Item Response Theory / Rasch Modeling Using Stata 14
In this webinar, Alan Acock, Ph.D., shares information from a new chapter on Item Response Theory (IRT) from the 5th edition of his book, A Gentle Introduction to Stata. IRT is widely used in computer adaptive testing, such as the SAT test, and offers many advantages over traditional summated scales such as Likert-type scales.
Stata 14 makes it very easy to estimate a variety of IRT models. For binary items there are 1-parameter (Rasch) models, 2-parameter IRT models, and 3-parameter models. For Likert-type items there are Graded Response Models (extension of 2-parameter IRT and Partial Credit Models (extension of 1-parameter Rasch model). Dr. Acock also introduces Differential Item Functioning theory.
(While there are also generalized partial credit models, rating scale models, nominal response models, and hybrid models that are easy extensions, this set of models will not be covered in the webinar.)
This webinar is intended for family researchers, graduate students, and advanced undergraduates. No experience with Stata, Rasch modeling, or IRT is needed.
Approved for 2 CFLE contact hours of continuing education credit.
|NCFR student members||$30|
|Individual use||NCFR professional members||$50|
|Classroom use||NCFR members||$115|
|Department use||NCFR members||$185|
Not an NCFR member? Join NCFR and save. You can immediately qualify for the member registration rate for these webinars by first purchasing an NCFR membership and then purchasing your webinar registration.
- use Item Response Theory (IRT) to develop new measures of attitudes, beliefs, and values
- use IRT evaluate the psychometric properties of existing measures
- compare IRT measures to summated type scales (e.g., Likert)
- run alternative IRT Models: 1 parameter models (Rasch models), 2 parameter models (IRT), and Graded Reponse Model — IRT for Likert Type Scales
- use drop-down menus
This is a hands-on webinar, and you'll need Stata 14 to participate; older versions of Stata will not work.
Register by Sept. 29 to receive a free 90-day trial version of Stata 14. You'll receive the free license, the activation code, and installation instructions one week prior to the webinar for use during the webinar, and you should install Stata 14 prior to the webinar so you can run the commands yourself.
You also will receive a practice dataset and a copy of the webinar text one week before the live webinar.
Acock, Alan C., (2015). Item Response Theory (IRT) — Draft of chapter that will appear in Acock, Alan C. (2016). A Gentle Introduction to Stata, 5th edition. College Station, TX: Stata-Press.
The 5th edition of this book will be available from Stata in early 2016. Note: Rasch/IRT is not introduced in earlier editions, as it became available in Stata 14.
About the Presenter
Alan Acock, Ph.D., is a University Distinguished Professor Emeritus of Family Studies in the School of Social and Behavioral Health Sciences at Oregon State University. He has a published extensively in family science and quantitative research methodology. His publications include seven books, 20 book chapters, and more than 150 refereed journal articles. In 2007, for his teaching and educational leadership, Dr. Acock received the Oregon State Alumni Distinguished Professor Award and NCFR's Ernest G. Osborne Award.
On-Demand Webinar Recording
You can watch the recording of this webinar at your convenience after the live webinar, even if you aren't able to attend the live event. The fee for this webinar is $30 for NCFR student members, $50 for NCFR members, $85 for nonmembers.
License for classroom use by one professor is available for $115 for NCFR members, $225 for nonmembers. License for departmental use (multiple professors) is available for $185 NCFR member / $370 nonmember. Departmental license for CFLE-approved programs is $155.
To purchase any webinar for classroom use, please fill out this webform.
Stata is a registered trademark of StataCorp LP, College Station, TX, USA, and the Stata logo is used with the permission of StataCorp. This webinar series is not sponsored by StataCorp and is not Stata certified.