Jordan Peterson has publicly discussed his own ability profile and is professionally familiar with psychometrics. The reported IQ around 150 is still a self report unless a named battery, score sheet, date and normative context are available for independent review.
A self reported estimate can be discussed, but it is not equivalent to a published score report.
0 The Short Answer
There is no published record of Jordan Peterson's IQ. Widely circulated clips and blog posts attribute to Peterson a statement that his measured IQ was in excess of 150, and that attribution now appears in hundreds of articles as though it were a citation. We could not locate a primary source for it: no named program, no episode, no date, no timestamp, no transcript. The attribution circulates without an origin, and that missing origin is the actual finding of this page.
What makes the case worth a full article rather than a one line dismissal is who the claim is about. Peterson is a clinical psychologist with a doctorate from McGill, taught in the Harvard psychology department, spent more than two decades as a professor at the University of Toronto, and co-authored peer reviewed research that administered standardized cognitive ability measures to real samples. He is not a bystander to psychometrics. He built instruments in it.
That is exactly why the case is instructive. It cleanly separates two things that readers constantly merge: expertise about how cognitive testing works, and a documented test record about oneself. The first is verifiable in Peterson's case, down to journal volumes and page numbers. The second does not exist in public, and no amount of the first can manufacture the second. A person who has published on the measurement of fluid and crystallized ability, speaking informally about a number they recall, is still producing a self-report. The rest of this page shows why a self-report is weak evidence even when it comes from someone who knows the field better than almost anyone repeating it.
Where ACIS stands
ACIS has never tested Jordan Peterson and has no connection to him or to any organization he is involved with. Everything here evaluates publicly available claims and published records. No figure on this page is an ACIS result. This page assesses evidence, not opinions, politics or personal conduct, and it does not speculate about anyone's health or private life.
1 Tracing the 150 Figure to Its Source
Run the standard provenance check on this number and the trail collapses faster than most. Search results split into three kinds of page: aggregator articles about celebrity intelligence, question and answer threads where users argue about the figure, and compilation videos with titles built around it. None of the three names an original broadcast. Each treats the number as already established, which is the signature of a claim that entered circulation without a first citation.
The most frequently referenced starting point is a November 2018 open thread on the blog Pumpkin Person devoted to Peterson's IQ. Read it directly and the claim arrives secondhand there too: the author states that a reader sent the information by email, describes the reader's account of Peterson saying his tested IQ was in excess of 150, and gestures toward a video link that does not appear in the post. So the most cited node in the chain is itself an unverified forwarded message. Every article downstream inherits that gap while presenting the number with increasing confidence, which is how folklore hardens.
Three features of the attribution should have slowed its spread. The wording is hedged rather than exact: "in excess of 150" is a boundary, not a score, and real score reports do not describe results that way. No instrument is named anywhere in the chain, and without a named battery and edition a three digit number has no scale attached to it. And no date appears, which matters more than people expect, since a score describes a person at a moment in their life rather than permanently.
Why did this particular number stick? Because Peterson talks about cognitive ability more than almost any other public intellectual, so a number attached to him feels topically appropriate in a way that a number attached to an actor does not. Both his supporters and his critics found the figure useful, one group as validation and the other as a target, and a claim that serves opposing camps gets repeated twice as often. The honest formulation is narrow and worth stating plainly: clips and posts attribute the statement to him, the attribution has no locatable primary source, and no test record has ever been published.
2 The Academic Record, Which Is Genuinely Verifiable
Unlike the IQ figure, Peterson's professional history leaves institutional footprints that anyone can follow. He completed a bachelor's degree in political science at the University of Alberta in 1982 and a second bachelor's degree in psychology at the same university in 1984. He earned a doctorate in clinical psychology at McGill University in 1991, supervised by Robert O. Pihl, with doctoral work on familial alcoholism and associated psychopathology. He then taught in the psychology department at Harvard University from 1993 to 1998, moved to the University of Toronto as a full professor in 1998, and remained there until retiring as professor emeritus. Institutional records like these are checkable in a way a number never is, which is the entire shape of the Dolph Lundgren case: a 1982 master's in chemical engineering from Sydney and a 1983 Fulbright to MIT have survived forty years of scrutiny, while the IQ figure bolted onto them has never produced a single document.
The publication record is where this page's argument actually lives, so it is worth naming specific papers rather than waving at a career:
DeYoung, Peterson and Higgins (2005), in Journal of Personality, volume 73, pages 825 to 858, studied the openness and intellect domain of personality using standard measures of cognitive ability alongside tasks tied to dorsolateral prefrontal function in a sample of 175 participants, and reported that both fluid and crystallized ability related positively to that domain and to no other personality trait.
DeYoung, Quilty and Peterson (2007), in Journal of Personality and Social Psychology, volume 93, pages 880 to 896, factor analyzed 75 facet scales from two major Big Five inventories in a community sample of 481 people, identified two aspects within each broad domain, and constructed and validated a 100 item instrument, the Big Five Aspect Scales, in a second sample of 480.
Morisano, Hirsh, Peterson, Pihl and Shore (2010), in Journal of Applied Psychology, volume 95, issue 2, pages 255 to 264, ran a randomized controlled goal setting intervention with 85 struggling university students and reported academic gains in the treatment group over four months.
Notice what that middle paper involves: item selection, factor structure, sample based validation and the construction of scales other researchers then used. Somebody who has done that work knows what a norm sample is, knows that a scale without published norms means nothing, and knows that instruments carry standard errors. This is precisely the background that makes an undocumented number attached to his name strange rather than routine, and it is the reason this article treats the case differently from a celebrity listicle entry.
3 Why Self-Reported IQ Is Weak Evidence, Even From a Psychologist
Start with the general case, because it is measurable. Freund and Kasten published a meta-analysis in Psychological Bulletin in 2012, volume 138, pages 296 to 321, on how well people's self-estimates of their own cognitive ability track their measured ability. The pooled correlation was about .34. That is a real relationship and clearly better than chance, and it is also far too loose to place any individual at a specific point on the scale. A correlation near .34 means self-estimates get the broad direction right and the magnitude badly wrong, in both directions, all the time. Even a screenshot beats a recollection: Joe Rogan's 127 at least points to a dated certificate from a named product, and Garry Kasparov's 135 to a supervised session a journalist watched happen.
Now the specific mechanisms that degrade an individual self-report, all of which apply regardless of the speaker's training:
The number usually comes from childhood. Most adults who believe they know their IQ learned it from a group administered school test decades earlier, often relayed by a parent or teacher rather than read off a report. Recall across thirty or forty years is reconstructive, and reconstruction of a flattering number drifts upward, not downward.
Older scales are not convertible. Ratio scoring, mental age divided by chronological age times 100, produced numbers with a different meaning and a different spread from modern deviation scoring at mean 100 and standard deviation 15. A childhood ratio figure of 150 does not translate into a modern 150. There is no clean conversion, and anyone treating the two as the same object is comparing units that were never the same.
The report is what gets dropped. A self-report transmits a single digit string and discards everything that gives the digit meaning: the instrument and edition, the administration date, the norm sample, the confidence interval, and the profile of index scores underneath the composite. What survives the retelling is the least informative part of the measurement.
Recall is selective by construction. People who took several tests over a lifetime tend to remember the best result. This requires no dishonesty at all. It is ordinary memory behaving the way memory behaves with self-relevant numbers.
Prior exposure inflates repeat performance. Clinicians train by administering and scoring cognitive batteries, which means seeing the items repeatedly. Familiarity with the materials raises scores on those materials, which is why professional testing tracks retest intervals in the first place.
Expertise does not neutralize a single one of those five. Understanding measurement theory improves the quality of a person's commentary about tests. It cannot retroactively produce a dated score report they never published, cannot repair a childhood memory, and cannot convert a ratio score into a deviation score. If anything the asymmetry cuts the other way: a psychometrically trained speaker is the one person capable of supplying the instrument, edition, date and index breakdown on request, so the continued absence of those details in every version of the claim is more informative than it would be coming from a novelist or an athlete.
There is a reason clinicians write reports instead of quoting numbers to each other. The report is the unit of evidence. The number is a summary of the report, and a summary detached from its document has lost the properties that made it interpretable. This is the single most useful idea to carry away from this page, and it applies to your own score just as much as to anyone famous. If you want to know what an IQ test actually measures, the answer is always found in a profile, never in one figure quoted from memory.
4 What a Score of 150 Would Actually Imply
Numbers this far out are easy to say and hard to picture, so here is the rarity ladder on the modern standard scale, mean 100 and standard deviation 15. A score of 150 sits three and a third standard deviations above the mean, which corresponds to roughly one person in 2,330. Rarity rises gently through the middle of the distribution and then accelerates hard, which is why claims in this zone deserve proportionally harder scrutiny than claims at 120.
IQ 100
The population median by construction. Half of the standardization sample scores above this point and half below.
IQ 120
Roughly the top 9 percent, about 1 person in 11. Common in graduate cohorts and in demanding professional work.
IQ 130
About the top 2 percent, roughly 1 in 44. The conventional cutoff for gifted identification in most school systems.
IQ 140
Roughly 1 in 261. Rare enough that a typical large workplace contains nobody at this level.
IQ 150
Roughly 1 in 2,330. Three and a third standard deviations up, and the figure attributed to Peterson in circulating clips.
IQ 160
Roughly 1 in 31,000, at or beyond the reporting ceiling of most professional adult batteries.
Two things follow from that ladder. First, 150 is genuinely exceptional but not physically implausible for a tenured research psychologist at a major university, which is what makes this claim harder to dismiss than the 160s and 170s handed out casually on celebrity lists. Roughly one in 2,330 describes a real population, and selective academic environments concentrate people from the upper tail. Second, plausibility is not evidence. A number can be entirely believable and still have no document behind it, and this one does not. Our percentile chart lays out the full conversion if you want to check any of these figures against your own arithmetic.
The distinction that matters here is not between a big number and a small one. It is between a claim that could be checked and a claim that cannot be. A documented 128 tells you something specific about one person's measured reasoning on a known date under known conditions. An undocumented 150 tells you only that a sentence has been repeated many times.
5 Uneven Profiles, and Why One Number Hides Them
People who describe their own cognition rarely describe a flat line. The common form of self-description is comparative: strong in one kind of thinking, weaker in another. That intuition is usually directionally sound, because index scatter is normal rather than exceptional. A modern battery reports separate composites for verbal comprehension, fluid reasoning, visual spatial processing, working memory, processing speed and quantitative reasoning, and a substantial share of people show gaps of a standard deviation or more between their highest and lowest index.
This is where composite scores mislead. A full scale figure is an average across domains, so two people with identical composites can have opposite profiles: one flat across every index, the other carrying a 30 point split between a very high verbal comprehension score and a mid range visual spatial or processing speed score. Those are different cognitive profiles with different practical consequences, and the composite erases the difference completely. When a battery reports large scatter, clinicians often treat the composite as less interpretable than the individual indices, which is the opposite of how single numbers get used in public conversation.
Two cautions belong here. The first is that a relative weakness inside a strong profile is still usually above average in absolute terms. Someone whose verbal comprehension sits far out in the tail and whose quantitative reasoning sits at the seventy fifth percentile has described a genuinely uneven profile, and both of those points are above the population median. Relative and absolute are different questions, and self-descriptions almost always answer the relative one while listeners hear the absolute one.
The second caution is the one this page has to observe about Peterson specifically. Secondary sources attribute various self-descriptions of an uneven profile to him, but those attributions trace back to the same chain of clips and forwarded accounts as the 150 figure itself, with no primary source we could locate. So this article makes no claim about which of his cognitive domains is stronger. That is not evasion, it is the same standard applied consistently: describing anyone's index profile requires the score report, and no score report exists here. If the general structure interests you, the split between fluid and crystallized ability is the cleanest place to see how two abilities can diverge sharply within the same person.
6 Admission Tests and Percentiles Are Not IQ Scores
A recurring move in arguments about famous people's intelligence is to convert an admission test result into an IQ figure. The move has a real empirical basis and an unreal application. Frey and Detterman published the standard reference in Psychological Science in 2004, volume 15, pages 373 to 378, reporting a correlation of .82 between SAT scores and a general ability factor extracted from the Armed Services Vocational Aptitude Battery in a sample of 917 participants from the National Longitudinal Survey of Youth, and .48 uncorrected against Raven's Advanced Progressive Matrices in 104 undergraduates. Admission tests clearly load on general ability. That much is settled.
The trouble starts at the individual level, and the field said so quickly. Brent Bridgeman published a comment in Psychological Science in 2005, volume 16, pages 745 to 746, arguing that the resulting individual IQ predictions were implausibly precise given what the underlying data could support. The general point generalizes past the SAT: a strong group level correlation permits statements about populations and produces wide error bands around any single person.
Three additional problems apply to graduate admission scores in particular. Range restriction is severe, since people who sit a graduate entrance exam are already a selected slice of the population, which compresses the observable variance and distorts any conversion built on it. Percentiles reference the test taking group, not the general public, so a ninety ninth percentile among graduate applicants and a ninety ninth percentile in a nationally representative norm sample are different quantities that share a label. And scoring systems change: the exam has been rescaled and restructured more than once, so a percentile from one era does not map cleanly onto the same percentile from another.
Put those together and the honest summary is that admission test performance supports a statement like "measured well above average on a test that loads heavily on general ability" and supports no specific three digit conversion for any individual. That is a real and useful conclusion. It is simply not the conclusion people want when they are trying to settle an argument with a number, and it is worth noticing that the conversion always seems to be attempted in the direction that produces a more impressive figure.
7 Ceilings, Error Bands and What Real Reports Look Like
Professional batteries do not report unbounded numbers. Mainstream adult instruments have historically capped their composite reporting around 155 to 160, and scores approaching that range require extended norms developed specifically for high scoring examinees. This is not an arbitrary convention. Standardization samples contain relatively few people at the extremes, so the norms thin out where the population thins out, and score precision degrades exactly where the most dramatic claims live.
Every properly reported score also arrives with a confidence interval rather than as a point. Test manuals publish standard errors of measurement for each composite, and reports typically present a band spanning several points around the obtained figure. A conscientious clinician writes something closer to "full scale composite in the range of 143 to 152, ninety five percent confidence" than a bare number, because the band communicates what the instrument can actually resolve. That precision loss is another thing that disappears the moment a score becomes a conversational fact.
Apply this to a claim of 150. If it were real and modern, it would sit in the zone where extended norms and careful documentation matter most, meaning it would be one of the better documented events in a person's testing history rather than a hedged remark with no paperwork attached. If it were instead a childhood ratio score, it would be a number from an incompatible scoring system that cannot be read on today's scale at all. Both of the plausible readings point to the same practical conclusion, and neither of them yields a figure anyone can use.
This is also the reason serious assessments report profiles with intervals. ACIS follows the same convention: results come back as a dated profile across separate cognitive domains with intervals around each estimate, because that is the form in which a measurement is interpretable. One floating number, however impressive, is not the output of a measurement process. It is what remains after the measurement has been thrown away.
8 Expertise Raises the Evidence Bar Rather Than Lowering It
The most common defense of the 150 figure runs like this: he is a psychologist, so he knows what an IQ score means, so his statement should be trusted. The reasoning inverts how evidential standards work.
Consider what a person with Peterson's training would have available if a documented result existed. They would know the instrument and edition, because clinicians record it. They would know the approximate date, because reports are dated. They would know the index profile, because batteries produce one, and they would be comfortable reporting confidence intervals, because their own published work used validated scales with known properties. Someone who co-constructed an instrument in Journal of Personality and Social Psychology can supply this level of detail without effort. When a claim from such a person contains none of it, the missing detail is more meaningful than it would be from someone who never handled a test protocol.
There is a second layer. Peterson has taught and discussed the psychometric prediction literature publicly for years, and that literature is real. Schmidt and Hunter's 1998 meta-analysis in Psychological Bulletin, volume 124, pages 262 to 274, is its central reference and estimated an operational validity near .51 for general mental ability predicting job performance across occupations. That estimate has since been revised: Sackett, Zhang, Berry and Lievens published a reanalysis in Journal of Applied Psychology in 2022, volume 107, pages 2040 to 2068, arguing that decades of range restriction corrections had been applied too aggressively and producing a substantially lower corrected estimate near .31. Cognitive ability remains among the better predictors available, at a strength that supports population level statements and leaves enormous individual variation unexplained. Our page on cognitive ability and job performance works through what those coefficients do and do not license.
The relevance to this page is direct. The very literature that makes cognitive measurement worth discussing is a literature about aggregates with wide individual scatter, documented instruments and stated error. Someone fluent in it is fluent in exactly the reasons an undocumented personal number should not be treated as a measurement. Knowing the field does not exempt a person from its standards. It is the strongest reason to hold them to those standards, and to hold everyone quoting them to the same.
9 Public Controversy Is Not Evidence in Either Direction
Peterson is a polarizing public figure, and searches for his name and IQ return a great deal of material that is really about something else. Supporters cite the number as confirmation that his arguments deserve deference. Critics attack the number, or mock it, as a way of attacking his conclusions. Both moves commit the same error, and this page rejects both. The population level research points the same way; as the evidence on IQ and political orientation sets out, whatever average differences appear depend on the country, the cohort and the wording of the scale, and none of them license a judgment about the individual in front of you.
Agreement with someone's views is not evidence about their measured cognitive ability. Neither is disagreement. Cognitive test scores predict performance on defined tasks under standardized conditions, and they carry no information whatsoever about whether a person's positions on culture, politics, religion or psychology are correct. Highly capable people hold views that later turn out to be wrong, and people of ordinary measured ability are right about difficult questions all the time. Any argument of the form "his ideas are good, therefore the number is real" or "his ideas are bad, therefore the number is inflated" is reasoning backward from a conclusion the arguer already held.
This page also declines two things that appear constantly in material about him and have no place in an evidence assessment. It does not speculate about his health, medical history or personal circumstances, which are not public data and would tell us nothing about a test score even if they were. And it does not use ridicule as a substitute for analysis. The question here is narrow and answerable: does a documented, dated, instrument identified IQ record exist in public? That question has the same answer whatever you think of him, which is precisely why it is the only question worth asking.
Applying a single standard consistently is what makes a verification method worth anything. A method that hardens against people you dislike and softens for people you admire is not a method, it is a preference wearing the costume of one. Every claim examined on this site gets the same four requirements, and this claim gets them too, no more and no less.
10 What Would Actually Settle This
Suppose someone wanted to resolve the question rather than argue about it. The requirements are concrete and short, and running the claim against them takes a minute.
A named instrument and edition. Not "an IQ test" but a specific battery and version, because norms, ceilings and score meanings shift between editions. Numbers from a ratio scale, a modern deviation battery and an unnormed web quiz are three different objects that happen to share digits.
A date of administration. Cognition is measured at a point in a life. A result obtained at twelve does not describe a person at fifty, and processing speed in particular changes substantially across adulthood even when verbal knowledge does not.
A defined norm sample. A standard score is a comparison against a reference population. Without knowing who that population was, and when it was collected, the comparison has no anchor and the number has no content.
A full score report with intervals. Separate index scores across domains, each with its confidence band, plus the examiner's notes on administration conditions. A naked composite with no profile behind it is the signature of a claim that never involved a testing session.
Hold the circulating 150 against those four and it satisfies none. Hold Peterson's academic record against the appropriate standard for what it claims to be, meaning institutional records of degrees, appointments and indexed peer reviewed publications, and it satisfies all of them cleanly. That contrast is the entire lesson. The same person can have an impeccably documented professional history and a completely undocumented personal number, because the two claims are made of different material and were never verified by the same process.
11 Running the Check Yourself
You can reproduce everything above for any public figure with a search engine and about ten minutes. Here is the procedure with this case as the worked example.
First, hunt the origin. Search the exact figure alongside the person's name and sort by oldest. Keep clicking backward until the results stop getting older. For this claim the trail runs into blog and forum territory around 2018 and stops at a post that itself credits a reader's email. When the earliest findable version of a claim is already secondhand, you have learned the decisive fact about it before minute four.
Second, hunt the instrument. Add testing vocabulary to the search: the names of the major adult batteries, plus terms like score report, examiner and administered. Genuine testing events surface these details immediately because the details are what made the event reportable. Here the searches return only more copies of the bare figure. A number that never appears next to the name of a test was not produced by one.
Third, look for the person's own dated words. Search their name with phrases like my IQ, tested and interview, and require a source you can open: a full episode, a transcript, a published book page. A compilation clip with no upload provenance is not a source, it is an edit. In this case no primary appearance surfaced, which is why this page describes the statement as attributed rather than as made.
Fourth, file the claim. Every celebrity intelligence claim lands in one of four bins: a documented test, an explicit refusal to claim a number, a retroactive estimate built from biography, or an unsourced attribution. This one files under unsourced attribution, with the unusual feature that the person it is attached to has a verifiable psychometric publication record. Run this drill three times and the genre stops persuading you. You will also notice how nearly empty the documented bin is, which is the real story behind every list of the highest IQs ever recorded.
12 Verdict: Documented Career, Undocumented Number
Jordan Peterson's IQ is not publicly known. The 150 figure circulating under his name is an attribution without a locatable primary source, names no instrument, carries no date, and has never been accompanied by a score report. The most cited origin point for it is a 2018 blog thread that credits an email from a reader and points to a video link the post does not contain. That is the whole evidentiary foundation, and it does not hold a three digit claim.
Everything else about his professional record checks out and needs no numerical decoration: doctoral training in clinical psychology at McGill, teaching at Harvard, a long professorship at the University of Toronto, and co-authored peer reviewed work in Journal of Personality, Journal of Personality and Social Psychology and Journal of Applied Psychology, including the construction of a personality instrument other researchers went on to use. Those are checkable facts with volumes and page numbers. The IQ figure is the one item in his public profile that fails every check, and it is the item repeated most often.
This verdict would change the day a dated report naming an instrument becomes public, and not before, regardless of how many more articles print the number. In the meantime, the more useful observation is that the number people argue about would tell them far less than the profile nobody ever asks for. Domain scores with intervals describe how a person actually thinks. A single remembered digit string describes nothing, no matter whose name is attached to it. If you have been comparing yourself against a figure you read in a headline, the alternative is to find out what a given score actually means and then measure rather than guess.
Publicly unknown. No test record, date or instrument has ever been published for him, so there is nothing to report beyond the fact that the question is open.
Did Jordan Peterson say his IQ is 150?
Circulating clips and posts attribute such a statement to him, but we could not find the original appearance behind any version of it. The safe description is attributed, not confirmed.
Where did the 150 number come from?
The most referenced starting point is a November 2018 blog thread that credits a reader's email and refers to a video link absent from the post. Later articles copied the figure from there.
Is he actually a psychologist?
Yes. He holds a doctorate in clinical psychology from McGill University, awarded in 1991, and spent his career in academic psychology departments.
Did he really teach at Harvard?
Yes. He was on the faculty of Harvard's psychology department from 1993 to 1998 before moving to the University of Toronto, where he became a full professor.
Has he published real psychometric research?
Yes. He co-authored the 2007 Journal of Personality and Social Psychology paper introducing the Big Five Aspect Scales, among other peer reviewed work involving validated instruments.
If he knows testing, shouldn't we trust his number?
Training makes someone better at explaining scores, not exempt from documenting their own. It also means he could supply the instrument and date easily, which makes their absence more telling.
How accurate are people's guesses about their own intelligence?
Freund and Kasten's 2012 meta-analysis in Psychological Bulletin found a pooled correlation near .34 between self-estimates and measured ability. That is well above chance and far too imprecise for individual placement.
Why do childhood scores cause so much confusion?
Older testing often used ratio scoring based on mental age, which produced figures on a different scale with a different spread. Those figures cannot be read as modern standard scores.
How rare would a score of 150 be?
About one person in 2,330 on a scale with mean 100 and standard deviation 15. Uncommon but not extraordinary within a selective academic population.
Is 150 within the range professional tests report?
It sits near the upper end. Mainstream adult batteries generally cap composite reporting around 155 to 160, and results that high call for extended norms built for the tail.
Can GRE or SAT results be converted into an IQ?
Not for an individual. Frey and Detterman showed a strong group level relationship in 2004, and Bridgeman's 2005 comment in the same journal argued the individual predictions were far too precise for the data.
Why does percentile wording mislead people?
A percentile is relative to whoever took that test. Ranking near the top among graduate school applicants is a narrower comparison group than ranking near the top of a national norm sample.
Can someone be far stronger in one cognitive domain than another?
Routinely. Gaps of a standard deviation or more between a person's highest and lowest index score are common, which is why batteries report each domain separately.
Does this page say which of his domains is strongest?
No. Descriptions of his profile trace to the same unsourced chain as the figure itself. Naming a person's relative strengths requires the report, and no report exists here.
Does a high composite mean every ability is high?
No. A composite averages across domains, so identical composites can sit on top of very different profiles, one flat and one with a thirty point internal split.
Does agreeing or disagreeing with his views tell us anything about his IQ?
Nothing at all. Test scores predict performance on defined tasks and say nothing about whether a person's positions are correct. Both directions of that argument are backward reasoning.
How strongly does cognitive ability predict work outcomes?
Schmidt and Hunter estimated an operational validity near .51 in 1998. Sackett and colleagues revised that downward toward .31 in 2022 after correcting how range restriction had been handled.
Why should a real score come with a confidence interval?
Because instruments have measurement error. Manuals publish standard errors for each composite, and competent reports present a band rather than pretending to point precision.
Does unknown imply anything negative about him?
No. Unmeasured in public simply means unmeasured in public. His documented research and teaching career stands independently of any score, and neither one substitutes for the other.
What would change this verdict?
A dated report naming the battery and edition, with index scores and intervals. One such document would settle the matter instantly, while additional articles repeating the figure change nothing.
14 Sources Behind This Page
No figure on this page should be taken on faith, including ours. The sources below are the primary records and the research needed to audit any IQ claim attached to a public figure.
Cox, C.M. (1926). The Early Mental Traits of Three Hundred Geniuses, full text at the Internet Archive. The historiometric study behind most IQ figures still attached to historical figures.
Frey, M.C. & Detterman, D.K. (2004). Scholastic Assessment or g? Psychological Science, 15(6). The study linking SAT scores to measured cognitive ability, and the reason SAT conversions are estimates rather than measurements.
Lubinski, D. & Benbow, C.P. (2006). Study of Mathematically Precocious Youth after 35 years. Perspectives on Psychological Science. What long term follow up of verified high scorers actually shows.
Voncken, L., Albers, C.J. & Timmerman, M.E. (2019). Improving confidence intervals for normed test scores. Behavior Research Methods. Open access. Documents the mean 100, SD 15 metric and the uncertainty that norming from samples adds to any score.
Pearson (2008). WAIS-IV Score Report sample. What a real report contains: every composite paired with a percentile rank, a 95% confidence interval and a qualitative description, never a bare number.
Pearson Clinical Assessment Scientific Council (2023). Standardized Clinical Assessment for Practitioners: A Primer. How standard scores, percentile ranks and the standard error of measurement are meant to be read together.
Pearson (2024). WAIS-5, Wechsler Adult Intelligence Scale, Fifth Edition. The current adult battery, covering ages 16:0 to 90:11 across five cognitive domains.
Crawford, J.R., Garthwaite, P.H. & Slick, D.J. (2009). On percentile norms in neuropsychology. The Clinical Neuropsychologist, 23(7), 1173-1195. Three definitions of a percentile coexist in practice and can return different results for the same score.
Take the assessment
You get a profile, not a number
ACIS measures six CHC domains across 20 subtests and reports each one with its own normed score and confidence interval, so you can see where you are strong and where you are not.
