摘 要
Martin和Tapia (1998)推廣Barnard (1947)構造拒絕域的精神，在多項分配母體下，提出以ESCM法建構拒絕域，雖然Martin和Tapia (1999)指出ESCM法具有最佳平均檢定力，但其程序實在太過於繁複且會有計算上的困難及拒絕域不唯一的問題。
在本篇論文中，我們以修正中間p-值改進中間p-值，修正中間p-值為一妥當p-值，且修正中間p-值法之計算較為簡單迅速也沒有拒絕域不唯一的問題。我們同時對中間p-值法、修正中間p-值法、卡方p-值法、修正卡方p-值法及ESCM法，在樣本總數n由3至30之下，進行實際顯著水準之數值分析。結果顯示，修正中間p-值法與ESCM法之實際顯著水準差距非常小－不超過且很接近指定顯著水準，在某些樣本數甚至比ESCM法更接近指定顯著水準。另一方面，中間p-值法在樣本數小時過於保守，卡方p-值法、修正卡方p-值法則超出指定顯著水準過多，因此實務上在樣本總數不是很大時，建議考慮使用修正中間p-值法。

參考文獻
Agresti, A. (1992). A survey of exact inference for contingency tables (with discussion). Statist. Sci. 7, 131-177.
Barnard, G. A. (1947). Significance tests for 2×2 tables, Biometrika, 34, 123-138.
Barnard, G. A. (1949). Statistical inference, Journal of the Royal Statistics Society, Series B, 11, 115-139.
Barnard, G. A. (1989). On alleged gains in power from lower p-values. Statist. Medicine 8, 1469-1477.
Barnard, G.. A.(1990). Must clinical trials be large? The interpretation of p-values and the combination of test results. Statist. Assoc. 89, 1012-1016.
Berger, R. L.and Boos,D. D. (1994). P-value maximized over a confidence set for the unisacne parameter .J. Amer Statist Assoc. 89, 1012-1016.
Berger, R. L. (1996). More powerful tests from confidence interval p-values. Amer. Statist . Assoc. 50 314-318.
Boschloo, R. D. (1970), Raised conditional level of significance for the 2×2 table when testing the equality of two probabilities, Statistica Neerlandica 24(1), 1-35
Fisher, R. A. (1935). The logic of inductive inference. Journal of the Royal Statistics Society, Series. A 98, 39-54.
Haber, M. (1986). A modified exact test for 2×2 contingency tables., Biom. J. 4, 455-463
Haber, M. (1987). A comparison of some conditional and unconditional exact tests for 2×2 contingency tables, Commun. in Stati.-Simul.16(4), 999-1013.
Hirji, K. F., Tan, S. J. and Elashoff , R. M. (1991). A quasi –exact test for comparing two binominal proportions. Statist. Medicine 10, 1137-1153.
Hwang, J.T. and Yang, M. C. (2001). An optimality theory for mid p-values in 2×2 contingency tables. Statist. Sinica, 11, 807-826.
Lancaster, H. O. (1961). Significance tests in discrete distributions. J. Amer Statist. Assco. 56, 223-234.
Lin, J. Y. and Yang, M. C. (2002). Validity of modified mid p-values in 2×2 contingency tables. Technical report.
Martin Andres, A. & Luna Del Castillo, J. D. & Herranz Tejedor, I. (1991). New critical regions for Fisher’s exact test, Journal of Applied Statistics, 18( 2), 233-254.
Martin Andres, A. & Silva Mato, A.(1994) Choosing the optimal unconditional test for comparing two independent proportions, Comp. Stat. and Data Analy., 17, 555-574
Martin Andres, A. & Herranz Tejedor, I. (1995). Is Fisher’s exact test very conservative? Comp. Stat. and Data Analy., 19, 579-591.
Martin Andres, A., Sanchez Quevedo, M. J, & Silva Mato, A. (1998). Fisher’s mid-p-value arrangement in 2×2 comparative trials, Comput. Statis. & Data Anal., 29(1), 107-115.
Martin Andres, A. & Tapia Garcia, J. M. (1998). On determining the p-value in 2×2 multinomial trials, Journal of Statistical Planning and Inference, 69(1), 33-49.
Martin Andres, A. & Tapia Garcia, J. M. (1999). Optimal unconditional test in 2×2 multinomial trials, Comp. Stat. and Data Analy., 31(3), 311-321.
Rhoades, H. M. and Overall, J. E., (1982): A sample size correction for Pearson chi- square in 2×2 contingency tables. Psych. Bull. 91, 418-423.
Suissa, S. and Shuster, J. J. (1985). Exact unconditional samples sizes for the 2×2 binomial trial. J. Roy. Statist. Soc. Ser . A 148(4), 317-327.
Silva Mato, A. & Martin Andres, A. (1995). Optimal unconditional tables for comparing two independent proportions, Biom. Journal, 37(7), 821-836.
Silva Mato, A. & Martin Andres, A. (1997). Simplifying the calculation of the p-value for Barnard’s test and its derivative. Statist Comput. 7, 137-143.
Tapia Garcia, J. M.& Martin Andres, A. (2000). Optimal unconditional critical regions for 2×2 multinomial trials. Journal of Applied Statistics, 27(6), 689-695
Upton, G. J. G.. (1982). A comparison of alternative tests for the 2×2 comparative trial. Journal of the Royal Statistics Society, Series. A 145, 86-105.
Upton, G. J. G.. (1992). Fisher’s exact test. Journal of the Royal Statistics Society, Series. A 155, 395-402.
Yates, F. (1984). Test of significance for 2×2 contingency tables. Journal of the Royal Statistics Society, Series. A 147, 426-463.