Monday, June 29, 2009

U of I Clout: On the performance of clouted students

Like most people, I am troubled by the University of Illinois clout scandal, but when I came across this passage in a recent Chicago Tribune article, I had to cringe:
Admissions dean Paul Pless reported that the school admitted at least 24 "SI," or special interest, students during a four-year span. He said they had lower grades and standardized test scores than the general applicant pool and they lagged behind their classmates once admitted. On average, they maintained a 2.86 grade point average during their first year compared with the 3.2 grade point average for the overall class, he said. One faced "formal disciplinary charges" and left the school.
From the article, it sounds like the clouted students are doing significantly worse than those people who would otherwise be admitted. After all, their GPA is 0.34 below the average student in their class. That's a B- in every class instead of a B. There's clearly a difference, but one could question how important that is.

But, the small difference between the clouted students and the average is not what concerns me. I am more concerned that the Law School and the Tribune make the wrong comparison. Who really expects the students who were bumped from the university in favor of a clouted student to be average? Those who were excluded from the law program in favor of the clouted students were excluded for a reason: they weren't as good as the other legitimate students in the program.

Their application credentials may have looked better than the clouted students' credentials, but the students who were bumped were not just randomly selected. Should we expect those students to obtain the average GPA? If we're really interested in how the clouted students performed, we should compare their performance to the 25th percentile. As a crude approximation, that's a much better comparison group, and it would actually tell us something about how unqualified the clouted students were compared with their actual alternative. Comparing to the average just stacks the deck.

To be fair, this is a messy scandal, and there's no justification for letting political influence sway admissions decisions at a public university. On the supporting points in the story, we don't need to stretch the truth with inappropriate statistical comparisons. The scandal looks bad enough without exaggeration.

No comments:

Post a Comment

Please feel free to share your ideas about this post in the open forum. Be mindful that comments in this blog are moderated. Please keep your comments respectful and on point.