A computer simulation was undertaken to determine the effects of using Huynh's single-administration estimates of the decision consistency indices for agreement and coefficient kappa, under conditions that violated the beta-binomial assumption. Included in the investigation were two unimodal score distributions that fit the model and two bimodal types that did not. While varying test length, mastery level, and sample size, estimates were calculated for both indices from two data sets, simulated to represent scores on equivalent forms. Then Huynh's method was applied to one of the sets to obtain the beta-binomial estimates. Statistics were computed to measure the magnitude and direction of bias resulting from the application of Huynh's technique. The results of the study did not provide conclusive evidence that Huynh's method is adversely affected by violation of the beta-binomial assumption. There was, however, clear indication that the estimates are strongly influenced by certain configurations of scores. (Author/PN)