<p>
</p>
<p>Well I did say it was an imperfect example, and that was because Harvard is already selecting outliers (for the most part).</p>
<p>
</p>
<p>I would, in general, agree with that description, and I have no doubt that these days, the makers of IQ tests strive to do just that. However, I also don’t believe that, even with their best efforts, they will ever truly be able to measure such an abstract concept with a single number without at least a semi-significant bias due to other factors.</p>
<p>In my mind, it is like trying to put a single number on how attractive a person is. Sure, guys will sit down and say “Oh, she’s a 9” but does it really mean anything? Is it scientific? No. People have tried to quantify these things, such as using face symmetry and other more quantitative measures, but it still can’t account for the more abstract portions of human attraction. With an IQ test, you will measure innate cognitive ability, but you will never be able to truly eliminate all the noise and all the other factors that would show up, even with the most objective of tests currently imaginable.</p>