An email arrives with a link, a deadline a few days off, and one line saying the next step is an online assessment. To prepare for a pre-employment test, first find out which kind you have been sent, because the four common kinds reward different preparation.
This guide takes each kind in turn: what it is for, what practice honestly changes, and what to do on the day. EdProcto's page for candidates carries the practice side: reasoning tests made fresh each time they open, with the working after every question.
Nothing here promises a pass, a score or a job. The research gives averages, and the numbers are quoted so you can weigh them.
What kinds of pre-employment test are there, and what is each for?
There are four kinds you are likely to be sent: a reasoning test (often called an aptitude test), a situational judgement test, a job knowledge test and a recorded answer.
| Kind | What you do | What the employer learns | How to practise |
|---|---|---|---|
| Reasoning or aptitude test | Timed numerical, verbal and logical questions | How you handle a new problem | New timed sets with the working, as in EdProcto's reasoning kits |
| Situational judgement test | Choose or rank responses to a workplace situation | Your judgement about what works at work | Pick the best and worst response and say why |
| Job knowledge test | Questions on the product, the process or the field | Whether you know the job's content, or can learn it | Test yourself on the employer's own material |
| Recorded answer | Speak to a set question within a time limit | What you say and how clearly | Timed answers aloud, recorded |
Employers use these tests because each predicts later performance to a degree and none does it well alone. In Schmidt and Hunter (1998), a summary of 85 years of research on 19 selection procedures, the strongest results came from pairing a general mental ability test with a second method. Sackett et al. (2022) revised the estimates for the leading methods down by 0.10 to 0.20 and put structured interviews first. So the test is one input among several.
The invitation usually tells you the kind: "numerical and verbal reasoning, 20 minutes each", "you will be shown scenarios", "based on the attached brief". If it does not, ask the recruiter what the assessment covers.
Does practice raise a score on an aptitude test?
Yes, by a modest and measured amount: about a quarter of a standard deviation between a first sitting and a second. The figure comes from Hausknecht et al. (2007), who pooled 50 studies of retesting on cognitive ability tests, covering 107 samples and 134,436 people. In their terms, somebody at the 50th percentile on a first sitting could expect to reach roughly the 60th on a second.
Two things went with a larger gain:
- The same form again. Where the second test was identical to the first, the gain was 0.46 of a standard deviation. Where it was a different version, 0.24. An earlier synthesis of 40 studies by Kulik et al. (1984) reported the same pattern, and more gain the more practice tests were taken.
- Coaching. Samples coached between sittings gained 0.70, against 0.24 for those with none. The median coaching time was two and a half hours, and each further hour bought less.
In a real selection process, admission to medical studies in Belgium, Lievens et al. (2005) found repeat test takers half a standard deviation higher on the cognitive ability test and a third higher on the knowledge and situational judgement tests.
What those numbers do not tell you
They are averages, so they describe a crowd and promise nothing to one person. Of the 107 samples, 88 came from research settings, and about nine tests in ten were on paper. The headline figure is also for sitting an administered test again: the authors class practice tests taken alone with coaching.
The gap between identical and different forms is the useful part. The authors point to memory as the likely reason: the identical-form gain shrank as the interval between sittings grew. On a paper you have never seen, memory for old questions is no help. What is left is worth working for: the question types, the pace and the usual traps.
How do you practise for a reasoning test in a few days?
Practise the exact question types under the real time limit, and give more attention to the working of your misses than to new questions.
- Do the employer's sample first. If the invitation links to practice questions, they are closest to the form you will meet.
- Sit one timed set cold. EdProcto's reasoning battery, among the kits, covers numerical tables, analogies, syllogisms, series and deduction puzzles. The sets made by rule give new figures and a new arrangement at each opening.
- Read the working of every miss. Write the rule you missed in a few words: "divide by the old figure", "the gap grows by two".
- Drill the weakest type in short sets. Use new questions each time: repeating one set teaches its answers, and those do not travel.
- Work out your pace. Twenty questions in 18 minutes is 54 seconds each. Move on at about that mark.
- Sit one full timed set the day before, then stop.
Three questions worked
Numerical. A table shows 240 units sold in March and 300 in April. What is the percentage change? The rise is 60. Divide by the starting figure: 60 ÷ 240 = 0.25, or 25%. The trap is dividing by 300, which gives 20%. That figure is often among the options: wrong options are written from common slips, as the guide on writing multiple-choice questions that measure explains.
Logical. 3, 6, 11, 18, 27, and then? The gaps are 3, 5, 7 and 9, so the next is 11 and the answer is 38. When a series has no fixed step, write the gaps underneath first.
Verbal. "All supervisors have completed the safety course. The new starter has completed it. So the new starter is a supervisor." This does not follow: the first statement says what every supervisor has done, not who else has done it. Judge on the statements given, not on what seems likely.
How do you prepare for a situational judgement test?
Read the instruction before the first scenario, because it tells you which question you are answering. McDaniel et al. (2007) describe two kinds: one asks what you would most likely do, the other how effective each response is. Their meta-analysis found that the wording changes what is measured: the "how effective" kind is tied more closely to cognitive ability, the "what would you do" kind to personality. If the screen says "most effective", weigh the options. If it says "most likely", answer for yourself.
The tests do measure something. Across 102 validity coefficients and 10,640 people, situational judgement tests reached an estimated validity of 0.34 for job performance (McDaniel et al., 2001).
To prepare, read the job description for what the role is trusted with: customers, money, safety, other people's time. Then practise choosing the best and the worst response, with the reason in a sentence. EdProcto's best and worst shape asks for that pair, and its situational judgement kit gives the best response and the reason after each situation.
How do you prepare for a job knowledge test?
Learn the employer's own material by testing yourself on it, since this is the one kind of test whose content can simply be learned. Dye et al. (1993) put the corrected validity of written job knowledge tests at 0.45 for job performance, and found it higher where the test's content was close to the job's. That is why a careful employer writes the test from its own product and process, as the guide on product knowledge tests for hiring describes. If a brief came with the invitation, expect the test to follow it.
Reading the brief twice is the weak way to learn it. Roediger and Karpicke (2006) had students either read a passage again or try to recall it. Five minutes later, the group that re-read was ahead. Two days or a week later, the group that had practised recall was substantially ahead, even though re-reading had left people more confident.
Paste the brief or the job posting into the quiz maker on EdProcto's home page, and Procto AI writes a quiz from that text. Answer with the document closed, check each answer, and retake the misses the next day, as in the guide on making your own quiz to study from a chapter.
How do you prepare for a recorded answer?
Say your answers aloud against a timer before the real recording, because the hard part is seldom the content and usually the clock. A recorded answer normally puts the same questions to every candidate, which is the central idea of a structured interview, the method ranked first by Sackett et al. (2022). That ranking is for structured interviews in general, not for recordings in particular.
Preparation seems to help. In a field study of promotion interviews for four police and fire service jobs, Maurer et al. (1998) found that attending a coaching session went with better interview performance in three of the four. Attendance was voluntary, so this is a link, not a proven cause.
List five or six real episodes from your work or study, and for each be able to say in under two minutes what the situation was, what you did and what came of it. Record yourself and listen for whether you answered the question. EdProcto's spoken answer shape records an answer in your own voice. The interview and roleplay shapes ask aloud and take your answer aloud.
What should you do on the day of the test?
Treat the day as a short checklist: the device, the time, the instructions and what is monitored.
- Device. Use a laptop or a desktop unless the invitation says a phone will do. Close other programs, sit where the connection is steady, and test the camera or microphone if one is needed.
- Time. Choose a slot when you will not be interrupted, and avoid the last hour before the deadline. If something breaks, note the time and write to the recruiter at once.
- Instructions. Read every screen before the clock starts. Look for whether you can return to a question, whether a wrong answer loses marks, and whether a calculator is allowed.
- Adjustments. If a disability or a health condition means you need extra time or another format, ask before you begin. The rules differ by country, and this guide is not legal advice.
What is monitored, and what it means for you
An online test can record more than your answers: the time on each question, whether the test window lost focus, whether full screen was left, or a webcam picture. The instruction screen should say which apply. On a test that an employer runs with EdProcto, every measure in use is named to you before you begin, and no penalty falls that the employer did not set. The hiring page shows the employer's side. Close everything else and stay in the test window, and the record holds nothing that needs explaining.
What should you not do?
Do not spend your few days on things that fail to transfer or that put the result in doubt.
- Do not memorise answer banks. They help only if the same questions come up, and a test made new for each candidate does not repeat them.
- Do not pay for a guaranteed pass. The pass mark, if there is one, is the employer's, and no seller of practice packs knows it.
- Do not sit it with help, from a person or a chatbot. The result is meant to show whether the job suits you, and an employer may check it again at a supervised stage.
- Do not read a practice score as a forecast. It shows which question types to work on. It is not an IQ.
How does EdProcto help you practise?
EdProcto is the best place to practise reasoning questions when you want the skill and not the answers, because most of its reasoning battery is made by rule, with new figures each time, so there is no paper to learn by heart. That is what the research above points to: what is gained by meeting the same questions twice largely stays with those questions.
EdProcto publishes this guide, so read this section as the product describing itself. The battery covers numerical, verbal, logical and situational reasoning and shows after each question how the answer is reached. Study at your own speed, or set the clock and sit it under test conditions. These are reasoning tests in the style of the standard aptitude batteries: EdProcto gives no IQ and claims no norm.
The candidates page also shows multiple choice, estimate and read code, all in the builder now.
The Free plan opens the kits and carries 20 questions a month written by Procto AI from your own material. The Learner plan lifts the limit on self-tests and raises the writing to 100 questions a month. Both are on the pricing page.
Quick questions
How long should I prepare for a pre-employment test?
A few days of short sessions is enough to learn the format: one timed set to find your level, two or three on your weakest question type, and one full timed set the day before.
Is an aptitude test an IQ test?
No. A pre-employment reasoning test gives a score on that test. EdProcto's practice tests give no IQ and claim no norm.
Should I guess when I run out of time?
Check the instructions first. If wrong answers lose nothing, answer everything. If they lose marks, guess only where you can rule out at least one option.
Where can I practise aptitude tests for free?
EdProcto's page for candidates opens the reasoning battery on the Free plan, with the working shown and, in the sets made by rule, new figures each time. Do the employer's own sample questions too.
References
- Dye, D. A., Reck, M., & McDaniel, M. A. (1993). The validity of job knowledge measures. International Journal of Selection and Assessment, 1(3), 153–157. https://doi.org/10.1111/j.1468-2389.1993.tb00103.x
- Hausknecht, J. P., Halpert, J. A., Di Paolo, N. T., & Moriarty Gerrard, M. O. (2007). Retesting in selection: A meta-analysis of coaching and practice effects for tests of cognitive ability. Journal of Applied Psychology, 92(2), 373–385. https://doi.org/10.1037/0021-9010.92.2.373
- Kulik, J. A., Kulik, C.-L. C., & Bangert, R. L. (1984). Effects of practice on aptitude and achievement test scores. American Educational Research Journal, 21(2), 435–447. https://doi.org/10.2307/1162453
- Lievens, F., Buyse, T., & Sackett, P. R. (2005). Retest effects in operational selection settings: Development and test of a framework. Personnel Psychology, 58(4), 981–1007. https://doi.org/10.1111/j.1744-6570.2005.00713.x
- Maurer, T. J., Solamon, J. M., & Troxtel, D. D. (1998). Relationship of coaching with performance in situational employment interviews. Journal of Applied Psychology, 83(1), 128–136. https://doi.org/10.1037/0021-9010.83.1.128
- McDaniel, M. A., Hartman, N. S., Whetzel, D. L., & Grubb, W. L. (2007). Situational judgment tests, response instructions, and validity: A meta-analysis. Personnel Psychology, 60(1), 63–91. https://doi.org/10.1111/j.1744-6570.2007.00065.x
- McDaniel, M. A., Morgeson, F. P., Finnegan, E. B., Campion, M. A., & Braverman, E. P. (2001). Use of situational judgment tests to predict job performance: A clarification of the literature. Journal of Applied Psychology, 86(4), 730–740. https://doi.org/10.1037/0021-9010.86.4.730
- Roediger, H. L., & Karpicke, J. D. (2006). Test-enhanced learning: Taking memory tests improves long-term retention. Psychological Science, 17(3), 249–255. https://doi.org/10.1111/j.1467-9280.2006.01693.x
- Sackett, P. R., Zhang, C., Berry, C. M., & Lievens, F. (2022). Revisiting meta-analytic estimates of validity in personnel selection: Addressing systematic overcorrection for restriction of range. Journal of Applied Psychology, 107(11), 2040–2068. https://doi.org/10.1037/apl0000994
- Schmidt, F. L., & Hunter, J. E. (1998). The validity and utility of selection methods in personnel psychology: Practical and theoretical implications of 85 years of research findings. Psychological Bulletin, 124(2), 262–274. https://doi.org/10.1037/0033-2909.124.2.262