ResearchPosted on rand.org 2004Published in: Advancing Educational Productivity: Policy Implications From National Databases / Edited By Susan J. Paik (Greenwich, CT: Information Age Publishing, 2004) p. 75-201
Large-scale, standardized achievement tests are used for a number of research and policy purposes. This chapter describes one method for improving the utility of information from large-scale tests. It describes a series of studies that examined the National Education Longitudinal Study of 1988 (NELS:88) mathematics and science tests and attempted to identify subscores that could be useful for understanding correlates of student achievement and patterns of group differences in achievement. Subscores for the math and science tests were identified through full information factor analyses and validated through interviews with a small sample of high school students. Investigations of relationships among performance on the subscores, student characteristics, and educational experiences revealed ways in which inferences about influences on achievement may vary as a function of the achievement measure used in the analysis. The chapter concludes with some recommendations for practitioners and policymakers who use data from standardized achievement tests.
This publication is part of the RAND external publication series. Many RAND studies are published in peer-reviewed scholarly journals, as chapters in commercial books, or as documents published by other organizations.
RAND is a nonprofit institution that helps improve policy and decisionmaking through research and analysis. RAND's publications do not necessarily reflect the opinions of its research clients and sponsors.
