Notes on the Atrocities
Like a 100-watt radio station, broadcasting to the dozens...

Thursday, February 06, 2003  

Bad Data

Today's LA Times published a poll they conducted on the topic of race and admissions under the headline "Bush's Opposition to Racial Preferences Gets Big Support."

Immediately, I knew something was wrong.

In a nation-wide telephone survey of just under 1400 people, the Times found that 55% disapproved of the University of Michigan admissions policy, compared to 27% who favored it. Or, as the Times wrote: “By a 2 to 1 majority, Americans approve of President Bush's call to strike down a race-based admissions policy at the University of Michigan and say that students should be judged only on their academic records.”

First, let’s take the claim itself. The numbers came from the following question:

“President Bush state publicly that he opposes a program of racial preferences for minority applicants at the University of Michigan. The Bush administration urged the Supreme Court to declare the University of Michigan’s admissions system unconstitutional. Do you approve or disapprove of the Bush administration’s decision to oppose the University of Michigan’s racial preference admissions policy?”

There are a number of problems here. First, the buzz word of “racial preferences.” Who thought this was going to produce unbiased data? The phrase—even for those who don’t recognize it as a Republican buzz word—almost demands an American reject it. We’re not a country of “preference”—we’re a country of merit. Second, the question is as much about whether respondents support the President as it is about race and admissions. And finally, the question is confusing, asking people whether they approved of Bush’s disapproval.

(A better question would have been something along the lines of: “The University of Michigan uses a number of criteria when considering applicants for admission. These include academic performance, athletic and artistic ability, geographic location, and family history with the University. Do you favor using race as one of these factors?)

Now, let’s think about the relevance of a poll like this. If the sample drawn was (as the Times claims) consistent with the population, it means that only 26% of respondents had a college degree. Are people with no experience of college qualified to determine admissions policies? More significantly, the question is a matter of law. Does a poll of random Americans contribute anything to a discussion of the legality of the policy? Obviously it does not.

All it really does is put the heat on the Supreme Court—heat that was turned up once already by the President. As I wrote back on January 15, the politicization of the judiciary doesn’t benefit anyone, and it knocks the whole checks and balances thing out of whack. It’s bad enough this was a bogus survey—it’s worse that it was even done in the first place. Shame on the Times.

posted by Jeff | 11:47 AM |
Blogroll and Links