Archive for Wednesday, June 18, 2008

SAT writing section predicts freshman grades

June 18, 2008


— The controversial new writing section of the SAT is a better predictor of college freshman grades than its math and critical reading sections, according to findings released Tuesday by the organization that administers the test.

The conclusion by the College Board partially rebuts critics who argue that the writing section isn't a good measure of writing skill and favors women and white students.

Overall, however, the studies found that the new and longer SAT test is no better at predicting success than the old one.

"I think the main point is that both tests are very valid," said Laurence Bunin, senior vice president of operations at the College Board, a non-profit in New York City.

The organization revamped the traditional SAT in March 2005 by adding a new section that included multiple-choice grammar questions and a 25-minute essay. It eliminated analogies and quantitative comparisons from the test and added shorter reading passages.

The test fee went up from $29.50 to $41.50, a $12 increase. The length went up from 3 hours to 3 hours 45 minutes.

The College Board's studies found the new SAT predicted college success better than high school grades for minority students. For white students, high school grades were a better predictor. A large majority - 69 percent - of the sample was white.

The organization examined the test scores of 151,316 students from 110 four-year colleges and universities in the United States.

Males, Hispanics, African Americans and Native Americans earned lower GPAs than their SAT scores predicted, the College Board found. Females, Asian Americans and Caucasians did better than their SATs forecasted.

"The SAT is a national fair benchmark at an era where grade inflation at the high school level is a major problem," said Bunin.

Robert Schaeffer, a spokesperson for FairTest, an advocacy group that opposes standardized testing, disagreed. He said that the studies' most important finding was that the more expensive and longer version of the SAT was no better than the old one.

"Their study is a devastating indictment of the new SAT," he said.


canyon_wren 9 years, 11 months ago

Interesting article. I think it is great that there is a way to show that GPAs are not good predictors for minorities--grade inflation really helps nobody. About time we were made aware of that. I wish there had been a bit more discussion about the writing part of the SAT and what criteria are used to "measure" that--which I thought would be included. Good writing has been de-emphasized for a number of years now. I have had occasion to read writing sections that were part of the GED exam and found those extremely depressing--even coming from students who did very well on the rest of that exam. Of course, I realize that attempting to compare work on a GED exam with the SAT is totally illogical but would still be interested in what they are finding on the SATs.

Commenting has been disabled for this item.