The purpose of this paper is to calculate the upper limit of the correlation between normal and dichotomous variables. An empirically obtained correlation should be evaluated in view of this limit, instead of the usual limit of Pearson correlation. (Author)

Investigated the effects of different types of risk information in a simulated decision-making task to test the predictions of protection motivation theory. College students (N=226) completed the task. Results showed outcome severity, efficacy of protection, and access to protection were related to protective decisions. (BH)

The fact that a group may subdivide into smaller subgroups is examined in the light of probability theory. (Author)

The characteristic shape of the normal distribution, one hump and two tails, is discussed. The reasoning is based on variation, combinations, probability, and logarithms. The purpose is to be able to answer some of the "whys" students might ask. (LS)

The Wilcoxon two-sample test used to examine judge bias. (MM)

A computational short cut suggested by Feldman and Klinger for the one-sided Fisher-Yates exact test is clarified and is extended to the calculation of probability values for certain two-sided tests when sample sizes are unequal. (Author)