Big Five Personality Tests in Hiring: Validity, Fairness, and Ethical Limits

7 days ago

19 min read

Author: Ukrainian Psychological Hub · Published: October 9, 2026 · Editorial Policy

Big Five personality tests can provide useful information during hiring when a particular instrument measures a job-relevant construct and the employer has credible evidence that its scores contribute to fair, accurate selection. The decision to use such a test requires a job analysis, instrument-specific validity evidence, a defensible scoring policy, accessibility, privacy safeguards, and continuing review. A Big Five score is a probabilistic research measure of personality tendencies; it is neither a professional license nor a diagnosis nor a forecast of what one candidate will do.

The U.S. Office of Personnel Management's assessment guidance recognizes that personality measures may predict aspects of job performance, while emphasizing that relevance depends on the job and that personality measures generally have less predictive validity than several other assessment methods. The U.S. Equal Employment Opportunity Commission addresses personality tests as employment selection procedures subject to applicable anti-discrimination law. This article examines the evidence and the obligations that arise when employers translate group-level associations into consequential decisions about individuals.

What does a Big Five hiring test actually measure?

The Big Five model describes five broad and approximately continuous domains of personality: openness to experience, conscientiousness, extraversion, agreeableness, and neuroticism, sometimes framed in instruments as emotional stability or negative emotionality. A questionnaire operationalizes these constructs through selected items and a scoring procedure. A proprietary employment assessment may sample the domains, selected narrower characteristics, or work-oriented variants. An employer needs to know precisely which construction was used before making claims about validity.

Conscientiousness broadly captures organization, diligence, dependability, and goal-directed effort. Extraversion concerns social engagement, assertiveness, and positive emotional engagement. Agreeableness addresses cooperation, compassion, and interpersonal orientation. Openness involves curiosity, imagination, and engagement with ideas and experience. Neuroticism is associated with tendencies toward negative emotional reactions; some selection research instead reports an emotional-stability score with a reversed direction. Domain names can differ between instruments, so a hiring report must preserve the specific scale definition and scoring direction.

A domain score is a location on a continuum according to a particular test and reference group. The Big Five does not partition applicants into five types, ten types, or a fixed set of ideal employee personalities. As our detailed guide to Big Five score interpretation explains, a percentile indicates relative standing in a defined normative sample, not an independent probability of work success. Employers should not create universal high/low cutoffs by borrowing labels from unrelated online tests.

Domains, narrower aspects, and facets also require care. The NEO family, Big Five Inventory-2, and Big Five Aspect Scales organize lower-level content differently. The existence of a conscientiousness result does not establish that the test measured every narrower behavior employers may be interested in. Read the instrument manual and the discussion of Big Five domains, aspects, and facets before equating one scale with a highly specific workplace behavior.

Are Big Five personality tests valid for predicting job performance?

At the broad research level, some personality traits have replicable associations with workplace outcomes. The foundational Barrick and Mount meta-analysis found conscientiousness-related associations across several occupational groups and performance criteria, while other Big Five dimensions varied more by job and outcome. That result supported the study of personality as one source of selection information, rather than a universal personality ranking of applicants.

A more recent meta-analytic reality check on conscientiousness reviewed 102 correlations from 23,305 participants and reported an overall observed mean correlation of approximately .17 with job performance. Subgroup estimates were roughly .18 in incumbent samples and .14 in applicant samples, with .18 in concurrent and .15 in predictive designs; those subgroup differences were not statistically established in the reported analyses. Crucially, only a small fraction of the studies used actual applicants followed in predictive designs. This makes generalization to every real selection setting a substantive empirical question.

The correlation of .17 should not be read as a 17% chance of making a good hire, a 17-point performance improvement, or an individual accuracy rate. Squaring a simple correlation gives a rough proportion of variation associated with a linear relation in that particular setting, and prediction performance can change with restriction of range, reliability, criterion quality, and applicant composition. A meta-analytic coefficient is a population-level summary across designs, samples, measures, and jobs. It cannot certify a vendor's test in a particular organization.

The peer-reviewed 2025 review of personality and job performance differentiates task performance, organizational citizenship behaviors, and counterproductive work behavior. It observes that personality associations vary across these outcomes and that situational demands shape trait expression. A person may have a similar underlying disposition in two jobs yet face different incentives, autonomy, training, leadership, and opportunities. An employer has to define the performance outcome it actually seeks.

Why job analysis comes before a personality questionnaire

Hiring validity begins with a description of the work, not a vendor's personality profile. A systematic job analysis identifies essential tasks, competencies, conditions, and observable indicators of successful performance. The OPM job-analysis guidance treats this as the foundation of assessment and selection. For example, a role requiring careful documentation and repeated quality checks creates a plausible connection to dependability, but the employer still needs evidence that its chosen score measures and predicts the relevant behavior.

Translate each proposed trait into a documented claim: which task matters, which behavior represents competent performance, why the construct might relate to that behavior, and what evidence supports the inference. Decide whether direct work samples, structured interviews, job-knowledge measures, or training assessments would provide more job-proximal evidence. A broad factor score may be useful, but it is not inherently superior to assessing the actual skill.

Job analysis also clarifies when a construct becomes inappropriate. A candidate's social confidence should not become a disguised requirement for a technically focused position where clear written communication is the central task. Likewise, preferences for social interaction are not a valid substitute for observing whether the candidate can handle customers effectively in structured exercises. Selecting for a preferred personality style without demonstrating job relevance can narrow access without improving selection quality.

Criterion validity, construct validity, and incremental validity

Criterion-related validity concerns the statistical relation between test results and a meaningful job outcome. A concurrent study measures existing workers and their current performance; a predictive study examines applicants and later performance. These designs answer related but different questions. The OPM assessment glossary defines both approaches and their use in selection research.

Construct validity concerns whether test responses support the interpretation that a particular psychological attribute was assessed. A questionnaire can show a recognizable Big Five factor structure while providing weak evidence for the employer's specific selection purpose. Conversely, a predictor can correlate with a criterion for reasons that do not support the intended personality interpretation. An ethical selection program needs both an interpretable measurement claim and relevant work-outcome evidence.

Incremental validity asks what the personality result contributes after the existing assessment process has already predicted. If a structured interview and work sample capture most of the useful information, the extra personality questionnaire may add little. The relevant comparison is not 'test versus no test' in isolation, but whether adding that test improves a complete hiring process in a demonstrable way. Claims about added predictive accuracy need evidence from the specific combination of methods and its evaluation sample.

Reliability matters because a score affected heavily by random error is less useful for ranking applicants. Fairness matters because a predictor that performs differently across groups or depends on inaccessible administration can distort selection even if its average correlation appears favorable. The technical details of general reliability belong to psychometrics; the key practical question here is whether the actual Big Five instrument has appropriate evidence for this exact selection decision. See our instrument-specific Big Five validity guide.

What the five traits may and may not tell an employer

Conscientiousness is the most consistently discussed broad domain in the job-performance literature. It is relevant to dependability and task follow-through, yet its relationship with performance varies by instrument, criterion, setting, and applicant conditions. A high score cannot guarantee reliability; a low score cannot demonstrate misconduct or incompetence. In the research, some effects are modest, and the actual task may require competencies best assessed through other methods.

Extraversion may be relevant where frequent social engagement or assertive influence constitutes an important part of the job, but 'more extraverted' is not a universally better employee. Jobs involving listening, focused analysis, or asynchronous collaboration may value different behaviors. Even in a sales or service role, a realistic work sample may evaluate communication more directly than a broad personality scale.

Agreeableness may relate to cooperative, courteous, or conflict-sensitive work behaviors, but it cannot be used as a moral measure of kindness or obedience. Important jobs may require principled disagreement, independent judgment, difficult feedback, and boundary-setting. An employer should evaluate the quality of those behaviors, rather than treating lower agreeableness as a defect or a license to infer risk.

Openness may correlate with learning-related outcomes or receptiveness to novel experiences in some contexts. It does not certify creativity, intelligence, innovation, or technical skill. A person may be highly curious without having the training for a role, or may be technically innovative despite a questionnaire profile that looks average. Job-relevant exercises supply more direct evidence for many specialized capabilities.

Neuroticism or reverse-scored emotional stability deserves special protection from overinterpretation. A trait score describing emotional response tendencies is not a psychiatric assessment, an anxiety diagnosis, a determination of fitness for duty, or a valid way to infer a person's private medical history. Its employment use requires especially careful review of job relevance, disability-related questions, medical-examination rules, accessibility, and data privacy. Separate clinical and neuroticism-specific intents belong to their own dedicated research areas.

Does a higher personality score mean a better hiring decision?

Selection decisions require a comparison with a specific job criterion, plus a clear account of error. An employer may legitimately ask whether scores add useful statistical information, but cannot convert a correlation between two variables into a deterministic ranking of individuals. Candidate A's modestly higher conscientiousness score does not establish that A will outperform B, especially when uncertainty, measurement error, interviews, skills, and context are considered.

Rank-ordering can magnify small score differences. If two applicants are close together and the instrument's measurement precision is limited, placing them into separate 'recommended' and 'not recommended' bands may create false certainty. A vendor should document the meaning of score differences near any decision boundary, the reliability relevant to that boundary, and the criterion evidence supporting the chosen procedure.

Prediction is also affected by the criterion used. Supervisor ratings can capture useful observations but may reflect halo effects or unequal opportunities. Objective outcomes sometimes depend on team assignments, geography, case difficulty, or work resources. Research associations may shift if performance is measured as customer satisfaction, safety, errors, production, turnover, or rule violations. The selection design must declare its intended criterion before examining apparent success.

Can applicants fake Big Five personality tests?

Applicants have incentives to present themselves favorably. The Birkeland and colleagues meta-analysis reported applicant–nonapplicant differences in several personality domains, with patterns that varied by the job. These findings support concern about impression management, while leaving open multiple explanations for differences between populations and contexts.

A particularly relevant meta-analysis of within-person comparisons involving actual applicants synthesized 20 studies and found that high-stakes responses were, on average, more socially desirable and showed somewhat less variation than lower-stakes responses. The existence of a mean response shift, however, does not establish that every applicant lies or that every personality assessment loses criterion validity.

Several approaches have been investigated, including warnings, response-quality indices, contextualized items, and forced-choice formats. A meta-analysis of forced-choice personality inventories found evidence of reduced susceptibility to faking for particular formats and research designs. Forced choice is not a blanket guarantee of honest measurement: its scoring method, whether it yields normative individual comparisons, and the exact item design must be validated for selection use.

An employer should not label a candidate dishonest because an algorithm flags 'inconsistent answers.' Inconsistency can arise from ambiguity, reading difficulty, disability-related circumstances, translation issues, or legitimately context-sensitive behavior. Our dedicated article on Big Five response bias and social desirability examines the underlying mechanisms and assessment alternatives in more detail.

Fairness: what does it mean for a personality test?

Fairness is broader than an average difference in trait scores. A procedure may require equitable access, consistent administration, relevant constructs, comparable score interpretation, scrutiny of prediction errors, and assessment of the entire selection system. The employer needs to ask who is measured accurately, whose scores may be distorted, who advances, and whether a less exclusionary method can accomplish the same job-related objective.

A small mean group difference cannot prove absence of adverse impact once cutoffs, weights, and combinations of measures are introduced. Likewise, a difference between populations on a score does not itself demonstrate that a particular test is biased. Relevant empirical questions include differential item functioning, measurement equivalence where applicable, differential prediction, recruitment funnel composition, accommodations, and observed selection rates. Each question needs appropriate data and methodology.

Bias can enter before the questionnaire is administered. Job advertisements, language requirements, recruitment channels, device accessibility, time pressure, and who can afford reliable connectivity may alter the applicant pool. It can also enter after the questionnaire, when interviewers interpret identical numerical scores differently because of assumptions about disability, age, gender, culture, or communication style.

US employers need to distinguish ethical fairness review from legal determinations. The EEOC's guidance on employment tests explains that intentional discrimination and unjustified adverse impact can raise federal legal issues. The applicable legal analysis depends on the protected characteristic, employment setting, jurisdiction, procedure, and factual record. A scientist's validation report is evidence for the process; it is not an automatic legal safe harbor.

The four-fifths rule and adverse-impact monitoring

Under the Uniform Guidelines on Employee Selection Procedures, the four-fifths or 80% rule is a screening rule of thumb for comparing selection rates across groups, not a universal definition of discrimination or a declaration that a procedure is fair. Small samples, statistical variability, and different decision stages complicate its interpretation. The federal-guideline explanations place the rule inside a broader disparate-impact analysis.

For a simplified example, if 40% of applicants in one group advance and 28% in another group advance, the ratio is .70. That is below .80 and warrants investigation under the rule of thumb. The calculation itself does not establish unlawful discrimination, and a ratio above .80 does not guarantee the selection practice is valid, nondiscriminatory, or ethically justified. Employers should examine counts, uncertainty, jobs, stages, and alternative procedures.

Document separately the outcomes of each stage: who was invited, who completed the test, who received an accommodation, who was scored, who was screened out, and who was ultimately hired. Investigate whether a personality measure contributes useful job-relevant information compared with a less exclusionary alternative. Avoid manipulating group scores or using protected-class-specific cutoffs; the EEOC notes restrictions under Title VII on race-, color-, religion-, sex-, or national-origin-based score adjustment.

Disability, medical examinations, and reasonable accommodation

Ordinary work-related personality questionnaires and clinical or medical examinations are not legally interchangeable. The EEOC's preemployment disability guidance explains that a psychological assessment can be treated as a medical examination if designed or used to reveal mental impairments or disorders. A nonmedical assessment of habits, preferences, or honesty may be treated differently. The actual content, purpose, administration, and interpretation matter.

Before a conditional job offer, the Americans with Disabilities Act generally restricts disability-related inquiries and medical examinations. After a conditional offer, different requirements apply, including consistency across individuals entering the same job category and protection of medical information. The EEOC's current public summary describes these stages. Employers should not assume that renaming a diagnostic screening tool a 'personality assessment' changes its legal character.

Reasonable accommodation can affect how an assessment is offered, completed, and interpreted. Employers must consider accessible formats, assistive technology compatibility, disability-related barriers, appropriate alternative processes, and the essential job skills actually being assessed. Accessibility review belongs in procurement and deployment, not merely in an appeal after automated rejection.

Applicants should understand which personal information is necessary, who evaluates it, how results influence decisions, and how to request an accommodation or report a technical problem. An employer should avoid asking candidates to disclose diagnoses in response to an ordinary workplace personality score. Such disclosure may be sensitive and unnecessary for the claimed selection purpose.

Why 'culture fit' is a weak justification

'Culture fit' is often used to mean compatibility with unwritten preferences about communication, sociability, or behavior. Without a defined job-related criterion, the label can turn subjective comfort into a selection rule. Employees can share an organization's values while displaying diverse personalities and interaction styles. A selection program should state observable behavioral requirements and test their relation to job success rather than reward resemblance to current staff.

An employer concerned about collaboration can evaluate specific teamwork behaviors in structured interviews, simulations, and references appropriate to the role. A general preference for extraversion, agreeable responses, or emotional composure can unnecessarily exclude candidates who would perform the work effectively. Culture-related claims need a defensible connection to the job rather than a retrospective story about why a score feels right.

Combining personality with other hiring evidence

The most defensible role for a personality questionnaire is usually supplementary. The OPM assessment-strategy guidance recommends standardized processes, current job analyses, strong validity evidence, and fair applicant treatment. A useful combined battery can include structured interviews, direct work tasks, credential verification where relevant, and other job-related assessments. The weighting of each component must be justified and checked as a whole.

A personality result can guide an interview question if the organization clearly separates the test's hypothesis from the candidate's observed answer. For example, a low score on a planning-related scale might prompt the same structured question that every candidate receives about managing competing deadlines, with a standardized behaviorally anchored scoring rubric. It should not become an invitation to cross-examine one applicant about an allegedly 'bad personality.'

When an organization has already adopted an effective selection battery, it should test the incremental contribution of the Big Five measure using out-of-sample data if feasible. The decision to add a measure should consider operational cost, time burden, applicant reactions, privacy, and adverse impact as well as predictive performance. A statistically detectable increment may still be too small or ethically costly to justify a new high-stakes data collection procedure.

Privacy and data governance in personality assessment

Personality responses are sensitive personal data in practical terms, even when a given law does not classify every work-style inventory as medical information. Individual answers can reveal deeply personal preferences and patterns, and secondary analyses may support inferences beyond the purpose for which information was collected. An ethical employer should collect only what is necessary for a disclosed, valid selection purpose.

A responsible data policy identifies the controller or responsible entity, purpose of collection, lawful basis where relevant, vendor recipients, cross-border transfers, retention period, access permissions, deletion process, security controls, and uses that are prohibited. It should establish whether vendors can reuse applicant responses to train models or build unrelated commercial profiles. Candidates should receive accessible, intelligible notices before testing, not discover these practices later.

Managers normally need a job-relevant decision summary, not unrestricted access to item-level answers or broad intimate personality narratives. Access should be role-based and logged. Test reports should not be casually copied into team chat, shared with future employers, or kept indefinitely for unspecified reasons. Retention obligations differ across jurisdictions; organizations should reconcile minimization with their legal recordkeeping duties.

Automated scoring, AI inference, and personality claims

AI can be involved in item selection, scoring, report generation, or statistical prediction from video, language, social media, and other digital traces. These are distinct activities. A validated self-report questionnaire is not equivalent to a model that infers a Big Five label from an applicant's face, recorded speech, or online profile. A high correlation with a questionnaire score is not, by itself, evidence of accurate job-performance prediction or procedural fairness.

For an AI-based personality inference system, employers need evidence of construct validity, representative validation samples, stability under changes in context, subgroup error rates, independent evaluation, data provenance, and performance in real applicant populations. Model accuracy on an internal dataset can deteriorate in new languages, job families, demographic groups, devices, and recording environments. More elaborate technology does not remove the need for job relevance and accountable selection criteria.

If a model claims that it 'knows' a candidate's personality from a brief interaction, the claim exceeds what a statistical association necessarily establishes. Separate the instrument's construct score, the machine's estimate of that score, and the relationship between either score and later work outcomes. Transparent human review and a meaningful way to contest errors are especially important when rejection is automated.

How to evaluate a vendor selling Big Five hiring tests

Start by requesting the exact instrument name, version, manual, licensing status, target population, constructs, item format, administration conditions, scoring algorithm, and normative reference group. Ask whether an applicant version was validated under high-stakes conditions, in the relevant language, on populations comparable to the intended candidates. A marketing statement that 'Big Five is scientifically proven' does not answer those questions.

Require a technical report with reliability estimates relevant to the intended decision; factor or other construct evidence; job-related criterion studies; study designs; sample sizes; confidence intervals; missing-data handling; potential conflicts of interest; and evidence for each proposed cutoff or score weighting. Examine whether performance claims come from incumbent or applicant samples, a concurrent or predictive design, and a directly relevant performance outcome.

Then request evidence about differential prediction, accessibility, reasonable accommodations, response distortion, group selection rates, and independent audit. For proprietary algorithms, require clear contractual audit rights and enough documentation for meaningful scrutiny. The employer retains responsibility for its process even when a vendor markets the tool as compliant.

Finally, map every proposed feature to a decision. If the report includes extensive 'leadership potential,' 'culture match,' 'resilience,' or 'retention risk' scores, ask whether those are independently defined and validated for the specific use. A Big Five foundation is not automatic validation of every derived prediction. Do not accept unverified cutoffs, private formulas described as 'scientific AI,' or compulsory clinical-style interpretations.

A responsible implementation sequence

1. Define the job and success criteria before selecting a questionnaire; document tasks, essential competencies, and foreseeable barriers. 2. Decide which personality construct could plausibly add value, and compare it against a direct, less intrusive way of measuring the relevant behavior. 3. Choose a named instrument and review its evidence in the intended language, applicant population, and selection context. 4. Predefine the decision rule, weights, cutoff rationale if any, and independent evaluation plan; avoid inventing universal personality thresholds. 5. Check the full assessment battery for prediction quality, accessibility, differential prediction where assessable, and selection-rate disparities. 6. Give candidates clear information, reasonable accommodations, an appropriate review channel, and accurate descriptions of how scores are used. 7. Restrict data access, prohibit undeclared secondary uses, and establish security, recordkeeping, retention, and deletion policies. 8. Monitor outcomes over time, reassess as jobs or applicant populations change, and suspend use if evidence or fairness concerns undermine the procedure.

These steps are an operational governance approach rather than a ready-made regulatory certification. The EEOC guidance, OPM job-analysis resources, and published validation literature supply the relevant starting points; sector-specific and state law may impose additional duties. Competent employment-law and assessment expertise should be part of the implementation.

Questions employers should answer before rejecting a candidate

Would the candidate's result be interpreted differently if the person were older, disabled, from another linguistic background, or less fluent in the assessment's dominant cultural assumptions? Is the same decision threshold supported for the job and the actual applicant population? Can the employer explain how a small score difference changes the prediction and what the prediction uncertainty is?

Could a structured interview, work sample, reasonable accommodation, or alternative administration reveal the capability more directly? Is the hiring manager tempted to treat a trait score as an explanation for motivation, honesty, mental health, or moral character? Does the process permit a reasoned appeal or review? These questions should be considered before a personality scale has meaningful exclusionary power.

Frequently asked questions

Are Big Five personality tests legal for hiring in the United States?

Employment-related personality tests can be used under applicable US law, but their design and deployment must comply with anti-discrimination and disability rules. The EEOC addresses selection practices with intentional discrimination or unjustified disparate impact; pre-offer medical examinations raise separate ADA restrictions. State and local rules can add requirements. Legality depends on the specific method, use, jurisdiction, and evidence.

Which Big Five trait is the strongest predictor of employee performance?

Conscientiousness has one of the most consistent broad-domain associations with overall performance across studied jobs, as described in Barrick and Mount's meta-analysis. Its value in an actual hiring process depends on the particular questionnaire, job, criterion, applicant population, and other selection methods. The magnitude of its average association does not establish one candidate's future performance.

How accurate is a Big Five hiring test?

There is no universal 'Big Five hiring accuracy' number. Accuracy depends on whether the instrument actually measures its intended traits, the quality of its scores, the job-related outcome, the validation design, and the way results enter a selection decision. A correlation in one study is not an accuracy percentage. See the separate article on accuracy and reliability of Big Five measures.

Can a candidate fail a Big Five personality test?

A personality dimension itself has no medically normal pass/fail threshold. Employers can establish selection rules, but any cutoff should have an appropriate, job-related empirical and legal justification for that particular use. A low trait percentile is not the same as a failed ability task, a diagnosis, or proof of unsuitable character.

Do personality tests discriminate against protected groups?

Potential adverse impact depends on the actual procedure and the populations being selected. The absence of large average differences in a study does not guarantee fairness in a weighted decision system. Employers should assess accessibility, score functioning, selection outcomes, and less exclusionary alternatives as appropriate, using the EEOC employment-testing guidance as a US federal starting point.

Can employers detect when applicants fake their answers?

Some methods can detect patterns consistent with response distortion, but no universal system reliably identifies individual dishonesty from a Big Five response profile. Candidate responses are affected by context, item interpretation, culture, and high-stakes motivation. Read the evidence on Big Five faking and response bias.

Should employers use the BFI-2 or NEO for hiring?

Those are named personality instruments, not interchangeable hiring certifications. Their item sets, scales, intended uses, manuals, scoring, norms, and validation evidence differ. Employers must verify authorization for commercial or employment use and obtain selection-context evidence for the specific version. A high-quality research scale is not necessarily validated for an employer's proposed exclusionary decision.

Is neuroticism an anxiety or mental-health diagnosis?

No. Neuroticism is a continuous personality construct. A trait result cannot establish a psychiatric diagnosis, and clinical or medical examinations trigger distinct safeguards. The EEOC explains when psychological examinations are treated as medical in its ADA preemployment guidance.

Can an employer reject an introvert for a customer-facing role?

An employer should assess demonstrable job-relevant communication behaviors and support its selection procedure with appropriate validity evidence. Extraversion scores alone do not show whether a candidate can communicate effectively with customers. A standardized role-play or behavioral interview may produce more direct evidence for the actual tasks.

Do forced-choice personality tests solve the problem of faking?

Some forced-choice formats show greater resistance to faking in research. That does not mean every forced-choice instrument generates fair, reliable, individually comparable selection scores. Its scoring method and validation evidence must be assessed for the intended employment setting, as reviewed in Martínez and Salgado (2021).

Can AI infer Big Five traits from interviews or digital footprints?

Statistical models may predict selected self-report trait scores from digital traces in a specified evaluation sample. That achievement is different from measuring an underlying construct with established validity and different again from predicting job success. Context shifts, privacy issues, demographic bias, and generalizability are substantial concerns. Employers should demand independent evidence for each inference and each decision use.

What should applicants ask about a personality assessment?

Applicants can reasonably request information about the test's purpose, how scores affect decisions, accommodations, who accesses responses, retention, and whether a human can review an apparent error. Specific access and legal rights vary by location and employer. A transparent employer should be prepared to explain job relevance without asserting that a score reveals a person's full character.

References

Barrick, M. R., & Mount, M. K. (1991). The Big Five personality dimensions and job performance: A meta-analysis. Personnel Psychology, 44, 1–26. https://doi.org/10.1111/j.1744-6570.1991.tb00688.x

Birkeland, S. A., Manson, T. M., Kisamore, J. L., Brannick, M. T., & Smith, M. A. (2006). A meta-analytic investigation of job applicant faking on personality measures. International Journal of Selection and Assessment, 14, 317–335. https://doi.org/10.1111/j.1468-2389.2006.00354.x

Derous, E., & Van Muylem, D. (2025). Personality in personnel recruitment. Current Opinion in Psychology, 66, 102117. https://doi.org/10.1016/j.copsyc.2025.102117

Hu, J., et al. (2021). Faking by actual applicants on personality tests: A meta-analysis of within-subjects studies. International Journal of Selection and Assessment, 29, 412–426. https://doi.org/10.1111/ijsa.12338

Martínez, A., & Salgado, J. F. (2021). A meta-analysis of the faking resistance of forced-choice personality inventories. Frontiers in Psychology, 12, 732241. https://doi.org/10.3389/fpsyg.2021.732241

Morgeson, F. P., Campion, M. A., Dipboye, R. L., Hollenbeck, J. R., Murphy, K., & Schmitt, N. (2007). Reconsidering the use of personality tests in personnel selection contexts. Personnel Psychology, 60, 683–729. https://doi.org/10.1111/j.1744-6570.2007.00089.x

Office of Personnel Management. (n.d.). Assessment glossary. https://www.opm.gov/policy-data-oversight/assessment-and-selection/assessment-glossary/

Office of Personnel Management. (n.d.). Designing an assessment strategy. https://www.opm.gov/policy-data-oversight/assessment-and-selection/assessment-strategy/

Office of Personnel Management. (n.d.). Job analysis. https://www.opm.gov/policy-data-oversight/assessment-and-selection/job-analysis/

Office of Personnel Management. (n.d.). Personality tests. https://www.opm.gov/policy-data-oversight/assessment-and-selection/other-assessment-methods/personality-tests/

Pletzer, J. L., & Abrahams, L. (2025). Personality and job performance: A review of trait models and recent trends. Current Opinion in Psychology, 102054. https://doi.org/10.1016/j.copsyc.2025.102054

Tett, R. P., Jackson, D. N., & Rothstein, M. (1991). Personality measures as predictors of job performance: A meta-analytic review. Personnel Psychology, 44, 703–742. https://doi.org/10.1111/j.1744-6570.1991.tb00696.x

Truxillo, D. M., Bauer, T. N., Campion, M. A., & Paronto, M. E. (2006). A field study of the role of Big Five personality in applicant perceptions of selection fairness, self, and the hiring organization. International Journal of Selection and Assessment. https://doi.org/10.1111/j.1468-2389.2006.00351.x

U.S. Equal Employment Opportunity Commission. (1995). Enforcement guidance: Preemployment disability-related questions and medical examinations. https://www.eeoc.gov/laws/guidance/enforcement-guidance-preemployment-disability-related-questions-and-medical

U.S. Equal Employment Opportunity Commission. (2007). Employment tests and selection procedures. https://www.eeoc.gov/laws/guidance/employment-tests-and-selection-procedures

U.S. Equal Employment Opportunity Commission. (n.d.). Pre-employment inquiries and medical questions & examinations. https://www.eeoc.gov/pre-employment-inquiries-and-medical-questions-examinations

Watrin, L., et al. (2023). The criterion-related validity of conscientiousness in personnel selection: A meta-analytic reality check. International Journal of Selection and Assessment, 31, 286–301. https://doi.org/10.1111/ijsa.12413

Psychological Encyclopedia