針對兩獨立二項母體的勝算比，我們通常會以區間估計的方式來探討勝算比。一般而言都是使用大樣本近似方法建構勝算比的信賴區間，但在中、小樣本時，此方法誤差會很大，故本文使用正確條件法及正確非條件法建構勝算比的信賴區間。由正確條件法所建構的信賴區間常具有保守性；由正確非條件法所建構的信賴區間會有最短的區間長度，其覆蓋機率會靠近名目水準 1-α 且不小於 1-α 。 For the interval estimation of the odds ratio in two independent binomial samples, the usual method to construct confidence interval is the large-sample approximation method. But, such a method will produce a confidence interval which has the actual significant level much larger than or equal to the nominal lever if the sample sizes are small or moderate. In this paper, we use the exact conditional approach and the exact unconditional approach to obtain a modified interval. Numerical studies show that confidence intervals based on the exact conditional approach can be conservative with small to moderate sample sizes. The modified confidence intervals based on the exact unconditional approach has shorter length, and its coverage probability is closer to and at least the nominal level.
