In language assessment, authenticity refers to the degree of correspondence between the characteristics of a test task and the features of real-world language use tasks. Defined formally by Bachman & Palmer (1996) as a test quality alongside validity, reliability, and practicality.
Authenticity is the match between test task characteristics and TLU domain task characteristics. The TLU domain is any specific real-world setting outside the test that requires the candidate to perform language use tasks, e.g., academic lectures, workplace emails, airport announcements.
Task characteristics compared across test and TLU domain include:
Bachman & Palmer (1996) split authenticity into two complementary qualities. Situational authenticity is the surface match between test-task and TLU-task characteristics: the test looks like the real-world task. Interactional authenticity is the engagement match: the task draws on the candidate's language ability, topical knowledge, and affective resources in the way real language use does. Both are necessary; either alone leaves room for tests that pass surface review but fail to measure what they claim.
A test with high authenticity:
A test with low authenticity may still be valid if it measures the right construct, but it risks negative washback and low stakeholder acceptance.
Perfect authenticity is impossible; tests necessarily simplify, standardise, and constrain. The goal is sufficient correspondence on the task characteristics that matter most for the construct being measured. Bachman & Palmer's framework provides a systematic checklist for evaluating and improving the match.