New Research: 16 Writing Interventions That Work
New research highlights that when it comes to writing instruction for middle and high school students, it’s important to get it, well, right.
“It does matter what you do when you teach writing because some things don't appear to work so well,” says Steve Graham, lead author of a new analysis of writing interventions (or “treatments”) for students in grades 6-12.
The meta-analysis was published in the Journal of Educational Psychology and used data from 406 previous experiments involving 52,529 students. While no interventions studied in this analysis had a negative effect on writing outcomes, some interventions didn’t have significant positive effects. For instance, “Just simply increasing how much students write in middle school or high school does not improve their writing,” Graham says.
On the bright side, however, many interventions did help students improve their writing, and writing practice paired with enhanced instruction could make a significant difference.
In their analysis, Graham and his co-authors identified 16 categories of writing interventions that had a positive impact on student writing and can serve as a guide to writing instructors. However, Graham cautions even though analysis of this kind provides teachers with tools that have a proven track record, there are no guarantees these strategies will work in all instances.
“Just because a writing practice worked in eight other teachers' classrooms and had a positive effect, it does not mean that it is for certain going to work in your classroom,” he says. “Because the conditions under which research studies are conducted are different from the ones that may exist in an individual teachers’ classroom.”
That’s why Graham advises instructors to use these 16 writing interventions as inspiration and to keep monitoring and adjusting their own practices based on what is and isn’t working for their students.
Note: This study, like most education studies, measured effect sizes in standard deviation, which is the average deviation from the mean score for a group of students studied. In education research, effect sizes are notoriously small, so anything approaching a standard deviation of 1 is significant. For instance, a student with a 1.5 below the mean would require significant intervention.
1. Comprehensive Writing Programs
Standard deviation: 0.47
These programs include those based on the process approach to writing, or what is sometimes called a writers’ workshop. Graham and his co-authors note this includes “extended opportunities for writing; writing for real audiences; engaging in cycles of planning, translating, and reviewing; personal responsibility and ownership of writing projects; high levels of student interactions and creation of a supportive writing environment; self-reflection and evaluation; personalized individual assistance and instruction; and, in some instances, systematic instruction.”
2. Strategy Instruction
Standard deviation: 0.76
This method of writing instruction involves explicitly teaching strategies for planning, revising, self-assessing, and/or editing text, the study authors note. Writing strategies range from processes, such as semantic webs to strategies designed for specific types of writing, such as stories.
3. Digital Writing Tools
Standard Deviation: 0.31
When students moved from using pen and paper to writing with the help of a traditional word processor over time, there was a marked improvement, Graham says. “The reasons for that are fairly obvious,” he adds. It’s easier to self-edit and move words around, and grammar and spellcheckers also help with the process. An even greater improvement was seen among students who had access to more advanced word processors, which might include the ability to add images and sound, or have gamified elements to help students learn to write.
4. Transcription Instruction
Standard deviation: 0.71
This positive effect was seen in lessons that included teaching spelling, handwriting, or keyboard use.
5. Computer-Assisted Instruction
Standard deviation: 0.32
This included teaching writing, spelling, and other lessons with the help of a computer program as well as technology provided personalized instruction. However, computer-generated feedback on writing, in and of itself, did not provide a benefit, Graham says. This research predated the rapid advances in generative AI over the past year, so computer-assisted benefits and limitations will need to be updated in the future.
6. Teaching Critical/Creative Thinking Skills for Writing
Standard deviation: 0.27
Teaching students critical thinking strategies improved writing as did teaching them how to add more creativity to their work. “Creativity could be exemplified in a study in which students were taught how to use metaphors, similes, etc., and critical thinking could be shown in a study in which students are analyzing their texts using questions that help them think critically about the veracity and value of the content they're looking at,” Graham says.
7. Emulating Good Models of Writing
Standard deviation: 0.46
Graham and his co-authors defined this as, “Examining one or more examples of model texts or models for carrying out writing processes and attempting to emulate these models when writing.” This is something many professional writers do intuitively, so it makes sense that it would help with student writers.
8. Feedback
Standard deviation: 0.34
“We found feedback makes a difference,” Graham says. This included instructor feedback as well as peer and group feedback but notably not self-assessment/feedback or computer-generated feedback, at least in the studies looked at for this analysis.
9. Goal Setting
Standard deviation: 0.44
Whether teacher-assigned or based on students’ own goals for for writing or learning writing skills and processes, goal setting seemed to have a measurable impact on writing success.
10. Prewriting Activities
Standard deviation: 0.49
“If you engage students in prewriting activities to gather or organize information – so it might involve discussion, or using some kind of organizer to generate and organize your ideas – writing gets better,” Graham says.
11. Grammar Instruction
Standard deviation: 0.77
This positive association was much stronger than in some previous research into grammar’s impact on writing, including work Graham has been involved in. He says the change is likely due to better methodology that eliminated less well-designed grammar interventions as well as other factors. Most of the studies involve teaching grammar in context, he adds. So it was not the old form of grammar instruction that involved fill-in-the-blanks exercises and decontextualized practice around specific contexts.
12. Sentence Instruction
Standard deviation: 0.73
“We found that teaching students how to create more complex sentences had a positive effect on students’ writing,” he says. “When you write a lot of your cognitive efforts and resources are engaged in taking your ideas, images, etc, and translating them into an acceptable sentence that conveys your intended meaning, and is going to be understandable to the reader. So when you teach kids to be more facile with construction, then there's a positive effect on your writing.”
13. Inquiry
Standard deviation: 0.92
“We don't have as much data on this, but the idea behind inquiry is that you're gathering information that you're analyzing, that will help you in terms of your writing assignments,” Graham says. The study notes this could include comparing and contrasting cases or collecting and evaluating evidence.
14. Observing Writers/Readers, Peer Assistance
Standard deviation: 0.41
Simply observing other writers, readers of writing, or teachers/peers as they model how to go about a writing process or skill, can also improve writing outcomes.
15. Summarization Instruction
Standard deviation: 0.49
This can take the form of either sharing summarization strategies or direct instruction in which you present a summary, you discuss it, your students practice, and you get feedback, Graham says.
16. Text Structure Instruction
Standard deviation: 0.39
Graham and his co-authors defined this as strategies in which teachers explicitly teach students knowledge about the purpose and/or structure of specific types of text, such as stories or persuasive texts. Once again, few writing instructors will be surprised this is an effective method.
