 Okay, you've made a histogram of your data and you're looking at it and it looks somewhat like a bell curve, but there's something going on over here and it makes me wonder are those extreme values or their outliers? Well you can guess and say they probably are because they're so much different than the rest of the data, but we don't know for sure unless we run a test and calculate the values that are outliers. I'll show you how to do that. Okay, I brought up the data that was responsible for that histogram. Let's start by creating a box plot or a box and whisker plot. I'm going to take this column there where the data is, control shift down arrow to make sure I drop to the bottom, get all the data, then go to insert, recommended charts, all charts, find box and whisker, click OK. We've got our basic chart, it's kind of hard to see because it's all squished up. I'm going to click on the y-axis, right-click, format axis, brings this open and set my minimum to be 4.5 to expand it and then close it, get out of the way. So we've got our basic box and whisker, I'm not going to pretty it up, but you can see we've got some points that are sticking out here and those are outliers. This box and whisker, if you remember, the box contains the data that falls in the interquartile range between Q1, which is the lower edge of the box, the first quartile, and Q3, which is the third quartile, the middle of the box is the median, which is Q2. These things are called fences, and they are delimiters. Any points that are outside the fences, above the fence, greater than the fence, or below the fence, are technically outliers. And the way that's calculated, it is the lower, in this case, Q1 plus 1.5 times the entire quartile range, which is the width of that box. And anything that is smaller than that is an outlier, anything greater than that is outlier. You can't just stop here because you don't know if you've got some double points, redundant points in there, or the same value more than once. So we need to dig into it a bit. I do it using descriptive statistics using Excel functions. We use the count, using the count function of 400, get the mean, standard deviation. We use the standard deviation of a sample, because this is a sample. Calculate the min, the max, the range, which is there between the max and the min. And then we calculate Q1 using the quartile.exe function. There's also a quartile inc function, but most people use the exe, and the arguments for it are the data range, E1 to E400, comma, 1, 1 indicates you want the first quartile, 3 indicates you want the third quartile. You could put a 2 in there to return the median if you wanted. The inner quartile range is just the difference in those two, Q1 minus, excuse me, Q3 minus Q1, which is 0.138. And I calculated the fences using the Q1 minus 1.5 times the inner quartile range, gives me 4.71, and then the upper fence is the Q3 plus 1.5 times the inner quartile, 5.265. And that's what those values are. So now let's take this information and see if we have any outliers here. First thing I want to do is to select that column and then go to data and then sort. And so it's sorted from smallest to largest. Now you could just manually inspect these. I like to use a conditional formatting to help me find out. I'm going to use, first of all, select that. Any cell that's less than the value of the lower fence, I want to be red. And then I'm going to use a greater than any cell that's greater than the upper fence. I want to be green, okay? And now I'm going to use the format painter and drag this down to format all my data. And you can see I've got an outlier there, of course, 4.6. And then down at the bottom I've got 4 more, 5.33, 5.33, 5.87, 5.9. That's one way to do it that's really fairly quick and foolproof.