sem
mugo
2 January 2024
#knitr::opts_chunk$set(echo = TRUE)
library(readxl)
library(lavaan)## This is lavaan 0.6-17
## lavaan is FREE software! Please report any bugs.library(psych)##
## Attaching package: 'psych'## The following object is masked from 'package:lavaan':
##
## cor2covlibrary(ggplot2)##
## Attaching package: 'ggplot2'## The following objects are masked from 'package:psych':
##
## %+%, alphalibrary(MVN)
library(semPlot)
library(semTools)## ## ################################################################################# This is semTools 0.5-6## All users of R (or SEM) are invited to submit functions or ideas for functions.## #################################################################################
## Attaching package: 'semTools'## The following objects are masked from 'package:psych':
##
## reliability, skewlibrary(GPArotation)##
## Attaching package: 'GPArotation'## The following objects are masked from 'package:psych':
##
## equamax, variminsem <- read_excel("C:/Users/USER/Desktop/sem/plsem/sem.xlsx")
head(sem)explore the data
structural model
subsetn<-data.frame(subset(sem,select = c(BT1,BT2,BT3,BT4,BT5,BT6,BT7,BD1,BD2,BD3,BD4,EMP1,EMP2,EMP3,EMP4,EMP5,CS1,CS2,CS3,EP1,EP2,EP3,EP4,LS1,LS2,LS3,LS4,LS5)))normality test
mnormal<-mvn(subsetn)
mnormal## $multivariateNormality
## Test HZ p value MVN
## 1 Henze-Zirkler 1.026133 0 NO
##
## $univariateNormality
## Test Variable Statistic p value Normality
## 1 Anderson-Darling BT1 7.2262 <0.001 NO
## 2 Anderson-Darling BT2 8.0671 <0.001 NO
## 3 Anderson-Darling BT3 6.6800 <0.001 NO
## 4 Anderson-Darling BT4 6.4183 <0.001 NO
## 5 Anderson-Darling BT5 6.1728 <0.001 NO
## 6 Anderson-Darling BT6 8.6698 <0.001 NO
## 7 Anderson-Darling BT7 7.5770 <0.001 NO
## 8 Anderson-Darling BD1 5.9732 <0.001 NO
## 9 Anderson-Darling BD2 8.7052 <0.001 NO
## 10 Anderson-Darling BD3 5.9169 <0.001 NO
## 11 Anderson-Darling BD4 6.8989 <0.001 NO
## 12 Anderson-Darling EMP1 5.2108 <0.001 NO
## 13 Anderson-Darling EMP2 6.4303 <0.001 NO
## 14 Anderson-Darling EMP3 4.7131 <0.001 NO
## 15 Anderson-Darling EMP4 7.8380 <0.001 NO
## 16 Anderson-Darling EMP5 5.5762 <0.001 NO
## 17 Anderson-Darling CS1 6.9341 <0.001 NO
## 18 Anderson-Darling CS2 7.1469 <0.001 NO
## 19 Anderson-Darling CS3 6.2791 <0.001 NO
## 20 Anderson-Darling EP1 7.5913 <0.001 NO
## 21 Anderson-Darling EP2 7.3530 <0.001 NO
## 22 Anderson-Darling EP3 7.2375 <0.001 NO
## 23 Anderson-Darling EP4 6.5793 <0.001 NO
## 24 Anderson-Darling LS1 9.0564 <0.001 NO
## 25 Anderson-Darling LS2 8.2974 <0.001 NO
## 26 Anderson-Darling LS3 8.5914 <0.001 NO
## 27 Anderson-Darling LS4 6.9281 <0.001 NO
## 28 Anderson-Darling LS5 5.8779 <0.001 NO
##
## $Descriptives
## n Mean Std.Dev Median Min Max 25th 75th Skew Kurtosis
## BT1 272 4.897059 1.307257 5 1 7 4 6.00 -0.41254431 -0.14823904
## BT2 272 4.871324 1.342598 5 1 7 4 6.00 -0.54057483 -0.13821529
## BT3 272 4.540441 1.550568 5 1 7 3 6.00 -0.24685876 -0.87882781
## BT4 272 4.757353 1.371811 5 1 7 4 6.00 -0.17498923 -0.57321521
## BT5 272 4.547794 1.416011 4 1 7 4 6.00 -0.10448505 -0.57213063
## BT6 272 5.150735 1.435990 5 1 7 4 6.00 -0.54708461 -0.43868563
## BT7 272 5.117647 1.500597 5 1 7 4 6.00 -0.62433083 -0.17711402
## BD1 272 4.588235 1.505074 5 1 7 4 6.00 -0.40835281 -0.38347131
## BD2 272 5.113971 1.418729 5 1 7 4 6.00 -0.64102981 -0.21095215
## BD3 272 4.514706 1.487886 5 1 7 3 6.00 -0.22263689 -0.68504836
## BD4 272 4.654412 1.341248 5 1 7 4 6.00 -0.11478964 -0.35230370
## EMP1 272 4.246324 1.628275 4 1 7 3 6.00 -0.24527887 -0.74671013
## EMP2 272 4.647059 1.447886 5 1 7 4 6.00 -0.41969844 -0.33308646
## EMP3 272 4.069853 1.687454 4 1 7 3 5.00 -0.05417392 -0.88787302
## EMP4 272 3.966912 1.345626 4 1 7 3 5.00 -0.18450837 -0.17021913
## EMP5 272 4.334559 1.476201 4 1 7 3 5.00 -0.20104328 -0.56563007
## CS1 272 4.448529 1.335410 4 1 7 4 5.00 -0.17728686 -0.06634707
## CS2 272 4.415441 1.300092 4 1 7 4 5.00 -0.17631873 -0.05861883
## CS3 272 4.426471 1.427884 4 1 7 4 5.25 -0.28794302 -0.24502398
## EP1 272 5.183824 1.388951 5 1 7 4 6.00 -0.43631709 -0.51113542
## EP2 272 5.106618 1.369014 5 2 7 4 6.00 -0.27766985 -0.78219332
## EP3 272 5.091912 1.388809 5 1 7 4 6.00 -0.33672045 -0.53268446
## EP4 272 4.988971 1.441308 5 1 7 4 6.00 -0.32707653 -0.53720876
## LS1 272 4.738971 1.239669 5 1 7 4 6.00 -0.60988129 0.29634604
## LS2 272 4.974265 1.249088 5 1 7 4 6.00 -0.48360113 0.04311961
## LS3 272 5.297794 1.309740 5 2 7 4 6.00 -0.54899072 -0.32742495
## LS4 272 4.941176 1.461762 5 1 7 4 6.00 -0.49879110 -0.37148784
## LS5 272 4.639706 1.554296 5 1 7 4 6.00 -0.38345201 -0.56146435correlation matrix
cmat<-round(cor(subsetn),2)
cmat[upper.tri(cmat)]<-""
cmat<-as.data.frame(cmat)
cmatnumber of factors
parallel<-fa.parallel(subsetn,fm="ml",fa="fa")
## Parallel analysis suggests that the number of factors = 7 and the number of components = NAparallel$fa.values## [1] 9.110306807 2.208726061 1.491265413 1.309686468 0.768926591
## [6] 0.526673413 0.314718465 -0.007758516 -0.052360241 -0.059610721
## [11] -0.091891890 -0.179201766 -0.228243066 -0.242869845 -0.263162154
## [16] -0.298329457 -0.307033491 -0.317601216 -0.329593839 -0.350375636
## [21] -0.372287692 -0.381583472 -0.413413631 -0.449660305 -0.504054385
## [26] -0.543895350 -0.589671435 -0.681460711sixfactor=fa(subsetn,nfactors = 6,rotate = "oblimin",fm="ml")
summary(sixfactor)##
## Factor analysis with Call: fa(r = subsetn, nfactors = 6, rotate = "oblimin", fm = "ml")
##
## Test of the hypothesis that 6 factors are sufficient.
## The degrees of freedom for the model is 225 and the objective function was 1.51
## The number of observations was 272 with Chi Square = 388.76 with prob < 7.2e-11
##
## The root mean square of the residuals (RMSA) is 0.03
## The df corrected root mean square of the residuals is 0.04
##
## Tucker Lewis Index of factoring reliability = 0.939
## RMSEA index = 0.052 and the 10 % confidence intervals are 0.043 0.06
## BIC = -872.55
## With factor correlations of
## ML3 ML1 ML6 ML2 ML4 ML5
## ML3 1.00 0.27 0.29 0.28 0.23 0.19
## ML1 0.27 1.00 0.43 0.46 0.37 0.31
## ML6 0.29 0.43 1.00 0.48 0.56 0.29
## ML2 0.28 0.46 0.48 1.00 0.51 0.49
## ML4 0.23 0.37 0.56 0.51 1.00 0.48
## ML5 0.19 0.31 0.29 0.49 0.48 1.00print(sixfactor,cut=.3,digits = 3,sort=T)## Factor Analysis using method = ml
## Call: fa(r = subsetn, nfactors = 6, rotate = "oblimin", fm = "ml")
## Standardized loadings (pattern matrix) based upon correlation matrix
## item ML3 ML1 ML6 ML2 ML4 ML5 h2 u2 com
## LS2 25 0.927 0.857 0.1430 1.01
## LS1 24 0.850 0.732 0.2675 1.01
## LS3 26 0.836 0.740 0.2601 1.06
## LS4 27 0.809 0.653 0.3471 1.04
## LS5 28 0.580 0.365 0.6354 1.04
## EP2 21 0.960 0.957 0.0433 1.01
## EP1 20 0.949 0.845 0.1545 1.01
## EP4 23 0.707 0.622 0.3777 1.06
## EP3 22 0.657 0.638 0.3617 1.21
## BT5 5 0.721 0.545 0.4551 1.08
## BT1 1 0.697 0.517 0.4830 1.03
## BT2 2 0.678 0.515 0.4855 1.06
## BT6 6 0.586 0.482 0.5183 1.08
## BT7 7 0.527 0.479 0.5214 1.36
## BT4 4 0.493 0.460 0.5399 1.41
## BT3 3 0.444 0.336 0.6636 1.71
## CS2 18 0.940 0.915 0.0853 1.00
## CS3 19 0.887 0.797 0.2033 1.01
## CS1 17 0.882 0.826 0.1741 1.01
## BD1 8 0.792 0.623 0.3766 1.02
## BD3 10 0.766 0.643 0.3570 1.01
## BD2 9 0.676 0.578 0.4215 1.11
## BD4 11 0.652 0.577 0.4225 1.07
## EMP1 12 0.722 0.534 0.4658 1.03
## EMP4 15 0.698 0.580 0.4203 1.10
## EMP3 14 0.686 0.435 0.5653 1.03
## EMP2 13 0.637 0.555 0.4455 1.16
## EMP5 16 0.411 0.483 0.5173 1.99
##
## ML3 ML1 ML6 ML2 ML4 ML5
## SS loadings 3.409 3.123 2.989 2.816 2.624 2.327
## Proportion Var 0.122 0.112 0.107 0.101 0.094 0.083
## Cumulative Var 0.122 0.233 0.340 0.441 0.534 0.617
## Proportion Explained 0.197 0.181 0.173 0.163 0.152 0.135
## Cumulative Proportion 0.197 0.378 0.551 0.714 0.865 1.000
##
## With factor correlations of
## ML3 ML1 ML6 ML2 ML4 ML5
## ML3 1.000 0.272 0.289 0.279 0.234 0.186
## ML1 0.272 1.000 0.433 0.458 0.370 0.308
## ML6 0.289 0.433 1.000 0.480 0.561 0.291
## ML2 0.279 0.458 0.480 1.000 0.514 0.491
## ML4 0.234 0.370 0.561 0.514 1.000 0.482
## ML5 0.186 0.308 0.291 0.491 0.482 1.000
##
## Mean item complexity = 1.1
## Test of the hypothesis that 6 factors are sufficient.
##
## df null model = 378 with the objective function = 19.082 with Chi Square = 4977.246
## df of the model are 225 and the objective function was 1.514
##
## The root mean square of the residuals (RMSR) is 0.029
## The df corrected root mean square of the residuals is 0.037
##
## The harmonic n.obs is 272 with the empirical chi square 168.285 with prob < 0.998
## The total n.obs was 272 with Likelihood Chi Square = 388.758 with prob < 7.24e-11
##
## Tucker Lewis Index of factoring reliability = 0.9392
## RMSEA index = 0.0516 and the 90 % confidence intervals are 0.043 0.0604
## BIC = -872.548
## Fit based upon off diagonal values = 0.993
## Measures of factor score adequacy
## ML3 ML1 ML6 ML2 ML4
## Correlation of (regression) scores with factors 0.966 0.984 0.924 0.976 0.932
## Multiple R square of scores with factors 0.934 0.968 0.855 0.952 0.868
## Minimum correlation of possible factor scores 0.867 0.936 0.709 0.903 0.736
## ML5
## Correlation of (regression) scores with factors 0.912
## Multiple R square of scores with factors 0.831
## Minimum correlation of possible factor scores 0.663plot(sixfactor)
cor.plot(subsetn)
cronchbar
cfa model
cfamodel1<-'BT=~BT1+BT2+BT3+BT4+BT5+BT6+BT7
BD=~BD1+BD2+BD3+BD4
EMP=~EMP1+EMP2+EMP3+EMP4+EMP5
CS=~CS1+CS2+CS3
EP=~EP1+EP2+EP3+EP4
LS=~LS1+LS2+LS3+LS4+LS5
'fit model
fit.cfamodel1<-cfa(cfamodel1,data=subsetn,estimator="MLR",mimic="Mplus")
summary(fit.cfamodel1,fit.measures=T,standardized=T,rsq=T)## lavaan 0.6.17 ended normally after 50 iterations
##
## Estimator ML
## Optimization method NLMINB
## Number of model parameters 99
##
## Number of observations 272
## Number of missing patterns 1
##
## Model Test User Model:
## Standard Scaled
## Test Statistic 603.531 534.890
## Degrees of freedom 335 335
## P-value (Chi-square) 0.000 0.000
## Scaling correction factor 1.128
## Yuan-Bentler correction (Mplus variant)
##
## Model Test Baseline Model:
##
## Test statistic 5190.329 4254.558
## Degrees of freedom 378 378
## P-value 0.000 0.000
## Scaling correction factor 1.220
##
## User Model versus Baseline Model:
##
## Comparative Fit Index (CFI) 0.944 0.948
## Tucker-Lewis Index (TLI) 0.937 0.942
##
## Robust Comparative Fit Index (CFI) 0.954
## Robust Tucker-Lewis Index (TLI) 0.949
##
## Loglikelihood and Information Criteria:
##
## Loglikelihood user model (H0) -11143.554 -11143.554
## Scaling correction factor 1.353
## for the MLR correction
## Loglikelihood unrestricted model (H1) -10841.788 -10841.788
## Scaling correction factor 1.180
## for the MLR correction
##
## Akaike (AIC) 22485.107 22485.107
## Bayesian (BIC) 22842.082 22842.082
## Sample-size adjusted Bayesian (SABIC) 22528.179 22528.179
##
## Root Mean Square Error of Approximation:
##
## RMSEA 0.054 0.047
## 90 Percent confidence interval - lower 0.047 0.040
## 90 Percent confidence interval - upper 0.061 0.054
## P-value H_0: RMSEA <= 0.050 0.153 0.769
## P-value H_0: RMSEA >= 0.080 0.000 0.000
##
## Robust RMSEA 0.049
## 90 Percent confidence interval - lower 0.041
## 90 Percent confidence interval - upper 0.057
## P-value H_0: Robust RMSEA <= 0.050 0.591
## P-value H_0: Robust RMSEA >= 0.080 0.000
##
## Standardized Root Mean Square Residual:
##
## SRMR 0.053 0.053
##
## Parameter Estimates:
##
## Standard errors Sandwich
## Information bread Observed
## Observed information based on Hessian
##
## Latent Variables:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## BT =~
## BT1 1.000 0.916 0.702
## BT2 1.030 0.078 13.284 0.000 0.943 0.704
## BT3 0.929 0.124 7.503 0.000 0.851 0.550
## BT4 0.999 0.099 10.129 0.000 0.915 0.668
## BT5 1.077 0.102 10.549 0.000 0.986 0.698
## BT6 1.091 0.116 9.440 0.000 0.999 0.697
## BT7 1.115 0.122 9.153 0.000 1.021 0.682
## BD =~
## BD1 1.000 1.149 0.765
## BD2 0.936 0.071 13.212 0.000 1.076 0.760
## BD3 1.037 0.065 15.921 0.000 1.192 0.802
## BD4 0.891 0.068 13.194 0.000 1.024 0.765
## EMP =~
## EMP1 1.000 1.145 0.705
## EMP2 0.923 0.078 11.872 0.000 1.057 0.732
## EMP3 0.922 0.090 10.192 0.000 1.056 0.627
## EMP4 0.867 0.081 10.699 0.000 0.993 0.739
## EMP5 0.877 0.093 9.425 0.000 1.004 0.681
## CS =~
## CS1 1.000 1.212 0.909
## CS2 1.023 0.040 25.698 0.000 1.240 0.955
## CS3 1.045 0.043 24.111 0.000 1.267 0.889
## EP =~
## EP1 1.000 1.260 0.909
## EP2 1.065 0.035 30.561 0.000 1.342 0.982
## EP3 0.851 0.061 13.883 0.000 1.073 0.774
## EP4 0.893 0.050 17.773 0.000 1.125 0.782
## LS =~
## LS1 1.000 1.058 0.855
## LS2 1.091 0.055 19.794 0.000 1.155 0.926
## LS3 1.051 0.096 10.996 0.000 1.112 0.851
## LS4 1.102 0.069 15.904 0.000 1.166 0.799
## LS5 0.875 0.090 9.675 0.000 0.926 0.597
##
## Covariances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## BT ~~
## BD 0.721 0.101 7.171 0.000 0.685 0.685
## EMP 0.489 0.080 6.154 0.000 0.466 0.466
## CS 0.626 0.093 6.747 0.000 0.564 0.564
## EP 0.621 0.095 6.531 0.000 0.538 0.538
## LS 0.345 0.071 4.842 0.000 0.356 0.356
## BD ~~
## EMP 0.802 0.112 7.167 0.000 0.609 0.609
## CS 0.791 0.112 7.060 0.000 0.568 0.568
## EP 0.643 0.106 6.080 0.000 0.444 0.444
## LS 0.345 0.088 3.921 0.000 0.283 0.283
## EMP ~~
## CS 0.831 0.114 7.282 0.000 0.599 0.599
## EP 0.596 0.099 5.998 0.000 0.413 0.413
## LS 0.317 0.098 3.238 0.001 0.261 0.261
## CS ~~
## EP 0.733 0.109 6.743 0.000 0.480 0.480
## LS 0.385 0.077 5.023 0.000 0.300 0.300
## EP ~~
## LS 0.368 0.094 3.919 0.000 0.276 0.276
##
## Intercepts:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .BT1 4.897 0.079 61.895 0.000 4.897 3.753
## .BT2 4.871 0.081 59.949 0.000 4.871 3.635
## .BT3 4.540 0.094 48.383 0.000 4.540 2.934
## .BT4 4.757 0.083 57.300 0.000 4.757 3.474
## .BT5 4.548 0.086 53.066 0.000 4.548 3.218
## .BT6 5.151 0.087 59.266 0.000 5.151 3.594
## .BT7 5.118 0.091 56.350 0.000 5.118 3.417
## .BD1 4.588 0.091 50.370 0.000 4.588 3.054
## .BD2 5.114 0.086 59.558 0.000 5.114 3.611
## .BD3 4.515 0.090 50.135 0.000 4.515 3.040
## .BD4 4.654 0.081 57.338 0.000 4.654 3.477
## .EMP1 4.246 0.099 43.089 0.000 4.246 2.613
## .EMP2 4.647 0.088 53.031 0.000 4.647 3.215
## .EMP3 4.070 0.102 39.850 0.000 4.070 2.416
## .EMP4 3.967 0.081 48.709 0.000 3.967 2.953
## .EMP5 4.335 0.089 48.516 0.000 4.335 2.942
## .CS1 4.449 0.081 55.041 0.000 4.449 3.337
## .CS2 4.415 0.079 56.116 0.000 4.415 3.403
## .CS3 4.426 0.086 51.221 0.000 4.426 3.106
## .EP1 5.184 0.084 61.666 0.000 5.184 3.739
## .EP2 5.107 0.083 61.632 0.000 5.107 3.737
## .EP3 5.092 0.084 60.579 0.000 5.092 3.673
## .EP4 4.989 0.087 57.192 0.000 4.989 3.468
## .LS1 4.739 0.075 63.163 0.000 4.739 3.830
## .LS2 4.974 0.076 65.799 0.000 4.974 3.990
## .LS3 5.298 0.079 66.833 0.000 5.298 4.052
## .LS4 4.941 0.088 55.852 0.000 4.941 3.387
## .LS5 4.640 0.094 49.322 0.000 4.640 2.991
##
## Variances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .BT1 0.864 0.119 7.243 0.000 0.864 0.507
## .BT2 0.906 0.107 8.481 0.000 0.906 0.505
## .BT3 1.672 0.158 10.572 0.000 1.672 0.698
## .BT4 1.038 0.123 8.411 0.000 1.038 0.554
## .BT5 1.025 0.120 8.526 0.000 1.025 0.513
## .BT6 1.057 0.118 8.994 0.000 1.057 0.514
## .BT7 1.201 0.137 8.771 0.000 1.201 0.535
## .BD1 0.936 0.113 8.252 0.000 0.936 0.415
## .BD2 0.848 0.116 7.336 0.000 0.848 0.423
## .BD3 0.786 0.102 7.682 0.000 0.786 0.356
## .BD4 0.744 0.095 7.801 0.000 0.744 0.415
## .EMP1 1.330 0.146 9.123 0.000 1.330 0.503
## .EMP2 0.971 0.115 8.426 0.000 0.971 0.465
## .EMP3 1.723 0.180 9.572 0.000 1.723 0.607
## .EMP4 0.818 0.099 8.301 0.000 0.818 0.454
## .EMP5 1.163 0.142 8.185 0.000 1.163 0.536
## .CS1 0.307 0.078 3.949 0.000 0.307 0.173
## .CS2 0.147 0.041 3.629 0.000 0.147 0.088
## .CS3 0.425 0.101 4.203 0.000 0.425 0.209
## .EP1 0.334 0.059 5.637 0.000 0.334 0.174
## .EP2 0.067 0.043 1.585 0.113 0.067 0.036
## .EP3 0.771 0.142 5.414 0.000 0.771 0.401
## .EP4 0.803 0.152 5.299 0.000 0.803 0.388
## .LS1 0.411 0.078 5.273 0.000 0.411 0.268
## .LS2 0.222 0.049 4.560 0.000 0.222 0.143
## .LS3 0.472 0.125 3.789 0.000 0.472 0.276
## .LS4 0.770 0.098 7.816 0.000 0.770 0.362
## .LS5 1.549 0.180 8.583 0.000 1.549 0.644
## BT 0.839 0.132 6.365 0.000 1.000 1.000
## BD 1.321 0.175 7.561 0.000 1.000 1.000
## EMP 1.312 0.188 6.986 0.000 1.000 1.000
## CS 1.469 0.159 9.261 0.000 1.000 1.000
## EP 1.588 0.138 11.541 0.000 1.000 1.000
## LS 1.120 0.149 7.520 0.000 1.000 1.000
##
## R-Square:
## Estimate
## BT1 0.493
## BT2 0.495
## BT3 0.302
## BT4 0.446
## BT5 0.487
## BT6 0.486
## BT7 0.465
## BD1 0.585
## BD2 0.577
## BD3 0.644
## BD4 0.585
## EMP1 0.497
## EMP2 0.535
## EMP3 0.393
## EMP4 0.546
## EMP5 0.464
## CS1 0.827
## CS2 0.912
## CS3 0.791
## EP1 0.826
## EP2 0.964
## EP3 0.599
## EP4 0.612
## LS1 0.732
## LS2 0.857
## LS3 0.724
## LS4 0.638
## LS5 0.356predict
pred<-predict(fit.cfamodel1)
head(pred)## BT BD EMP CS EP LS
## [1,] 0.4986380 0.9518435 -0.81464907 -0.4132022 -1.03434255 -3.1463068
## [2,] -0.6071380 -0.4902541 -0.38141687 -1.3082481 -0.13072573 -0.8675342
## [3,] -1.9079688 -1.7512811 -1.39078648 -1.2012873 -2.80567036 -0.9558520
## [4,] 1.4453643 -0.1115977 -0.55667130 0.3458193 0.01982881 -1.5611947
## [5,] 0.8128071 0.3573468 -0.03157638 0.1281985 -0.72243830 0.7326089
## [6,] -0.2078168 -0.3213221 -0.69389260 -1.3334496 -2.10653603 -0.9947009inspect
inspect(fit.cfamodel1,"cor.lv")## BT BD EMP CS EP LS
## BT 1.000
## BD 0.685 1.000
## EMP 0.466 0.609 1.000
## CS 0.564 0.568 0.599 1.000
## EP 0.538 0.444 0.413 0.480 1.000
## LS 0.356 0.283 0.261 0.300 0.276 1.000discriminant validity
condisc<-function(x){std.loadings<-inspect(x,what = "std")$lambda}
condisc## function(x){std.loadings<-inspect(x,what = "std")$lambda}model
model1<-'BT=~BT1+BT2+BT3+BT4+BT5+BT6+BT7
BD=~BD1+BD2+BD3+BD4
EMP=~EMP1+EMP2+EMP3+EMP4+EMP5
CS=~CS1+CS2+CS3
EP=~EP1+EP2+EP3+EP4
LS=~LS1+LS2+LS3+LS4+LS5
#regression
CS=~LS
CS=~BT+BD+EMP+EP
'fit.semmodel<-sem(model1,data=subsetn,estimator="MLR",mimic="Mplus")summary
summary<-summary(fit.semmodel,fit.measures=T,standardized=T,rsq=T)semPaths(fit.semmodel, whatLabels = "est", style = "lisrel", intercepts = FALSE)
#library(seminr)
#measurement model
#mm<-constructs(composite("BD",multi_items("BT",1:7)),composite("BD",multi_items("BD",1:4)),composite("CS",multi_items("CS",1:3)))
#sm<-relationships(paths(from = c("BT","BD"),to=c("CS")))
#estimate<-estimate_pls(data=sem,measurement_model = mm,inner_weights=path_weighting,structural_model = sm,missing = mean_replacement,missing_value = "-99")
#library(restriktor)
# lavaan syntax
#model.lav <- "y1 ~ 1 + a*x1 + b*x2 + c*x3"
#fit.lav <- sem(model.lav, data = PoliticalDemocracy)
# restriktor syntax
#fit.restr <- goric(fit.lav, hypotheses = list(h1 = 'a > b > c'), comparison = #"complement")
#fit.restr
#h1 <- 'a1 > a2 > a3; b1 > b2 > b3'
#model <- '
#y1 ~ 1 + a1*x1 + b1*x2
#y2 ~ 1 + a2*x1 + b2*x2
#y3 ~ 1 + a3*x1 + b3*x2
#'
#iht(model = model, data = PoliticalDemocracy, constraints = h1,
#R = 1000, double.bootstrap = "no", parallel = "snow", ncpus = 8)
#The advantage of the goric(a) is that you can directly evaluate your order-restricted hypothesis against its complement (i.e., not h1).
#fit.lav <- sem(model, data = PoliticalDemocracy)
#goric(fit.lav, hypotheses = list(h1 = h1), comparison = "complement")
#From the results you can conclude that hypothesis h1 is 'some value' more supported than its complement.
#See the Vignette in restriktor for "Guidelines interpretation GORIC(A) output" or feel free to contact me.