An Investigation of Effective Methods for Using Writing in Secondary Math and Science to Improve Student Outcomes

Gaps in writing, math, and science achievement (e.g., SAT and NAEP scores) continue to highlight inequity among groups of U.S. students. Over the last few decades, research has documented the role that writing plays in student learning. Further, prior studies show that effective writing instruction can improve students' content knowledge of math and science, as well as their skills in general writing composition and writing that is specific to the math and science disciplines. For secondary students to leave high school prepared to engage in post-secondary education or careers in math and science, they should be frequently writing in all disciplines. Writing instruction in math and science is an essential area of research to ensure equitable K-12 and college experiences, and to better prepare all students in ways that provide opportunities to pursue STEM career pathways. Despite this, the requirement to teach writing in math and science disciplines (per national and state standards) is relatively new, and the field needs guidance for how to implement evidence-based practices for writing in math and science. While a strong literature base supports the efficacy of secondary writing instruction generally, few syntheses have specifically focused on identifying the effectiveness of writing instruction in math and science for students with and without diverse learning needs. Thus, a meta-analysis is needed to understand how, for whom, and under what conditions writing instruction in secondary math and science can be most effective. This project is a meta-analysis in the area of secondary (grades 6-12) math and science (M/S) writing instruction. Exploring research from 25 years, this three-year project will: a) determine the overall mean effect of M/S writing instruction on students’ M/S achievement, M/S-specific composition, and general composition; b) identify moderators of M/S writing instruction effects; and c) evaluate the reporting quality and risk of bias for included studies. This synthesis will include studies that employed experimental or quasi-experimental group design and investigated the effectiveness of writing instruction that had the purpose of improving students’ math/science content knowledge, writing skill (M/S-specific writing or general writing), or both. Using random-effects multilevel multivariate meta-regression with robust variance estimation, this meta-analysis addresses limitations of previous reviews by focusing on secondary students in M/S learning contexts, including instruction for all writing types and purposes (e.g., not limiting included studies to only writing to learn activities), and attending to methodological limitations of previous reviews. Given the recent focus on discipline-specific writing, and the role that M/S learning plays in school-based achievement and adult outcomes, now is a crucial time to conduct this meta-analysis. Findings from this meta-analysis will provide directions for the next steps in research on teaching and learning in math and science. The findings will also provide practical implications, such as effective methods for designing writing instruction and training for pre- and in-service teachers, which can ultimately help improve student learning.