Jay P. Greene's Blog

(Guest Post by Collin Hitt)

Charter schools push out students who fall behind academically – that’s how they achieve superior results, right?* This is one of the main claims of charter school opponents, a claim that calls the moral character of charter school employees into question.

Charter schools – especially “no excuses” schools – thrive off of promises to recruit and educate students who would otherwise be left behind. Poor kids. Minority kids. Kids with low test scores who can’t read. A movement that promises to educate these kids only to slyly turn them away in favor of their higher-performing peers, is a movement of charlatans.

A forthcoming article looks at the issue. Ron Zimmer and Cassandra Guarino,in Education Evaluation and Policy Analysis, examine the transfer-out patterns in a large urban school district with nearly 60 charter schools. It is by far the largest, high-quality study to ever look at the issue.

Their main finding:

The coefficient estimates suggest that low-performing students at TPS schools are 1% to 5% more likely (at statistically significant margin) to transfer than above-average students, although the statistical significance may be achieved in part due to the large sample size. Low-performing students are neither more nor less likely to transfer out of charter schools.

They find no evidence that charter schools are pushing out their lowest-performing students. But wait – did anybody else catch that? – what did they write about traditional public schools? From the next paragraph of their study (emphasis added):

This suggests that low-performing students are more likely to transfer out of a [traditional public school] than a charter school. But again, the differences are relatively small—about 5%. Overall, the results across all models provide no evidence that low-performing students are more likely to exit a charter school than a high-performing student or a low-performing student in a TPS.

This study shows that in one very large urban district, there’s no evidence that charter schools are pushing out their bottom performers. The same cannot be said of the district’s traditional public schools. Let’s be charitable to the traditional public schools in that district; perhaps the findings that their lower-performers are more likely to leave is a result of the fact that those parents are seeking to leave for charter schools. Or perhaps this finding is just a fluke with Zimmer and Guarino’s model, and there’s really no difference at all in the exit rates of low performers in the district’s charter and traditional schools. Those are plausible ways to interpret the findings. (Can anyone think of plausible, less charitable ways to view these findings?)

Now ask yourself, had Zimmer and Guarino instead arrived at the opposite results – charter schools are more likely than traditional public schools to see their lowest performing students leave – would charter opponents be this generous in their interpretation? My guess is that the web would be somewhat more atwitter about this study, which deserves better coverage than it’s getting.

* The most rigorous research on charter schools uses random assignment. In the intent-to-treat components of these studies, students are considered charter students if they’re ever offered a seat in a charter school. Regardless of whether the students accept the offer, and regardless of whether students transfer out of charter schools at a later date, they are still considered charter school students in the ITT analysis. Yet random assignment studies consistently find positive results for charter schools, especially in urban areas where they are most likely to encounter disadvantaged students. Even if charter schools were pushing out their students – which apparently they’re not doing, according Zimmer and Guarino – it couldn’t explain these gains.

This entry was posted on Wednesday, September 11th, 2013 at 8:03 am and is filed under Uncategorized. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.