+ Reply to Thread
Results 1 to 2 of 2

Thread: Which analysis more suitable?

  1. #1
    Points: 13, Level: 1
    Level completed: 25%, Points required for next Level: 37

    Posts
    1
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Which analysis more suitable?




    Dear all,
    My data consists of 8000 rows and 14 columns, as follows:
    The rows correspond to quadruplicate measurements of 2000 variables and columns represent different non time-related conditions.
    I want to identify variables with statistically significant size difference in any pairwise comparison between conditions.
    I would like to find the maximum difference for a variable among two different conditions of it, but not comparing the variable’s values of two different rows.
    A two-way between subjects Anova and Tukey test calculates the mean of each column and not the mean of each variable in every column. On the other hand, one-way Anova post-hoc analysis with Tukey must be done 2000 times. In that case, I am not sure if FDR is the appropriate way for multiple testing correction.
    I would only like a comparison among all pairwise column combinations within each variable with an average difference and a p-value for each comparison.

    Thank you in andvance

  2. #2
    Points: 1,050, Level: 17
    Level completed: 50%, Points required for next Level: 50

    Posts
    17
    Thanks
    1
    Thanked 4 Times in 4 Posts

    Re: Which analysis more suitable?


    O.o a bit more information on the subjects would be nice. It sounds almost like a random regression analysis which can tease out the fixed effects of the individual 2000 against the random effects of itself and against the covariates (the measurements).

    At least that's what I think your data sounds like it needs.

+ 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