Open Collections

Test validation and Bayesian statistical frameworks to estimate the magnitude and corresponding uncertainty of washback effects of test preparation

This monograph describes a framework for test validation that synthesizes construct theories and argument-based approaches bringing washback (also described as consequences or impact) into the foreground of the validation practices. This framework is well suited for tests immersed mainly in a construct theory and whose validation practices have allied with test validity theory as codified in the Standards for Educational and Psychological Testing, where washback has been in the background. A Bayesian method is described and used in a washback study of test preparation. The monograph closes with a critical review of washback studies for nationally mandated and regulated tests. [A revised file was uploaded on 08-17-2023 with minor revisions to Section 5.]

Text

UBC Psychometric Research Series

Vancouver : University of British Columbia Library

10.14288/1.0435197

Education, Faculty of; Educational and Counselling Psychology, and Special Education (ECPS), Department of

Reviewed

http://creativecommons.org/licenses/by-nc-nd/4.0/