“用R进行结构方程模型分析”的版本间的差异
来自OBHRM百科
Lichaoping（讨论 | 贡献） （→脚本与注释） |
Lichaoping（讨论 | 贡献） |
||
| （未显示同一用户的2个中间版本） | |||
| 第2行： | 第2行： | ||
<pre>library(lavaan) # 调用lavaan包，如果没有安装，需要先安装 | <pre>library(lavaan) # 调用lavaan包，如果没有安装，需要先安装 | ||
curwd=setwd("c:\\program files\\mplus\\mplus examples\\user's guide examples") # 设定数据文件所在的目录为工作目录，如果脚本与数据文件在同一目录，则不需要设定 | curwd=setwd("c:\\program files\\mplus\\mplus examples\\user's guide examples") # 设定数据文件所在的目录为工作目录，如果脚本与数据文件在同一目录，则不需要设定 | ||
| − | semdata <- read.table("ex5.11.dat") | + | semdata <- read.table("ex5.11.dat") # 读取数据文件到cfadata，cfadata是自己命名，可以随便定 |
| − | names(semdata) <-c(paste("y", 1:12, sep="")) | + | names(semdata) <-c(paste("y", 1:12, sep="")) # 给变量命名 |
| − | semmodel <- ' | + | semmodel <- ' # 设置模型，semmodel还是自己命名 |
| − | f1=~ y1+y2+y3 | + | f1=~ y1+y2+y3 # 测量模型，用=~ |
f2=~ y4+y5+y6 | f2=~ y4+y5+y6 | ||
f3=~ y7+y8+y9 | f3=~ y7+y8+y9 | ||
f4=~ y10+y11+y12 | f4=~ y10+y11+y12 | ||
| − | f3 ~ f1+f2 | + | f3 ~ f1+f2 # 路径 |
f4 ~ f3 | f4 ~ f3 | ||
| − | f1 ~~ f2 | + | f1 ~~ f2 # f1与f2相关 |
' | ' | ||
| − | semfit <- sem(semmodel, semdata) | + | semfit <- sem(semmodel, semdata) # 计算，semfit为自己命名，方便后面调用 |
| − | summary(semfit,fit.measures="TRUE") | + | summary(semfit,fit.measures="TRUE") # 显示SEM的总体结果 |
| − | fitMeasures(semfit,fit.measures="all", baseline.model=NULL) | + | fitMeasures(semfit,fit.measures="all", baseline.model=NULL) # 显示所有拟合指数 |
| − | standardizedSolution(semfit) | + | standardizedSolution(semfit) # 显示标准化的结果</pre> |
==结果== | ==结果== | ||
| + | <pre>> summary(semfit,fit.measures="TRUE") # 显示SEM的总体结果 | ||
| + | lavaan (0.5-23.1097) converged normally after 35 iterations | ||
| + | |||
| + | Number of observations 500 | ||
| + | |||
| + | Estimator ML | ||
| + | Minimum Function Test Statistic 53.704 | ||
| + | Degrees of freedom 50 | ||
| + | P-value (Chi-square) 0.334 | ||
| + | |||
| + | Model test baseline model: | ||
| + | |||
| + | Minimum Function Test Statistic 1524.403 | ||
| + | Degrees of freedom 66 | ||
| + | P-value 0.000 | ||
| + | |||
| + | User model versus baseline model: | ||
| + | |||
| + | Comparative Fit Index (CFI) 0.997 | ||
| + | Tucker-Lewis Index (TLI) 0.997 | ||
| + | |||
| + | Loglikelihood and Information Criteria: | ||
| + | |||
| + | Loglikelihood user model (H0) -9646.960 | ||
| + | Loglikelihood unrestricted model (H1) -9620.108 | ||
| + | |||
| + | Number of free parameters 28 | ||
| + | Akaike (AIC) 19349.920 | ||
| + | Bayesian (BIC) 19467.929 | ||
| + | Sample-size adjusted Bayesian (BIC) 19379.056 | ||
| + | |||
| + | Root Mean Square Error of Approximation: | ||
| + | |||
| + | RMSEA 0.012 | ||
| + | 90 Percent Confidence Interval 0.000 0.032 | ||
| + | P-value RMSEA <= 0.05 1.000 | ||
| + | |||
| + | Standardized Root Mean Square Residual: | ||
| + | |||
| + | SRMR 0.029 | ||
| + | |||
| + | Parameter Estimates: | ||
| + | |||
| + | Information Expected | ||
| + | Standard Errors Standard | ||
| + | |||
| + | Latent Variables: | ||
| + | Estimate Std.Err z-value P(>|z|) | ||
| + | f1 =~ | ||
| + | y1 1.000 | ||
| + | y2 1.183 0.102 11.612 0.000 | ||
| + | y3 0.938 0.085 11.065 0.000 | ||
| + | f2 =~ | ||
| + | y4 1.000 | ||
| + | y5 0.870 0.086 10.105 0.000 | ||
| + | y6 0.891 0.089 10.024 0.000 | ||
| + | f3 =~ | ||
| + | y7 1.000 | ||
| + | y8 0.872 0.060 14.570 0.000 | ||
| + | y9 0.882 0.060 14.782 0.000 | ||
| + | f4 =~ | ||
| + | y10 1.000 | ||
| + | y11 0.826 0.096 8.595 0.000 | ||
| + | y12 0.682 0.085 7.975 0.000 | ||
| + | |||
| + | Regressions: | ||
| + | Estimate Std.Err z-value P(>|z|) | ||
| + | f3 ~ | ||
| + | f1 0.563 0.072 7.849 0.000 | ||
| + | f2 0.790 0.086 9.160 0.000 | ||
| + | f4 ~ | ||
| + | f3 0.473 0.057 8.342 0.000 | ||
| + | |||
| + | Covariances: | ||
| + | Estimate Std.Err z-value P(>|z|) | ||
| + | f1 ~~ | ||
| + | f2 -0.030 0.055 -0.545 0.586 | ||
| + | |||
| + | Variances: | ||
| + | Estimate Std.Err z-value P(>|z|) | ||
| + | .y1 1.033 0.092 11.236 0.000 | ||
| + | .y2 0.795 0.101 7.900 0.000 | ||
| + | .y3 1.137 0.093 12.266 0.000 | ||
| + | .y4 1.151 0.104 11.097 0.000 | ||
| + | .y5 0.950 0.083 11.497 0.000 | ||
| + | .y6 1.056 0.090 11.747 0.000 | ||
| + | .y7 0.954 0.088 10.800 0.000 | ||
| + | .y8 0.945 0.079 11.976 0.000 | ||
| + | .y9 0.896 0.077 11.658 0.000 | ||
| + | .y10 1.202 0.118 10.178 0.000 | ||
| + | .y11 0.916 0.085 10.751 0.000 | ||
| + | .y12 1.071 0.083 12.933 0.000 | ||
| + | f1 0.884 0.121 7.311 0.000 | ||
| + | f2 0.888 0.130 6.853 0.000 | ||
| + | .f3 0.550 0.091 6.054 0.000 | ||
| + | .f4 0.555 0.103 5.403 0.000 | ||
| + | |||
| + | > fitMeasures(semfit,fit.measures="all", baseline.model=NULL) # 显示所有拟合指数 | ||
| + | npar fmin chisq | ||
| + | 28.000 0.054 53.704 | ||
| + | df pvalue baseline.chisq | ||
| + | 50.000 0.334 1524.403 | ||
| + | baseline.df baseline.pvalue cfi | ||
| + | 66.000 0.000 0.997 | ||
| + | tli nnfi rfi | ||
| + | 0.997 0.997 0.953 | ||
| + | nfi pnfi ifi | ||
| + | 0.965 0.731 0.997 | ||
| + | rni logl unrestricted.logl | ||
| + | 0.997 -9646.960 -9620.108 | ||
| + | aic bic ntotal | ||
| + | 19349.920 19467.929 500.000 | ||
| + | bic2 rmsea rmsea.ci.lower | ||
| + | 19379.056 0.012 0.000 | ||
| + | rmsea.ci.upper rmsea.pvalue rmr | ||
| + | 0.032 1.000 0.054 | ||
| + | rmr_nomean srmr srmr_bentler | ||
| + | 0.054 0.029 0.029 | ||
| + | srmr_bentler_nomean srmr_bollen srmr_bollen_nomean | ||
| + | 0.029 0.029 0.029 | ||
| + | srmr_mplus srmr_mplus_nomean cn_05 | ||
| + | 0.029 0.029 629.494 | ||
| + | cn_01 gfi agfi | ||
| + | 710.020 0.983 0.973 | ||
| + | pgfi mfi ecvi | ||
| + | 0.630 0.996 0.219 | ||
| + | > standardizedSolution(semfit) # 显示标准化的结果 | ||
| + | lhs op rhs est.std se z pvalue | ||
| + | 1 f1 =~ y1 0.679 0.035 19.612 0.000 | ||
| + | 2 f1 =~ y2 0.780 0.033 23.913 0.000 | ||
| + | 3 f1 =~ y3 0.637 0.036 17.802 0.000 | ||
| + | 4 f2 =~ y4 0.660 0.037 17.694 0.000 | ||
| + | 5 f2 =~ y5 0.644 0.038 17.049 0.000 | ||
| + | 6 f2 =~ y6 0.633 0.038 16.617 0.000 | ||
| + | 7 f3 =~ y7 0.766 0.026 29.561 0.000 | ||
| + | 8 f3 =~ y8 0.723 0.028 25.835 0.000 | ||
| + | 9 f3 =~ y9 0.736 0.027 26.914 0.000 | ||
| + | 10 f4 =~ y10 0.645 0.043 15.068 0.000 | ||
| + | 11 f4 =~ y11 0.625 0.043 14.508 0.000 | ||
| + | 12 f4 =~ y12 0.521 0.045 11.573 0.000 | ||
| + | 13 f3 ~ f1 0.454 0.047 9.663 0.000 | ||
| + | 14 f3 ~ f2 0.639 0.044 14.533 0.000 | ||
| + | 15 f4 ~ f3 0.595 0.049 12.263 0.000 | ||
| + | 16 f1 ~~ f2 -0.034 0.062 -0.546 0.585 | ||
| + | 17 y1 ~~ y1 0.539 0.047 11.464 0.000 | ||
| + | 18 y2 ~~ y2 0.392 0.051 7.695 0.000 | ||
| + | 19 y3 ~~ y3 0.594 0.046 13.026 0.000 | ||
| + | 20 y4 ~~ y4 0.564 0.049 11.467 0.000 | ||
| + | 21 y5 ~~ y5 0.586 0.049 12.057 0.000 | ||
| + | 22 y6 ~~ y6 0.600 0.048 12.453 0.000 | ||
| + | 23 y7 ~~ y7 0.413 0.040 10.385 0.000 | ||
| + | 24 y8 ~~ y8 0.478 0.040 11.827 0.000 | ||
| + | 25 y9 ~~ y9 0.459 0.040 11.404 0.000 | ||
| + | 26 y10 ~~ y10 0.583 0.055 10.548 0.000 | ||
| + | 27 y11 ~~ y11 0.610 0.054 11.343 0.000 | ||
| + | 28 y12 ~~ y12 0.728 0.047 15.522 0.000 | ||
| + | 29 f1 ~~ f1 1.000 0.000 NA NA | ||
| + | 30 f2 ~~ f2 1.000 0.000 NA NA | ||
| + | 31 f3 ~~ f3 0.405 0.056 7.280 0.000 | ||
| + | 32 f4 ~~ f4 0.646 0.058 11.195 0.000</pre> | ||
2017年3月13日 (一) 11:40的最新版本
脚本与注释
library(lavaan) # 调用lavaan包，如果没有安装，需要先安装
curwd=setwd("c:\\program files\\mplus\\mplus examples\\user's guide examples") # 设定数据文件所在的目录为工作目录，如果脚本与数据文件在同一目录，则不需要设定
semdata <- read.table("ex5.11.dat") # 读取数据文件到cfadata，cfadata是自己命名，可以随便定
names(semdata) <-c(paste("y", 1:12, sep="")) # 给变量命名
semmodel <- ' # 设置模型，semmodel还是自己命名
f1=~ y1+y2+y3 # 测量模型，用=~
f2=~ y4+y5+y6
f3=~ y7+y8+y9
f4=~ y10+y11+y12
f3 ~ f1+f2 # 路径
f4 ~ f3
f1 ~~ f2 # f1与f2相关
'
semfit <- sem(semmodel, semdata) # 计算，semfit为自己命名，方便后面调用
summary(semfit,fit.measures="TRUE") # 显示SEM的总体结果
fitMeasures(semfit,fit.measures="all", baseline.model=NULL) # 显示所有拟合指数
standardizedSolution(semfit) # 显示标准化的结果
结果
> summary(semfit,fit.measures="TRUE") # 显示SEM的总体结果
lavaan (0.5-23.1097) converged normally after 35 iterations
Number of observations 500
Estimator ML
Minimum Function Test Statistic 53.704
Degrees of freedom 50
P-value (Chi-square) 0.334
Model test baseline model:
Minimum Function Test Statistic 1524.403
Degrees of freedom 66
P-value 0.000
User model versus baseline model:
Comparative Fit Index (CFI) 0.997
Tucker-Lewis Index (TLI) 0.997
Loglikelihood and Information Criteria:
Loglikelihood user model (H0) -9646.960
Loglikelihood unrestricted model (H1) -9620.108
Number of free parameters 28
Akaike (AIC) 19349.920
Bayesian (BIC) 19467.929
Sample-size adjusted Bayesian (BIC) 19379.056
Root Mean Square Error of Approximation:
RMSEA 0.012
90 Percent Confidence Interval 0.000 0.032
P-value RMSEA <= 0.05 1.000
Standardized Root Mean Square Residual:
SRMR 0.029
Parameter Estimates:
Information Expected
Standard Errors Standard
Latent Variables:
Estimate Std.Err z-value P(>|z|)
f1 =~
y1 1.000
y2 1.183 0.102 11.612 0.000
y3 0.938 0.085 11.065 0.000
f2 =~
y4 1.000
y5 0.870 0.086 10.105 0.000
y6 0.891 0.089 10.024 0.000
f3 =~
y7 1.000
y8 0.872 0.060 14.570 0.000
y9 0.882 0.060 14.782 0.000
f4 =~
y10 1.000
y11 0.826 0.096 8.595 0.000
y12 0.682 0.085 7.975 0.000
Regressions:
Estimate Std.Err z-value P(>|z|)
f3 ~
f1 0.563 0.072 7.849 0.000
f2 0.790 0.086 9.160 0.000
f4 ~
f3 0.473 0.057 8.342 0.000
Covariances:
Estimate Std.Err z-value P(>|z|)
f1 ~~
f2 -0.030 0.055 -0.545 0.586
Variances:
Estimate Std.Err z-value P(>|z|)
.y1 1.033 0.092 11.236 0.000
.y2 0.795 0.101 7.900 0.000
.y3 1.137 0.093 12.266 0.000
.y4 1.151 0.104 11.097 0.000
.y5 0.950 0.083 11.497 0.000
.y6 1.056 0.090 11.747 0.000
.y7 0.954 0.088 10.800 0.000
.y8 0.945 0.079 11.976 0.000
.y9 0.896 0.077 11.658 0.000
.y10 1.202 0.118 10.178 0.000
.y11 0.916 0.085 10.751 0.000
.y12 1.071 0.083 12.933 0.000
f1 0.884 0.121 7.311 0.000
f2 0.888 0.130 6.853 0.000
.f3 0.550 0.091 6.054 0.000
.f4 0.555 0.103 5.403 0.000
> fitMeasures(semfit,fit.measures="all", baseline.model=NULL) # 显示所有拟合指数
npar fmin chisq
28.000 0.054 53.704
df pvalue baseline.chisq
50.000 0.334 1524.403
baseline.df baseline.pvalue cfi
66.000 0.000 0.997
tli nnfi rfi
0.997 0.997 0.953
nfi pnfi ifi
0.965 0.731 0.997
rni logl unrestricted.logl
0.997 -9646.960 -9620.108
aic bic ntotal
19349.920 19467.929 500.000
bic2 rmsea rmsea.ci.lower
19379.056 0.012 0.000
rmsea.ci.upper rmsea.pvalue rmr
0.032 1.000 0.054
rmr_nomean srmr srmr_bentler
0.054 0.029 0.029
srmr_bentler_nomean srmr_bollen srmr_bollen_nomean
0.029 0.029 0.029
srmr_mplus srmr_mplus_nomean cn_05
0.029 0.029 629.494
cn_01 gfi agfi
710.020 0.983 0.973
pgfi mfi ecvi
0.630 0.996 0.219
> standardizedSolution(semfit) # 显示标准化的结果
lhs op rhs est.std se z pvalue
1 f1 =~ y1 0.679 0.035 19.612 0.000
2 f1 =~ y2 0.780 0.033 23.913 0.000
3 f1 =~ y3 0.637 0.036 17.802 0.000
4 f2 =~ y4 0.660 0.037 17.694 0.000
5 f2 =~ y5 0.644 0.038 17.049 0.000
6 f2 =~ y6 0.633 0.038 16.617 0.000
7 f3 =~ y7 0.766 0.026 29.561 0.000
8 f3 =~ y8 0.723 0.028 25.835 0.000
9 f3 =~ y9 0.736 0.027 26.914 0.000
10 f4 =~ y10 0.645 0.043 15.068 0.000
11 f4 =~ y11 0.625 0.043 14.508 0.000
12 f4 =~ y12 0.521 0.045 11.573 0.000
13 f3 ~ f1 0.454 0.047 9.663 0.000
14 f3 ~ f2 0.639 0.044 14.533 0.000
15 f4 ~ f3 0.595 0.049 12.263 0.000
16 f1 ~~ f2 -0.034 0.062 -0.546 0.585
17 y1 ~~ y1 0.539 0.047 11.464 0.000
18 y2 ~~ y2 0.392 0.051 7.695 0.000
19 y3 ~~ y3 0.594 0.046 13.026 0.000
20 y4 ~~ y4 0.564 0.049 11.467 0.000
21 y5 ~~ y5 0.586 0.049 12.057 0.000
22 y6 ~~ y6 0.600 0.048 12.453 0.000
23 y7 ~~ y7 0.413 0.040 10.385 0.000
24 y8 ~~ y8 0.478 0.040 11.827 0.000
25 y9 ~~ y9 0.459 0.040 11.404 0.000
26 y10 ~~ y10 0.583 0.055 10.548 0.000
27 y11 ~~ y11 0.610 0.054 11.343 0.000
28 y12 ~~ y12 0.728 0.047 15.522 0.000
29 f1 ~~ f1 1.000 0.000 NA NA
30 f2 ~~ f2 1.000 0.000 NA NA
31 f3 ~~ f3 0.405 0.056 7.280 0.000
32 f4 ~~ f4 0.646 0.058 11.195 0.000
