+ Reply to Thread
Results 1 to 2 of 2

Thread: Variation of Mantel test?

  1. #1
    Points: 327, Level: 6
    Level completed: 54%, Points required for next Level: 23

    Posts
    3
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Variation of Mantel test?




    Hello. I am new to this forum but it looks like it will be a great reference!

    For my thesis I used Mantel tests to compare non-parametric behavioral data against genetic relatedness data. So, I had behavior rates in a square matrix and compared them against genetic relatedness values in a square matrix. The matrices had the same lay out (all individuals were in the same cells of the matrix). This told me if there was a relationship with the behavior rates and genetic relatedness between individuals.

    Now, I have taken the same matrices and broken the individual animals up into 2 groups based on their location. So I have Group A and Group B. I need to compare the different behavioral matrices of Group A with A and then Group A with B to see if their location influenced their behavior rates. There are different numbers of individuals within Groups A and B. With a Mantel test the matrices have to be the same size, I also believe they are the same individuals represented in the 2 compared matrices. The test I need would be comparing matrices of 2 different groups against each other. Does anyone know of such a test? Or does anyone have other advice?

    Hopefully this is clear, let me know if anything needs clarified!

  2. #2
    TS Contributor
    Points: 12,227, Level: 72
    Level completed: 45%, Points required for next Level: 223
    rogojel's Avatar
    Location
    I work in Europe, live in Hungary
    Posts
    1,470
    Thanks
    160
    Thanked 332 Times in 312 Posts

    Re: Variation of Mantel test?


    hi,
    assuming that the Mantel test is the right one for your research question I would try the following technique, related to the bootstrap method.


    I would pick two random samples ( with replacement) from the two populations so, that the samples satisfy the requirements for the test, run the test and calculate the the test statistic. Repeat the process a large number of times (500? 1000?) and calculate the average of the the test statistic and report that, together with the method used to obtain the result.

    I am not 100% sure that the bootstrap can be recommended for this test and/or whether it will be accepted in your thesis. Maybe someone else will also chime in and help. A negative argument is never a strong one, but I see no reason why this should not work .

    regards
    rogojel

+ Reply to Thread

           




Tags for this 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