By design, randomized experiments (XPs) rule out bias from confounded selection of participants into conditions. Quasi-experiments (QEs) are often considered second-best because they do not share this benefit. However, when results from XPs are used to generalize causal impacts, the benefit from unconfounded selection into conditions may be offset…

This report describes the purposes of benchmark assessments and provides recommendations for selecting and using benchmark assessments--addressing validity, alignment, reliability, fairness and bias and accessibility, instructional sensitivity, utility, and reporting issues. We also present recommendations on building capacity to support schools'…

Guidelines for test developers and users are provided to insure that construction and selection of test instruments are conducted fairly. In addition to development and selection, issues of interpretation of scores; provision of information to test takers; and prevention of test bias based on race, gender, or ethnicity are addressed. Twenty-one…

This paper discusses whether or not revisions of the Scholastic Aptitude Test (SAT) and the American College Test (ACT) have created such significant differences between the two tests that a student could conceivably score significantly higher on one than the other. The SAT has been revised to meet the needs of an increasingly diverse student…

Because of concerns about the validity of tests used in education and employment, a consultation on June 16, 1989, focused on tests of ability, achievement, and other skills. Invited experts were asked to address a set of issues common to both education and employment testing, primarily related to test construction procedures and how to establish…