非常感谢大家可以提供帮助,我把问题一一描述一下:
Notes for Model (Default model)
Computation of degrees of freedom (Default model)
| Number of distinct sample moments: | 528 |
| Number of distinct parameters to be estimated: | 85 |
| Degrees of freedom (528 - 85): | 443 |
Result (Default model)
Minimum was achieved
Chi-square = 870.756
Degrees of freedom = 443
Probability level = .000
小于0.05,拒绝了虚无假设,说明模型与数据不匹配,我通过什么方法可以改变这种状况?
删除一些观测变量可以吗.
Estimate
S.E.
C.R.
P
Label
F
<---
E
.683
.111
6.145
***
par_38
F
<---
B
.190
.109
1.748
.080
par_41
F
<---
C
-.010
.096
-.108
.914
par_43
F
<---
D
.035
.171
.206
.837
par_45
F
<---
A
.089
.087
1.024
.306
par_46
H
<---
A
.159
.118
1.340
.180
par_36
H
<---
F
.386
.174
2.219
.026
par_37
H
<---
E
.659
.209
3.159
.002
par_39
H
<---
B
-.084
.149
-.564
.573
par_40
H
<---
C
-.302
.130
-2.329
.020
par_42
H
<---
D
.163
.225
.726
.468
par_44
这里接受的假设只有4个。但是如果我用回归分析的话就会有7个左右。这也是我疑惑的地方。希望坛友解答。下面是一些指标——和模型。
Model
NPAR
CMIN
DF
P
CMIN/DF
Default model
85
870.756
443
.000
1.966
Saturated model
528
.000
0
Independence model
32
12432.423
496
.000
25.065