ruqianl / comapr Goto Github PK

• the outlier dots are mixed with the actual points. For the boxplot version. Perhaps remove the outlier dots.

Hi Ruqian,
Would it be possible to add a function that detects interference? E.g. in order to make an "expected" distrubution for the null hypothesis of no interference:

1. take all double crossovers and randomise their labels
2. do this e.g. 1000 times to generate a distribution
3. calculate the p-value is = P(B ≤ b) = (b+1)/ (m+1) as per doi 10.2202/1544-6115.1585

I'm not sure if it's possible to make direct comparisons between groups.

Happy to discuss this request further. Thanks for considering it.