Example: Henry - random forest model

Both alternatives suggest that the most important variables are gender and age, followed by class.

Source: Figure 11.2