Just because two phenomena are associated, one does not necessarily cause the other. A frequent reasoning fallacy exploited by partisans of about any cause.

Very important to remember when applying any sort of social science, such as psychology.

Both my high school and college psych courses provided me with the 'ice cream causes drowning' example:

Drowning deaths are most frequent during the months when ice cream sales are at their peak.

From this fact alone, one can make the causality go two ways:
  1. By eating ice cream, you are more likely to get a cramp and drown while swimming.
  2. When people drown, the mourners are more likely to console themselves by going out for ice cream.
Both of these theories are pretty shaky. What is more likely is that we did not look at enough information:

3. During the summer, more people are likely to go swimming, and drown. People are also more likely to buy ice cream.

In this model, note that drowning and ice cream sales do not effect each other, but both are effected by a common cause.

Another example, from an early statistics class of mine:

The number of arrests for prostitution in a given year in Sydney correlate strongly with the number of graduates from Harvard Business School that year.

Make what you want of that. Whatever you make of it, you will be wrong.

Log in or registerto write something here or to contact authors.