Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

What sigma value to use?

  1. Mar 8, 2015 #1
    I'm working on a project where I have about 120,000 sources and my supervisor has told me to work out what significance to use for taking results.

    I had set it at 3 sigma. But there's roughly 1 in 500 chance that we will get a 3 sigma result by chance.

    I've been looking for a table to compare the sigma with probability of random occurrence, but I don't think I'm typing the right thing into google or something because I can't find what I'm after!

    Also, any recommendations on what I should be going for would be great, 1 in 1,000,000 chance? more or less? I'm not really sure what I should do here. Any advice would be greatly appreciated, my understanding of statistics is lacking :(
     
  2. jcsd
  3. Mar 8, 2015 #2

    mfb

    User Avatar
    2016 Award

    Staff: Mentor

    What do you mean with 120,000 sources?
    What what is the "significance to use for taking results"? Why don't you look at the distribution of deviations, and see if it agrees with the expectations?

    For normal distributions, wikipedia has a table,and WolframAlpha also understands the concept.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook