在考慮多元資料的母體平均數之比較問題,通常假設資料服從多項分配下進行統計分析。但是當資料具有相關性時,根據此假設所得到的推論是不正確的。 本文將Royall and Tsou(2003)提出的強韌概似函數之方法,應用在多組母體平均數之比較,利用多項分配為實作模型,並且經過適當修正的概似函數,在考慮資料具有相關性情況下的正確之統計推論方法。 The aim of this paper is to extend the robust likelihood technique introduced by Royall and Tsou(2003) to the comparison of seven population means under the correlated multinomial data. More specifically the score test derived on basis of multinomial method is adjusted to become robust, and we also propose the unbiased moment covariance estimators in this thesis.