Generalizability Theory
2000; Elsevier BV; Linguagem: Inglês
10.1016/b978-012691360-6/50019-7
Autores Tópico(s)Advanced Statistical Modeling Techniques
ResumoGeneralizability theory is a random sampling theory for examining the dependability of measurement procedures that has been heralded by many psychometricians as "the most broadly defined psychometric model currently in existence." This chapter provides a nontechnical overview of generalizability theory and illustrates its use as a comprehensive method for designing, assessing, and improving the dependability of measurement procedures. Throughout the chapter, example cases and situations are explained in detail to enable the reader verify calculations and replicate the results. In addition, references are made to computer packages. The chapter also enables readers to use generalizability theory procedures in their own research. The chapter begins with an overview of the fundamentals of generalizability theory, with particular focus on the simple one-facet model, which is the most common measurement procedure used. The basic concepts extended to multifaceted and multivariate measurement designs, and some new extensions to the basic generalizability theory approach are introduced. The chapter concludes by providing an overview of computer programs that can be used to conduct generalizability analyses.
Referência(s)