Prioritizing Conventions Instruction from Error Patterns
Ms. Alvarez teaches four sections of seventh-grade ELA; the target class has 26 students. Over five weeks she collected three timed persuasive essays and coded conventions errors. Nineteen students wrote at least one sentence fragment in body paragraphs; fourteen produced comma splices; eight had subject–verb agreement errors with indefinite pronouns; twenty-two stated a clear thesis. Only three students made all three sentence-boundary errors. The district rubric weights ideas 40%, organization 30%, conventions 20%, and voice 10%. The next unit lasts two weeks, includes one required revision cycle, and meets in 45-minute periods. Ms. Alvarez can provide one whole-class mini-lesson per day and two small-group pull-out sessions per week. She wants to improve conventions without sacrificing the revision cycle.
1. Which convention should Ms. Alvarez prioritize for whole-class instruction, and which students should receive small-group support? Use the data to justify your choices.
Read the worked answer
Prioritize sentence fragments for whole-class instruction. Fragments appear in 19 of 26 essays (73%), far more than comma splices (14/26, 54%) or subject–verb agreement (8/26, 31%). Fragments also damage clarity more directly than a single agreement slip, and they are the error most likely to recur across the revision cycle.
Do not spend whole-class time on subject–verb agreement yet. Only eight students show that error, and they are not the same eight who struggle most with fragments. A whole-class lesson would reteach most students unnecessarily and consume time needed for revision.
Group by error profile, not by overall writing level. Pull-out group A: students with repeated fragments plus comma splices; focus on sentence-boundary decision-making using mentor sentences. Pull-out group B: students with comma splices only; focus on joining clauses. Students with subject–verb issues get an editing-checklist item and brief conferencing, not a full lesson.
Protect the revision cycle. Use the whole-class mini-lessons to front-load fragment work before the first revision, so students apply the lesson to their own drafts.
2. A colleague argues that subject–verb agreement should be taught first because it is 'basic grammar.' How should Ms. Alvarez respond, and what risk does the colleague's approach create?
Read the worked answer
She should agree that agreement matters but point to the incidence data: eight students, not the class. Teaching it whole-class first would spend scarce instructional time on an error that most students do not make, while fragments remain unaddressed in nearly three-quarters of the essays.
The risk is twofold. First, opportunity cost: every day spent on a low-incidence error is a day not spent on the error pattern that most affects clarity. Second, misdiagnosis: treating 'basic' as 'first' assumes all students need the same sequence, when the data show clustered, uneven needs.
A defensible compromise is to embed agreement in the editing checklist and in small-group conferencing for the eight students, while keeping whole-class time on fragments and comma splices. The colleague's concern can be honored without letting it drive the whole-class sequence.
3. What evidence would show that the conventions intervention worked, and what evidence would NOT be sufficient?
Read the worked answer
Sufficient evidence: compare the rate of the targeted error in a new, timed persuasive essay before and after instruction, using the same coding scheme. Also examine whether students corrected fragments during the required revision without teacher identification. A useful metric is the percentage of sentences that are complete in body paragraphs, not just a count of errors.
Not sufficient: a grammar quiz on fragments; a clean final draft after peer editing; or the teacher's impression that lessons went well. Those do not show transfer to independent writing.
Why: the goal is not to recognize a fragment in isolation but to produce complete sentences under normal writing conditions. A quiz can improve without changing composing behavior, and a corrected draft can reflect peer or teacher editing rather than student learning.