| Session Title: Understanding Terminology in Multi-level Modeling for Program Evaluation |
| Demonstration Session 518 to be held in Calhoun Room on Friday, November 9, 9:25 AM to 10:10 AM |
| Sponsored by the Quantitative Methods: Theory and Design TIG |
| Presenter(s): |
| Caroline Wiley, University of Arizona, crhummel@u.arizona.edu |
| Mei-kuang Chen, University of Arizona, kuang@u.arizona.edu |
| Julius Najab, George Mason University, jnajab@gmu.edu |
| Abstract: In program evaluation, especially in the area of education, multilevel modeling is often used for data analysis to deal with the nested data structures. However, the terminology related to multilevel modeling, also known as linear mixed models, is somewhat confusing. For example, random coefficient model, slopes-as-outcomes model, random effect model, hierarchical linear model, variance component model, and value-added model are all slightly different models in the family of linear models. In this demonstration, these commonly used statistical terms will be presented in the context of program evaluation and conducting and interpreting statistical analyses. Understanding the meaning of these terms is pertinent to planning evaluation and using various software packages. |