Types of Pre-Employment Assessment: 8 Kinds Compared (2026)

The short answer

The eight types of pre-employment assessment, compared

Ordered by how well each predicts job performance in the 2022 re-analysis by Sackett, Zhang, Berry and Lievens. Validity runs from 0 to 1 and higher is better. The group gap is the Black-White difference in applicant scores, and it is where legal challenges start.

Type	Validity	Group gap	Candidate time	How it is priced
Structured interview	.42	.23	45 to 60 minutes	Your own time. Interview-kit software is optional.
Job knowledge test	.40	.54	20 to 40 minutes	Per candidate, or inside a platform subscription.
Skills test or work sample	.33	.67	30 to 90 minutes	Credits or per candidate. TestDome charges $7 to $20 a candidate.
Cognitive ability test	.31	.79	12 to 30 minutes	Per candidate, or an annual licence.
Integrity test	.31	.10	15 to 30 minutes	Per report.
Personality questionnaire	.25	−.07	6 to 20 minutes	Monthly by respondents, per report, or a headcount licence.
Situational judgement test	See note	See note	20 to 30 minutes	Usually built for you, by quote. Clevry lists an off-the-shelf SJT at £40 a test.
Physical or role-specific test	Not in the 2022 table	Varies by test	Varies	Per test, usually through an occupational health provider.

Note. We print a figure only where we have checked it against the 2022 paper. Situational judgement tests are in that paper, but we could not confirm the value, so the cell is left without one. Physical tests were not part of it. The personality row is conscientiousness with work-worded items; with general wording it is .19. Candidate times are typical ranges from vendor documentation. Prices are from our read of 28 vendors' pricing pages.

Why these numbers look low

The figures most guides quote were corrected in 2022

For two decades the reference table was Schmidt and Hunter's 1998 review, which put cognitive ability and work samples above .50. In 2022 Sackett and colleagues showed that a statistical correction had been applied too generously across those studies. Redone with care, most methods fell by .10 to .20.

The order changed with them. Structured interviews moved to the top, cognitive ability dropped from first to sixth, and years of experience came out at .07. Assessment still works. The claim that any single test predicts performance above .50 does not.

The eight, one at a time

What each type of pre-employment assessment measures

01

Structured interview

Job-relevant behaviour, using the same questions in the same order for every candidate, scored against written anchors.

The strongest single method in the 2022 estimates, and the one you already run. What does the work is the structure: fixed questions, a rating scale, and scoring before the panel talks. Results vary a lot with how well the interview is built, so an interview scored from memory is an unstructured one.

Best for: Every hire.
Watch out: It only counts if you score against the anchors.

02

Job knowledge test

Whether someone already knows the body of knowledge the job depends on: tax rules, a programming language, clinical procedure.

Close behind the interview, but only where the role has real content to know on day one. It cannot be used for entry-level roles you intend to train, and it rewards people who have done the job before over people who would learn it quickly.

Best for: Experienced hires into technical or regulated work.
Watch out: Useless where you plan to teach the content anyway.

03

Skills test or work sample

Performance on a real slice of the job: a coding task, a writing brief, a bookkeeping problem.

The method candidates rate as fairest, because it looks like the work. It also took the largest downward revision in 2022, from .54, and it carries the second-widest group gap in the table. Many vendor pages still quote the older figure. Fix the scoring rubric before anyone sits it, or the result drifts back into opinion.

Best for: Any role you can cut down to a short, fair task.
Watch out: Long unpaid tasks cost you good applicants.

04

Cognitive ability test

Reasoning with unfamiliar information, and how quickly someone learns.

Long described as the best single predictor at .51. The 2022 re-analysis put it at .31, sixth among the methods it covered. It still earns a place in complex roles with a steep learning curve, but it has by far the widest group gap of any common method, which is where adverse-impact challenges begin. Write down why the job needs that level of reasoning.

Best for: Complex roles where people must absorb new material fast.
Watch out: The highest adverse-impact risk here.

05

Integrity test

Dependability, rule-following, and the pull towards counterproductive behaviour such as theft or absenteeism.

As predictive as a cognitive test with almost none of the group gap, which makes it the quiet strong performer of the eight. The cost is on the candidate side: people find the questions intrusive, and a clumsy rollout reads as suspicion. Explain what is being measured and why the role needs it.

Best for: Cash-handling, safety-critical and high-trust roles.
Watch out: The method candidates like least.

06

Personality questionnaire

How someone tends to work: planning, persistence, pressure, and how they deal with people. Usually the Big Five.

Conscientiousness measured with work-worded items reaches .25, with almost no variation between studies. With general wording it reaches .19. That is modest, and it comes with essentially no group gap and a view of things a skills test cannot see, such as how a person treats colleagues. Use it to decide what to ask in the interview, never as a pass mark. The instruments differ a great deal, and we compare them on a separate page. See the personality tests employers use, side by side.

Best for: Role fit, interview questions, onboarding.
Watch out: Self-report. Applicants present themselves favourably.

07

Situational judgement test

What a candidate chooses to do in realistic dilemmas drawn from the job.

Useful for customer-facing and people-heavy roles, where the easy response and the right one diverge. Quality depends almost entirely on the job analysis behind the scenarios. One written from real incidents in your own workplace measures judgement. A generic one measures reading comprehension.

Best for: Customer service, supervision, values-sensitive work.
Watch out: Off-the-shelf versions measure very little.

08

Physical or role-specific test

Strength, stamina, vision, or another capacity the work cannot be done without.

The category with the heaviest legal care. Each requirement has to be tied to an essential function of the job, applied to everyone in the role, and in the United States anything medical in nature waits until after a conditional offer. Sex differences on strength tests are large, so the job-relatedness record matters more here than anywhere else.

Best for: Safety-critical and physically demanding roles.
Watch out: The highest disability-law exposure of the eight.

Start from the problem

Which type of assessment for which hiring problem

Find the sentence that sounds like your last bad hire.

The problem	Start with	Add
People we hire can do the work, then clash or leave	Personality questionnaire	Structured interview that probes the two or three traits the report flags
We cannot tell whether applicants have the skill	Skills test or work sample	A fixed scoring rubric, agreed before anyone sits it
The role has a steep learning curve	Cognitive ability test	A written record of why the job needs it, and a lower-gap method beside it
The role handles cash, safety or confidential data	Integrity test	Structured interview and reference checks
Front-line judgement calls go wrong	Situational judgement test built from your own incidents	Structured interview
Interviewers disagree, or hire on instinct	Structured interview	Nothing. Fix this before buying anything
Too many applicants to read	A short personality or job knowledge screen	A person who reads the ranking. No automatic rejection
The work is physically demanding	Physical ability test, after a conditional offer	A documented link to essential job functions

Two or three, not five

How to combine assessments. Buy information you do not already have.

Sensible combinations of methods average a validity of about .47, down only slightly from .51 under the older estimates. No single method gets there alone.

01

Pick methods that do not overlap

A second cognitive test tells you what the first one did. A personality questionnaire beside a skills test tells you something new. Each method should earn its place with information the others lack.

02

Put the interview at the centre

It is the strongest method and it is free. Use test results to decide what to ask, then score the answers. A personality report is most useful as the source of those questions.

03

Trade a little validity for fairness

Weighting methods so that selection rates meet the four-fifths ratio costs about .09 of validity in the 2024 matrix by Berry and colleagues. That is a small price for a process you can defend.

Candidates judge the process too. Across 86 samples, interviews and work samples were rated most fair and personality and integrity questionnaires least (Hausknecht, Day and Thomas, 2004). Say what you are measuring, why the role needs it, and how long it takes.

Legal basics

Keeping pre-employment testing legal and fair

Job-related

Under the EEOC's Uniform Guidelines, every scored test is a selection procedure. Write down why each thing you measure matters for this role, before you send anything.

Watched for adverse impact

If one protected group passes at under four-fifths the rate of another, the burden moves to you. Vendor marketing does not count as evidence of validity.

Not medical

The ADA holds medical examinations until after a conditional offer. Clinical questionnaires count as medical. Normal-range work inventories do not.

Decided by a person

New York City's Local Law 144 requires a bias audit for automated decision tools, and the EU restricts purely automated decisions. Keep a human in the decision.

General information, not legal advice. The rules for personality tests, including the case law, are set out in are personality tests legal for hiring? The UK position is on our assessment tools page.

If you have been asked to take one

This guide is written for employers, but the same facts apply to you. A personality questionnaire has no right answers, and answering as yourself is the approach that serves you, because the job that fits the real you is the one you will want to keep. If you would like to see how one reads before you sit it, our work personality test is free and gives you your own report, and so does the free personality test on our homepage.

Questions people ask

Pre-employment assessment: questions and answers

Q.01

What is a pre-employment assessment test?

It is any standardised, scored method used before a hiring decision to measure something that predicts job performance: skill, knowledge, reasoning, personality, judgement or integrity. Standardised is the important word. Every candidate gets the same questions under the same conditions and is scored the same way, so you compare people with each other instead of comparing the impressions they left.

Q.02

What is a pre-employment testing tool, and do I need one?

A pre-employment testing tool is software that delivers one or more of the eight types online, scores them and ranks the results. You need one for skills, cognitive, integrity and personality tests, because scoring them by hand is slow and unreliable. You do not need one for the strongest method, the structured interview, which takes a question list and a rating scale. Prices and billing units vary widely, and most vendors do not publish them.

Q.03

What are selective hiring assessments?

Selective hiring means choosing carefully from a large pool against criteria set in advance, and the assessments that support it are the ones on this page. The practice that makes hiring selective is order: decide what the role needs, choose two or three methods that measure those things, fix the scoring, and only then look at candidates. Criteria adjusted after meeting the applicants are preferences, not selection.

Q.04

What are the best pre-employment assessments for predicting long-term job performance and retention?

For performance, the 2022 estimates put structured interviews first (.42), then job knowledge tests (.40), work samples (.33), and cognitive and integrity tests (.31 each). Retention is a different outcome. It is predicted better by fit between the person and the job, which is what personality and interest measures describe, and by integrity tests, which predict absenteeism and counterproductive behaviour. A combination covers both.

Q.05

How do I choose the right pre-employment assessment for my company?

Start from the job rather than the catalogue. Write down the two or three things that separate people who succeed in the role from people who do not, pick the type that measures each, and avoid buying two that measure the same thing. Then check the billing unit against how you hire: per candidate if you hire rarely, a subscription if you hire all year. For most office roles, a structured interview plus one test is enough.

Q.06

What is the difference between pre-employment testing and pre-employment screening?

Screening is the wider term: everything checked before an offer, including background checks, references and right-to-work documents. Testing means the scored, predictive instruments described here. Screening confirms facts about the past. Testing estimates how someone will perform. The legal rules differ too: background checks fall under consumer-reporting law in the United States, and tests fall under the rules on selection procedures.

Q.07

Are pre-employment tests legal?

Yes, when they are job-related and given to every candidate the same way. In the United States the Uniform Guidelines on Employee Selection Procedures treat any scored test as a selection procedure, so if it screens out a protected group at a much higher rate you must be able to show it measures something the job requires. Medical examinations wait until after a conditional offer. This is general information, not legal advice.

Keep reading

Related guides

About this page

Michael Hodge, founder of SeeMyPersonality, holds a BSc in Psychology from the University of Wollongong and has run online personality measurement since 2003. More about the author.

We make a tool in one of the eight categories, the personality questionnaire, and this guide reports that category's figure the same way as the others: sixth of the six we could verify. Validity figures are averages across many studies and describe methods in general, not any product.

Written and reviewed by Michael Hodge Updated 18 September 2026 Disclosure we sell a personality assessment for hiring

Sources (8)

If personality is one of your two or three

Build one for a real role and read a sample result first.

About a minute, no account. It writes the interview questions from the gaps it finds.

Build a hiring assessment Comparing vendors? See who publishes a price.

The 8 types of pre-employment assessment, and what each predicts