In 1960, Milton Friedman wrote Freedom and Capitalism. That book proposed education vouchers whereby money would follow children rather than go to local education agencies. Among the people who read that book was Ronald Reagan (Friedman told me he actually did read the book). Reagan entered the White House in 1980 with an education agenda of vouchers, tuition tax credits, school prayer and the demolition of the federal department of education. In a sweet irony, he almost refused to accept "A Nation At Risk," provided by his secretary of education, because it addressed none of these issues. In the White House schism, Ed Meese and other conservatives urged Reagan to reject the document. Jim Baker and Mike Deaver prevailed, arguing that while Meese was right, there was too much good stuff in the booklet to campaign on to let it die.

Since 1980, vouchers have become a more and more visible aspect of education reform (see EDDRA2). Now Martin Carnoy and Patrick McEwan have this to say about vouchers: they don't work.

In 1980, inspired by Friedman's writing, Chile initiated a national system of vouchers. Chile also tests kids in grades four and eight and gathers background info on families. Carnoy and McEwan found that while the established religious schools had students that tested about three percent higher than public schools, the scores of students in the public schools were higher than those in what C & M call "commercial schools"--the schools that sprang up after the voucher policy was instituted. This in spite of the fact that the public schools lost 20% of their clients to the commercial and religious schools and are now serving poorer families (confirming in this regard the predictions of voucher opponents).

Looking in areas that had the greatest concentration of private schools, C & M were unable to find any evidence that competition had improved the performance of public schools. If market accountability is what makes schools good, then it should work the same in Chile as in the U.S.," said Carnoy. It didn't work. What did work was a government program that provided more money and teacher training to the nation's 900 lowest performing schools.

If you walk out of your house to the nearest intersection and ask 100 people what's happening to test scores, I predict that 99 of them will say "They're declining" (please feel free to test this empirically).

Dour accounts abound. Typical is the commentary of Stephan Thernstrom, Winthrop Professor of History at Harvard University and co-author with his wife Abigail of America in Black and White. In the introduction to the fall, 1998 issue of The Concord Review, a quarterly out of Concord, MA that publishes essays of high schoolers, Thernstrom had this to say:

Whites, 13 points, from 293 to 306
Blacks, 22 points, from 235 to 257
Hispanics, 12 points, from 249 to 261
Total, 11 points, from 283 to 294

510 510 510 510 510 510 510 430 430 430 ------------ Average of 10 Scores: 486	Consider the 500's in the left column to be SAT scores for white students and the 400 to be the SAT scores of minority students. In the right column, consider the 510's to be the SAT scores of white students and the 430's to be the SAT scores of minority students. White students have gained 10 points. Minority students have done better, gaining 30 points. However, in the left column, minority students are only 10 percent of the sample and in the right column they are 30 percent of the sample. The impact is that their improving-but-still-lower scores actually depress the overall average. Although hypothetical, this is a fairly accurate historical representation of what happened with SAT scores. When we look at the 90th and 95th percentiles by ethnicity, we find that only two of the 18 scores fell (three ethnicities by three subjects by two points in time). The 95th percentile for whites in reading swooned from 358.9 to 358.4 and mathematics "plummeted" from 357.8 in 1978 to 357.7 in 1996. For minorities the scores are all up. If we add in the 36 salient scores from ages 9 and 13, the other NAEP assessment ages, we find all of them up, some in the 20-30 point range.
About Thernstrom's contention that we spend more and more money and maybe don't have anything to show for it, the following can be said: 1) most of the new money since 1969 has gone into areas that one would not expect result in test score gains, e.g., special education. 2) There is little reason to expect gains in NAEP scores from such expenditures. NAEP is connected to no one's curriculum save Kentucky's. And as Archie Lapointe, the former executive director of NAEP used to say, the biggest problem with NAEP is keeping the kids awake for the test. It is an assessment that, until recently, was of no consequence to anyone and no one took it seriously. Now that it is being used somewhat by certain governors and legislators for bragging rights, NAEP's integrity is being compromised. If the Good Professor has any comments on the above, I will let you know. NOTE: Asians are not included in the ethnic analysis because until recently they constituted too small a proportion to generate a reliable estimate in the NAEP sampling system. Return to Top of Page
© 1999 Gerald Bracey Last updated June 11, 1999	Web Services by