Ask a teacher what revision means in their classroom, and the answer is often a checklist: fix run-ons, correct apostrophes, swap in stronger verbs. Ask a writing researcher the same question, and the answer points somewhere else entirely. Revision and editing are not the same skill, and most classroom revision protocols train the wrong one.

Real revision requires re-seeing the argument, not re-reading the sentence. A student who corrects ten comma splices has improved the surface of a draft without touching its reasoning, its evidence, or its structure. That distinction is not a matter of opinion. It is one of the most consistent findings in composition research over the past four decades.

Nancy Sommers tested it directly. In her 1980 study comparing student writers to experienced adult writers, she had both groups revise drafts and tracked exactly what kind of changes they made. Student writers overwhelmingly made lexical substitutions—swapping one word for another to avoid repetition—while leaving the underlying structure of their argument untouched. Experienced writers did something different: they reread their drafts for the relationship between what they intended to say and what the draft actually said, then restructured paragraphs, cut sections, and reordered claims to close that gap. Sommers called this the difference between revising and editing, and found that most students had never been taught the first one.

The distinction holds up under direct intervention testing. MacArthur, Philippakos, and Ianetta tested a revision-focused strategy curriculum on 276 students across 19 developmental writing classes at two universities, comparing it against standard instruction for a full semester. Students taught to evaluate and reorganize their own arguments produced markedly stronger essays, with an effect size of 1.22 on overall quality. Grammar instruction in the same classes showed no comparable gain. Time spent teaching students to find and fix structural problems in their own arguments moves writing quality. Time spent on usage drills does not.

Zoi Traga Philippakos’s 2023 synthesis of writing-instruction research reaches the same conclusion from a wider survey of the field: students revise more successfully when instruction explicitly separates meaning-level concerns from surface-level concerns, and when they are taught to identify problems in their own drafts rather than apply corrections handed to them. Four decades after Sommers, the finding has not aged out. It has been replicated.

This gap shows up most clearly in how revision gets assessed. A rubric that credits a “revised draft” for having fewer errors than the first draft is measuring editing, not revision. A rubric that asks whether the writer’s central claim is easier to locate, better supported, or more clearly organized in the second draft is measuring the skill Sommers and Traga Philippakos describe. Few classroom revision rubrics make that distinction explicit, which means two students doing entirely different cognitive work can receive identical credit for “revising.”

None of this makes editing worthless. Conventions matter, and a draft riddled with errors is harder to read regardless of how strong its argument is. The problem is conflation. When a revision assignment consists of a proofreading checklist, it is functioning as an editing assignment, whatever it is labeled. Students complete it, the teacher checks a box, and the part of the writing process that research identifies as separating competent writers from developing ones never gets practiced.

The fix is not more time. It is a different question asked at the revision stage. Instead of asking what needs to be corrected, the question has to be what the draft is actually arguing, and whether every paragraph serves that argument. That question forces a student back into the structure of the piece rather than its surface. It is harder to answer, harder to grade, and it is the only version of revision the research supports calling by that name.

Further Reading