How is a split-half reliability measurement performed?


Quick Answer

A half-test reliability measurement can be performed by splitting the test questions into two groups, such as odd-numbered questions and even-numbered questions, and then comparing the results. The scores obtained from the two groups are correlated, and if the scores obtained from both halves demonstrate a high positive reliability coefficient, or correlation, it can generally be assumed that the test is reliable.

Continue Reading

Full Answer

Reliability in a testing procedure refers to the degree of consistency that can be expected when the test is administered under varying circumstances. In addition to the split-half reliability method of measuring test consistency, other measurement procedures include the inter-observer, test-retest and parallel-forms methods.

Before relying upon a test to asses important research hypotheses and questions, researchers will subject their tests to one or more reliability measurement methods to help ensure the accuracy and consistency of the results.

Learn more about Statistics

Related Questions