Monday, June 21, 2010

SAT is Racist: Solution? Make The Test Harder.

In a recent CD post, I featured some claims that the SAT test is sexist and "rigged to favor boys," even though the College Board's Board of Trustees is headed by a woman (Youlonda Copeland-Morgan, pictured above), and has slightly more women (16 members) than men (15 members).  Now there's some new evidence that the SAT is also racially biased, and I guess therefore "rigged to favor whites," even though the current Chair of the Board of Trustees of the CollegeBoard (which owns the SAT) is black (Ms. Copeland-Morgan pictured above) and Lester Monts, the past chairman, is black.  

And this part of the new research seems especially confusing:

There's a concept on test-taking research known as "differential item functioning" (DIF). A DIF question is one on which students "matched by proficiency" and other factors, have variable scores predictably by race, on selected questions.  On some of the easier verbal SAT questions, two studies find that the DIF favored white students, but on some of the most difficult verbal questions, the DIF favored black students. The white advantage is larger such that the studies suggest scores for black students are being held down by the way the test is scored, and that a shift to favor the more difficult questions would benefit black test-takers.


At 6/21/2010 7:58 PM, Blogger juandos said...

From the new evidence link: 'In other words, the College Board says that American society is unfair, but the SAT is fair. And while many educators question that fairness of using a test on which wealthier students do consistently better than less wealthy students, research findings that directly isolate race as a factor in the fairness of individual SAT questions have, of late, been few'...

Oh dear! Yet another 'women and minorities hardest hit' banner headline in the making...

Well let's not bother with more difficult questions in these 'politically correct times' let's just go with the Port Chester solution...

At 6/21/2010 8:50 PM, Blogger JamesD'Troy said...

What I really dislike about these reports is how they skew the article to narrowly focus upon the standard Leftist-agitator narrative of 'Whites as oppressor-blacks as victim' while ignoring the reasons why Whites and Asians score significantly better than Blacks and Hispanics.

One of the letter writers from the link raised issues similar to what I was thinking after reading the article.

"Why are Asian Americans the highest performers on the SAT I test (both Math and Verbal sections combined), on the average, despite the fact that they face the obstacles of cultural and language differences, as well as social and economic disadvantages, such as lower parental incomes and parental education. Over 70% of the Asian American SAT I test takers are either immigrants or the children of immigrants.

According the the studies and the data from the CC, the poorest Asian Americans from families in the lowest income levels with parents of less than a high school education outperform blacks on the SAT I test, on the average, from families in the highest income levels with parents with a college degree or more. This is the racial gap that is independent of the social economic gap between the races and ethnic groups. Asian Americans even outperform whites when these parameters are examined.

One cannot explain the better performance of Asian Americans over whites and blacks due to the alleged racial biases of the SAT I. How can a study of the SAT I test exclude the Asian American group, the highest performers (especially on the Math and even on the Verbal sections). If this test is racially biased against blacks, then it must be biased against Asian Americans, but they still outperform whites, as well as blacks."

At 6/21/2010 8:50 PM, Anonymous DeeBee9 said...

I sure wish those who claim that the SAT is biased would cite those questions that they believe are biased. As I recall the "aptitude" SAT tests (quite different from the "achievement" SAT tests) from years ago, it would be interesting to see where the bias could be, especially in the math aptitude test. But, for some reason, examples of biased questions are never given. It's silly to claim bias based on results without citing the biased questions that lead to those results.

At 6/21/2010 9:46 PM, Anonymous Anonymous said...

JunkScience Mom tears the whole story down over at Big Journalism:

Junk Science - The Hidden Truth Behind Claims Of Racial Bias In the SATs

Leftists love the lie. It's what they live for.

At 6/21/2010 10:12 PM, Blogger Unknown said...

After a few years of tutoring students I've realized these studies are worthless. "Hard" questions are just those that most people miss. Most of the hard questions are quite easy, given some test preparation. For math, these questions usually involve strange symbols that express basic math concepts, or finding patterns that are easy to spot if you know what to look for. In the verbal sections the hard questions require students to read boring passages, or to have a grasp of grammar somewhat stronger than what a native speaker should have intuitively. Thus black students missing easy questions but getting the hard ones could simply be the result of different speech patterns among racial groups (for example, many people often use "them" instead of "him" or "her"). The big difference between Asian, whites, and blacks might be test prep. I've always had a disproportionately high number of Asian students and only one black student.

At 6/21/2010 11:00 PM, Blogger Jason said...

If you don't like the answer, change the question.

Sooner or later, we will all realize we've destroyed an expanding economy and a way of life because we were unable to confront simple truths and brutal reality.

The SAT and EXCUSES of gender and racial bias is just another sad chapter in the de-construction of America.

At 6/30/2010 12:39 PM, Anonymous Anonymous said...

People are dismissing the results rather casually and out-of-hand.

I, too, am skeptical of these kinds of claims b/c of their poor (often laughable) track record, but I'd at least want to read more about them before I dismiss them as just another attempt to statistically engineer a finding of bias.

In any case, I certainly am curious to read how they determine whether two students are "matched by proficiency..."

And an interesting point that I haven't seen mentioned yet is that (at least 10 years ago, when I read quite a bit about the SATs) the racial disparity is greater in the MATH section, and this study shows NO problems with the MATH section!! If the verbal sections have a few problematic questions, one could argue that it makes sense to focus more on the MATH scores, which would be DETRIMENTAL to blacks.

So it's damned-if-you-do, damned-if-you-don't

At 7/12/2010 3:21 PM, Blogger Unknown said...

This is utter bs. I am Latina and I aced my SAT. I've just read a bunch of similar articles that talk about these studies. It is pseudo science, and we are not being told all the facts. I agree with JunkScienceMom


Post a Comment

<< Home