Summer Break and the Achievement Gap
Note: Andrew McEachin, an assistant professor of educational policy analysis and program evaluation at NC State University, is guest posting this week.
I'm going to keep this short and sweet. This week, I've focused on various aspects of teacher and school accountability. Today I am going to take the conversation in a slightly different direction. From the Coleman Report forward, we have known that the vast majority (upwards of 75 percent) of students' achievement is attributable to outside-of-school factors. And yet when the general public hears statistics like the achievement gap between the wealthiest and the poorest students in the US has grown by 50 percent in the past thirty years, their gut reaction is to blame teachers and schools. Are there areas for improvement within the public education systems? Of course. Is the public education system to blame for this widening gap? Absolutely not. Can the public education system as a whole close the achievement gap between the wealthiest and the poorest students? No, I do not believe that schools alone can close this gap. It is this last point that I will focus on today. For an awesome primer on the subject, I highly recommend reading Class and Schools by Richard Rothstein.
The past year or so I have spent a lot of time thinking and writing about the role students' summers contribute to their academic development (along with Dr. Allison Atteberry at UVa). Why am I talking about summers on a blog that focuses on education? The research on summers (here and here, for example) shows us three important trends. First, achievement gaps among class and racial/ethnic lines already exist at the start of kindergarten. Second, students learn at roughly the same rate during the school year. This means that the gaps do not grow significantly while students are in school. Third, achievement gaps widen during the summer period when students are not under the direction of teachers and schools. Simply put, schools do a pretty good job keeping the achievement gaps that they did not create at bay.
Wait, then why did I spend three days talking about various approaches to, and problems with, teacher and school accountability policies? Herein lies a distinction that is often ignored. Policies and practices can affect students in two ways. First, they can raise the overall level of student outcomes. Second, they can focus on a specific group or groups in an effort to narrow achievement gaps. I think accountability policies do a fairly good job at the former. The jury is still out whether they can be used to significantly affect the latter. In fact, I think there are few, if any, education policies, practices, or reforms that can significantly narrow achievement gaps at scale. As a simple thought experiment, think about the most transformative educational experience in your lifetime. Now think about the barriers that would keep that from happening at scale. Also, think about how that experience would influence kids from all backgrounds. It is hard to imagine an experience that wouldn't have near equal effects for all kids. In short, that experience may improve learning for all students, but it will not likely close the gap. It is clear that schools are not the main driver of the achievement gaps along racial/ethnic and socioeconomic lines.
The summer research also tells us something that, much to my surprise, has been largely ignored in policy and research. If we know that achievement gaps widen over the summer, that students are not randomly assigned to schools, and that we only measure students' achievement each spring, then the school performance measures we use in accountability policies are likely biased--especially against schools serving larger shares of traditionally under-served students. In fact this is true. In a working paper with Dr. Atteberry, we find that there is a significant amount of bias against these schools when they are held accountable for student achievement growth using Student Growth Percentiles or Value-add based on spring-to-spring achievement growth. If you instead use only the amount of learning that occurs during the school year in growth models, removing the summer period, these schools are much more likely to perform at the same level as their wealthier peers. Federal and state policymakers, who by and large continue to push growth and/or value-add evaluation models, should take note that this small adjustment can significantly reduce the bias in these evaluation systems.
I want to thank Rick again for the opportunity to appear on his blog. I have no idea how he does this on a regular basis. Regardless of whether you've agreed or disagreed with me, found me aggravating or interesting, I hope that something in these four posts got you to stop and think a bit more about education policy or practice, even if it's just five minutes. Please feel free to contact me at firstname.lastname@example.org or follow me on twitter @ajmceachin if you'd like to continue this conversation.