Lies, Damned Lies, and (Small Schools) Statistics

A study of the city’s small high schools illustrates the challenges of making and assessing policy in the accountability era. An editorial about the study underlines the failures of the fourth estate to exercise critical judgment.

The closing of 20 large high schools and creation of 132 small high schools is one of many initiatives Joel Klein has taken to close the achievement gap. The theory is that traditional “comprehensive” high schools, often with 3,000 or more students, are too large to offer the individual attention and accountability that traditionally disadvantaged students need to make it to college. The fight over Jamaica High School this past spring was, in a far less academic sense, a debate whether “big schools” serve our children’s needs.

Into this fray comes the MDRC, a research group focused on poverty. Their report’s headline cuts right to the chase: “Transforming the High School Experience: How New York City’s New Small Schools Are Boosting Student Achievement and Graduation Rates.” This conclusion is interesting because several prior studies of the effort were not positive. The Gates Foundation, which put close to $2 billion into small schools, was so uncertain of the schools’ impact that it stopped funding small schools projects in 2008.

But MDRC claims that “SSCs [Small Schools of Choice] increase overall graduation rates by 6.8 percentage points, which is roughly one-third the size of the gap in graduation rates between white students and students of color in New York City.” MDRC says its research “provid[es] clear and reliable evidence that, in roughly six years, a large system of small public high schools can be created and can markedly improve graduation prospects for many disadvantaged students.”

Chancellor Klein seized on the MDRC’s “landmark” report as evidence of the “tragic” impact of the court ruling preventing the closing of Jamaica High and 18 other schools the city had tried to shut down this year.

Alas, beyond the headlines of MDRC’s 159-page report, it seems the “transformation” is incomplete. In a footnote we read:

Although the estimated effect of SSCs on the overall high school graduation rate is statistically significant, estimates of SSC effects on graduation rates by type of diploma (p = 0.07) miss the standard of statistical significance established for this study (p = 0.05). Thus, comparisons of effects across diploma types are suggestive only.

What, dear reader, does this mean?  There are three high school diplomas in our fair state. A local diploma is so meaningless that is set to be eliminated. A Regents diploma could (at least) get you to the CUNY system. Earlier this week, Commissioner of Education David Steiner said the Advanced Regents Diploma should really be the standard.

Thus, the MRDC can only “suggest” students are more likely to graduate with a Regents or Advanced Regents Diploma. Keep in mind that SSC’s “improved” graduation rate is 44 percent for these diplomas types, statistically identical to the graduation rate for other schools. Nearly six in 10 students in small schools of choice still face extremely long odds in graduating from college.

An alternate measure of college readiness is graduates’ scores on the Regents Exam (anything above a 65 is passing). There is no difference in the percentage of students achieving a score of 75 or higher on the Math A Regents exam (22.2 percent for SSCs vs. 22.8 percent for control group), although there is a significant difference for the English Regents exam (34.1 percent in SSCs vs. 28.8 percent in other schools).

This is what passes for “clear and reliable evidence” that SSCs are closing the achievement gap? I can’t speak to MDRC’s motives in making claims that seem to far exceed what their data allow, but as parents and taxpayers, we have to stay alert to public relations pabulum peddled as proven, practical solutions.

Alas, over at the New York Times, someone fell for the report’s headline. The Times says the MRDC study “validates the small school policies of the Bloomberg administration” and should “encourage Mayor Michael Bloomberg to replace more large failing schools and districts elsewhere to follow New York City’s example.” Maybe they read a different part of the report than I did?

Chancellor Joel Klein’s commitment to addressing the failure of many disadvantaged kids to graduate is commendable. I wish he and his staff would find silver bullets everywhere they turn. But wishing it so and having it be so are two different things. The Times editorial board does its readers no favors when they fail to get beyond the surface of a complex issue.

Small Schools of Choice are an expensive way to boost graduation rates. They require more administrators, and more central support to gain traction. Just as with proposals to improve outcomes by reducing class sizes, I suspect that the benefits of SSCs don’t outweigh their costs. This may change as SSCs graduate more classes and more data becomes available, but for the moment it’s hard to argue SSCs are a “proven” solution.

About our First Person series:

First Person is where Chalkbeat features personal essays by educators, students, parents, and others trying to improve public education. Read our submission guidelines here.