When do I use the Q-Test?

Question

Do I use the q-test whenever I'm find the mean and standard deviation of a set of numbers? For instance, I have a set of numbers: 15.13, 15.14, 10.83, and 15.35. The q-test says to throw out the 10.83, so do I indeed throw out the 10.83 when find the mean and standard deviation?

This is for a chemistry class.

Answer 1

it is used to throw out outliers in a set of data.

you know the formula, do your math.

Answer 2

You only throw something out when you have a specific reason for doing so. That specific reason can't have anything to do with the data itself -- like, "that one value looks really weird, and will throw my results off."

UPDATE:

When you throw things out, it is because you do not want to consider them, and it is as if they don't exist. Therefore, N is the number of observations that you will consider. Otherwise, you would, in effect, be replacing the outlier with 0, which is likely to be an even bigger outlier.

But note that, from a purely statistical view (I know squat about chemistry), ignoring observations should only be done if there is good reason to believe that the observation was gathered in error. The only other reason would be because the procedure itself is designed specifically to isolate outliers.