What Hiring Managers Should (and Shouldn't) Use Personality Tests For

In 1991, Murray Barrick and Michael Mount ran a meta-analysis across 117 studies and 23,994 working adults. The question was simple: do personality traits actually predict who performs well on the job? The answer they landed on still shapes how industrial psychologists think about hiring three decades later ¹.

The short version: yes, but only one trait predicts performance across almost every job, and the size of the effect is smaller than most hiring managers assume.

The long version is the part worth getting right, because the gap between "personality predicts performance" and "we should screen candidates out by personality score" is where most hiring teams get into trouble — legally, ethically, and statistically.

What Barrick and Mount actually found

Across all jobs and all five Big Five traits, only one trait was a meaningful predictor of job performance everywhere: Conscientiousness. Average correlation around 0.22 with overall job performance ¹. That is not huge — it explains roughly 5% of the variance in who performs well — but it is real and it replicates.

The other traits showed up too, but in job-specific ways:

• Extraversion predicted performance in sales and management roles, where talking to people is most of the job.
• Openness predicted who would do well in training and learn new skills quickly.
• Agreeableness predicted teamwork-heavy roles but had near-zero correlation with individual performance.
• Emotional Stability (low Neuroticism) predicted performance in high-stress roles.

Later meta-analyses have largely confirmed the pattern ². Conscientiousness is the workhorse. The others matter for specific job families.

What this means in practical hiring terms

Personality tests can add real predictive power on top of interviews, especially for entry-level roles where structured interview signal is weak. The research-backed uses are narrow:

• Conscientiousness as a tiebreaker between equally qualified candidates, not as a knockout filter.
• Trait-job fit for sales, leadership, and team roles, where job-specific traits add signal.
• Self-awareness conversations during onboarding, helping a new hire understand their own working style.

Notice what is missing from that list: rank-ordering candidates, screening out applicants below a cutoff score, or making any single hire-or-no-hire call based on a personality result.

Why cutoffs are where it goes wrong

The seductive use of a personality test in hiring is the cutoff: anyone scoring below the 40th percentile on Conscientiousness gets dropped. It feels efficient. It is also a bad idea for three reasons.

1. The effect size does not support it. A correlation of 0.22 means many high performers score below the cutoff and many low performers score above it. You are mostly filtering noise.

2. Adverse impact is a legal problem. In the US, EEOC guidance treats any selection procedure that disproportionately screens out a protected group as creating legal exposure unless you can document job-relatedness. Personality tests can have adverse impact, and the burden of proof falls on the employer ³.

3. Candidates fake. When test results are tied to a job offer, people answer differently than they would in a low-stakes setting. Meta-analyses show measurable score inflation in high-stakes contexts, especially on Conscientiousness and Emotional Stability ⁴. The trait you most want to measure is the one most likely to be distorted.

The legitimate uses, restated carefully

A personality assessment in hiring is most defensible when it is:

• One input among several, never the deciding one.
• Used to predict trainability and fit, not to exclude.
• Validated for the specific role, not borrowed from a generic vendor.
• Discussed openly with the candidate, not used as a secret filter.

The most useful pattern I have seen is this: use the test result as a starting point for a conversation in the final interview. "Your profile suggests you may prefer structured environments — does that match how you have described your best work?" That kind of use respects the candidate, surfaces real information, and avoids the legal traps that cutoff-style use creates.

What candidates should know

If you are on the candidate side and asked to take a personality test, three things are worth knowing.

First, the result is not the whole picture. A reasonable employer will treat it as one input.

Second, you do not need to game it. The "right" answers are not consistent across roles, and obvious gaming (every answer at the extreme) shows up on the impression-management scales most tests include.

Third, if you are asked to take a test that feels invasive — questions about mental health history, very personal life areas, or anything that crosses into clinical territory — that is worth asking about. Standard Big Five and HEXACO instruments do not include those items. Tests that do are different instruments and may have different legal status.

What hiring managers should actually do

If you run hiring and you want to use personality science responsibly:

• Read what the test measures. A vendor's marketing page is not the validation literature.
• Anchor on Conscientiousness for most roles, and add job-specific traits only where the research supports them.
• Use bands, not cutoffs. "This candidate scored in the moderate-low range on Conscientiousness — worth probing in the interview" beats "scored below 40th percentile — rejected."
• Document job-relatedness. If you ever have to defend the use of the test, you want a written rationale that ties the traits measured to the actual job tasks.
• Re-validate every couple of years. A test that predicted performance five years ago in a role that has since changed may no longer predict it.

The single biggest mistake hiring teams make with personality tests is treating them as a precision instrument when the research describes them as a directional one. Used as direction, they help. Used as precision, they create false confidence and real risk.

Personality tests do not tell you who will succeed. They tell you who fits a statistical pattern that, on average, slightly predicts success. That is genuinely useful — and it is also genuinely limited. The hiring teams that get the most out of them are the ones that hold both truths at the same time.

Try a Big Five assessment to see what a candidate-side experience looks like →

References