Over 1000 people took the test. Statistically speaking, it should have included about 50 sociopaths.
Not if LessWrong values truthseeking activities more than the general population, or considers lying/truth-fabrication a greater sin than the general population does, or if LessWrong just generally attracts less sociopaths than the general population. If over 1000 fitness enthusiasts take a test about weight, the statistics re: obesity are not going to reflect the general population’s. Considering the CRT scores of LessWrong and the nature of this website to admire introspection and truthseeking activities, I doubt that LW would be reflective of the general population in this way.
Lies are more than untrue statements; at least, in the context of self-reports, they are conscious manipulations of what one knows to be true. Someone might think they know their IQ because they’ve taken less reliable IQ tests, or because they had a high childhood IQ, or because they extrapolated their IQ from SAT scores, or for a host of other reasons. In this case they haven’t actually lied, they’ve just stated something inaccurate.
Someone could put an IQ when they have no idea what their IQ is, yes, in the sense that they have never taken a test of any sort and have no idea what their IQ would be if they took one, even an inaccurate one. I don’t think many people here would do that, though, because of the truthseeking reasons mentioned earlier.
Mensa is a club not a professional IQ testing center. They’re not even legally allowed to give out scores anymore. Their test scores are not considered to be accurate.
Mensa doesn’t need to be a professional IQ testing center for their normings to be accurate, however. I am also not sure how not accounting for learning disorders would seriously alter IQTest.dk’s validity over self-reports.
However, it’s inaccurate to say that because someone puts their number in the box from IQTest.dk that they’re “equally flawed” to the other intelligence questions. Someone who self-reports an IQ number, any number, may not know if that number was obtained using accurate methodology. It may be an old score from childhood, and childhood IQ scores vary wildly compared to adult IQ scores. It may be an extrapolation from SAT scores, as I mentioned above. There are a number of ways in which self-reported IQ differs from reported IQtest.dk IQ.
LessWrong is going to eat you alive, honey. Get out while you’re ahead.
This reads as unnecessarily tribalistic to me. I take it you think I am an undiscriminating skeptic? In any case, cool it.
I’d expect Less Wrongers to be more likely to be sociopaths than average. We’re generally mentally unusual.
Yeah, I am perfectly aware that the IQ score I got when I was three wasn’t valid then and certainly isn’t now. The survey didn’t ask “What’s a reasonable estimate of your IQ?”.
Not if LessWrong values truthseeking activities more than the general population, or considers lying/truth-fabrication a greater sin than the general population does, or if LessWrong attracts less sociopaths than the general population. If over 1000 fitness enthusiasts take a test about weight, the statistics re: obesity are not going to reflect the general population’s. Considering the CRT scores of LessWrong and the nature of this website to admire introspection and truthseeking activities, I doubt that LW would be reflective of the general population in this way.
That is why I used the wording “statistically speaking”—it is understood to mean that I am working from statistics that were generated on the overall population as opposed to the specific population in question. You are completely ignoring my point which is that you have chosen a position which is going to be more or less impossible to defend. That position was:
I don’t think anyone on Less Wrong has lied about their IQ.
It’s considered very rude to completely ignore someone’s argument and nit pick at their wording. That is what you just did.
Lies are more than untrue statements; at least, in the context of self-reports,
Now it’s like you’re trying to make up a new definition of the word lying so you can continue to think your ridiculous assessment that:
To lie about your IQ would mean you’d have to know to some degree what your real IQ is
By the common definition of the word “lie” producing a number when you do not know the number definitely does qualify as a lie. You’re not fooling me by trying to make a new definition of the word “lie” in this context. This behavior just looks ridiculous to me.
Mensa doesn’t need to be a professional IQ testing center for their normings to be accurate, however.
But they do need to provide a professional IQ testing service if they want their norms to mean something. The iqtest.dk might turn out to be a better indicator of visual-spatial ability than IQ, or it might discriminate against autistics, which LW might have an unusually large number of (seeing as how there are a lot of CS people here).
However, it’s inaccurate to say that because someone puts their number in the box from IQTest.dk that they’re “equally flawed” to the other intelligence questions.
Here you go twisting my wording. I specifically said:
In that way, they’re equally flawed to the other intelligence questions...
The only reason I’m responding to you is because I am hoping you will see that you need to do more work on your rationality. Please consider getting some rationality training or something.
The general population would contain 50 sociopaths to 1000; I don’t think LessWrong contains 50 sociopaths to 1000. Rationality is a truth-seeking activity at its core, and I suspect a community of rationalists would do their best to avoid lying consciously.
I am not sure what “the common definition of the word ‘lie’” is, especially since there are a lot of differing interpretations of what it means to lie. I know that wrong answers are distinct from lies, however. I think that a lot of LessWrong people might have put an IQ that does not reflect an accurate result. But I doubt that many LessWrong people have put a deliberately inaccurate result for IQ. Barring “the common definition” (I don’t know what that is), I’m defining “stating something when you know what you are stating is false” as a lie, since someone can put a number when they don’t know for sure what the true number is but don’t know that the number they are stating is false either.
I don’t know what you mean by “mean something” with respect to Mensa Denmark’s normings. They will probably be less accurate than a professional IQ testing service, but I don’t know why they would be inaccurate or “meaningless” by virtue of their organization not being a professional IQ testing service.
The only way I can think of in which the self-reported numbers would be more accurate than the IQTest.dk numbers is if the LW respondents knew that their IQ numbers were from a professional testing service and they had gone to this service recently. But since the test didn’t specify how they obtained this self-report, I can’t say, nor do I think it’s likely.
IQTest.dk uses Raven’s Progressive Matrices which is a standard way to measure IQ across cultures. This is because IQ splits between verbal/spatial are not as common. It wouldn’t discriminate against autistics, because it actually discriminates in favor of autistics; people with disorders on the autism spectrum are likely to score higher, not lower.
I’m not sure how the bolding of “in that way” bolsters your argument. Paraphrased, it would be “in the way that the user types the IQ score into the survey box themselves, the IQTest.dk questions are equally flawed to the other intelligence questions.” But this neglects to consider that the source of the number is different; they are self-reports in the sense that the number is up to someone to recall, but if someone types in their IQTest.dk number you know it came from IQTest.dk. If someone types in their IQ without specifying the source, you have no idea where they got that number from—they could be estimating, it could be a childhood test score, and so on.
Please consider getting some rationality training or something.
Remarks like these are unnecessary, especially since I’ve just joined the site.
Not if LessWrong values truthseeking activities more than the general population, or considers lying/truth-fabrication a greater sin than the general population does, or if LessWrong just generally attracts less sociopaths than the general population. If over 1000 fitness enthusiasts take a test about weight, the statistics re: obesity are not going to reflect the general population’s. Considering the CRT scores of LessWrong and the nature of this website to admire introspection and truthseeking activities, I doubt that LW would be reflective of the general population in this way.
Lies are more than untrue statements; at least, in the context of self-reports, they are conscious manipulations of what one knows to be true. Someone might think they know their IQ because they’ve taken less reliable IQ tests, or because they had a high childhood IQ, or because they extrapolated their IQ from SAT scores, or for a host of other reasons. In this case they haven’t actually lied, they’ve just stated something inaccurate.
Someone could put an IQ when they have no idea what their IQ is, yes, in the sense that they have never taken a test of any sort and have no idea what their IQ would be if they took one, even an inaccurate one. I don’t think many people here would do that, though, because of the truthseeking reasons mentioned earlier.
Mensa doesn’t need to be a professional IQ testing center for their normings to be accurate, however. I am also not sure how not accounting for learning disorders would seriously alter IQTest.dk’s validity over self-reports.
However, it’s inaccurate to say that because someone puts their number in the box from IQTest.dk that they’re “equally flawed” to the other intelligence questions. Someone who self-reports an IQ number, any number, may not know if that number was obtained using accurate methodology. It may be an old score from childhood, and childhood IQ scores vary wildly compared to adult IQ scores. It may be an extrapolation from SAT scores, as I mentioned above. There are a number of ways in which self-reported IQ differs from reported IQtest.dk IQ.
This reads as unnecessarily tribalistic to me. I take it you think I am an undiscriminating skeptic? In any case, cool it.
I’d expect Less Wrongers to be more likely to be sociopaths than average. We’re generally mentally unusual.
Yeah, I am perfectly aware that the IQ score I got when I was three wasn’t valid then and certainly isn’t now. The survey didn’t ask “What’s a reasonable estimate of your IQ?”.
That is why I used the wording “statistically speaking”—it is understood to mean that I am working from statistics that were generated on the overall population as opposed to the specific population in question. You are completely ignoring my point which is that you have chosen a position which is going to be more or less impossible to defend. That position was:
It’s considered very rude to completely ignore someone’s argument and nit pick at their wording. That is what you just did.
Now it’s like you’re trying to make up a new definition of the word lying so you can continue to think your ridiculous assessment that:
By the common definition of the word “lie” producing a number when you do not know the number definitely does qualify as a lie. You’re not fooling me by trying to make a new definition of the word “lie” in this context. This behavior just looks ridiculous to me.
But they do need to provide a professional IQ testing service if they want their norms to mean something. The iqtest.dk might turn out to be a better indicator of visual-spatial ability than IQ, or it might discriminate against autistics, which LW might have an unusually large number of (seeing as how there are a lot of CS people here).
Here you go twisting my wording. I specifically said:
The only reason I’m responding to you is because I am hoping you will see that you need to do more work on your rationality. Please consider getting some rationality training or something.
The general population would contain 50 sociopaths to 1000; I don’t think LessWrong contains 50 sociopaths to 1000. Rationality is a truth-seeking activity at its core, and I suspect a community of rationalists would do their best to avoid lying consciously.
I am not sure what “the common definition of the word ‘lie’” is, especially since there are a lot of differing interpretations of what it means to lie. I know that wrong answers are distinct from lies, however. I think that a lot of LessWrong people might have put an IQ that does not reflect an accurate result. But I doubt that many LessWrong people have put a deliberately inaccurate result for IQ. Barring “the common definition” (I don’t know what that is), I’m defining “stating something when you know what you are stating is false” as a lie, since someone can put a number when they don’t know for sure what the true number is but don’t know that the number they are stating is false either.
I don’t know what you mean by “mean something” with respect to Mensa Denmark’s normings. They will probably be less accurate than a professional IQ testing service, but I don’t know why they would be inaccurate or “meaningless” by virtue of their organization not being a professional IQ testing service.
The only way I can think of in which the self-reported numbers would be more accurate than the IQTest.dk numbers is if the LW respondents knew that their IQ numbers were from a professional testing service and they had gone to this service recently. But since the test didn’t specify how they obtained this self-report, I can’t say, nor do I think it’s likely.
IQTest.dk uses Raven’s Progressive Matrices which is a standard way to measure IQ across cultures. This is because IQ splits between verbal/spatial are not as common. It wouldn’t discriminate against autistics, because it actually discriminates in favor of autistics; people with disorders on the autism spectrum are likely to score higher, not lower.
I’m not sure how the bolding of “in that way” bolsters your argument. Paraphrased, it would be “in the way that the user types the IQ score into the survey box themselves, the IQTest.dk questions are equally flawed to the other intelligence questions.” But this neglects to consider that the source of the number is different; they are self-reports in the sense that the number is up to someone to recall, but if someone types in their IQTest.dk number you know it came from IQTest.dk. If someone types in their IQ without specifying the source, you have no idea where they got that number from—they could be estimating, it could be a childhood test score, and so on.
Remarks like these are unnecessary, especially since I’ve just joined the site.
In principle, one could make up a number or insert a number other than what they got. But I don’t think a nontrivial fraction of respondents did that.