Proof of why correlation is not always causation

Imagine a hotel with 100 people in 100 rooms, 1 person in 1 room. Everyone finds a pill along with a slip of paper saying: “One of you will not take the pill. And one of you will die.”
The next morning, you find that one of the 100 people didn’t take the pill. But, one of the people who took the pill died.
This shows how misleading the slip is: just because these two things happen at the same time, that doesn’t mean that not taking the pill kills you.

Leave a Comment

Comments (2)

  1. Another example:

    Nearly all of Dr Terrett’s students are at CEFR B2 or higher according to their recent English tests. Therefore, Dr Terrett must be an excellent teacher 👀

    It’s a false conclusion, but can you explain why it is false using correlation vs causation?

  2. 😵

    Maybe list it out in bullet points for clarity?

    Any images to go with that? A diagram of the rooms, pills, etc. might help as a visual aid.

    I usually use the witch doctor story…

    Two people are sick. One visits the witch doctor and is cured; the other doesn’t visit the witch doctor and isn’t cured. Does this mean the witch doctor’s magic medicine worked? Have we taken into account the placebo effect?

    Later, a villager points out all the people who have visited the witch doctor and who are now cured. Now we have more cases linking the witch doctor’s magic medicine to healthy results, does this prove anything?

    What other data would we need to be able to draw meaningful conclusions about the efficacy of the witch doctor’s magic medicine? What would we need to know to be able to accept the witch doctor’s intervention as genuine medicine?