Does a Summer Reading Program Based on Lexiles Affect Reading Comprehension?

Regional need and study purpose. A decline in student reading skills over the summer months is well documented (Allington and McGill-Franzen 2003; Bracey 2002; Heyns 1987; Luftig 2003; Malach and Rutter 2003). Summer reading loss disproportionately affects students with lower reading proficiency (Puma et al. 1997) and economically disadvantaged students (Cooper et al. 1996; Alexander, Entwisle, and Olson 2007). In the Southwest Region, the percentage of economically disadvantaged students has been rising. This study builds on research examining the effect of summer reading programs that supply students with books matched on reading level, interest area, or both on summer reading loss (Allington et al. 2010; Butler 2010; Crowell and Klein 1981; Kim 2006; Kim 2007; Kim and Guryan 2010; Kim and White 2008).

Intervention description. This intervention consisted of sending students 8 books—matched to reading level and interests—by mail over the summer, along with reminder postcards, one a week for six weeks. Students' reading levels were gauged using Lexile scores, based on grade 3 spring 2009 Texas Assessment of Knowledge and Skills (TAKS) results. The reading difficulty of the books was also obtained using the Lexile scale. Students' reading interests were identified using a student interest survey disseminated with the parent consent form for the study.

Study design and period. The study focused on economically disadvantaged students in grade 3 in spring 2009 whose reading level was below the 50th percentile nationally. Four medium-to-large districts in Texas were selected to participate in the study. The intervention took place during summer 2009, with data collection during spring 2009 (pre-intervention) and in fall 2009 (post-intervention).

Key outcomes and measures. This study was intended to assess the impact of a voluntary summer reading program on student reading skills. Reading skills were assessed post-intervention using the Scholastic Reading Inventory (SRI), a measure of reading comprehension.

Data collection approach. All data collection took place in 2009. In June, the districts provided REL Southwest researchers with the student interest surveys, student demographic data, and TAKS Lexile measures and reading scale scores from the spring administration of the TAKS. The researchers matched books to student reading levels and interests. The SRI, administered in fall 2009 to intervention and control group students, was used to assess posttest reading skills. And in fall 2009, all study participants were surveyed about their summer reading activities, summer school participation, and home reading material.

Analysis plan. The study's primary hypothesis was that the intervention group students would have higher average (Lexile) scores on the SRI than would control group students. Preliminary data analyses were conducted before testing the primary hypothesis. Preliminary analyses included an outlier analysis and careful examination of the demographic characteristics and baseline equivalence of the two study groups. The primary hypothesis was tested using an ordinary least-squares model comparing student outcomes in the intervention and control groups.

Principal investigators. Chuck Wilkins, PhD, Director of Statistics and Evaluation Research, REL Southwest; Russell Gersten, PhD, Instructional Research Group, REL Southwest; Lauren E. Decker, PhD, Researcher, REL Southwest; Leslie Grunden, PhD, Researcher, REL Southwest; Sarah Brasiel, Researcher, PhD, REL Southwest; Kim Brunnert, PhD, Senior Researcher, REL Southwest; and Madhavi Jayanthi, EdD, Senior Research Associate, Instructional Research Group.

Click here for full printable version