Fifty-six Toronto (Ontario, Canada) seventh-grade and eighth-grade learning-disabled students whose handwriting was very difficult to read were randomly assigned to either an experimental or comparison group. Experimental group students were loaned a portable computer to use freely at school and at home during the course of the experiment. Computer-written and hand-written compositions done in time-controlled situations were collected 14 times over an 8-month period. These pieces were scored holistically and analytically. Data were also collected on experimental students' self-reported, parent-reported, and teacher-reported use of the computers. Results indicated that: (1) half of the experimental students made frequent use of the computers at home and at school, and (2) a majority of these frequent users (and far more of them than of either the infrequent users or the comparison subjects) made significant improvement in the quality and quantity of their written work. Differences in improvement in quality and quantity of writing between frequent users and others were statistically significant. No significant differences were, however, observed in surface features such as spelling and punctuation. Assessment tools used and 12 references are appended. (Author/JDD)