Topics Covered
The Linear Equation
Break down Y = β₀ + β₁X + ε — intercept, slope, and error, demystified.
Least Squares
How R finds the best-fit line by minimizing the sum of squared residuals.
Mean-Centering
Why centering your predictor makes β₀ interpretable and reduces collinearity.
Key Concepts
- Interpreting coefficients: what a one-unit change in X really means for Y
- R² as proportion of variance explained — and its limitations
- Residual plots for checking model assumptions
- The lm() function in R: fitting, summary, and confidence intervals
Homework
- Fit a simple regression predicting reaction time from stimulus intensity
- Mean-center the predictor and compare the intercept before and after
- Plot the regression line with confidence bands using ggplot2