A fairly large number of different methods exist for evaluating comparisons or null hypothetical contrasts. Because it is difficult to choose among them, this paper develops a 2 by 2 classification of the methods. The first dimension of the classification is decision-based error rate vs. experimentwise error rate. The second dimension is planned vs. post-hoc contrast. Most of the generally useful and better-known methods are considered; t, Bonferroni, Dunnett, Duncan Multiple Range, Rodger, Scheffe, Tukey, and Newman-Keuls Multiple Range. Each method is classified and discussed. If one has very specific questions to be answered, and a very good idea of what the true population means are, use carefully planned tests; otherwise use post-hoc methods. If one believes that a null hypothesis of equal means may be true and that one should have a conventionally low probability of rejecting it in error, use experimentwise error rate, but keep the number of means being considered as small as possible; otherwise use decision-based error rates. (Author/BW)