Outliers

Deze pagina lezen in het Nederlands.

This page was translated from the Dutch-language original using AI and has not been fully reviewed. Where the two differ, the Dutch version takes precedence.

Which analysis should I choose if leaving out outliers leads to a different conclusion?

I ran my parametric test twice, once with and once without the outliers. This results in a different conclusion. What does that mean? Would this mean that, when a difference is found in the sensitivity analysis, a non-parametric test should be used after all for a correct analysis of the data? Or is it more valuable to describe and report both conclusions after a parametric test? I assume that when a test (for example ANOVA) becomes significant because of an outlier, that result is not valid. But conversely, when a result becomes significant without the outlier, the result tells me that there is a difference in the population, but that the outlier has concealed that difference. So, in case of a difference, should I switch to non-parametric, or describe both parametric results?

If including or excluding the outlier gives a different conclusion, it means that you do not have a robust analysis. This means it is not valid (it does not matter whether it was significant with or without the outliers). The advice, then, is indeed to use a non-parametric test. If all goes well, the conclusion of that test will not change depending on whether or not you include a few outliers. Ultimately, of course, you include all the data that was collected (unless you clearly see that the measurement went technically wrong or something similar).

---

Finished reading? You can go to the Overview of all topics, or to Choosing a test for help picking a suitable test or analysis. Want to learn more about biostatistics? Then follow the AMC e-learning Practical Biostatistics. Did you find something on these pages that isn't correct? Does a link not work? Or would you like to contribute to the wiki? Then get in touch with us.

This biostatistics wiki is an initiative of the former statistics helpdesk of Amsterdam UMC, location AMC. Amsterdam UMC staff can request statistical support via intranet.