0

for a project I need to get a mean each time, this mean I then use it to compare it to a distribution under the null hypothesis H0. However sometimes some values (few) are much larger than all the other values, so if I use a basic arithmetic mean these values will greatly impact the final mean.

Here is an example

enter image description here

Here if I do an arithmetic mean I get about 90, while in reality we can see that the vast majority of values on the axis are around 50.

Do you have an idea of an average that would be used in this kind of case?

chippycentra
  • 193
  • 6
  • 1
    If each dot is an observation then the large number of high values on the right hand side are not obvious errors and should be taken into account in your descriptive statistics. You have a right-skewed distribution – Henry Jan 23 '21 at 18:12
  • @Henry yes but as I said Henry I need one overall statistic that represent the most the values, and if I simply do an arithmetic mean then I do not get a representativ value at all. What do you think about ```Median``` or ```Geometric mean```wich one should be the best? – chippycentra Jan 23 '21 at 18:15

0 Answers0