Location - Room 491, East Mall, The University of British Columbia, 1961 East Mall, Vancouver, BC

Course Fee - 495 plus HST until May 30, 2010; $595 plus HST after May 30.
Includes two refreshment breaks each day. Participants are responsible for their own lunches. Fees are not refundable after June 15, 2010.

Instructors

Dr. Bruno D. Zumbo, is Professor of Measurement, Evaluation, & Research Methodology (MERM), with affiliations in the Department of Statistics and the Institute of Applied Mathematics at UBC. He has done fundamental research in both theoretical and applied psychometrics and statistical science (mathematical and statistical methods). His research interests include procedures for evaluating variable importance in statistical models; performance characteristics of standard procedures (both parametric and nonparametric) under non-standard conditions; measurement theory (including axiomatic measurement theory, classical test theory and item response modeling), educational measurement, and the foundations of statistics. He has developed a secondary research program on measurement, program evaluation, and methodological issues in quality of life, subjective well-being, and social science research.

Yan Liu is a doctoral student in the MERM Program at UBC. Her research focuses on modeling of large-scale data, outliers, and multivariate statistics.  Dr. Amery Wu is a post-doctoral research fellow in the health sciences at UBC. She is interested in psychometric methods and multivariate statistics.

Course description

This short course will focus on the fundamentals of multilevel modeling with an eye towards providing the participants with the foundations to learn more advanced multilevel models. Each day will consist of a series of lectures and computer demonstrations covering the theory and practice of multilevel modeling methods. Participants will learn how to design multilevel studies, estimate and interpret random effects, model longitudinal data, and conduct multi-parameter tests. Multilevel models have many applications across the fields of psychology, sociology, education, business, health as well as public administration, and psychosocial health research broadly defined.

 This five-day intensive short course consists of sessions dealing with:

• The fundamentals of multilevel modeling building from a background in linear regression

• Limitations of traditional statistical methods, and the strengths of multilevel models

• The contexts in which multilevel models will be most useful; nested data such as kids in classrooms or change and growth studies

• The focus will be on linear mixed effects models, sometimes called random coefficient models, hierarchical linear models, and mixed models

• Examples primarily will come from the fields of education, psychology, and the psycho-social health sciences

• Some discussion and description of multilevel models in measurement and assessment

• Hands-on examples will be worked out and the participants will be able to run models with widely available software

Each day, morning lectures will be followed by afternoon lab sessions, using multilevel modeling software.

Prerequisite - Minimum knowledge needed by course participants: at minimum a background in linear regression is essential. Participants will be expected to be familiar with simple and multiple linear regression concepts and practice.

Additional materials/software requirements - Participants should bring a Windows laptop. We will load the required software for the interactive sessions and datasets will be provided. If you do not have access to a laptop with Windows, please let us know when you register and we will provide one for your use.

Who should attend - Graduate students, researchers and faculty members in the social, behavioral and health sciences, as well as epidemiologists and public health workers who are interested in developing and applying multilevel statistical models.

Sponsored by Population Data BC, in partnership with the Division of Continuing Studies, University of Victoria