Open AccessPASS‐FAIL RELIABILITY FOR TESTS WITH CUT SCORES: A SIMPLIFIED METHOD
Author(s) -
Breyer F. Jay,
Lewis Charles
Publication year - 1994
Publication title -
ets research report series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.235
H-Index - 5
ISSN - 2330-8516
DOI - 10.1002/j.2333-8504.1994.tb01612.x
Subject(s) - reliability (semiconductor) , statistics , test (biology) , consistency (knowledge bases) , mathematics , index (typography) , binary number , internal consistency , cut off , test score , arithmetic , standardized test , psychometrics , computer science , discrete mathematics , paleontology , power (physics) , physics , quantum mechanics , world wide web , biology
A single‐administration classification reliability index is described that estimates the probability of consistently classifying examinees to mastery or nonmastery states as if those examinees had been tested with two alternate forms. The procedure is applicable to any test used for classification purposes, subdividing that test into two half‐tests, each with a cut score, where the sum of the two half‐test cut scores is equal to the cut score for the total test. The application of this pass‐fail consistency index to binary scored objective tests, nonbinary scored performance tests, and tests containing both binary and nonbinary scored questions is presented. A calculation example is provided together with look‐up tables.