Model Selection Using Visualizations

Learn how visualizing the data can help in model selection.

When should we use an interaction model vs. a parallel slopes model? Recall that we fit both interaction and parallel slopes models for the outcome variable yy (teaching score) using a numerical explanatory variable x1x_1 (age) and a categorical explanatory variable x2x_2 (gender recorded as a binary variable). We compared these models using the figure below:

Get hands-on with 1200+ tech skills courses.