The inductive reasoning test: what an employer's pattern task measures

0 The short answer

An inductive reasoning test is a short timed task in which you infer the rule that links a set of shapes, symbols or sequences and apply it to the next case, and it samples induction, the core narrow ability of fluid reasoning (Gf) in the CHC model. SHL's fact sheet for its Verify Interactive version lists 18 minutes, a maximum of 15 questions and an adaptive format (read October 29, 2026), and the test reaches you through an employer's invitation, so no outside page can show you a score on it. ACIS is one option for seeing your own induction on adult norms, through the Matrix Reasoning subtest that sits in every form (Quick 15 dollars and about 45 minutes, Optimized 30 dollars and about 110 minutes, Full Scale 50 dollars and about 175 minutes, prices read October 29, 2026), but it is not that test, does not prepare you for it and does not predict it. The decisive limit is that a second attempt at the same kind of task typically scores higher, so a rising score is not by itself proof of a stronger ability.

1 What Is an Inductive Reasoning Test, and What Does SHL's Version Contain?

An inductive reasoning test is a timed, mostly nonverbal puzzle format in which the correct answer is the one that fits the rule you infer from the examples on the screen, and publishers and employers use several labels for related tasks, among them the inductive-logical reasoning that one bank lists for its applicants. SHL sells it as Verify Interactive Inductive Reasoning. Its 2022 fact sheet, which we opened on October 29, 2026, says the assessment measures a candidate's ability to identify patterns within pieces of information and generate innovative solutions, that it tests the ability to infer rules, patterns or sequences and then use that information to predict outcomes, and that the task presents information as shapes and alphanumeric sequences.

The same fact sheet lists the logistics in a table. Allowed time is 18 minutes, the maximum number of questions is 15, the test is taken in one sitting, the question format is described as interactive, activity based and adaptive, the product category is Ability and Aptitude, and the test is designed for an unsupervised environment. The fact sheet says the adaptive technology offers shorter testing and administration times, the ability to test unsupervised and a more precise score. It describes the test as used for candidates applying to mid to senior level positions that require inductive reasoning ability, and it names troubleshooting, creative problem solving and innovative idea generation as sample tasks of jobs that may require it, with Product Development, IT, Professional Services and Management as example roles. Adaptive, in the usual sense, means the questions you receive depend on how you answered earlier ones, which is how a fixed maximum can sit beside the shorter testing time the fact sheet claims. SHL's cognitive ability page, read the same day, describes the inductive test as working out the pattern in a set of images and identifying the correct next image, and says its cognitive assessments ask fewer questions because of a partial scoring method.

Those are SHL's own descriptions of one product at one date. Two cautions follow. First, the fact sheet does not say how scores are reported, what comparison group they are set against or how an employer uses them, and we will not guess. Second, the figures belong to the Verify Interactive version: an invitation that names a different version, or a different publisher, may differ in time, question count and format. The page on the SHL test family goes through the publisher's range, and this page stays with the inductive task.

Employers name the task differently. The UBS graduate program page, read October 29, 2026, lists four online assessments, namely verbal reasoning, numerical reasoning, inductive-logical reasoning and a culture match, and says which ones an applicant receives depends on the business area. It says they must be completed within seven days of the invitation email and cannot be retaken. That is one employer's rule, not a rule of the test, but it shows what to check first: how many attempts there are and by when.

Where does induction sit among the neighboring labels? The page on abstract reasoning tests covers the wider family of figure based reasoning tests and where their overlap with IQ ends. The page on diagrammatic reasoning tests covers a neighboring employer format, and the page on deductive reasoning tests covers the opposite direction of inference, from stated premises to a conclusion that must follow. For the general category that all of these belong to, see the page on pre-employment cognitive tests. We describe only SHL's format, because it is the one publisher page we opened.

2 Which Tests Sit Near It? A Comparison of the Real Options

No other option on this list is a substitute for the employer's test: they differ in who administers them, whom they compare you with and what you receive, and the table lets you see why. It holds only what we could read on a publisher or company page on October 29, 2026. Where a page is silent, the cell says "not stated". Practice material sold by third parties is left out of the table on purpose: those sellers are not sources about intelligence, and a candidate can find them without our help.

SHL's practice page is the publisher's own. It offers two inductive practice versions, one mobile enabled with a drag and drop format and one in a standard multiple choice format, and states no time limit or number of questions for either. It warns that the practice difficulty may not exactly reflect the test you will take, and that you are not permitted to share your results with any third party. The publisher gives you a way to see the interface, and says in its own words that the practice is not a map of your test.

Three things stand out in the table. Only the professional battery and ACIS state whom you are compared with; the employer test's fact sheet is silent on it. Only the professional battery is individually administered. And the minutes are not comparable: 18 minutes of one adaptive task and 175 minutes of twenty subtests answer different questions.

3 What Is Induction in the CHC Model?

In the CHC model induction is one of three narrow abilities under fluid reasoning (Gf), the ability to observe a phenomenon and work out the principles or rules that determine how it behaves, and that is exactly the demand an inductive reasoning test sets. Hanif Akhtar's 2022 review of fluid reasoning measurement in Buletin Psikologi, which we opened on October 29, 2026, summarizes Schneider and McGrew's definition of Gf as the deliberate but flexible control of attention to solve novel problems that cannot be solved by relying only on previously learned knowledge. The same review lists what such thinking includes: identifying relations, forming concepts, generating and testing hypotheses, drawing inferences, classifying, generalizing, extrapolating and transforming information. Two features of that definition decide how the test should be read. The problem is novel, so a fact you already know cannot carry you to the answer. And attention is controlled, so the task depends on keeping the right things in mind while you search.

The three narrow abilities are named in the same review and in a 2018 paper by Anita Lubis and Dewi Maulina on building a figural inductive reasoning test: induction, general sequential (deductive) reasoning and quantitative reasoning. Lubis and Maulina, who quote Schneider and McGrew (2012, page 112) for the definition of induction, add that the field commonly treats induction as the core narrow ability of Gf and that matrices are among the tasks used to measure it. That wording is theirs, and it is a statement about how the literature frames induction, not a measured ranking. The practical consequence is simple. When a publisher says "inductive reasoning", the CHC name for what is being sampled is induction within Gf, and when it says "deductive reasoning" the name is general sequential reasoning, the narrow ability for reasoning logically from stated premises. The page on deductive reasoning tests takes up that second ability on its own terms.

CHC names matter here because everyday words blur the two. In ordinary speech, "inductive" can mean generalizing from examples, and in a test it means that too, with one difference that is easy to miss: an item is built so that one intended rule fits all the evidence on the screen, and the keyed answer is the one that follows from that rule. Our reading is that the task is inductive in how you reach the answer, because the rule is hidden and must be inferred, and deterministic in how it is scored, because the designer fixed the rule in advance.

Induction is not confined to shapes. The ACIS technical manual makes the point about its own tasks when it says that the same item format can involve more than one process, and that a semantic relation item may involve both lexical knowledge and induction. It is also why a reader should not treat "inductive reasoning" and "pattern recognition" as synonyms; the page on pattern recognition tests separates what the pattern format measures from what the label suggests. For the wider map of broad abilities that Gf belongs to, the guide to the CHC model and the page on cognitive domains in IQ testing set out each one with its CHC name.

4 What Happens in the Mind When You Find the Rule?

The best process evidence comes from a 1990 analysis of Raven matrices, which found that higher scorers differ from lower scorers mainly in two things: inducing abstract relations and managing a large set of goals in working memory. Carpenter, Just and Shell, in Psychological Review, analyzed the processes in the Raven Progressive Matrices test using verbal protocols, eye fixation patterns and errors. They expressed their theory as a pair of computer simulation models that performed like the median or the best college students in their sample. They report that the processing characteristic common to all subjects was an incremental, reiterative strategy for encoding and inducing the regularities in each problem, and that the processes distinguishing individuals were primarily the ability to induce abstract relations and the ability to manage a large set of problem solving goals dynamically in working memory. We read the abstract, not the full text, so we report the abstract's claims and nothing finer.

What does this say about an inductive reasoning test? Three cautious points. First, the study analyzed a matrix test, with a model benchmarked against college students. It describes the processing behind that test and does not describe SHL's adaptive format, so anything we say about a different test is an extension, and we label it as one. Second, the incremental strategy fits what a rule finding task asks of anyone: find one regularity, check it, add the next. A harder item does not demand a different kind of thinking so much as more regularities held together, and the technical manual describes its own Matrix Reasoning in those terms, saying difficulty rises as items combine more rules operating at once across rows and columns. Third, the two distinguishing processes are not one thing: inducing relations gives the task its name, and managing goals is the working memory demand that rides along with it. The Fluid Intelligence Test page presents the working memory dispute as a dispute and not as a settled answer.

The three induction subtests in ACIS differ in what they ask the person to do, and the differences are worth knowing even for a reader who never takes ACIS. In Matrix Reasoning, per the manual, the examinee sees an incomplete visual matrix and selects the option that completes it; the material is nonverbal and the rules are never stated, so they must be induced from the arrangement. The Matrix Reasoning page adds that a weak result can reflect difficulty discovering rules, comparing several visual relations or holding several possible hypotheses at once. In Visual Number Series, the examinee sees a sequence of numbers with one value missing and types the value that continues the rule; because the answer is produced and not chosen, the examinee cannot work backward from options. The Visual Number Series page places it between Gf and quantitative reasoning (Gq) and notes that the numbers are the vehicle and the demand is rule discovery. In Complex Relations, per the manual, the examinee completes an abstract analogy by supplying both missing terms, so the rule linking the given pair must first be induced and then applied twice. The Complex Relations page calls the demand relational integration. Number series have their own history in aptitude testing, which the page on number series tests traces, including why a major battery demoted the format. We describe these formats and not their items, and nothing here claims that an employer's task looks like any of them.

5 How Does Inductive Reasoning Relate to Raven-Style Matrices and to g?

Inductive reasoning is the task family that Raven-style matrices made famous, and fluid reasoning is the broad ability that sits closest to the general factor, but a single matrix test is not g and a score on one is not a measure of general intelligence. The Raven tradition is the reason "find the rule in the grid" became the default picture of an intelligence test. The page on Raven's Advanced Matrices covers that test, its sets and its scoring in its own right, and this page uses the tradition only for what it shows about induction. Lubis and Maulina note that matrices are among the tasks used to measure inductive reasoning, and Carpenter and colleagues describe the Raven test as a widely used, nonverbal test of analytic intelligence.

On the link to g, the literature is stronger than a casual reader might expect and also more contested. Akhtar's abstract says fluid reasoning has been shown to be statistically indistinguishable from general cognitive ability. In the body of the review he reports that Gustafsson (1984) found the second order fluid reasoning factor to be identical to the g factor, and he also notes that Blair's 2006 review from developmental neuroscience argued that Gf and g should be distinguished. A balanced summary has both sides: in factor analytic work Gf and g can be hard to separate, and other lines of evidence suggest they are not the same thing. We rely on Akhtar's account of those two papers and did not open them. The explainer on the g factor sets out what g is and is not, and the page on the Fluid Reasoning Index explains why factor analyses dispute how cleanly a fluid reasoning score stands apart.

None of this licenses reading a matrix result as general intelligence. A test is one sample of a construct, with its own format, its own error and its own specific demands. How much general intelligence a single matrix test captures is the question the pattern recognition page, mentioned above, works through with the numbers, and the page on what IQ measures explains why a Full Scale IQ is built from several domains instead of one. The same caution applies to "culture fair" claims. Figural content reduces the dependence on vocabulary and schooling in a language, and the technical manual, describing its own nonverbal composite, says it is not culture free because no cognitive test is. The page on culture-fair IQ tests goes through what figural formats do and do not remove. For an employer's test the same logic holds: shapes lower one barrier and leave others in place.

6 Why Do Employers Test It, and What Does the Evidence Support?

Employers use a test like this because cognitive ability measures are an established part of selection research, but the evidence behind that practice concerns selection procedures in general, a 2022 synthesis lowered the validity estimates for most of them by .10 to .20 points, and nothing on SHL's fact sheet gives a validity figure for this particular test. The fact sheet's stated purpose is an interactive, business relevant approach to predict workplace performance. That is a design claim by the publisher, and it names jobs that may call for inductive reasoning: troubleshooting, creative problem solving and innovative idea generation. Whether a person's skill at finding the rule in a set of figures predicts those activities in a given job is a question of evidence, and the fact sheet does not supply it.

The broader evidence has been revised downward in recent years. Sackett, Zhang, Berry and Lievens (2022), in the Journal of Applied Psychology, revisited earlier meta-analytic estimates of validity in personnel selection and concluded that common corrections for range restriction had often overcorrected, so that the validity of many selection procedures for predicting job performance had been substantially overestimated. Their abstract reports that most of the procedures that ranked high in earlier summaries remain high in rank, with mean validity estimates reduced by .10 to .20 points, that structured interviews emerged as the top ranked procedure, and that the authors paired validity estimates with information on mean Black and White subgroup differences. They conclude that selection procedures remain useful and that predictor and criterion relationships are considerably lower than previously thought. We read the abstract, not the tables, so we do not quote a validity figure for cognitive tests from that paper. The page on IQ and job performance reviews the general mental ability literature in more detail, and the page on pre-employment cognitive tests covers what such tests are and how they sit in a hiring process.

What follows for the person sitting the test? Three things, none of them advice about tactics. First, a cognitive test is rarely the only input: the 2022 synthesis ranks structured interviews at the top, a reminder that a hiring decision is built from several procedures. Second, a group level validity coefficient describes a relationship across many people and cannot say what your own score means for your own future performance. Third, the Standards for Educational and Psychological Testing (AERA, APA and NCME, 2014), which we read in the open access edition, say in Standard 11.2 that evidence of validity based on test content requires a thorough and explicit definition of the content domain of interest, and in the accompanying comment that for an employment test the job content domain should be described in terms of the tasks performed or the knowledge, skills, abilities and other characteristics required on the job. That standard is addressed to the employer and the publisher. A candidate cannot audit it, but can see that "this role needs inductive reasoning" is a claim someone should be able to support.

One more distinction keeps expectations honest. An employer inductive test is built for a selection purpose, not to estimate a person's IQ, which is why the rest of this page never treats a result on it as an IQ.

7 Can You Practice? Free Tests, Examples, Answers and What Retests Show

Practice and retesting raise scores on cognitive ability tests: across 122 studies a second attempt added about a third of a standard deviation, and the gain was larger on an identical form than on an alternate form, so a rising score shows practice has done something without proving the ability has changed. The search terms "inductive reasoning test practice", "free", "examples" and "with answers" all come down to the same question, and the answer has two parts: what is available, and what practice does.

What is available. The publisher itself offers practice versions of its inductive test, as the comparison table shows, and SHL says they may not reflect the difficulty of the real one. Third parties sell practice material for employer tests, and we neither link to nor cite them: they are commercial sellers, and reproducing or redrawing items from any test is outside what this page does, so we describe formats and no more. Answer keys deserve a plain word. The live test is described as adaptive, so the questions one candidate sees depend on that candidate's earlier answers, and a list of answers to someone else's questions would not map onto your own sitting. That inference follows from the meaning of adaptive testing and is ours, not SHL's. The Standards speak to fairness here. Standard 8.1 says information about test content and purposes that is available to any test taker before testing should be available to all test takers, and its comment adds that some organizations may supplement it with additional training or coaching. Standard 1.7 says that if performance is claimed to be essentially unaffected by practice and coaching, the propensity for performance to change with these forms of instruction should be documented, and its comment says materials written for test takers should provide practical guidance about the value of test preparation activities, including coaching.

What practice does. The most recent large synthesis is Scharfen, Peters and Holling (2018), in Intelligence. It pooled 174 samples from 122 studies, with 786 test outcomes and an overall sample size of 153,185, applying strict criteria on study design, participant age and health, and the tests included. They found significant retest effects and no further score gains after the third administration. The effect is expressed as a standardized mean change, a gain in standard deviation units, and the table gives the pooled values we read in the paper.

Read the rows as a curve, not a ladder. Most of the gain comes early, and the paper reports that between the third and fourth administrations the change was 0.02, with an interval of minus 0.020 to 0.062, so no further gain. On a scale with a standard deviation of 15, 0.33 of a standard deviation would be about 5 points, which is our arithmetic and a translation of convenience: the studies in the meta-analysis use many different tests and scales. Four further points matter for a candidate. Identical forms produced larger first to second gains (0.372) than alternate forms (0.226), a difference the authors report as significant. Reasoning tasks, the category that includes figural matrices, showed a first to second gain of 0.326 across 68 samples, close to the overall figure, and the authors report that differences between cognitive operations were not significant in the comparisons they tested, with one exception late in the sequence. The longer the interval, the smaller the effect, but weakly: the authors estimate a decrease of 0.0008 standard deviations per week, which they illustrate as 0.04 standard deviations over 50 weeks. And the studies are retests of the same abilities, so they speak to what a repeated attempt does and not to what any particular employer's test will ask.

Why do scores rise? The meta-analysis is descriptive and the question has several answers: familiarity with the format, reduced anxiety, memory for specific items on identical forms and test specific strategies. The authors discuss these in their introduction and results, and we do not rank them. For a reader who may meet an inductive test twice, a second score on the same kind of task is not comparable with a first as if the two measured the same thing, and the page on retaking an IQ test explains why the second score runs higher. This page sells no preparation.

8 Does Training Improve Inductive Reasoning?

Training studies show that people can be taught a procedure for inductive reasoning and then score higher on inductive tasks, but the strongest meta-analysis is about children, the adult trial evidence is about older adults, and a gain on a trained task family is not the same as a higher general ability. The reader of an employer's test usually wants to know whether a few hours of work can change the result, and the research answers a different question: what sustained, structured training does.

Klauer and Phye (2008), in Review of Educational Research, present a prescriptive theory of inductive reasoning and a meta-analysis of 74 training experiments with nearly 3,600 children. The training teaches a procedural strategy for making comparisons. The authors hypothesized that it would increase fluid intelligence test performance and improve academic learning, and they report that both hypotheses were confirmed. They also report two moderating effects: training effects on intelligence test performance increased over time, and transfer to academic learning was greater than transfer to intelligence test performance. They state that the results cannot be explained by placebo or test coaching effects, and they conclude that children of a broad age range and intellectual capacity benefit. That is the authors' conclusion about a particular trained procedure in children. It is not a finding about adults preparing for a hiring test, and it does not say that a different kind of practice has the same effect.

For adults, the largest randomized evidence comes from the ACTIVE trial in older adults. Ball and colleagues (2002), in JAMA, randomized 2,832 volunteers aged 65 to 94 to ten session group training in memory, in reasoning (the ability to solve problems that follow a serial pattern; 705 people), or in speed of processing, or to a no contact control. Each intervention improved its targeted ability compared with baseline, with effects that lasted to two years; 74 percent of the reasoning trained participants showed reliable improvement immediately after training, and booster training enhanced the gains. The authors found no training effects on everyday functioning at two years. Ten years on, Rebok and colleagues (2014), in the Journal of the American Geriatrics Society, report that the reasoning intervention maintained its effect on its targeted ability, with an effect size of 0.23 (99 percent interval 0.09 to 0.38), and that participants in each intervention group reported less difficulty with instrumental activities of daily living (reasoning, effect size 0.38, 99 percent interval 0.02 to 0.74). The mean baseline age was 73.6.

How far do these studies carry? Three limits. The participants were children in one meta-analysis and adults aged 65 and over in the other, so neither is a sample of job candidates. The outcomes concern the targeted ability as each trial defines it, so the results show that the targeted skill can improve and be maintained, and they leave open how far it spreads to other abilities. And a training gain over weeks is a different thing from a practice effect over two sittings, which is what the retest studies measure. The page on whether you can increase your IQ takes these questions up across abilities.

9 How Should You Handle the Sitting Itself?

The conditions of an unsupervised, adaptive, 18 minute sitting are mostly practical, and the first job is to learn the rules before you start: SHL's fact sheet describes one sitting that can be taken on any device, and an employer may allow no second attempt. The fact sheet says the Verify Interactive suite is built with mobile first technology and can be deployed on any device type, including mobile phones, and that the inductive test is designed for an unsupervised environment. The UBS page shows how much an employer can add on top: a seven day window and no retake. Those two facts together mean that the invitation, not the test, tells you what you can do if something goes wrong.

The Standards give a candidate a useful list of what to expect to be told. Standard 8.2 says test takers should be provided in advance with as much information as is consistent with valid responses about the test, the testing process, the intended test use, scoring criteria, testing policy, availability of accommodations and confidentiality protection. Its comment adds that test takers should be made aware of any imposed time limits, that for computer administrations they should be shown samples of the interface and given an opportunity to practice with it, and that they should usually be told whether and when they will have access to their results. Standard 6.5 adds that test takers should be given appropriate instructions, practice and other support necessary to reduce construct irrelevant variance, so not knowing how a drag and drop screen works should not cost you points. The practical reading is to ask the employer for the publisher, the time limit, the number of attempts, the accommodation process and what happens to your result, and to use the publisher's own practice version to see the interface, bearing in mind SHL's warning that its difficulty may differ.

Our own, unsourced suggestions are modest: a device and a connection you trust, a place without interruptions and a time when you are not tired, because 18 minutes is short. Speed of working is built into the sitting by the time limit, and processing speed is a separate CHC ability with its own measures; the page on the processing speed test covers that ability on its own. The comment to Standard 6.10 says test users are primarily responsible for avoiding misinterpretation of scores, and the next section says what a result can and cannot carry.

10 How Do You Read an Inductive Reasoning Score?

A score on an inductive test means little until you know whom you are compared with, how precise the score is and what it is built from, and those same three questions apply to every cognitive score you meet. On the first, SHL's fact sheet is silent on the comparison group, so we cannot tell you what a result is set against. In general, the group matters because a score is a position relative to a reference group: the page on how IQ scores are normed explains how norm groups are built, the page on IQ score versus percentile explains why a standard score and a percentile are different statements, and the explainer on standard deviation 15 shows what a point on that scale means.

On precision, the Standards for Educational and Psychological Testing (AERA, APA and NCME, 2014) are direct. Standard 2.3 says that for each total score, subscore or combination of scores that is to be interpreted, estimates of relevant indices of reliability and precision should be reported, and its comment warns that total scores can be acceptably consistent while subscores have unacceptably low reliability. Standard 6.10 says that when score information is released, the interpretation should describe in simple language what the test covers, what scores represent, the precision of the scores and how they are intended to be used, and its comment notes that precision might be depicted by error bands or likely score ranges. For a short test of one narrow ability, the implication is that a single number is a point inside a band, and the band is wider than the number suggests. The page on reliability and validity explains what a reliability coefficient does and does not say about one person.

The third question is what the score is built from. A narrow task samples one ability in one format, and a battery samples several. The ACIS pages on its subtests say the same: one subtest should never be read as the whole construct, and the pattern across tasks is usually more informative than any isolated score. That is also why we hold this page to the claim in its title. An inductive reasoning test samples induction, an employer reads that sample for its own purpose, and an adult who wants a picture of their own reasoning needs a wider set of scores than one 18 minute task can give.

11 How Can an Adult See Their Own Inductive Reasoning?

An adult who wants to see their own inductive reasoning on adult norms can look at Matrix Reasoning, which is in every ACIS form, and at the Inductive Reasoning Index, which the technical manual builds from Visual Number Series, Matrix Reasoning and Complex Relations and which is reported on the Full Scale form. The manual says the index "focuses on Induction through Visual Number Series, Matrix Reasoning, and Complex Relations" and that it is useful when the interpretive question concerns rule inference and abstraction across novel stimuli. Its form matrix lists the index under the Full Scale form only. All three subtests also feed the Fluid Reasoning Index, together with Figure Weights and Logic Grid, which is the broader fluid reasoning picture.

The forms, with prices read on October 29, 2026 that can change: Quick costs 15 dollars, has six subtests in three domains including Matrix Reasoning and takes about 45 minutes; Optimized costs 30 dollars, has 13 subtests in five domains including Matrix Reasoning, Figure Weights and Logic Grid and takes about 110 minutes; Full Scale costs 50 dollars, has all 20 subtests in six domains and takes about 175 minutes. Visual Number Series and Complex Relations are only in the Full Scale form. Custom Test lets a buyer pick any subtests at 5 dollars each, and every subtest returns its scaled score. Each form reports a Full Scale IQ with its 95 percent confidence interval, adult norms cover ages 16 to 90, breaks are allowed, progress is saved between subtests and a buyer has 30 days to finish. The free trial needs no card and opens a different set of subtests in the Full Scale form, so it is a way to see the format and the report, not a way to get an induction score.

Scores arrive on standard scales: indices and the Full Scale IQ with a mean of 100 and a standard deviation of 15, and subtest scaled scores from 1 to 19 with a mean of 10 and a standard deviation of 3, together with percentiles. If you want to compare your induction with your other abilities, the page on index score differences explains how to read a gap between two indices. If you are choosing where to start, the page on IQ tests for adults and the page on the cognitive ability test are the entry points.

The honest framing is narrow. The formats differ from an employer's: the manual describes Matrix Reasoning as choosing from image options, Visual Number Series as typing a missing value and Complex Relations as supplying two missing terms of an analogy. ACIS is not SHL's test, uses different items, a different interface and different norms, and nothing in this page claims that taking it changes, predicts or rehearses an employer score. What it offers is an adult normed look at induction, as a person's own information.

12 Where ACIS Sits, and What a Buyer Should Do

ACIS is an online, unsupervised, English language battery that can show an adult their own induction on adult norms, and it is not an employer test, a clinical instrument or a predictor of either. The research page on the site states the limit literally: ACIS "is not intended for clinical diagnosis, hiring decisions, accommodation determinations or admission to high IQ societies" (read on October 29, 2026). It is also not a preparation for any employer test, and this page does not suggest giving an ACIS report to an employer. The technical manual says verbal and knowledge tasks are affected by English proficiency and educational exposure, and all instructions and items are in English. Those are limits of the instrument, not footnotes.

The Standards for Educational and Psychological Testing (AERA, APA and NCME, 2014) explain why the limit is right. Standard 1.1 says no test permits interpretations that are valid for all purposes or in all situations and that each recommended interpretation for a given use requires validation. The APA Guidelines for Psychological Assessment and Evaluation (2020), whose text we read, say validation evidence presented by developers primarily addresses the stated purposes and the cohorts like the standardization sample, and that when a test is used for purposes that differ from those the developer investigated, psychologists are encouraged to evaluate the validity evidence for that specific use. An ACIS score is evidence about your standing on ACIS tasks against adult norms. It is not evidence about how an employer's adaptive task scored you, and using it that way is exactly the unvalidated use those documents warn about. Standard 2.3 applies as well: each score you interpret needs its own precision, which is why a report gives intervals and why one subtest is read with more caution than a battery.

What strengths remain? Adult norms from 16 to 90, a Full Scale IQ with a 95 percent confidence interval, six indices, a one time payment with no subscription, breaks, saved progress and 30 days to finish, a free trial with no card and example reports you can open before paying. The refund promise is conditional and the site words it so: "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."

What should a buyer do? If you have an invitation from an employer, read it first: the publisher, the time limit, the number of attempts, the accommodation process and what happens to your result. Use the publisher's own practice version to see the interface, and treat everything else as optional. If you want your own profile of induction and the abilities around it, start with the free trial or an example report to see what a result looks like, then choose the form that covers what you want to know: Matrix Reasoning is in all three, and the Inductive Reasoning Index needs the Full Scale form. Compare it with the other routes in the page on where to take an IQ test before you decide, and read the technical manual for the documentation behind the scores.

13 Sources Behind This Page

Every figure above traces to one of the following, and each is linked at the point where it is used. Statements marked as our arithmetic are calculations on published figures and are not attributed to the authors. Pages from publishers and employers were read on October 29, 2026.

• American Educational Research Association, American Psychological Association and National Council on Measurement in Education. Standards for Educational and Psychological Testing. American Educational Research Association, 2014. Standards 1.1, 1.7, 2.3, 6.5, 6.10, 8.1, 8.2 and 11.2. Open access edition.
• American Psychological Association, APA Task Force on Psychological Assessment and Evaluation Guidelines. APA Guidelines for Psychological Assessment and Evaluation. American Psychological Association, approved March 2020. Guidelines PDF.
• Akhtar H. Measuring fluid reasoning and its cultural issues: A review in the Indonesian context. Buletin Psikologi, 2022, volume 30, issue 2, pages 276 to 288. DOI 10.22146/buletinpsikologi.74475.
• Ball K, Berch DB, Helmers KF, and colleagues. Effects of cognitive training interventions with older adults: A randomized controlled trial. JAMA, 2002, volume 288, issue 18, pages 2271 to 2281. DOI 10.1001/jama.288.18.2271.
• Carpenter PA, Just MA, Shell P. What one intelligence test measures: A theoretical account of the processing in the Raven Progressive Matrices Test. Psychological Review, 1990, volume 97, issue 3, pages 404 to 431. DOI 10.1037/0033-295X.97.3.404.
• Klauer KJ, Phye GD. Inductive reasoning: A training approach. Review of Educational Research, 2008, volume 78, issue 1, pages 85 to 123. DOI 10.3102/0034654307313402.
• Lubis AD, Maulina D. Development of an Indonesian figural-inductive reasoning test for high school students based on the Cattell-Horn-Carroll theory. Proceedings of the 1st International Conference on Intervention and Applied Psychology (ICIAP 2017), Advances in Social Science, Education and Humanities Research, volume 135, Atlantis Press, 2018. DOI 10.2991/iciap-17.2018.3.
• Pearson. Wechsler Adult Intelligence Scale, Fifth Edition (WAIS-5), product page. Pearson Assessments, 2024 edition. Product page.
• Rebok GW, Ball K, Guey LT, and colleagues. Ten-year effects of the Advanced Cognitive Training for Independent and Vital Elderly cognitive training trial on cognition and everyday functioning in older adults. Journal of the American Geriatrics Society, 2014, volume 62, issue 1, pages 16 to 24. DOI 10.1111/jgs.12607.
• Sackett PR, Zhang C, Berry CM, Lievens F. Revisiting meta-analytic estimates of validity in personnel selection: Addressing systematic overcorrection for restriction of range. Journal of Applied Psychology, 2022, volume 107, issue 11, pages 2040 to 2068. DOI 10.1037/apl0000994.
• Scharfen J, Peters JM, Holling H. Retest effects in cognitive ability tests: A meta-analysis. Intelligence, 2018, volume 67, pages 44 to 66. DOI 10.1016/j.intell.2018.01.003.
• SHL. SHL Verify Interactive Inductive Reasoning, Assessment Fact Sheet. SHL and its affiliates, 2022. Fact sheet PDF.
• SHL. Cognitive ability tests, product page. SHL cognitive assessments.
• SHL. Practice tests and assessments, SHL Direct candidate page. SHL practice page.
• UBS. Graduate Talent Program, early careers page. UBS page.

14 Frequently Asked Questions