Tag: SAT

Squandering Assessment Test

By
Neal McCluskey

Yesterday the annual summary of SAT—formerly Scholastic Aptitude Test—scores came out, and the news was once again disheartening. Indeed, average reading scores hit a record low, and math remained stagnant. Writing scores also dipped, but that part of the test has only existed since 2006.

There are important provisos that go with drawing conclusions about the nation’s education system using the SAT. Most notably, who takes it is largely self-selected, and growing numbers of people sitting for it—some of whom might not have bothered in the past—could lower scores without indicating the system is getting worse. That said, as the chart below shows, no likely amount of self-selection or changing test-takers can account for the overwhelming lack of correlation between spending and scores. Per-pupil outlays have taken off like a moonshot while scores have either sat on the runway, or even burrowed down a bit.

Sadly, this corresponds to the results from long-term National Assessment of Educational Progress exams—which are nationally representative—for 17-year-olds. Again, as the following chart reveals, spending has skyrocketed while scores have, um, decidedly not skyrocketed.

There are factors that make comparing year-to-year SAT scores imprecise. But the trend clearly reinforces what we should already know: we get almost no return for our education “investment.”

Topics: 
Education and Child Policy, General

College Board’s SAT Drop Spin Doesn’t Hold Up

By
Andrew J. Coulson

Nationwide verbal SAT scores fell to their lowest level in years on the most recent administration of the test, and the College Board, which administers the SAT, has an explanation:

Average SAT scores fell slightly for 2011 high-school graduates, as the number of test takers and the proportion of minority students grew, according to a report released on Wednesday by the College Board, which owns the test.

The idea—which has been offered as an explanation of earlier declines—is that the overall average score can fall even if the performance of every participating group was stable or improving—if the groups that tend to score lower comprise a larger share of the total test-taking population than they did in the past. And, indeed, minority students (who often score below white students) now comprise a larger share of the test taking population than ever before.

So: case closed? Nope. If you actually look at the score breakdown for the major race/ethnicity groups (see chart) you’ll notice that only white students’ scores held constant from last year. The scores of all the minority groups declined. And, since 1996, white students’ scores have been flat, those of Asian students have risen appreciably, and those of Hispanic and African American students have declined.

Since there has not been any government program targeted exclusively at improving the achievement of Asian students, these data don’t exactly bolster confidence in the effectiveness of either state or federal education policy. If we want to see improved educational productivity, we might just want to look at more free enterprise education systems that offer schools the freedoms and incentives that actually make it happen.

Topics: 
Education and Child Policy

SAT/ACT Factoid Debunked

By
Neal McCluskey

There’s been a factoid making the rounds during the Wisconsin union standoff that you may have seen. I’m not sure what the ultimate source of the factoid is, but here’s the meat of it as reiterated by a blogger for The Economist:

Only 5 states do not have collective bargaining for educators and have deemed it illegal. Those states and their ranking on ACT/SAT scores are as follows:

South Carolina – 50th
North Carolina – 49th
Georgia – 48th
Texas – 47th
Virginia – 44th

If you are wondering, Wisconsin, with its collective bargaining for teachers, is ranked 2nd in the country.

Now, aside from the factoid, if true, providing no real insight into whether collective bargaining is good or bad for education – there are myriad variables at work other than collective bargaining, none of which does this control for – but the factoid itself is highly dubious. Again, it is hard to find the original source for this, but I looked up 2009 ACT and SAT state rankings, and at the very least it seems highly unlikely that Virginia ranks 44th out of all states. According to the ACT ranking, for instance, Virginia places 22nd, and on the SAT (assuming the linked to list is accurate – I’m doing this fast), it ranked 33rd. It’s hard to see how those would be combined for a 44th place overall finish.

How about the Wisconsin second place-finish? Well, that is accurate for the SAT, but notably only 5 percent of Wisconsin students took the SAT – a negligible rate. On the ACT, which is the main test taken in the Badger State, Wisconsin finished 13th – not bad, but hardly great.

So what does this tell you? Not that collective bargaining is educationally good or bad – like I said, you just can’t get there from here – but that you have to be very careful about your sources of information. Unfortunately, that seems especially true when you’re dealing with education.

Topics: 
Education and Child Policy

Future Teachers Most Likely to Cheat in College?

By
Andrew J. Coulson

The current issue of the Chronicle of Higher Education features a story by a professional ghost-writer of college student papers. One passage in particular caught my eye:

it’s hard to determine which course of study is most infested with cheating. But I’d say education is the worst. I’ve written papers for students in elementary-education programs, special-education majors, and ESL-training courses. I’ve written lesson plans for aspiring high-school teachers, and I’ve synthesized reports from notes that customers have taken during classroom observations. I’ve written essays for those studying to become school administrators, and I’ve completed theses for those on course to become principals….

This is of course the weakest of anecdotal evidence and no one should take it as gospel (particularly the seminary students who apparently also contract out papers to the same ghost writer). But let’s say, for the sake of argument, that it’s true—that ed school students are the most common consumers of fraudulent papers. How could we explain that?

There’s no reason to believe that future teachers are any more ethically deficient than their peers in other fields, so that’s an unlikely explanation. Could it be that ed school students are less well prepared for college? Certainly it’s an uncomfortable truth that the SAT scores of those applying to ed school (both undergraduate and graduate) consistently rank below those of applicants to most other college programs. But it is also widely acknowledged that the academic standards of ed schools are commensurately below those of other college disciplines, so future teachers shouldn’t have any more difficulty completing their assignments than students in other fields.

But there is one way in which education is fundamentally different from every other college discipline: it’s the only one whose students will go on to work in a government monopoly industry. Not only is the hiring process of public school systems less focused on identifying candidates’ academic excellence, there is evidence that it is actively hostile to excellence (e.g., that principals are less likely to hire top-scoring candidates from elite colleges than candidates from less rarefied institutions). What’s more, compensation for public school teachers is generally a function of time served (over which teachers have no control) and degrees conferred (over which they do). This has created demand on the part of teachers for graduate degrees—not necessarily for the acquisition of advanced skills, but for the diplomas themselves, which  amount to valuable cash prizes.

Again, we can’t know from a single ghost-writer’s experience if ed school students systematically cheat more in college than their peers in other fields, but we certainly shouldn’t be surprised if they do. We’ve organized education in this country in a way that decouples skill and performance from compensation, and instead couples compensation to the mere trappings of higher learning (e.g., masters degrees). We’ve created a powerful financial incentive for existing and future teachers to cheat. Maybe not such a good idea.

Hat tip: Bill Evers.

Topics: 
Education and Child Policy