Spin Off Logic Problem From Genius-Religion Debate

[David Sklansky]

Regardless of the subject, what are we to make of the fact that, lets say, the percentage of people with an IQ of x who believe Y to be true, is x/6? This is a statistic after a survey. Not a scientific formula

Of course we must stipulate that the above statistic is not known to anybody before they come to their conclusion about Y.

Since no one has an IQ of 300, there is no group of people who are better than even money to believe Y. And there is no guarantee that the formula would hold for them anyway. Of the very smartest people alive about 30% believe Y. In other word even among the very smartest people, at least 70% believe Y isn't true. (Remember again that the results of this survey were not known when opinions were first expressed.)

On the other hand there is the unmistakable pattern that the smarter you are, the more likely you are to believe Y.

Armed with this information, but with no information as to what Y is about, are you justified in believing that Y is probably true?

PS To avoid muddying the waters, I will specify that anyone who believes Y to be true does so with the same degree of high certainty. Likewise those who don't believe Y. Perhaps 95%.

[bunny]

As long as Y is the kind of thing you score points for getting right in an IQ test.

[hitch1978]

[ QUOTE ]
As long as Y is the kind of thing you score points for getting right in an IQ test.

[/ QUOTE ]

Like seeing a multiple choice question, assigning probability's and opting for the most likely you mean?

[David Sklansky]

[ QUOTE ]
As long as Y is the kind of thing you score points for getting right in an IQ test.

[/ QUOTE ]

This is far from clear. Notice that if you only surveyed very smart people you would have no reason to bet Y was correct.

[bunny]

[ QUOTE ]
[ QUOTE ]
As long as Y is the kind of thing you score points for getting right in an IQ test.

[/ QUOTE ]

Like seeing a multiple choice question, assigning probability's and opting for the most likely you mean?

[/ QUOTE ]
Yeah, if Y is something like that, then people who score highly on IQ tests are more likely to get it right. I'm not convinced that scoring highly on an IQ test means you're more likely to get any old question right - why would anyone think that?

It's probably true that there are a class of problems for which IQ-successful strategies are not optimal.

[Phil153]

This is an excellent point that invalidates some of the points made in the other thread - namely, the one about geniuses in the 1600s made by Mendacious.

I'd agree with the Y-truth correlation on the stipulation that the subject matter required to form the opinion is broadly understandable and available to all and not highly esoteric. It may also need to be limited to a single culture/country. I can think of some limited domains of knowledge where the smartest experts can get it wrong for reasons of groupthink or similar, where the gut feelings of the less intelligent are more likely to be correct. But that's certainly not the case for religion, evolution, political theories, social theories and so on.

[bunny]

[ QUOTE ]
[ QUOTE ]
As long as Y is the kind of thing you score points for getting right in an IQ test.

[/ QUOTE ]

This is far from clear. Notice that if you only surveyed very smart people you would have no reason to bet Y was correct.

[/ QUOTE ]
Sorry - I read too quickly. Didnt notice that most of them thought it was incorrect. (EDIT: Despite the fact you went out of your way to make it abundantly clear)

[Phil153]

[ QUOTE ]
Yeah, if Y is something like that, then people who score highly on IQ tests are more likely to get it right. I'm not convinced that scoring highly on an IQ test means you're more likely to get any old question right - why would anyone think that?

[/ QUOTE ]
?

Do you really think that the group of people who can:

- Correctly recognize patterns
- Think quickly and with agility
- Understand mathematical relationships
- Understand logical relationships
- Have good reading comprehension
- Have a solid vocabularly,

are going to be outperformed on questions of fact or considered opinion by the group of people who:

- Can't recognize patterns as well
- Can't think as quickly or with agility
- Don't understand mathematical relationships as well
- Don't understand logical relationships as well
- Don't have as good a reading comprehension
- Don't have as solid a vocubularly

bunny: WTF?

[bunny]

No I think the first group will outperform the second on a huge number of such questions. I do think there are likely to be questions for which the IQ-maximising strategies are not optimal (similar to the groupthink you mentioned above). In that small subset of questions, high IQ will be a liability.

EDIT: Also the problem with IQ tests is that they test all of those things together. Someone in the second group may be much, much better than me in all of the categories you list except for vocabulary (for example). I will then score a higher IQ than them in a test situation even though they will be more likely to get an IQish question right in non-test conditions (once it was patiently explained to them in simple language)

[Phil153]

Let's hear them.