Assessment and Evaluation of Learning
What this domain tests
Precision with assessment vocabulary, quality criteria, item analysis, and the design of assessment in Ghanaian SHS: 30% school-based assessment and 70% external examination for WASSCE.
Core distinctions
Measurement (numbers), assessment (gathering evidence), evaluation (judging against standards). Formative assessment happens during learning and feeds back into teaching; summative assessment comes at the end of a unit or term; diagnostic assessment precedes teaching to locate gaps; placement decides a level or class. Assessment for learning is formative in purpose; assessment of learning is summative.
Quality criteria
Validity (measures what it claims - content, construct, criterion), reliability (consistency - test-retest, inter-rater), fairness, practicability and washback. A test can be reliable without being valid; it cannot be valid without being reasonably reliable.
Item analysis
- Difficulty index: p = proportion answering correctly. Acceptable band about 0.3–0.7.
- Discrimination index: D = (correct in upper group − correct in lower group) ÷ group size. D ≥ 0.3 is good; 0–0.2 needs revision; negative means a faulty item or a miskeyed answer.
- Distractor analysis: a distractor chosen by many high scorers signals a problem with that option.
Worked example
Forty students sat an item; 25 chose the key. Of the top 10, 8 were correct; of the bottom 10, 3 were correct.
Reasoning: p = 25 ÷ 40 = 0.63, inside the acceptable band. D = (8 − 3) ÷ 10 = 0.50, strong discrimination. Verdict: keep the item. If D had been −0.2, check the key first, then the wording, then the distractor set.
Table of specifications
Build one by crossing content areas with Bloom levels and weighting by teaching time, so coverage is planned rather than accidental.
Ghanaian specifics
SBA is continuous, criterion-referenced against indicators and recorded; end-of-term examinations are summative; WASSCE reports grades A1–F9, with F9 a fail. Feedback must be specific, timely and actionable - 'good work' is not feedback.
Traps that cost marks
- Using 'evaluation' for a single quiz.
- Claiming a reliable test must be valid.
- Reading negative discrimination as a good sign.
- Describing formative assessment as unrecorded and therefore unimportant.
- Using norm-referenced language where the curriculum is criterion-referenced.