Assessment
Assessment
This domain tests how teachers select, design, interpret, and use assessments. Items cover formative and summative purposes, validity, reliability, fairness, feedback, grading, and accommodations.
Core rules
- Purpose first: formative assessment informs instruction; summative assessment documents learning at a point in time. A graded quiz can still be formative if the teacher reteaches from it.
- Validity means the assessment measures the intended construct; reliability means consistency across occasions, raters, or items. A scale can be reliable but invalid for measuring writing quality.
- Align methods to objectives: selected response for knowledge, performance tasks and rubrics for complex skills, portfolios for growth over time.
- Feedback should be specific, timely, and actionable, tied to criteria; grades should not mix behavior and achievement unless the standard includes it.
- Accommodations change access without changing the construct (extended time, read-aloud when reading is not the target); modifications change the construct or expectation.
- Avoid bias: review language, contexts, and examples for cultural familiarity; use a table of specifications to sample content fairly.
Worked example
After a fractions unit, 14 of 25 students miss items that require regrouping when subtracting mixed numbers. The teacher does not curve the quiz. She sorts the exit tickets, reteaches with fraction strips to a small group, and gives others a problem-solving extension. Reasoning: the pattern is a shared misconception, so the data should change instruction; a curve would hide the gap and make the summative grade uninformative.
Traps
Do not confuse validity with reliability; do not treat accommodations as cheating; do not use only one assessment method for a complex objective; do not average practice work into a final grade without checking what the grade represents.