Math IQ test for adults: what a normed one scores, and why math is two abilities

0 The short answer

A math IQ test is a normed measure of quantitative ability, and it has to say which of two abilities it scores: learned mathematics (quantitative knowledge, Gq) or reasoning with quantities (which the Cattell-Horn-Carroll model files within fluid reasoning, Gf). A clinician can give one inside a Wechsler or Woodcock-Johnson battery, employers use timed numerical reasoning tests that are not IQ tests, and many free "math IQ" pages online are quizzes with no stated norm group. ACIS is one concrete option: Arithmetic and Mathematical Achievement are in the Full Scale form only (50 dollars, about 175 minutes), or in the Custom Test for 10 dollars, which returns the Quantitative Reasoning Index and two scaled scores (prices read October 28, 2026, and they can change). The decisive limit is that it is online and unsupervised, so it is not a clinical or diagnostic instrument and it is not for hiring or school decisions.

1 Is There Such a Thing as a Math IQ?

There is no single math IQ: a test that asks mathematical questions can be sampling learned schoolwork, reasoning with numbers, or the memory and attention that mental calculation uses, and a normed battery keeps those apart. The word "math" in a test name hides at least three different demands. The first is acquired knowledge of number facts, operations and procedures, which the Cattell-Horn-Carroll (CHC) model calls quantitative knowledge, Gq, and which instruction builds. The second is reasoning about quantities and relations in a problem the person has not met before, which the CHC literature files with fluid reasoning, Gf, as the narrow ability quantitative reasoning. The third is working memory (Gwm), the capacity to hold a number while transforming it, which mental arithmetic loads heavily.

The page on the CHC model lays out the broad and narrow abilities, and the page on what IQ measures shows where they sit in a full battery. McGrew's 2009 paper in Intelligence is the account of how CHC theory joins Carroll's three stratum survey of human cognitive abilities to the older Cattell-Horn Gf-Gc tradition, and that joining is the reason quantitative skill carries two homes in the model. The full argument for why the name "quantitative reasoning" is slippery belongs to the dedicated page on the distinction between the two quantitative abilities, so this page takes the result and applies it to the question a test taker asks: which test, which score, and what does the number mean?

Publishers tag their own tasks this way. The Woodcock-Johnson V brochure from Riverside Insights, read October 28, 2026, labels its Number Series test with both Gq and Gf, its Applied Problems and Calculation tests with Gq, and its Math Facts Fluency test with Gs and Gq, and it groups the achievement tests into math clusters that carry the Gq label. The same document labels Concept Formation with Gf alone. In other words, a major publisher itself treats number series as a task that straddles the two abilities, and tags calculation and applied word problems as quantitative knowledge.

For a reader who wants one practical rule, it is this. If a test is mostly school mathematics, its score is largely an achievement score, and it will rise with instruction. If a test is mostly new rules over quantities or visual structure, its score is closer to reasoning, and it will track a Full Scale IQ more closely. A good report tells you which one you are looking at, which is the single most useful thing to check before you pay for any test that has "math" in its name. The page on fluid intelligence tests covers the second ability more broadly, and the technical manual is the place to read how ACIS classifies each of its tasks.

2 What Do Free Math IQ Tests Online Actually Give You?

Many pages that rank for "math IQ test" are quizzes, and a quiz can be entertaining without being a measurement: a score needs a norm group, a stated reliability and a defined scale before it means anything about you. The searches people type include "math iq test free", "math iq test online" and "math iq test questions", and a live search on October 28, 2026 for "math iq test for adults online" returned mostly entertainment quizzes, puzzle sites and prep pages rather than test manuals. Four checks separate a measurement from a quiz.

First, is there a stated comparison group? An IQ score is a position in a distribution of people, normally on a scale with a mean of 100 and a standard deviation of 15, and the page on how IQ scores are normed explains why the same raw performance gives different numbers against different groups. A quiz that converts a raw count into "your IQ" with no group behind it is a formula, not a norm. Second, is the score reported with its uncertainty? Third, does the page say which abilities the items sample? Fourth, are the tasks protected from circulation, so that the score reflects the person rather than rehearsal?

That fourth point answers the search for "math iq test with answers". Real normed items are not published with their answers, and they should not be. The Standards for Educational and Psychological Testing (AERA, APA and NCME, 2014), which are open access on the Standards website, state in Standard 6.7 that test users have the responsibility of protecting the security of test materials at all times. A published item and key can be learned, and a learned item no longer measures what it was written to measure. This page therefore reproduces no item from any test. It describes formats instead, and a published answer sheet for a "math IQ test" is a strong sign that the number it produces is a score on a worksheet.

The same logic explains why a quiz you can retake with the same items inflates over time. The meta-analysis by Scharfen, Peters and Holling in Intelligence, which covered 174 samples from 122 studies, found significant retest effects and no further score gains after the third administration, and it reported that the size of the effect depends on the cognitive ability operation and content, on whether the forms are equivalent, on the retest interval and on age. A free page that lets you try again is measuring how fast you learn that page, and a reader weighing a free start can compare the options on the page about the free IQ test.

None of this makes every free quiz worthless as practice for enjoyment. It makes it unfit for the question most readers actually have, which is where their quantitative ability sits relative to other adults.

3 What Are the Real Options for Taking a Math IQ Test?

There are five real routes, and they differ most in what they norm against and who supervises: a clinician with a full battery, a clinician with an achievement test, an employer numerical test, a free quiz, and a normed online battery. The table sets them side by side. Prices and times are dated, taken from the pages named, and they can change.

The Pearson WAIS-5 product page, read October 28, 2026, gives the age range, the 45 and 60 minute administration times and the kit prices, and it lists a new quantitative reasoning index among the indexes. The page on WAIS-5 subtests explains which of its subtests build the Full Scale IQ. The University of Warwick careers page on selection tests, read October 28, 2026, says numerical tests are timed, usually 20 to 30 minutes, built from tables and graphs, word problems and number series, and normally set at GCSE level mathematics. It describes them as measuring numeracy aptitude, which is a narrower claim than intelligence.

Three points follow. A school grade or a classroom quiz measures mastery of a curriculum and is a different object again. A clinician is the route when a result must be used for a decision about you, such as an educational plan or a clinical question, and the page on where to take an IQ test lists every route with its trade offs. An employer test is not a substitute for either, because the employer owns the result and it was built to rank applicants for a role. A normed online battery is the route for a reader who wants an informed estimate of where quantitative ability sits, at a price and effort a clinic visit does not match, with the limits in the last section of this page. The page on the best online IQ tests compares the online options on the same axes.

4 How Do Clinical Batteries Test Quantitative Ability?

Clinical batteries sample quantitative ability in two different places, as a reasoning task inside the cognitive battery and as an achievement test beside it, and the placement tells you which ability is being scored. The Wechsler tradition has long carried a spoken arithmetic task. Pearson's WAIS-5 page lists Arithmetic in the kit and describes the fifth edition as providing ten secondary subtests beyond the primary set, with a new quantitative reasoning index among its features. The page does not describe the Arithmetic task further, so this page does not either, beyond saying that the ACIS Arithmetic subtest is a spoken mental calculation task, as its own page and the technical manual describe it.

The Woodcock-Johnson V divides the work differently. In the brochure read on October 28, 2026, the cognitive tests include Number Series tagged Gq and Gf, and the achievement tests include Calculation, Applied Problems and Math Facts Fluency, with Math Calculation Skills, Math Problem Solving and Brief Math clusters tagged Gq. That arrangement matters for interpretation. A person can post a strong Number Series result with a weak Calculation result and the pattern is informative, not contradictory: it says that rule finding over quantities is stronger than recall of conventional procedure.

The practical consequence is that "math IQ" in a clinician's report is usually a statement about one cluster, not about general ability. The cluster is read beside the rest of the profile. The page on cognitive domains explains how a battery arranges domains, and the page on the general ability index explains a composite that leaves out working memory and speed, which helps when a profile is uneven.

ACIS follows the same split. Its two quantitative subtests form the Quantitative Reasoning Index, and its technical manual lists a composite called CQRI that combines Mathematical Achievement, Figure Weights, Arithmetic and Visual Number Series, the four tasks where quantity does the work whether the demand is knowledge or reasoning. A CQRI is available when its four subtests are available. The point of the composite is the same as the point of the clinical clusters: to read learned mathematics and quantitative reasoning together without pretending they are the same ability.

5 What Formats Does a Math IQ Test Use?

A math IQ test uses three formats, and each loads a different mix of abilities: spoken mental arithmetic, written math problems, and quantitative fluid tasks that look like puzzles. Describing the formats is as far as this page goes, because showing an item would defeat the test.

Spoken mental arithmetic

The first format presents a word problem aloud and asks for a mental answer. The Arithmetic subtest page describes it as mental calculation under structured conditions, partly number knowledge and partly holding information in mind long enough to transform it. The technical manual adds that the examinee hears the problem, that no written copy remains on screen, and that the terms must therefore be held while the answer is computed, so the format adds a retention demand to the arithmetic itself. It lists 26 audio problems, 30 seconds to respond to each, and an end after three consecutive errors. By our arithmetic, 26 problems at 30 seconds each means the response time cannot exceed 13 minutes, plus the time the audio itself takes. The task needs working sound, which the practical section returns to.

This format is the one most sensitive to working memory and attention. The manual says Arithmetic is sensitive to attention, anxiety, working memory and calculation fluency, and that a low Arithmetic score with an adequate Mathematical Achievement score may point to the memory and attention load rather than to weak mathematics.

Written math problems

The second format is a set of written problems from school mathematics. The Mathematical Achievement subtest page describes tasks that emphasize learned quantitative knowledge, familiar procedures and school based content. In the technical manual, it is 25 problems in 35 minutes, multiple choice with six options, with a reference table of formulas available throughout so that the score reflects applying procedure rather than recalling a formula, scratch paper permitted, calculators not permitted, and an explicit "I do not know" response so that omissions stay distinguishable from errors. The manual calls it the most achievement oriented quantitative subtest in ACIS and says it must be read as a mixture of cognitive ability and learned academic exposure.

Quantitative fluid tasks

The third format hides the arithmetic inside a reasoning problem. The Figure Weights subtest page describes tasks in which the person must discover the quantitative rule that balances a visual arrangement, and it places the task between fluid reasoning and quantitative reasoning in CHC terms. The Visual Number Series page describes inferring the rule that generates a sequence and projecting it forward, a task whose core demand is induction rather than a memorized procedure. In ACIS, Figure Weights and Visual Number Series sit in the Fluid Reasoning Index with Matrix Reasoning, Logic Grid and Complex Relations, not in the Quantitative Reasoning Index, which is why a person who wants the full quantitative picture should read both indexes.

The sibling pages go deeper on the neighbouring formats: the page on the number series test covers sequences as a test format, the pattern recognition test page covers rule finding across shapes and figures, and the page on the logic IQ test covers deduction. A reader who wants to avoid arithmetic altogether will find the choices on the page on an IQ test without math.

6 Which ACIS Subtests Measure Quantitative Ability, and What Do They Return?

ACIS measures quantitative ability with two subtests in its Quantitative Reasoning Index and two fluid tasks that lean quantitative, and the two index subtests are available in the Full Scale form and the Custom Test only. The battery has 20 subtests in six CHC domains, reported as a Full Scale IQ and six primary indices, and the Quantitative Reasoning Index (QRI) is built from Arithmetic and Mathematical Achievement. Scores use the standard scale for indices (mean 100, standard deviation 15) and scaled scores from 1 to 19 for subtests (mean 10, standard deviation 3), with percentiles and a 95 percent confidence interval. The page on the standard deviation of 15 explains why that number matters.

The forms matter for this reader. Quick (15 dollars, about 45 minutes) and Optimized (30 dollars, about 110 minutes) do not include either quantitative subtest. Full Scale (50 dollars, about 175 minutes) includes both, along with Figure Weights and Visual Number Series. Arithmetic and Mathematical Achievement are two of seven subtests found only in the Full Scale form. The Custom Test costs 5 dollars per subtest, lets you pick any subtests, returns the full index score for any domain you complete and a scaled score for every subtest, and at 10 subtests costs the same as the Full Scale form. Picking Arithmetic and Mathematical Achievement therefore costs 10 dollars, and it returns the QRI and two scaled scores. It does not return a Full Scale IQ, because that needs the Full Scale battery. All prices were read on October 28, 2026 and can change.

Other terms apply to every form: one time payment with no subscription, breaks allowed, progress saved between subtests, and 30 days to complete. The guarantee reads, in the words of the site, "Full refund within 5 days if the report does not deliver the features described at checkout or a technical issue prevents you from accessing your results." The free trial needs no card, and the page on a cognitive ability test describes the broader battery that sits around the quantitative pair.

A buyer who wants the quantitative subtests and a Full Scale IQ in one report should compare the Full Scale form with a Custom selection of the same size, and the answer is the Full Scale form once the selection passes 10 subtests. A buyer who only wants to know whether the quantitative index differs from the rest of the profile can read the example reports before paying: the example Full Scale report shows how an index, its percentile and its confidence interval appear.

7 How Does a Math Score Relate to Your Full Scale IQ?

A math score and a Full Scale IQ are related but far from interchangeable: the best pooled estimate puts fluid reasoning and mathematics at a correlation of .41, which leaves most of the variation in each unshared. The estimate comes from the meta-analysis by Peng, Wang, Wang and Lin in Psychological Bulletin, which pooled 680 studies, 793 independent samples and more than 370,000 participants. It reported a correlation of .38 between fluid intelligence and reading and .41 between fluid intelligence and mathematics, a stronger link for mathematics than for reading, a stronger link for complex skills than for foundational ones, and relations that grew stronger with age. It also reported that fluid intelligence and reading or mathematics predicted each other over development, even after controlling for initial performance, so the direction of influence runs both ways.

A correlation of .41 squares to about .17, so fluid reasoning accounts for roughly 17 percent of the variance in mathematics in that pooled estimate. That squaring is our arithmetic, not a figure the authors report, and it describes groups rather than any one person. The page on math ability and IQ sorts the further studies by design, age and measure, including the exam based cohort evidence, and it is the place to read the full evidence. This page takes the practical reading of it.

For a test taker the reading is as follows. A high quantitative score makes a high Full Scale IQ more likely, and a low one does not rule it out, because the quantitative score mixes reasoning with taught skill, working memory and, as the next section shows, anxiety. Tilts are common. The page on verbal versus quantitative ability reviews the evidence on people who are stronger in one than the other, and the page on index score differences shows how often the six indices of a single adult differ by a large gap, which is normal and not a defect of the test.

Where does the quantitative index sit in the ACIS Full Scale IQ? The Full Scale IQ is built from all 20 subtests in the Full Scale form, so Arithmetic and Mathematical Achievement contribute along with the other 18. A Custom selection of the two quantitative subtests returns the QRI and two scaled scores, but no Full Scale IQ. If the question is whether the quantitative index is high or low relative to the rest of a profile, the useful comparison is the QRI beside the Fluid Reasoning Index and the Verbal Comprehension Index in the same report, and the manual's own guidance is that the QRI reading is strongest when Mathematical Achievement and Arithmetic converge. A reader who wants an overall estimate with working memory and speed set aside can read about the composites on the page on the general ability index.

8 Does Math Anxiety Lower a Math IQ Score?

Math anxiety and math performance are reliably linked in the research, with an average correlation of negative .28, but a correlation does not settle which causes which, and it never tells you what happened in one person's session. Barroso, Ganley, McGraw, Geer, Hart and Daucourt published the largest update in Psychological Bulletin. It pooled 747 effect sizes from research conducted between 1992 and 2018 and reported a small to moderate, negative and statistically significant correlation of negative .28 between math anxiety and math achievement. The abstract reports that the relation held across the moderator subgroups except assessments measuring the approximate number system, that it was present in childhood and persisted into adulthood, that it was weaker in Grades 3 to 5 and in postsecondary samples, and that published studies reported significantly stronger correlations than unpublished ones, which the authors read as large negative effects being underreported.

The working memory account explains why anxiety might matter more for some formats than others. Ashcraft and Kirk reported in the Journal of Experimental Psychology: General that people high in math anxiety had smaller working memory spans, especially on a computation based span task, and that when they did mental addition while holding a memory load, the reduced capacity went with slower responses and more errors. The authors describe the effect as a transitory disruption of working memory and suggest that central executive processes are involved. That is the demand Arithmetic makes: a problem is heard, nothing stays on screen, and the terms must be held while the answer is computed.

The ACIS technical manual makes the same connection from the test side. It says Arithmetic is sensitive to attention, anxiety, working memory and calculation fluency, and the page for Mathematical Achievement says a weaker result can reflect gaps in instruction, anxiety, speed, working memory or unfamiliar notation. Read together, those statements suggest a practical comparison: a person with a clearly lower Arithmetic score than Mathematical Achievement score, and also a low score in the working memory domain, should consider the memory and attention route before concluding anything about mathematics. That is an interpretive heuristic drawn from the manual's guidance, not a measured result about any person.

Three cautions apply. The pooled correlation is an average over many samples and measures, and it does not mean anxiety lowers any given reader's score. The studies mostly measure math achievement, not an IQ index. And a test does not diagnose anxiety: a score pattern cannot tell you that you have it. For the broader question the page on anxiety and IQ covers what is known across tests.

9 Do Education and Practice Change a Math Score?

Yes, both do, and they should: quantitative knowledge is the part of the score that instruction builds, so a Gq score is expected to move with schooling, while practice on the same items moves it for a different and less useful reason. Two bodies of evidence apply. The first is on education. Ritchie and Tucker-Drob's meta-analysis in Psychological Science pooled 142 effect sizes from 42 data sets with more than 600,000 participants and reported approximately 1 to 5 IQ points for an additional year of education, drawing on quasi-experimental designs such as school entry age cutoffs and changes in compulsory schooling. Those are IQ points on whole tests, not math subtests. It would be reasonable to expect a subtest that is partly achievement to be at least as sensitive to schooling, but that is our inference, and the study does not report it.

The second is on practice. The Scharfen meta-analysis described above found that retest effects exist, that they differ by ability operation and content, and that they stop growing after the third administration. Seeing the same items twice teaches the items. That is why a published item bank with answer keys is a poor basis for a measurement and why a person who takes a test a second time should expect a higher result that overstates the change in ability.

What does this mean for someone preparing? Reviewing arithmetic procedures, percentages, ratios and algebra before a test of learned mathematics is legitimate, because the construct is learned knowledge, but it changes what the score says: it becomes a statement about how current your mathematics is. Reviewing for a reasoning task such as a number series task is less direct, because the construct is novel rule finding, and the page on the number series test treats that format. Whatever you do, write down the date and conditions of each administration, because a profile read two years later is only comparable if you remember whether you reviewed first.

10 How Do You Read a Math Score?

Read a quantitative result as a scaled score for each subtest and an index for the pair, each with a percentile and a confidence interval, and trust the index more than either subtest because one task carries more error than a combination. ACIS reports subtests as scaled scores from 1 to 19 with a mean of 10 and a standard deviation of 3, and indices on the scale with a mean of 100 and a standard deviation of 15. A scaled score of 13 is one standard deviation above the mean, which on a normal curve is about the 84th percentile, and that conversion is general arithmetic of the normal distribution, not an ACIS statistic. The page on IQ score versus percentile explains why the two cannot be averaged, and the IQ score chart maps scores to ranges. The tools for percentile conversion and for how rare a given IQ is turn an index into a number you can compare.

The confidence interval is the honest part of the report. A score is an estimate with error around it, and the Standards for Educational and Psychological Testing (2014) say in Standard 6.10 that when score information is released, interpretations should describe in simple language what the test covers, what scores represent, the precision or reliability of the scores, and how the scores are intended to be used. Standard 2.3 adds that estimates of reliability and precision should be reported for each total score, subscore or combination of scores that is interpreted, and its comment warns that subscores may have unacceptably low reliability even when total scores are consistent. That is the formal reason to give a single subtest less weight than the index it belongs to. The page on reliability and validity explains the two terms with examples, and the technical manual is where ACIS documents its own.

Three habits help. Compare the interval, not the point: two scores whose intervals overlap are not clearly different. Compare the pair, not one subtest: if Arithmetic and Mathematical Achievement agree, the QRI is well supported, and if they diverge, the manual's guidance on working memory and instruction applies. And compare the profile, not a number in isolation: the quantitative index means most beside the fluid reasoning and verbal indexes, and the page on processing speed tests is a reminder that speed can move a score on any timed task.

A last point on labels. Neither the QRI nor a scaled score is a "math IQ" in the sense of a single number that summarizes mathematical genius. The QRI says how this person performed on two quantitative tasks relative to adults in the norm group, with a stated margin. Nothing in it predicts a career in mathematics, and nothing in it is a verdict on effort or potential.

11 What Should You Set Up for an Unsupervised Math Test?

The quiet, controlled conditions of a clinic are missing online, so the test taker has to supply them: working sound, a quiet room, scratch paper and a pencil, and no calculator. Arithmetic presents its items as audio and needs working sound, so test the speakers or headphones before starting. The manual says Mathematical Achievement permits scratch paper and does not permit calculators, and that a table of formulas stays available. Using a calculator on a task that forbids one would turn a quantitative knowledge score into something else, and the resulting index would describe the calculator.

The APA's Guidelines for Psychological Assessment and Evaluation (2020) say, in Guideline 8, that test selection, scoring and administration should reflect the appropriate normative comparison, situational influences, effort and standardized administration. A clinician controls those. An online test taker has to control them personally. In practice that means taking each subtest in one sitting, silencing notifications, and not stopping in the middle of an Arithmetic problem. ACIS allows breaks and saves progress between subtests, with 30 days to complete, so the sensible plan is to take the audio and timed subtests when rested and to spread the session across days if needed.

Two other conditions bear on this test in particular. The problems are in English, and the manual says verbal and knowledge tasks are affected by English proficiency and educational exposure, which applies to word problems read aloud. The Arithmetic subtest ends after three consecutive errors, so an early lapse in attention is more costly than a late one. The second condition is mood and state. A test taken after a bad night, or in a state of strong worry about math, will show it, and the section on anxiety above is the reason to wait a day if you notice yourself unsettled.

These are habits, not guarantees. No home setup replicates supervised administration, and no rule on this page changes that. Where a result will be used for a decision about you, a clinician is the route, as the page on the IQ test for adults explains when it compares the ways an adult can get a score.

12 Where ACIS Sits, and What a Buyer Should Do

ACIS is a sensible way to see where your quantitative ability sits relative to adult norms and to the rest of your profile, and it is the wrong tool for any decision that needs a supervised, clinical or admissions grade score. ACIS sells these tests, so treat this section as a disclosure and check it against the rest of the page. Its strengths are specific: a normed index with a 95 percent confidence interval, two quantitative subtests that separate a spoken, working memory loaded task from a written achievement task, a Custom option that costs 10 dollars for exactly those two, and a published technical manual.

Its limits are equally specific. It is online and unsupervised. It is not a clinical or diagnostic instrument, and it cannot diagnose dyscalculia, anxiety or any other condition. It is not for hiring, for school accommodations or for admission to high IQ societies. It is in English only. Its quantitative subtests are in the Full Scale form and the Custom Test only, and they do not return a Full Scale IQ unless the Full Scale battery is completed. And the Mathematical Achievement subtest is, by the manual's own words, a mixture of cognitive ability and learned academic exposure.

The Standards for Educational and Psychological Testing (AERA, APA and NCME, 2014) are the benchmark for how a score should be read. Standard 12.10 says that in educational settings a decision with major impact on a student should consider not just scores from a single test but other relevant information, and the APA's 2020 guidelines say in Guideline 7 that individual performance on tests is one piece of an assessment placed in context. A single online math result, however well normed, should be treated the same way: as one input.

What should a buyer do? If the question is curiosity or self knowledge about quantitative ability, pick Arithmetic and Mathematical Achievement in the Custom Test for 10 dollars, take them with working sound and no calculator, and read the QRI with its interval beside the scaled scores. If the question is where your whole profile sits, the Full Scale form at 50 dollars gives all six indices and a Full Scale IQ, and the free trial, which needs no card, shows how the reports read before you pay. If the result will be used for a decision about you, book a clinician and take the table in this page to the appointment. For readers hoping for a number above the usual range, the page on high range IQ tests covers scores above the usual range, and the page on a gifted test for adults covers confirming where an adult stands.

13 Sources Behind This Page

Every figure above is traceable to one of the following, and each is linked at the point where it is used. The squared correlation of about .17, the 13 minute ceiling on Arithmetic response time and the 84th percentile for a scaled score of 13 are our own arithmetic on published figures and are labelled as such in the text.

• Peng P, Wang T, Wang C, Lin X. A meta-analysis on the relation between fluid intelligence and reading/mathematics: Effects of tasks, age, and social economics status. Psychological Bulletin, 2019, volume 145, issue 2, pages 189 to 236.
• Barroso C, Ganley C, McGraw A, Geer E, Hart S, Daucourt M. A meta-analysis of the relation between math anxiety and math achievement. Psychological Bulletin, 2021, volume 147, issue 2, pages 134 to 168.
• Ashcraft M, Kirk E. The relationships among working memory, math anxiety, and performance. Journal of Experimental Psychology: General, 2001, volume 130, issue 2, pages 224 to 237.
• Scharfen J, Peters J, Holling H. Retest effects in cognitive ability tests: A meta-analysis. Intelligence, 2018, volume 67, pages 44 to 66.
• Ritchie S, Tucker-Drob E. How much does education improve intelligence? A meta-analysis. Psychological Science, 2018, volume 29, issue 8, pages 1358 to 1369.
• McGrew K. CHC theory and the human cognitive abilities project: Standing on the shoulders of the giants of psychometric intelligence research. Intelligence, 2009, volume 37, issue 1, pages 1 to 10.
• American Educational Research Association, American Psychological Association, National Council on Measurement in Education. Standards for Educational and Psychological Testing, 2014, Standards 2.3, 6.7, 6.10 and 12.10.
• American Psychological Association. APA Guidelines for Psychological Assessment and Evaluation, approved by the APA Council of Representatives, March 2020, Guidelines 7 and 8.
• Pearson Assessments. Wechsler Adult Intelligence Scale, Fifth Edition product page, read October 28, 2026: age range, administration times, kit prices and indexes.
• Riverside Insights. Woodcock-Johnson V brochure, read October 28, 2026: test names and CHC ability tags.
• University of Warwick Careers. Selection tests, read October 28, 2026: format, timing and level of employer numerical tests.

14 Frequently Asked Questions