# Damuri Ajashi, you are a moronic troglodyte of a crackpot **URL:** \ **Category:** The BBQ Pit\ **Created:** [February 17, 2012, 3:28am UTC](https://boards.straightdope.com/t/damuri-ajashi-you-are-a-moronic-troglodyte-of-a-crackpot/613099 "2012-02-17T03:28:21Z")\ **Posts on this page:** 1\ **Showing post:** 142
**Author:** ![FixMyIgnorance](https://avatars.discourse-cdn.com/v4/letter/f/7feea3/32.png) [@FixMyIgnorance](https://boards.straightdope.com/u/FixMyIgnorance)\ **Post date:** [February 24, 2012, 2:55pm UTC](https://boards.straightdope.com/t/damuri-ajashi-you-are-a-moronic-troglodyte-of-a-crackpot/613099/142 "2012-02-24T14:55:44Z")
Okay. > [@](#): > > I’m not sure what your point is here. The claim is that the number of Asian admittees increased significantly, not that that the number went “skyrocketing.” Do you dispute this claim? A simple yes or no will do. Point is, no, it doesn’t increase significantly. It also largely depends on how admissions are conducted. If all your school cares about are grades and SATs, then yes, you’ll see higher populations of Asians/Jews/etc assuming they want to go to that particular school. But most top schools don’t admit solely by grades/SATs. > [@](#): > > By the same reasoning, I could claim that there is insufficient proof that smoking cigarettes reduces life expectancy. After all, there might be some missing variable being overlooked by epidemilogical studies. And it’s your burden to rule out such a missing variable. No. There is plenty of proof that smoking causes life expectancy to drop. We have access to more data, more of the variables involved, well-randomized test groups, and plenty of highly-successful prediction models that can explain a vast majority of the variance and reveal significant relations between smoking and damage to health. Just because you can always point to a nonzero gap in variance doesn’t mean what you _can_ explain suddenly goes out the window. There can always be missing variables. The point is whether or not the variables you _do_ have are sufficient in capturing enough to yield predictive results. There’s a big difference between a model that captures 10% of the variance and one that captures, say, 95%. > [@](#): > > For me, the reason is that (apparently) Asian admittees have significantly higher test scores than people of other racial groups. > > … > > Of course none of this proves anything, but Occam’s Razor and common sense points strongly to one simple conclusion. Except the common-sense conclusion is not necessarily well-grounded because “common sense” would tell most people that higher test scores always = higher chance of admission, when this is not the case. > [@](#): > > I don’t see why this would result in higher SAT scores for Asian admittees. Can you explain? Diminishing marginal utility = the gains with each increase matter less and less. The “gain” in admissions chances is greater between a 2000 and a 2100 versus a 2300 and a 2400. This also makes sense because of the way the test is curved. At the upper end, the slightest mistakes can tank your score. As you get lower and lower on the chain, mistakes are easier to absorb without the same loss in score. It’s curved. The point being made here is that for top schools, the fact that Asians have higher scores doesn’t necessarily mean much. The marginal gains are tiny, so it doesn’t make much sense to just naively hold SAT levels constant and compare admission rates between races. You’ll get really misleading results that way that in no way necessarily imply anti-Asian discrimination. It just means the extra points Asians get on the SAT don’t help their chances all that much when both Asians and whites alike have great scores. Admissions officers aren’t going to split hairs over the difference. > [@](#): > > Would claim that the bar is roughly the same for Asians and non-Asians who are non-athletetic recruits and come from the same geographic area and socio-economic status? Not necessarily. But that still doesn’t imply anti-Asian discrimination. For instance, if I am a white person in a largely Asian population, I could be of similar socioeconomic status but have an easier time with admissions because I’m on the under-represented side from a less-diverse region. That would be a homogenous trend selected against for the sake of diversity. > [@](#): > > Homogeneous in terms of race? Or something else? Homogenous in terms of _any_ trend you could lace through a particular race or any other clustered metric for that matter. > [@](#): > > What besides what I listed? Already answered this like 2-3 times. > [@](#): > > The fact that you need to start throwing personal insults at this point suggests to me that cognitive dissonance is starting to build in your mind. No. This is the BBQ Pit. You’re supposed to call each other names here, twat. > [@](#): > > I asked a yes or no question, and you did not give a yes or no answer. My question was not an unfair one; it did not contain any presuppositions. You’re quibbling over a simple yes/no while claiming to want to understand the position more. It would have been misleading to say “no, that’s pretty much not it” because it implies that your answer was wrong. It wasn’t wrong, it was just incomplete. That’s why I said “Sort of/yes and no” and gave _more detail_ which is by no means _evasion_ and you’d have to be a pretty thickskulled fuckup to think that. --- _[View the full topic](https://boards.straightdope.com/t/damuri-ajashi-you-are-a-moronic-troglodyte-of-a-crackpot/613099)._