Category Archives: non-parametric statistics

7. Assumptions of Statistical Tests

“All models are incorrect.  Some are useful.” George Box *** When you do a statistical test, you are, in essence, testing if the assumptions are valid.  We are typically only interested in one, the null hypothesis.  That is, the assumption … Continue reading

Posted in assumptions, Biostatistics, heteroscedasticity, non-normality, non-parametric statistics, Statistics, Uncategorized | 14 Comments

7a. Assumptions of Statistical Tests: Ordinal Data

In my ‘7. Assumptions of Statistical Tests‘, there was a glaring omission (at least glaring for me): no discussion of ordinal data.  I felt that I let my readers down, by glossing over this issue.  Time to repair my quite … Continue reading

Posted in ANOVA, assumptions, Biostatistics, non-normality, non-parametric statistics, Ordinal, Statistics | 2 Comments

8. What is a Power Analysis?

People who haven’t the time to do things ‘perfectly’, always have time to do them over. Measure once, cut twice.  Measure twice, cut once. ‘My god, you’ve conclusively proven it.  Time equals Money.’ *** Cost for Failure Every CEO I’ve … Continue reading

Posted in Biostatistics, Dichotomization, Effect Size, Key Comparison, Non-Inferiority, non-inferiority hypothesis, non-parametric statistics, noninferiority hypothesis, p-values, Power, Statistical, Statistics, Treatment Effect | Leave a comment

10. Parametric or non-parametric analysis – Why one is almost useless

… ‘If you lost your watch in that dark alley, why are we looking here?’  ‘Well, <hic> there’s light here.’  (old chestnut) *** In my last blog, I stated that we should avoid dichotomizing as it throws away a lot … Continue reading

Posted in assumptions, Biostatistics, Dichotomization, Effect Size, heteroscedasticity, Interval, non-normality, non-parametric statistics, Ordinal, p-values, Power, Statistical, Statistics | 2 Comments

19. A Reconsideration of My Biases

No virtual observations were harmed in the running of this study. A man with one watch knows what time it is.  A man with two is not sure, at least for the Wilcoxon test. And don’t try this at home, … Continue reading

Posted in ANOVA, Effect Size, non-normality, non-parametric statistics, p-values, Power, t-test | 2 Comments

23. Small N study to publish or not?

The following question was sent to me, I thought it useful enough for a full elaboration: Submitted on 2014/05/12 at 8:23 am Dr. Fleishman, I am so happy I found your site. I have been trying to decide how to best … Continue reading

Posted in assumptions, Confidence intervals, Design, Effect Size, non-parametric statistics, Ordinal, p-values, Parallel Group, Power, t-test, Treatment Effect | 1 Comment