Question

3. According to Wikipedia, if X and Y are correlated, there are many explanations:

1. X causes Y; (direct causation)

2. Y causes X; (reverse causation)

3. X and Y are consequences of a common cause, but do not cause each other;

4. X causes Y and Y causes X (bidirectional or cyclic causation);

5. X causes Z which causes Y (indirect causation);

6. There is no connection between X and Y; the correlation is a coincidence.

A well-known study published in Nature (Quinn et al. 1999) showed a high correlation between children who sleep with a light on and later development of myopia. In response to this research, another group of researchers looked at a large data set collected on child myopia (Zadrick et al 2000, Nature) and did not find a relationship between room lighting and development of myopia. Instead, they found a strong relationship between myoptic parents and children who later develop myopia.

(You can and probably should discuss your answers to the following with other teams.)

Which category does Experiment 1 fall into and why?

Which category does Experiment 2 fall into and why?

Why do you think the experiments did not agree?

Answer 1

The first experiment showed a correlation between children who sleep with a light on and later development of myopia. But, the second experiment showed no relationship between them. This means that the correlation observed was simply a coincidence, maybe because of sample bias, or experiment design.

Hence Experiment 1 falls into category 6.

The second experiment showed a relationship between myopic parents and children who later developed myopia. This correlation is probably a cause of genetic inheritance, though it is simply an educated guess, hence it is a case of causation.

Hence Experiment 2 falls into category 1.

The reason experiments did not agree could be selection bias, insignificant sample size, or simply bad experiment design.

Comment in case of any doubts.