Association for Public Policy Analysis & Management

However, almost all that is currently known about summer learning stems from either research studies conducted in the 1970s and 1980s (Heyns, 1978; Entwisle & Alexander, 1992) or studies that only focused on early elementary grades (e.g., Quinn et al., 2016; von Hippel et al., 2018). As a result, very little is known about the relationship between summer and inequality in upper elementary and middle school years. Using national math and reading assessment data from NWEA, this study contributes to the literature on summer learning in three ways: (1) documenting patterns of school year gains and summer gains/drops from kindergarten all the way through 8th grade, (2) using a nationally-weighted dataset containing over two million unique students across the United States, the largest dataset by far to ever be used to examine this topic, and (3) examining how summer learning rates differed by student race/ethnicity and district poverty.

We found that students lost ground on average during each summer between kindergarten to 8th grade, with the average summer drops representing a loss of 12 to 29% of the school year gains in reading and 14 to 35% of the school year gains in math. We also find that there is far more variability in summer learning rates than school year learning rates, since almost all students show positive growth during the school year but students display a range of gains and losses in the summer. Looking at racial/ethnic differences in monthly learning rates, Black students show significantly slower average growth during the school year than White students, but also less summer loss on average than White students, holding constant district SES. A similar pattern is seen for Hispanic students in math, though the results are less consistent in reading. Asian students gained more on average in math than White students during both the school year and summers, leading to an average score advantage by the later grades.

Given that our findings highlight the prevalence of summer setback in both elementary and middle school, policymakers interested in addressing summer loss should not concentrate only on solutions for students in early grades. Additionally, most of the variation in summer learning is observed to be within schools, indicating that opportunities for summer learning are not concentrated within certain communities and can be expanded locally to target students who are most vulnerable for learning loss.