What is another option when setting cutoffs from a data set? I ran a set of normals in a new assay. I want to set negative and positive cutoffs. I know that I can take the mean plus 2 SD. Are there any other ways?

Hi Assay Queen,
There are a lot of ways to determine cutoffs, but it really depends on what you are trying to do. Can you be more specific when you are "setting cuttofs from a data set"?

At any rate, some standard favorites for determining statistical measures are:

Student's t-test.

Wilcoxon test.

If you want to get really hard core into cutoffs you can start thinking about using Bayesian techniques, clustering algorithms or even defining support vectors. But my guess is that you are probably looking for something much simpler, like one of the first two listed above.


Again, with sparse details in the original post, I'm not sure which one will apply to you, but hopefully that gets the wheels turning.

Thanks, jonatmudd!

I have tested known normals and postives in a new assay. I am looking at this data set trying to determine what my negative, positive and indeterminate range should be.