Aryadoust V., Raquel M. (Eds.) Quantitative Data Analysis for Language Assessment. Volume II: Advanced Methods

New York, USA: Routledge, 2020. — 260 p. — (Routledge Research in Language Education). — ISBN 978-1-138-73314-5.Количественные Данные Оценки ЯзыкаQuantitative Data Analysis for Language Assessment Volume II: Advanced Methods emonstrates advanced quantitative techniques for language assessment. The volume takes an interdisciplinary approach and taps into expertise from language assessment, data mining, and psychometrics. The techniques covered include Structural Equation Modeling, Data Mining, Multidimensional Psychometrics and Multilevel Data Analysis.Volume II is distinct among available books in language assessment, as it engages the readers in both theory and application of the methods and introduces relevant techniques for theory construction and validation. This book is highly recommended to graduate students and researchers who are searching for innovative and rigorous approaches and methods to achieve excellence in their dissertations and research. It is also a valuable source for academics who teach quantitative approaches in language assessment and data analysis courses.Advanced item response theory (IRT) models in language assessment
Applying the mixed Rasch model in assessing reading comprehension
Multidimensional Rasch models in first language listening tests
The log-linear cognitive diagnosis modeling (LCDM) in second language listening assessment
Application of a hierarchical diagnostic classification model in assessing reading comprehension
Advanced statistical methods in language assessment
Structural equation modeling to predict performance in English proficiency tests
Student growth percentiles in the formative assessment of English language proficiency
Multilevel modeling to examine sources of variability in second language test scores
Longitudinal multilevel modeling to examine changes in second language test scores
Nature-inspired data-mining methods in language assessment
Classification and regression trees in predicting listening item difficulty
Evolutionary algorithm-based symbolic regression to determine the relationship of reading and lexicogrammatical knowledge