Thursday, March 15, 2007

The Freedom to Reject the Best

Ludwig von Mises Institute article by Jim Fedako

A new study suggests that private schools are not inherently better than public schools. Surprised? Enough people were such that the study, funded by the US Department of Education, has created a stir in the education arena, as well as in the national news. But I want to argue that the results are meaningless, and for reasons not having to do with the methodology employed in the study.

The authors of Comparing Private Schools and Public Schools Using Hierarchical Linear Modeling analyzed math and reading scores of nearly 7,000 public schools and more than 500 private schools on the 2003 National Assessment of Educational Progress at the fourth and eighth grades. NAEP was the chosen assessment tool since it is considered to be the national achievement test and is used to assess student academic performance against national standards.

Though the title sounds impressive, the findings simply suggest a conclusion. Nothing has really been proven and no new truths exposed. I could begin by questioning the whole concept of empirical studies that suggest this or suggest that. I could ask, "What truths have been brought to light by any study that is couched in such a vague qualifier?" I could attack all the assumptions that went into the model and then list those that did not. Had I gone that route, I hopefully would have raised enough doubt in the reader that the study would be discarded as worthless.

But the real error here is more philosophical than empirical. Studies such as these simply show that a deeper ill exists, a malaise caused by government interventionism.

Consider Consumer Reports

The popular magazine reviews consumer goods based on a proprietary set of standards. They test, analyze, test, analyze, etc., until they are satisfied as to the quality of the products under review. CR then assigns individual product ratings and notes one product as a best buy. Though most Americans accept CR's results as being of excellent quality, the noted best buy is not usually the market best-seller. Yes, I will occasionally look at CR prior to purchasing a good, but I almost never buy the best buy. I agree that the CR results are scientifically valid based on their standards, but that doesn't mean I am in the market for the scientifically valid, CR best-buy product.

continue reading ...

1 comment:

Anonymous said...

A scholarly rebuttal of “Comparing Private Schools and Public
Schools Using Hierarchical Linear Modeling”:


http://www.ksg.harvard.edu/pepg/PDF/Papers/PEPG06-02execSummary.pdf

Don't hold your breath waiting for the New York Crimes to give this Harvard piece ANY coverage.