我也很想知道 有高手能解嗎

是不是还可以这样处理？1）、delete if missing (listwise or pairwise); or:
2）、regress the variable which has missing values with a set of explanatory variables, then impute the missing values using the regression equation.

上面说的是比较简略、甚至有些粗暴的办法；详细的可以参考：
Treatment of Missing Data
By David C. Howell
http://www.uvm.edu/~dhowell/StatPages/More_Stuff/Missing_Data/Missing.html