William C. Ward - Böcker

This book brings together psychometric, cognitive science, policy, and content domain perspectives on new approaches to educational assessment -- in particular, constructed response, performance testing, and portfolio assessment. These new assessment approaches -- a full range of alternatives to traditional multiple-choice tests -- are useful in all types of large-scale testing programs, including educational admissions, school accountability, and placement. This book's multi-disciplinary perspective identifies the potential advantages and pitfalls of these new assessment forms, as well as the critical research questions that must be addressed if these assessment methods are to benefit education.

Although computer-based tests (CBT) have been administered for many years, improvements in the speed and power of computers coupled with reductions in their cost have made large-scale computer delivery of tests increasingly feasible. CBT is now a common form of test delivery for licensure, certification, and admissions tests. Many large-scale, high-stakes testing programs have introduced CBT either as an option or as the sole means of test delivery. Although this movement to CBT has, to a great extent, been successful, it has not been without problems. Advances in psychometrics are required to ensure that those who rely on test results can have at least the same confidence in CBTs as they have in traditional forms of assessment. This volume stems from an ETS-sponsored colloquium in which more than 200 measurement professionals from eight countries and 29 states convened to assess the current and future status of CBT. The formal agenda for the colloquium was divided into three major segments: Test Models, Test Administration, and Test Analysis and Scoring. Each segment consisted of several presentations followed by comments from noted psychometricians and a break-out session in which presenters and discussants identified important issues and established priorities for a CBT research agenda. This volume contains the papers presented at the colloquium, the discussant remarks based on those papers, and the research agenda that was generated from the break-out sessions. Computer-Based Testing: Building the Foundation for Future Assessments is must reading for professionals, scholars, and advanced students working in the testing field, as well as people in the information technology field who have an interest in testing.

This book brings together psychometric, cognitive science, policy, and content domain perspectives on new approaches to educational assessment -- in particular, constructed response, performance testing, and portfolio assessment. These new assessment approaches -- a full range of alternatives to traditional multiple-choice tests -- are useful in all types of large-scale testing programs, including educational admissions, school accountability, and placement. This book's multi-disciplinary perspective identifies the potential advantages and pitfalls of these new assessment forms, as well as the critical research questions that must be addressed if these assessment methods are to benefit education.

Although computer-based tests (CBT) have been administered for many years, improvements in the speed and power of computers coupled with reductions in their cost have made large-scale computer delivery of tests increasingly feasible. CBT is now a common form of test delivery for licensure, certification, and admissions tests. Many large-scale, high-stakes testing programs have introduced CBT either as an option or as the sole means of test delivery. Although this movement to CBT has, to a great extent, been successful, it has not been without problems. Advances in psychometrics are required to ensure that those who rely on test results can have at least the same confidence in CBTs as they have in traditional forms of assessment. This volume stems from an ETS-sponsored colloquium in which more than 200 measurement professionals from eight countries and 29 states convened to assess the current and future status of CBT. The formal agenda for the colloquium was divided into three major segments: Test Models, Test Administration, and Test Analysis and Scoring. Each segment consisted of several presentations followed by comments from noted psychometricians and a break-out session in which presenters and discussants identified important issues and established priorities for a CBT research agenda. This volume contains the papers presented at the colloquium, the discussant remarks based on those papers, and the research agenda that was generated from the break-out sessions. Computer-Based Testing: Building the Foundation for Future Assessments is must reading for professionals, scholars, and advanced students working in the testing field, as well as people in the information technology field who have an interest in testing.