Can you study for an IQ test?

Short answer: You can practice for an IQ test, but you cannot study your way to a higher IQ. If you have taken a similar test before, your score usually rises by about a quarter of a standard deviation, roughly 4 IQ points by our calculation from a 2007 meta-analysis of 134,436 people (Hausknecht and colleagues). Research finds that gain comes mostly from familiarity and strategy rather than a rise in intelligence. So the fair way to prepare is to know what to expect, sleep, and take the test once, rested and honest.

What helps and what does not

The lists below sum up the research. Each point is sourced in the sections that follow.

What helps your score show your real level:

1. Knowing the format and the time limit from the test's own instructions, so nothing on the day is a surprise.
2. A warm-up item or practice puzzle, if the test gives one before the timed part.
3. A normal night's sleep and a normal meal, as one child clinic tells parents.
4. A quiet place, glasses if you need them, and no interruptions.
5. Telling the examiner in advance about an illness or a condition that could affect your performance.

What does not help, or makes the score mean less:

1. Drilling sample items until the format feels easy. The score goes up, but it then partly measures practice.
2. Retaking the same test to chase a higher number. Retest gains are well documented and are not a gain in ability.
3. Brain-training apps. Large reviews find little or no transfer to reasoning tests.
4. Looking up the items of a real clinical test. The APA's ethics code asks psychologists to keep test materials secure, and the APA's guide for patients warns that practicing ahead of a psychological assessment "usually backfires".

How much practice raises an IQ score

Taking a test a second time usually raises the score. Researchers call this the practice effect or retest effect.

The best-measured number comes from a 2007 meta-analysis of 107 samples and 134,436 people. It found an adjusted average gain of .26 standard deviations on cognitive ability tests (Hausknecht and colleagues, Journal of Applied Psychology). The standard deviation on an IQ test is approximately 15 points, as the US Supreme Court noted in Hall v. Florida (2014), so that is about 4 points by our calculation. Hausknecht and colleagues also found larger effects when practice came with coaching and when the identical test was used again.

A 2025 review in the Journal of Intelligence breaks these effects down further (Sommer and Arendasy):

• Same test again: average gains of 0.37 to 0.42 standard deviations with identical retest forms, against 0.22 to 0.23 with alternate forms.
• New items that look alike: when a new form changes only surface features of the items, the gain is often almost as large as with the identical test.
• Practice on sample items: when test-takers practice sample items on their own, with or without feedback, scores rose by 0.22 to 0.40 standard deviations (d = 0.22 to 0.40). The review calls this practice-based training, and the effect grows with how similar the practice items are to the real ones.

On a matrix test, one study found more. Writing in The Conversation in July 2024, psychologists Giovanni Sala and Fernand Gobet note that taking a common nonverbal reasoning test twice raised scores by "roughly the equivalent of eight IQ points". That study gave 35 university students two short-form tests from Raven's Advanced Progressive Matrices on two days about a week apart (Hayes, Petrov and Sederberg, Intelligence, 2015). Treat eight points as one small study's result. The typical gain is about half that.

Clinical tests show the effect too. In the WAIS-5 manual's retest study of 201 people tested about 29 days apart, the Full Scale IQ rose by 3.7 points for ages 16 to 69 and by 2.3 points for ages 70 to 90, as Winter, Trudel and Kaufman report (Journal of Intelligence, 2024). The same authors report gains of 4.5 to 5 points on the WAIS-IV. A 2012 study gave the WAIS-IV twice to 54 young adults, 3 or 6 months apart, and found Full Scale IQ rose about 7 points (Estevis, Basso and Combs, The Clinical Neuropsychologist).

How long you wait matters, but less than you would hope. The 2025 review says retest effects are larger after a short interval, "they remain notable even with time intervals of several years", and they grow with each retest until they level off (Sommer and Arendasy).

Why a practice gain is not a gain in intelligence

A higher second score looks like progress. The studies below trace much of that rise to strategy and familiarity.

• Strategy explains much of it. Hayes and colleagues tracked eye movements while 35 people solved Raven's matrices on two days. They found "that one-third of the variance of score gains could be attributed to test-taking strategy alone", and once that was accounted for, the remaining gains were no longer significant (Intelligence, 2015).
• The authors of a WAIS-IV study read it as practice. In their 2012 retest study, they wrote: "These gains reflect practice effects instead of genuine intellectual changes, which may lead to errors in clinical judgment" (Estevis and colleagues, 2012).
• It blurs what the test measures. Sala and Gobet write that after practice, the score partly measures your expertise at that test, and "practising on a cognitive test essentially makes the test results un-interpretable" (The Conversation, July 2024).
• It can overstate ability. Schneider and Sparfeldt write that score increases after retesting, coaching and training typically "do not reflect an enhanced dispositional intelligence" and can lead to an overestimate of a person's ability (Journal of Intelligence, 2021).

One publisher puts it more loosely. Riverside Insights, which publishes the CogAT school ability test, tells parents that reasoning ability "can be affected by growth, practice, and motivation" (read 24 October 2026). The two views fit together: practice can raise a reasoning test score without raising general intelligence.

There is one thing with good evidence of a lasting effect, and it is slow. A 2018 meta-analysis of over 600,000 participants found education raised cognitive abilities by "approximately 1 to 5 IQ points for an additional year of education" (Ritchie and Tucker-Drob, Psychological Science). That is a year of schooling. A week of practice tests is something else. Our article on whether you can increase your IQ covers that research in more detail.

Do brain-training apps help?

Not with an IQ test. A 2016 review of the studies that brain-training companies cite found "extensive evidence that brain-training interventions improve performance on the trained tasks" but "little evidence that training enhances performance on distantly related tasks" (Simons and colleagues, Psychological Science in the Public Interest).

Later meta-analyses agree. Across working-memory training, video games, music, chess and exergames, the effect on unrelated tasks was essentially zero, between −0.03 and 0.02, once the analysis corrected for publication bias (Gobet and Sala, Perspectives on Psychological Science, 2023). You get better at the game you practice, and that is all.

How to prepare fairly

Fair preparation means arriving able to show what you can do. It never means knowing the answers. Here is what sources support, read on 24 October 2026.

• Read the instructions you are given. Know how many items there are, whether there is a time limit, and whether you can go back. The APA-hosted test takers' guidelines ask you to "Follow the test instructions you are given and represent yourself honestly during the testing" (Joint Committee on Testing Practices, 1998).
• Do the warm-up, if there is one. If the test offers a practice item before the scored part, use it to learn the format.
• Sleep and eat as usual. A child clinic at James Madison University tells parents to encourage a good night's sleep and breakfast before an evaluation. The research is more specific than the advice: a 2010 meta-analysis of 70 articles on going without sleep for up to 48 hours found it hurt simple attention badly while its effect on reasoning accuracy was small and not significant (Lim and Dinges, Psychological Bulletin). A timed test still needs your attention.
• Choose a quiet time and place. For an online test, that means no other tabs, calls or people asking you things.
• Speak up about anything that may affect you. The same guidelines say you should tell the examiner in advance if you need an accommodation or have "a physical condition or illness that may interfere with your performance on the test".
• Don't panic about nerves. A 2018 synthesis of 238 studies found test anxiety was linked with lower scores on standardized tests, with a small to moderate effect (von der Embse and colleagues, Journal of Affective Disorders). Those studies looked at school and university tests rather than IQ tests in particular. Knowing the format is the plainest way to take some of the surprise out of the day.

Before a clinical IQ test

If a psychologist is testing you, don't look for the test items. The APA's guide for patients says "practicing ahead of time usually backfires", because trying to take a test in a certain way can make the results inconsistent (APA, updated August 2022). The same guide says psychological testing is not something you need to study for.

The tests themselves are protected. The APA Ethical Principles of Psychologists and Code of Conduct, Standard 9.11, says "Psychologists make reasonable efforts to maintain the integrity and security of test materials and other assessment techniques" (read 24 October 2026). That is why real clinical items are not published, and why this article has none.

You can ask about retakes. The APA-hosted test takers' guidelines say testing professionals should tell you, on request, how much your score might change if you take the test again, including the error in the measurement itself (Joint Committee on Testing Practices, 1998). Our guide to the WAIS IQ test explains what that adult clinical test includes, and how long an IQ test takes covers the appointment.

Retaking an IQ test: how soon?

There is no single waiting period for every test. What the sources say, read on 24 October 2026:

• American Mensa lets candidates who fail to qualify "retest after eight weeks". Before April 2024 it limited retakes, and it says the change follows direction from the test publishers (American Mensa, 8 April 2024).
• Public schools evaluating a child under the federal IDEA rules may reevaluate "not more than once a year", unless the parent and school agree otherwise (34 CFR 300.303). That rule covers special education evaluations in general, including more than IQ tests.
• Research shows the gain shrinks with a longer gap but does not vanish. Estevis and colleagues found similar gains at 3 months and at 6 months (2012).

If you retake an online test, read the second score as a retest. It will likely read higher for the reasons above, and comparing it with your first score is not a fair comparison.

Preparing a child for an IQ test

Parents can prepare a child without drilling. The James Madison University clinic says that preparing a child "helps to decrease their fears about the evaluation" and suggests telling a school-age child that it will look at their strengths and weaknesses (read 24 October 2026). The clinic also asks parents not to change a child's medication in the two weeks before, except antibiotics and cold medicine.

Sala and Gobet write that a child taking a test such as the Cognitive Abilities Test a second time is likely to do better than the first time (The Conversation, July 2024). If a school or psychologist is testing your child, ask them what to expect. Our guide on where to take an IQ test covers how school evaluations work.

Can you prepare for the Mensa test?

Mensa offers practice, with limits. American Mensa's online practice test takes 30 minutes and "will not qualify you for membership", and it is scored only for people 14 and older (read 24 October 2026). Mensa International's free Mensa IQ Challenge has 35 puzzles in 25 minutes, invites you to take it as many times as you want, and says it is "for entertainment purposes only" (read 24 October 2026).

The admission test itself is supervised, by local volunteers or at approved testing facilities (American Mensa). Sala and Gobet note that standardized tests such as Mensa's are not publicly available, because practicing the same test teaches the logic behind its questions (The Conversation, July 2024).

What a score after practice can tell you

Every IQ score has a margin of error. In Hall v. Florida (2014), the US Supreme Court described the standard error of measurement as reflecting a test's imprecision and said "an individual score is best understood as a range". The court gave five points on either side as an example.

Even the best clinical tests carry this error. On the WAIS-5, the reliability of the overall score (the FSIQ) averaged 0.97 across the age range, as Winter, Trudel and Kaufman report from the manual (2024). By our calculation, that gives a standard error of about 2.6 points and a 90% range of about 4.3 points either side of the score. A retest gain of 3 or 4 points sits inside that range, so a small rise after practice tells you very little.

Three things to keep in mind:

• A first sitting is the fairest estimate you will get from a given test.
• A second sitting on the same or a similar test reads high on average.
• Two scores from different tests can differ even with no practice at all, which our article on whether IQ tests are accurate explains.

If you want a first sitting on a matrix test, the WhatIQ test has "24 picture puzzles, about 15 minutes", with an untimed warm-up puzzle before the timed part (whatiq.com, read 24 October 2026). It gives a score on the standard IQ scale with the range it most likely sits in, and calls the result "an estimate from 24 puzzles, not a clinical assessment".

Frequently asked questions

Should I practice before an IQ test?

Get to know the format, but don't drill. Reading the instructions and doing any warm-up item the test offers is fair. Practice on sample items raises scores by 0.22 to 0.40 standard deviations, a 2025 review in the Journal of Intelligence found. A 2021 study in the same journal says such gains typically reflect test-specific skill or strategy rather than higher intelligence (Schneider and Sparfeldt). The APA warns that practicing ahead of a psychological assessment usually backfires.

How much can practice raise an IQ score?

About 4 points on average, by our calculation from a 2007 meta-analysis of 134,436 people that found a gain of .26 standard deviations (Hausknecht and colleagues). The WAIS-5 manual's retest study found 3.7 points for adults under 70. Sala and Gobet put the gain in one small 2015 study on short Raven's tests at roughly eight IQ points (The Conversation, July 2024). Gains are larger with the identical test.

How soon can you retake an IQ test?

It depends on who gives it. American Mensa lets candidates who fail retest after eight weeks (read 24 October 2026). Public schools may reevaluate a child under IDEA not more than once a year unless the parent and school agree. Practice effects shrink with time but can remain notable even after several years, a 2025 review found, so a retake still reads high.

Do brain-training apps raise your IQ?

The research says no. A 2016 review found brain training improves the trained tasks but gives little evidence of gains on distantly related tasks (Simons and colleagues). A 2023 review of meta-analyses found the effect on unrelated tasks was essentially zero once publication bias was corrected (Gobet and Sala). You improve at the game. Reasoning in general stays where it was.

Can you study for the Mensa test?

Mensa offers an online practice test, 30 minutes long, that does not qualify you for membership (American Mensa, read 24 October 2026). The supervised admission test is not public. Practice may make the format familiar, but any gain reflects the retest effect rather than higher intelligence. Candidates who fail can retest after eight weeks, under a rule American Mensa adopted in April 2024.

Does a higher score on a retake mean my IQ went up?

Usually not. A 2012 study that retested 54 adults on the WAIS-IV found gains of about 7 points and wrote that they reflect practice effects instead of genuine intellectual changes (Estevis and colleagues, 2012). A clinical score also carries an error of a few points either way. Compare first sittings with first sittings, and read any retake as a retest.