The Effects of Writing on Learning in Science, Social Studies, and Mathematics: A Meta-Analysis

Steve Graham, Sharlene A. Kiuhara, Meade MacKay

Research output: Contribution to journalArticlepeer-review

94 Scopus citations


This meta-analysis examined if students writing about content material in science, social studies, and mathematics facilitated learning (k = 56 experiments). Studies in this review were true or quasi-experiments (with pretests), written in English, and conducted with students in Grades 1 to 12 in which the writing-to-learn activity was part of instruction. Studies were not included if the control condition used writing to support learning (except when treatment students spent more time engaging in writing-to-learn activities), study attrition exceeded 20%, instructional time and content coverage differed between treatment and control conditions, pretest scores approached ceiling levels, letter grades were the learning outcome, and students attended a special school for students with disabilities. As predicted, writing about content reliably enhanced learning (effect size = 0.30). It was equally effective at improving learning in science, social studies, and mathematics as well as the learning of elementary, middle, and high school students. Writing-to-learn effects were not moderated by the features of writing activities, instruction, or assessment. Furthermore, variability in obtained effects were not related to features of study quality. Directions for future research and implications for practice are provided.

Original languageEnglish (US)
Pages (from-to)179-226
Number of pages48
JournalReview of Educational Research
Issue number2
StatePublished - Apr 1 2020


  • instruction
  • learning
  • mathematics
  • science
  • social studies
  • writing
  • writing-to-learn

ASJC Scopus subject areas

  • Education


Dive into the research topics of 'The Effects of Writing on Learning in Science, Social Studies, and Mathematics: A Meta-Analysis'. Together they form a unique fingerprint.

Cite this