ABSTRACT
This chapter looks at some of the simple statistical methods and procedures that may be useful for preliminary analysis of data. The statistical theory behind attempts to answer this question is quite complex in parts, and often in significance testing in real life situations it is uncertain that the necessary theoretical assumptions are met for such tests to be valid. An additional issue in significance testing concerns whether one should make so-called one-sided or two-sided tests. This is again a somewhat philosophical discussion, and really depends on the assumptions one makes before performing the test. The basic hypothesis in the several statistical methods described is always that there is no real difference between the groups, and that the observed difference is a random effect.
