Abstract
This review examines which types of instruction in writing-to-learn lead to effects on insight and topic knowledge in different disciplines, in grades 5–12 and in higher education. Forty-three empirical studies have been selected to answer this question. Four types of instruction are distinguished. Three of them are based on hypotheses proposed by Klein (1999) about the cognitive processes involved in writing-to-learn: Forward Search, Genre Writing and Backward Search. The fourth type, Planning Only, arises from the literature reviewed. Results of the studies show that about two thirds of the (quasi) experimental studies lead to positive effects on insight and topic knowledge for the four types of instruction. However, given the small number of experimental studies conducted, no firm conclusions can be drawn for three types of instruction. For the remaining type of instruction, Genre Writing, a larger number of studies provide positive evidence. Suggestions for future research are discussed.
Context and implications
Rationale for this study
It is known that instruction is needed for promoting learning by writing. The question is which specific types of instruction can be used for inciting the writing-to-learn process.
Why the new findings matter
The finding that the four types of instruction, especially Genre Writing, may promote learning, can serve as a reference for judging or formulating instruction.
Implications for educational researchers and policy makers
The outcome may be interesting for education, because it entails that one from four types of instruction, that is, Genre Writing, shows the largest positive effects. Subject teachers often are hesitant to use writing-to-learn in their teaching practice, therefore the findings may provide clearness about how to instruct it. They can use Genre Writing instruction for composing writing tasks and select the preferred content of this instruction. The authors indicate mechanisms underlying the process of writing-to-learn that may be stimulated by instruction. For researchers it can be interesting to study these mechanisms, for instance in think-aloud studies.