The reliability coefficient that is approximately equal to the average of all the possible split-half...

The Kuder-Richardson coefficient (specifically KR-20 and KR-21) is used to estimate the internal consistency reliability of a test and is mathematically equivalent to the average of all possible split-half reliabilities. It is most often used with tests whose items are scored dichotomously (e.g., correct/incorrect). “Kuder-Richardson formulas provide an index of reliability that represents the average of all possible split-half coefficients for a test.”