Assessment, Measurement and Evaluation of Learning Outcomes
Three terms that must not blur
Assessment is the process of gathering evidence of learning. Measurement is assigning numbers or categories to that evidence. Evaluation is judging worth or merit against criteria - deciding what the evidence means and what to do next. A quiz produces measurement; interpreting the pattern of results against the CLO and deciding on remediation is evaluation.
Working vocabulary
Formative assessment is used to inform learning while it is happening; summative assessment certifies achievement at the end. Diagnostic assessment establishes starting points. OBE is criterion-referenced: performance is compared with the standard, not with peers. Quality criteria for any instrument include validity (the evidence supports the inference intended), reliability (consistency across markers, occasions and items), fairness, practicality and authenticity. A table of specifications or test blueprint distributes items across CLOs and cognitive levels so that no single outcome or level dominates. Rubrics may be analytic (criteria scored separately) or holistic (single overall judgement).
Traps
- Using assessment and evaluation interchangeably in a question stem.
- Confusing validity with reliability. A bathroom scale that reads two kilograms heavy every time is highly reliable and invalid; a scale that gives a different reading each time is unreliable and cannot be valid.
- Supposing formative assessment must be ungraded, or that summative feedback cannot feed forward.
- Building tests without a blueprint and then claiming alignment.
Rule of thumb
Ask: what inference do I want to make, and does this evidence support it? Then ask: would a second marker, on a second day, reach the same conclusion?