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: 6848719" data-attributes="member: 16814">Multiquote incoming -- I took the weekend off.They aren't interval data.  Do some checking on the source material, there are many points of data that clearly state that IQ is ranking data, and has no definable interval.No, I'm reading your posts very clearly, you're just not following my argument.  It is, again, simply that the distribution of IQ scores, despite what you may think that means, is invalid for comparison to other distributions.  Period.  The forced spreading of the distribution to match relative occurrence in the general population distorts the distance between scores.  IE, the actual distance between a 120 and a 130 IQ doesn't match the distance between any two numbers of a 3d6 spread.  Comparing the two as if you can draw a valid conclusion is false, and you shouldn't do it.Quote the numbers of IQ distribution when I've clearly stated that those very numbers are a bodge, a false narrative set up to provide a way to do some kinds of useful comparison using stats only within the IQ dataset shows that you're completely failing to understand my point.  Those numbers are bullsh*t, pure and simple, and cannot be used as justification to compare to any other distribution.  Just because you can perform a mathematical operation on a given set of numbers does not mean that the answers have meaning.  Averaging ordinal data is one such situation, and that's at the core of the false distribution of IQ scores. Only in the sense that I'm saying the 3d6 mapping argument is deeply flawed to the point that it's not even wrong.  Max's claimed mapping of INT x 10 = IQ has exactly as much validity as the 3d6 to IQ distribution mapping -- they're both made up models that have little to no bearing on anything real.  I disagree with Max's mapping, for the record, and don't find it valuable.  But I don't quite find it the perversion of statistics that the 3d6 argument is, and, as I've previously noted, I have issues when it comes to bad stats.  It's a known character flaw, but I still indulge it.So does the 3d6 roll: only 1 in 216 (= (1/6)^3) people will have INT 18.But 1 in 216 is approximately 0.5%; hence an 18 INT is far more common, in the imagined population of D&D PCs, than is 180 IQ in the real population of human beings.As far as I can tell, nothing in the above reasoning requires treating IQ, or INT, as measuring some quantity; nor does the notion of a "normal distribution" play a role in the reasoning; nor does the reasoning require treating the mean or SD of any data set as meaningful in the way that you are objecting to.I'm not a mathematician or statistician, but to me it seems to be a rather straightforward argument about likelihoods.(And since writing this reply, I see that   [MENTION=6787503]Hriston[/MENTION] has said the same thing as above in post 816.)</blockquote>IQ scores provide a ranking, and also indicate a likelihood - so IQ 180 means that X% of people will be at or above that IQ (for some relatively small X, many decimal places less than 1).[/QUOTE]Theoretically, yes, the model has been built to do exactly this.  But, tell me, how much smarter than someone with an IQ of 170 is someone with a 180?  How much smarter are they than someone with a 50 IQ?  You cannot say, because IQ is ordinal -- it's nothing more than numerical ranking.  Higher scores are smarter (nominally) than lower scores.  That's all you can really say with that data set.  The forced mapping to population doesn't mean that IQ has any kind of even distribution of intelligence across that population, just that the rankings picked are evenly spaced.I'll go back to the race argument. You can also say that 1% of the people will be in the top place of a race, and X% will be in the middle place, and Y% will be in the bottom place (places are really bins, in this case, not discrete finish positions).  But you cannot use this data to say anything more than the top place is faster than the middle place is faster than the bottom place.  Even if it turns out relatively accurate in predicting the % population that will be in each bin, the size of each bin is arbitrary and the distance between bins is arbitrary.  This means that you can't compare this distribution with any other distribution because, and this is important, distributions only have meaning if the data they model are interval or rational.  3d6 is a rational data set.  It's distribution is very useful in many things.  But, to the case at hand, someone that has an 18 intelligence is exactly 1 higher than someone that has a 17 intelligence.  They are 1/18 more smart.  This statement is factually correct.  As is someone with an 18 intelligence is exactly six times smarter than someone with a 3 intelligence.  If you're having cognitive difficulties agreeing with the above paragraph, don't feel bad -- it makes absolutely no sense.  But this is because we, as people, think of intelligence as something generally immeasurable.  Like IQ, we can note that someone is smarter than someone else, but we cannot quantify it.  This is why IQ is the way it is -- ordinal data that just ranks, it doesn't measure.  But 3d6, being rational, DOES apply concrete meaning to the numbers.  18 is three times 3, period.  But for intelligence, we'd never actually say that.This is the crux of my argument.  If you map a 3 on 3d6 to a 50 IQ and a 18 on 3d6 to a 150 IQ, you are saying that 150 IQ is six times smarter than 50 IQ because 18 is six times larger than 3.  This is clearly impossible, and this is directly because, despite the fancy math terms and the apparent similarities between the two distributions, they are fundamentally not the same thing.Statistics lies to you, and does so by cleverly swapping the data for a parameter of the data, and then doing lots of math on that parameter.  It's very easy to forget that the thing being mathed isn't the data, but the parameter, and parameters are almost always an arbitrary choice.  There's nothing preventing you from choosing a parameter and/or test when you apply statistics, and those choices can be woefully misleading if you forget that what you did isn't your data.  Stats can be, and is, a powerful tool, but you always need to remember to carefully go back, note your choices of test and parameters, and make sure you didn't fool yourself and convince yourself you did a good thing because you did math, and math is always right.  Math does tend to be always right, but people use math, and people are wrong all of the time.  Math doesn't fix that.I like this post.  I have some minor quibbles here and there -- mostly in statements about the camps that are too strong or slightly off, but, overall, strong post.  The one major quibble I have I'll highlight here:I don't care a lick, either.  I may care if the INT 5 character solves the puzzle, though.I don't expect my players to be stupid even if they play a stupid character.  I don't discourage table talk (it's not worth it to me), and my players often share ideas between them and then the appropriate character(s) will take action.  Sometimes, this does mean that a 5 INT character does something smart, often it means they don't, and a higher INT character gets to do the smart thing.  Really depends on the dynamic at the moment - my players work this out themselves, I don't have to have any input.  But, on the average, the 5 INT character will not be solving many hypothesized puzzles.  His/her/its player can participate as much as they'd like.I actually find this works well to manage spotlight time, too.  The high INT characters get to have spotlight time because they effect the solution to tricky puzzles (in my game more often mysteries than actual puzzles) that are often worked out by the players as a whole instead of the smart players getting that spotlight time regardless of their character's INT scores.  I agree, but this post and a following post I didn't quote seems to be going down the route of treating acting smart despite a low INT as 'not roleplaying'.  I have issue with that definition of 'not roleplaying'.  Whether or not that's bad roleplaying, though is really up to the table at hand.  There are no globally objective measures of good or bad for something as subjective as roleplaying, but there very much can be local measures of it.  Hey, now, I may disagree occasionally with [MENTION=42582]pemerton[/MENTION], but I don't consider him to have expertise on having a 5 INT.  He's clearly at least a 14.Not quite.  It's still an appeal to authority even if the person has relevant authority if the form of the argument is that they are correct because of their authority.  You're correct that it's not automatically fallacious -- a bad argument isn't necessarily a wrong one and informal fallacies only speak to bad arguments (formal ones are the ones automatically wrong) -- but the general way to tell if a fallacy is in place is if the fallacy appears in place of an argument.  "I'm a doctor, so you're wrong" is an appeal to authority, even if the argument is correct.  In that case, it's still a A2A, it's just not a fallacious one.  Now, "I'm a doctor, and because of that I know these things (provided), and those things are in contradiction to your claims," is not an A2A.  Yes, the person cited their relevant authority, but they used that authority as bone fides for the the counter arguments they provided.  Hence, not A2A.[/QUOTE]

Verification