jswu167 发表于 2013-3-21 21:20 
分析思路:
1、如果3组举例,可以设置1与2,1与3,2与3,分三次卡方检验;
2、P值确定,分为两种,两两比较 ...
请教朋友jswu167 :
1. P值的确定:如果6组,各组与模型组比较一遍——即5次,那P=0.05/5吗?各组间比较一遍——即15次,那P=0.05/15? 怎么感觉怪怪的,应该都按P=0.05/15进行吧,当进行任两组间比较时都默认被拆成了15次,只是根据研究目的选不选用到比较对象中来而已啊?
2. 如理论频数存在0怎么办?直接用Fisher's 确切概率法吗?
但会出现:
>Warning # 3211
>On at least one case, the value of the weight variable was zero, negative,
>or missing. Such cases are invisible to statistical procedures and graphs
>which need positively weighted cases, but remain on the file and are
>processed by non-statistical facilities such as LIST and SAVE.
不知道怎么办啊……
3. 选用 Fisher's 确切概率法 没有卡方值怎么回事啊?(n=21 <40,所以用Fisher's, 但Value那一项没有值……)
组别 * D5妊娠情况 Crosstabulation
D5妊娠情况 Total
妊娠 未妊娠 妊娠
组别 正常组 Count 10 1 11
% within 组别 90.9% 9.1% 100.0%
模型组 Count 7 3 10
% within 组别 70.0% 30.0% 100.0%
Total Count 17 4 21
% within 组别 81.0% 19.0% 100.0%
Chi-Square Tests
Value df Asymp. Sig. (2-sided) Exact Sig. (2-sided) Exact Sig. (1-sided) Point Probability
Pearson Chi-Square 1.485(b) 1 .223 .311 .256
Continuity Correction(a) .439 1 .508
Likelihood Ratio 1.531 1 .216 .311 .256
Fisher's Exact Test .311 .256
Linear-by-Linear Association 1.414(c) 1 .234 .311 .256 .221
N of Valid Cases 21
a Computed only for a 2x2 table
b 2 cells (50.0%) have expected count less than 5. The minimum expected count is 1.90.
c The standardized statistic is 1.189.
恳请各位大侠帮忙啊!谢谢!!