Analyzing Lines of Fit
Overview
Goal. Use residuals and residual plots to judge whether a linear model fits data, interpret a correlation coefficient's strength and direction, and explain why correlation does not prove causation.
Residual (definition). The difference between an actual y-value in the data and the value predicted by the model: residual = actual − predicted. A residual plot graphs the residuals against x; a good linear fit leaves residuals scattered randomly around 0.
Lesson Notes
Flip through the notes one page at a time using the buttons below.
Lesson Video
Watch the lesson video.
Interactive Activity
Goal. Judge a line of fit by its residuals, read a correlation coefficient, and separate correlation from causation.
Residual (definition). The difference between an actual y-value and the value predicted by the model: residual = actual − predicted.