Assessment and Data-Informed Decision Making
What this domain tests
Assessment and Data-Informed Decision Making tests how you choose, design, interpret, and respond to evidence of student literacy learning. Prompts may ask you to evaluate an assessment, analyze a data set, give feedback, or adjust instruction.
Core rules and concepts
- Formative assessment is ongoing evidence used to adapt teaching; summative assessment evaluates achievement at a point in time. The distinction is purpose, not format.
- Validity: does the assessment measure the intended construct? Reliability: consistency across raters, tasks, and occasions. A reliable rubric can still be invalid for the wrong construct.
- Analytic rubrics give trait-level information (claim, evidence, organization, language) that can drive mini-lessons; holistic scores are efficient but less diagnostic.
- Feedback should be timely, specific, and actionable, focused on the task and process rather than praise or ego.
- Data triangulation: no high-stakes decision should rest on one score. Consider multiple measures, standard error of measurement, and opportunity-to-learn.
Worked example
A teacher scores 28 argument essays with a four-trait rubric. The class average is strong on claim but weak on evidence commentary: students quote sources without explaining relevance. Rather than reteaching the whole essay, the teacher plans a two-day lesson using annotated mentor paragraphs and a revision checklist focused on the warrant. Reasoning: the trait data localizes the problem; the intervention targets the reasoning gap, and a quick re-score of one paragraph checks whether the lesson transferred.
Traps that cost marks
- Using a single test score to place or label a student.
- Confusing reliability with validity.
- Giving feedback that is only praise or a grade.
- Teaching to the format of a standardized test at the expense of the construct.
- Ignoring bias in items, texts, and scoring norms.