In my research I am using a surrogate approach (number of surrogates=100)
and I making a comparison among indeces (as correlation
indeces computed with a second known signal ) from my original data and from surrogates data , respectively. I would like to verify if the index
value from my original data is very "different" (in statistical sense)from
those surrogates. To detect and to quantify this difference I thought to use an outliers-non parametric test (that is, considering the index from original data an outlier value in comparison with the index from surrogates) . Is it a correct approach? What type of test can I use?

thank you for your help!