**Due to the rapidly-changing situation with COVID-19, this training has been cancelled.**
Workshop 3: Building a Regression Model
Once a regression model is selected, available data are then used to build a model to help us learn about relationships of interest. Important considerations for model building include variable coding, identifying confounding and interactions and using model selection criteria.This workshop will focus on developing a conceptual understanding of the process of model building and interpreting adjusted effect measures and interaction terms.
At the end of this workshop the audience will be able to:
- Define variables for modelling purposes
- Identify key components of confounding and interaction identification
- Select the ‘best’ regression model using appropriate model fit criteria
- Interpret adjusted regression model estimates and corresponding confidence intervals
About the NB-IRDT Workshop Series
This one-day workshop is offered as a part of a three-part series intended for individuals who are familiar with statistics and want to learn about using regression modelling:
- March 12 - Workshop 1: Describing and Comparing Populations
- April 16 - Workshop 2: Selecting a Regression Model
- May 14 - Workshop 3: Building a Regression Model
Each workshop can be taken on its own; it is not necessary to take all three workshops. The workshops use population-based data for guided examples but methods taught are applicable to data collected for many purposes, such as in clinical research or survey studies. This workshop series is led by Dr. Sandra Magalhaes, an epidemiologist and researcher at the New Brunswick Institute for Research, Data and Training (NB-IRDT). Each workshop is catered. Seating is limited. This session is being offered in English only.