Research Methods Series

TREC@WUSTL hosts a Research Methods Series that is open to the entire Washington University community. 

2014-2015 SERIES

2013-2014 SERIES
"Thank R It's Friday"
Click to read the flyer 

Session One
Many projects require the calculation of sample size before collecting the actual data as a means of achieving desired power, effect size, or significance level. This is particularly true in funded medical research where the collecting of patients for study is expensive and time-consuming. This seminar covers the basics of sample size calculations including: meeting statistical objectives, modifications for clustered and stratified samples, difficult issues with randomized clinical trials, and comparing model results.  The emphasis is on the primary statistical principles rather than cookbook use of websites.
Presenter: Jeff Gill, PhD
Department of Political Science, College of Arts & Sciences, Washington University in St. Louis
Division of Biostatistics and Division of Public Health Sciences, Washington University School of Medicine

Session Two
“Efficient Statistical Methods in Case-Control Studies”
This seminar addresses common statistical challenges in case-control studies and why the logistic regression model is often more appropriate in case-control studies based on theoretical justifications. Also introduced will be the most up-to-date methods for: fitting any model beyond the logistic regression, methods applicable when the disease prevalence rate is known in the population, strategies for multi-stage sampling, and methodologies for gene-environmental interaction.
Presenter: Jihyun Song, PhD
Postdoctoral Fellow, TREC@WUSTL