Introducing a Computer-Adaptive Testing System to a Small School District

Abstract

A computer-adaptive test (CAT) is a relatively new type of technology in which a computer program “intelligently” selects and presents questions to examinees according to an evolving estimate of achievement and a prescribed test plan. A well written CAT can be expected to efficiently produce student achievement estimates that are more accurate and more meaningful than a typical teacher-generated paper and pencil (P&P) test with a similar number of questions. Although this method of testing sounds good in theory, many schools and districts are waiting for positive examples of practical applications and observable benefits before adopting a CAT. This chapter begins by describing the essential elements of meaningful measurement in education and the features of a typical CAT. Next, we describe the Measures of Academic Progress (MAP) system of the Northwest Evaluation Association (NWEA; 2004) and observations made during the introduction of this system into a small semirural school district. Finally, as independent observers, we provide a set of recommendations to help guide other districts as they consider the potentials of implementing a CAT system to guide instruction within their schools.