In the context of assessment, reliability is understood as the consistency with which an assessment performs its function. For example, a highly-reliable assessment of maths would be the expectation that an individual scored consistently well over multiple instances of the assessment; if the same child took the same assessment twice in the same day, the outcome would be broadly similar.

It is good practice in using the term reliability to follow it with the phrase ‘…for the purpose of X’. For instance, teachers and school leaders might talk about ‘the reliability of the Year 6 maths exam for the purpose of assessing pupils’ progress in maths during Year 6’.