COURSE ANNOUNCEMENT: 

************************************************************************
**********************************
The Strategy of Data Analysis in Epidemiologic and Public Health
Research
************************************************************************
**********************************

Instructor:     	Prof. dr. David G. Kleinbaum,  Emory University,
USA

Date: 		May 22-23-24,  2006
Place: 		Maastricht University, The Netherlands

In epidemiologic and public health research, an appropriate data
analysis requires a comprehensive strategy including various key
elements. 
Generally, the analysis starts with simple and stratified analysis, and
then proceeds with more complex multivariable techniques such as
linear and logistic regression analysis, survival analysis or analysis
of repeated measurements. While the separate elements of the analysis 
are dealt with in a variety of courses, the use of a comprehensive
analytic strategy is often neglected. This course emphasizes the
"strategy" 
of data analysis, and discusses a variety of issues and approaches in
this regard. The discussion will be placed primarily within the context 
of the analysis of an exposure-disease relationship using mathematical
modeling approaches, with emphasis on multiple logistic regression
analysis.

Study objectives
1. The student learns how a data analytic strategy can be developed,
from simple and stratified analysis to multivariable techniques 
relating the simultaneous influence of various independent variables to
the outcome.
2. The student learns how to incorporate multiple logistic regression
analysis in the context of the analytic strategy.

Target group and basic knowledge required
All PhD students and senior researchers, dealing with data analysis in
epidemiologic research. For a fruitful participation in the course, 
participants should be familiar with epidemiologic methods and
regression analysis.

Course schedule
During the first two days, lectures will be given by Professor Kleinbaum
in the morning and the afternoon. In tutorial groups, exercises 
will be discussed to bring the theory into practice and to test the
acquired insights and skills. During the third day an SPSS training will
be 
given in the morning and afternoon. 

Covered topics include: 
1. Overview of Analysis Strategy: Crude, Stratified, Modeling Approaches

2. The EVW model: why important (logistic regression emphasis). 
3. Logistic Regression for Matched Data 
4. General Modeling Strategy for an EVW logistic model (logistic
regression emphasis). 
5. Screening when there are a lot of E and C variables (logistic
regression emphasis). 
6. Modeling strategy with several exposure variables (logistic
regression emphasis) 
7. Collinearity Diagnosis and Assessment (for different types of models)

8. Overview of Analysis of Correlated Data.

Fee:
Euro 1150,-  including bed and breakfast (2 nights)
Euro 950,-    excluding lodgings

Information:
Dr. Frans Tan / Mrs. Hanny Prick
University of Maastricht
P.O.Box 616 - 6200 MD Maastricht, The Netherlands
Phone: +31-43-388.2319; Fax: +31-43-367.1458
E-mail: [log in to unmask] / [log in to unmask]
Website: www.researchschoolcare.nl
************************************************************************
**********************************