The Statistical Sign Test

Abstract This paper presents and illustrates a simple statistical test for judging whether one of two materials or treatments is better than the other. The data to which the test is applied consist of paired observations on the two materials or treatments. The test is based on the signs of the differences between the pairs of observations. It is immaterial whether all the pairs of observations are comparable or not. However, when all the pairs are comparable, there are more efficient tests (the t test, for example) which take account of the magnitudes as well the signs of the differences. Even in this case, the simplicity of the sign test makes it a useful tool for a quick preliminary appraisal of the data. In this paper the results of previously published work on the sign test have been included, together with a table of significance levels and illustrative examples.