Descripciones mejoradas de Syndetics:

Using familiar concepts from classical measurement methods and basic statistics, Hambleton and colleagues introduce the basics of item response theory (IRT) and explain the application of IRT methods to problems in test construction, identification of potentially biased test items, test equating, and computerized-adaptive testing. The book also includes a thorough discussion of alternative procedures for estimating IRT parameters, such as maximum likelihood estimation, marginal maximum likelihood estimation, and Bayesian estimation in such a way that the reader does not need any knowledge of calculus to follow these explanations. Including step-by-step numerical examples throughout, the book concludes with an exploration of new directions in IRT research and development.

Tabla de contenidos provista por Syndetics

• Background
• Concepts, Models, and Features
• Ability and Item Parameter Estimation
• Assessment of Model-Data Fit
• The Ability Scale
• Item and Test Information and Efficiency Functions
• Test Construction
• Identification of Potentially Biased Test Items
• Test Score Equating
• Computerized Adaptive Testing
• Future Directions of Item Response Theory

Reseñas proporcionadas por Syndetics

CHOICE Review

Item response theory (IRT) is increasingly replacing classical measurement theory for the various psychometric procedures required by large-scale test development efforts. Although the underlying fundamentals of IRT are straightforward, the sophisticated computations necessary for IRT applications require mathematical skills that many students do not possess. This book is intended for individuals with no background in IRT, and demands only modest statistical skills by the user; mathematical derivations have been avoided. Hambleton is recognized as a leader in IRT applications with numerous papers, articles, and books on the subject. Both the purpose and content of this book are similar to Frank B. Baker's The Basics of Item Response Theory (CH, Feb'86). Hambleton and his colleagues, however, have provided a more detailed and current treatment. Their coverage and evaluation of various parameter-estimation models and computational software are more complete, model fit to the data is stressed, and there are chapters on detecting item bias, computerized adaptive testing, and future directions for IRT. For the novice, however, this book lacks the visual and interactive appeal provided by the integrated microcomputer disk that accompanies Baker's book. Appropriate for advanced undergraduates, graduate students, and faculty.-P. G. Romine, Tennessee Technological University