Wednesday, Sep. 03, 2008

SAT scores aren’t great, terrible or terribly significant

WITH SCHOOL TEST results season now under way, we’re in for weeks — months, actually — of bickering over what the numbers mean. So before we get too far along, let’s step back and remember some basics:

• Year-to-year changes don’t tell us a lot; it’s the trend that we need to watch.

• Those year-to-year changes become even less meaningful as the size of the groups being compared decreases, because just a handful of really high or really low scores can cause what look like dramatic shifts in the averages of small groups.

• Comparing raw numbers can be misleading, because we’re all accustomed to grading on a 100-point scale.

And most important: Of all the test scores that will be thrown at us between now and the end of the year, nothing tells us less about how well our state is doing than the SAT scores. There are several reasons for this: The test isn’t taken by all students or even by a random sample of students, but by a self-selected group. The test isn’t designed to measure school performance or even to measure students’ knowledge; it’s designed to predict how students will perform as college freshmen — and many people think it does this badly.

Unfortunately, SAT scores are the first ones released each fall, they’re the ones people pay the most attention to, and that attention tends to fixate on those year-to-year changes.

Average scores in South Carolina increased 2 points this year, to 1,461, while the national average remained unchanged at 1,511. Public school critics were quick to note that the average score for public school students actually dropped, and that it was large gains by private school students that caused our improvement. True enough — and if you look back, you’ll see that public schools out-improved private in other years.

What you’ll also see is that our state’s average in reading and math has increased by 34 points in the past decade (writing is in only the third year of testing) — the fastest improvement in the nation. Leave out private schools, and our 30-point gain is still second-best among public school students nationally. More significantly, not only are we closing the gap, but the gap isn’t nearly as large as it looks. Those totals are out of a possible 2,400 points, which means that students nationally are getting 63 percent of the answers correct, compared to 61 percent for S.C. students.

The most disappointing result was among African-American test-takers, whose average dropped by 5 points, while white students’ scores improved by 6 points. It’s even tougher to know what to make of this than of overall year-to-year changes, since it occurred while the number of black students earning college credit through Advanced Placement exams increased.

But what we do know — from a variety of measures — is that our state still has work to do to close the racial achievement gap that runs throughout the educational system. And that we’ve still got work to do to keep kids in school until they graduate, and then to get them to go on to higher education. And that while we’re making tremendous strides, we’ve still got work to do to make sure elementary, middle and high school students are learning as much as their peers in the rest of the nation — and as much as they need to learn.

We’ll find out more about just how much work we have to do in coming weeks, when we get a look at the results of tests that actually measure such things.