Gillespie, A., & Graham, S. (2014). A meta-analysis of writing interventions for students with learning disabilities. Exceptional Children, 80(4), 454–473.
Writing is an essential skill that is a prerequisite for participation in our 21st century economy and civic society. Employers of white-collar workers indicate that they take into account how well one writes when hiring and promoting workers, and 80% of blue-collar workers report that writing is part of their job (National Commission on Writing in America’s Schools and Colleges, 2005). Writing is often used to gauge students’ content understanding (Graham & Hall, 2016). Finally, evidence suggests that writing about what they read improves students' understanding of the material and that teaching students how to write enhances their general reading comprehension, reading fluency, and word reading (Graham & Hebert, 2011).
Many students who have learning disabilities (LD) or are at risk for LD struggle with almost every aspect of writing. When compared with their typically developing peers, the writing of students with LD contains fewer ideas; is more poorly organized; lacks basic structural elements; involves less diverse vocabulary; is less legible; includes more errors involving spelling, grammar, and usage; and is of poorer overall quality (Graham, Collins, & Rigby-Wills, 2015). Although we know that writing is important and that students with and at risk for LD need more support in developing writing skills, questions remain about the best way to teach writing to students with LD.
Gillespie and Graham (2014) conducted a meta-analysis of writing interventions for students in grades 1–12 with documented LD to determine (a) whether writing interventions are effective for students with LD and (b) whether specific types of writing interventions are more effective than others. Meta-analysis is an approach in which the findings of all located studies on a particular topic are systematically extracted and coded, and the effects from these studies are analyzed to answer specific research questions. Meta-analysis is different from a regular review of previous research studies. Meta-analysis attempts to find all published studies on a particular topic and then analyze the effects from those studies statistically so that we know not just whether a particular intervention works, but also how well it works—in other words, the effect of the intervention on the academic outcome of interest.
Gillespie and Graham located and analyzed 43 studies. They used the effects in these studies to answer research questions regarding the impact of writing interventions on writing quality outcomes. Results provided strong support for administering writing interventions to students with LD, particularly strategy instruction, dictation, goal setting, process writing, and—above all—strategy instruction that incorporates self-regulated strategy development (SRSD) and any instruction that includes explicit instruction and modeling (rather than only procedural facilitation or prewriting prompts).
Key findings: The writing quality of students with LD improved through intervention. Writing interventions had a significant positive impact on the writing quality of students with LD, resulting in an average effect size of 0.74, which represents a large, positive standardized difference between students who did and did not receive an intervention. Of the 43 total studies, 38 produced positive effects favoring writing intervention over a comparison condition that was most frequently described as "business-as-usual" writing instruction. These effects were found across all grades, with most involving students in grades 4 to 8. Effects did not differ depending on study quality, study publication year, type of publication (dissertation, technical report, or peer-reviewed journal), or the type of instructor delivering the intervention.
Key findings: At least four studies were located and included in the meta-analysis for each of six types of writing instruction: strategy instruction, dictation, procedural facilitation, prewriting, goal setting, and process writing. Teaching students to plan, write, and revise using strategy instruction significantly improved the quality of writing of students with LD, with an average effect size of 1.09. Effects were more pronounced if strategy instruction was taught via SRSD. Writing strategy instruction taught via SRSD had an average effect size of 1.33, whereas writing strategy instruction without SRSD had an average effect size of 0.76. An effect size of 1.33 suggests that writing strategy instruction using SRSD resulted in large improvements in writing quality for children with LD.
SRSD involves explicit instruction in genre-specific writing strategies through a series of criterion-based lessons that emphasize self-regulation and mastery learning (Harris, Graham, Mason, & Friedlander, 2008). The primary focus of SRSD is on teaching students task-specific writing strategies. Students learning how to write a persuasive essay might learn and memorize the first-letter mnemonic "POW" (Pick an idea, Organize notes, Write and say more) + "TREE" (Topic sentence, Reasons, Explain reasons, Ending). During the SRSD instructional sequence, the instructor develops students’ background knowledge about the features of a given writing genre, introduces the mnemonic, models use of the mnemonic, guides students in memorizing the mnemonic, supports students’ use of the mnemonic while writing, and finally provides opportunities for students to practice using the mnemonic to plan and write independently. Simultaneously, students are taught self-regulatory procedures that include goal setting, self-monitoring, self-instruction, and self-reinforcement. Students use self-talk to facilitate strategy use, monitor and graph their success in achieving writing goals, compare their preinstructional performance with their performance during instruction, and learn to credit their success to the use of the target strategies.
Short (1- to 3-day) dictation interventions also improved the writing quality of students with LD. Students who dictated their compositions into a tape recorder or to a scribe showed greater writing improvements than students who composed by hand, with an average effect size of 0.55, a medium-sized effect. Similarly short (2- to 6-day) goal-setting interventions also improved the writing quality of students with LD (average effect size = 0.57). In these interventions, the instructor developed the goals—students in one study could choose from a menu of instructor-developed goals. Finally, process writing improved students’ writing quality (average effect size = 0.43). Process writing consisted of cycles of planning, drafting, revising, editing, and publishing writing; sustained time for writing for authentic purposes and authentic audiences; and instruction conducted in mini-lessons to target students’ writing needs as they arose.
In contrast, prewriting interventions (which involved activities such as brainstorming or using a graphic organizer to generate and organize ideas for writing) and procedural facilitation interventions (which included supports such as verbal prompts or cue cards that facilitated planning, writing, or revising compositions) did not produce statistically significant positive effects on students’ writing quality. An additional important finding was that the impact of writing interventions that included instruction (average effect size = 0.93) was significantly larger than the impact of writing interventions with minimal or no instruction (average effect size = 0.22). Thus, simply providing students with prompts, graphic organizers, or procedures to use while writing (without providing explicit instruction, modeling, and guided practice) appears to be less effective for students with LD.
Results of this meta-analysis suggest the benefit of writing interventions for students with LD in the elementary and secondary grades. These interventions are most effective when they are in the form of strategy instruction, dictation, goal setting, and/or process writing. Writing strategy instruction, and particularly strategy instruction taught via SRSD, had the biggest impact on writing quality. Writing interventions that included explicit instruction, modeling, and guided practice were more effective that interventions that provided prewriting and/or procedural prompts with minimal instruction.
Clearly, explicit writing instruction using an effective program such as SRSD is most beneficial for improving writing in children and adolescents with LD. However, this meta-analysis also shows that some relatively simple techniques such as dictation are moderately effective and can be implemented relatively quickly and easily. One implication for practice would be to immediately introduce these easy-to-implement techniques while learning how to fully implement a highly effective program such as SRSD.
Gillespie and Graham note that good writing involves the simultaneous use of multiple skills (including planning, transcribing, revising, and editing) and that students with LD tend to struggle with many of these skills. The authors call for future research evaluating comprehensive writing programs and multicomponent interventions that involve teaching a wide range of writing skills to students with LD.
Graham, S., Collins, A., & Rigby-Wills, H. (2015). A meta-analysis examining the writing characteristics of students with learning disabilities and normally achieving peers. Manuscript submitted for publication.
Graham, S., & Hall, T. E. (2016). Writing and writing difficulties from primary grades to college: Introduction to the special issue. Learning Disability Quarterly, 39(1), 3–4.
Graham, S., & Hebert, M. (2011). Writing-to-read: A meta-analysis of the impact of writing and writing instruction on reading. Harvard Educational Review, 81, 710–744.
Harris, K. R., Graham, S., Mason, L. H., & Friedlander, B. (2008). Powerful writing strategies for all students. Baltimore: Brookes.
National Commission on Writing in America’s Schools and Colleges. (2005). Writing: A powerful message from state government. New York, NY: College Board.