+ Reply to Thread
Results 1 to 1 of 1

Thread: Non parametric test help - Grad student

  1. #1
    Points: 4, Level: 1
    Level completed: 7%, Points required for next Level: 46

    Thanked 0 Times in 0 Posts

    Non parametric test help - Grad student

    I have a data set of 6 different element's values (i.e. Aluminum= 0.25, Copper= 0.06, etc.) over a two year period. I am trying to see if there are any differences in the element's values between the two years and location where they were collected. The samples were collected at different cities around Lake Michigan (Chicago, Milwaukee, and Manitowoc). I am using JMP Pro11 to complete my analysis.

    Example of my data table:

    Sample type - Location collected - Year - Aluminum
    Water - Chicago - 2014 - 2.00
    Water - Chicago - 2015 - 0.25
    Water - Milwaukee - 2014 - 1.15
    Water - Milwaukee - 2015 - 0.05
    Water - Manitowoc - 2014 - 0.89
    Water - Manitowoc - 2015 - 0.035

    I checked to see if my data was Normal, and it wasn't (even when I log transformed my data). I then went on to perform a Non-parametric test, Wilcoxon test, and saw that it produced a Prob>Chisq value of 0.0307* (which I understand this test is telling me that I have significant difference between my values when compared to year and location sampled).

    I then proceeded to conduct a Wilcoxon each pair test to see between which years and locations the element's values were significant. I thought this was the next logical step, but when I ran the Wilcoxon each pair test, it said none were significant. I've tried the Steel-dwass All Pairs test and Dunn All Pairs test as well with the same result. I don't know what else I can do.

    I've attached an example of my Wilcoxon test, as well as my Wilcoxon each pair test.

    Thank you for your help.
    Attached Files

+ 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