 |
 |
 |
 |
 |
 |
|||||
|
Fundamentals of Item Response Theory
Five-day Course • June 15–19, 2009• Lawrence, Kansas Sponsored by the Quantitative Psychology Training
Program of the Department of Psychology at the ObjectivesItem response theory (IRT) has become increasingly popular in the past few decades in a wide variety of fields. Whether it is being used to create computerized adaptive tests for health related quality of life, allow for seamless equating in K-12 educational testing, or to create more precise measurement instruments in psychological testing, IRT has become an essential feature of the modern measurement landscape. This course is roughly divided into three sections. The first section will provide a brief overview of classical test theory (CTT), focusing on the assumptions typically made in CTT and how IRT allows us to relax those assumptions. The second portion of the class will cover the basic IRT models (2- & 3-parameter logistic, graded response model, etc.), with the emphasis on understanding the parameters and the basic concepts involved. This second part will also focus on software and how to estimate these models in practice. Finally, we will briefly discuss several advanced topics to provide an overview of the wide variety of potential uses for IRT. Topics in this section will include: measurement invariance, computerized adaptive testing, linking, equating, and multidimensional IRT models. AudienceThis course is structured for someone with at least a working knowledge of statistics on par with what is typically available in an introductory statistics sequence. Working examples will be drawn from a wide array of fields including education, psychology, and health outcomes. The course will be beneficial for individuals wishing to learn about the concepts of IRT as well as to those who want to be able to conduct their own IRT analyses. InstructorMichael C. Edwards, PhD, is an assistant professor of quantitative psychology at The Ohio State University. He received his PhD in 2005 from the L. L. Thurstone Psychometric Laboratory at the University of North Carolina at Chapel Hill under the direction of Dr. David Thissen. His dissertation, A Markov chain Monte Carlo approach to Confirmatory Item Factor Analysis, was awarded the 2006 Psychometric Society Dissertation Award. His research interests include multidimensional IRT, computerized adaptive testing, and general measurement problems in the social sciences. He consults for a wide variety of organizations ranging from the American Physical Therapy Association to the College Board. Software and Computer SupportParticipants in the course are expected to bring their own laptops. The primary software for the course is MULTILOG, which can be obtained at http://www.e-academy.com/irt. Note that this is a free (but fully functional) 15-day trial version. Individuals interested in purchasing MULTILOG should visit http://www.ssicentral.com/ordering/index.html. Currently an academic-priced version of MULITLOG can be obtained for $250. All files (data, syntax, output, etc.) used in the course will be made available. Two other software packages we will use can be obtained free of charge. IRTLRDIF can be found at http://www.unc.edu/~dthissen/dl.html. CEFA can be found at http://faculty.psy.ohio-state.edu/browne/software.php. SyllabusCOMMING SOON!Contact InformationAny questions regarding course content should be directed to Michael Edwards at edwards.134@osu.edu.
|
Please send questions or comments concerning this Web site to Donald Gay.
