Testing for association in 2 x 2 contingency tables with very small sample sizes.

Applied statistic textbooks generally recommend the use of chi-square tests of homogeneity and independence with 2 x 2 contingency tables only when the expected frequency of each cell is five or more. Recent research has shown this rule-of-thumb criterion to be unnecessarily restrictive, but has not explored the accuracy of the chi-square tests when the total number of observations is less than 20 or when expected frequencies fall well below one-the primary issues considered in this article.