Teacher Evaluation as Trojan Horse: The Case for Teacher-Developed Assessments

摘要

In his focus article "How Is Testing Supposed to Improve Schooling?" Ed Haertel distinguishes between seven uses of educational tests as a function of the intended action and what or who will be influenced by the intended action. He then applies Mike Kane's interpretive argument approach (Kane, 2006) as a basis for speculating about the validity of intended actions. A recurring theme in the article is that validating the premise that testing has improved schooling will require much more attention to issues of score extrapolations and decision making. In this comment I expand upon the section of Haertel's article that focused on testing for accountability. More specifically, I take up the management role of testing within the context of recent policy shifts toward high-stakes teacher evaluation.