Slides are here:
To prepare for this class:
A. Please read the following:
- from PDQ Statistics, pages 53-61 “Simple and Multiple Linear Regression”
B. Please consider the following questions for discussion in class:
- How is interpretation of regression coefficients different for continuous variables compared with dichotomous or categorical variables?
- How is it the same?
- What methods can we use to visualize the relationship between a continuous variable and a categorical variable?
- Why should we always visualize our data before jumping into a regression model?
C. Get your radon dataset ready for analysis
- Ensure that you know how to create a data subset in Deducer
Objectives of this class:
A. Hypothesis generation for continuous variables
- Which variables in the radon dataset are continuous?
- How might they be associated with the radon concentrations?
- How can we test our hypotheses?
B. Visualization of continuous variables by other continuous variables
C. Testing and quantifying the association between two continuous variables
- Pearson’s correlation coefficient
D. Multiple linear regression
- What happens when you have more than one variable?
- Unadjusted and adjusted effect estimates
- Interpretation of coefficients
- Interpretation of test statistics
- Approaches to variable selection
Last updated on February 9, 2017 @5:01 pm