+ Reply to Thread
Results 1 to 2 of 2

Thread: chi square test to test difference for 5x2

  1. #1
    Points: 906, Level: 16
    Level completed: 6%, Points required for next Level: 94

    Thanked 0 Times in 0 Posts

    chi square test to test difference for 5x2

    My data has two groups people, VV and NV, and I want to test the difference in their occupation. The occupation were 5 groups: farm/fh, farmer/fh, professional, skw, wcollar. I used chi-square test in SAS and the results show there is no difference between the VV and NV groups. The code is as follows:

    proc freq data=vvnv;
    tables Occupation * status / fisher;

    The P-value is >0.05, so there is no difference between the two groups. But when I looked at the table, it seems there are some difference, for example, VV group has only two types of occupation, 50% (10 people) were farmer/fh, and 50% (10 people) were skilled workers. But NV has 5 types of occupation, 55% skilled workers, and 34% farmer/fh, 9% white collar etc. VV has 0 in the three other types.

    My question is: the overall 5x2 chi-square test showed no difference, but I break it into 5 of 2x2 chi-square to test each occupation type?

  2. #2
    Omega Contributor
    Points: 38,334, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Not Ames, IA
    Thanked 1,185 Times in 1,146 Posts

    Re: chi square test to test difference for 5x2

    You ran a fishers exact test which is more applicable given your sample counts. Splitting them up may be an idea, but you would be conducting 5 tests instead of 1, which increases your false discovery rate. Your overall problem here is your small sample size. You are likely under powered to find a difference that could be of interest.
    Stop cowardice, ban guns!

+ Reply to Thread


Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts

Advertise on Talk Stats