Psychometric Test for Job Applications: What Employers Are Measuring

The email lands on a Tuesday with a link and a 72 hour deadline. Three sections, 55 minutes, do not refresh the browser. A psychometric test for a job is a standardised assessment that scores you against a comparison group on something the employer believes predicts performance at work. That might be reasoning ability, stable personality traits, judgement in work situations or your output on a sample of the actual job.

Which of these you are sitting changes how you should approach it. Most advice treats them as one thing, which is why so much of it is unhelpful.

What are the main types of psychometric test for a job?

There are four families, and each one rewards a different kind of preparation.

Many large employers run aptitude and personality sections in the same sitting, sometimes with a situational section between them. Check the instruction screen for each section, because the rules on timing and guessing often change from one to the next.

What does a psychometric test actually predict?

Less than older textbooks claimed. Sackett, Zhang, Berry and Lievens reexamined the classic validity figures in the Journal of Applied Psychology in 2022 and argued that earlier estimates had been overcorrected. On their revised numbers, structured interviews came out on top at .42, job knowledge tests at .40, work samples at .33 and general cognitive ability at .31. Most mean validity estimates dropped by .10 to .20 from the figures that had been quoted for decades.

Two practical points follow. A correlation of .31 means a cognitive test explains under a tenth of the differences in later job performance, so any single test is a weak signal about one person. And careful employers know this, which is why they combine a test with an interview and often a work sample. Your aptitude score is one column in a spreadsheet, sitting next to other evidence.

Work samples are the one family where preparation and performance are the same activity. A coding task, a case study or a mock sales call is a direct look at the work, which also makes it the fairest thing on the list to be judged on.

Why do personality tests have no right answers but still catch people out?

There is no correct score on a personality questionnaire. There is a profile the employer decided suits the role, and checks built into the instrument to see whether your answers hang together.

Applicants do inflate their scores, and by measurable amounts. Birkeland and colleagues pooled 33 studies in the International Journal of Selection and Assessment in 2006 and found job applicants scored higher than non applicants on every Big Five dimension they tested. The gaps were largest for conscientiousness, at .45 of a standard deviation, and emotional stability, at .44, with much smaller differences for extraversion and openness. For some jobs, such as sales, the pattern shifted toward the traits applicants assumed the role wanted.

Test publishers defend against this in several ways. Social desirability scales include statements almost nobody can honestly agree with in full, and a high score on those gets your profile flagged. Consistency checks ask about the same trait several ways, including reverse worded items. Forced choice formats make you pick between two equally attractive statements, so you cannot simply agree with everything.

The part test preparation sites rarely mention is what happens when faking works. You get placed in a role chosen for a version of you that does not exist. Shape your answers to look like a highly structured, detail focused operations person because the advertisement implied it, and you may spend eighteen months in work that grinds against how you actually operate. Answer as your normal working self and answer quickly. Overthinking each item is what produces the inconsistencies the software is built to spot.

How do situational judgement tests work?

A situational judgement test gives you a short workplace scenario and a set of possible responses, then asks you to rank them, rate how effective each one is, or pick the best and worst. Your choices are scored against a key built from expert judgement or from what strong performers in the role chose.

Read the instruction wording closely, because it changes the test. A question asking what you should do is testing knowledge of good practice. A question asking what you would do is testing your instinct, and some employers prefer that version for exactly that reason.

Memorising answer patterns does not work, because the key is specific to each scenario and the options are written so that several look defensible. What works is knowing what the role puts first. A manufacturing job with safety risks wants you to stop the line and escalate. A hospital puts patient safety ahead of team harmony. Read the job description, the stated values and any published competency framework, then answer from inside that frame.

Can you improve your score by practising aptitude tests?

Yes, and the evidence is unusually clear. Hausknecht, Halpert, Di Paolo and Moriarty Gerrard pooled 107 samples covering 134,436 test takers in the Journal of Applied Psychology in 2007. Scores rose by an average of .26 standard deviations from one sitting to the next, and the gains were larger when the same test form was used again and when practice included coaching.

Be clear about what that buys you. Practice gives you familiarity with the answer screen, the clock and the fact that most numerical reasoning items are percentage change or ratio arithmetic set inside a table. It does not change your underlying reasoning ability, and working through hundreds of items reaches diminishing returns quickly. Two or three timed runs in the real format do most of the work.

Two mechanics are worth understanding before the day. Many aptitude tests are adaptive, meaning the software estimates your level after each answer and serves the next item to match. If the questions keep getting harder, you are probably doing well. Also check whether wrong answers are penalised. Most tests do not penalise them, and in that case a blank is strictly worse than a guess. In high volume campus assessments, where cut offs exist to reduce numbers, unanswered items are the most common avoidable loss.

What actually helps with test anxiety?

Sitting a timed assessment three months into a job search, with money running short, is not a neutral experience. The anxiety is reasonable, and being told to relax does not help.

Ray Hembree combined 562 studies in a meta analysis in the Review of Educational Research in 1988. He concluded that test anxiety causes poorer performance, as opposed to simply appearing alongside it, and that a range of treatments reduce it, with better test results following those reductions. The practical reading is that anxiety is worth treating as a real cause of lost marks.

On a Tuesday evening that means a few concrete steps. Do one timed run under real conditions so the format stops being new. Sit the test at the hour you think best. Remove the problems that turn into panic, such as an unstable connection, a housemate walking in or a laptop on low battery. If your anxiety is severe enough to be a health condition, ask the employer in writing about adjustments before the test window opens. Extra time, screen reader support and an alternative format are common requests.

What should you do if you fail a psychometric test?

1. Ask for feedback in writing. Some employers send a profile summary, some send nothing, and asking costs you nothing.
2. Ask about the retest policy. Many set a wait of six or twelve months, and the practice effect means a second sitting tends to score higher.
3. Work out which family you fell short on. Missing an aptitude cut off is a different problem from a personality mismatch, and only the first is worth practising for.
4. Check whether it was a fixed cut off or a ranking. In high volume graduate hiring, rejection often means you sat below a line set by that year of applicant numbers.
5. Avoid drawing big conclusions. One employer comparison group and cut off tell you little about the next one.

A rejection at the assessment stage is a weak signal about you and a fairly good signal about how that employer designs its process. Responsible employers use a test as one input among several and never as the only filter, which is what the validity figures above require. The guide to the candidate screening process shows where testing usually sits among the other stages, and AI interview questions by role covers the interview that often follows.

Sources

1. Journal of Applied Psychology, Sackett, Zhang, Berry and Lievens, Revisiting meta analytic estimates of validity in personnel selection, 2022
2. International Journal of Selection and Assessment, Birkeland and colleagues, A meta analytic investigation of applicant faking on personality measures, 2006
3. Journal of Applied Psychology, Hausknecht, Halpert, Di Paolo and Moriarty Gerrard, Retesting in selection: A meta analysis of coaching and practice effects for tests of cognitive ability, 2007
4. Review of Educational Research, Hembree, Correlates, causes, effects, and treatment of test anxiety, 1988