Quantitative Social Science Initiative Colloquium - Ben Soltoff

Mixed-Effects Modeling for Grouped/Clustered Data
When Dec 09, 2013
from 12:10 PM to 01:15 PM
Where B001 Sparks - the 'Databasement'
Contact Name
Contact Phone 814-867-2720
Add event to calendar vCal
iCal
Traditional heirarchical or multilevel model techniques do not apply to repeated measurement data where individual units are grouped into multiple categories which are not nested. Mixed-effects modeling is one technique for conducting inference and making generalizable findings about relationships between covariates and the outcome of interest. This approach provides several advantages over other analytic approaches regarding the treatment of missing data, repeated observations, and identifying unit-specific effects.
Bio:
Benjamin is a fourth-year Ph.D. candidate in the Department of Political Science, with specializations in American politics and political methodology. His research interests focus on the role of institutions and how they influence individual political behavior. His dissertation explores judicial behavior on state supreme courts, examining how variations in procedural rules and selection systems influence case selection and decisions on the merits. His work makes use of automated text classification procedures and multilevel modeling techniques.
Light lunch will be served.