Debra Viadero reported last week on a paper by Dave Stuit and Tom Smith finding that charter school teachers are 230% more likely to leave the field (an updated version actually pegs it at 237%) at the end of the year than are teachers in traditional public schools, based on data from 2003-04.
Given that both writers are from Vanderbilt and that I find the topic interesting, I thought I'd look into it further. It's a conference paper -- not a finished paper that's been peer-reviewed and published in a journal -- so I'm not going to get into significant detail. While I'm sure it's not perfect and that it will undergo further revisions, I will say that the methodology seems pretty straightforward and that I trust both of the writers to investigate things rigorously and interpret them correctly, so my guess is that when the final version is released most of these figures will remain about the same. That said, let's get to it.
The most striking finding was that charter school teachers were more than twice as likely to leave the field as are teachers in traditional public schools -- the raw numbers are 14.1% and 7.0% (please note that the 230% estimate is based on log-odds ratios calculated using Hierachical General Linear Models controlling for clustering in schools*). They also calculated that charter school teachers were 113% as likely to transfer to another school.
They also found turnover nearly twice as high in schools that were new start-ups versus schools that had been converted to charter schools -- which makes sense because a new start-up would have more growing pains and instability than would a converted school.
But the most important question is why charter school teachers were more likely to leave the profession. Based on survey responses from those who left, charter school teachers were about twice as likely to report leaving for better salary and benefits (46%/22%), dissatisfaction iwth school (51%/24%), or due to school staffing action (40%/20%), and more than three times as likely to report leaving to pursue additional coursework outside education (27%/8%). They were also about one-third as likely to report leaving due to retirement (14%/38%).
In short, it seems that more teachers are leaving to move on to bigger and better things. This makes sense if we consider how many charter schools pluck people to serve as almost missionaries for a few years (think KIPP and their use of TFA fellows). I would expect that charter schools employ more people who are teaching as a way to give back to the community for a few years and fewer people that want to make a career out of teaching than are traditional public schools. Indeed, if we look at the demographics, charter schools employ more people under the age of 30 (34%/20%) and fewer over the age of 50 (18%/29%) than do traditional public schools. Teachers are also much less likely to be certified (67%/90%), and one would imagine that investing the time and effort into getting certified means that a teacher both intends to stay longer and will now be more likely to stay longer.
I've only just scratched the surface of all the paper examines and discusses, but I think the difference in attrition between sectors is worth thinking through. I could see the interpretation that this means charter schools need to work harder to retain teachers. I could also see an argument that attrition will hold back charters from achieving the success they could. But my main interpretation would be that charters tend to employ different types of teachers than do traditional public schools -- and these teachers, put simply, are less likely to stay in the profession for a long time (which could be exacerbated by the amount of time and effort required to teach in some of these charter schools). And I think the largest question it raises is about the replicability and scalability of charter schools.
If we continue to expand the number of charter schools and replicate those that have been more successful, will they continue to operate by employing teachers that don't make a career out of teaching? If so, how many people are out there that are both capable of being good teachers and want to teach for a few years?
*there was some confusion over how the 230% figure (now 237%) was derived on the Inside School Research site. After consultation with the author it's been explained that it was done thusly: a log-odds ratio of 3.368 was calculated. An odds ratio of 1 would mean that teachers were equally likely to leave the profession from both sectors. A ratio of 3.368 means that those in charter schools were 3.368-1 (=2.37, move the decimal point so that it turns into 237) percent more likely to leave.