can I compare p-value of permutation test with that of a t-test?

From: Jiti (sine_arc_at_hotmail.com)
Date: 12/28/04

  • Next message: Jiti: "Re: question : Type I error control" 
    Date: Tue, 28 Dec 2004 02:47:43 +0000 (UTC)

    This question relates to comparing the significance of the results
    obtained from parametric tests and that from permutation test. Lets
    assume that I conduct a two-tail single sample t-test to compare
    the means of samples obtained under conditions A and B for n subjects.
    And then I conduct a permutation test using the difference in means
    of A and B as my statistic and calculate the p-value from the
    distribution of this statistic from all simulated samples.
    Is it valid to compare the p-value obtained in the t-test and
    permutation test?
    Thanks again in advance...
    Jiti

  • Next message: Jiti: "Re: question : Type I error control"