The effects of different tasks on children's process and product variables in writing

The main problem of the reported study was to examine the relation between writing tasks (pictures vs, texts) and children's strategies used in the writing process, on one side, and their writing products (narratives): on the other. Twenty 12-year-old children were sampled randomly as subjects for the experiment. The experiment was organised as a 2 x 2 x 2 three-way factorial design with condition (task variable), group, and gender used as independent variables. After having examined the intercorrelations among a preliminary sample of dependent measurements the number of variables was reduced such that the final sample of dependent measures consisted of three groups of variables: 1) three general strategy variables (attention, feedback, and persistence), 2) six domain-specific strategies, and 3) five measures of the writing products used for the final data analyses. The results of the multivariate analyses indicated that tasks presented as text (input) had significant effects on concentration and persistence (general strategies) and on writing time, frequency of pauses in writing, time for long and short pauses (domain-specific strategies), whereas pictorial input had no substantial effects on the writing process variables under research. On the other hand, pictorial material had significant effects on two of the product variables: accuracy and formal aspects of the writing product. Few significant findings were caused by the group and the gender factors used in the study. (C) 1999 Elsevier Science Ltd. All rights reserved.