This paper outlines the theoretical framework of a two-year two cohort study of inter-related modules to model reading in 300 grades 4, 5 and 6 readers, and describes the specially designed microcomputer programs written in Turbo Pascal to implement this modular approach. In brief, the modular approach postulates the interrelated phonological, morphological modules in processing words and sentence/paragraph processing. These three modules or components are subserved by 10 measurable tasks all implemented on the microcomputer (IBM-PC compatible) with reaction time measures as indices of the different facets of verbal efficiency. Statistical analyses of the large data sets for Phase 1 show the linear structural equation modelling fits the data reasonably well. In particular, the morphology task consistently explains between 32% to 44% of the variance in reading performance in the three grades. The findings lend themselves to specific approaches to habilitation in the different components or sub-components.
Keyword Computer literacy - phonology - morphology - microcomputers - sentence processing
This study was supported by a grant (No. 410-86-0048) from the Social Sciences and Humanities Research Council of Canada awarded to Che Kan Leong. Simon Lock and Jacky Lau were responsible for the microcompurter programming. Edward Danskin, Melvin Kelley, Donald Simmons, Louise Sellinger, Kristin Strauss and Tara Wiley carried out the experimentation with individual children in the schools. The cooperation of the principals, the teachers and the 
subjects
was greatly appreciated. Requests for reprints should be addressed to Dr. C. K. Leong, Department for the Education of Exceptional children, University of Saskatchewan, Saskatoon, Canada, S7N OWO.