Principles of Test Theories
Based on a tremendous increase in the development of psychometric theories in the past decade -- ranging from techniques for criterion-referenced testing to behavioral assessment, generalizability, and item response theory -- this book offers a summary of core issues. In so doing, it provides a comprehensive survey of reliability, validity, and item analysis from the perspectives of classical true-score model, generalizability theory, item response theory, criterion-referenced testing, and behavioral assessment. Related theoretical issues such as item bias, equating, and cut-score determination are also discussed. This is an excellent text for courses in statistics, research methods, behavioral medicine and cognitive science as well as educational, school, experimental, counseling/social, clinical, developmental, and personality psychology.
What people are saying - Write a review
We haven't found any reviews in the usual places.
Other editions - View all
3-parameter ability levels applied approach assess Bayesian modal estimation behavior biased chapter classical theory common conceptual construct validity content validity correlation criterion measure criterion-referenced testing criterion-related validity crossed design cut score D-study scenario derivative discrimination discussed Equation examinees example facet G-coefficient generalizability theory guessing Hambleton Hence indicate item correctly item parameters item response theory item score ith item judges likelihood function linear transformation math aptitude mathematical matrix maximum likelihood estimation measurement error measurement procedure methods nonmasters norm-referenced number of items object of measurement observation session observed score obtained ogive function p-value parallel tests assumptions Pearson's prior distribution probability proportion psychometric random Rasch model raters raw score reliability and validity reliability coefficient sample scale score variance Simpson's paradox single-facet slope Specifically standard error statistical subjects Suen test score total score true score true variance values variance components Venn Diagram zero