sign test A non-parametric method to test the null hypothesis that a sample is selected from a population with median m. If the null hypothesis is true, the sample of size n is expected to have an equal number of observations above and below m, and the probability that r of them are greater than m has the binomial distribution B(n,0.5). The null hypothesis is rejected if the value lies in the critical region determined by the chosen significance level.
The test may also be used to compare the medians of two populations from paired data. |