This study used multigroup structural equations to evaluate the possibility that a theory-driven, evidence-based, yearlong reading program for sixth-grade struggling readers moderated the interrelationships among elements of the simple model of reading (i.e., listening comprehension, word reading, and reading comprehension; Hoover & Gough, 1990). The specific interest was in the relation of theory, program, and evaluation. The motivating assumptions were that 1) a well-designed, theory-based program affects performance in predictable ways and that 2) treatment effects may be present even when group differences in posttest means are not robust. The analysis sample comprised 327 students, 113 in the business-as-usual condition and 214 in treatment. Students were pretested in the fall of sixth grade and posttest data were collected in the fall of seventh grade. There were 217 cases in the posttest sample, 47 comparison students and 170 treatment students. The findings support the possibility that treated sixth-grade students improved in response to an intensive, yearlong intervention, when conceptualizing change in terms of predictable interrelationships of important underlying skills, rather than in terms of group mean differences at posttest. Specifically, the results suggest that verbal knowledge is less proximal to the reading comprehension of students who have become proficient in the use of text processing and reading comprehension strategies.
Roberts, G., Fletcher, J. M., Stuebing, K. K., Barth, A. E., Vaughn, S., & Leroux, A. J. (2012). Treatment effects for adolescent struggling readers: An application of moderated learning and individual differences. Learning and Individual Differences. Advance online publication. doi:10.1016/j.lindif.2012.09.008