Genetic diversity is the basic and most important component of biodiversity. It is essential for the effective conservation and utilization of genetic resources to accurately estimate genetic diversity of the targeted species and populations. This paper reports analyses of genetic diversity of a wild soybean population using three molecular marker technologies (AFLP, ISSR and SSR), and computer simulation studies of randomly selected subsets with different sample size (5-90 individuals) drawn 50 times from a total of 100 wild soybean individuals. The variation patterns of genetic diversity indices, including expected heterozygosity (He), Shannon diversity index (/), and percentage of polymorphic loci (P), were analyzed to evaluate changes of genetic diversity associated with the increase of individuals in each subset. The results demonstrated that (1) values of genetic diversity indices of the same wild soybean population were considerably different when estimated by different molecular marker techniques; (2) genetic diversity indices obtained from subsets with different sample sizes also diverged considerably; (3) P values were relatively more reliable for comparing genetic diversity detected by different molecular marker techniques; and (4) different diversity indices reached 90% of the total genetic diversity of the soybean population quite differently in terms of the sample size (number of individuals) analyzed.When using the P value as a determinator, 30-40individuals could capture over 90% of the total genetic diversity of the wild soybean population. Results from this study provide a strong scientific basis for estimating genetic diversity and for strategic conservation of plant species.