The choice of an appropriate sample size is a neglected topic in termite sex ratio research because the literature contains a variety of sample sizes with most < 50 individuals. We compared the effect of sample size (from 25 - 800) on matching the ratio obtained from several Reticulitermes flavipes (Kollar) populations of 1600 individuals. This study showed that a sample size of < 50 is too small to estimate the actual sex ratio of a population. A sample size of at least 100 termites is required to provide ±10% error in a census of Reticulitermes sex ratios. The results should improve confidence in estimates of termite colony sex ratios, and we propose a standard sample size for such studies.