The IQ of Jewish prisoners

In 1992 the journal Psychological Reports published an interesting paper called “JEWISH INMATES: AN ATYPICAL PRISON GROUP” by DONALD I. TEMPLER and PENNY A. JACKSON

The study design was as follows:

All inmates were in the Southern Desert Correctional Center, a medium security prison in whch most convicted felons in the southern half of Nevada are processed and incarcerated. The identities of the 13 Jewish inmates were obtained from prison records and word of mouth; it appears that these men constituted the total population of Jewish inmates.
The quantitative part of this study focused upon two different sorts of information in the records of inmates. One was Raven’s Standard Progressive Matrices percentile based on norms in Table SPM 25 of the manual (Raven,
Court, & Raven, 1986). Because Jews in general tend to score higher than non-Jews on tests of mental ability, it was hypothesized that Jewish inmates would have higher scores on the matrices thin non-Jewish inmates. It was
also hypothesized that Jewish inmates would have fewer violent crimes.

The unstructured interviews were held with a Jewish custody staff Lieutenant, a Jewish Correctional Case Worker Supervisor, and 10 of the 13 Jewish inmates. In addition, a director of a half-way house for the Jewish exfelons in Los Angeles was interviewed by telephone.

Here are the results:

One of the Jewish inmates did not have a Raven’s score or percentile in his records. For the 12 Jewish men who did, nine (75%) were over the 80th percentile of the Matrices manual, but only 27% of the 1477 general population inmates were over the 80th percentile (x2 = 14.03, p< 0.001). Four (31%) of the 13 Jewish inmates and 578 (52%) of the 1477 general population inmates had one or more violent felony convictions (x2 = 2.35, p<0.20).

The 80th percentile is IQ 113 (white norms) so assuming a normal curve and a standard deviation of about 15 for both the Jewish and Gentile inmates, these results suggest the average Jewish prisoner scored IQ 123 and the average Gentile prisoner scored 104.

However the Raven norms were probably about 50 years too old, and it’s well known that Raven ability increased precipitously over the 20th century because of prenatal nutrition and/or schooling (the Flynn effect). It’s also well known that prisoners average IQs around 90. So let’s say the scores were inflated by about 14 IQ points. This would suggest Jewish prisoners have a mean IQ of 109 (19 points higher than their Gentile counterparts!).

Of course with only 12 Jews in the study, one shouldn’t overinterpret these results.

The study is also odd because the Jewish IQ advantage is seen on a non-verbal test where Jews typically have less of an advantage. On the other hand, the Raven aimed to measure higher level conceptual abstract ability that transcends the traditional verbal vs spatial dichotomy.

The authors noted that the Jewish prisoners did not strongly identify with Judaism and had middle class backgrounds. They also showed high achievement motivation.

The authors write:

The reason(s) the Jewish inmates scored higher on the Raven Matrices are not certain. Nevertheless, it should be pointed out that the test is a nonverbal and purported culture-fair test with content that appears less relevant
to formal education than most standard tests of intelligence. Further, Jewish culture is heavily saturated with verbal learning and performance. A large number of Jews do not enter occupations dependent on spatial relations such
as those of machinists, mechanics, draftsmen, and mechanical engineers. To ascribe automatically the high Raven scores entirely to education of high quality or ephemeral characteristics would be simplistic and is unwarranted
at this time.

On the other hand, one of the reasons the Raven has been so sensitive to the Flynn effect is that scores likely can be propped up by education. This seems counter-intuitive because the test is so culturally reduced, but because there’s no time limit and the task is so abstract, educated people likely approach the test with much more confidence, interest, and persistence. By contrast gypsies, who have a subculture quite opposite from Jews when it comes to education, typically dropping out of school in 8th grade, have adult Raven IQs around 70 and complain that it gives them a headache. It would be most interesting to compare the IQs of Roma raised by Ashkenazim and vice versa.

Big Brained PhD student wants to know his IQ part 2

I recently got an email from a very big brained (literally) PhD student who wanted to know his IQ. Based on your education, estimated brain size, and ethnic background, I estimated you would score 119 on an IQ test (or the equivalent on a college admission test).

However I didn’t need to estimate because you had already told me what your scores were. You “took the SAT twice in 2006 and scored 680V/660Q the first time and a 720V/630Q the second time.”

In the early 2000s, I estimate IQ = 23.835 + 0.081(SAT score), so these were the equivalent of scoring IQ132 and 133 respectively (U.S. norms).

You also say “I took the, which is essentially a Raven’s progressive matrices test, in high school and got a 124.” The problem is I don’t know if has valid norms.

You also state:

I’ve taken the GRE a few times and my score in 2010 was 630V/650Q. I took a practice test years later and scored 163V/155Q. After studying for a couple of months, I took the GRE two more times and was able to get a 170V/152Q/4A the first time and a 169V/161Q/4.5A the second.

So your combined (V + Q) scores on the new GRE (318 to 330) seem to equate to an IQ of 134 to 145! This is considerably higher than your IQ as derived from the SAT even though both tests measure similar skills (reading & math). I wonder if most people score higher on the GRE than the SAT because university is inflating their scores beyond their true ability.

You also state:

I took the online Wonderlic when I was in my early 20s and got a 31 on it. I took it again last year and got a 30.

This fits with my suggestion that thirty-somethings deserve a 1 point age bonus on the Wonderlic. A score of 31 = IQ 118 (remarkably close to what I estimated from your education and brain size, but now I’m just data mining)

You also write:

I’ve also recently taken the Open-psychometrics IQ test and scored a 137 Verbal/134 Spatial/121 Memory with a calculated 127 Full-scale IQ. The designers claim the norming is based on scores from proctored IQ tests and admissions test scores of users. I also took Antjuan Finch’s SAT/IQ test and I scored a 30/30 on the verbal section and, I believe, a 26/30 on the Quantitative.

I don’t recall Antjuan having a quantitative test (maybe you mean his non-verbal test).


Assuming full disclosure, your test scores range from 118 (wonderlic) to about 140 (the mid-point of your GREs). Assuming these two tests correlate about 0.8 if the general U.S. population took both, this suggests a composite IQ of 131. This is 11 points higher than expected for a black American PhD with an assumed 1534 cc brain size.

But then your background is far more elite than the average black American’s. You write:

My parents are black African immigrants. My dad doesn’t have a formal college education but learned about computers on his own/on the job as they were becoming more important for business infrastructure and joined the IT revolution at the ground floor. When he retired, he was a director of IT at a cybersecurity company.

My mother has a community college degree in nursing but decided to focus on being a stay-at-home mother rather than having a job. When my brother and I were old enough to watch ourselves, she began to take on new jobs which she excelled at. Her jobs have included being a computer lab instructor for my school (not computer science, just teaching proper operation and security for them), a small business owner (though the 2008 financial crisis hit us hard), and an interior designer/event planner. She also manages our family’s finances. You’ll probably get a kick out of this: people have actually talked about getting a very Oprah-esque vibe from her.

Big brained PhD student wants to know his IQ: Part 1

A reader sent me the following email:

Hello PumpkinPerson,

I have been reading your blog off-and-on for the last 5-6 years and I’ve seen that you’ve started to estimate reader IQs from their histories and test scores and I was curious what you would estimate mine to be. I’ve written a history that I think contains the majority of the most pertinent information for your approximation. I’m sorry if I wrote too much but I wanted to give as much information as possible.

Biography: I am black male in my early 30s who is currently in his 4th year of a PhD program in psychology (my program is not elite at all and is roughly #100 in the US grad school ranking for my field) after getting my BA in the same from a top 60 college. I grew up upper-middle class and have lived in multiple countries but am mono-lingual and have spent most of my life in the US. I could read the bible at the age of 5 and I was able to read at a college level by the time I was in middle school. My math ability has lagged behind notably, however. I was better than most of my peers at math but I was not truly exceptional by any means. My family is religious but I have been an atheist since I was 14.

About 520,000 black Americans become old enough to get a PhD every year, but of these, only 2,512, did so in 2019. This puts the median black PhD in the top 1,256, or one in 414 level (+2.8 SD level). Assuming a 0.68 correlation between IQ and years of education, we’d expect your IQ to be 0.68(+2.8 SD) = +1.9 SD above the black mean.

Given that U.S. blacks have a mean WAIS-IV IQ of 87.7 with an SD 14.4 (U.S. norms), +1.9 SD = IQ 115.

According to page 374 of Charles Murray’s Coming Apart, blacks with PhDs or professional degrees average IQ 112.2 (U.S. norms)

You also write:

Physical characteristics and health: I am 5’10 and my head circumference is around 60.5cm. I think this should be taken with a grain of salt though since my head is very long and fairly narrow which should deflate actual brain volume. I was very overweight until I was around 20 years old at which point, I lost a lot of weight and have vacillated between 160-185 since and am roughly 180lbs right now. I have a history of depression and anxiety and have diagnosed epilepsy.

605 mm circumference implies a 192.5 mm cranial diameter, however because your circumference was measured in-vivo, I subtract 11 mm for fat and skin around the skull. This reduces it to 181.5 mm which implies a true circumference of 570.2 mm. Below is the curve predicting brain size from head circumference:

Source: Figue 3 in Jørgensen, J. B., & Quaade, F. (1956). External cranial volume as an estimate of cranial capacity. American Journal of Physical Anthropology, 14(4), 661–664.

The above curve predicts you’d have a cranial capacity of about 1676 cc according to an equation I created from said curve:

cranial capacity = 0.0080(head circumference)^2 – 1.9(head circumference) + 158

However because of ethnic and gender differences in head shape, I would multiply this figure by 0.915, reducing it to 1534 cc. This is still over 227 cc above the mean for young black men circa 1980 and probably about 174 cc above young black men today, putting you in the 97 percentile (+1.91 SD). Such a large brain may help explain your epilepsy because of the enhanced electrical activity.

Assuming a 0.4 correlation between IQ and within-sex brain size, this predicts an IQ of 0.4(+1.9) = +0.76 SD above the black mean, or IQ 99.


Estimated IQ of black American PhD: 115

Estimated IQ of black American with very large (1534 cc) brain size: 99

Estimated IQ of black American with both: 119

So 119 would be what I’d expect based on your two most salient biodemographic IQ correlates, but in part 2 we’ll discuss your actual test scores and see how close this prediction comes.

Jensen’s default hypothesis

In his book The g Factor (pg 169), Arthur Jensen wrote:

The broad heritability of IQ is about 0.40 to 0.50 when measured in children, about 0.60 to 0.70 in adolescents and young adults, and approaches 0.80 in later maturity.

Rushton & Jensen write:

…the default hypothesis, is that genetic and cultural factors carry the exact
same weight in causing the mean Black–White difference in IQ as they do in
causing individual differences in IQ, about 80% genetic–20% environmental by

Elsewhere they write:

Jensen (1998) demonstrated that Black-White IQ differences typically increase with age (because genetic influences become stronger over the life span). He used Shuey’s (1966) compendium to document that the average
Black-White difference was 0.70 standard deviations in early childhood, 1.00 standard deviations in middle childhood, and 1.20 standard deviations in early adulthood…

…Until the results of several such studies allow reassessment of the situation, the best estimate of Black-White convergence over the past 100 years is between 0 and 3.44 IQ points—a maximum effect size of 0.23—well within the predictions of our estimated heritability of .80 for the Black-White g difference in the United

Rushton & Jensen seem to be saying that because IQ is only 20% environmental (among U.S. whites), the 15 to 18 IQ point black-white U.S. IQ gap could at most, only be expected to close by 20% (3 to 3.6 points), thus reducing the gap to only 12 to 14.4 points .

One problem with this theory is that 80% means the proportion of IQ variance explained by genes so you need to take the square root to get the regression line predicting genetic IQ from phenotypic IQ which means differences would actually be 90% genetic, allowing only a 10% decline in the black-white IQ gap, assuming you believe a) heritability is 80%, and b) the default hypothesis is true.

White Lotus (2021 TV series)


I didn’t think I’d like this show but by the second episode, I found myself utterly addicted. It’s one of the best TV shows I have ever seen.

The HBO series is set at a Hawaii hotel called The White Lotus and focuses on several of the guests and a couple key staff members. There’s the prissy gay Hotel manager who smiles to everyone’s face but badmouths them behind their backs, the spoiled yuppie who can’t understand why is new bride would want an identity of her own, the woke university student who brings a minority friend on the family vacation to prove she’s morally superior to her parents, and the lonely ageing rich lady who can’t enjoy her wealth because she’s constantly whining about her childhood.

If American Beauty exposed the façade of the perfect American family, then White Lotus shows us the dark side of the perfect American vacation.

The show reminded me of the most luxurious vacations I have ever taken: When you first see everyone they seem so perfect, but as you get to know them, the dysfunction is legion. It made me nostalgic for the last time I’d been to Hawaii; the delicious outdoor dinners overlooking the sun setting into the ocean as the wind blew in my hair; the smell of tiki torches everywhere.

You’ll fall in love with the characters and the humour is up to the minute. In one scene the lonely rich lady agrees to meet with a strange man because she thinks, what a wonderful person he must be to work for “BLACK LIVES MATTER”. Much to her chagrin she discovers that the BLM he works for is actually BUREAU OF LAND MANAGEMENT.

In one scene, a woman worries that her son is actually at a disadvantage these days being a white heterosexual male (“no one seems to care about them”) only to be scolded by her woke daughter for saying something so cringeworthy. When the woman tries to defend herself, the daughter cuts her off again “MOM CRINGE!”.

In another scene, the mother scolds her woke daughter and minority friend for badmouthing Hillary Clinton, disgusted that the woke generation’s girls could turn on such a feminist icon. The scene is ironic because the woke daughter looks like a young Hillary:

woke university student with her minority friend
Young Hillary Clinton

If you get HBO-on demand, I would definitely watch this series from the start.

Monstrous (2020)

pumpkin person rating: 7/10

I couldn’t let Friday the 13th pass without posting something horror related. I recently watched the 2020 horror film Monstrous on Shudder. The film is about two women who have never met before (one is lesbian, the other bisexual) who go on a log cozy road trip to the lesbian’s cottage where a lot of pretty young women have gone missing. Some suspect they’re being killed by a serial killer, others suspect it’s sasquatch. The bisexual risks it all to find out the truth.

Well worth watching if you’re fan of monster movies, slashers, wilderness, or if you just get off on watching pretty women make out. 🙂

Watch the best Canadian horror film of all time, for free

The Dark Hours (2005) is the best Canadian horror film of all time in my opinion. Although the film was a commercial flop (reportedly grossed only $423 USD) and has a mediocre rating on IMDB, it is one of the most atmospheric, interesting, well-acted, and well-cast films I have ever seen. How sad to see such great art playing for free on youtube but you might as well watch before they take it down.

The film revolves around Samantha, a doctor at some sort of hospital for the criminally insane who discovers she has a brain tumor and decides to spend a relaxing weekend at her Winter cottage with her underemployed husband and younger sister. But things get weird when a diminutive Ashkenazi stranger (who reminds me of commenter Pill) is invited in to warm up by the fire.

I suggest hooking your computer to a television so you can enjoy this on a big screen. While some viewers might find this film a little too slow-paced, ambiguous and anti-climatic, the dream-like quality and distinctly Canadian vibe makes it an enjoyable experience.

Yet another reader wants to know his IQ

A reader sent me the following email:

Dear PP, 
I recently stumbled upon your blog and find it quite fascinating. I’ve been interested in IQ and psychometrics for many years and have taken many online assessments and standardized exams to gauge my own IQ. Interestingly enough, however, it seems that the more tests I take, the less conclusive my IQ range becomes due to high variances in scores. 
This is why I’m turning to you, an expert on this topic, to request an estimation. For one, in elementary school I was tested for the gifted program and unfortunately rejected. This tells me that my childhood IQ was beneath 130, which I believe is the cutoff for most school gifted programs.

Yes, but the mere fact that you were tested for gifted suggests you were probably close. Perhaps 125 (U.S. norms).

The reader continues:

However, all the standardized exams that I’ve taken (which you’ve claimed to be valid proxies for IQ) seem to tell a different story.  My scores are as follows: 
SAT (Taken 2017) – Math 800, Verbal 720 
ACT (Taken 2017) – 34 Composite 
I was also ranked 6/945 in my high school class for GPA. I figured this might be relevant given the large sample size and the notion that academic performance correlates with IQ. Since my high school was in a lower-middle class area, we can assume a relatively normally distributed sample with an average akin to the American population average. All the students ranked ahead of me were Asian, and I am South Asian myself (Indian). 

A post-2016 SAT score of 1520 equates to an IQ of about 132. I’ve done very little research on the ACT but apparently a 34 on the SAT is like 1535 on the post-2016 SAT which equates to an IQ of 133.

The reader continues:

I’m also currently an undergraduate (a double major in mathematics and economics) and have been studying for the GRE. My diagnostic score for the GRE, with little to no preparation, was a 157 Verbal and 166 Quantitative. I’ve since improved to a 163 Verbal and 167 Quantitative for my second practice exam. 

Averaging across both sittings, you scored 160 V and 167 Q for a composite of 327. Preliminary research suggests this equates to an IQ of 143.

Averaging the SAT, ACT & GRE together, your mean IQ equivalent on college admission tests is 136. Normally I don’t endorse the averaging of tests but these particular tests are so similar that it’s not worth the trouble of considering them different tests.

Meanwhile, your score on an official IQ test was probably around 125.

Given that official IQ tests correlate no more than 0.7 with college admission tests, your score on a composite of both types of tests would be about 134.

I had a couple follow-up questions which the reader answered in another email:

my socioeconomic background is generally lower-middle class. My high school was located in the suburbs but had many students from the nearby urban area…
One caveat to my socioeconomic status, which applies particularly to my situation, was that I grew up in a single-parent household. Both my parents were highly educated (held graduate degrees) but my father unfortunately passed away when I was young.

I’m very sorry to hear that, but at least he lived long enough to pass on his high IQ genes through you.

This obviously influenced my household income. 
As for my ethnic background, I come from a Brahmin ancestry. Although I’m not a practicing Hindu, I recognize that this could have played a role in my generally well-educated familial background. 


A reader’s Wechsler profile

I recently got an email from a college age man who was concerned that his VERBAL KNOWLEDGE INDEX (as I call it) was in the genius range (on both the children and adult version of the Wechsler) despite the rest of his cognitive profile being mediocre or low. When he asked his psychiatrist to interpret the results there was no helpful reply, so a friend of his suggested he contact me.

I’m not a psychiatrist so my opinions are for entertainment purposes only.

The first thing I did was correct all his subtest scores for norm inflation because the WISC-IV norms were 11 years old when he was tested and the WAIS-IV norms were a decade old. The sources I used were pg 240 of James Flynn’s Are We Getting SMARTER? and this table found here:

Such corrections are approximate because one can’t always assume that the rate of norm inflation can be extrapolated beyond the dates from which we have data and some subtests are so new, their rate of inflation had to be estimated using similar tests. In some cases there was norm deflation (see Coding in the table above, aka Digit Symbol).

The next thing I did was substitute the four index scores used on the WISC-IV and WAIS-IV with the five index scores used on the WISC-V. Again this gives only approximate results because the WISC-V index scores were built exclusively on WISC-V data and you’re not allowed to just substitute different versions, and in some cases I had to substitute subtests or adjust for not having the right number of subtests.

Nonetheless, the five factor model is so superior to the four-factor model, that for entertainment purposes only, I did it anyway.

The other liberty I took was calculating his overall IQ, by weighting all five indexes equally (the WISC-V gives equal weight to all core subtests, but more core subtests fall under some indexes than others).

WISC-IV (2013)WAIS-IV (2016)
 Similarities 17.23 18.4
 Vocabulary 15.67 17.1
 (Information)  12.55
 (Comprehension) 11 
 Block Design 5.56 6.73
 Visual Puzzles  
 Matrix Reasoning 8.4 13.45
( Picture Concepts) 12.4 
 Digit Span 7.89 7.73
 (Arithmetic)  7
 (Letter number sequence) 8.89 
 Digit Symbol 8.66 7.81
 Symbol Search 6.66 10.81
 OVERALL IQ 98 106

The first thing we notice is remarkable stability as we move from the children’s scale (2013) to the adult scale (2016). In 2013 his overall IQ was slightly below the U.S. mean of 100; in 2016 he scored slightly above, and even this modest increase might be partly explained by practice effect.

In 2013 his profile was verbal > abstract > working memory > processing > spatial. In 2016, verbal > abstract > processing > spatial > working memory. In other words there is a near perfect 0.95 correlation between his cognitive profile in 2013 and 2016, despite the fact that different versions of the Wechsler (with different questions) were used on each date.

Vertical reliability vs horizontal reliability

Reliability (not to be confused with stability) is typically measured by dividing all the items on a test in half in some random way (e.g. odd vs even numbered items). If the total score on all the odd number items correlates well with one’s score on all the even numbered items, this suggests your score was reliable, because it internally self-replicates. The reliability of the Wechsler scales are so high at the full-scale level that they are said to have a standard error of only 2 points, meaning in 2/3rd of all cases, one’s score is within 2 points of one’s “true” score and in 95% of cases, one’s score is within 4 points of one’s true score.

But what is true score? True overall score is the overall score one would get on the Wechsler if we could make every subtest infinitely long, yet factor out fatigue, practice effects, and ageing.

However I propose an alterative definition of true overall score: the overall score one would get on the Wechsler if we could increase the number of subtests to infinity, yet factor out fatigue, practice effects, and ageing. But since many subtests redundantly measure the same functions, what we really want to do is increase the number of index scores to if not infinity, then the maximum number that exist within the human mind. Let’s call this horizontal true score, to distinguish it from the typical definition of true score, which we can call vertical true score.

To measure horizontal true score, imagine we were doing a poll of the average IQ in a given school. If we tested five students, and they had an average IQ of 80 with an SD of 10, then the standard error of our poll would be 10 divided by the square root of our sample size.

Now instead of trying to find the average IQ of different people in a school, we’re instead trying to find the average index score of different talents within the same mind. Once we find the standard error of the average index score, we could convert it to standard error of overall IQ (because index scores are imperfectly correlated, one’s composite score on multiples indexes tends to be more extreme than one’s average index score). Multiplying the standard error by 1.96 and then adding and subtracting it from the overall IQ gives the 95% confidence band.

Based on the amount of scatter, I calculate that all we can say with 95% certainty is that the subject’s true horizontal overall score is anywhere from IQ 73 to IQ 122 (in 2013) and anywhere from IQ 78 to IQ 133 (in 2016).

So the subject is either very smart, or very not-smart but we can’t be more precise than that without running several more tests.

Career advise

Subject will be disapointed that I didn’t devote more content to this section, but there’s only so much career advise to be gained from Wechsler test results. He is a genius at overall verbal knowledge, but only slightly above average at social knowledge (comprehension), slight below average at processing and poor at working memory and arithmetic. His social knowledge is not poor (and given the low reliability of a single subtest, should not be over-interpreted), and when combined with his genius overall verbal knowledge, he would probably have a competitive edge. I suggest he pursue whatever he field he is passionate about, as long as it’s not closely related to law, STEM, clerical or visual technical skills because these require working memory, arithmetic, processing speed and spatial ability respectively.

Guest post by Illuminaticatblog

The following is a guest article written by Illuminaticatblog. The views in this article do not necessarily reflect those of Pumpkin Person:

Many commenters on Pumpkin’s blog think very differently. This has led many to call each other autistic. But there is no scientific consensus on what autism is otherwise autism would not be confused with schizophrenia so often. MBTI the personality system does not help. A new system does. It was made by Digibro the Anime YouTuber. Two axis exist.

Lexical thinking is formalized rules thinking. Contrast (impressionistic thinking) is informal rules or patterns that are informal i.e. patterns that are new and hard to explain. creativity does not follow a rules system if new otherwise it is just calculation.

linear vs lateral thinking is easy to explain. multiple vs singular train of thought. they can be conscious or unconscious.

The only true autistic type on the chart would be the human calculator. They will have such a narrow focus that they will only do something if it is complete. They will not deviate and this makes relationships hard because relationships are not a collection of parts that can be categorized.

I am a newtype. I have high lateral thinking and high impressionistic thinking. I think in multiple ways at once and new stuff is coming into my mind all the time.

Here is a chart on blog commenters and their type.