What does internal consistency measure in a test?

Internal consistency measures the degree to which items within a test are correlated with one another, indicating how well they assess the same underlying construct or ability. This correlation among items suggests that they are consistently measuring the same concept. For a test to have strong internal consistency, it means that participants' responses to different items are reliably similar, which reinforces the test’s validity in measuring what it intends to measure.

In the context of test construction and evaluation, internal consistency is often assessed using statistical methods like Cronbach's alpha, which quantifies the extent to which items correlate. If a test shows high internal consistency, it implies that the items are cohesively working together to provide a reliable measure of the construct of interest.

Other aspects such as overall test length, item difficulty, or response frequencies do not directly reflect the coherence among the test items. Therefore, they do not pertain to internal consistency as effectively as the intercorrelations among the test items do.