The Unsuitability of Goodman and Kruskal’s Lambda Measure of Association for Taxi Analysis of Multiple-Choice Question Difficulty Taxonomies
Abstract
TaxI analysis of published multiple-choice question bank difficulty taxonomies produces classification matrices relating measured or observed question difficulty to published difficulty level, i.e., the accuracy of thee published taxonomy. Where there is a preponderance of questions in one of the published categories, an anomaly in the Goodman and Kruskal lambda measure of association renders it unsuitable for TaxI classification matrices The present study explains that anomaly and illustrates its unsuitability are explained.
Downloads
Published
2022-03-26
Issue
Section
Innovations and Future Directions in Education
License
This work is licensed under a Creative Commons Attribution-NoDerivatives 4.0 International License.