Three years ago, before any of my kids had reached the age to take the SATs, I noticed an interesting piece by Charles Murray on the tests. Murray is always interesting, of course, but I was curious about his take on the SAT because his views on IQ are well known.
Murray argued that the SAT should be scrapped. His case was not (no surprise) the usual indictment of the tests as culturally biased. Instead, he argued that the SAT is unnecessary and, in some ways, counterproductive.
The SAT began as a way for colleges to identify bright students from less than stellar high schools and give them an opportunity. Admission committees might discount excellent grades from inferior schools, but scores on an "aptitude" test (they later changed the word to "assessment" to avoid the accusation that the SAT was measuring IQ) could be revealing. Murray suggests that he used to think his own performance on the test was what got him into Harvard.
But a study by Saul Geiser and Roger Studley from the University of California seemed to show that the SATs contributed little to predicting a student's success in college, whereas achievement tests and high school grades were more reliable. "Those of us who thought that the SAT was our salvation were probably wrong ... our scores on achievement tests would have conveyed about the same picture to college admissions committees as our scores on the SAT conveyed."
The reality, Murray wrote, is that smart kids tend to do well on tests, whether a pop quiz or the AP exams. But whereas the SAT was originally designed to flag kids who might otherwise have been missed by college admissions committees, it has today become a "corrosive symbol of privilege." Everybody now believes, according to Murray, that wealthier parents can purchase higher scores for their kids through expensive coaching. And while Murray points out that this is not so (coaching adds, at most, a couple of dozen points, according to three studies), it is the case that children of college-educated (and graduate degreed) parents walk away with the best scores. Everyone else is found wanting. "All who enter an SAT testing hall feel judged by their scores," Murray writes.
While the idea of junking those stressful, laborious three-hour tests has its appeal, there are reasons to resist.
If achievement tests were substituted for the SAT, all of the cultural and psychological baggage of the high-stakes test would simply switch over. All would continue to "feel judged by their scores" -- just on a different test. (And Murray may overestimate the importance people attach to scores.)
Additionally, Murray doesn't account for the important male/female difference in test performance, particularly on aptitude tests. (Cards on the table: I write as a parent of three boys.) For whatever reason, during the past 30 years, our society has seen girls outperforming boys at every level of education. The average high school GPA for girls is 3.09. For boys, the average is 2.86. About one quarter more boys than girls drop out of high school, and boys are three times as likely to be expelled. Girls do significantly better at reading proficiency in all grades. And in math, traditionally a male preserve, the two sexes are tied. Women now earn 58 percent of bachelor's degrees and 60 percent of master's degrees in the U.S.
Something is going on. It may be the significant attention the educational establishment has lavished on girls, the lure of video games, the lack of fathers in so many homes, the fact that boys mature more slowly than girls, or maybe none of those. But we do know that whatever may be inhibiting them from excelling in high school as much as girls, boys do score proportionately better on the SATs.
In 2010, a total of 382 students scored a perfect 2400. Of these, 206 were boys, and 176 were girls. (If the writing test is omitted, 1,305 students got a 1600 -- 820 boys and 485 girls.) Among those who scored a 2350, 341 were boys, and 266 were girls. The same rough ratios hold (with one exception) for all of the scores in the top 10 percentiles. At the 90th percentile and below, some of the girls' scores are higher than the boys'. And in the middle range, it's a mixed bag.
So long as college requires mental ability, the SATs will remain a signal that boys with less than perfect high school records may be late bloomers or perhaps were ill served by their schools. But scrapping one of the few remaining avenues for talented boys to show, yes, their aptitude, seems unwise.