A caveat on the use of a revised bootstrap algorithm

A recently suggested revision of Efron's bootstrap algorithm for small bivariate samples discards bootstrap samples having fewer than 3 distinct points in order to prevent degeneracies in Pearson'sr. Examples are given showing that this approach may severely underestimate the variance of sample moments when sample sizes are very small. This paper should have appeared in the September 1984 issue, however, due to some unknown reason the final stages of the publication process were not completed. We regret any inconvenience to the authors and our readers.—Editors. This research was supported by NIMH grant number 37188.