MPS outperforms Voucher schools on state tests

by folkbum

Thomas J. Mertz lets us know that the Legislative Audit Bureau's analysis of test score data from the first year of an "accountability" project has been released. (Full report--27 pages of .pdf.) Mertz reproduces this graphic, which pretty much says it all:

(Click for the larger version if you cannot read the numbers.)

Now, we already had an idea that this is what the data would look like. In February, the university researchers conducting the longitudinal study released their first report on exactly these data and concluded, as the table above seems to indicate, that voucher schools and voucher students do not automatically outperform the public schools. The key new information in the LAB's report released yesterday is the addition of the nationally-normed test data, which I'll get to in a second. But I do want to make a handful of quick points about the WKCE data:

1. MPS has aligned its curriculum to the state standards which are, in theory, what the state test tests. Voucher schools may be doing a good job of teaching things not covered by the state standards--which may be either good or bad depending on how you feel about our standards.
2. These data represent aggregates. Individual students will still thrive or fail in different environments for different reasons, and anecdotal evidence suggests that there are indeed students who "make it" at voucher schools who wouldn't have in MPS. There are also plenty of world-class students in MPS.
3. These data are still the "baseline" year, based on tests given two years ago in November of 2006. They show that in year zero of the study, MPS outperforms voucher schools. We should have, any moment now, the researchers' report on the 07-08 test data, which may or may not show the same results.
4. The WKCE data here (as well as the nationally-normed test data below) tell you nothing about how well an individual school does on the testing. Lucky for you, MPS has a full suite of downloadable "report cards" for each school (also available at the schools or central office on paper) so that you can see how an individual school performs on the state tests, and how those scores break down by race, sex, poverty, special needs, and so on. (From the MPS Portal, click on "Schools" in the left sidebar; each school's profile contains a link to the last few report cards.) Voucher schools are not required to provide that information to you or to the researchers for this study, so you have no way to know whether your neighborhood voucher school does any better with your tax dollars than your neighborhood public school.
5. Speaking of special needs students, check out the ratios in Table 5; less than 10% of voucher students tested were identified as having disabilities, while more than 20% of the matched MPS sample were and more than 25% of the random sample were. And the random sample performed best!
6. Test scores are not the best or only measure of student achievement, but they do provide an equal basis for comparison.

Here's the graphic of nationally-normed reading test scores at the voucher schools that submitted them:

(Again, click for a bigger image.)

I selected reading because it was the first graph (there's also math and science), not because it makes voucher schools look low-performing. The other scores are not any better--voucher students at these schools are generally performing well below the national average. (MPS does not administer these tests--by law, it administers the NCLB-approved state WKCE test--so there is no comparison.) This last point reinforces what I have been saying all along, so often that you can probably now say it with me: The problems in the Milwaukee Public Schools are not school problems as much as they are Milwaukee problems, something I tried again to get at a couple of weeks ago. Yes, there are things MPS can and should be doing better, and goodness knows I bust my own behind every day to help my students beat the odds. But the solution to low educational achievement in Milwaukee is not simply the dismantling of MPS or the transplanting of MPS students into private schools.

Who knows; maybe the long-term data will prove me wrong about that, although after almost 20 years of vouchers, you'd think that they would be beating MPS by now if they were really the solution.

(I wrote this post instead of a response to Patrick McIlheran's column this morning about a potential boarding school in Milwaukee. The success of such boarding schools in urban settings around the country reinforces my thesis above: If you remove students from a disruptive environment outside of school, they can learn better in school. Ironically, McIlheran has long been pro-voucher, though the data seem clear (to date) that vouchers alone don't do enough to create large-scale change. A 400-seat boarding school won't do much for the other 84,600 students in MPS, either.)