在生物統計學或其他的領域中，常常會接觸到具有相關性的二元資料，這些相關性可能是由基因、環境、重複測量或是時間造成的。而分析二元資料時，大部份的情形都是使用邏輯斯迴歸模型。 Tsou (2006a) 提出分析具有相關性之二元資料的強韌概似函數法。此強韌分析法，在大樣本之情形下，即使二元資料彼此有相關或分配不是二項分配時，也能對迴歸參數作正確之推論。 本文之目的是將該文章中所計算出的修正項，在不同的架構之下做進一步的推算，以釐清修正項與自變量及相關性間的關係。 Correlated data are commonly encountered in biostatistics and other research fields. The correlation may come from the genetic heredity, familial aggregation, environmental heterogeneity, or repeated measures. Tsou(2006a) proposed a parametric robust likelihood technique that provides parametric robust inferences for correlated binary data. Specifically, Tsou(2006a) modified the binomial working model to become robust. With large samples the adjusted binomial likelihood is asymptotically legitimate for correlated binary data. In this work, the adjustments that corrected the binomial working model are inspected in details to reveal how the covariates affect the magnitude of adjustments.
