積液型中耳炎好發於學齡前的兒童，嚴重的話需要進行外科手術將通氣管置入耳鼓以幫助積液排出。然而，通氣管會因為自然的脫落或是耳內的積液堵塞導致中耳炎不斷的復發。因此，研究者想要藉由建立檢定統計量來檢測實驗組與對照組發生發炎情形的風險是否一致，了解在術後的給予藥物治療是否能夠延長通氣管的使用時間，以減少發炎的情形發生。在本文中，將會比較在兩個不同檢定統計量之下其檢定力的大小，並且進一步找出最佳的樣本比例配置。而在實際資料的部分，也會針對耳科學的資料，比較兩個檢定統計量的檢定力大小。

In the clinical trials, researchers tend to compare the efficacy of the medicine by hazard. For clustered right-censored data, Manatunga and Chen (2000) proposed a test statistics based on the ratio of hazard. However, different test statistics may lead to different power. As the result, the goal of this research is to compare the power of two different test statistics. Moreover, we find out the best proportion allocation based on the different parameters combination. In addition, we demonstrate the proposed method using an otology study conducted by Le and Lindgren (1996).

1.Huster, W. J., Brookmeyer, R., and Self, S. G., “Modelling paired survival data with covariates.” Biometrics, 145-156. (1989)
2.Lachin, J. M., and Foulkes, M. A., “Evaluation of sample size and power for analyses of survival with allowance for nonuniform patient entry, losses to follow-up, noncompliance, and stratification.” Biometrics, 507-519. (1986)
3.Le, C. T., and Lindgren, B. R., “Duration of ventilating tubes: a test for comparing two clustered samples of censored data.” Biometrics, 328-334. (1996)
4.Manatunga, A. K., and Chen, S., “Sample Size Estimation for Survival Outcomes in Cluster‐Randomized Studies with Small Cluster Sizes.” Biometrics, 56(2), 616-621. (2000)
5.Royall, R. M., “Model robust confidence intervals using maximum likelihood estimators.” International Statistical Review/Revue Internationale de Statistique, 221-226. (1986)
6.Su, P. F., “Sample size determination for clustered right-censored data using copula models.” Journal of Statistical Computation and Simulation, 86(4), 696-706. (2016)
7.Yan, J., “Enjoy the joy of copulas: with a package copula.” Journal of Statistical Software, 21(4), 1-21. (2007)