This is the third possibility for bivariate relationships. Again, we have some standard widely used tools: the boxplot for visualization, and comparing means/medians for an initial exploration of the effect of the categories in the average values of a numerical feature.
Firstly, let's remember what a boxplot is. Although its basic construction can vary between different software tools, it is usually constructed as:
- The graph starts at the minimum value (bottom horizontal line).
- Then, the box starts at the 25th percentile. The horizontal line inside the box corresponds to the median (50th percentile), and the top of the box corresponds to the 75th percentile.
- Finally, it ends at the maximum ...
