Project Activities
Phase I of this project included search, screening, and selection. Researchers conducted a systematic search of relevant writing studies that provide sufficient information to be included in this synthesis. In Phase II, the research team coded studies, estimated study effect sizes, and completed data analysis to evaluate the included studies.
Structured Abstract
Setting
The sample of 378 studies were conducted in the U.S. as well as 23 other countries around the world. All writing interventions were implemented in typical school settings.
Sample
The sample of studies in the meta-analysis included 378 eligible studies that included 3,727 effect sizes. Inclusion criteria for the studies were: (a) participants in kindergarten through grade 5, (b) the study investigated a writing intervention that involved more than 50% of instruction time devoted to writing, (c) the study included at least one writing outcome measure, (d) studies used randomized control trials (RCT) or quasi-experimental designs with pretest, and (e) the study was reported in English. Included studies were published from 1931 to 2023.
Researchers reviewed and synthesized empirical evidence from writing interventions. This meta-analysis systematically explored (a) what interventions are effective on what writing outcomes, (a) for whom (e.g., students with disabilities, second language learners), and (c) under what conditions (e.g., dosage, instructor) in kindergarten through grade 5.
Research design and methods
The meta-analysis methods included two project phases. Project Phase I—Search, Screening, and Selection identified all published and unpublished studies that met eligibility criteria using the following systematic search methods: electronic databases searches, reference harvesting, journal hand searches, examination of previously conducted reviews, and consultation with prevalent national and international writing researchers. The research team began by screening titles and abstracts and then reviewed full-text articles to select studies for inclusion. Project Phase II—Data Extraction and Analysis involved rigorous coding of all eligible studies and extracting data to calculate standardized mean difference effect sizes (and associated variance). The research team completed independent reliability coding of 20% of the studies that met inclusion criteria. Agreement on screening decisions and study coding exceeded 0.91 for each project phase. The research team conducted data analyses using robust variance estimation techniques and systematically explored 21 identified writing interventions (including 15 subcategories of writing interventions) across 15 writing outcomes for elementary students.
Control condition
All eligible studies included a group that received a contrasting comparison condition such as Business as Usual (BAU) instruction, intervention in another academic subject (e.g., mathematics, reading), or alternative writing activities (e.g., standard grammar curriculum). Studies could also include more than one writing treatment (e.g., one spelling intervention and one handwriting intervention) if the study designated a specific control or included a comparison condition (e.g., BAU).
Key measures
Learner outcomes in this meta-analysis were the standardized mean difference effect sizes between intervention and control conditions across 15 writing outcomes that included measures of: (1) products, (2) process, (3) skills, (4) knowledge, and (5) motivation. Across these learner outcomes, the research team examined associated variance and explored moderators (learner populations and instructional conditions) of study effect sizes. This included exploration of intervention effects for students with disabilities and at-risk for disabilities. Other moderators of instructional conditions included intervention dosage, instructor, and measures of study quality related to intervention implementation (e.g., method for measuring fidelity).
Data analytic strategy
The research team calculated separate standardized mean difference effect sizes between writing interventions and control conditions for each writing outcome. Researchers then used robust variance estimation to estimate separate mean effect sizes for each intervention type and writing outcome. In addition, the research team conducted systematic moderator analyses to examine heterogeneity of effect size variability across writing interventions, outcomes, populations of students, and intervention conditions.
Key outcomes
The main findings of this project are as follows:
- Findings from the meta-analysis support the following recommendations for teaching writing to elementary students:
- Write, but writing is not enough
- Support students as they write
- Teach foundational writing skills
- Teach writing strategies
- Teach creativity, critical thinking and imagery
- Teach summary writing
- Enhance students’ writing knowledge
- Apply twenty-first century writing tools
- Write across the curriculum
- Connect writing and reading instruction
- Create a motivating writing environment
People and institutions involved
IES program contact(s)
Project contributors
Products and publications
Project website:
Publications:
Graham, S., Collins, A. A., & Ciullo, S. (2024). Evidence-based recommendations for teaching writing. Education 3-13, 1–14.
Collins, A. A., Rigby-Wills, H., & Ciullo, S. (2024). Quality Writing Instruction for Students in the Elementary Grades. In B. L. Hott & P. Williamson (Eds.) Quality Instruction and Intervention for Elementary Educators (pp. 88-106). Rowman & Littlefield.
Inside IES Research Blog
Collins, A.A. (2023, January 18) Research to Accelerate Pandemic Recovery in Special Education: Grantee Spotlight Blog Series Featuring Dr. Alyson Collins. Inside IES Research.
Available data:
The researchers developed a database of writing intervention effects to better understand on what writing outcomes, for whom, and under what conditions various writing interventions work best for students in kindergarten through grade 5. The database is expected to be publicly available in 2025 in the data repository, Texas State University Dataverse Repository https://dataverse.tdl.org/dataverse/txst.
Related projects
Questions about this project?
To answer additional questions about this project or provide feedback, please contact the program officer.
