Estimating Latent Ability from the Nondestructive Test Results When the Test Items Difficulties Are Unknown Beforehand

• Activity Number: 612
• Type: Contributed
• Date/Time: Wednesday, August 12, 2015 : 2:00 PM to 3:50 PM
• Sponsor: Quality and Productivity Section
• Author(s): Emil Bashkansky* and Vladimir Turetsky
• Companies: ORT Braude College and ORT Braude College
• Keywords: Maximal Likelihood ; Latent ability ; Item response ; Testing ; Statistical inference ; Test item difficulty

Abstract

A new approach to evaluation of binary test results when checking a one-dimensional ability is proposed. We consider the case where a qualitatively homogeneous population of objects is tested by a set of non-destructive test items having different, but unknown beforehand levels of difficulty, and we need to evaluate both the intrinsic abilities of these objects and the levels of difficulty of the test items. We assume that the same scale invariant item response model applies to all members of the tested population of objects under study (OUTs). In the context of the paper, OUT can mean an electronic component, examinee, program unit or material under test, etc. An algorithm for solving the above mentioned problem, applicable for engineering testing, is proposed. It combines several already developed methods, such as item response theory, maximum likelihood estimation, method of flow redistribution and others. This combination allows building of an acceptable logical/numerical scheme for evaluation of testing results. It includes some stages converging to a solution determining levels of difficulty, abilities and distribution of these abilities among the tested population.