Giving Books to students, Self-selected or Not, Prevents Summer Learning Loss Among Low-Income Students

Erin T. Kelly, C. Andrew Aligne. Internal Medicine-Pediatrics, University of Rochester, Rochester; Pediatrics, University of Rochester, Rochester.

BACKGROUND: Reading proficiency is a critical skill and an important determinant of health. Summer learning loss, the “summer slide,” refers to the academic loss, or failure to make continued gains, during summer vacation, which is worst among low-income students. One intervention in high-poverty primary schools in Florida provided students with self-selected books at the end of the year, dramatically improving students' reading achievement (Allington 2010).

OBJECTIVE: Improve literacy among Rochester City School students by providing free self-selected books to stem summer learning loss.

DESIGN/METHODS: Eighteen children in one 2^nd grade class at an urban primary school each received 13 self-selected books at the end of the school year; 20 children in the other control class received books from the current community program (a few unselected books mailed to them over the summer). All students had literacy assessments (Developmental Reading Assessment) in the spring and the following fall. In the spring of 2014, this intervention was expanded: four classes of kindergarten through second grade students received 13 self-selected books. The control intervention was modified, based on ethical concerns, given the success of the prior year intervention, to include some student selection of books and batch delivery at the end of the school year; one first and one second grade class in the same school served as modified controls.

RESULTS: The pilot showed a statistically significant improvement in reading score over the summer in students who received books (paired t-test, p-value 0.01), and no statistically significant change in reading scores among the control class of students (paired t-test, p-value 0.22). With replication on a larger scale and modification of the control intervention, there was no significant difference between the reading scores of students in the two groups (n=67 intervention, n=20 in control). Both groups saw overall average gains in reading scores and had low rates of “summer slide,” with more than 75% of students maintaining or improving their reading, compared to an average summer learning loss of one month seen in prior studies (meta-analysis, Cooper 1996).

CONCLUSIONS: Providing books to students over the summer was associated with improved reading scores overall. The results of the modified control group in the expanded intervention suggest that receiving some books, even with more limited self-selection, may stem the summer reading loss.