One of the most difficult concepts to explain to a stats student is the idea behind hypothesis testing and the logic that goes with it. I have tried several different techniques, but one of the most useful is to compare it to a court of law in the US. In stats the sample data is innocent (of being different from the population) until proven guilty. When we reject the null hypothesis, we find the sample "guilty" of being different from the population. If we use this thought process, it may give some people another option for understanding how a statistical hypothesis test works.