The General Aptitude Test Battery explained: nine aptitudes, one review, one law

0 The short answer

The General Aptitude Test Battery (GATB) is a multiple aptitude battery that the U.S. Employment Service built in the 1940s, with 12 separately timed tests that combine into nine aptitude scores, and the O*NET Ability Profiler that grew out of it was retired on November 20, 2021. The 1989 National Research Council review judged the GATB adequate in psychometric quality apart from two serious flaws, weak test security and severe speededness, and the Civil Rights Act of 1991 later made it unlawful, in selecting or referring applicants for employment or promotion, to adjust employment test scores on the basis of race, color, religion, sex or national origin. None of the pages we opened offers a self administered online GATB, so an adult who wants to see a profile of broad abilities has to look elsewhere: ACIS is one concrete option, at 15 dollars for Quick (about 45 minutes), 30 dollars for Optimized (about 110 minutes) and 50 dollars for Full Scale (about 175 minutes), one time, prices read October 29, 2026. The decisive limit is that ACIS is not the GATB and has no motor or dexterity tasks, so it cannot show what the battery's apparatus tests measured.

1 What Was the General Aptitude Test Battery, and Who Used It?

The GATB was a federal employment test: the U.S. Employment Service decided between 1942 and 1945 to build a general battery that could screen for many occupations, and it became operational in 1947. The account comes from the 1989 report of the Committee on the General Aptitude Test Battery of the National Research Council, which John A. Hartigan chaired and Alexandra K. Wigdor directed as study director, and the two edited the volume. The report is titled Fairness in Employment Testing: Validity Generalization, Minority Issues, and the General Aptitude Test Battery and is open on the website of the National Academies. Its fourth chapter, which we read on October 29, 2026, says the Employment Service drew on about 100 occupation specific tests developed since 1934 and identified a small number of basic aptitudes that appeared relevant to many jobs. From its inventory of job specific tests it selected the ones that best measured each of nine aptitudes, and the result became the operational GATB in 1947, edition B-1002, in two forms, A and B. Form A was reserved for Employment Service offices, while Form B was used for validation research and for retesting and was available to other authorized users. Two more forms, C and D, arrived in 1983.

Who sat it matters for how the scores should be read. The report's Summary says the GATB was used for vocational counseling and job referral by state administered Employment Service offices, also called Job Service offices, in some 1,800 communities. Chapter 4 adds that it was also made available to high schools, technical schools, labor union apprenticeship programs and public and private vocational rehabilitation services, and that it remained virtually unchanged for most of more than 40 years. In plain terms, counseling is a person looking at their own results, while referral, as the Summary describes the proposed system, is an office ranking registrants for an employer's job order, and the 1989 review concerned the second function.

The page on aptitude versus intelligence explains how an aptitude test differs in purpose from an intelligence test, and the page on the history of IQ testing places the battery in the longer line from Binet to Wechsler.

This page covers only the GATB and the tools derived from it, not other tests that share a similar name.

2 What Are the Nine GATB Aptitudes?

The GATB reports nine aptitudes, each with a letter code: general learning ability (G), verbal (V), numerical (N), spatial (S), form perception (P), clerical perception (Q), motor coordination (K), finger dexterity (F) and manual dexterity (M). The National Research Council report lists them with the Employment Service's own definitions, and the Japanese edition published by the Japan Institute for Labour Policy and Training lists the same nine. The table below gives the aptitude, the committee's definition in condensed form, and the tests that feed it according to the committee's description and the O*NET Ability Profiler User's Guide, which we opened on October 29, 2026.

Two details in that table deserve attention. First, the committee notes that four of the nine aptitudes, clerical perception, motor coordination, finger dexterity and manual dexterity, involve speed of work as a major component. Nearly half of the battery's profile, then, rests on how fast a person works as well as on how accurately, which is one reason a GATB profile cannot be read as a profile of reasoning. Second, general learning ability is not a separate test. It is a composite built from three of the others, the vocabulary, the arithmetic reasoning and the three dimensional space tests, and it therefore overlaps with the verbal, numerical and spatial scores by construction.

These labels are the Employment Service's, not those of the Cattell-Horn-Carroll model, which the page on the CHC model explains, and a later section maps them onto broad abilities.

3 How Was the GATB Built, Timed and Scored?

The GATB consists of 12 separately timed tests, eight on paper and four on apparatus, each scored as the number answered correctly with no correction for guessing, and several of them carry time limits so severe that the committee called out the battery's speededness as one of its two serious flaws. The committee's description of the structure, in its fourth chapter, names the 12 tests. The paper tests are Name Comparison, Computation, Three-Dimensional Space, Vocabulary, Tool Matching, Arithmetic Reasoning, Form Matching and Mark Making. The apparatus tests are Place, Turn, Assemble and Disassemble. Two of the paper tests, Name Comparison and Mark Making, and all four apparatus tests are intended to measure the aptitudes that involve speed of work as a major component.

The formats can be described without reproducing anything. Name Comparison asks whether each pair in two long columns of names is the same or different, 150 names in 6 minutes. Three-Dimensional Space asks which of four drawings of solid objects could be made from a flat piece of metal that is bent or rolled, 40 items in 6 minutes. Vocabulary asks which two of four words have the same or opposite meanings, 60 items in 6 minutes. Mark Making asks for three pencil marks in each of 130 small boxes in 60 seconds. The apparatus tests use a pegboard, on which pegs are moved and then turned over, and a second board on which small parts are assembled and taken apart. This page shows no item from any of them.

The time limits explain the criticism. The committee's Summary says that many subtests have such severe time limits that an average applicant can expect to complete only one third of the test, and that such tests are eminently coachable: test takers can learn strategies that raise their performance, for example by randomly filling in the remaining blanks in the last minute. When a test is built so that nobody can finish, the score partly measures how fast a person works and how well they use the clock. The Standards for Educational and Psychological Testing make a related technical point in the introduction to their chapter on reliability and precision. They state that when a test is designed to reflect rate of work, internal consistency estimates of reliability, particularly by the odd even method, are likely to yield inflated estimates for highly speeded tests.

Each test yields a raw number correct, and the raw scores are converted and combined into the nine aptitude scores. The O*NET Ability Profiler User's Guide states that the aptitude scores are scaled to a mean of 100 and a standard deviation of 20 in the GATB norming population. That is not the scale used by most intelligence tests, which the page on the 15 point standard deviation explains, so a GATB aptitude score of 120 cannot be read directly as an IQ of 120. If both scales describe a normal curve, a score one standard deviation above the mean is 120 on the GATB scale and 115 on the intelligence scale, which is our arithmetic and which assumes the norm groups are comparable. The page on how IQ scores are normed explains why that assumption is the hard part.

4 Why Did the National Research Council Review the GATB in 1989?

The Department of Labor sought the advice of the National Academy of Sciences because the Employment Service was promoting the GATB as the basis for referring job seekers to all jobs in the U.S. economy, and the Department of Justice had challenged one element of that plan, the use of within group percentile scores. The committee's Summary describes the system, which it calls the VG-GATB Referral System, where "VG" stands for validity generalization, the idea of extending the validities found for some 500 jobs to all other jobs in the U.S. economy. As conceived by Employment Service staff, virtually all registrants would be tested, virtually all job orders would be filled on the basis of GATB scores, and candidates would be referred in rank order of those scores. GATB scores would be computed as percentile scores within each of three racial or ethnic groups: Black, Hispanic and other.

The Summary says the Department of Labor, faced with that legal challenge, sought the advice of the National Academy of Sciences, and a National Research Council committee was asked to study within group scoring and to evaluate validity generalization and its application to the GATB. The department asked whether the pivotal role envisioned for the GATB was technically justified, whether the anticipated economic benefits were realistic, and what the effects might be on veterans, people with handicapping conditions, employers and job seekers.

The report was not a verdict on employment testing in general. It reviewed one proposal, to make a single federal battery the gateway for referral across the economy, and it framed the tension openly. The Summary acknowledges that rank ordered scores would screen out a large proportion of Black and Hispanic candidates and expose employers to legal action under civil rights law, while score adjustments to mitigate that effect create vulnerabilities to charges of reverse discrimination. It states that the committee could not make definitive statements about the legality of race conscious scoring methods, and it cautions that data from intelligence testing were misused in the early twentieth century in ways that fed racial and ethnic prejudices.

The page on whether IQ tests are biased covers the measurement side of fairness in cognitive tests in general; this page keeps to what the committee and the statute say about this battery.

5 What Did the 1989 Review Find About Validity, Security and Speed?

The committee concluded that the GATB was adequate in psychometric quality except for two serious flaws, weak test security and severe speededness, and that its validities for supervisor ratings were modest, in the range of .2 to .4, so it should supplement other referral methods and not replace them. The Summary reports that the GATB was supported by some 750 criterion related validity studies covering about 500 jobs, which relate scores to a measure of job performance, typically supervisor ratings of job incumbents and in some cases grades in a training course.

After correction for sampling error, the correlations of GATB based predictors with supervisor ratings were in the range of .2 to .4. The average validity of aptitude composites, corrected for criterion unreliability, was about .25 in studies conducted since 1972 and about .35 in the older studies, against values of .5 or higher in the Employment Service's technical reports. The committee said it had more data, that the more recent studies produced noticeably lower validities, and that it did not accept the magnitude of the corrections for criterion unreliability and range restriction that the technical reports applied.

That dispute is the kind of issue the Standards for Educational and Psychological Testing (AERA, APA and NCME, 2014), which we read in their open access edition, address in the chapter on workplace testing. Standard 11.8 says that people who conduct and interpret predictor criterion studies should identify artifacts such as errors of measurement and range restriction, and its comment says that when corrections are applied both corrected and uncorrected values should be presented along with the rationale. Standard 11.5 says that decisions about local studies should be grounded in knowledge of relevant research, and its comment recognizes that a large, consistent cumulative literature can support a predictor criterion relationship. The 1989 committee accepted the general thesis of validity generalization, that results of validity studies can be generalized to many jobs not studied, but urged generalizing only to appropriately similar jobs, and it saw no evidence that the battery was a valid predictor for all 12,000 jobs in the economy. The page on IQ and job performance covers the later meta-analytic literature.

A validity of that size means scores carry some information about average performance in a group of workers while leaving much of the variation in performance unpredicted. The committee judged that GATB scores can provide useful screening information but are not strong enough for the GATB to become the sole means of filling all job orders, that its best use is to supplement current methods, and that forcing people to take it as a condition of receiving job placement services serves no one's best interests.

Weak security followed from only two current alternate forms and from administration under a variety of protocols by a variety of organizations, so the committee expected the forms to become available outside government channels once doing well on the test mattered for getting a job. Speededness followed from the severe time limits described above, which make the test coachable. The committee found the GATB neither markedly superior nor inferior to other batteries, such as the Armed Services Vocational Aptitude Battery (ASVAB), in predictive validity and reliability, but it said the GATB does not compare well with the ASVAB in test security, the production of new forms, the strength of its normative data and the severe time limits.

The committee also called the often repeated claim that the Employment Service's use of the GATB would produce a gain of 79.36 billion dollars unfounded, because it rested on overestimates of validities, of the variability of worker productivity and of the selectivity of employers.

6 What Was Within Group Score Adjustment, and What Does the Civil Rights Act of 1991 Say?

Within group score adjustment meant converting each applicant's GATB score to a percentile computed only among members of the applicant's own racial or ethnic group, and section 106 of the Civil Rights Act of 1991 now makes it unlawful, in selecting or referring applicants for employment or promotion, to adjust employment test scores on the basis of race, color, religion, sex or national origin. The facts below come from the 1989 report and from the text of the statute on the website of the Equal Employment Opportunity Commission, which we opened on October 29, 2026. They are history and law, not a statement about any person who sits a test.

The mechanism, as the committee's Summary describes it, was this. The Employment Service computed GATB scores as percentile scores within each of three groups, Black, Hispanic and other, and referred candidates to an employer from the highest adjusted score downward. The Summary says the purpose of these score adjustments, which serve to erase group differences in test scores, was to mitigate the adverse effects that rank ordering on test scores would otherwise have on the employment opportunities of minority job seekers. The Summary notes that the Department of Labor adjusted the scores of Black, Hispanic and in some cases Native American applicants.

The committee's reasoning is set down below as the Summary states it. It asked whether false-acceptance and false-rejection rates differ between majority and minority groups of applicants. Its analysis found that, because prediction is imperfect and because there are average group differences in test scores, selecting by rank order of scores without adjustment screens out in greater proportions the minority applicants who could have performed successfully, while including in greater proportions the majority applicants who turn out not to perform successfully. The Summary adds that these effects are a function of high and low test scores, not of racial or ethnic identity. On that basis the committee judged that the disproportionate impact of selection error provided scientific grounds for adjusting minority scores, and it recommended the continued use of score adjustments for Black and Hispanic applicants in choosing whom to refer. It analyzed two adjustment methods, within group percentile scores and a performance based method, and said both were race conscious and would virtually eliminate the adverse impact of the GATB at current validity levels. It also stated that others would have to decide whether those scientific reasons were compelling in the realms of public policy and law. These statements describe how a selection rule performs for groups of applicants. They do not describe any person's ability, and this page makes no claim about the reasons for average group differences in test scores.

Section 106 of the Civil Rights Act of 1991, titled "Prohibition Against Discriminatory Use of Test Scores", amended section 703 of the Civil Rights Act of 1964 (42 U.S.C. 2000e-2) by adding a new subsection (l). The text in the original act reads: "It shall be an unlawful employment practice for a respondent, in connection with the selection or referral of applicants or candidates for employment or promotion, to adjust the scores of, use different cutoff scores for, or otherwise alter the results of, employment related tests on the basis of race, color, religion, sex, or national origin." The Legal Information Institute, which we also opened on October 29, 2026, prints the same subsection in the United States Code and records that Public Law 102-166, section 106, added it, with the act dated November 21, 1991. The text does not name the GATB. It applies to employment related tests in general and to three acts, adjusting scores, using different cutoff scores and otherwise altering results, when done on the listed bases.

The sources we opened do not state why Congress chose that wording, so this page does not claim that the review caused the law. Side by side, the record is that in 1989 a committee described within group percentile scoring and recommended continued adjustment, and in 1991 the statute made adjusting employment test scores on those bases unlawful for the parties it covers. This page is not legal advice and does not interpret how courts have applied the provision.

7 What Came After the GATB? The O*NET Ability Profiler

The O*NET Ability Profiler, a career exploration tool of the U.S. Department of Labor, was developed from the GATB through a research project to improve its technical and operational qualities, and it was retired on November 20, 2021. The sentence that links the two is in the Ability Profiler User's Guide, a Department of Labor document that we opened as a PDF on October 29, 2026. It says the Ability Profiler was developed as the result of a research effort to improve the GATB, which had been used nationwide since 1947 by state employment service offices for occupational counseling and applicant referral, and that new versions of the GATB were developed periodically to guard against overexposure, to ensure test security and to allow retesting.

The User's Guide lists what changed, and several items on the list match the problems the 1989 review named. Its primary improvements were reducing test speededness and susceptibility to coaching, revising scoring procedures, developing items free from bias, assembling tests as parallel to each other as possible, improving the aesthetics of the tests and revising answer sheets and other materials. It says other specific objectives were incorporated based on recommendations by the National Academy of Sciences. The Form Matching test was deleted, which reduced total testing time without reducing validity according to the guide, and Tool Matching was renamed Object Matching.

The structure that remained was smaller: 11 separately timed exercises, six on paper (Arithmetic Reasoning, Vocabulary, Three-Dimensional Space, Computation, Name Comparison and Object Matching) and five psychomotor (Mark Making, Place, Turn, Assemble and Disassemble), the latter optional according to the guide. It lists nine abilities: Verbal Ability, Arithmetic Reasoning, Computation, Spatial Ability, Form Perception, Clerical Perception, Motor Coordination, Manual Dexterity and Finger Dexterity. That list splits numerical aptitude in two and has no separate score for general learning ability, a difference the pages we read do not explain.

The purpose changed along with the design. The guide states that the changes to the score report and the interpretation reflect a focus on comparing an individual's profile of abilities with the profiles of occupations, rather than using cutoff scores. The report compares the relative shape of an examinee's results, as a bar graph, with the typical ability profile for O*NET occupations, and the Department's user agreement authorizes the tools for career exploration, career planning and career counseling purposes only. That is a different use from the referral system the 1989 committee reviewed. The guide also gives the conditions for taking it: examinees must be at least 16 years old, must be able to read English at the sixth grade level or higher, and should wait six months before retaking it.

The retirement is documented on the program's own pages. The Ability Profiler page says that on November 20, 2021, the O*NET Ability Profiler was retired and that technical assistance is no longer available. The archived materials page says the materials previously distributed for administration, scoring and training are provided as is and should be used for research purposes only, and the career exploration tools page lists the Ability Profiler as retired with materials archived, alongside several other retired tools. The pages we opened do not describe a replacement for the Ability Profiler, so this page does not name one.

8 Can an Adult Take the GATB Today, and What About Practice Questions?

No page we opened offers the GATB as a self administered online test: the U.S. tool developed from it is retired, the Japanese edition is an administered battery with apparatus, and an adult who is asked by an employer, agency or program to take a GATB should ask that organization which edition it uses. Three pieces of evidence define the landscape. The first is the U.S. one, above: the Ability Profiler is retired and archived. The second is Japan. The Japan Institute for Labour Policy and Training publishes a page, in Japanese, on the Ministry of Health, Labour and Welfare edition of the General Aptitude Test Battery, which we opened on October 29, 2026. It says the tool measures the same nine aptitudes, with 11 paper and pencil tests and four apparatus tests, for people from junior high school age to about 45 years, in 45 to 50 minutes for the paper tests and 12 to 15 minutes for the apparatus tests, answered as fast and accurately as possible within the time limit. It also says the institute does not sell the tool, which is provided through the Employment Problems Research Association. The page gives no revision date, price or norm description, so none appears here.

The third is Canada. A 2025 article by Uttl, Sikma and Tat in Psychological Injury and Law, which we read in its published abstract through its DOI, describes the GATB Canadian Edition as derived from the United States Employment Service GATB but normed on a Canadian general working population sample. The authors report that university students tested on that edition some 40 years after it was normed scored on average about 1 standard deviation, the equivalent of 15 IQ points, below the 1985 norms, while the same students averaged 103 on the Shipley-2, which was normed in 2008 on a US sample, and they conclude that the Canadian norms have become outdated. We read only the abstract, and it is one study of students, so it is a warning about the age of norms and not a measurement of any group. The American Psychological Association's Guidelines for Psychological Assessment and Evaluation (2020), which we opened on October 29, 2026, make the general point in the rationale for Guideline 8, which says that the agedness of the norms and the continued relevance and definition of the constructs being measured can be particularly important considerations in test selection. We did not open a provider's page, so we do not say who offers the Canadian edition.

Many websites sell GATB practice material, and we neither link them nor treat them as sources, because they are sold by third parties and any resemblance to the real battery cannot be checked from this page. What research can say is how retesting works. A meta-analysis by Hausknecht, Halpert, Di Paolo and Moriarty Gerrard in the Journal of Applied Psychology (2007), which we read through its PubMed abstract, covered 50 studies, 107 samples and 134,436 participants of practice effects on cognitive ability tests. It reported an adjusted overall effect size of .26, with larger effects when practice came with coaching and when identical forms were used. For a speeded battery, where the 1989 committee thought coaching strategies could raise scores, familiarity with the format and the clock can matter, and a retest score is not a clean second measurement. The Ability Profiler guide's six month wait before retaking is consistent with that concern, although the guide does not say it rests on retest research.

An adult who wants a profile without a GATB can compare supervised and online routes on the page about where to take an IQ test, see what a normed battery returns on the page about IQ tests for adults, and read about employers' timed tests from the candidate's side on the page about IQ tests for job interviews. The sibling pages on the mechanical reasoning test and the error checking test cover two more employer formats, the second a speeded checking format of the family to which Name Comparison belongs.

9 How Do the Nine Aptitudes Line Up With Broad Abilities, and What Can ACIS Show?

Four of the nine GATB aptitudes, verbal, numerical, spatial and clerical perception, have a recognizable counterpart among the broad abilities of the Cattell-Horn-Carroll model and a related ACIS index, but ACIS has no task for form perception, motor coordination, finger dexterity or manual dexterity, and general learning ability has no single counterpart. The map below is our reading of the committee's and the Employment Service's definitions against the CHC names used on the page about the domains of cognition. It shows what overlaps and is not a claim that a GATB score and an ACIS index are interchangeable, because the tasks, time limits, norm groups and scales all differ.

Take the four counterpart rows in turn. For verbal ability, the page on verbal IQ tests explains what a verbal score measures, and the ACIS Verbal Comprehension Index covers five subtests, more breadth than the GATB's single Vocabulary test of word relations. For numerical ability, the page on quantitative reasoning explains why the name covers two abilities, and the two subtests behind the Quantitative Reasoning Index, Mathematical Achievement and Arithmetic, are only in the Full Scale form; Arithmetic is presented as audio and needs working sound. The page on verbal versus quantitative ability covers what a gap between the two means, which matters because the GATB reported them separately.

For spatial ability, the page on the spatial IQ test describes the three ACIS spatial subtests and what the index returns. The Quick form has no spatial subtest, and the Optimized form includes Spatial Comprehension and Layer Rotation. The sibling page on the spatial reasoning test is the place to read about that name as an aptitude test format.

For clerical perception the overlap is partial. Name Comparison is a perceptual speed task with clerical content. The page on processing speed tests explains the ability behind speed measures, and the ACIS subtests Coding and Symbol Search sample speeded visual work, with Symbol Search described on its page as rapid visual comparison and decision speed. Neither is a name checking task, and the Coding page notes that results can reflect motor speed, visual tracking and attention. That is the nearest ACIS comes to the GATB's speed aptitudes, and it is not a motor or dexterity test.

10 What Can a Nine Score Aptitude Profile Still Tell an Adult?

A profile of separate aptitude scores can show where a person is relatively stronger, but one study of the GATB database found that the difference between two abilities added almost nothing to predicting job performance beyond ability level, so the profile is better used to explore than to decide. The most direct recent evidence uses the GATB database. Kato and Scherbaum, in the Journal of Intelligence (2023), examined ability tilt, a measure of the differential strength between two specific abilities, in a large sample from the GATB database, and the abstract, which we opened on October 29, 2026, reports that tilt related to job performance in the expected direction for 27 of 36 tilt and job combinations, with a mean effect size of .04 when the tilt matched job requirements. The mean incremental validities for tilt were .007 over g and .003 over g and specific abilities. The authors describe the results as limited evidence that ability tilt may be a useful predictor in addition to ability level.

Read that way, the finding is consistent with the 1989 committee's cautions. The GATB had modest validities as a whole, and in the one study we read the tilt between two abilities added little at the level of a population. That does not mean a person's strengths are irrelevant to the work they would enjoy or find easy. It means that, on this evidence, a score difference between two aptitudes is a weak basis for a prediction about performance. The page on IQ tests for career guidance covers how an adult can use ability information when exploring careers, and the page on average IQ by profession covers the occupational averages that people often ask about, with the selection effects that make them hard to use.

Three habits keep a profile honest. Each of the nine scores carries error, and small gaps between two scores are the least stable information in it, as the page on reliability and validity explains. A percentile compares a person with a norm group, and the page on IQ score versus percentile explains why percentiles and scores cannot be averaged or swapped. And the age of the norms matters, which the Canadian edition study illustrates and Guideline 8 of the APA guidelines names as a consideration.

11 Which Real Options Give an Adult an Aptitude Profile, and How Do They Compare?

No option listed here is an online GATB: the Japanese edition and the WAIS-5 are administered batteries, the O*NET Ability Profiler is retired, and ACIS is a different, online battery that costs 15 to 50 dollars. The table uses only the pages we opened on October 29, 2026, and where a page is silent, it says so.

A practical comparison follows from the prices. A profile that mirrored the four cognitive aptitudes with a counterpart would use all five Verbal Comprehension subtests, both Quantitative Reasoning subtests, all three Visual Spatial subtests and both Processing Speed subtests: 12 subtests, which would cost 60 dollars through the Custom Test at 5 dollars each (our arithmetic) against 50 dollars for the Full Scale form with all 20. At 10 subtests the Full Scale form costs the same as the Custom Test, so it is the better value for anyone who would choose 10 or more. The page on cognitive ability tests compares the options by purpose.

12 Where ACIS Sits, and What a Buyer Should Do

ACIS is an online, unsupervised, non clinical, English only battery that gives an adult a profile of broad abilities, and it is not the GATB, is not a substitute for it and has no motor or dexterity tasks. On the page it publishes for researchers, ACIS for Research, the site says that ACIS "is not intended for clinical diagnosis, hiring decisions, accommodation determinations or admission to high IQ societies". That sentence describes the right use of an ACIS report: a way to see your own profile of broad abilities, read for yourself. The documentation is on the ACIS technical manual, which says that verbal and knowledge tasks are affected by English proficiency and educational exposure.

What the Standards for Educational and Psychological Testing (AERA, APA and NCME, 2014) say about workplace testing explains why the separation matters. Standard 11.1 says that before an employment test is developed and implemented, a clear statement of the intended interpretations of scores for specified uses should be made, with validation designed to determine how well that has been achieved for all relevant subgroups. Standard 11.4 says that when multiple test scores or test scores and nontest information are integrated to make a decision, the role played by each should be clearly explicated and the inference made from each source of information should be supported by validity evidence. That is why the limit quoted above matters. The American Psychological Association's Guidelines for Psychological Assessment and Evaluation (2020) add, in Guideline 6, that psychologists should choose measures with sufficient validity evidence for their uses and fair and appropriate for the purpose, population, setting and context, and, in Guideline 8, that the age of the norms matters. An adult can apply the same questions as a buyer: what is the score for, who is it compared with, and how old is that comparison.

The strengths are specific. ACIS reports a Full Scale IQ and six primary indices on the standard scale with a mean of 100 and a standard deviation of 15, scaled subtest scores from 1 to 19, percentiles and a 95 percent confidence interval, with adult norms covering ages 16 to 90. Breaks are allowed, progress is saved between subtests, and a form can be completed within 30 days. Digit Span, Alphanumeric Sequencing and Arithmetic present their items as audio and need working sound. The purchase is one time, and the guarantee reads: "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 limits are as plain: it is unsupervised, English only, and does not measure what the GATB's apparatus tests measured.

A sensible next step depends on the question. If an organization told you to take a GATB, ask which edition it uses, whether the results go to anyone and how long they stay valid. If you simply want to see your own profile of verbal, numerical, spatial and speed abilities, start the free trial, which needs no card and keeps finished trial subtests if you later unlock a form that includes them, open the example report of the form you are considering, and then choose among Quick, Optimized, Full Scale and the Custom Test using the arithmetic above. If your question is about careers, read the career guidance page first, and keep in mind what the 1989 committee wrote about any single test: it provides only partial information and works best alongside other sources.

13 Sources Behind This Page

Every third party fact and figure on this page comes from the documents below, opened or verified on October 29, 2026, and any figure we calculated ourselves is labeled in the text as our arithmetic. Where we read only an abstract, the entry says so. The O*NET and Pearson pages were read as published that day, and prices and policies on commercial pages change, so check the page before you pay.

1. Hartigan, J. A., and Wigdor, A. K., editors; National Research Council, Committee on the General Aptitude Test Battery (1989). Fairness in Employment Testing: Validity Generalization, Minority Issues, and the General Aptitude Test Battery. Washington, DC: National Academy Press. John A. Hartigan, chair; Alexandra K. Wigdor, study director. Summary (pages 1 to 16) and Chapter 4 (pages 73 to 80) read on October 29, 2026. DOI 10.17226/1338; Summary at https://www.nationalacademies.org/read/1338/chapter/2; Chapter 4 at https://www.nationalacademies.org/read/1338/chapter/6
2. U.S. Equal Employment Opportunity Commission. Civil Rights Act of 1991 (Original Text), Public Law 102-166, section 106, "Prohibition Against Discriminatory Use of Test Scores". Read on October 29, 2026. https://www.eeoc.gov/civil-rights-act-1991-original-text
3. Legal Information Institute, Cornell Law School. 42 U.S. Code section 2000e-2, subsection (l), with the amendment note for Public Law 102-166, section 106. Read on October 29, 2026. https://www.law.cornell.edu/uscode/text/42/2000e-2
4. National Center for O*NET Development, U.S. Department of Labor, Employment and Training Administration. O*NET Ability Profiler User's Guide, PDF. Introduction, exercises, development overview, equating study and administration conditions. Read on October 29, 2026. https://www.onetcenter.org/dl_files/AP_zips/AP-UG-deskp.pdf
5. O*NET Resource Center. O*NET Ability Profiler, retirement notice. Read on October 29, 2026. https://onetcenter.org/AP.html
6. O*NET Resource Center. O*NET Ability Profiler Archived Materials. Read on October 29, 2026. https://www.onetcenter.org/reports/AP_Archive.html
7. O*NET Resource Center. O*NET Career Exploration Tools. Status of the Ability Profiler and other tools. Read on October 29, 2026. https://www.onetcenter.org/tools.html
8. Japan Institute for Labour Policy and Training (JILPT). General Aptitude Test Battery (GATB), Ministry of Health, Labour and Welfare edition, tool page in Japanese. Aptitudes, tests, target group and time. Read on October 29, 2026. https://www.jil.go.jp/institute/seika/tools/GATB.html
9. Uttl, B., Sikma, K., and Tat, M. (2025). General Aptitude Test Battery (GATB): Use of Obsolete Tests, Data, and Beliefs Causes Harm. Psychological Injury and Law, 18(3), 199 to 214. Abstract only, verified through Crossref on October 29, 2026. https://doi.org/10.1007/s12207-025-09535-w
10. Hausknecht, J., Halpert, J., Di Paolo, N., and Moriarty Gerrard, M. (2007). Retesting in selection: A meta-analysis of coaching and practice effects for tests of cognitive ability. Journal of Applied Psychology, 92(2), 373 to 385. Abstract read on PubMed, DOI verified through Crossref on October 29, 2026. https://doi.org/10.1037/0021-9010.92.2.373
11. Kato, A., and Scherbaum, C. (2023). Exploring the relationship between cognitive ability tilt and job performance. Journal of Intelligence, 11(3), 44. Abstract read through Crossref on October 29, 2026. https://doi.org/10.3390/jintelligence11030044
12. American Educational Research Association, American Psychological Association, and National Council on Measurement in Education (2014). Standards for Educational and Psychological Testing. Washington, DC: AERA. Introduction to the chapter on reliability and precision, and Standards 11.1, 11.4, 11.5 and 11.8. Open access edition: https://www.testingstandards.net/open-access-files.html
13. American Psychological Association (2020). APA Guidelines for Psychological Assessment and Evaluation. Guidelines 6 and 8. Read as a PDF on October 29, 2026. https://www.apa.org/about/policy/guidelines-psychological-assessment-evaluation.pdf
14. Pearson Assessments (2026). WAIS-5 (Wechsler Adult Intelligence Scale, Fifth Edition), product page. Ages, administration times, qualification level and index scores. Read on October 29, 2026. https://www.pearsonassessments.com/en-us/Store/Professional-Assessments/Cognition-%26-Neuro/Wechsler-Adult-Intelligence-Scale-%7C-Fifth-Edition/p/P100071002

14 Frequently Asked Questions