Evaluating subsampling methods for estimating numbers of Varroa jacobsoni mites (Atari: Varroidae) collected on sticky-boards

Abstract

Six subsampling methods for estimating the number of Varroa jacobsoni (Oudemans) mites on sticky-boards were evaluated. Each method was based on stratified random sampling and was evaluated on 3 groups of boards: those with an undetermined number of mites, those with >500 mites, and those with >1,000 mites. A working compromise between efficiency and precision was obtained by randomly sampling half of the cells in each of 96 two-cell strata. For this method and these 3 groups of boards, estimates of the number of mites on individual boards were within 38.46, 5.93, and 5.03%, respectively, of the true value 95% of the time. Evaluation of this method on a different set of 120 boards gave similar results. Using the same method, estimates of the total number of mites on 3 boards from the same colony were within 3.33% of the true value 95% of the time. These results demonstrate that subsampling can reduce the amount of time required to determine the number of mites on a sticky-board by approximately one half.