Twenty-four children with writing problems were given instruction in handwriting automaticity, spelling strategies, and the composing process (plan, write, review, revise) in 14 one-hour individual tutorials during the summer between third and fourth grade. Half the children (8 boys, 4 girls) received extra practice in composing, while half the children (8 boys, 4 girls) received special training in orthographic and phonological coding. Hierarchical linear modeling of growth curves was used to compare the treatment groups to a non-contact control group (10 boys, 5 girls) on a standard battery at pretest, midtest, posttest, and the two treatment groups with each other on probe measures of handwriting, spelling, and composition in each tutorial session. The treatment groups improved at a faster rate than the control group on some measures of handwriting, spelling, and composition (fluency and quality) in the standard battery, but Verbal IQ did not predict rate of improvement. Differences were found between the two treatment groups in some probe measures of writing and a motivation variable (work avoidance). Repeated-measures ANOVA was used to compare treatment groups to a non-contact control group at pretest, midtest, posttest, and follow-up. Differences between the treatment and control groups favoring the treatment groups were maintained at 6-month follow-up on some handwriting, spelling, and composition (quality) measures. Individual differences were found in learner characteristics prior to treatment and in response to the same treatment. The importance of affect and motivation as well as cognitive variables is emphasized.