Mapping real world IQ to D&D INT stat?

[Armored Walrus]

Just curious if anyone knows of any work that's been done to map the statistical distribution of IQ's among the real world human population compared to the statistical distribution of likely attribute scores if you just rolled 3d6 for each person on earth, so that it can be said that, for example, 8 Int = IQ of 80, 14 INT = IQ of 120 or whatever?

[DracoKnight]

Just curious if anyone knows of any work that's been done to map the statistical distribution of IQ's among the real world human population compared to the statistical distribution of likely attribute scores if you just rolled 3d6 for each person on earth, so that it can be said that, for example, 8 Int = IQ of 80, 14 INT = IQ of 120 or whatever?

I forget when or where I saw it, but someone tied to WotC noted that Einstein had a 14 INT.

[Breashios]

I wouldn't try. The way I see intelligence (overall intelligence) it involves most of your stats. Charisma is a portion of intelligence or at least how intelligent you appear and how you interact with social puzzles (challenges, questions, etc.). Your wisdom might impact how you perceive a question or relationships in people and things. For instance, "Which of these two are most alike?" questions. INT is just one aspect of your overall intelligence and while it may be the one that has the largest impact in scoring an IQ test, I always assume all the mental and social stats would play a role. And then there are feats. Are you lucky?

[Armored Walrus]

I'm mostly curious because I don't buy into the (edit: seemingly) widely held conception that 8 INT = dumb. My guess is that 8-12 INT is within "normal" range, and it's only above and below that that you become truly special. In one way or the other.

[Doug Lampert]

I'm mostly curious because I don't buy into the (edit: seemingly) widely held conception that 8 INT = dumb. My guess is that 8-12 INT is within "normal" range, and it's only above and below that that you become truly special. In one way or the other.

3d6 has a standard deviation of ~2.96, IQ has a standard deviation of ~15 (depending on test, but it's supposed to be 15).
3d6 averages 10.5, IQ is supposed to average 100, but most tests actually average a bit higher (all the children are above average).

This gives a conversion formula of almost exactly Int*5+50.

So if you assume IQ and Int correspond by a roughly linear formula and measure the same thing then Int 8 is an IQ of 90 or so, which is perfectly functional.

[coolAlias]

I'm mostly curious because I don't buy into the (edit: seemingly) widely held conception that 8 INT = dumb. My guess is that 8-12 INT is within "normal" range, and it's only above and below that that you become truly special. In one way or the other.

Mechanically speaking, an 8 is just barely worse than a 10, which is the average, and a 12 is just barely better.

Take a DC 10 Int check, for example. The average commoner will succeed 55% of the time; the 'smart' one 60%, and the 'dumb' one 50%. That's not exactly a huge gap in success, but it is enough to make the smarter character actually feel smarter.

So, if I was playing an Int 8 character, I wouldn't feel the need to play them as significantly dumber than an average human. Perhaps I would always let the rest of the party discuss making a plan and only chime in with my thoughts at the end to simulate being a little slower, or refrain from using big words when speaking, etc., without doing anything to restrict my actual intellectual contribution to the game.

[MrStabby]

Just compare bell curves would be an ok start. Unfortunately this only works with randomly rolled stats and even then is tied to method of rolling. I.e. an 18 or more has a probability of 1/(6x6x6) if you roll 3 dice (use binomial for picking 3 from more), then look that up on your I Q curve of choice.

As it is, this may end up being far from what we observe in games where 2/3 of characters dump int (guess that is selection bias for going adventuring)

[Waterdeep Merch]

It's messy, and IQ doesn't really communicate intelligence all that well. Besides all the regular problems with IQ, the intelligence attribute in D&D applies to rote memorization, recollection, and observation.

IQ would actually better apply to wisdom, as wisdom represents problem solving and application. That's what the IQ test attempts to measure.

[Naanomi]

If you want to assume that the 3d6 roll mapped well to the standard distribution assumed by the Stanford-Binet (which I wouldn't but it is the only way to make this conversation meaningful) the map is as follows:

INT = IQ
3 - 57
4 - 66 <- Diagnostically eligible for a 'mental retardation' diagnosis under the DSM IV-TR
5 - 72
6 - 78
7 - 83 <- Statistically identifiable as 'below average'
8 - 88
9 - 93
10 - 98
11 - 102
12 - 107
13 - 112
14 - 117 <- Statistically identifiable as 'above average'
15 - 122
16 - 128
17 - 134
18 - 143 <- Meets Mensa requirement for 'genius'

If I attempt to 'extrapolate the curve' on a simple statistical model (bad math in the behavioral sciences but I can do it on paper for dubious results) I find...

1 - 16 <- virtually no cognitive engagement, below 20 and learning is generally considered impossible due to massive deficits in processing and memory
2 - 42 <- the estimated 'IQ' of most 'smart' animals (monkeys, dolphins, crows)... still minimal language impairment present
19 - 158 <- about Einstein levels
20 - 184 <- just beyond measuring in the normal battery of IQ tests

The scale gets even more wonky beyond 20... 21 putting me in the 230 IQs somewhere, beyond what is even estimated for human capability (which makes sense I guess)

[Cl0001]

That's one thing I don't like about people playing d&d. Players act like any stat below a 10 is somehow incompetent at completing the simplest tasks which involve that skill. But in reality, as stated above, stats between 8-12 are average. That being said, I have a hard time finding Einstein's intelligence to be as low as 14, unless everyone in the game has a heightened average intelligence. Because, if Einstein, one of the smartest people on earth was only a 14, can anyone ever achieve a 20?

[RSP]

This comparison also doesn't really take skills (or rather Proficiency in skills) into account.

The +Prof Bonus goes a long way towards what you know/remember about a given subset of information.

Very few rolls are pure Int and it's perfectly within the rules to have a dumped Int yet still be really good at skills.

[Zanos]

3d6 has a standard deviation of ~2.96, IQ has a standard deviation of ~15 (depending on test, but it's supposed to be 15).
3d6 averages 10.5, IQ is supposed to average 100, but most tests actually average a bit higher (all the children are above average).

This gives a conversion formula of almost exactly Int*5+50.

So if you assume IQ and Int correspond by a roughly linear formula and measure the same thing then Int 8 is an IQ of 90 or so, which is perfectly functional.

This is what I go with. IQ tests kind of collapse at more extreme ranges than that anyway.

[obryn]

I remember reading that INT=IQ*10 back in the AD&D days.

But IQ is a really questionable real-world measure with a whole host of problems.

[Findulidas]

I thought the whole notion of IQ-testing had been refuted and therefore IQ was no longer used in any reputable sense.

[Armored Walrus]

If you want to assume that the 3d6 roll mapped well to the standard distribution assumed by the Stanford-Binet (which I wouldn't but it is the only way to make this conversation meaningful) the map is as follows:

INT = IQ
3 - 57
4 - 66 <- Diagnostically eligible for a 'mental retardation' diagnosis under the DSM IV-TR
5 - 72
6 - 78
7 - 83 <- Statistically identifiable as 'below average'
8 - 88
9 - 93
10 - 98
11 - 102
12 - 107
13 - 112
14 - 117 <- Statistically identifiable as 'above average'
15 - 122
16 - 128
17 - 134
18 - 143 <- Meets Mensa requirement for 'genius'

If I attempt to 'extrapolate the curve' on a simple statistical model (bad math in the behavioral sciences but I can do it on paper for dubious results) I find...

1 - 16 <- virtually no cognitive engagement, below 20 and learning is generally considered impossible due to massive deficits in processing and memory
2 - 42 <- the estimated 'IQ' of most 'smart' animals (monkeys, dolphins, crows)... still minimal language impairment present
19 - 158 <- about Einstein levels
20 - 184 <- just beyond measuring in the normal battery of IQ tests

The scale gets even more wonky beyond 20... 21 putting me in the 230 IQs somewhere, beyond what is even estimated for human capability (which makes sense I guess)

Rough as you feel this is, I find it enlightening.

As you said, 21 is beyond human capability, only nonhumans or those using some sort of magic can achieve it. 2 being monkey/dolphin smart and as low as 7 still being statistically valid as "normal" are things I wish people would keep in mind when trying to roleplay.

As other have pointed out, of course, there are other considerations that shade how "smart" someone is, which is, IMO, all the more reason not to go "8 Int, I'm dumb."

[Naanomi]

I thought the whole notion of IQ-testing had been refuted and therefore IQ was no longer used in any reputable sense.

It is still part of the diagnostic process of intellectual disabilities of various types. It is better at differentiating between low levels of IQ than high levels, and the non-verbal measurement tools have some decent validity for that purpose even for young age or highly impacted individuals.

There is also a bit more validity in some of the other 'not Stanford-Binet' Intelligence measuring tools, like the Weschler and a few of the other 'multiple intelligences' models

That being said, the numbers above are fun to look at but not particularly useful for a variety of reasons (not the least of which is that IQ and DnD intelligence don't correspond directly)

[Armored Walrus]

I thought the whole notion of IQ-testing had been refuted and therefore IQ was no longer used in any reputable sense.

I'm not trying to use it in any reputable sense. :P

[Findulidas]

As I suspected then. I dont think they use IQ for anything in my country. Im pretty sure Ive read that mensa dont really take it all that seriously either.

I'm not trying to use it in any reputable sense. :P

Fair enough, Im probably just objecting as a knee jerk reaction due to the fact that there are still many people who dont seem to know this.

[Zanos]

I thought the whole notion of IQ-testing had been refuted and therefore IQ was no longer used in any reputable sense.

What IQ measures is IQ. It doesn't necessarily mean intelligence, but it is a test to assess certain aspects related to intelligence, and is directly curved such that 100 is the median.

So while I wouldn't say that low IQ makes you dumb, it does make you bad at the stuff IQ tests measure.

I dont think they use IQ for anything in my country. Im pretty sure Ive read that mensa dont really take it all that seriously either.

Mensa is explicitly a high IQ society, so I would think they take it seriously.

[PeteNutButter]

As I suspected then. I dont think they use IQ for anything in my country. Im pretty sure Ive read that mensa dont really take it all that seriously either.

Fair enough, Im probably just objecting as a knee jerk reaction due to the fact that there are still many people who dont seem to know this.

IQ is very much a real thing. In part it measures a persons ability to take IQ tests, but it is also carries over to some real world applications.

It is commonly refuted as people dislike some of the conclusions that people can incorrectly draw from IQ, such as some people are incapable of doing x or y and the like. While people ought to be careful to type-cast people by a number, it is also wrong to discard the idea that some people have more cognitive ability than others.

I think for the purposes of D&D the best thing to say is the smartest person you know probably doesn't break 16 int. A super genius is probably an 18. Average is 8-12. No need to involve IQ.

[willdaBEAST]

IQ is very much a real thing. In part it measures a persons ability to take IQ tests, but it is also carries over to some real world applications.

It is commonly refuted as people dislike some of the conclusions that people can incorrectly draw from IQ, such as some people are incapable of doing x or y and the like. While people ought to be careful to type-cast people by a number, it is also wrong to discard the idea that some people have more cognitive ability than others.

I think for the purposes of D&D the best thing to say is the smartest person you know probably doesn't break 16 int. A super genius is probably an 18. Average is 8-12. No need to involve IQ.

The issue is that an IQ test skews terribly in favor of Western educational practices and overemphasizes pattern recognition over things like emotional intelligence. I agree that it has some value, but it should never be your only evaluation of a person's intelligence.

I take exception to the idea that dolphins have an intelligence of 2, they're probably smarter than an ogre. Also there's a huge difference between monkeys and apes when it comes to intelligence.

[Naanomi]

True I greatly generalized in that chart up there, though animals (even very smart ones) have big differences (both struggles and superiorities compared to humans) that the Stanford-Binet is woefully underequiped to detect so it may not make big distinction between a monkey and an ape's performance on the test.

Anyways, various intelligence tests are pretty important in my work (special education) in terms of eligibility and sometimes in differentiation between specific disabilities. It is far from the only piece of evidence we rely on but it is a useful piece of information... reading a file and seeing 'IQ 60' on a test results gives me a lot of idea about what kind of student to expect on my observation

[Armored Walrus]

The issue is that an IQ test skews terribly in favor of Western educational practices and overemphasizes pattern recognition over things like emotional intelligence. I agree that it has some value, but it should never be your only evaluation of a person's intelligence.

I take exception to the idea that dolphins have an intelligence of 2, they're probably smarter than an ogre. Also there's a huge difference between monkeys and apes when it comes to intelligence.

None of which is really germane to the purpose of me asking about this; which was to refute the characterization of a "dumped" 8 INT character as being "dumb."

IQ is a useful enough tool to discuss the point I was trying to make. A player who portrays their character with 8 INT talking like "Me smash you face, you not good person" is way overplaying it. 8 INT is 'normal' intelligence. These are not people who just sit drooling in a corner until it's time to get up and swing their sword.

[Zanos]

The issue is that an IQ test skews terribly in favor of Western educational practices and overemphasizes pattern recognition over things like emotional intelligence.

I have to inquire as to when people without a western education would be exposed to IQ testing in significant quantities. Even countries that aren't Western like Japan and China have very Western educational systems these days.

Emotional intelligence is very difficult to measure on it's own. I think you'd struggle for any kind of multiple choice test for that.

I agree that it has some value, but it should never be your only evaluation of a person's intelligence.

Agreed.

[Naanomi]

Yes the hulk-speak or 'bad disability emulation' is unfortunate. You only expect to see the basics of lexicon limitations at around IQ 70 (so... INT 5?) and effective communication and syntax (albeit sometimes with limited vocabulary) as low as 40 (INT 2?).

Though... by the same token, if you played an Orc and rolled a '3', this IQ chart would put you in the completely nonfunctional range at INT 1

[Zanos]

Yes the hulk-speak or 'bad disability emulation' is unfortunate. You only expect to see the basics of lexicon limitations at around IQ 70 (so... INT 5?) and effective communication and syntax (albeit sometimes with limited vocabulary) as low as 40 (INT 2?).

Though... by the same token, if you played an Orc and rolled a '3', this IQ chart would put you in the completely nonfunctional range at INT 1

Not sure about 5th, but in older editions characters with an intelligence lower than 3 weren't considered to be playable characters.

[Naanomi]

Not sure about 5th, but in older editions characters with an intelligence lower than 3 weren't considered to be playable characters.

Not explicitly in 5e. I would say that most players play 'stupid characters' like someone with an IQ around 40... INT 2 by my extrapolated chart there

[Armored Walrus]

Not explicitly in 5e. I would say that most players play 'stupid characters' like someone with an IQ around 40... INT 2 by my extrapolated chart there

Yep, whereas, the reality is closer to "your 8 INT character is probably pretty close to as smart as you are."

Although maybe dumbing down these characters is the consequence of the fact that your 20 INT wizard is being played by a player of only normal intelligence, so to appear to be a supernatural genius, the 8 INT guy has to be dumb.

[Unoriginal]

QUOTE=Armored Walrus;22120812]Just curious if anyone knows of any work that's been done to map the statistical distribution of IQ's among the real world human population compared to the statistical distribution of likely attribute scores if you just rolled 3d6 for each person on earth, so that it can be said that, for example, 8 Int = IQ of 80, 14 INT = IQ of 120 or whatever?[/QUOTE]

There is no correlation between IQ and D&D 5e0's INT.

I'm mostly curious because I don't buy into the (edit: seemingly) widely held conception that 8 INT = dumb. My guess is that 8-12 INT is within "normal" range, and it's only above and below that that you become truly special. In one way or the other.

Why would you not buy into it? It's true.

Well, 8 is just "not very bright", slightly under your average joe's capacities.

[Frozen_Feet]

In AD&D first edition, Intelligence was explicitly called out as analogous to IQ. Later editions are vague about designer purpose, but for most part you can presume Intelligence in D&D is meant to measure the same things as IQ. As IQ distribution is modeled with a bell curve, and 3d6 rolls create a bell curve, it is trivial to match these and earlier poster have already shown how.

As for how widely used IQ is in the real world? Its original use was to identify school children in need of special education. In that use, it has largely been replaced by more specific tests meant to diagnose specific learning impairments. General intelligence tests (general intelligence being what IQ is meant to measure) are perennial favorites of armies and workplaces around the world, but the kicker is, they're usually not graded using IQ. For the most part, while IQ correlates well with a number of things, it usually better to just measure those things directly.

IQ continues to be used as a statistical tool for large scale studies, but it's less because it's a good measure and more because alternative measures suck even worse.