PD24-03: Multilevel Modeling With International Large-Scale Assessment Databases Using the HLM Software Program

Session Type: Professional Development Course

Abstract

Data from international large-scale assessments (ILSAs) reflect the nested structure of education systems and are, therefore, well suited to multilevel modeling (MLM). However, because these data come from complex cluster samples, there are methodological aspects that researchers need to understand when using MLM, including the need to use sampling weights and multiple achievement values for accurate parameter and standard error estimation. Using recently released ILSA data from TIMSS 2019, this course will teach participants how to conduct MLM, with an emphasis on the assessment design features of ILSAs (e.g., TIMSS, PIRLS, and PISA) and implications for MLM analysis.
Participants will learn how to specify two-level models using HLM software as well as about model comparison, centering decisions and their consequences, and the resources available for doing three-level models. Time will be allotted for hands-on exercises, with instructors available to mentor and answer participants' questions. Step-by-step demonstrations of each practice item will also be provided. Interactive activities will provide research networking opportunities throughout the course. Participants should have a solid understanding of Ordinary Least Squares (OLS) regression and a basic understanding of MLM. Prior experience using a statistical software program, such as Stata, R, or SPSS, is helpful. Prior knowledge of ILSAs—and prior experience using their databases or HLM—is not required. So that they can fully participate in the hands-on exercises, participants without an HLM8 license will be informed prior to the workshop how to temporarily access HLM8, which works in Windows and Parallels Desktop on Macs.

Sub Unit

• Professional Development and Training Committee

Director

• Amy H. Rathbun, American Institutes for Research

Instructors

• Francis H. Lim Huang, University of Missouri

• Saki Ikoma, American Institutes for Research

• Sabine Meinck, IEA Hamburg

• B. Jasmine Park, American Institutes for Research

• Yuan Zhang, American Institutes for Research