Doing linear tests (like ANOVA or t-tests) on proportional data (values between 0 and 1) is difficult since the distributions of these values are not strictly Gaussian, especially when the proportions are near 0 or 1. The Rationalized Arcsine Transform linearizes the proportions and converts them to "rational arcsine units". The linear tests can then be performed on the RAU values. (p=0.5 roughly corresponds to a rau of 50).
RAU(p) computes the rationalized arcsine transform for a proportion value p (0 <= p <= 1). p can also be a vector of proportion values.
Gautam Vallabha (2019). Calculate Rationalized arcsine transform (https://www.mathworks.com/matlabcentral/fileexchange/16139-calculate-rationalized-arcsine-transform), MATLAB Central File Exchange. Retrieved .