| Definition | This is about how well an assessment actually measures what it is intended to measure and how precisely an assessment measures a pupil's level of ability, depth of knowledge, mastery of skill, or whatever the test or teacher's rating scale is intended to measure. The process of assessment rarely, if ever, results in exact measures of learning, for a variety of reasons. A single score from a single test used on a single occasion is unlikely to be a precise measure of a pupil's level of knowledge, ability or skill. Test reliability can be increased by increasing the length of the test or narrowing the focus – this, though, risks compromising content validity and curriculum relevance. Standardised tests are designed to be highly reliable technically, but they are not always able to fully reflect the taught curriculum. |
|---|
