Untimed IQ test: what taking away the clock really changes in a score

0 The short answer

A fully untimed, norm referenced IQ test is rare, because time limits are normally part of the standardized procedure that the norms were built on, but tests with generous or no limits on reasoning tasks do exist, and the right choice depends on whether you want to measure how well you can solve a hard problem or how quickly you can solve it. The real options are a psychologist giving a Wechsler battery, a matrix test given untimed in research, free untimed online screeners of 30 to 33 questions, and a timed normed online battery such as ACIS, which reports a Full Scale IQ and up to six indices for 15, 30 or 50 dollars one time (prices read on October 29, 2026) in about 45, 110 or 175 minutes. The decisive limit is that ACIS is not an untimed test: its subtest pages state per item clocks, section limits or fixed speed windows, and it is online, unsupervised, not clinical and in English only.

1 What Does an Untimed IQ Test Mean, and Does a Truly Untimed One Exist?

An untimed IQ test can mean three different things, a session with no clock at all, a test with limits so generous that almost nobody runs out of time, or a test that sets item cutoffs but gives no credit for speed, and only the first is untimed in the strict sense. The wording matters because the three are treated differently in measurement theory. The Standards for Educational and Psychological Testing, the joint document of the American Educational Research Association, the American Psychological Association and the National Council on Measurement in Education, defines a timed test as one given to test takers who are allotted a prescribed amount of time to respond. It defines speededness as the extent to which scores depend on the rate at which work is performed as well as on the correctness of the responses, and it adds that the term is not used to describe tests of speed (Standards for Educational and Psychological Testing, 2014, glossary). A test can therefore be timed and still be only slightly speeded.

The older and more useful pair of terms is speed test and power test. The APA Dictionary of Psychology describes a power test as one intended to calculate the participant's level of mastery of a topic under conditions of little or no time pressure, with items that become progressively more difficult (APA Dictionary of Psychology, power test). Kyllonen and Zu, in a review of response time in ability measurement, describe the same split as two performance dimensions, speed and level. Speed has typically been measured with simple speeded tests where the outcome is the number correct in a given time, and level with complex power tests where the outcome is the number correct under liberal or no time limits (Kyllonen and Zu, 2016, Journal of Intelligence). The page on why IQ tests are timed works through what the clock contributes to each format, so this page does not repeat that; it asks the next question, which is what an adult should do with the distinction when choosing a test.

Practical definitions of "unspeeded" exist, and they show how far an untimed test is from a literal absence of a clock. Kyllonen and Zu report that one definition treats a test as completely unspeeded when no test taker's standing would change if more time were given, and that Educational Testing Service policy has considered a form of a major standardized test unspeeded if all test takers finish at least 75 percent of the items and at least 80 percent respond to all items. Oliver Wilhelm and Ralf Schulze give a different threshold in their study of speeded and unspeeded reasoning: time constraints supposedly introduce speededness when less than 90 percent of a sample completes all items (Wilhelm and Schulze, 2002, Intelligence). Neither number is a rule for every test; they show that "untimed" in practice is a matter of degree.

Our own way to sort the claims you will meet is this. A test that states no time limit and no session cap is untimed in the strict sense. A test that states a generous section limit, or an automatic advance after a long interval, is a power test with a safety valve. A test that counts the number of items finished in a fixed window is a speed test, whatever else it is called. Reading which of the three a product is, from its own instructions, is the first step in comparing it with anything else.

2 Which IQ Tests and Subtests Are Timed, and Why?

Most reasoning tests are given under some time constraint, Pearson's table for the fourth edition Wechsler adult scale names a stopwatch or item time limit for Block Design, Visual Puzzles, Figure Weights and Arithmetic, and the reasons are as much about standardization and efficiency as about measuring speed. Wilhelm and Schulze open their paper by noting that most reasoning tests are administered under some type of time constraint, and they give two reasons. The first is that test performance is conceptualized as output per time unit, so both the time required and the accuracy of cognitive processes are of interest. The second is that some limit is required for pragmatic purposes in group testing so that administration proceeds efficiently.

For a named battery, the most direct public document is Pearson's own telepractice table for the Wechsler Adult Intelligence Scale, Fourth Edition. For Block Design it tells the examiner to use a stopwatch. For Visual Puzzles, Figure Weights and Picture Completion it says the examiner uses a stopwatch and must ensure the examinee stops at the item time limit, and it gives the same instruction for Arithmetic (Pearson, WAIS-IV telepractice considerations, Table 1). The same table lists no stopwatch for Similarities, Vocabulary, Information, Comprehension or Matrix Reasoning. That is a statement about what the table says and not about every rule in the manual, and it concerns the fourth edition. Pearson's comparison flyer for the fifth edition (2024) lists a 45 minute administration for the seven subtest Full Scale IQ and about 60 minutes for the ten primary index subtests, against 70 minutes for the core subtests of the fourth edition, and introduces a processing speed subtest called Naming Speed Quantity (Pearson, WAIS-5 comparison flyer). We did not open the fifth edition administration manual, so this page makes no claim about which fifth edition subtests carry item limits; the page on the WAIS-5 subtests describes the battery.

The Standards explain why a time limit is not an arbitrary add on. In their discussion of standardization they list uniform directions, specified time limits, specified room arrangements, use of proctors and consistent security procedures as the conditions implemented so that differences in administration will not inadvertently influence the performance of some test takers relative to others. A time limit is a controlled condition in the same sense as the wording of the instructions. If the norm group had a limit and you do not, your raw score is not being compared with theirs under the same conditions, which is the central problem for any untimed score read against timed norms.

There is also a measurement reason that is easy to miss. Standard 4.14 says that for a test that has a time limit, test development research should examine the degree to which scores include a speed component and should evaluate the appropriateness of that component given the domain the test is designed to measure. Its comment adds that, where speed is not a meaningful part of the target construct, time limits should be set so that examinees have adequate time to demonstrate the targeted knowledge and skill. In other words, a responsible publisher is expected to ask whether the clock belongs in a given task. That expectation is why a vocabulary task and a symbol matching task can sit in one battery with very different time rules, and why the page on how long an IQ test takes reports session length by instrument.

3 Which Options Does an Adult Have, and How Do They Compare?

There are four realistic routes to a test with little or no time pressure, and they differ less in how long the clock runs than in whether the score is tied to a stated reference group. The table compares them using only documents we opened. Prices and facts were read on October 29, 2026 and can change. The two free sites are named as market facts, are not linked, and are described only by what their own pages state; we make no judgment about their accuracy.

Three points follow from the table. First, "untimed" appears only on the unsupervised free screeners, and those are also the options whose norm samples are least described on their own pages. That is a pattern in four rows, not a law. Second, the only supervised, documented-norm route, a Wechsler battery, has item limits on several subtests, so a person who wants no clock at all cannot get it from the most established instrument. Third, the page on where to take an IQ test lists every route in person and online, so this page stays on the timing question.

4 What Does a Time Limit Change in What a Reasoning Task Measures?

A time limit adds a speed component to a reasoning score, and the published studies we opened agree on that direction while differing in design, so the fair summary is that timed and untimed reasoning tests overlap strongly but are not interchangeable. Three studies bear on the question, and each is described here by its design because the design decides what it can show.

Wilhelm and Schulze argued that time constraints introduce a spurious mental speed contribution to reasoning performance, so that speeded reasoning measures are a compound of working memory capacity and mental speed. They tested 277 participants with 12 reasoning tasks, six of them time constrained, and 9 mental speed tasks, and 90 further participants worked on both batteries under timed conditions. Mental speed correlated higher with speeded than with unspeeded reasoning measures, and two speeded reasoning tests correlated higher with each other than a speeded and an unspeeded test did. The variance of a speeded reasoning factor could be explained completely by a linear function of unspeeded reasoning and mental speed. Their conclusion was that speeded reasoning tests are likely to lead to overestimates of the relation between mental speed and reasoning ability (Wilhelm and Schulze, 2002, Intelligence). Read narrowly, this is a finding about correlations among tests: it says the clock inflates the apparent tie between speed and reasoning, not that either kind of test is the better measure of intelligence.

Adam Chuderski compared working memory with fluid reasoning, written Gf in the Cattell-Horn-Carroll model that the page on the CHC model explains. A broad working memory factor built from four functions, storage capacity, attention control, relational integration and updating, shared 83.4 percent of variance with Gf tested under time pressure. A reanalysis of previous data with the same model showed only 58.2 percent shared with virtually untimed Gf tests. In the timed tests only the easiest, early items contributed to the working memory and Gf correlation, whereas in the untimed tests the hardest, late items were also linked with Gf. The author suggested that measuring "fast" intelligence primarily taps working memory functions, whereas "slow" intelligence also depends on other cognitive processes beyond working memory (Chuderski, 2015, Personality and Individual Differences). Note the design: the untimed figure comes from a reanalysis of earlier data, so the two percentages come from different samples and tests and do not show what one person's score would do if the clock were removed. A later corrigendum to the paper concerns the funding statement only.

Philip Vernon and Lida Kantor ran the cleanest experiment of the three. One hundred and thirteen high school students were randomly assigned to take an intelligence test, the Multidimensional Aptitude Battery, under timed or untimed conditions, and all then took eight reaction time tests. Combinations of reaction times predicted IQ scores about equally well in both groups, and the correlations of single reaction time tests with Verbal and Full Scale scores were about equal across conditions, but correlations with Performance scores were higher in the timed condition. The authors concluded that timed and untimed intelligence tests impose different information processing demands, but that the speed with which people cope with these demands is equally important in both conditions (Vernon and Kantor, 1986, Intelligence). That last clause is worth keeping: removing the clock does not remove speed from thinking, it removes the penalty for taking the time you need.

Taken together, these studies support two plain statements. In the samples studied, a clock raised the weight of working memory and mental speed in reasoning scores, and in the timed condition of the Chuderski data the link to working memory came from the easier items. The effect is a matter of degree, not a switch. The page on fluid intelligence tests covers what Gf tasks measure in general, and it places Gf next to the other broad abilities.

5 Does Taking Away the Clock Raise Your Score?

No source we opened gives a general number of IQ points gained by removing a time limit, so any claim of the form "you will score ten points higher untimed" is unsupported, and the more useful question is whether the score you get can still be read against the norms. The studies above compare correlations and shared variance, and the abstract of the one randomized experiment reports how reaction time predicts IQ under each condition, not the average score difference between conditions. One definition of an unspeeded test that Kyllonen and Zu review is also framed in terms of standing, whether anyone's place in the group would change with more time, and not in terms of raw points.

There is a reason to expect raw scores to rise when a clock that cuts people off is removed: someone who would have run out of time now answers more items. That is our reasoning, not a finding. What it does not tell you is whether the resulting number is comparable. A raw score is converted to a standard score through a table, and the table is valid for the conditions under which the norm group was tested. Add time that the norm group did not have and the same table can over-credit the result; remove time the norm group did have and it can under-credit it. This is the same logic that sits behind the page on how IQ scores are normed.

A second study speaks to how well a shortened time condition tracks an untimed one. Hamel and Schmittmann compared a 20 minute timed version of the Raven Advanced Progressive Matrices with the untimed version in first year psychology students. The test normally takes 40 to 60 minutes, and the authors note that it can be administered as a 30 or 40 minute timed test or in short versions. They concluded that the 20 minute timed version proved to be an adequate predictor of the untimed score (Hamel and Schmittmann, 2006, Educational and Psychological Measurement). "Predictor" is the operative word. It means the two sets of scores were related closely enough that one forecast the other in that sample. It does not mean the two sets of scores were equal, and it does not mean a person scored the same number of points on each. For the matrix format itself, the page on Raven's 2 describes the commercial successor to the Raven tests.

If you are weighing a retest on a different timing condition, treat it as a retake of a different procedure, not as a repeat. The page on retaking an IQ test covers what a second attempt can and cannot show, and the same caution applies with more force when the second attempt also changes the clock.

6 How Are Untimed Reasoning Tests Normed?

A norm referenced score is only as comparable as the match between the way you took the test and the way the norm group took it, so an untimed test needs a norm sample that was also tested without a clock, or at least a published rule for converting between conditions. The American Psychological Association describes norm referenced tests as standardized so that test takers are evaluated in a similar way no matter where they live or who administers the test (APA, Understanding psychological testing and assessment). The Standards turn that into duties. Standard 4.15 says the directions for administration should be clear enough that others can replicate the administration conditions under which the data on reliability, validity and, where appropriate, norms were obtained, and that allowable variations should be described. Standard 6.1 says test administrators should follow carefully the standardized procedures specified by the test developer. A time limit, or its absence, is one of those procedures.

What this means for the free untimed screeners in the table is easy to state and hard to check. The page of one 30 question screener describes the IQ scale in general terms but, as we read it on October 29, 2026, describes no norm sample for its own test. The page of a 33 question screener says its score is placed on a scale with mean 100 and standard deviation 15 using WAIS-IV bands and compared with people of the same age and country, and in the same text says the reference point is fixed and not re-estimated from the people who take the test; the page does not specify the normative sample. Those are statements about what the pages say. They do not prove that the scores are wrong, and they do not show that they are right. They show what a reader can and cannot check, which is the point of a norm sample. The page on the standard deviation of 15 explains why the scale itself is only a convention that any test can print, and why the sample behind it is what gives the number meaning.

For the same reason, matrix tests given untimed in research are not a shortcut to an untimed IQ. Hamel and Schmittmann's design compares administrations in a university sample, which shows how the timed and untimed procedures relate but is not a general adult norm table. Where an untimed administration is not part of a published manual, the safest reading of an untimed number is as a raw performance level on that test, not as an IQ on the standard scale.

A practical checklist follows from this, framed as questions to put to any untimed product before relying on its number. Does it state a reference group and the date and size of that sample? Does it say whether the group took the test with or without a clock? Does it print an interval around the score, not a single figure? Does it say what scoring rule produced the raw score? A product that answers all four in its own documentation is giving you something you can evaluate. One that answers none is giving you a sorting game. The page on reliability and validity explains why the third question matters.

7 Which Abilities Depend on the Clock? The Construct, Precisely

Processing speed (Gs) is the one broad ability in which the clock is the construct, while fluid reasoning (Gf), visual processing (Gv), quantitative knowledge (Gq) and comprehension knowledge (Gc) are abilities of level that a clock can contaminate to different degrees. Kyllonen and Zu review how the Cattell-Horn-Carroll structure treats this. They report that separate speed and level factors of fluid ability, crystallized ability and visual perception ability have been identified, and that a separate broad speediness factor, Gs, is measured by numerical facility, perceptual speed and rate of test taking tasks. In plainer terms, you can be fast or slow at a reasoning task and strong or weak at its hardest items, and those are partly separate things.

The ACIS subtest pages and technical manual say where speed is meant to count, and they are worth reading before choosing a form. The two processing speed subtests, Symbol Search and Coding, are timed tasks in which the number of correct responses in a fixed window is the score, and the page on the processing speed test covers those tasks and how an index is read from them, while the page on processing speed as an IQ domain covers the ability itself. In the other domains the clock is a pacing device, and the pages say so in different ways. The Arithmetic page lists working memory load, slow calculation, limited practice and misunderstanding the verbal problem among the reasons for a weaker result, which means speed of calculation is one of several things the score can reflect. The technical manual says Layer Rotation is not a processing speed task, because the per item limit controls pacing while the dominant demand is figural transformation, and it adds that the speed of the transformation still contributes to the score. Its description of Visual Puzzles says the task can be affected by speed, visual scanning and strategy.

Presentation matters as much as the clock. Digit Span and Alphanumeric Sequencing present items as audio at a set pace, one number or letter per second according to their pages, so the pace of presentation is fixed by the recording and not by the person. Arithmetic is also audio. That is a different kind of time pressure from a countdown over a visual item.

The consequence for a reader who dislikes time pressure is specific. In the studies above, fluid reasoning is the ability where timed and untimed versions were compared and found to differ in what they tap, working memory is the ability most entangled with the clock in the Chuderski comparison, and processing speed is the ability a clock free test cannot measure at all. A single total score blends them. A profile that separates them is more informative for this question, and the page on index score differences shows how common sizable gaps among indices are in ordinary people. For adults whose reasoning is high and whose speed is lower, the page on the twice exceptional test for adults covers testing for adults whose profile is uneven.

8 How Is Each ACIS Subtest Timed? A Plain Answer

ACIS is not an untimed test: 17 of its 20 subtest pages state a clock of some kind, and the other three state no time rule, which is not the same as having none. The count is our arithmetic on the pages as we read them on October 29, 2026. The table gives each rule in the words of the subtest's own page, and says "not stated" where the page is silent. We describe timing only as the page does; anything else about a subtest, including its item content, is not disclosed.

Sorted by kind of rule, which is again our reading, the table shows three patterns. Nine subtests carry a clock on every item: Vocabulary, Information, Similarities, Matrix Reasoning, Figure Weights, Visual Number Series, Complex Relations, Visual Puzzles and Layer Rotation. Five set one limit for a whole section, which leaves the person free to spend time unevenly across items: Antonyms, Synonyms, Logic Grid, Mathematical Achievement and Spatial Comprehension. Two are fixed windows in which the count of correct responses is the score, Symbol Search and Coding. One, Arithmetic, mentions a per item timer in its second section. Three state no time limit on their pages, and two of those present items at a fixed audio pace of one per second.

The pages for the subtests this article discusses most are worth opening before paying. Matrix Reasoning is the fluid reasoning matrix task with the per item clock and the shared extra time bank. Figure Weights is a balance scale task with a 45 second item clock. Visual Puzzles uses two clock lengths, a shorter one for the first seven items. Layer Rotation carries no words at all, according to the technical manual, and is timed per item. Spatial Comprehension is the one visual spatial subtest with a single 25 minute limit for its whole length.

The forms follow from the table. The Quick form, 15 dollars, has six subtests and no processing speed subtest, but Matrix Reasoning, Figure Weights, Vocabulary and Similarities each carry a per item clock. The Optimized form, 30 dollars, adds Antonyms and Synonyms, Logic Grid, Spatial Comprehension, Layer Rotation, Coding and Symbol Search. The Full Scale form, 50 dollars, has all 20. Breaks are allowed between subtests, progress is saved, and you have 30 days to complete a form, so the session as a whole is not under a clock even though each subtest is.

9 How Can an Adult Who Wants Less Time Pressure Choose a Test?

An adult who wants less time pressure should decide first whether the goal is to measure reasoning level or reasoning speed, then read each candidate's time rule before paying, and then insist that the score be compared with a group that took the test the same way. Five checks cover most of the decision, and they apply to any test, not only to ACIS.

The first check is the goal. If the question is how hard a problem you can solve given the time you need, a power style task such as a long section limit or an open matrix test fits it better than a per item countdown. If the question is how efficiently you handle simple tasks, a timed processing speed measure fits it better, and the processing speed pages linked earlier are the right place for that. Treating one as the other is the commonest error in this topic. The second check is the actual rule. Open the instructions or the subtest page and write down whether the limit is per item, per section or a fixed window. The table above shows how different those three are inside one battery.

The third check is the reference group, as the norming section set out. A free untimed screener can be a pleasant way to see how you handle reasoning problems at your own pace, and the page on IQ tests with no sign up covers tests that need no registration, but a number from it should be read as an estimate unless its page documents a norm sample. The fourth check is whether the report separates abilities. A report with an index per domain lets a slower pace show up where it belongs, in the speed index, instead of leaking into every score. ACIS also lists a General Ability Index among its composites, which the technical manual describes as emphasizing higher order reasoning and knowledge while reducing dependence on working memory and processing speed; the page on the General Ability Index explains when that composite is used and when the Full Scale IQ answers the question better. The fifth check is not to mix numbers. A timed score and an untimed score do not average, and the page on IQ score versus percentile explains why even two scores on the same scale cannot be averaged casually.

A reader who wants to choose subtests rather than take a preset form has one more option. The Custom Test costs 5 dollars per subtest; any domain you complete returns its full index score and every subtest returns its scaled score, while the Full Scale IQ needs the Full Scale battery. Someone who only wants the verbal and reasoning indices could take those subtests alone and see the clock rules on each page first. That choice has a cost, which is that without the full battery there is no Full Scale IQ and the profile is partial.

Nearby decisions are covered elsewhere. The page on IQ tests for adults compares the main routes for a general adult reader, and the page on IQ tests that return results instantly covers tests that return a score immediately. If your interest is the top of the range, the page on high range IQ testing is the better starting point. Accommodations such as extra time for a disability are decided by qualified professionals and institutions; this page does not advise on them, and ACIS is not an instrument for them.

10 How Should You Read a Score From a Test With a Clock?

A score from a timed test is read the same way as any norm referenced score, as a scaled score, an index and a percentile with an interval around it, but the interval and the comparison group both assume the timing was the same as in the norm sample. ACIS reports indices and the Full Scale IQ on the standard scale with a mean of 100 and a standard deviation of 15, scaled subtest scores from 1 to 19 with a mean of 10 and a standard deviation of 3, percentiles and a 95 percent confidence interval.

Precision depends on the procedure. The Standards, in their chapter on reliability and precision, say that reliability applies to a particular assessment procedure and is likely to change if the procedure is changed in any substantial way, that shortening an assessment is likely to lower reliability and that lengthening it with comparable tasks is likely to raise it. They also note that the standard error of measurement can be used to generate confidence intervals around reported scores. Changing a time limit changes the procedure, so a confidence interval published for the timed version does not automatically carry over to an untimed one. The same chapter observes that when a test is designed to reflect rate of work, internal consistency estimates, particularly by the odd even method, are likely to yield inflated estimates of reliability for highly speeded tests.

For a single subtest, treat the scaled score as a coarse reading. One task samples one narrow ability, and a lower scaled score on a speeded task can reflect speed, attention, device or motivation as well as the ability named. An index built from several subtests is steadier, and the Full Scale IQ is steadiest, which is the logic of the Standards passage on lengthening an assessment. For the reader who is asking about time pressure, the useful comparison is within the report: set the speed index beside the reasoning indices and see whether the gap is large, and then check how often gaps of that size occur in ordinary people before reading anything into yours. The technical manual, linked at the ACIS technical manual, is the place for the documentation behind ACIS scores.

11 What Practical Conditions Matter in an Unsupervised Timed Session?

In an unsupervised session you control the conditions that an examiner would otherwise control, and the clock makes some of them matter more: the device, the sound, the room, the breaks and what you do when a countdown runs out. The technical manual states the general point, that speeded tasks can be affected by input latency, screen size, visual clarity, distraction, fatigue and motivation, and that working memory tasks can be affected by audio conditions, attention and rehearsal strategy.

Concretely, use a device whose screen shows the whole task and whose input does not lag. The Symbol Search and Coding pages ask for landscape orientation on a phone, and Coding stays locked horizontally on mobile so the full workspace fits on screen. Make sure sound works before you begin: Digit Span, Alphanumeric Sequencing and Arithmetic present their items as audio and cannot be taken without working sound. Choose a quiet room and silence notifications, because the subtest pages describe countdowns that run while you work. Read the aids rule on each page: most subtest pages state that no external aids such as calculators, notes, scratch paper or drawing tools are allowed, while the Mathematical Achievement page allows pen and scratch paper and a reference table of formulas, with no calculators.

Take advantage of the structure the platform gives you. Practice items come first on several subtests and are not scored, and the Symbol Search, Coding and Logic Grid pages say the clock starts after them. Breaks are allowed between subtests and progress is saved, so you can split a long form across sessions within the 30 days. Because the clock rules apply inside each subtest as its page states them, decide to rest before you press Begin on the next one. If a per item countdown runs out on an item you could have solved, the sensible response is to move on, not to carry the frustration into the next item, because the next item has its own clock.

None of this makes an unsupervised session equivalent to a supervised one: a psychologist can see that you were distracted and can stop a session, and a browser cannot. The page on the free IQ test covers what the free trial offers before you decide on a form.

12 Where ACIS sits, and what a buyer should do

ACIS is a timed, normed, online battery that gives an adult a Full Scale IQ and up to six indices with a 95 percent confidence interval, and it is the wrong choice for someone who needs an untimed administration, a supervised session, a clinical conclusion or a score for an institution. We sell ACIS, so treat this section as a disclosure and check it against the rest of the page.

The strengths are specific. Every form reports a Full Scale IQ with its interval, and the Full Scale form measures all 20 subtests in six CHC domains. Prices are one time, 15, 30 or 50 dollars for the Quick, Optimized and Full Scale forms as read on October 29, 2026, with about 45, 110 or 175 minutes of testing, breaks between subtests and 30 days to finish. A free trial with no card is available once per person. 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." It is a conditional guarantee and not an unconditional money back promise.

The limits are equally plain. ACIS is online and unsupervised. It is not a clinical or diagnostic instrument and is not for hiring, school accommodations or admission to high IQ societies. It is in English only, and the technical manual says verbal and knowledge tasks are affected by English proficiency and educational exposure. It is timed: the table above lists each subtest's rule, and no form of ACIS is an untimed test. The Standards for Educational and Psychological Testing (AERA, APA and NCME, 2014) are the reference for reading any score. Their standard on administration says to follow the developer's standardized procedures, which for any timed test includes its clocks, and their standard on speed says a test should be examined for how far its scores reflect speed and whether that is appropriate to the domain measured. The APA's public guidance adds that norm referenced tests are standardized so that people are evaluated in a similar way, and that licensed clinical psychologists are the professionals trained to administer assessments and interpret the results. If you need a diagnosis or a decision, that is the route.

The next step depends on the reader. If you want to see how you handle the tasks at their stated pace without paying, start the free trial with no card and open an example report. If you want less pressure than a per item countdown allows, read the table and consider the Custom Test, accepting that the Full Scale IQ needs the full battery. If you want an untimed score on the standard scale with a documented norm group, the honest answer from this page is that we did not find such a product, and that a supervised Wechsler battery is the documented route even though several of its subtests carry item limits. Whatever you pick, compare numbers only with numbers that came from the same procedure.

13 Sources Behind This Page

Every figure above is traceable to one of the following, and each is linked at the point where it is used, except the two free screeners, which are named in the comparison table and described only by what their own pages state on October 29, 2026. ACIS prices, forms and timing were read on the live ACIS site and technical manual on October 29, 2026 and can change. The counts of subtests by kind of time rule are our arithmetic on the subtest pages.

• 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 (open access edition; glossary, chapter 2, standards 4.14, 4.15 and 6.1).
• American Psychological Association. Understanding psychological testing and assessment. apa.org, page last updated August 18, 2022.
• American Psychological Association. APA Dictionary of Psychology, entry "power test". apa.org, entry updated April 19, 2018.
• Chuderski A. The broad factor of working memory is virtually isomorphic to fluid intelligence tested under time pressure. Personality and Individual Differences, 2015, volume 85, pages 98 to 104. A corrigendum (2017) adds the missing funding information.
• Hamel R, Schmittmann V. The 20-Minute Version as a Predictor of the Raven Advanced Progressive Matrices Test. Educational and Psychological Measurement, 2006, volume 66, issue 6, pages 1039 to 1046.
• Kyllonen P, Zu J. Use of Response Time for Measuring Cognitive Ability. Journal of Intelligence, 2016, volume 4, issue 4, article 14.
• Pearson. Specific telepractice considerations by WAIS-IV subtest (Table 1). Pearson Clinical, read October 29, 2026.
• Pearson. WAIS-IV vs. WAIS-5: what is new and improved (US version flyer). Pearson Clinical, read October 29, 2026.
• Vernon P, Kantor L. Reaction time correlations with intelligence test scores obtained under either timed or untimed conditions. Intelligence, 1986, volume 10, issue 4, pages 315 to 330.
• Wilhelm O, Schulze R. The relation of speeded and unspeeded reasoning with mental speed. Intelligence, 2002, volume 30, issue 6, pages 537 to 554.

14 Frequently Asked Questions