Which reliability measure checks for homogeneity through different half-splits of a test?

The measure that checks for homogeneity through different half-splits of a test is split-half reliability. This approach involves dividing a test into two halves and then determining the correlation between the scores on these halves. The underlying principle is that if the two halves yield similar results, the test is considered reliable, indicating that the items on the test measure a consistent construct.

This type of reliability is particularly useful because it provides a straightforward method to assess internal consistency without requiring a larger sample size or additional testing occasions. By checking the homogeneity of the items, researchers can ascertain whether the test is measuring a single latent trait or construct effectively. This method is foundational in both educational and psychological assessments, ensuring that tests are stable and reliable across different sections.

In contrast, other measures such as cronbach's alpha focus more specifically on the internal consistency of all items and involve a more complex calculation than simply splitting the test in half. Test-retest reliability assesses the stability of test scores over time by comparing results from the same participants at two different points in time, while interrater reliability measures the extent to which different raters or judges agree in their assessments. These approaches serve different purposes in reliability testing and do not assess the concept of homogeneity through split-half comparisons.