Beretvas, Susan Natasha2013-12-052017-05-112017-05-112013-05May 2013http://hdl.handle.net/2152/22569textThis report focuses on introducing two statistical models for dealing with data involving complex social structures. Appropriate handling of data structures is a concern in the context of educational settings. From base single-level data to complex hierarchical with cross-classifications and multiple-memberships, we explain and demonstrate their distinction and establish appropriate regression models. Real data from the National Center for Education Statistics (NECS) is used to demonstrate different way of handling a cross-classified data structure as well as appropriate models. Results will be presented and compared to examine the practical operation for each model.application/pdfen-USHierarchical modelCross-classificationMultiple-membershipHandling complex multilevel data structures2013-12-05