Summary BBS2007 - Linear regression analysis

BBS2007: “Linear and Logistic regression and Repeated Measures Analysis”


Lecture 1

Linear regression
analysis
Bjorn Winkens
Methodology and Statistics
Maastricht University



Methodology and Statistics | University of Maastricht © Bjorn Winkens 2020

, Content
• Simple linear regression analysis (repetition BBS1003)
– Relation with independent-samples t-test and correlation analysis

• Multiple linear regression analysis I
– Correction for other variables: interpretation and tests
– Dummy variables
– Assumptions

• Multiple linear regression analysis II
– Effect-modification: interpretation and tests
– Top-down procedure
– Confounding

2

, Applications (1)
Study goals:
• Let X = independent/explanatory variable, Y = numerical
dependent/outcome variable
• Association between X (numerical) and Y simple (univariable)
• Assess group effect (X; binary) on Y linear regression (1)

Y = dependent variable = numerical and X = independent variable =
categorial or numerical
Examples:
• Association between age (X) and systolic blood-pressure (Y)
• Effect of treatment versus control (X) on decrease in blood pressure (Y)

(1)
Note: univariate = one outcome variable (Y); univariable = one explanatory variable (X)
3

, Applications (2)
Study goals:
• Association between X1,…, Xp and Y
• Determine which risk factors are multiple
independently related with Y (multivariable)
• Assess group effect, controlling linear
for other variables regression (1)
Example:
• Are age (X1), height (X2), weight (X3), sex (X4), and smoking (X5)
independently related to systolic blood pressure (Y)?
Multi-variable: >1 independent variable
(1)
Note: multivariate = more than one outcome variable or repeated measures (Y1, Y2,…)
4