TCS Daily

Let's Pretend that Public Schools Work

By James Freeman - September 4, 2000 12:00 AM

This week the College Board announced that SAT math scores are at their highest level in 30 years. On the surface, this seems to be a sign that America's K-12 schools are on the right track, and that the need for school reform is overblown. Guess again.

I don't know why the College Board wants to participate in a political spinning of test results, but nobody should know better than the testing organization how misleading it is to claim that math scores are at a 30-year high. The test has changed a great deal since 1970, and in the 1990s, the changes were particularly dramatic. Kids were allowed to use calculators for the first time and both the math and verbal tests were redesigned to emphasize different skills.

In addition, SAT scores were artificially inflated beginning in 1995. As you may recall, that year the testing service "re-centered" the test to ensure that 500 was the average score on each part of the test - math and verbal. Basically, since the scores range from 200 to 800 on each test, the College Board wanted 500 to be the average score. But most kids were scoring below 500, so the College Board began adding points to each kids' results to bring the averages up - roughly 75 points on the average verbal score and 20 points on the math.

To make historical comparisons, the College Board says that they have re-centered all the scores back to 1967. But, perhaps because of different record-keeping in different eras, the organization used one statistical method to re-center scores from 1987-1995, another for the period 1972-1986, and still another for the years before 1972, which are "based on estimates," according to the College Board. Forgive me if I'm skeptical about their ability to make apples-to-apples comparisons. In fact, this is why they should never alter the scoring method -- without a fixed standard it's difficult to interpret the results.

Grade inflation is not just a problem at the College Board. Schools have gotten into the habit as well. The College Board released its report on test scores this week along with answers from student questionnaires. Particularly striking is that a full 40% of students report that they have grade averages of A- or better.

Even beyond the grade inflation, and even after adjusting the test, the results still say that U.S. kids are scoring below the levels of the 1960s. Not a good sign in an economy that's growing more competitive and knowledge-intensive. One interesting finding this year is that foreign students - who make up 3% of SAT test-takers - scored higher than the averages on both the math and the verbal. Foreign kids, many of whom are dealing with English as a second or third language, are beating us on our own test?

In fact, the foreign kids seeking admission to American colleges are probably among the highest achievers in their home countries, so it may not be fair to beat up on our schools for these results. But it's very clear from other data that we're not delivering first-class K-12 education, compared with other countries. In the most recent international test, American high school students finished 18th out of 21 nations in math and science literacy, and the test didn't even include any Asian countries. In a separate test for students taking advanced physics, our kids finished dead last. In a 1998 Public Agenda poll, 52% of college professors said that the students they observe lack the skills necessary to succeed in college.

What's the solution? Competition. It works in our economy, and it works in our college and university system. As long as the K-12 public schools enjoy a legally-enforced monopoly on public education dollars, it's naive to think that they'll deliver first-class results.

So Tim Draper, a Silicon Valley venture capitalist, is backing a campaign to pass Proposition 38 in California. This measure would give every family a $4,000 voucher per year for each child to spend on the school of their choice. There would be no tax increase, and since California averages $7,400 in per-pupil spending, the public schools would get to keep $3,400 for each student who chooses to leave the public schools.

So every family, rich and poor, would get to pick the best schools for their kids, and the public schools would have more resources to educate fewer students. It's a winning idea, and could be a model for other states.

Predictably, the teachers unions are opposing the initiative, but if public schools are really delivering improved results, then they have nothing to fear from competition, right?

TCS Daily Archives