Three different models were used to analyze data for use in binomial (infested/noninfested) sampling plans for the bird cherry-oat aphid, Rhopalosiphum padi (L.). Two of the models gave comparable results, but a third fit less well, especially for proportions >60% infested tillers. Sample sizes for the different models were also calculated and compared with sample size for direct counts. The incidence count method described in this paper should improve aphid sampling efficiency.