Regression Analysis I

May 25, 2017
9:00am-12:30pm

The Regression Analysis workshop is designed to deepen and expand understanding of linear regression modeling. The workshop will cover the basics of simple and multiple linear regression, with emphasis on topics commonly encountered in research, including

• types of research questions for which regression models are likely provide useful information
• model assumptions—their implications and ramifications of violations
• determining the validity of a model
• interpretation of regression parameters, especially
• in multiple predictor models
• after transformation
• in models including interactions
• tests of subsets of regression parameters
• variable selection and model building, contrasted with model and parameter testing.

Emphasis on practical issues to help researchers better apply regression analysis to address research questions and better understand and report results.

Prerequisites: Familiarity with material typically covered in an introductory statistics course, including graphical displays, mean and standard deviation, normal distributions, t-tests and confidence intervals and simple linear regression.

 

Instructor: Dr. Scott Richter is Professor in the Department of Mathematics and Statistics and Director of the UNCG Statistical Consulting Center. He teaches undergraduate and graduate level courses in statistical methodology, and consults extensively with researchers across campus. More information can be found at http://www.uncg.edu/mat/people/people.php?username=sjricht2.

Workshop files:

Regression workshop 2017 slides

Workshop analyses SAS code