SAT Scores Drop: Wanted One Whipping Boy

Jay Matthews reports today in WaPo on the sizeable drop in SAT scores, particularly on reading (5 points). The led reads, "The average reading score on the SAT for last spring's high school graduates had its biggest drop in 31 years, the College Board disclosed this morning in a report that decried a decline in composition and grammar teaching in U.S. high schools."

Call me incredulous, but should teaching really take the fall for a test that continues to be re-normed, and re-designed: a test that according to many is not a measure of any set curriculum, but is instead designed to assess skills necessary for success in college (Marchant and Paulson, 2005)? The SAT predicts above all else the future performance of a college freshman's first semester. The SAT has been equated to an intelligence test, measuring students’ ability to learn, not mastery of what was learned (Gose, Selingo, & Brownstein, 2001). ETS loves to distance themselves from any IQ or g comparisons, but doesn't wholly discount them either. And Princeton Review makes promises of 100 point increases on SATs for its paying customers. Basic SAT coaching increases scores by 30 points. See the Boston Globe for solid discussion.

The other fall guy, ETS said in their press conference, is not as many kids taking the test twice. The decline in multiple test taking dropped from 43.8 percent for the class of 2005 to 46.5 percent in the class of 2006. More discussion and less blind following would benefit us all. It may be true that schools should continue to improve composition teaching, but curriculum is a piss poor whipping boy for a brand new SAT that more kids failed take twice.