We propose an extension of the Wilcoxon–Mann–Whitney test to compare two groups when the outcome variable is latent. We empirically demonstrate that the test can have superior power properties relativ
The dataset contains a refined version of data originally extracted from the 2019 Spanish Google Books corpus. It is related to the upcoming publication Guerrero Montero J. and Blythe R. A (in submiss
Wald confidence intervals for a binomial proportion are known to perform poorly. Adjusted Wald intervals achieve better performance by adding in c successes and c failures (for an approp