I'm having non-parametric data and plan to use Bonferroni adjustment for repeated comparison. Is it suitable? Is it the same way to perform adjustment by dividing the no. of comparison?
PS: there is no such thing like "non-parametric data". Data subject to statistical analyses is numerical, and it may either be discrete or continuous. What you mean is presumably that the (conditional) frequency distribution of the data can not be described well by some known parametric probability model.
Bonferroni correction is your only option when applying non-parametric statistics (that I'm aware of). Or, actually, any test other than ANOVA. A Bonferroni correction is actually very simple. Just take the number of comparisons you want to make, then multiply each p-value by that number.
2. Which Statistical test is most applicable to Nonparametric
It seems SPSS also provides some nonparametric tests for this (haven't ... Dear Sheejamol, you can use Bonferroni correction for multiple comparisons . ... For post hoc tests, Mann-Whitney U Test, is good, But, with a correction to adjust for the ...
3. Bonferroni correction on multiple Kruskal-Wallis tests - Cross ...
Jan 14, 2015 - You would not use the Bonferroni adjustment on the Kruskal-Wallis test itself. The Kruskal-Wallis test is an omnibus test, controlling for an overall false-positive rate. You would use the Bonferroni for post hoc Dunn's pairwise tests. Indeed, Dunn introduced the "Bonferroni" adjustment.
Do a relevant paired test for each pair (e.g. t test, Mann-Whitney, Wilcoxon) and adjust the sig values with a Bonferroni adjustment. NB it is not permitted to do a ...