Correlation and Causation

Events that coincide with each other are not necessarily caused by each other. Consider for instance:

• Pure coincidence: Brazil’s GDP highly correlated with population of rabbits in Australia.
• Direct causation: Smoking causes lung cancer.
• Reverse causation: Number of firemen fighting a fire correlates with the size of the fire. It would be incorrect to conclude that firemen cause fires to spread. Similarly, it would be incorrect to conclude that using a wheelchair is dangerous, because a high proportion of people who use them have had an accident.
• Bi-directional or cyclic causation (A causes B and B causes A): Predator numbers affect prey numbers, but prey numbers, i.e., food supply, also affect predator numbers.
  Cyclic causation is often seen in the economy. For instance, people tend to spend less and save more when the economy is weak, causing the economy to decline further.
• Third factor (the common-causal variable) causes both A and B. Spike in sales of soft drinks correlates with incidence of beach drowning deaths. Drinking soft drinks is not causing the drowning deaths. A third factor, hot weather is causing people to consume more soft drinks, and it also the reason why more people are going to beaches to swim.