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: 6850352" data-attributes="member: 16814">Feel free to do so, but if you read some of the literature, it will say this very thing.  It turns out this model, while wrong, is useful in some cases, though.  Those cases are all when comparing within the model, and not when comparing the model to other models.No, I misspoke.  18 is one higher than 17, not 1/18 higher.But you can't do that.  The rational data provided by 3d6 means that it actually has a mathematical mean.  You can average the numbers and it means something because the numbers have value and an equal spacing from each other.  The average of 10 and 12 is 11, for instance.  But this isn't true of IQ data.  You can't average IQ data.  The average of a 100 IQ and a 120 IQ isn't a 110 IQ.  The scale doesn't work like that.  The values of IQ are rankings, without assigned value and without equal spacing.  You can't average those numbers.  So you can't determine a valid mean.  Without a valid mean you can't determine a standard deviation.  Without a mean and a standard deviation, you can't form a distribution, normal or otherwise.  So, instead, they pretend you can do these things with IQ for convenience.  A 50 IQ isn't as rare as it is in the population because of statistics, the statistics have been fudged to create that distribution, both by adjusting, as necessary, what 50 IQ means and by pretending you can actually do stats on IQ data.So your statement that you're just comparing the rarity of a 50 IQ with a 3 on 3d6 is false because the 50 IQ doesn't have a statistically determined valid rarity (this is technically an incorrect usage, stats don't have rarities, but it's useful enough in this context), it has an assigned one.</blockquote>

Verification