Research

One study (Denton, Anthony, Parker, & Hasbrouck, 2004) reviewed by the WWC investigated the effects of a modified version of Read Naturally on an English language learner sample. Although there was random assignment to treatment groups, three students assigned to the control group were reassigned to the treatment group, and three students assigned to the treatment group were reassigned to the control group one week after the study had begun (as requested by the participating schools). Therefore, this study was determined to be a quasi-experimental design that met the WWC evidence standards with reservations. In addition, data from three students in the comparison group were eliminated from the analysis because of exposure to Read Naturally in their classroom, and no data were eliminated from analysis in the treatment group. Although this created differential attrition rates between the study groups (10% attrition in the comparison group and 0% attrition in the treatment group), the authors were able to demonstrate post-attrition equivalence between groups on the pretest.

Denton, Anthony, Parker, & Hasbrouck (2004). The study that examined Read Naturally included 60 participants. The Read Naturally intervention group received English language pull-out tutoring during the school day in addition to their regular English instruction. The control group received only their regular English language pullout instruction.

Extent of evidence

The WWC categorizes the extent of evidence in each domain as small or medium to large (see the What Works Clearinghouse Extent of Evidence Categorization Scheme). The extent of evidence takes into account the number of studies and the total sample size across the studies that met WWC evidence standards with or without reservations. ⁴

The WWC considers the extent of evidence for Read Naturally to be small for reading achievement. No studies that met WWC standards with or without reservations addressed mathematics achievement or English language development.