Plotting Data
Correlation vs. Causation
Examples
Example 1 Suppose a cup is turned upside down and liquid falls on the floor. There is no other possible reason that the liquid is on the floor other than the cup was turned upside down. Causation has one independent variable (whether or not the cup is upside down) and one dependent variable (whether or not the liquid falls on the floor.) It can be inferred that if there is liquid on the floor, the cup must have been turned upside down.
Example 2 Correlation means that two things are related, but not necessarily cause-and-effect. In this case, there are two dependent variables. For instance, temperature vs. number of ice cream trucks on the street. These are a correlation because as temperature increases, so does the number of ice cream trucks. However, there is no causation - the increase in temperature does not cause the number of ice cream trucks to increase. This is because there are two dependent variables. If you said "increasing temperature causes an increase in ice cream trucks" it would also mean that "an increase in ice cream trucks causes an increase in temperature." This is clearly a false statement.
Example 2 Correlation means that two things are related, but not necessarily cause-and-effect. In this case, there are two dependent variables. For instance, temperature vs. number of ice cream trucks on the street. These are a correlation because as temperature increases, so does the number of ice cream trucks. However, there is no causation - the increase in temperature does not cause the number of ice cream trucks to increase. This is because there are two dependent variables. If you said "increasing temperature causes an increase in ice cream trucks" it would also mean that "an increase in ice cream trucks causes an increase in temperature." This is clearly a false statement.