0

I investigate survival until the following year (0,1) and I wish to test if the variance in survival for two or more groups are significantly different from each other.

I read that the Fligner-Killeen test is a non-parametric test which is very robust against departures from normality but is it correct (valuable technique for publication) to use it on binary data?

EmMo
  • 61

1 Answers1

0

Surely, if the probability of survival differs between groups, you do not need to test for differences in variance. The variance of a binomially distributed random variable with success probability p is np(1-p). If the probabilities differ then, other than the special case p[a]=1-p[b] for groups a and b, the variances must also differ.

S Ellison
  • 111
  • I have ran MCMCglmm model (family = "categorical") with one random effect on survival and with the categorical variables I was interested in investigating but the results revealed no significant differences in the probability of survival between the groups.

    The exploration of the raw data suggested differences in variance between the group that is why I was interested to do a Fligner-Killeen test. Is that a correct procedure?

     – EmMo Feb 18 '16 at 15:52