Fourteen Proficient Scores, Four Students Who Cannot Explain Their Control Group
Ms. Lindqvist teaches ninth-grade biology. Her 27 students completed a lab-report performance task scored on a 4-point analytic rubric covering question, design, evidence, and conclusion. Fourteen of 27 scored 3 or 4 overall. When she interviewed six of those fourteen, four could not explain why they chose their control group, though their written reports described it correctly. Two students with IEPs scored 2 despite giving strong oral explanations of their design when she questioned them. Her gradebook policy requires entering the task as a summative score this week, and the department requires summative scores to reflect the rubric as written.
1. Decide what goes in the gradebook this week, and justify the decision against the department rule and the evidence in front of her.
Read the worked answer
Decision: enter the rubric scores as written - the department rule is legitimate and a teacher who quietly adjusts scores upward creates a defensible-looking grade built on an indefensible process. But enter them as one evidence source, not as her conclusion about what students know, and say so in the comment field and to her team.
Reasoning: the scores are valid measures of written lab-report quality. They are not valid measures of design understanding, because four students reproduced a correct control-group description without being able to explain it. Her oral interviews are better evidence of understanding and worse evidence of written communication. Neither measure is wrong; they answer different questions.
Practical move: the task stays summative for the writing standards it actually assesses. She logs the interview results separately and uses them to plan - four students need a design-check conference, not a re-teach of report format. If her gradebook allows multiple evidence entries, she records the interview as a second, formative entry so the record shows both.
Why not average them: averaging a written score and an oral score into one number destroys the diagnostic information she needs and produces a number that measures neither skill.
What would change the call: if the rubric's design criterion explicitly required justification of the control group rather than description of it, then the four students' reports would not have earned those points, and the problem is rubric-task misalignment - fixed by revising the rubric, not by revising scores.
2. How should she handle the two students with IEPs whose written scores and oral explanations diverge, and what should she change in the task itself?
Read the worked answer
On the two students: their pattern - strong oral reasoning, weak written report - is exactly the profile an IEP accommodation is designed for, and it should have been anticipated at planning, not discovered at grading. She does not retroactively change the rubric. She checks the IEPs for whether a graphic organizer, scribing, or oral-defense option was specified and not provided. If an accommodation was missed, that is a procedural error and the score should be re-determined with the accommodation in place, documented. If no such accommodation exists, the data are evidence for the next IEP review, and in the meantime she records the written score and notes the oral evidence.
Task revision: the rubric rewarded a described design; the standard requires an explained one. She revises the design criterion to require a sentence justifying why the control group isolates the variable, and she adds a short oral-defense component - three minutes per student across two lab days, or a recorded option - so the task measures understanding of design rather than fluency of written description.
Validity check before reusing the task: she uses the interview protocol on a broader sample next time, because four of six is too high a rate of unexplained correct answers to treat as idiosyncratic. If the same pattern appears in another class, the task prompt is cueing a formulaic answer, and the fix is the prompt, not the students.
3. Describe how she should use the whole set of results to plan the next unit, and what evidence would tell her the change worked.
Read the worked answer
Pattern to act on: 14 of 27 reached proficient on the written product, but the interview sample suggests roughly two-thirds of those cannot explain the reasoning behind a correct answer. That is a scoring-validity problem, not a grading problem - the task is rewarding the production of a correct form.
Planning implications: (1) rewrite the design and conclusion criteria to demand justification, not description; (2) build in oral or written explanation checkpoints mid-unit, so she is not learning in week four that students cannot explain week two; (3) add a control-group comparison routine that students use in every lab, so justification becomes a habit rather than a task-specific performance; (4) plan the oral-defense component into the calendar and staffing rather than bolting it on at grading.
Evidence the change worked: during the next unit she samples five students who score 3 or 4 on written work and interviews them. If three or more can explain their control group, the revised criteria are doing their job. She also compares the rate of unexplained-correct responses across the two tasks; the target is a clear drop from four of six. If the rate holds, the cueing is in the prompt wording, and no amount of rubric rewriting will fix it.
What she tells her team: the pair of findings - written proficiency up, explanation unchanged - is a validity warning about the common task, and worth raising departmentally, since a rubric that cannot distinguish form from reasoning will produce the same misleading data in every classroom that uses it.