用item1和item2,item3,item4,item5做两两相关性分析。得到如下分别
|  | item2 
 | item3 
 | item4 
 | item5 | item6 
 | 
| item1 | 0.6722693 | -0.795771 | -0.448187 | 0.4689706 | 0.4734231 | 
|  | item2 
 | item3 
 | item4 
 | item5 
 | item6 | 
| item1 | 0.6722693 | -0.795771 | -0.448187 | 0.4689706 | 0.4734231 | 
我的理解是,item1和item3有负相关。
然后再用item1和2,3,4,5,6一起做回归分析,得到如下结果,却又显示,item1和item5,6强相关,和item3一般相关,请问怎么解释这个现象啊?急
| SUMMARY OUTPUT |  |  |  |  |  |  |  | 
|  |  |  |  |  |  |  |  |  | 
| Regression Statistics |  |  |  |  |  |  |  | 
| Multiple R | 0.9093522 |  |  |  |  |  |  |  | 
| R Square | 0.8269214 |  |  |  |  |  |  |  | 
| Adjusted R Square | 0.7403821 |  |  |  |  |  |  |  | 
| Standard Error | 0.0223087 |  |  |  |  |  |  |  | 
| Observations | 16 |  |  |  |  |  |  |  | 
|  |  |  |  |  |  |  |  |  | 
| ANOVA |  |  |  |  |  |  |  |  | 
|  | df | SS | MS | F | Significance F |  |  |  | 
| Regression | 5 | 0.0237778 | 0.0047556 | 9.5554454 | 0.0014428 |  |  |  | 
| Residual | 10 | 0.0049768 | 0.0004977 |  |  |  |  |  | 
| Total | 15 | 0.0287546 |  |  |  |  |  |  | 
|  |  |  |  |  |  |  |  |  | 
|  | Coefficients | Standard Error | t Stat | P-value | Lower 95% | Upper 95% | Lower 95.0% | Upper 95.0% | 
| Intercept | 0.3926463 | 0.1210249 | 3.2443431 | 0.0088044 | 0.122986 | 0.6623066 | 0.122986 | 0.6623066 | 
| item2 
 | 0.040421 | 0.052* | 0.7627556 | 0.4632221 | -0.077656 | 0.1584975 | -0.077656 | 0.1584975 | 
| item3 
 | -0.194501 | 0.4143798 | -0.469378 | 0.6488665 | -1.117796 | 0.7287951 | -1.117796 | 0.7287951 | 
| item4 
 | -0.044412 | 0.0340965 | -1.302528 | 0.2219337 | -0.120383 | 0.0315601 | -0.120383 | 0.0315601 | 
| item5 
 | 0.0023168 | 0.0010004 | 2.3158239 | 0.0430781 | 8.772E-05 | 0.0045458 | 8.772E-05 | 0.0045458 | 
| item6 
 | 0.3999183 | 0.1705215 | 2.3452654 | 0.0409701 | 0.0199726 | 0.779864 | 0.0199726 | 0.779864 | 
版主cora033:有以下可能性(1) item 2,3,4,5,6之间有较强相关性,模型存在多重共线性。(2)item 1和2,3,4,5,6之间不是线性关系。(3)模型的形式设定不当,可变换模型形式,如双对数模型,半对数模型,双曲线模型等等。(4)变量的形式设定不当,可以通过取对数等形式。