Post Conference Workshop:
Item Response Theory and Modeling: A Latent Variable Modeling Approach
May 24, 2018
Michigan State University
This one-day workshop is based on a latent variable modeling approach to item response theory (IRT) and item response modeling (IRM). The workshop commences with a non-traditional approach to IRT and IRM that is based on their essential connections to other behavioral measurement methodologies. These include classical test theory (CTT), (nonlinear) factor analysis, and logistic regression. The flaws of earlier treatments of CTT in the context of multiple IRT and IRM discussions are pointed out, which in addition to misrepresenting CTT effectively shift away research attention to more specific rather than more generally applicable modeling and analytic procedures. Resent research on the connections between classical test theory and IRT/IRM is then reviewed. Using a measurement invariance examination based approach, a multiple testing method for differential item functioning is discussed that deals with limitations of an existing widely utilized procedure. A readily applicable method for studying essential unidimensionality of multi-item measuring instruments is then obtained as a byproduct of this method. Item response models with covariates are finally discussed. Use of the popular software Mplus and Stata is made repeatedly and on a few occasions IRTPRO, flexMIRT, and R are utilized. The workshop is based on an integrative approach to measurement in the behavioral and social sciences and emphasizes throughout the links between IRT and IRM on the one hand and other measurement and modeling methodologies related to them on the other hand. Register for the post-conference or the 2018 Modern Modeling Methods Conference, click here.
This workshop will be held on Thursday, May 24, 2018 from 9:00-5:00pm in Laurel Hall, Room 102.
0. Resources for course. Latent variables and their relevance for Item Response Theory
(IRT) and Item Response Modeling (IRM).
1. Introduction and overview of IRT/IRM.
2. An example of item response modeling.
3. Popular unidimensional IRT models.
4. Information functions.
5. Differential item functioning (DIF).
6. Introduction to multidimensional IRT/IRM.
7. Extensions and limitations of some current IRT/IRM applications.
8. Conclusion and outlook.
Note. Participants in the workshop will benefit significantly from the keynote address by the
instructor on May 23rd.
Raykov, T., & Marcoulides, G. A. (2017). A course in item response theory and modeling with Stata. College Station, TX: Stata Press.
Raykov, T., & Marcoulides, G. A. (2016). On the relationship between classical test theory and item response theory: From one to the other and back. Educational and Psychological Measurement, 76, 325-338.
Raykov, T., & Marcoulides, G. A. (2017). On studying common factor dominance and approximate unidimensionality in multi-component measuring instruments with discrete items. Educational and Psychological Measurement (in press).
Bio for Tenko Raykov
Tenko Raykov‘s research spans a variety of areas with many practical applications, including latent variable and structural equation modeling, measurement and scale construction and development, multilevel modeling, longitudinal data modeling, analysis of incomplete data sets (missing data), latent class analysis (finite mixture analysis), survival and duration analysis, as well as item response theory and modeling. Recent projects include methods for reliability and validity estimation, examining population heterogeneity, survival analysis, missing data analysis, longitudinal modeling, model fit assessment, and measurement invariance. Professor Raykov has published more than 100 peer-reviewed articles in numerous academic journals, including Structural Equation Modeling, British Journal of Mathematical and Statistical Psychology, Multivariate Behavioral Research, Applied Psychological Measurement, Journal of Educational and Behavioral Statistics, and Educational and Psychological Measurement. He has also written several textbooks (with G. A. Marcoulides): A First Course in Structural Equation Modeling (2006), An Introduction to Applied Multivariate Analysis (2008), Introduction to Psychometric Theory (2011), Basic Statistics: An Introduction with R (2012), and A Course in Item Response Theory and Modeling With Stata (2017).
Register for the post-conference or the 2018 Modern Modeling Methods Conference, click here.