Baseball Crank
Covering the Front and Back Pages of the Newspaper
August 11, 2006
POLITICS: About That Schools Study...

The Department of Education released a National Assessment of Student Progress (NAEP) study last month comparing performance of students in private vs. public schools nationwide, based upon tests conducted in 2003; the study compared a total sample of over 6,900 public school fourth grades and over 530 private school fourth grades, with a similar sample for 8th grades. As the study itself notes, NAEP tests "typically show a higher average score for private school students than for public school students." (at 7) This study, however, found that when you adjust for various characteristics of the student body, the usual advantage of private schools seemed to disappear: the average performances of private and public school students were close enough as to be, for statistical purposes, identical.

Predictably, this finding was trumpeted by many on the left who oppose private school choice, on the theory that it showed that there is no benefit to sending kids to private instead of public school. (See here and hereand here for samples from the blogosphere, and here and here for a big-media pundit and the New York Times making the same claim). In fact, this argument overreads the results of the study and entirely misses the point of the case for school choice.

Read on.

I should note that I am building here on the observations of Megan McArdle and Stuart Buck (hat tip to Jon Henke of QandO), both of whom have focused on the central problem with the study, which is the comparison of mean (i.e., average) scores. For a deeper look at the study's methodological problems (including the fact that the study selects out smaller private schools that may be superior, partly for that reason, to gargantuan public schools) and an alternative study showing private school advantages, see the Cato Institute's blog.

1. The Study Itself Notes Its Limitations

From the executive summary:

When interpreting the results from any of these analyses, it should be borne in mind that private schools constitute a heterogeneous category and may differ from one another as much as they differ from public schools. Public schools also constitute a heterogeneous category. Consequently, an overall comparison of the two types of schools is of modest utility. The more focused comparisons conducted as part of this study may be of greater value. However, interpretations of the results should take into account the variability due to the relatively small sizes of the samples drawn from each category of private school, as well as the possible bias introduced by the differential participation rates across private school categories.

There are a number of other caveats. First, the conclusions pertain to national estimates. Results based on a survey of schools in a particular jurisdiction may differ. Second, the data are obtained from an observational study rather than a randomized experiment, so the estimated effects should not be interpreted in terms of causal relationships. In particular, private schools are "schools of choice." Without further information, such as measures of prior achievement, there is no way to determine how patterns of self-selection may have affected the estimates presented.

(Emphasis added).

2. Not All Schools Are Average

As Buck points out, "children who are likely to be eligible for vouchers do not attend schools that equal the nationwide average." The entire point of school choice is to ensure that every student who could benefit from picking a private school (or another public school) over his or her current school has a chance to do so. You don't knock down the case for doing so by showing that the average public school is equal to the average private school - to make the case against choice on grounds that it won't provide a better education, you need to show that every public school is at least equal to every private school in its immediate geographic area. Otherwise, you are consigning kids in one school district to a bad school simply because somebody else doesn't need an alternative.

3. Means Are Not Medians

Any decent statistical study will give you both means and medians (i.e., the 50th percentile, the student right in the middle), and the study from start to finish speaks only of mean results. Without detail on the median results, there's a distinct possibility that the best public schools are pushing up the averages, concealing a greater number of truly failed schools in the public school sample. It is cold comfort to parents in Bedford-Stuyvesant that kids in Scarsdale are pulling up the average, or for that matter that the best kids in one school are well ahead of the worst.

4. Competition Makes Schools Better

A basic principle of markets is that competition improves services by the monopolist, not just the competitor. In wealthy communities, school choice already exists - if the public school doesn't serve the kids, the parents will take them elsewhere. Thus, public schools in such communities already benefit from choice because they have to keep up with private schools to retain their students. In fact, the wealthiest communities tend to have public schools that are particularly academically successful, because people who are focused on education will move into those districts (ask any real estate agent about this) and willingly pay more taxes to improve the school. Even with the effort at weighted averages, including such schools in the study undermines the entire effort to compare private schools to true monopoly public schools where the kids don't have meaningful alternatives.

5. Let The Parents Decide

If it is really true that there is no advantage to private schools, then does that mean that all the parents paying to get private educations for their kids are fools who could be getting an equally good education for free? Conservatives are not so arrogant or collectivist as to assume that a government study of a national average is a better judge of each kid's needs than his or her own parents. At the end of the day, each student is different - no student is the New Average Man. If you give people choices and those choices are no better than what they have, they won't go anywhere. If liberals believe that there really is nothing gained by kids leaving for private schools, what are they afraid of?

Republicans shouldn't be cowed by those who argue that an aggregate national study is a substitute for a parent's own knowledge of their child's needs. On this issue, as with so many other issues that don't involve the taking of a human life, we should remain the pro-choice party.

Posted by Baseball Crank at 8:30 AM | Politics 2006 | Comments (2) | TrackBack (0)

I am a fine kind of liberal, as I am for school choice. For plain simple American capitalistic values: competition is a good thing. Would Judge LAndis have ordered the breakup of the school systems and impose a fine?

My kids go to public schools. In my town, they are very good, and they do fine. I know quite a few (when you coach teams, you meet everyone) who either go to the local Catholic school, also very good, and a few priovate schools--not bad, but you pay a lot for I'm not sure what.

Want good schools? Have concerned parents who keep coming in. That, way more than class size is a determining factor. When you have school choice, evolution comes into play: the parents who are concerned will actively look for the best place, and stay involved. Leaving the kids who can fend the least with the least. But that will happen no matter what anyway. And I don't like defining all kids norms down because too many people give birth but don't parent.

Posted by: Daryl Rosenblatt at August 11, 2006 10:27 AM

Another problem with that study is that it only tested 7 year olds. I'm willing to bet the gap widens by age 16 or 17.

Posted by: Tom G at August 14, 2006 9:54 AM
Site Meter 250wde_2004WeblogAwards_BestSports.jpg