Nearly all the statistical functions implemented in EASISTAT are as described in Statistical Methods in Medical Research by P. Armitage and G. Berry (2nd edition, Blackwell Scientific Publications, 1987). The user is advised to consult this source for a detailed account of the theoretical basis of the tests and the circumstances under which they should be used. It is assumed that the user has at least a basic knowledge of statistics, and only brief descriptions of the tests will be given in this manual, which does not replace a statistics textbook.
The significance values given are generally two-tailed, except for a few of the nonparametric tests for which it is expected that only small samples may be available and for which one-tailed significance values are output. If a test is one-tailed this is made clear when the p value is output. In practice this means that sometimes you may want to double the one-tailed probability value given (i.e. halve its significance) in order to produce a two- tailed significance value, depending for example on whether or not you had a prior hypothesis for the direction of the difference. Or you can just quote the result, say it is one-tailed and let people draw their own conclusions.