For example, if I asked what is the next number in the sequence …, 1, 2, 3,…, you would, probably, postulate 4. But if we revealed a little more information: 1, 1, 2, 3, …, then you would, hopefully, postulate 5. Inductive reasoning is not always a guarantee that you will come to the correct conclusions. Not so long
ago, a doctor someplace noticed that people who smoked tobacco died prematurely more often than not, so he concluded that smoking kills people.
What he didn’t understand is that (unlike pure mathematical sequences), life isn’t so neatly defined. Instead, the doctor found a correlation. Yes, there is a strong relationship between smoking and premature death, but it is not an exact (i.e., one-to-one) relationship, for there are lots of people who die prematurely without smoking and plenty who live too long although they smoked their entire lives. In other words, like the mathematical example above, we need to have as much data as possible to make proper hypotheses. Please don’t misquote me; I’m not advocating smoking, for only an idiot would bet his life on a 50-50 wager. I am, however, advocating informed, intelligent decision making when it comes to interpreting data.