The outlier when you’re dealing with data, is typically the piece of information that doesn’t really fit the standards of everything else. And in an experiment, it can be a sign that something didn’t go right, and in a test it would be the person who destroyed the curve for everybody else. It’s basically the one piece of data that sort of doesn’t fit in the normal distribution of everything else.
To find it, what we do is, we first want to arrange our data in some sort of order, typically smallest to largest. I’m going to rearrange these numbers, so we have 50, 56, 56, 65, 71 and then 87. Once you have your numbers arranged logically, just look at the difference between each number. So the 50 and 56 we have a difference of 6, 56 to 56, obviously that’s a difference of nothing, 56 to 65 is 9, 65 to 71 is 6 and then 71 to 87 is 16.
The outlier is going to be the one that sort of doesn’t fit in. Here we have some decent size gaps, but everything is a gap less than 10, short of this last little jump which went up 16. In this case, the outlier is the one that doesn’t fit, that mould which is going to be the 87 on the end.
Just line them up in ascending order, and to find the outlier just look for large gaps. Typically it’s going to occur at the high end or the low end.