Assessment and Data-Driven Decision Making
Assessment and Data-Driven Decision Making (GTET 2) tests whether you can select, interpret, and act on assessment evidence responsibly. Questions cover formative and summative purposes, reliability and validity, item statistics, and data-team decisions.
Core rules
- Formative assessment occurs during learning and informs instruction; summative assessment reports achievement at the end.
- Reliability is consistency; validity is measuring the intended construct. A test can be reliable but invalid.
- Criterion-referenced scores compare performance to a standard; norm-referenced scores compare to a group. Percentile rank is not percent correct.
- Item difficulty p is the proportion correct. Discrimination D shows how well high scorers outperform low scorers. A very easy item with near-zero D provides little information.
- Data protocols: examine patterns by standard and subgroup, form a hypothesis, choose one instructional change, then recheck.
Worked example
Item analysis shows p = .92 and D = .05 on a fractions item. Interpretation: almost everyone answered correctly, and strong and weak students performed alike, so the item yields little information about learning. Action: keep it as a warm-up check but replace it on the summative form with an item requiring fraction comparison. Reasoning: high ease plus low discrimination means the item is too easy and does not separate students who have the skill from those who do not.
Traps that cost marks
- Reading a percentile rank as percent correct.
- Claiming a test is valid because it is reliable.
- Reteaching an entire unit after one missed item.
- Making placement decisions from a single score.
- Ignoring subgroup patterns that reveal an access problem.