Common wisdom says that there are three ways to not tell the truth: lies, damned lies, and statistics. This might be true, but more often it is just misuse of statistics that brings about ridiculous results. Here is just one example (through Panda’s Thumb):
An ID guy states that the alleged difference between the chimp genome and the human genome of 1% is absolutely wrong. He does a simple thing: gets a 30 nucleotide chunk of one genome and looks for an exact match in the other genome. He can find it only in (approximately) 65% times. So the real difference should be 35%? 🙂
Simple problem for students: what is the chance that among 30 experiments there will be at least one success if the probability of success is about 0.01? (that is, what is the chance that chunks of length 30 will contain at least one “wrong” letter).