Written and maintained by CASRAI Editorial Board
Last updated
The Personality Assessment Inventory (PAI) is a 344-item self-report questionnaire for adults aged 18 to 89, developed by Leslie C. Morey and published by PAR, Inc. It yields 22 nonoverlapping scales (4 validity, 11 clinical, 5 treatment consideration and 2 interpersonal). This page is for researchers, students and methods reviewers who need to understand what the instrument measures, what is known about its reliability, how access is controlled, and how to describe it in a methods section. It is a measurement reference, not a way to assess yourself or anyone else.
What is the Personality Assessment Inventory?
The PAI is a multiscale, objective (fixed-response) instrument that screens for clinical syndromes, response distortion and treatment-relevant features in adults. According to its publisher, it has 344 items, 22 nonoverlapping scales, and ten of those scales carry conceptually derived subscales. The items are written at about a fourth-grade reading level, and PAR lists administration time at roughly 25 to 55 minutes, generally under an hour. It is available in print (with a hand-scoring answer sheet) and digitally. The standard reference is Morey’s professional manual (Morey, L. C. (2007). The Personality Assessment Inventory professional manual, 2nd ed. PAR).
It belongs to the same family of multiscale inventories as the Millon Clinical Multiaxial Inventory (MCMI), but it is a different instrument with its own scale structure. It should not be confused with the public-domain International Personality Item Pool (IPIP), which is a free item bank for building trait scales rather than a licensed clinical test.
Structure: the 22 PAI scales
The scales fall into four groups. Respondents answer each item on a four-point scale: false/not at all true, slightly true, mainly true, very true.
| Group | Scales (code) | Count |
|---|---|---|
| Validity | Inconsistency (ICN), Infrequency (INF), Negative Impression (NIM), Positive Impression (PIM) | 4 |
| Clinical | Somatic Complaints (SOM), Anxiety (ANX), Anxiety-Related Disorders (ARD), Depression (DEP), Mania (MAN), Paranoia (PAR), Schizophrenia (SCZ), Borderline Features (BOR), Antisocial Features (ANT), Alcohol Problems (ALC), Drug Problems (DRG) | 11 |
| Treatment consideration | Aggression (AGG), Suicidal Ideation (SUI), Stress (STR), Nonsupport (NON), Treatment Rejection (RXR) | 5 |
| Interpersonal | Dominance (DOM), Warmth (WRM) | 2 |
The group counts (4, 11, 5 and 2, totalling 22) are published by PAR; the scale names and codes follow the manual’s labelling as reproduced in the research literature. Because the scales are nonoverlapping, no item contributes to more than one scale score. Researchers interested in how multiscale structures are examined empirically can compare this design with exploratory factor analysis of item pools.
Reliability and validity evidence
Paulino et al. (2024), in a validation study of the Portuguese adaptation, summarise the manual’s US evidence: mean internal consistency (Cronbach’s alpha) of roughly 0.81 to 0.86 across the full scales in census-matched, college and clinical samples, and test-retest reliability averaging about 0.82 across the 22 scales. In their own Portuguese sample (n = 900), alphas were adequate for the clinical, treatment and interpersonal scales but low for several validity scales (reported range 0.08 to 0.68).
That pattern is worth understanding rather than treating as a defect. Validity scales such as Infrequency and Inconsistency are built to detect unusual or careless responding, so most people are expected to answer them in a similar low-endorsement way. Internal consistency is therefore weak evidence for them, and low alpha on such a scale does not by itself mean the scale fails at its purpose. When you report reliability for a sample, report each scale you analyse separately and do not substitute the manual’s figures for your own estimates.
For the underlying concepts, see the CASRAI guides on psychometrics and criterion validity, which explain how correlations with outside criteria are used to support an instrument’s interpretation.
Access, qualification and licensing
PAI test online, free, or as a PDF?
There is no legitimate free online version or downloadable PDF of the PAI items. PAR lists the PAI at Qualification Level C, its tier for tests that require advanced training in psychological assessment, and the item content and scoring are copyrighted. For that reason this page does not reproduce questions, item text or scoring keys. Research teams that want to use the PAI obtain it from the publisher, who sets the qualification and licensing terms.
Can the PAI be used for self-diagnosis?
No. Scores are interpreted by trained professionals alongside interview, history and other data, and a scale elevation is a screening signal, not a diagnosis. If you are concerned about your own mental health, speak to a licensed clinician.
Reporting the PAI in research
A methods section should name the instrument and edition, the mode of administration (print or digital), the scales analysed, and how invalid protocols were handled. A template follows; adapt it to the study you actually ran.
Example (illustrative wording, not drawn from a real study): Participants completed the Personality Assessment Inventory (PAI; Morey, 2007), a 344-item self-report measure yielding 22 nonoverlapping scales. Responses were collected [on paper / digitally]. Validity scales were inspected before analysis, and [state your exclusion rule and the number of protocols affected]. Cronbach’s alpha in the present sample was [report per scale].
Reference entry: Morey, L. C. (2007). The Personality Assessment Inventory professional manual (2nd ed.). PAR.
FAQ
How many items and scales does the PAI have?
344 items and 22 nonoverlapping scales: 4 validity, 11 clinical, 5 treatment consideration and 2 interpersonal (PAR).
How long does the PAI take?
PAR lists 25 to 55 minutes, and says most people finish in under an hour.
Who can administer it?
PAR sets the PAI at Qualification Level C, so purchasers must show the required training and credentials.
Are there PAI example questions or a free PDF?
Not from a legitimate source. The items are copyrighted and restricted to qualified purchasers.
Sources
- PAR, Inc. Personality Assessment Inventory (PAI) product page, https://www.parinc.com/products/PAI (read 2026-11-07).
- Paulino, M. et al. (2024). Psychometric properties of the Portuguese version of the Personality Assessment Inventory: normative data and reliability. Frontiers in Psychology, 15, 1359793. https://doi.org/10.3389/fpsyg.2024.1359793 (read 2026-11-07).
- Morey, L. C. (2007). The Personality Assessment Inventory professional manual (2nd ed.). PAR.
