What Actually Works When You Try to Assess Student Writing Before the Final Draft
Most teachers I know hand out a rubric and hope for the best. The students glance at it, write whatever feels comfortable, and submit a draft that completely ignores every criterion on that page. This is why formative assessment for writing exists as a practice, not just a document. The goal is to catch issues while there is still time to fix them, rather than discovering after submission that the argument never really landed. Traditional grading evaluates the final product. Formative assessment for writing evaluates the process, the reasoning, and the gaps between what the student intended and what actually made it onto the page. You are looking for evidence of learning in progress. A single essay graded at 72 percent tells you almost nothing about why that student struggled. A writing conference where they read their own introduction aloud and you ask, "What were you trying to argue here?" tells you everything. I spent three years collecting drafts and returning them with heavy red-pen corrections before I realized I was doing exactly the wrong thing. Students read the marks, got discouraged, and made superficial changes just to satisfy the surface-level edits. The deeper structural problems never got addressed. The turnaround happened when I shifted to using a single sticky note per draft that asked one targeted question instead of listing twelve corrections. That single question approach forced the student to think, not just comply. It cut my response time per draft from roughly twenty minutes down to about six, and the quality of revision improved noticeably.
The Practical Workflow
Here is how I structure this in a standard high school or college writing classroom without it consuming my entire week. The first step is a pre-writing checkpoint. Before any drafting begins, students submit a one-paragraph thesis statement and a three-bullet outline of their supporting points. You scan this quickly. If the thesis is vague or the outline skips over a key counterargument, you flag it immediately. Catching a weak thesis at this stage saves hours of rework later. Most students will not voluntarily revise a thesis once they have written two thousand words around it, so this early intervention matters significantly more than late-stage feedback. The second step is a mid-draft check-in. By this point, students should have a complete first draft, but it does not need to be polished. The expectation is that the ideas are visible even if the mechanics are rough. At this stage, I use targeted conferences lasting two to four minutes each. The conference follows a simple script: the student reads their own work aloud, I take notes on where they stumble or where the logic breaks down, and then we discuss one or two specific issues rather than trying to cover everything at once. One or two issues is the ceiling, not the floor. More than that and the student walks away overwhelmed with nowhere to actually begin.
The third step is a structured peer review session. This is where most people mess up. Peer review only works if the reviewers have extremely specific criteria to evaluate against, and those criteria must align with the actual grading rubric. A generic "give feedback on organization" prompt produces generic, useless feedback. Instead, peers look for whether each paragraph contains a clear topic sentence and whether the evidence presented in each paragraph directly supports that topic sentence. These are binary observations. The reviewer does not need to be an expert. They simply check for presence or absence of required elements. The final step is a reflective cover sheet. When the student submits the revised draft, they fill out a short form answering three questions: what changed from the first draft, what remains unclear to them, and what specific part of the revision took the most effort. This forces metacognition. It also gives you a direct signal about where to focus your attention during the final evaluation phase.
Get the Full Details
Common Pitfalls That Derail the Process
The most common pitfall is feedback overload. Returning a draft with comments on grammar, structure, argumentation, word choice, and citation format all at once creates paralysis. The student does not know which layer to address first. I limit my written comments to a maximum of three per draft, and those three must be about ideas, not mechanics. Grammar issues get circled and noted at the margin with a reference to a handbook section. The idea-level feedback gets the real attention. A less obvious but equally damaging pitfall is treating formative assessment as low-stakes with no accountability. If the mid-draft checkpoint and the peer review session carry zero weight toward the final grade, students will treat them as optional busywork. The simplest fix is to allocate ten to fifteen percent of the total assignment grade to the formative process itself, broken into small increments: five percent for the pre-writing checkpoint, five percent for the mid-draft conference, and five percent for the reflective cover sheet. This signals that the process matters without turning the entire exercise into a bureaucratic hoop-jumping contest. Another frequent issue is the assumption that all students respond to the same type of feedback. Some learners thrive on direct, explicit instruction. Others shut down under it. I learned this the hard way with a student who produced remarkably thin drafts despite repeated substantive conferences. The problem was not that the student lacked ability. The problem was that my conferences focused entirely on macro-level arguments while the actual barrier was paragraph-level coherence. Once I switched to using sentence combining exercises as a diagnostic tool, the blockage cleared within two sessions. Different writers present different hidden bottlenecks, and you cannot identify them through generic feedback alone.
When This Approach Breaks Down
Formative assessment for writing requires a teacher-to-student ratio that makes it impractical in some settings. If you are managing a class of forty or more students and you attempt full mid-draft conferences for every single learner, you will either burn out or reduce the conferences to meaningless thirty-second check-ins that accomplish nothing. In large cohort courses, the workshop model becomes more viable than individual conferences. You structure the class period around in-progress writing, with you rotating through tables and giving small-group feedback to three or four students at a time while the rest of the class continues working. This scales better but still requires disciplined time management. The second breakdown scenario involves subjects where writing is secondary to the learning objective. In a chemistry lab report course, for example, the priority is experimental design and data interpretation, not rhetorical sophistication. Formative assessment still has a role here, but the criteria shift entirely toward clarity of method description and accuracy of data presentation rather than argumentation or stylistic nuance. Using a literature-focused rubric on a science report will confuse both you and the students.
A Simple Tool You Can Build Yourself
I do not rely on commercial assessment platforms because they tend to overcomplicate a process that does not require software. Instead, I use a shared spreadsheet with four columns: student name, draft stage (pre-writing, mid-draft, final), feedback focus area (one phrase maximum), and completion status. The feedback focus area column forces me to be deliberate about what I am addressing. I cannot write a paragraph of comments and paste it into a single cell. I have to distill my observation down to its essence, which usually improves the quality of my feedback on its own. The spreadsheet also serves as a tracking record. When I look back at the end of the semester, I can see which students received repeated feedback on the same issue across multiple drafts. That pattern identification is genuinely useful for adjusting my instruction in subsequent terms. It also helps during parent-teacher conferences because I have concrete evidence of progress trajectories rather than relying on memory or impression. The underlying mechanism here is straightforward: assess the writing process at multiple checkpoints, give limited and targeted feedback, require student reflection, and hold the process accountable through minor grade weight. It is not elegant. It is not particularly exciting to explain at a conference. But it produces measurably better drafts and gives you information about student understanding that a single summative grade never will.
