The Relative Contributions of Research-Based Composition Activities to Writing Improvement in the Lower and Middle Grades

In a benchmark meta-analysis of experimental research findings from 1962 to 1982, Hillocks (1986) reported the varying effects of general modes of instruction and specific instructional activities (foci) on the quality of student writing. The main purpose of the present study was to explore the relative effectiveness of those modes and foci using a non-experimental methodology and a new group of 16 teachers and 275 students in grades 1, 3–6, and 8. Teachers who had attended a summer writing institute reported on 17 different instructional variables that were primarily derived from the meta-analysis during each week of a ten-week treatment period that occurred at the beginning of the next school year. A pre- and post- treatment large-scale writing assessment was used with a prompt that allowed latitude in student choice of topic and extra time for prewriting and/or revision. Large gains in quality and quantity were found in the lower grades (1, 3, and 4) and smaller gains were found in the middle grades (5, 6, and 8). The demographic variables of SES, primary language, residence, and gender were found to have small and/or insignificant relationships to gains. Teacher-determined combinations of instructional variables and their relationship to gains in quality were investigated through factor analysis while controlling for pretreatment individual differences. Only one combination of activities was associated with large gains, and it was interpretable as the environmental mode of instruction. This combination included inquiry, prewriting, writing about literature, and the use of evaluative scales.