Upgrade your account to a Community Supporter account and remove most of the site ads.

Reply to thread

Message: <blockquote data-quote="Ovinomancer" data-source="post: 6846210" data-attributes="member: 16814">Again, you misunderstand.  I wasn't saying anything about Maxperson's argument, I was saying that you can't dismiss Maxperson's argument based on your statistical argument because your statistical argument has as much to do with Maxperson's claims as unicorn farts.Personally, I don't agree with or really care about Maxperson's argument re: INTx10 = IQ.  It's just not interesting to me.  I do, however, have a nutty interest in seeing statistical misuse corrected.  I do this not to be pedantic (well, not entirely) but because stats are routinely misunderstood and improperly used by our society and so a little education on how not to be suckered by mathemagic shows is helpful (IMNSHO, of course).  So I'm pointing out that you're improperly applying statistics and trying to explain why.  Please re-read my multiple posts on the inappropriateness of using means, SDs, and normal distributions with ordinal data.No, they don't.  The can't even possess a mean that is meaningful in any way.  That someone did the math that you would use to find a mean in a non-ordinal data set and provided it as a mean doesn't magically make that have meaning.  It's something done improperly because the bad model that results has some value in evaluating IQ scores against IQ scores within the model.  It is utterly false that an ordinal set of data can posses a true mean, a true standard deviation, or be a true normal distribution.  It's double extra utterly false to use the mathemagic tricks to fake those thing as if they can be compared to other distributions in any useful or meaningful way.  They cannot.Let me try to explain this again.  Person A has an IQ of 50.  Person B has an IQ of 100.  Person C has an IQ of 150.  If these values were nominal, you could say that B was twice as smart as A, and C was three times as smart as A, but only half again as smart as B.  But IQ is ordinal, which means we cannot tell how much difference there is between numbers, just that the numbers are ranked in order.  So Person B is smarter than Person A, but we cannot say how much smarter.  I would be very inappropriate to say twice a smart, for instance, because the only information we can glean is the B is smarter to some degree than A.  C is the smartest in this little group, but we can't say how much smarter.  He may be twice as smart as B but only 2 and a half times smarter than A.  Or he may be barely smarter than B.  We cannot tell.  Because of this, even though I could do the math to average the three scores and say that the mean is 100, that's has no value because I don't know that C is the same distance from B as A is, so saying B is the average value isn't possible -- the true average value may be anywhere between A and C.If you can grasp this, you'll see why I keep saying that you stats with IQ is meaningless (outside of the set, and within only in a limited way) and you shouldn't make arguments using those stats.</blockquote>

Verification