Of the hundreds of people who clicked on the TAVIS link, 70 submitted their answers (excluding obvious retakes from the same person). The mean TAVIS is 9.79 with a standard deviation of 2.25. The distribution looks approximately normal.
However previous research suggests Pumpkin Person readers are approximately 2 standard deviations smarter than the average American. Assuming my readers have the same variability as the general U.S. population (it’s actually higher but that leads to absurd conclusions) and assuming TAVIS scores are roughly linear, we get the following IQ equivalents:
TAVIS 0 = IQ 65 Educable (mild) Retardation
TAVIS 1 = IQ 72
TAVIS 2 = IQ 78 (Borderline retardation )
TAVIS 3 = IQ 85 (Dull)
TAVIS 4 = IQ 92
TAVIS 5 = IQ 99 (U.S. average)
TAVIS 6 = IQ 105
TAVIS 7 = IQ 112 (U.S. University graduate average)
TAVIS 8 = IQ 119
TAVIS 9 = IQ 125 (Borderline genius)
TAVIS 10 = IQ 132 (Mild genius)
TAVIS 11 = IQ 139
TAVIS 12 = IQ 145 (Moderate genius)
TAVIS 13 = IQ IQ 152
TAVIS 14 = IQ 159
TAVIS 15 = IQ 166 (Promethean)
TAVIS 16 = IQ 172 (Mega society)
Much more research is needed to confirm these IQ equivalencies but it’s a start. In future article(s), I’ll explore equipercentile equating TAVIS scores with scores on well normed tests like the SAT.
Pumpkin Person 
70 testees??? I guess most of your readers don’t comment
https://en.wikipedia.org/wiki/1%25_rule
Over 300 people have now taken the TAVIS
will we see an updated norm soon? I got 14 but the iq equivalent seems incredibly high
Yes there are some major updates coming. Now that well over 300 people having taken it, I’m getting a much clearer picture of the rarity of high scores.
I solved item 18 after I finished the test, but I guess I can’t have a redo because then everyone else would have to be allowed a second try, which would change the norms
correct
Wow, this is a shockingly accurate!
My TAVIS score was 8 which is IQ 119.
My Wonderlic test score is equivalent to an IQ of 120.
On WAIS-IV, my Verbal Comprehension Index (VCI) was 117 (though my Working Memory Index – WMI, was 141, which technically makes my overall Verbal IQ much higher I think).
Honestly, pretty good!
sadly, the above is actually a plot of penis length in cms for koreans.
I thought it was a plot of what I would guess your IQ was based on your comments.
Mug of Pee cant make a single joke without having it take place at the expense of others.
its kind of sad how humanity is so intent on pinning people against each other instead of valuing acceptance. everything seems to be a competition to everyone.
i apologize. that should’ve been “tavis smiley knows what lurks in the PANTS of korean men.”
most comedy is making fun of someone LOADED. get a sense of humor.
gary larson made bank making fun of animals. why is that acceptable?
you are a non-playing character MOP.
Owlson, you #%&*#*% what do you know about all that?
^^^ and they say Republicans arent retarded….hahah. i laugh at your stupidities.
Who the hell is republican here owlson?
PP, is my WAIS-IV cognitive profile weird?
WMI (141) > PRI (127) > PSI (124) > VCI (117)
My best subtest was Digit Span (scaled score of 18). My lowest subtest was vocabulary (scaled score of 12).
1 sd gap between WMI and PRI/PSI and almost 2 sd gap between WMI and VCI.
Is it normal or maybe ADHD or something? No one ever explained to me my results since I took the test as a volunteer for psych grad students who are practising test administration. At the end, I asked for my results and was given them, though no one actually explained anything to me about my cognitive profile and what it means. Is there anything you can tell me?
Definitely not ADHD. More like the exact opposite. You did best on the indexes on which ADHDs do worst.
I have difficulty focusing on anything useful, but still do best on processing speed, which I think is usually a deficit in ADHD. I’m extremely good at routines and generally decent at applying information I already know, but I struggle to retain new information. Following a Starcraft match is beyond me. Maybe when I’m more fit, it’ll be easier
What’s the exact opposite of ADHD then 😅? Mb its the difference of ADHD-C vs ADHD-PI? Is there anything you can tell me about my cognitive profile since I was quite surprised when I first saw it. I thought my VCI would be much much higher and my PSI much much lower. I’m not sure what to make of it.
Btw, I can somewhat relate to what Ganzir wrote above.
i had thought i wouldve been diagnosed with ADHD as well. i am very inattentive and have a difficult time retaining new info like Ganzir had mentioned as well as not being overwhelmed by new information.
i have a really hard time doing word problems. i think this is because i cant distinguish between relevant and irrelevant information in the text. i become overwhelmed by reading the question or at least attempting to that it flies over my head. sucks.
Owlson, you &*#$%^ who cares ?
Who the he’ll is a republican here, owlson?
What was your FSIQ?
My FSIQ is 130 on WAIS-IV.
Oh and I just realized when I wrote my comment above, I accidentally swapped PRI and PSI. My PSI is the one that’s 127, and my PRI is 124. My bad.
So my cognitive profile is
WMI (141) > PSI (127) > PRI (124) > VCI (117)
FSIQ 130.
Did your psychologist separated your scores between CPI (cognitive proficiency index, which is the composite of PSI and WMI) and GAI (composite of VCI and PRI) since there’s more than 1SD difference between your WMI and PRI? If so, what is your CPI and GAI?
Also, I highly doubt you have ADHD, since this disorder impairs your CPI by a significant amount, and you seem to have a high CPI (130+).
Hey, ya know, I think I made a careless error on one of the easy ones, and with that my score would have been 13 and my predicted iq would be about where my pre 1995 sat score would have put me. Either way, a very close correlation. Anyway in life I’m no genius, borderline retard, to be honest!
Use the standardized test data to get a better iq norm!
Good work, PP. I do feel like me performance (9) is not as high as 125 but I’ll take it. I wonder if the time limit plays an important role in the norming.
Thanks. I feel like the norms are a bit generous too. It will be interesting to see if they hold up when I try other methods.
I received a 132 based on the ten i got right. that is probably where my verbal actually is.
my spatial is probably around average.
I am glad I reached such a score despite the culture thing. I didn’t know New York was [redacted by pp, 2022-03-22] obvious answer znd it’s probably the same for 3 other questions
High IQ people tend to be, on average, more liberal leaning. This holds true for ultra high IQ individuals as well.
But every now and then, I see someone who has a super high IQ (160+ sd.15) who is extremely conservative. They are anti-vax, don’t believe in climate change, anti everything.
Why is this? Predisposition from personality traits? Life experience?
Pumpkin, if you score below average on the visual-spatial portion of the WAIS (Block Design, Visual Puzzles, Object Assembly (WAIS-III)) but are otherwise high in g, would the Raven’s Matricies underestimate your fluid reasoning ability?
Perhaps, but not by much. It doesn’t load strongly on the spatial factor. It’s more conceptual.
I’m trying to estimate my Raven IQ but the only data point I have that isn’t contaminated with practice effects was when I first took IQtest.dk at 14.5 yrs old and scored 108, which I’ll just call 110 because ppl who take online tests tend to be above avg. It’s Flash so it’s now obsolete. The HTML version is: https://mensa.dk/iqtest/
Most people who take the test are probably 16+ and 16 seems to be where the Raven plateaus so extrapolating 110*(16/14.5) = 121.
This also puts 75th percentile for a 16 y.o. at 90th percentile for a 14.5 year old.
Do you think this is a reasonable estimate?
i do terrible on Ravens tests (around average or somewhat below). i can do all the easy ones and then the harder ones start to accumulate and put me under a lot of stress.
Is Jimbob McGee the same person as Rahul who used to post senseless questions here from time to time? sounds like it.
The Declaration was originally drafted by the Marquis de Lafayette, in consultation with Thomas Jefferson.
What’s the average for non-native speakers?
Haven’t calculated it yet, but it’s kind of useless information. Whatever language bias they suffer is likely negated by the fact that they are self-selected for higher IQ so their scores would both underestimate their IQ and underestimate the culture bias
You could see which test items are most native language-biased and exclude them from the norms, but perhaps too small samples and more work than it’s worth.
By far my highest score on any test of intelligence. I guess there are bound to be extreme outliers if one takes enough tests. I figure this score is about as high above my ‘true’ I.Q. as my worst score is below my ‘true’ I.Q.
Pumpkin,
because of my comatose psychosis, I am on the schizophrenic spectrum. My processing speed and working memory are really low. This happens in schizophrenics because they are taxed trying to maintain a sense of reality. When I push my working memory I get headaches. It is extreme burnout.
In light of the similarity to this test would you be willing to do a post on the Miller Analogies Test?
Some comments on the test after thinking about it these two days. I think the questions jump from relatively simple to very abstruse and challenging rather abruptly. You are saying Santa clause[redacted by pp, 2022-03-22} and then you are suddenly doing [redacted by pp, 2022-03-22] in your head, lol! I think the test should be 30 questions and there should be a smoother transition with a set of intermediate items. That way you could achieve a more graded distribution. Questions of the sort where the knowledge is more common, but the relationship is more difficult to tease out, perhaps on the level of the ghost/ [redacted by pp, 2022-03-22} question.
I agree. I originally designed the harder questions to be “ungoogleable,” and I figured that the easier items should be solvable to anyone who has access to the internet. I didn’t put a lot of effort into moderately difficult items because I wasn’t sure of how to make them immune to cheating. Should have had more trust.
Oh I see what you mean, they are ungoogle able. I guess that’s why it’s hard to write good test questions! If it were easy everybody would be doing it! Lol! The lions/wolves question was particularly brilliant, I thought.
Article III – The principle of any sovereignty resides essentially in the Nation. No body, no individual may exercise any authority which does not proceed directly from the nation.
Article X – No one may be disquieted for his opinions, even religious ones, provided that their manifestation does not trouble the public order established by the law.
Article XIII – For the maintenance of the public force and for the expenditures of administration, a common contribution is indispensable; it must be equally distributed to all the citizens, according to their ability to pay.
Interesting remits. I scored a 10. My WAIS results were
Verbal 121
WM 97 brutal, I know
PSI 120
PRI 106
FSIQ 114
My highest subtest on WAIS was 99th percentile in vocabulary, so perhaps this test loads on whatever aspect of cognition is involved in that area
I got 8 on this (119 IQ equivalent). My WAIS-IV scores are:
Verbal Comprehension 117
WMI 141
PRI 127
PSI 124
My best subtest was Digit Span (scaled score of 18). My worst subtest was vocabulary (scaled score of 12).
Also my Wonderlic Test score (basically a verbal test) was equivalent to an IQ of 120.
I agree that this test seems like a pretty good approximation for Verbal Comprehension.
Interesting! I would kill to have your working memory
Article I – Human Beings are born and remain free and equal in rights. Social distinctions can be founded only on the common good.
Article IV – Liberty consists of doing anything which does not harm others: thus, the exercise of the natural rights of each man has only those borders which assure other members of the society the fruition of these same rights. These borders can be determined only by the law.
Article VI – …All the citizens, being equal in its eyes, are equally admissible to all public dignities, places, and employments, according to their capacity and without distinction other than that of their virtues and of their talents.
adolph hitler reed says neo-liberalism has exhausted itself and next will be fascism or social democracy/democratic socialism for americastan.
there is no resurgence. it’s the media lying. i paraphrase reed.
Hehehe I got 16. Thanks for the little ego boost today. I submitted my 2016 SAT score as 1370 but I also took the older one and scored 2030 if that’s of interest. Nearly hit a perfect verbal on the retake (790+680=1480) but missed something. I’ve moved on now though ☹️.
Yes I saw that a second person has now scored 16 (out of what is now 173 respondents)
The math error in my post acts as a testament to my verbal-math imbalance 😂. 1470, not 1480. I’ve also posted as nemonous before btw though probably not often enough anyone would recognize me.
I’m too far removed from the test maker’s culture to know about [redacted by pp, 2022-03-23]. However I can’t even answer 13 so it’s not like I’d ever excel. I am however really curious about 14 and 21. Any way to get the answers and explanations on those? Perhaps am email if it needs to stay private?
I’ll leave that up to Teffec since he’s the test’s author.
You can give the answers to anyone who emails you. Lmk if you don’t understand the logic to any of them. 21 I think is a very good measure of mathematical reasoning. 14 may be a little esoteric, but it’s hard to tell from the inside looking out.
“Esoteric” lol
Hey pumpkin, please email me the answers. I will put my real email in the box. Thanks!
Not yet. Too many people are still taking the test to risk a leak.
Thanks PP!
A few points:
There may be a bit of selection bias in that people who scanned the items and submitted scores were those most likely to do well.
The test may have been a bit too speeded. Several of the questions I think took more than ~25 seconds to solve. The time limit obviously raised the ceiling but took away from the “eureka” factor I was going for with some of the more abstruse items.
A few questions were mathematical and/or spatial. That was probably bad design on my part, as that may confound what was otherwise a solid measure of verbal knowledge and ingenuity. Are you able to itemize the questions, or are you able to see only composite scores?
Thanks to everyone who tried and gave feedback!
Teffec, good points!
The reason I decided to impose a time limit is if I ever wanted to do a gold standard norming of this test (giving it to a representative sample of Canadians/Americans) I suspect most people would give up well before 10 minutes or practical considerations would prevent us from spending more time. Time limits might also reduce cheating because less time to google.
I would suggest re-administering it untimed, but some of the people who did the timed version would have inferred the right (or at least wrong) answer from their score (at least one person took this to an extreme).
The general rule in psychometrics is that, if lengthening or removing the time limit changes the rank-order of candidates, the time limit was too short.
correct
Yeah, the time limit was not a factor for me at all.
I see the answers, not just the total score. At some point I’ll post a ranking of the items by percent passed.
PP, how do you think this test compares with the PATMA?
I think they’re quite complementary. This test measures mostly verbal IQ while the PATMA measures mostly math IQ.
Pumpkin where are my responses to that crude idiot Hindu? he deserves to be put in his place!
theres so much [redacted by pp, 2022-03-28] already and you cant spare me the opportunity to capitalize on it? wow.
You two have already insulted each other enough. I don’t want the comment section littered with insults.
okay fair. i just want to say to him that being autistic and Indian is not a good look. maybe its accepted by society because they love sheep just like him…but overall itll limit him far more than he can even imagine.
i finished the first 10 questions in about 3 minutes with minimal effort, then quit the test when i encountered a question that i couldn’t solve; putting me at 132. i got 125 on the wais-iv and 112 on the verbal portion of the wais. i had performance anxiety and didn’t follow the instructions properly on some of the subtests… guessing 130’s fsiq
I was pretty surprised to see what my 12 equated to. It felt like I’d left a few correct answers on the table that I would’ve figured out on a day I wasn’t so tired and stressed out.
PATNA score 116 anD TAVIS score 112 (not english speaker) and Gestalt 114. What could be good estimate for my IQ?
*PATMA
118
Why a higher score than he got on any of those three tests?
Because being bright in three very different areas makes you very bright overall.
ah ye I hadn’t considered that they were tests for different factors
What’s the formula for multiple regression that a Promethean taught you? I’ve seen you mention it, but you’ve never explained it. That, or you explained it, but the numbers didn’t seem to work out.
Not a formula per se; more like matrix reasoning. I’ll revisit it in my imminent article about Dr. Phil
Assuming a correlation between each of those tests of 0.6 I think the formula would be:
(16+14+12)/square root of (1+1+1+2(0.6)+2(0.6)+2(0.6)) = 16 points above the mean.
tank u pp, ur de bestest
scored 7
tend to do about average on vocabulary tests
20 ACT first time (14) 1320 SAT first time (16) 1510 SAT second time (17)
most semi-validated IQ tests return a score of 110-140 for me, most in the mid 20s
Cool, got a 12. Yet another high verbal score yet I doubt you would know it talking to me at this point. I consistently seem to score in the 135-145 range yet the best validated tests I take seem to center around 120. I strongly suspect that the latter is the true measure of my real IQ.