Three Day Workshop in Advanced Multilevel Modeling

John Poe, a doctoral candidate in Political Science and research methodologist with the UK Center for Public Health Services and Systems Research, will be conducting a three day progressive workshop in multilevel modeling.

The workshop will include two morning sessions (9:00-10:30, 10:45-12:15) and two afternoon sessions (1:00-2:30, 2:45-4:15) each day.

Among the topics to be covered are: introduction to MLM, fit and diagnostics, issues with data structure, network models, spatial models, MLM with panel data, MLM with cross-section data, linear and non-linear models, and Bayesian approaches.

Location is TBA.

What are multilevel models?
Many kinds of data have a hierarchical or clustered structure that should be accounted for when making inferences. For example, units of observation that share a similar group (nesting) are more likely to be similar than observations chosen at random. These units of observation may be further grouped or nested at higher (and meaningful) levels, such as geographic areas, institutions, or organizations. 

Multilevel models enable us to estimate residual components at each level in the hierarchy. In short, this powerful modeling technique permits us to better estimate standard errors and to make more accurate inferences. Multilevel models also permit us to develop and test hypotheses about group effects at multiple levels, while simultaneously estimating the effects of individual and group-level predictors. One can clearly see how having knowledge of and a grasp on multilevel modeling is important for researchers across the social, health, and behavioral sciences.

The workshop series features depth and breadth in instruction on multilevel modeling that is difficult to find at UK, and in general, and will be of great benefit to many researchers on campus.

You don't want to miss this great opportunity! Be sure to reserve your seat by registering below! 

Contact Information
Enter your Quantitative Initiative for Policy and Social Research username.
Enter the password that accompanies your username.