Answers are grounded in this paper and cite their sources.
Reliability is important in the academic judgements that higher education institutions pass to employers, graduate schools and other stakeholders. However, there are well-known limits to the levels of reliability that can be reached at reasonable cost and without sacrificing validity. It is argued that a systemic approach to the assessment of student learning shows how resources may be freed up to invest in securing more reliable assessments where they are desirable and might reasonably be had.
No takes yet. Share an insight, caveat, or question.
Peter T. Knight (2000) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: