Since the early 1970s, many educators have touted writing as a means of enhancing learning. Several reasons have been suggested for this purported enhancement: that writing is a form of learning, that writing approximates human speech, that writing supports learning strategies. Alternatively, some researchers have cautioned that the educative effects of writing may be contingent on the contexts in which it occurs. The research on writing's effects on learning is ambiguous. This meta-analysis of 48 school-based writing-to-learn programs shows that writing can have a small, positive impact on conventional measures of academic achievement. Two factors predicted enhanced effects: the use of metacognitive prompts and increased treatment length. Two factors predicted reduced effects: implementation in Grades 6-8 and longer writing assignments. (Contains 2 figures, 5 tables, and 1 note.)