One of the many interesting things Freakonomics talked about was the difference between correlation and causation.  Correlation means that two or more things are related; causation means that one thing caused another thing.  We humans tend to confuse the two and jump to faulty conclusions that cause lots of problems later on.  For example, let's say the headlines read, "People who eat chocolate daily are three times as likely to develop arthritis."  Our tendency is to assume that daily chocolate consumption causes arthritis.  However, if we actually looked at the data, we might find that people who eat chocolate daily have a tendency to overeat in general, whereby they gain excess weight, causing undue stress on their knees and ankles, causing arthritis.  So daily chocolate consumption and arthritis are correlated.  Wikipedia (as usual) has more. #science