Premium
Bias and Bias Correction Method for Nonproportional Abilities Requirement (NPAR) Tests
Author(s) -
Ip Edward H.,
Strachan Tyler,
Fu Yanyan,
Lay Alexandra,
Willse John T.,
Chen ShyhHuei,
Rutkowski Leslie,
Ackerman Terry
Publication year - 2019
Publication title -
journal of educational measurement
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.917
H-Index - 47
eISSN - 1745-3984
pISSN - 0022-0655
DOI - 10.1111/jedm.12204
Subject(s) - interpretability , test (biology) , scope (computer science) , set (abstract data type) , computer science , cognition , psychology , cognitive psychology , econometrics , artificial intelligence , mathematics , paleontology , neuroscience , biology , programming language
Test items must often be broad in scope to be ecologically valid. It is therefore almost inevitable that secondary dimensions are introduced into a test during test development. A cognitive test may require one or more abilities besides the primary ability to correctly respond to an item, in which case a unidimensional test score overestimates the primary ability and creates interpretability problems. In this article, we demonstrate the nonproportional abilities requirement, a phenomenon with which secondary abilities are more required for difficult items. A novel and practical method for correcting bias in the primary ability is proposed and illustrated using a real data set from an international assessment. Simulation data are also used to evaluate the performance of the method.