ABSTRACT
The public credibility and legal defensibility of the accountability system, as it applies to all educators, including those in the “non-tested subjects and grades”, will depend on the technical quality of the assessments that are employed, how the different indicators are combined to yield an overall rating, and the consequences tied to each rating. Optimization is the attempt to attain the goals while respecting the constraints. The quality of the assessment is then judged by the degree to which optimization has been achieved. Similarly, a transition to computer-based delivery can substantially expand the range of item types that can be employed and, through adaptive testing, minimize differences in measurement error across the score scale. It can also reduce scoring costs through the use of expert systems to replace human graders. The chapter also presents an overview on the key concepts discussed in this book.
