Two variables are associated (and not independent) if some of the variability in one variable can be explained by the other variable.
City | Boat | Car |
---|---|---|
Columbus | 20 | 300 |
Portsmouth | 120 | 50 |
You cannot summarize the battery usage in the two cities in the same way. In Portsmouth, people use more batteries in boats than in cars. However, in Columbus, people use many more batteries in cars than in boats. The variables are associated because the distribution of observations for vehicle differs depending on the city.