64.010 - Hierarchical Linear Modeling With Large-Scale International Databases

Session Type: Professional Development Course

Abstract

Hierarchical linear modeling (HLM) has rapidly grown in popularity as the method of choice for analyzing data that is hierarchically structured, or nested. Data from large-scale international studies is very well suited for HLM analyses, as the sampled students are nested within classrooms and/or schools and nested within countries. However, not only are these databases unique in terms of the kind of data they contain, but there are methodological aspects of the databases that a researcher needs to understand (e.g., sampling weights and multiple achievement values) in order to correctly use HLM with the databases. The purpose of this course is to teach participants how to do HLM with data from large-scale international studies. The target audience includes graduate students, emerging researchers, and continuing researchers who have at least a basic level of proficiency using HLM software and who wish to learn how to do HLM using data from one or more large-scale international studies, including PIRLS, PISA, and TIMSS. Participants should bring their own laptop computer with HLM 7 loaded onto it. The content of the course will include a presentation on the design of international studies and databases, and implications for HLM analysis. Following this presentation will be a hands-on demonstration focused on the preparation of an international dataset for analysis with HLM. The remainder of the course will include a hands-on demonstration walking participants through some HLM analyses. Participants will be able to access course material ahead of time (including PowerPoint presentations and data files) via Dropbox.

Sub Unit

• Professional Development and Training Committee

Instructors

• David C. Miller, American Institutes for Research

• Francis Howard Lim Huang, University of Missouri

• Leslie Rutkowski, University of Oslo

• Andres Sandoval-Hernandez, International Association for the Evaluation of Educational Achievement

• Anna K. Chmielewski, OISE/University of Toronto