Toward the Peaceful Coexistence of Test Developers, Policymakers, and Teachers in an Era of Accountability

Abstract

The authors describe a potential disconnect between research and practice in literacy assessment and instruction. They organize discussion around professionally recognized standards for the evaluation of educational assessments and assessment practices. These standards address both technical aspects of tests (e.g., validity; reliability; standardized administration, scoring, and reporting; scales; norms; and score comparability) as well as more pragmatic concerns about the use of tests (e.g., fairness in testing, testing individuals from diverse linguistic backgrounds). In particular, the article focuses on the tension that has emerged in an era of accountability between demands for scientifically rigorous assessment in the area of early literacy and the practical day-to-day needs of teachers who must put assessment data to use in designing and delivering instruction. Increasing demands on teachers' implementation of scientifically based reading assessment and instruction often result in redundancy in testing and loss of instructional time, and an overreliance on externally imposed measures that may sacrifice validity for reliability, while providing little useful information that informs instruction. The authors argue that assessments can be technically sound in ways that preserve the theoretical integrity of reading development and instruction and provide the flexibility and instructional transparency that teachers need.