The IQ test for non-native English speakers: what your English moves and what it leaves alone

0 The short answer

You can take an IQ test in English as a second language, but the language moves some parts of the score far more than others, so the choice that matters is between a test in your own language, a nonverbal test, and an English test read index by index. Research on cognitive test performance finds that language abilities influence scores, and the professional Standards say a test should be given in the language most relevant to its purpose. ACIS is online, unsupervised and English only, and reports a Culture Fair composite on its Optimized and Full Scale forms and a Nonverbal composite on Full Scale only, but neither makes a score equivalent across languages.

1 Can You Take an IQ Test in English if English Is Not Your First Language?

Yes, nothing prevents you from taking an IQ test in English, and the real question is which parts of the result your English moves and which it leaves alone. An intelligence test is a set of tasks, and the tasks do not depend on language equally. A task that asks you to define a word depends on how many English words you know. A task that asks you to complete a grid of shapes depends on very little English once you understand the instruction. The score you receive combines both kinds of task, and the second kind is easier to defend when English is not your first language.

This page is written for adults who read English comfortably in daily life: students and professionals who study or work in English, expatriates, and people who watch and read in English by choice. It is not written for someone who can follow English only with a dictionary open. If that describes you, the answer is shorter: wait for a test available in your own language, because a test whose instructions you must decode measures the decoding. The reader in between, who reads English well but did not grow up in it, is the one the rest of this page serves.

It helps to separate four ways language can touch a score. The first is instruction comprehension: understanding what the task asks. The second is verbal knowledge, the stock of words and facts that the CHC model calls crystallized ability, abbreviated Gc. A test of vocabulary or general information measures what you have learned in the language of the test, so a second language speaker can know a concept well and still lack the English word for it. The third is written or spoken production: typing a definition in English is a different act from thinking it in another language. The fourth is speed: a timed item that a native reader finishes in a few seconds may take a second language reader longer. Our page on what IQ measures and the CHC model explain the abilities behind these distinctions, and the sections below take each in turn.

A fair reading of the evidence is therefore neither "a second language invalidates an IQ test" nor "language does not matter." It is a gradient. At one end sit vocabulary and general knowledge, which the language moves most. At the other end sit tasks built from shapes, patterns and spatial relations, which move less, though not to zero, as the section on nonverbal tests explains. A good decision follows the gradient: choose the form, read the profile by index, and treat any single number with the caution the Standards for Educational and Psychological Testing ask for.

This page is about taking a test as a person. It does not compare average scores of language groups, which the page on average IQ by language covers, and it does not claim that any ACIS score is equivalent across languages.

2 How Does Language Reach a Test Score? Linguistic Demand and Cultural Loading

Two properties of a test matter for a second language reader: how much language the task demands and how much culture-specific knowledge it assumes, and research suggests that the examinee's own language ability is the stronger of the two influences. Psychologists who assess students from other language backgrounds use a classification called the Culture-Language Interpretive Matrix, or C-LIM, which places each cognitive test along those two dimensions, cultural loading and linguistic demand, so that an examiner can judge how much a given score may reflect language and culture instead of the ability the test is meant to sample.

Whether the matrix classifications are accurate is an empirical question. Cormier, McGrew and Ysseldyke (2014) tested it in a study of the Woodcock-Johnson III cognitive tests published in the Journal of Psychoeducational Assessment. They noted that little empirical research supported the matrix recommendations, used the test's standardization sample to estimate how cultural loading and linguistic demand influenced performance on 20 of its tests, and reported results that supported a re-classification of the matrix. The practical reading is modest: the idea that tasks differ in language dependence is sound, but the published labels for individual tests were not exactly right, so any ranking of subtests by language load, including the one on this page, is a judgment and not a measurement.

A second study asked a sharper question. Cormier, Bulut, McGrew and Kennedy (2022) examined whether the language of the test directions or the examinee's own language ability explains performance. In their analysis in the Journal of Intelligence, they used a mixed-effects model with test characteristics, namely the linguistic demand of the directions, and examinee characteristics, namely expressive and receptive language ability. Their abstract reports that language abilities had a significant influence on cognitive test performance, whereas test characteristics did not influence performance after accounting for language abilities. The authors drew a practice implication: assess expressive and receptive language before administering, scoring and interpreting cognitive scores for English learners.

Three limits keep this finding in proportion. The sample was school-age students from the Woodcock-Johnson IV normative sample, not adults taking an online test. The authors state that it did not contain many English learners. And it concerns one test family, not ACIS. What transfers is the logic: if the examinee's language ability, and not the wording of the directions, drives the effect, then simplifying instructions helps less than choosing tasks that need less language.

The Standards for Educational and Psychological Testing, published by the American Educational Research Association, the American Psychological Association and the National Council on Measurement in Education in 2014, use second language examinees as an example of construct-irrelevant variance. In the chapter on fairness in testing, they note that vocabulary more complex than needed on an employment test may be a source of such variance for English language learners. Construct-irrelevant variance is score variation caused by something other than the ability the test claims to measure. On a reasoning task, English that stands in the way is irrelevant. On a vocabulary task, English is the construct itself. That distinction organizes the rest of this page.

3 What Does a Second Language Do to Vocabulary, Fluency and Speed?

A second language reliably lowers measured English vocabulary and semantic fluency even in fluent bilinguals, which is why vocabulary and general knowledge scores are the least portable part of an English IQ test. The evidence here is stronger than for any other claim on this page, because several studies agree.

Bialystok and Luk (2012) pooled English receptive vocabulary scores from 797 monolingual and 808 bilingual participants aged 17 to 89 across 20 studies, in Bilingualism: Language and Cognition. The distribution of scores was unimodal in both groups, but the mean standard score of the monolinguals was significantly higher than that of the bilinguals. In children the pattern is the same and better characterized. Bialystok, Luk, Peets and Yang (2010) analyzed 1,738 children aged 3 to 10 and found a consistent receptive vocabulary difference between bilinguals and monolinguals, and their preliminary analyses suggested that the difference did not change with different language pairs and was largely confined to words relevant to a home context rather than a school context. That detail matters for test takers: the gap is not a general weakness in language, it reflects which words were learned in which language.

For adults in an academic setting, Portocarrero, Burright and Donovick (2007) compared bilingual college students born abroad with monolingual English speakers in Archives of Clinical Neuropsychology. The bilinguals' English vocabulary was in the average range, yet they had lower receptive and expressive English vocabularies than their monolingual peers, despite having enough English to attend a competitive university. Age of arrival was moderately correlated with vocabulary, so those who arrived younger did better. Phonetic fluency was similar between groups, but semantic fluency was significantly lower in the bilingual group. Their result is a useful corrective for a common assumption: being admitted to an English language university does not mean an English vocabulary test will place you where a monolingual peer would be.

Two cautions apply. These studies compare group means and say nothing about any one person, whose English may be richer than that of the average monolingual. And a lower English vocabulary score does not mean a lower vocabulary overall: a bilingual's knowledge is split across languages, so an English-only test counts only the English part. That is why a verbal score for a second language speaker answers the narrow question "how much English word knowledge is there" and not the broad question "how much verbal ability is there." Our page on the verbal IQ test goes through the verbal score in more detail, and the Vocabulary subtest page says that a weaker result may reflect limited exposure or second language effects.

Speed needs an honest label. We did not find a study that measures how much a second language slows an adult on a timed online item, so the following is reasoning, not a finding. If reading an item takes longer, the time left for thinking is shorter, and the effect is largest on text-heavy tasks. The Standards recognize the issue: they say that when a developer recommends different time limits for people from diverse linguistic backgrounds, field testing should establish them instead of allowing an arbitrary multiple of the standard time. Timed, text-heavy items are the first place to look when a second language profile has a surprising dip.

4 Which Kinds of IQ Assessment Can a Non-Native English Speaker Choose, and What Do They Cost?

A non-native English speaker has six realistic routes, and they differ less in price than in the language of the test, who gives it and what the result may be used for. The table compares them with facts and prices from pages we opened on October 13, 2026. Where a page gave no price, the cell says so instead of guessing, and prices can change.

Three details sit behind the table. First, the WAIS-5. Pearson's product page for the WAIS-5 lists an age range of 16:0 to 90:11, a qualification level of C, and kit prices that start at 1,784 dollars. It also states that the test was normed on individuals whose primary language is English and recommends a caveat in summary reports that the WAIS-5 was normed on a sample of primary English speakers. That is the publisher of the best-known examiner-given test telling its users that language background is a limit. A psychologist who gives you a WAIS-5 in English still has to interpret your score with that limit in mind, which is why a psychologist who works in your language or knows how to use alternative measures is worth more than the brand of the test.

Second, Mensa. American Mensa's testing page listed Local Group testing proctored by volunteers at 60 dollars and private testing at more than 400 approved facilities at 99 dollars, a qualifying score at or above 98 percent of the general population, a photo ID with proof of date of birth, and one to two hours to complete. It showed promotional prices ending on October 15 and October 31 (year not stated), so we list standard prices. It did not mention English proficiency or a culture fair option. That option appears on a local chapter's testing page, which said the Culture Fair test is primarily focused on those for whom English is not a primary language, that one must contact the membership coordinator beforehand, and that both options carry a 60 dollar processing fee. That is one chapter's statement, so check with your own chapter or national organization. Our page on the Mensa IQ test covers what the society accepts.

Third, Raven's 2. Pearson's page for Raven's Progressive Matrices, Second Edition lists ages 4 to 90, a qualification level of B, and a complete print kit from 435.20 dollars. It says the test minimizes the impacts of language skills and cultural differences and can be given to people who are not language proficient. Those are the publisher's claims, and the next section explains how far to take them. The kit is sold to qualified professionals, who give the test.

Two other routes need a sentence each. A test translated into your language is a different test until someone shows otherwise: the Standards say that when a test is translated and adapted, developers or users must describe the methods and document evidence for the validity of the interpretations. A free quiz offered in many languages that shows no such method and no norms is entertainment, and our page on free versus validated IQ tests explains what a free number does not carry. A psychologist's fee depends on the clinic, the country and the question, so we print no figure we could not source. For supervised versus unsupervised testing, see professional versus online IQ tests and the professional IQ test guide.

5 Do Nonverbal and Culture Fair Tests Solve the Problem?

Nonverbal tests reduce the language problem and are the best available adjustment for a second language reader, but the research says they are culture-reduced, not culture-free, and they sample fewer abilities than a full battery. The reduction is real. A matrix or a spatial puzzle needs an instruction and little else, so a person who understands the instruction is not asked to know English words. That is why publishers describe these tests as less dependent on language, and why the nonverbal IQ test page and the page on the culture fair IQ test explain them as a reduced-language option for people who do not share the test's language.

The qualification is also real. Gonthier (2022) reviewed the psychological and ethnographic literature in Cognitive Research: Principles and Implications and argued that the assumption that visuo-spatial tests are culture-fair, because they are nonverbal, is dubious at best. Her review reports that the perception, manipulation and conceptualization of visuo-spatial information differs significantly across cultures in ways relevant to intelligence tests, and she outlines a model of how those differences can affect seven steps of the solution process for Raven's matrices. Her conclusion is that a number of cultural assumptions appear deeply ingrained in visuo-spatial reasoning tests, to the extent that it disqualifies the view of such tests as intrinsically culture-fair. Her target was the use of such tests for group comparisons, and that is not our question, but the point carries over: a nonverbal test removes the English, not the culture of the task.

For an individual reader the consequences are three. First, treat "nonverbal" as "less dependent on English," which is the claim the evidence supports, and not as "fair." Second, remember what is lost: a nonverbal composite leaves out crystallized knowledge, Gc, so it describes reasoning, spatial ability, memory and speed, not the verbal knowledge an all-round score includes. Third, the label is a convention. ACIS's own report text says its Culture Fair composite lowers verbal knowledge dependence relative to the Full Scale IQ but is not culture free, because no cognitive test is, and describes its Nonverbal composite as a reduced verbal profile summary and not a culture free absolute measurement. Read any composite with those names that way.

Pearson's page says Raven's 2 "minimizes the impacts of language skills and cultural differences," which is the language of reduction. Our page on whether IQ tests are biased takes up the wider fairness debate, and the page on reliability and validity explains why validity evidence is specific to a group and a purpose, which is the standard a nonverbal claim must also meet.

6 Is Testing in Your Own Language Always Better?

Testing in the language in which you are most proficient is the professional default, but it is not automatic, because the Standards allow that the language of your schooling or training can be the more appropriate one, and a translated test is only as good as the evidence behind the translation. The Standards for Educational and Psychological Testing state in Standard 3.13 that a test should be administered in the language that is most relevant and appropriate to the test purpose. The accompanying comment says that, except where the purpose is to determine proficiency in a particular language, test takers should be tested in the language in which they are most proficient, and adds that in some cases a person's most proficient language may not be the language in which they were instructed or trained in relation to the tested constructs, so that administering the test in the language of instruction may be more appropriate.

That nuance matters for adults. A person who trained as an engineer in English may handle quantitative and technical material better in English than in the language spoken at home, and the opposite can hold for a person who moved abroad as an adult. The American Psychological Association's ethics code points the same way. In Standard 9.02, psychologists are told to use assessment instruments whose validity and reliability have been established for use with members of the population tested, and to describe the strengths and limitations of results when that has not been established, and to use assessment methods appropriate to an individual's language preference and competence, unless the use of an alternative language is relevant to the assessment issues. Those lines govern psychologists, not you as a consumer, but they are the yardstick a clinician applies and the one you can ask a clinician about.

Translation is the usual proposed remedy, and the evidence standard is demanding. Standard 3.12 requires developers or users of a translated and adapted test to describe the methods and document evidence for the validity of the interpretations. The comment to Standard 3.14 notes that differences in word meaning, familiarity, frequency, connotations and associations make it difficult to compare scores from non-standardized translations with English-language norms. The International Test Commission publishes guidelines for translating and adapting tests (second edition, International Journal of Testing, 2018) as the professional reference. A browser's machine translation of an English test is not an adapted test, and using it during a test also changes the conditions under which the score was normed.

Standard 3.14 adds that where an interpreter is needed, standardized procedures apply, and that unless a test was normed with interpreters, using one may alter the measurement. That concerns examiner-given settings, but it shows how seriously the field treats any third party between the test and the examinee.

What follows for choice? If your purpose carries consequences, such as a school placement, a work decision, a legal matter or a clinical question, choose a licensed psychologist who works in your language or who can document how they handled the language, and ask which test and which norms they use. If your purpose is personal understanding, an English test you can read well, interpreted index by index, is a reasonable option. If your purpose is society membership, go to the society's own test, and for a nonlanguage format ask the chapter. Our page on the accurate IQ test explains how to judge whether a score is real, and the page on average IQ in Germany shows what a national comparison involves for readers in a non-English country.

7 How Good Does Your English Need to Be for an English IQ Test?

No published threshold tells an adult how much English is enough for an English IQ test, so the honest answer is a self-check you run before you pay, built on the finding that your own language ability, not the wording of the directions, drives the effect. We searched for a validated cutoff, such as a proficiency level at which scores stop being distorted, and did not find one for adults on an unsupervised online test. Cormier and colleagues (2022) recommend assessing expressive and receptive language before interpreting cognitive scores for English learners, which implies that no fixed number replaces measuring the person. What follows is our rule of thumb, not a research result, and it is meant to help you decide whether to start.

Run five checks. First, can you read a paragraph of instructions in English once, then explain in your own words what you were asked to do? A task you must reread several times costs time you will not get back on a timed item. Second, do you read news, work documents or articles in English without translating them? Daily reading is a more direct sign of comfort than the number of years you studied the language. Third, can you type a short English sentence explaining what a word means without searching for the words to do it? Three ACIS verbal subtests ask for typed answers, so production matters. Fourth, how do you feel about the clock? The subtest pages state limits such as 90 seconds per item on Similarities, 45 seconds per item on Complex Relations and 30 seconds per item on Information. Fifth, can you follow spoken English from headphones at a steady pace? Several working memory tasks play audio.

If you pass four or five of the checks, an English test read index by index is a reasonable choice. If you pass two or three, favor the tasks that need less English and treat the verbal indices as a measure of English word knowledge. If you pass fewer, a test in your own language, with a psychologist if the stakes are high, is the better route, and you lose nothing by waiting.

The free trial exists for this purpose. ACIS offers a free trial with no card, so you can see how the instructions and timed screens read before you buy a form. Watch two things in it: how long you take to understand an instruction, and whether a timer runs out while you are still reading.

8 Which ACIS Subtests Lean on English, and Which Lean on It Less?

Six of the 20 ACIS subtests lean heavily on English, five lean on it moderately, and nine lean on it little once you understand the instructions, which is our reading of each subtest page and not a measured property. The table uses what the subtest pages say about format, typed versus selected answers and time limits, and it adds our judgment of language load. We have not measured these loads, and as the 2014 study of the Culture-Language Interpretive Matrix showed, published classifications of this kind needed revision, so use the table as a guide to where to look and not as a guarantee. All instructions are in English for every subtest.

Several rows deserve comment. Antonyms and Synonyms remove the typing but not the vocabulary, so a second language reader who recognizes a word can still choose well, while the typed subtests add production effort. That difference is useful when you read your report. Complex Relations is a typed verbal analogy, which is why we place it in the high column even though ACIS groups it with fluid reasoning. The Arithmetic page itself lists misunderstanding the verbal problem among the reasons for a weaker result.

The moderate rows differ in cause. Digit Span and Alphanumeric Sequencing use spoken audio, so listening comprehension matters, although the content is digits and letters. Alphanumeric Sequencing applies an order to letters, so a reader whose first language uses another script should look at its page. Spatial Comprehension uses compass points written as N, NE, E and so on, so a reader whose language abbreviates a direction differently should check that page. The page for Mathematical Achievement does not describe its wording load, so judge it in the trial.

The low rows are the ones the page text supports as language-light. The Layer Rotation page says the instruction is carried by a mark on the layer and not by a sentence, so the task places very little load on language. The Matrix Reasoning page says the content is abstract and usually nonverbal, and the Symbol Search page says the format keeps the demand on search efficiency and not on language. Even these keep the English of the opening instructions.

9 What Do the ACIS Nonverbal and Culture Fair Composites Report, and Which Form Includes Them?

The Nonverbal composite is built from eight subtests with minimal verbal content, the Culture Fair composite from the subtests outside the Verbal Comprehension Index (15 on Full Scale, 9 on Optimized), and ACIS's own report calls neither of them culture free. The home page of ACIS lists the subtests behind each. Nonverbal draws on Matrix Reasoning, Figure Weights, Visual Puzzles, Layer Rotation, Logic Grid, Visual Sequence, Symbol Search and Coding. Culture Fair draws on all the subtests except the five in Verbal Comprehension, which means it keeps Complex Relations, Arithmetic, Mathematical Achievement, Digit Span and Alphanumeric Sequencing, tasks the table above rates higher in language load than the Nonverbal set.

That difference is the main thing to understand. The report describes the Culture Fair composite as emphasizing reasoning, working memory and processing efficiency with reduced reliance on acquired verbal content, which lowers verbal knowledge dependence relative to the Full Scale IQ but is not culture free. It describes the Nonverbal composite as covering reasoning, visual spatial ability, working memory and speed with minimal reliance on verbal content, a reduced verbal profile summary and not a culture free absolute measurement. So if your concern is English, the Nonverbal composite is the cleaner of the two, and the Culture Fair composite is the broader of the two. We do not state precision figures for either here; the technical manual documents how each is built and scored, and its figures are there for you to read in context.

Which form carries these scores? The Full Scale form reports both composites; it contains all 20 subtests in six domains, takes about 175 minutes with breaks allowed and costs 50 dollars one time (prices read on October 13, 2026). The Optimized form, 30 dollars, has 13 subtests in five domains and reports a Culture Fair composite built from its nine subtests outside Verbal Comprehension, but no Nonverbal composite. The Quick form, 15 dollars, has six subtests in three domains and reports neither. The reason the Nonverbal composite cannot come from the shorter forms is simple: it draws on Visual Puzzles and Visual Sequence, and neither Quick nor Optimized contains either of them. Quick holds two verbal comprehension subtests out of six, and Optimized holds four out of 13, so for a second language reader the shorter forms put proportionally more weight on English.

That is a product fact, not a sales argument. Full Scale takes the longest, and reading and writing in a second language can be tiring, so plan sessions: the product allows breaks, gives 30 days to complete and carries a 5 day quality guarantee. If you want only a quick look at reasoning, the shorter forms exist, but they will not give you the Nonverbal composite. Our page on the Full Scale IQ explains what the headline number combines, and its report describes the six indices.

One caution completes the picture: the composite names promise less than they sound. A Nonverbal composite describes the tasks included and does not state that the score is unaffected by language or culture, a point Gonthier's review makes for visuo-spatial tests in general. And the Full Scale IQ still includes the verbal subtests, so for a second language reader it may sit below the Nonverbal composite partly because of English. That is why the next section teaches you to read the two together, and why we never call either number your real IQ.

10 How Should You Read an ACIS Report When English Is Your Second Language?

Read the report as a profile and compare indices with each other, because a gap between the verbal index and the others is a pattern that a second language can produce but cannot prove. The report gives indices and the Full Scale IQ on the standard scale, where the mean is 100 and the standard deviation is 15, and it gives percentiles and a 95 percent confidence interval. Scaled subtest scores run from 1 to 19, with a mean of 10 and a standard deviation of 3. Our pages on what IQ scores mean and the standard deviation of 15 explain the scale, and the page on IQ score versus percentile explains why the two should not be averaged or confused.

Use a sequence of five comparisons. First, set the Verbal Comprehension Index beside the Fluid Reasoning, Visual Spatial and Working Memory indices. If the verbal index is lower than the rest, a second language is one plausible reason, because the research above finds lower English vocabulary and semantic fluency in bilinguals, but it is not the only reason, and people have uneven profiles for many reasons unrelated to language. A gap is a question, not an answer.

Second, compare the Nonverbal composite with the Full Scale IQ. If the Nonverbal composite is higher, part of the difference may be English, and part may simply be that the two composites contain different tasks and so draw on different abilities. The Full Scale IQ includes crystallized knowledge and the Nonverbal composite does not. Neither is your real IQ. Each answers a different question, and our page on the General Ability Index shows the same logic for a composite that reduces working memory and speed instead of language.

Third, compare subtests that ask for the same kind of knowledge in different ways. Antonyms and Synonyms ask you to choose, and Vocabulary, Information and Similarities ask you to type. If you do well when you can choose and less well when you must write, production in English may be costing you. This is our hypothesis for how to read a profile, not a tested rule, so treat it as a prompt for a closer look, not as a conclusion.

Fourth, look at the time limits. A subtest with a short per-item clock, such as Complex Relations at 45 seconds, can punish reading time, and if its score is below its neighbors in the same index, reading speed is one candidate explanation.

Fifth, read the confidence interval and not only the point. The interval is the range within which the score would be expected to fall on repeated measurement, so a difference between two indices that is smaller than their uncertainty should not be interpreted. A score is an estimate, which is why the interval matters.

Then stop. A report of this kind describes how you performed on this day, under these conditions, in this language. It is not a clinical or diagnostic result, and it should not be used for hiring, school accommodations or admission to high IQ societies. A group finding, such as lower mean English vocabulary among bilinguals, says nothing about you as an individual. The next section turns the reading into a choice.

11 Which Test Should You Choose for Which Purpose?

Match the test to the purpose: personal understanding allows an English test read by index, society membership follows the society's own route, and any decision with consequences belongs to a psychologist who can handle your language. The table puts the same advice in one place.

Three notes explain the rows. The curiosity row is where ACIS fits, and the reason for the Full Scale form is the Nonverbal composite, not a general preference for the longest option. If the Nonverbal composite is not what you need, there is no reason to pay for 20 subtests. For the society row, the page on high IQ society requirements shows that societies publish their own admission routes, and ACIS is not for admission to them, so the Culture Fair option on one Mensa chapter's page is the one to ask about. For the consequences row, the choice of a psychologist who works in your language is the standard advice in the professional documents quoted above, and the cost is the cost of a clinical service, which we could not source.

The country row deserves a sentence. If you live outside the English-speaking world, national norms in your language answer a different question from the one an English test answers, and our country pages, such as the one on average IQ in Germany, show how national comparisons are framed. Readers in English-speaking countries who are non-native speakers face the same question as everyone on this page, and the sibling guides for readers in Canada, college students, seniors and former gifted kids cover other situations that can combine with a second language.

12 Where ACIS Sits, and What a Buyer Should Do

ACIS is a normed, online, unsupervised, English-only cognitive assessment that reports six indices and a Full Scale IQ, it is not a clinical or diagnostic instrument, and it does not claim that its scores are equivalent for readers whose first language is not English. We state the limits as plainly as the strengths because the Standards for Educational and Psychological Testing (AERA, APA and NCME, 2014) make that disclosure part of responsible practice. Standard 3.15 says that developers and publishers who claim a test can be used with examinees from specific subgroups are responsible for providing the information that supports appropriate interpretation for those subgroups. We make no such claim for readers of other first languages, and this page is therefore a guide to using the instrument cautiously, not evidence that it works equally for everyone. Standard 3.16 says that when credible research indicates scores for some subgroups are differentially affected by construct-irrelevant characteristics, test users should use the test only for subgroups for which there is sufficient evidence of validity to support the intended use. The research reviewed above, which finds that language abilities influence cognitive test performance, is the reason we advise against using any English test for a consequential decision about a second language reader.

What ACIS offers is concrete. It reports 20 subtests in six domains as a Full Scale IQ and six primary indices, on the standard scale, with percentiles and a 95 percent confidence interval, and its adult norms cover ages 16 to 90. The Optimized form reports a Culture Fair composite, and the Full Scale form adds a Nonverbal composite to it; these are the scores most relevant to a second language reader. There is no subscription: Quick costs 15 dollars, Optimized 30 dollars and Full Scale 50 dollars, one time, as read on October 13, 2026, with a free trial that needs no card, 30 days to complete and a 5 day quality guarantee. The technical manual is the documentation to read before you decide.

What ACIS does not offer is equally concrete. It is online and unsupervised, so there is no examiner to notice that you misread an instruction or to adapt the pace. It is English only: every instruction, every verbal item and every typed answer is in English, and there is no translated version. It is not a clinical or diagnostic instrument, and it is not for hiring, for school accommodations or for admission to high IQ societies. Where the American Psychological Association's ethics code asks psychologists to use assessment methods appropriate to a person's language preference and competence, that is a duty on a clinician and a standard against which to judge a clinical service. ACIS is not that service.

What should a buyer do? If your purpose is personal understanding and you read English comfortably, start the free trial and notice how long the first instructions take to understand. If it feels comfortable, the Full Scale form is the one that reports the Nonverbal composite, and the reading guide above tells you how to set it beside the Full Scale IQ. If your purpose carries consequences, or your English is limited, go to a psychologist who works in your language, or to the society's own route, and keep this page as a list of questions to ask: Which language will the test use? Which norms apply to someone like me? How will you handle my second language in the interpretation? Those questions are the practical form of what the Standards ask.

13 Sources Behind This Page

Every figure on this page comes from one of the sources below, opened or confirmed on October 13, 2026, or from the ACIS pages linked in the text.

• American Educational Research Association, American Psychological Association and National Council on Measurement in Education (2014). Standards for Educational and Psychological Testing. Washington, DC: AERA. Standards 3.12 to 3.16 and the chapter on fairness in testing, read in the open access edition.
• American Mensa. Testing. Page read October 13, 2026.
• American Psychological Association (2017). Ethical Principles of Psychologists and Code of Conduct. Standard 9.02, Use of Assessments, in the version amended effective June 1, 2010 and January 1, 2017.
• Bialystok, E., and Luk, G. (2012). Receptive vocabulary differences in monolingual and bilingual adults. Bilingualism: Language and Cognition, 15(2), 397 to 401.
• Bialystok, E., Luk, G., Peets, K. F., and Yang, S. (2010). Receptive vocabulary differences in monolingual and bilingual children. Bilingualism: Language and Cognition, 13(4), 525 to 531.
• American Mensa, Cincinnati chapter. Testing. Local chapter page read October 13, 2026.
• Cormier, D. C., Bulut, O., McGrew, K. S., and Kennedy, K. (2022). Linguistic influences on cognitive test performance: Examinee characteristics are more important than test characteristics. Journal of Intelligence, 10(1), 8.
• Cormier, D. C., McGrew, K. S., and Ysseldyke, J. E. (2014). The influences of linguistic demand and cultural loading on cognitive test scores. Journal of Psychoeducational Assessment, 32(7), 610 to 623.
• Gonthier, C. (2022). Cross-cultural differences in visuo-spatial processing and the culture-fairness of visuo-spatial intelligence tests: an integrative review and a model for matrices tasks. Cognitive Research: Principles and Implications, 7(1), article 11.
• International Test Commission (2018). ITC guidelines for translating and adapting tests (second edition). International Journal of Testing, 18(2), 101 to 134.
• Pearson Assessments. Raven's Progressive Matrices, Second Edition (Raven's 2). Product page read October 13, 2026.
• Pearson Assessments. Wechsler Adult Intelligence Scale, Fifth Edition (WAIS-5). Product page read October 13, 2026.
• Portocarrero, J. S., Burright, R. G., and Donovick, P. J. (2007). Vocabulary and verbal fluency of bilingual and monolingual college students. Archives of Clinical Neuropsychology, 22(3), 415 to 422.

14 Frequently Asked Questions