profiles showed significantly more elevated scores among mild injury patients on 2 scales, Somatization and Depression, and more elevated scores among moderate-to-severe patients on 2 scales, Antisocial Features and Alcohol Problems. We consider unique contributions of the PAI for the psychological assessment of head injury.