Scoring Performance Assessments Based on Judgements focuses on the applications of Generalizability Theory to Performance Assessment. The author developed an objective method to examine the dependability of the scoring of large-scale performance assessments by humans. This book presents principles in identifying common missing data patterns inherent in assessment designs. These principles can be used to guide the design of assessment procedures including those used in large-scale testing programs, observations, and structured interviews
CONTENT
1 Introduction -- 2 Literature in Variance Component Estimations, Large-Scale Assessments, and Missing Data -- 3 Methodology -- 4 Validations of the Subdividing Method -- 5 Conclusions, Discussions, and Future Directions -- 6 Computer Programs and Rater Design Diagnosis -- Appendices -- References -- Author Index
