Using test scores to evaluate and hold school teachers accountable in New Mexico

Abstract

For this study, researchers critically reviewed documents pertaining to the highest profile of the 15 teacher evaluation lawsuits that occurred throughout the U.S. as pertaining to the use of student test scores to evaluate teachers. In New Mexico, teacher plaintiffs contested how they were being evaluated and held accountable using a homegrown value-added model (VAM) to hold them accountable for their students’ test scores. Researchers examined court documents using six key measurement concepts (i.e., reliability, validity [i.e., convergent-related evidence], potential for bias, fairness, transparency, and consequential validity) defined by the Standards for Educational and Psychological Testing and found evidence of issues within both the court documents as well as the statistical analyses researchers conducted on the first three measurement concepts (i.e., reliability, validity [i.e., convergent-related evidence], and potential for bias).