« Mapping Education | Main | In and Out: Charter School Transfers »

April 27, 2010

Charter School Stability

This is the first post in a series that looks at data from charter schools’ Basic Education Data System (BEDS) reports. This data was provided to us by the New York State Education Department (NYSED) via a Freedom of Information Law (FOIL) request. A full spreadsheet with the data we used is available here.

One of the largest issues in the charter school debates has been accusations that charters “counsel out” students who have learning disabilities or who do not adhere to the schools’ strict codes of conduct. While we haven't found comprehensive statistics that track individual students enrolled in charter schools from year to year, the BEDS reports include a “student stability” number that is relevant to this issue. Student stability counts the number of students who are currently enrolled in the highest grade that the charter serves who were also enrolled in the school last year. For instance, if a charter school serves students in kindergarten through 8th grade, the student stability number would look at the number of current 8th graders who were also 7th graders last year.

We found that, on average, charter schools retain 84% of their students, compared to 93% for traditional public schools city-wide. This percentage has remained constant for the past three years but the percentage at individual schools varies widely. Some schools, such as the Beginning with Children Charter School and the Harbor Sciences and Arts Charter School, experience almost no attrition. Others, such as Harlem Day Charter School and the John V. Lindsay Wildcat Academy, consistently lose more than one third of their class. And for many charter schools whose highest grade was 9th, the attrition was noticeably high, probably because many of their eighth graders chose to go to other, perhaps more well-known, high schools.

To better visualize the data, we have created a map that shows all of the charter schools that had applicable data. The size of the dot corresponds to the percentage of students that left the school, and if you hold your mouse over the dot you will be able to see relevant information such as grade studied, number who stayed in the school, and number who left. You can zoom in on certain districts, choose to look at the stability ratio for specific grades, or click to see the stability number of specific schools by using the menu to the right. Unfortunately, this data is only for the 2008-2009 school year – to see the numbers for 2007-2008, you will have to look at the spreadsheet.

It is important to note that these stability numbers only look at one grade in a particular charter school. Furthermore, the BEDS data, while vetted by the NYSED, is not without its flaws, which include the timing of the report (charter schools must report their numbers in mid-October) as well as the lack of substantial follow up by the relevant parties who collect the data.

Nevertheless, we believe that this information provides important insight into charter school stability. As always, we welcome your feedback for ways we can improve and build on this report.


TrackBack URL for this entry:

Listed below are links to weblogs that reference Charter School Stability:


I have admired a lot of your recent work, but I think you're off the mark here.

Attrition in so large a city as ours is driven as much by residential change as parent choice. To compare to averages citywide almost certainly understates the attrition you use as a counterfactual.

I would like to see the annual retention rate by district or neighborhood to get a sense of the variance around your 93% numbers.

Also, Wildcat is un ungraded school for students who are overage and underaccredited and they are almost certainly moving the average significantly, either due to data irregularities or actually high attrition. What is the charter average without Wildcat's data?

Why is student stability only looked at in the terminal grade? That seems to be a very limited defintion of stability. I suggeset you look at the number of students who enter in a beginning grade and how many remain to graduate. Tak a look at the DOS's testin gdata and see how many kids are tested in third grade and how many make it to later grades to be teseda gain

Is it a surprise that retention rates are higher at district-run schools? Many students have no choice but to attend their local zoned school, whereas charter schools are schools of choice where parents get to vote with their feet if they are not satisfied with the school's performance.

Thanks for all the feedback!
B - You're completely right - I neglected to look at the stability numbers district by district which could change the results. I just went through the demographic data and updated it (if you click on the link in the post you should be able to get the spreadsheet) and the numbers appear to be relatively consistent district to district. For instance, District 5 has a stability percentage of 91%. Districts 7,8,9 (South Bronx) had percentages of 92%, 90%, and 91% respectively. As for Wildcat, I took it out for both years and the percentage changed slightly, to 85% for both years. This has also been updated in the spreadsheet.
JG - I would LOVE to look at the number of students who enter and who remain until graduation - unfortunately, I haven't been able to track down those statistics. I'm not sure that charters file them with the state or with the DOE, but if you know otherwise, please do let me know. In terms of enrollment/testing data, it's complicated by the fact that many charters can admit more students as their numbers decline. For instance, if a charter has 15 kindergartners and 5 leave at the end of the school year, the school may decide to admit 5 more students for next year's 1st grade class. Thus, it would look as if the numbers stayed constant, but there's no way of telling if those are the same students. Does that make sense?
Gideon - That's certainly a plausible theory for why the retention rate might be higher. I don't claim to know why the numbers differ - it would be great if I did know, though!

love you not because of who you are,but because of who I am when I am with you.

The comments to this entry are closed.