This project is designed to test the efficacy of a fully developed intervention, Collaborative Strategic Reading (CSR), with adolescent struggling readers. During a 10-year period, CSR has been evaluated by using quasi-experimental designs and has yielded positive outcomes for students with learning disabilities, students at risk for reading difficulties, average- and high-achieving students, and English language learners. This project meets the need for randomized controlled trials to more rigorously assess the effectiveness of CSR with adolescent struggling readers.
Research is ongoing.
Teachers implement CSR with struggling adolescent readers.
Students in the control condition are provided with school-designed, or “business-as-usual,” reading intervention to determine whether the CSR curriculum is more effective than other interventions generally provided in middle schools.
Measures of Key Outcomes
Measures of key outcomes include the following: Woodcock Johnson III Passage Comprehension, Test of Word Reading Efficiency, Gates-MacGinitie Reading Tests, and a strategy use measure.
Primary Research Method
This study uses a multisite cluster randomized design. Students school-identified as experiencing reading difficulty are randomly assigned to classes, and classes are randomly assigned to treatment and comparison conditions. Teachers’ classes are the units of assignment as well as analysis.
Data Analytic Strategy
A series of hierarchical linear models are estimated to investigate the effect of treatment on vocabulary and comprehension. Average pretest scores (per class) and student-level pretest scores are used as covariates to evaluate the moderating effect of pretreatment differences in reading ability on CSR’s efficacy.
In each participating middle school, approximately six teachers of reading intervention or English language arts classes participate in the study. Participating students are school-identified as experiencing reading difficulty and scheduled for a reading intervention class. From this pool, the project randomly selects the sample of student participants.