Building a Validity Argument for a Listening Test of Academic Proficiency

Over the years, various approaches to validation have emerged in psychological and educational assessment research, which can be classified into traditional approaches and modern approaches. Traditional approaches view validity as a multicomponential concept including, for example, content, construct, and predictive validity, while modern approaches conceptualize it as a unitary concept evaluated through argumentation. Drawing on the modern approach, this book builds a validity...
More Description

Over the years, various approaches to validation have emerged in psychological and educational assessment research, which can be classified into traditional approaches and modern approaches. Traditional approaches view validity as a multicomponential concept including, for example, content, construct, and predictive validity, while modern approaches conceptualize it as a unitary concept evaluated through argumentation. Drawing on the modern approach, this book builds a validity argument for an International English Language Testing System (IELTS) listening test sample. The book provides some insights into the listening sub-skills that the test engages, the psychometric dimensionality of the test, variables that predict item difficulty parameters, bias across age, nationality, test experience, and gender, as well as predictive-referenced evidence of validity. A variety of techniques including the Rasch model and structural equation modelling are used to answer the research questions and to build a validity argument framework; this argument organizes the thematically related findings into a coherent treatment of the validity of the listening test. The book presents the first treatment of validity argument and related analytical tools in one volume and maps the psychometric/statistical analysis tools onto the validity argument framework. It also provides an extensive literature review of listening comprehension, validation, and psychometric modeling and proposes both methods for developing and validating self-assessment instruments and novel approaches to improving the quality of language assessments.