A Few More Points About Charter Schools And Extended Time

A few weeks ago, I wrote a post that made a fairly simple point about the practice of expressing estimated charter effects on test scores as “days of additional learning”: Among the handful of states, districts, and multi-site operators that consistently have been shown to have a positive effect on testing outcomes, might not those “days of learning” be explained, at least in part, by the fact that they actually do offer additional days of learning, in the form of much longer school days and years?

That is, there is a small group of charter models/chains that seem to get good results. There are many intangible factors that make a school effective, but to the degree we can chalk this up to concrete practices or policies, additional time may be the most compelling possibility. Although it’s true that school time must be used wisely, it’s difficult to believe that the sheer amount of extra time that the flagship chains offer would not improve testing performance substantially.

To their credit, many charter advocates do acknowledge the potentially crucial role of extended time in explaining their success stories. And the research, tentative though it still is, is rather promising. Nevertheless, there are a few important points that bear repeating when it comes to the idea of massive amounts of additional time, particularly given the fact that there is a push to get regular public schools to adopt the practice.

We’re talking about massive amounts of additional time. The first and most simple point, which I have made before, is to remember that, when it comes to big name charter chains, we’re not just talking about an additional week or two of school. We’re talking about additional months. By adding 2-3 hours to the regular public school day, which is usually 6-7 hours, most of the big name chains offer 30-50 percent more time in school, the regular public school equivalent of up to four months (regular public schools, of course, could not achieve this increase without lengthening the school day, even if they eliminated the summer and all vacations).

To be clear, this is not a criticism. Quite the contrary; many such schools serve children in high-poverty neighborhoods who begin schooling far behind their more advantaged peers. And keeping struggling students in school longer might be one way to help them catch up. But let’s not forget that, in these cases, “extended time” means a lot of extension.

Large-scale time extension is a somewhat blunt instrument. This brings me to my second point on this issue, one that is deliberately provocative. It might be posed as a question: After 20 years, is this the biggest concrete policy lesson that we’ve learned from experiments with charter schools? That keeping children in school for an extra 2-4 months can increase their test scores?

I consider myself neither a supporter nor opponent of charter schools. So, when I hear so much talk about “charter school innovation," I don’t doubt that many of these schools are doing great things, just as I know there are plenty of regular public schools doing the same. Still, massive time extension is among the only actual policies with any record of association with charter success, and you can hardly consider it a new or novel idea. That doesn't mean it's not an important lesson; sometimes blunt instruments are needed. But it's not really "innovative" at all.

We shouldn't oversimplify or kid ourselves about the cost and labor supply issues. The third and perhaps most practical point I think bears on this discussion over charter schools and extended time is whether it can be "scaled up" significantly - i.e., adopted by regular public schools. The first issue here is cost. I’ve heard some rumblings that teachers' unions “block” extended time, which is why charter schools are the only ones able to get it done. The basic argument here is that unions, specifically via collective bargaining laws, would require extended time proposals to be negotiated, and that many union leaders would either refuse it or require prohibitively large salary increases.

First, to be completely honest, as a supporter of unions, my impulse response is: If the proposal here is to increase teachers' -- or any group of employees’ -- working days by 30-50 percent, and compensate them inadequately for that huge expansion of their job duties, then I for one am glad that collective bargaining laws are around.

More to the point, though, the constraints here would seem to be a lot more complicated. Any serious extension of time is expensive. The big charter chains receive millions in private donations, and they use this money in part to extend their days and years. If you want to debate whether collectively-bargained salary schedules should be frontloaded, or pension expenditures reallocated to salary, as these measures might attract more people willing to work 9-10 hours a day, that’s a fine debate to have, but the bottom line is that most charter schools, for all their autonomy and union-free workplaces, can’t afford 8-9 hour days anyway, and that’s in no small part because doing so costs money, and they don’t get enough in private donations.

In addition, there is a very obvious labor supply issue here, one that thoughtful charter supporters are more than willing to recognize: There are some talented teachers who are willing to work 10-12 hours a day, and in some cases be on call via cell phone even when they’re home, without making a whole lot of money for it. But, in any given labor market, their ranks would probably thin out pretty quickly.

So, I think there is plenty of room for thoughtful, targeted extensions of the school day in U.S. public schools, but the difficulties here are considerable.

Extended time is not going to work for all (maybe even most) students. This brings me to my fourth and final point about charter schools and extended time in general: There is no empirical basis for it, and so it’s just a hunch, but it’s difficult to believe that 8-9-hour school days would work for all children. In fact, I would tend to doubt that it would be worth the cost for most children, regardless of background.

The big name charter chains, such as KIPP, benefit from a school choice system in which they accept all comers (usually via lotteries), but many of those who don't thrive under their model leave and seek out a different school. Charter critics call this cheating or “creaming." I don’t quite see it that way. I see it as follows: KIPP and similar schools offer a particular model of schooling, and those who aren’t doing well under it should leave, while those who thrive should stay. That’s the whole premise of school choice.

That said, while I'm always open to reviewing evidence to the contrary, I suspect that the subset of students for whom 8-9 hours days would be worth the additional cost is limited, perhaps even relatively small. It would be smart to keep that in mind when you hear grandiose proposals to extend school time, based on the results of a small group of chains dispersed throughout the nation.

- Matt Di Carlo


I appreciate your nuanced dissection of the issues here, and I agree with you that collective bargaining is a necessary counterweight to employers who would seek significant increases in work without otherwise addressing compensation. And keep in mind, American teachers for the most part already spend so much time directly working with students that they generally lack adequate time for all of their other responsibilities. I don't have the citation, but have seen several times information showing that in leading nations, teachers actually spend 1-2 hours per day on their own learning, collaboration, etc. So, I think the solutions we should look at involve bringing more people into schools. Pre-service teachers might have a role to play here as part of a modified induction and training approach. Community organizations might be engaged to provide some services at school, during this extended day. Of course, to prevent setting off battles with unions, we should be engaging them from the start in coming up with creative solutions, not viewing them as obstacles.


No. Remember, research by Mass2020 has found when the extra 2-4 months were done by "regular" schools, no effect. Moreover, hours don't explain the variation within KIPP. Some have many hour "extra hours" than others. Moreover, hours don't explain the variation by charter markets. Boston charters on average are not much diff than New Orleans or NYC. But Boston charters way better results. I would argue that the lesson learned is the totality of No Excuses schools. And that starts not with longer day, but with a focus on school culture, and on obsessive measurement, and on human capital. The Hoxby thing you linked to probably picks up No Excuses schools which happen to have more hours. * * * I'm not aware of a study that compares charters with long hours, but wouldn't that be the test? a. Regular charters with long hours, who are not No Excuses schools versus b. No Excuses charters, with similar hours And then find which ones create what size gains for kids? If I understand, your hypothesis is the groups would be similar. My hypothesis is that b would far outperform a.