#Reading in the data
library(haven)
## Warning: package 'haven' was built under R version 4.3.3
csek12 <- as.data.frame(read_sav("RegRep_K12 Sex Ed_De-Id Data_Clean_11-24-23.sav"))
#Making variables nominal
csek12$Vote <- factor(csek12$Vote, levels = c(1:3))
csek12$Gender_3cat <- factor(csek12$Gender_3cat, levels = c(1:3))
csek12$Gender_5cat <- factor(csek12$Gender_5cat, levels = c(1:5))
csek12$Race_cat <- factor(csek12$Race_cat, levels = c(1:7))
csek12$SO <- factor(csek12$SO, levels = c(1:6))
csek12$Geo <- factor(csek12$Geo, levels = c(1:4))
csek12$Parent <- factor(csek12$Parent, levels = c(1:3))
#setting "I don't know" responses to NA
csek12$PolAff <- ifelse(csek12$Politic == 8, NA, csek12$Politic)
#setting "I would abstain (not vote)" responses to NA
csek12$Vote1 <- ifelse(csek12$Vote == 3, NA, csek12$Vote)
CSEK12_Items <- as.data.frame(csek12[,c(5:13, 15:36, 38:45)])
csek12$highered <- ifelse(csek12$Edu > 5, 1, 0)
csek12one <- '
CSE =~ Q2_1 + Q2_2 + Q2_3 + Q2_4 + Q2_5 + Q2_6 + Q2_7 + Q2_8 + Q2_9 + Q3_1 + Q3_2 + Q3_3 + Q3_4 + Q3_5 + Q3_6 + Q3_7 + Q3_8 + Q3_9 + Q3_10 + Q4_1 + Q4_2 + Q4_3 + Q4_4 + Q4_5 + Q4_6 + Q4_7 + Q4_8 + Q4_9 + Q4_10 + Q5_1 + Q5_2 + Q5_4 + Q5_5 + Q5_6 + Q5_7 + Q5_8 + Q5_9 + Q5_10 + Q5_11'
#outliers for one factor model
gcd.cfa.one <- faoutlier::gCD(CSEK12_Items, csek12one)
gcd.df <- as.data.frame(print(gcd.cfa.one, ncases = 75))
## gCD
## 23 20.8330234
## 162 19.6241643
## 89 18.1736515
## 135 9.4883661
## 278 7.7142933
## 18 7.4100895
## 282 7.0051095
## 64 6.5176744
## 250 6.1831421
## 111 5.7671950
## 117 5.5295826
## 269 4.5024798
## 169 4.4364703
## 101 4.2983051
## 60 3.8662917
## 284 3.6514077
## 152 3.5844975
## 290 3.4134331
## 93 3.3317468
## 222 3.1529251
## 272 3.1196155
## 271 2.9078272
## 297 2.8088492
## 73 2.6349612
## 283 2.5234844
## 3 2.4651199
## 29 2.3244764
## 154 2.3146595
## 134 2.3130229
## 146 2.2926201
## 286 2.1304443
## 199 2.1091443
## 241 2.0634007
## 274 2.0118393
## 262 2.0089802
## 80 1.9815271
## 185 1.9368994
## 295 1.7910419
## 175 1.7086479
## 53 1.5969983
## 178 1.5650120
## 202 1.5167497
## 147 1.4703360
## 238 1.3888787
## 50 1.3717834
## 240 1.3439534
## 99 1.3137057
## 276 1.2982603
## 54 1.2923161
## 55 1.2923161
## 153 1.2923161
## 148 1.2817452
## 296 1.2361155
## 223 1.2295029
## 17 1.1779550
## 125 1.1777589
## 144 1.1762872
## 128 1.1644040
## 252 1.1593186
## 233 1.1473196
## 122 1.1357495
## 214 1.0916477
## 264 1.0822014
## 232 1.0510886
## 66 0.9741224
## 208 0.9695800
## 110 0.9620269
## 88 0.9578064
## 235 0.9456809
## 167 0.9072802
## 251 0.9057366
## 279 0.8889584
## 217 0.8590883
## 257 0.8537280
## 226 0.8525605
plot(gcd.cfa.one)
dfb.cfa.one <- as.data.frame(gcd.cfa.one$dfbetas)
library(writexl)
## Warning: package 'writexl' was built under R version 4.3.3
write_xlsx(dfb.cfa.one, "dbf.cfa.one.xlsx")
library(rstatix)
## Warning: package 'rstatix' was built under R version 4.3.2
##
## Attaching package: 'rstatix'
## The following object is masked from 'package:stats':
##
## filter
md.cfa.one <- faoutlier::robustMD(na.omit(CSEK12_Items, csek12one, method = "mve")) #issue with IQR; Q4_8 has an IQR of 0 (first q, median, third q and max all 5); use alternative function
## Error in cov.rob(data, method = method, ...): at least one column has IQR 0
md.cfa.one.1 <- mahalanobis_distance(CSEK12_Items)
md.cfa.one.outliers <- data.frame(md.cfa.one.1$mahal.dist, md.cfa.one.1$is.outlier)
#removal of participants who were considered outliers on more than one method of detection
csek12.nooutliers <- csek12[-c(3, 18, 23, 29, 50, 54, 55, 60, 64, 73, 80, 89, 99, 101, 111, 117, 128, 134, 135, 146, 152, 153, 154, 162, 169, 175, 178, 185, 199, 202, 240, 241, 250, 262, 269, 271, 272, 276, 278, 282, 283, 284, 286, 290, 295, 296, 297),]
library(lavaan)
## Warning: package 'lavaan' was built under R version 4.3.3
## This is lavaan 0.6-18
## lavaan is FREE software! Please report any bugs.
library(semTools)
## Warning: package 'semTools' was built under R version 4.3.2
##
## ###############################################################################
## This is semTools 0.5-6
## All users of R (or SEM) are invited to submit functions or ideas for functions.
## ###############################################################################
#Checking fit of one factor model when outliers are removed
csek12one.fit.nooutliers <- lavaan::cfa(csek12one, data = csek12.nooutliers, std.lv = TRUE, missing = "FIML", estimator = "MLR")
summary(csek12one.fit.nooutliers, fit.measures = TRUE, standardized = TRUE) #at face value, fit looks worse w/ no outliers; does have smaller AIC and BIC
## lavaan 0.6-18 ended normally after 187 iterations
##
## Estimator ML
## Optimization method NLMINB
## Number of model parameters 117
##
## Number of observations 251
## Number of missing patterns 2
##
## Model Test User Model:
## Standard Scaled
## Test Statistic 2625.418 1567.869
## Degrees of freedom 702 702
## P-value (Chi-square) 0.000 0.000
## Scaling correction factor 1.675
## Yuan-Bentler correction (Mplus variant)
##
## Model Test Baseline Model:
##
## Test statistic 12416.542 7035.629
## Degrees of freedom 741 741
## P-value 0.000 0.000
## Scaling correction factor 1.765
##
## User Model versus Baseline Model:
##
## Comparative Fit Index (CFI) 0.835 0.862
## Tucker-Lewis Index (TLI) 0.826 0.855
##
## Robust Comparative Fit Index (CFI) 0.872
## Robust Tucker-Lewis Index (TLI) 0.864
##
## Loglikelihood and Information Criteria:
##
## Loglikelihood user model (H0) -5746.181 -5746.181
## Scaling correction factor 2.352
## for the MLR correction
## Loglikelihood unrestricted model (H1) NA NA
## Scaling correction factor 1.771
## for the MLR correction
##
## Akaike (AIC) 11726.362 11726.362
## Bayesian (BIC) 12138.840 12138.840
## Sample-size adjusted Bayesian (SABIC) 11767.936 11767.936
##
## Root Mean Square Error of Approximation:
##
## RMSEA 0.104 0.070
## 90 Percent confidence interval - lower 0.100 0.067
## 90 Percent confidence interval - upper 0.109 0.074
## P-value H_0: RMSEA <= 0.050 0.000 0.000
## P-value H_0: RMSEA >= 0.080 1.000 0.000
##
## Robust RMSEA 0.090
## 90 Percent confidence interval - lower 0.084
## 90 Percent confidence interval - upper 0.096
## P-value H_0: Robust RMSEA <= 0.050 0.000
## P-value H_0: Robust RMSEA >= 0.080 0.997
##
## Standardized Root Mean Square Residual:
##
## SRMR 0.046 0.046
##
## 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
## CSE =~
## Q2_1 0.695 0.061 11.419 0.000 0.695 0.833
## Q2_2 0.577 0.057 10.152 0.000 0.577 0.624
## Q2_3 0.541 0.046 11.635 0.000 0.541 0.854
## Q2_4 0.638 0.051 12.594 0.000 0.638 0.797
## Q2_5 0.610 0.056 10.874 0.000 0.610 0.778
## Q2_6 0.551 0.062 8.840 0.000 0.551 0.782
## Q2_7 0.547 0.067 8.198 0.000 0.547 0.842
## Q2_8 0.529 0.056 9.506 0.000 0.529 0.729
## Q2_9 0.464 0.060 7.673 0.000 0.464 0.678
## Q3_1 0.618 0.051 12.036 0.000 0.618 0.913
## Q3_2 0.592 0.054 10.882 0.000 0.592 0.889
## Q3_3 0.543 0.052 10.382 0.000 0.543 0.875
## Q3_4 0.621 0.051 12.223 0.000 0.621 0.913
## Q3_5 0.554 0.053 10.458 0.000 0.554 0.878
## Q3_6 0.438 0.052 8.483 0.000 0.438 0.721
## Q3_7 0.706 0.066 10.746 0.000 0.706 0.786
## Q3_8 0.751 0.073 10.309 0.000 0.751 0.657
## Q3_9 0.824 0.067 12.314 0.000 0.824 0.630
## Q3_10 0.553 0.057 9.719 0.000 0.553 0.692
## Q4_1 0.615 0.053 11.548 0.000 0.615 0.921
## Q4_2 0.604 0.053 11.476 0.000 0.604 0.809
## Q4_3 0.577 0.057 10.171 0.000 0.577 0.933
## Q4_4 0.572 0.058 9.829 0.000 0.572 0.910
## Q4_5 0.511 0.058 8.833 0.000 0.511 0.810
## Q4_6 0.620 0.055 11.291 0.000 0.620 0.898
## Q4_7 0.743 0.058 12.846 0.000 0.743 0.835
## Q4_8 0.482 0.058 8.256 0.000 0.482 0.808
## Q4_9 0.370 0.069 5.329 0.000 0.370 0.392
## Q4_10 0.590 0.054 10.972 0.000 0.590 0.866
## Q5_1 0.609 0.048 12.667 0.000 0.609 0.808
## Q5_2 0.525 0.054 9.774 0.000 0.525 0.770
## Q5_4 0.528 0.056 9.413 0.000 0.528 0.880
## Q5_5 0.436 0.045 9.696 0.000 0.436 0.721
## Q5_6 0.503 0.052 9.589 0.000 0.503 0.811
## Q5_7 0.536 0.054 9.898 0.000 0.536 0.848
## Q5_8 0.383 0.066 5.802 0.000 0.383 0.457
## Q5_9 0.475 0.049 9.700 0.000 0.475 0.843
## Q5_10 0.491 0.060 8.153 0.000 0.491 0.678
## Q5_11 0.555 0.058 9.515 0.000 0.555 0.906
##
## Intercepts:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .Q2_1 4.482 0.053 85.122 0.000 4.482 5.373
## .Q2_2 4.406 0.058 75.507 0.000 4.406 4.766
## .Q2_3 4.657 0.040 116.589 0.000 4.657 7.359
## .Q2_4 4.510 0.051 89.290 0.000 4.510 5.636
## .Q2_5 4.526 0.050 91.369 0.000 4.526 5.767
## .Q2_6 4.625 0.045 103.927 0.000 4.625 6.560
## .Q2_7 4.665 0.041 113.797 0.000 4.665 7.183
## .Q2_8 4.546 0.046 99.227 0.000 4.546 6.263
## .Q2_9 4.614 0.043 106.823 0.000 4.614 6.743
## .Q3_1 4.622 0.043 108.150 0.000 4.622 6.826
## .Q3_2 4.645 0.042 110.452 0.000 4.645 6.972
## .Q3_3 4.681 0.039 119.611 0.000 4.681 7.550
## .Q3_4 4.633 0.043 107.852 0.000 4.633 6.808
## .Q3_5 4.689 0.040 117.841 0.000 4.689 7.438
## .Q3_6 4.722 0.038 122.874 0.000 4.722 7.777
## .Q3_7 4.490 0.057 79.154 0.000 4.490 4.996
## .Q3_8 4.311 0.072 59.765 0.000 4.311 3.772
## .Q3_9 4.076 0.083 49.359 0.000 4.076 3.115
## .Q3_10 4.470 0.050 88.556 0.000 4.470 5.590
## .Q4_1 4.637 0.042 109.984 0.000 4.637 6.942
## .Q4_2 4.562 0.047 96.842 0.000 4.562 6.113
## .Q4_3 4.705 0.039 120.420 0.000 4.705 7.601
## .Q4_4 4.697 0.040 118.501 0.000 4.697 7.480
## .Q4_5 4.709 0.040 118.337 0.000 4.709 7.469
## .Q4_6 4.641 0.044 106.470 0.000 4.641 6.720
## .Q4_7 4.466 0.056 79.573 0.000 4.466 5.023
## .Q4_8 4.749 0.038 126.219 0.000 4.749 7.967
## .Q4_9 4.363 0.060 73.161 0.000 4.363 4.618
## .Q4_10 4.629 0.043 107.638 0.000 4.629 6.794
## .Q5_1 4.534 0.048 95.344 0.000 4.534 6.018
## .Q5_2 4.629 0.043 107.638 0.000 4.629 6.794
## .Q5_4 4.721 0.038 124.579 0.000 4.721 7.863
## .Q5_5 4.665 0.038 122.159 0.000 4.665 7.711
## .Q5_6 4.701 0.039 120.065 0.000 4.701 7.578
## .Q5_7 4.681 0.040 117.206 0.000 4.681 7.398
## .Q5_8 4.546 0.053 85.952 0.000 4.546 5.425
## .Q5_9 4.745 0.036 133.424 0.000 4.745 8.422
## .Q5_10 4.570 0.046 100.012 0.000 4.570 6.313
## .Q5_11 4.705 0.039 121.692 0.000 4.705 7.681
##
## Variances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .Q2_1 0.213 0.036 5.848 0.000 0.213 0.306
## .Q2_2 0.522 0.084 6.244 0.000 0.522 0.610
## .Q2_3 0.108 0.017 6.456 0.000 0.108 0.270
## .Q2_4 0.233 0.044 5.284 0.000 0.233 0.365
## .Q2_5 0.243 0.040 6.115 0.000 0.243 0.395
## .Q2_6 0.193 0.045 4.261 0.000 0.193 0.389
## .Q2_7 0.123 0.024 5.185 0.000 0.123 0.291
## .Q2_8 0.247 0.038 6.463 0.000 0.247 0.469
## .Q2_9 0.253 0.042 6.092 0.000 0.253 0.540
## .Q3_1 0.076 0.014 5.450 0.000 0.076 0.166
## .Q3_2 0.093 0.025 3.793 0.000 0.093 0.210
## .Q3_3 0.090 0.023 3.999 0.000 0.090 0.234
## .Q3_4 0.077 0.018 4.278 0.000 0.077 0.167
## .Q3_5 0.091 0.018 4.932 0.000 0.091 0.228
## .Q3_6 0.177 0.038 4.685 0.000 0.177 0.481
## .Q3_7 0.309 0.046 6.643 0.000 0.309 0.382
## .Q3_8 0.742 0.128 5.811 0.000 0.742 0.568
## .Q3_9 1.032 0.133 7.765 0.000 1.032 0.603
## .Q3_10 0.333 0.075 4.430 0.000 0.333 0.521
## .Q4_1 0.068 0.018 3.742 0.000 0.068 0.152
## .Q4_2 0.193 0.047 4.130 0.000 0.193 0.346
## .Q4_3 0.050 0.010 5.105 0.000 0.050 0.130
## .Q4_4 0.068 0.018 3.805 0.000 0.068 0.172
## .Q4_5 0.136 0.029 4.765 0.000 0.136 0.343
## .Q4_6 0.093 0.015 6.072 0.000 0.093 0.194
## .Q4_7 0.239 0.039 6.188 0.000 0.239 0.302
## .Q4_8 0.123 0.023 5.366 0.000 0.123 0.347
## .Q4_9 0.755 0.108 7.025 0.000 0.755 0.846
## .Q4_10 0.116 0.024 4.838 0.000 0.116 0.250
## .Q5_1 0.197 0.036 5.547 0.000 0.197 0.347
## .Q5_2 0.189 0.047 4.040 0.000 0.189 0.407
## .Q5_4 0.081 0.019 4.221 0.000 0.081 0.225
## .Q5_5 0.176 0.029 6.042 0.000 0.176 0.480
## .Q5_6 0.132 0.023 5.804 0.000 0.132 0.343
## .Q5_7 0.113 0.024 4.692 0.000 0.113 0.282
## .Q5_8 0.556 0.109 5.107 0.000 0.556 0.791
## .Q5_9 0.092 0.016 5.901 0.000 0.092 0.290
## .Q5_10 0.283 0.046 6.105 0.000 0.283 0.540
## .Q5_11 0.067 0.015 4.543 0.000 0.067 0.179
## CSE 1.000 1.000 1.000
dynamic::cfaOne(csek12one.fit.nooutliers, plot = TRUE) #still does not meet dynamic fit indices
## Warning: Returning more (or less) than 1 row per `summarise()` group was deprecated in
## dplyr 1.1.0.
## ℹ Please use `reframe()` instead.
## ℹ When switching from `summarise()` to `reframe()`, remember that `reframe()`
## always returns an ungrouped data frame and adjust accordingly.
## ℹ The deprecated feature was likely used in the dynamic package.
## Please report the issue at <https://github.com/melissagwolf/dynamic/issues>.
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_lifecycle_warnings()` to see where this warning was
## generated.
## Your DFI cutoffs:
## SRMR RMSEA CFI
## Level 1: 95/5 .024 NONE NONE
## Level 1: 90/10 -- .027 .987
## Level 2: 95/5 .025 .037 .976
## Level 2: 90/10 -- -- --
## Level 3: 95/5 .026 .047 .962
## Level 3: 90/10 -- -- --
##
## Empirical fit indices:
## Chi-Square df p-value SRMR RMSEA CFI
## 2625.418 702 0 0.046 0.104 0.835
##
## The distributions for each level are in the Plots tab
## [[1]]
##
## [[2]]
##
## [[3]]
csek12one.mod4 <- '
CSE =~ Q2_1 + Q2_2 + Q2_3 + Q2_4 + Q2_5 + Q2_6 + Q2_7 + Q2_8 + Q2_9 + Q3_1 + Q3_2 + Q3_3 + Q3_4 + Q3_5 + Q3_6 + Q3_7 + Q3_8 + Q3_9 + Q3_10 + Q4_1 + Q4_2 + Q4_3 + Q4_4 + Q4_5 + Q4_6 + Q4_7 + Q4_8 + Q4_9 + Q4_10 + Q5_1 + Q5_2 + Q5_4 + Q5_5 + Q5_6 + Q5_7 + Q5_8 + Q5_9 + Q5_10 + Q5_11
#covariances
Q2_3 ~~ Q2_4
Q2_3 ~~ Q2_5
Q2_3 ~~ Q5_5
Q2_3 ~~ Q5_6
Q2_4 ~~ Q2_5
Q2_4 ~~ Q5_5
Q2_4 ~~ Q5_6
Q2_5 ~~ Q5_5
Q2_5 ~~ Q5_6
Q5_5 ~~ Q5_6
Q4_4 ~~ Q4_5
Q4_4 ~~ Q4_6
Q4_5 ~~ Q4_6
Q4_7 ~~ Q4_8
Q2_8 ~~ Q2_9
'
#checking fit of one factor model + all covariance sets added when outliers are removed
csek12one.mod4.fit.nooutliers <- lavaan::cfa(csek12one.mod4, data = csek12.nooutliers, std.lv = TRUE, missing = "FIML", estimator = "MLR")
summary(csek12one.mod4.fit.nooutliers, fit.measures = TRUE, standardized = TRUE) #fit looks worse w/ no outliers, does have smaller AIC and BIC
## lavaan 0.6-18 ended normally after 209 iterations
##
## Estimator ML
## Optimization method NLMINB
## Number of model parameters 132
##
## Number of observations 251
## Number of missing patterns 2
##
## Model Test User Model:
## Standard Scaled
## Test Statistic 2401.021 1452.349
## Degrees of freedom 687 687
## P-value (Chi-square) 0.000 0.000
## Scaling correction factor 1.653
## Yuan-Bentler correction (Mplus variant)
##
## Model Test Baseline Model:
##
## Test statistic 12416.542 7035.629
## Degrees of freedom 741 741
## P-value 0.000 0.000
## Scaling correction factor 1.765
##
## User Model versus Baseline Model:
##
## Comparative Fit Index (CFI) 0.853 0.878
## Tucker-Lewis Index (TLI) 0.842 0.869
##
## Robust Comparative Fit Index (CFI) 0.889
## Robust Tucker-Lewis Index (TLI) 0.880
##
## Loglikelihood and Information Criteria:
##
## Loglikelihood user model (H0) -5633.982 -5633.982
## Scaling correction factor 2.386
## for the MLR correction
## Loglikelihood unrestricted model (H1) NA NA
## Scaling correction factor 1.771
## for the MLR correction
##
## Akaike (AIC) 11531.965 11531.965
## Bayesian (BIC) 11997.325 11997.325
## Sample-size adjusted Bayesian (SABIC) 11578.869 11578.869
##
## Root Mean Square Error of Approximation:
##
## RMSEA 0.100 0.067
## 90 Percent confidence interval - lower 0.095 0.063
## 90 Percent confidence interval - upper 0.104 0.070
## P-value H_0: RMSEA <= 0.050 0.000 0.000
## P-value H_0: RMSEA >= 0.080 1.000 0.000
##
## Robust RMSEA 0.085
## 90 Percent confidence interval - lower 0.078
## 90 Percent confidence interval - upper 0.091
## P-value H_0: Robust RMSEA <= 0.050 0.000
## P-value H_0: Robust RMSEA >= 0.080 0.895
##
## Standardized Root Mean Square Residual:
##
## SRMR 0.045 0.045
##
## 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
## CSE =~
## Q2_1 0.695 0.061 11.375 0.000 0.695 0.833
## Q2_2 0.579 0.057 10.103 0.000 0.579 0.626
## Q2_3 0.538 0.047 11.532 0.000 0.538 0.851
## Q2_4 0.633 0.051 12.354 0.000 0.633 0.790
## Q2_5 0.604 0.057 10.683 0.000 0.604 0.770
## Q2_6 0.549 0.063 8.748 0.000 0.549 0.778
## Q2_7 0.546 0.067 8.168 0.000 0.546 0.840
## Q2_8 0.524 0.056 9.386 0.000 0.524 0.722
## Q2_9 0.460 0.061 7.568 0.000 0.460 0.672
## Q3_1 0.620 0.051 12.165 0.000 0.620 0.917
## Q3_2 0.593 0.054 10.945 0.000 0.593 0.890
## Q3_3 0.543 0.052 10.370 0.000 0.543 0.875
## Q3_4 0.623 0.051 12.300 0.000 0.623 0.915
## Q3_5 0.556 0.053 10.561 0.000 0.556 0.882
## Q3_6 0.438 0.051 8.521 0.000 0.438 0.722
## Q3_7 0.706 0.066 10.719 0.000 0.706 0.786
## Q3_8 0.752 0.073 10.298 0.000 0.752 0.658
## Q3_9 0.825 0.067 12.313 0.000 0.825 0.631
## Q3_10 0.552 0.057 9.687 0.000 0.552 0.691
## Q4_1 0.617 0.053 11.613 0.000 0.617 0.923
## Q4_2 0.603 0.053 11.452 0.000 0.603 0.808
## Q4_3 0.579 0.057 10.230 0.000 0.579 0.935
## Q4_4 0.570 0.058 9.775 0.000 0.570 0.908
## Q4_5 0.503 0.058 8.669 0.000 0.503 0.799
## Q4_6 0.616 0.055 11.145 0.000 0.616 0.893
## Q4_7 0.743 0.058 12.803 0.000 0.743 0.836
## Q4_8 0.481 0.058 8.240 0.000 0.481 0.806
## Q4_9 0.367 0.070 5.266 0.000 0.367 0.389
## Q4_10 0.591 0.054 10.989 0.000 0.591 0.867
## Q5_1 0.609 0.048 12.666 0.000 0.609 0.808
## Q5_2 0.526 0.054 9.820 0.000 0.526 0.772
## Q5_4 0.529 0.056 9.468 0.000 0.529 0.882
## Q5_5 0.431 0.045 9.510 0.000 0.431 0.712
## Q5_6 0.504 0.052 9.644 0.000 0.504 0.813
## Q5_7 0.534 0.054 9.850 0.000 0.534 0.843
## Q5_8 0.382 0.066 5.768 0.000 0.382 0.456
## Q5_9 0.474 0.049 9.667 0.000 0.474 0.842
## Q5_10 0.488 0.060 8.075 0.000 0.488 0.675
## Q5_11 0.557 0.058 9.560 0.000 0.557 0.909
##
## Covariances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .Q2_3 ~~
## .Q2_4 0.025 0.014 1.829 0.067 0.025 0.156
## .Q2_5 0.018 0.015 1.234 0.217 0.018 0.111
## .Q5_5 0.012 0.015 0.784 0.433 0.012 0.086
## .Q5_6 0.009 0.013 0.663 0.508 0.009 0.073
## .Q2_4 ~~
## .Q2_5 0.119 0.032 3.672 0.000 0.119 0.483
## .Q5_5 0.034 0.020 1.728 0.084 0.034 0.162
## .Q5_6 -0.031 0.012 -2.649 0.008 -0.031 -0.175
## .Q2_5 ~~
## .Q5_5 0.039 0.021 1.905 0.057 0.039 0.185
## .Q5_6 -0.032 0.015 -2.177 0.029 -0.032 -0.175
## .Q5_5 ~~
## .Q5_6 0.014 0.017 0.817 0.414 0.014 0.089
## .Q4_4 ~~
## .Q4_5 0.023 0.013 1.773 0.076 0.023 0.233
## .Q4_6 0.014 0.011 1.322 0.186 0.014 0.172
## .Q4_5 ~~
## .Q4_6 0.047 0.015 3.221 0.001 0.047 0.402
## .Q4_7 ~~
## .Q4_8 0.007 0.019 0.362 0.718 0.007 0.041
## .Q2_8 ~~
## .Q2_9 0.129 0.033 3.957 0.000 0.129 0.508
##
## Intercepts:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .Q2_1 4.482 0.053 85.122 0.000 4.482 5.373
## .Q2_2 4.406 0.058 75.507 0.000 4.406 4.766
## .Q2_3 4.657 0.040 116.589 0.000 4.657 7.359
## .Q2_4 4.510 0.051 89.290 0.000 4.510 5.636
## .Q2_5 4.526 0.050 91.369 0.000 4.526 5.767
## .Q2_6 4.625 0.045 103.927 0.000 4.625 6.560
## .Q2_7 4.665 0.041 113.797 0.000 4.665 7.183
## .Q2_8 4.546 0.046 99.227 0.000 4.546 6.263
## .Q2_9 4.614 0.043 106.823 0.000 4.614 6.743
## .Q3_1 4.622 0.043 108.150 0.000 4.622 6.826
## .Q3_2 4.645 0.042 110.452 0.000 4.645 6.972
## .Q3_3 4.681 0.039 119.610 0.000 4.681 7.550
## .Q3_4 4.633 0.043 107.852 0.000 4.633 6.808
## .Q3_5 4.689 0.040 117.841 0.000 4.689 7.438
## .Q3_6 4.722 0.038 122.959 0.000 4.722 7.780
## .Q3_7 4.490 0.057 79.153 0.000 4.490 4.996
## .Q3_8 4.311 0.072 59.766 0.000 4.311 3.772
## .Q3_9 4.076 0.083 49.359 0.000 4.076 3.115
## .Q3_10 4.470 0.050 88.556 0.000 4.470 5.590
## .Q4_1 4.637 0.042 109.984 0.000 4.637 6.942
## .Q4_2 4.562 0.047 96.842 0.000 4.562 6.113
## .Q4_3 4.705 0.039 120.420 0.000 4.705 7.601
## .Q4_4 4.697 0.040 118.501 0.000 4.697 7.480
## .Q4_5 4.709 0.040 118.336 0.000 4.709 7.469
## .Q4_6 4.641 0.044 106.470 0.000 4.641 6.720
## .Q4_7 4.466 0.056 79.573 0.000 4.466 5.023
## .Q4_8 4.749 0.038 126.219 0.000 4.749 7.967
## .Q4_9 4.363 0.060 73.161 0.000 4.363 4.618
## .Q4_10 4.629 0.043 107.638 0.000 4.629 6.794
## .Q5_1 4.534 0.048 95.344 0.000 4.534 6.018
## .Q5_2 4.629 0.043 107.638 0.000 4.629 6.794
## .Q5_4 4.721 0.038 124.580 0.000 4.721 7.863
## .Q5_5 4.665 0.038 122.159 0.000 4.665 7.711
## .Q5_6 4.701 0.039 120.065 0.000 4.701 7.578
## .Q5_7 4.681 0.040 117.206 0.000 4.681 7.398
## .Q5_8 4.546 0.053 85.952 0.000 4.546 5.425
## .Q5_9 4.745 0.036 133.424 0.000 4.745 8.422
## .Q5_10 4.570 0.046 100.012 0.000 4.570 6.313
## .Q5_11 4.705 0.039 121.692 0.000 4.705 7.681
##
## Variances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .Q2_1 0.213 0.037 5.800 0.000 0.213 0.306
## .Q2_2 0.519 0.083 6.244 0.000 0.519 0.608
## .Q2_3 0.111 0.017 6.372 0.000 0.111 0.277
## .Q2_4 0.240 0.045 5.303 0.000 0.240 0.375
## .Q2_5 0.251 0.040 6.206 0.000 0.251 0.408
## .Q2_6 0.196 0.046 4.239 0.000 0.196 0.395
## .Q2_7 0.124 0.024 5.168 0.000 0.124 0.294
## .Q2_8 0.253 0.039 6.426 0.000 0.253 0.479
## .Q2_9 0.257 0.042 6.149 0.000 0.257 0.548
## .Q3_1 0.073 0.013 5.536 0.000 0.073 0.160
## .Q3_2 0.092 0.024 3.793 0.000 0.092 0.208
## .Q3_3 0.090 0.023 3.910 0.000 0.090 0.234
## .Q3_4 0.075 0.018 4.139 0.000 0.075 0.163
## .Q3_5 0.088 0.018 4.930 0.000 0.088 0.222
## .Q3_6 0.176 0.038 4.667 0.000 0.176 0.478
## .Q3_7 0.309 0.046 6.676 0.000 0.309 0.383
## .Q3_8 0.741 0.128 5.787 0.000 0.741 0.567
## .Q3_9 1.031 0.134 7.719 0.000 1.031 0.602
## .Q3_10 0.334 0.075 4.441 0.000 0.334 0.523
## .Q4_1 0.066 0.018 3.659 0.000 0.066 0.148
## .Q4_2 0.193 0.047 4.127 0.000 0.193 0.347
## .Q4_3 0.048 0.009 5.258 0.000 0.048 0.126
## .Q4_4 0.069 0.018 3.824 0.000 0.069 0.176
## .Q4_5 0.144 0.030 4.747 0.000 0.144 0.362
## .Q4_6 0.097 0.016 6.079 0.000 0.097 0.203
## .Q4_7 0.239 0.039 6.187 0.000 0.239 0.302
## .Q4_8 0.124 0.023 5.309 0.000 0.124 0.350
## .Q4_9 0.757 0.107 7.051 0.000 0.757 0.849
## .Q4_10 0.115 0.024 4.818 0.000 0.115 0.249
## .Q5_1 0.197 0.036 5.517 0.000 0.197 0.347
## .Q5_2 0.188 0.047 4.018 0.000 0.188 0.404
## .Q5_4 0.080 0.019 4.147 0.000 0.080 0.223
## .Q5_5 0.181 0.030 6.055 0.000 0.181 0.494
## .Q5_6 0.131 0.023 5.765 0.000 0.131 0.340
## .Q5_7 0.116 0.026 4.532 0.000 0.116 0.289
## .Q5_8 0.556 0.109 5.109 0.000 0.556 0.792
## .Q5_9 0.092 0.016 5.921 0.000 0.092 0.291
## .Q5_10 0.285 0.047 6.126 0.000 0.285 0.545
## .Q5_11 0.065 0.015 4.465 0.000 0.065 0.175
## CSE 1.000 1.000 1.000
dynamic::cfaOne(csek12one.mod4.fit.nooutliers, plot = TRUE) #still does not meet dynamic fit indices
## Your DFI cutoffs:
## SRMR RMSEA CFI
## Level 1: 95/5 NONE NONE NONE
## Level 1: 90/10 .024 NONE NONE
## Level 2: 95/5 .024 .03 .985
## Level 2: 90/10 -- -- --
## Level 3: 95/5 .024 .039 .975
## Level 3: 90/10 -- -- --
##
## Empirical fit indices:
## Chi-Square df p-value SRMR RMSEA CFI
## 2401.021 687 0 0.045 0.1 0.853
##
## The distributions for each level are in the Plots tab
## [[1]]
##
## [[2]]
##
## [[3]]
#invariance testing with outliers removed
#examining invariance by school aged children or not when outliers are removed
csek12.schoolage1 <- subset(csek12.nooutliers, Parent != 3) #sub-setting to remove participants who do not have children
csek12one.config.nooutlier <- cfa(csek12one.mod4, data = csek12.schoolage1, std.lv = TRUE, missing = "FIML", estimator = "MLR", group = "Parent") #configural model
## Warning: lavaan->lav_mvnorm_missing_h1_estimate_moments():
## The smallest eigenvalue of the EM estimated variance-covariance matrix
## (Sigma) is smaller than 1e-05; this may cause numerical instabilities;
## interpret the results with caution.
## Warning: lavaan->lav_model_vcov():
## The variance-covariance matrix of the estimated parameters (vcov) does not
## appear to be positive definite! The smallest eigenvalue (= -1.536235e-14)
## is smaller than zero. This may be a symptom that the model is not
## identified.
summary(csek12one.config.nooutlier, fit.measures = TRUE)
## lavaan 0.6-18 ended normally after 156 iterations
##
## Estimator ML
## Optimization method NLMINB
## Number of model parameters 264
##
## Number of observations per group:
## 1 54
## 2 55
## Number of missing patterns per group:
## 1 1
## 2 1
##
## Model Test User Model:
## Standard Scaled
## Test Statistic 5373.220 5339.097
## Degrees of freedom 1374 1374
## P-value (Chi-square) 0.000 0.000
## Scaling correction factor 1.006
## Yuan-Bentler correction (Mplus variant)
## Test statistic for each group:
## 1 2629.175 2612.478
## 2 2744.044 2726.618
##
## Model Test Baseline Model:
##
## Test statistic 9704.117 8467.492
## Degrees of freedom 1482 1482
## P-value 0.000 0.000
## Scaling correction factor 1.146
##
## User Model versus Baseline Model:
##
## Comparative Fit Index (CFI) 0.514 0.432
## Tucker-Lewis Index (TLI) 0.475 0.388
##
## Robust Comparative Fit Index (CFI) 0.517
## Robust Tucker-Lewis Index (TLI) 0.479
##
## Loglikelihood and Information Criteria:
##
## Loglikelihood user model (H0) -2316.961 -2316.961
## Scaling correction factor 2.089
## for the MLR correction
## Loglikelihood unrestricted model (H1) NA NA
## Scaling correction factor 1.181
## for the MLR correction
##
## Akaike (AIC) 5161.922 5161.922
## Bayesian (BIC) 5872.438 5872.438
## Sample-size adjusted Bayesian (SABIC) 5038.232 5038.232
##
## Root Mean Square Error of Approximation:
##
## RMSEA 0.231 0.230
## 90 Percent confidence interval - lower 0.225 0.224
## 90 Percent confidence interval - upper 0.238 0.237
## P-value H_0: RMSEA <= 0.050 0.000 0.000
## P-value H_0: RMSEA >= 0.080 1.000 1.000
##
## Robust RMSEA 0.229
## 90 Percent confidence interval - lower 0.223
## 90 Percent confidence interval - upper 0.236
## P-value H_0: Robust RMSEA <= 0.050 0.000
## P-value H_0: Robust RMSEA >= 0.080 1.000
##
## Standardized Root Mean Square Residual:
##
## SRMR 0.080 0.080
##
## Parameter Estimates:
##
## Standard errors Sandwich
## Information bread Observed
## Observed information based on Hessian
##
##
## Group 1 [1]:
##
## Latent Variables:
## Estimate Std.Err z-value P(>|z|)
## CSE =~
## Q2_1 0.503 0.111 4.535 0.000
## Q2_2 0.397 0.140 2.832 0.005
## Q2_3 0.508 0.087 5.811 0.000
## Q2_4 0.633 0.114 5.554 0.000
## Q2_5 0.476 0.140 3.396 0.001
## Q2_6 0.380 0.079 4.788 0.000
## Q2_7 0.413 0.101 4.108 0.000
## Q2_8 0.401 0.070 5.688 0.000
## Q2_9 0.282 0.064 4.368 0.000
## Q3_1 0.461 0.048 9.697 0.000
## Q3_2 0.466 0.071 6.525 0.000
## Q3_3 0.397 0.034 11.752 0.000
## Q3_4 0.472 0.066 7.199 0.000
## Q3_5 0.428 0.078 5.459 0.000
## Q3_6 0.342 0.080 4.247 0.000
## Q3_7 0.508 0.134 3.794 0.000
## Q3_8 0.507 0.169 3.000 0.003
## Q3_9 0.874 0.177 4.947 0.000
## Q3_10 0.430 0.076 5.671 0.000
## Q4_1 0.425 0.055 7.680 0.000
## Q4_2 0.465 0.069 6.749 0.000
## Q4_3 0.375 0.070 5.323 0.000
## Q4_4 0.319 0.054 5.911 0.000
## Q4_5 0.298 0.062 4.783 0.000
## Q4_6 0.466 0.097 4.800 0.000
## Q4_7 0.748 0.137 5.474 0.000
## Q4_8 0.408 0.146 2.783 0.005
## Q4_9 0.185 0.082 2.263 0.024
## Q4_10 0.352 0.052 6.833 0.000
## Q5_1 0.496 0.070 7.040 0.000
## Q5_2 0.382 0.066 5.781 0.000
## Q5_4 0.362 0.071 5.094 0.000
## Q5_5 0.346 0.064 5.396 0.000
## Q5_6 0.287 0.079 3.627 0.000
## Q5_7 0.310 0.062 4.992 0.000
## Q5_8 0.311 0.123 2.528 0.011
## Q5_9 0.284 0.085 3.322 0.001
## Q5_10 0.324 0.076 4.250 0.000
## Q5_11 0.381 0.040 9.523 0.000
##
## Covariances:
## Estimate Std.Err z-value P(>|z|)
## .Q2_3 ~~
## .Q2_4 -0.000 0.038 -0.012 0.991
## .Q2_5 0.034 0.041 0.834 0.404
## .Q5_5 -0.017 0.036 -0.487 0.627
## .Q5_6 0.027 0.023 1.156 0.248
## .Q2_4 ~~
## .Q2_5 0.154 0.066 2.327 0.020
## .Q5_5 -0.011 0.032 -0.345 0.730
## .Q5_6 -0.046 0.029 -1.585 0.113
## .Q2_5 ~~
## .Q5_5 0.062 0.038 1.649 0.099
## .Q5_6 -0.021 0.041 -0.526 0.599
## .Q5_5 ~~
## .Q5_6 0.010 0.038 0.253 0.800
## .Q4_4 ~~
## .Q4_5 0.008 0.023 0.329 0.742
## .Q4_6 -0.030 0.021 -1.414 0.157
## .Q4_5 ~~
## .Q4_6 0.060 0.042 1.431 0.153
## .Q4_7 ~~
## .Q4_8 -0.009 0.055 -0.170 0.865
## .Q2_8 ~~
## .Q2_9 0.080 0.065 1.240 0.215
##
## Intercepts:
## Estimate Std.Err z-value P(>|z|)
## .Q2_1 4.463 0.093 47.795 0.000
## .Q2_2 4.352 0.131 33.130 0.000
## .Q2_3 4.611 0.085 54.553 0.000
## .Q2_4 4.444 0.110 40.376 0.000
## .Q2_5 4.519 0.107 42.168 0.000
## .Q2_6 4.593 0.089 51.670 0.000
## .Q2_7 4.667 0.079 59.397 0.000
## .Q2_8 4.556 0.081 55.948 0.000
## .Q2_9 4.685 0.068 68.483 0.000
## .Q3_1 4.648 0.070 66.341 0.000
## .Q3_2 4.648 0.075 62.142 0.000
## .Q3_3 4.722 0.061 77.475 0.000
## .Q3_4 4.648 0.079 58.651 0.000
## .Q3_5 4.704 0.077 61.140 0.000
## .Q3_6 4.685 0.078 60.224 0.000
## .Q3_7 4.389 0.115 38.032 0.000
## .Q3_8 4.037 0.166 24.384 0.000
## .Q3_9 3.796 0.202 18.811 0.000
## .Q3_10 4.463 0.104 42.991 0.000
## .Q4_1 4.648 0.075 62.142 0.000
## .Q4_2 4.556 0.089 50.927 0.000
## .Q4_3 4.778 0.062 76.637 0.000
## .Q4_4 4.778 0.057 84.450 0.000
## .Q4_5 4.796 0.061 78.960 0.000
## .Q4_6 4.704 0.081 57.878 0.000
## .Q4_7 4.426 0.127 34.786 0.000
## .Q4_8 4.741 0.088 54.126 0.000
## .Q4_9 4.519 0.093 48.337 0.000
## .Q4_10 4.722 0.061 77.475 0.000
## .Q5_1 4.593 0.089 51.670 0.000
## .Q5_2 4.722 0.066 71.182 0.000
## .Q5_4 4.796 0.061 78.960 0.000
## .Q5_5 4.741 0.075 63.272 0.000
## .Q5_6 4.778 0.062 76.637 0.000
## .Q5_7 4.778 0.062 76.637 0.000
## .Q5_8 4.741 0.075 63.272 0.000
## .Q5_9 4.796 0.061 78.960 0.000
## .Q5_10 4.667 0.079 59.397 0.000
## .Q5_11 4.759 0.058 81.802 0.000
##
## Variances:
## Estimate Std.Err z-value P(>|z|)
## .Q2_1 0.217 0.054 4.024 0.000
## .Q2_2 0.774 0.285 2.717 0.007
## .Q2_3 0.128 0.035 3.640 0.000
## .Q2_4 0.254 0.065 3.904 0.000
## .Q2_5 0.394 0.112 3.531 0.000
## .Q2_6 0.282 0.111 2.535 0.011
## .Q2_7 0.163 0.050 3.226 0.001
## .Q2_8 0.197 0.063 3.116 0.002
## .Q2_9 0.173 0.069 2.512 0.012
## .Q3_1 0.053 0.017 3.052 0.002
## .Q3_2 0.085 0.029 2.974 0.003
## .Q3_3 0.043 0.015 2.921 0.003
## .Q3_4 0.116 0.060 1.934 0.053
## .Q3_5 0.136 0.064 2.114 0.035
## .Q3_6 0.210 0.090 2.340 0.019
## .Q3_7 0.461 0.121 3.803 0.000
## .Q3_8 1.224 0.286 4.279 0.000
## .Q3_9 1.436 0.315 4.565 0.000
## .Q3_10 0.397 0.259 1.534 0.125
## .Q4_1 0.122 0.060 2.012 0.044
## .Q4_2 0.216 0.091 2.371 0.018
## .Q4_3 0.069 0.022 3.124 0.002
## .Q4_4 0.071 0.024 2.995 0.003
## .Q4_5 0.110 0.046 2.397 0.017
## .Q4_6 0.139 0.047 2.948 0.003
## .Q4_7 0.315 0.078 4.030 0.000
## .Q4_8 0.248 0.069 3.592 0.000
## .Q4_9 0.438 0.105 4.151 0.000
## .Q4_10 0.076 0.030 2.577 0.010
## .Q5_1 0.181 0.075 2.401 0.016
## .Q5_2 0.091 0.031 2.911 0.004
## .Q5_4 0.068 0.024 2.804 0.005
## .Q5_5 0.183 0.075 2.457 0.014
## .Q5_6 0.127 0.037 3.476 0.001
## .Q5_7 0.114 0.047 2.425 0.015
## .Q5_8 0.206 0.075 2.764 0.006
## .Q5_9 0.119 0.037 3.240 0.001
## .Q5_10 0.228 0.094 2.433 0.015
## .Q5_11 0.038 0.012 3.109 0.002
## CSE 1.000
##
##
## Group 2 [2]:
##
## Latent Variables:
## Estimate Std.Err z-value P(>|z|)
## CSE =~
## Q2_1 0.834 0.119 7.013 0.000
## Q2_2 0.714 0.123 5.804 0.000
## Q2_3 0.698 0.122 5.736 0.000
## Q2_4 0.726 0.109 6.660 0.000
## Q2_5 0.611 0.127 4.814 0.000
## Q2_6 0.713 0.132 5.400 0.000
## Q2_7 0.615 0.133 4.610 0.000
## Q2_8 0.561 0.137 4.087 0.000
## Q2_9 0.619 0.135 4.572 0.000
## Q3_1 0.742 0.110 6.717 0.000
## Q3_2 0.695 0.124 5.608 0.000
## Q3_3 0.616 0.113 5.462 0.000
## Q3_4 0.740 0.115 6.433 0.000
## Q3_5 0.692 0.100 6.888 0.000
## Q3_6 0.639 0.106 6.026 0.000
## Q3_7 0.721 0.106 6.835 0.000
## Q3_8 0.842 0.115 7.301 0.000
## Q3_9 0.867 0.141 6.149 0.000
## Q3_10 0.662 0.094 7.044 0.000
## Q4_1 0.733 0.116 6.345 0.000
## Q4_2 0.715 0.115 6.229 0.000
## Q4_3 0.740 0.117 6.346 0.000
## Q4_4 0.695 0.125 5.578 0.000
## Q4_5 0.645 0.145 4.459 0.000
## Q4_6 0.716 0.119 6.015 0.000
## Q4_7 0.734 0.100 7.354 0.000
## Q4_8 0.633 0.112 5.670 0.000
## Q4_9 0.369 0.174 2.120 0.034
## Q4_10 0.667 0.132 5.062 0.000
## Q5_1 0.671 0.120 5.574 0.000
## Q5_2 0.551 0.127 4.350 0.000
## Q5_4 0.690 0.135 5.127 0.000
## Q5_5 0.564 0.111 5.079 0.000
## Q5_6 0.582 0.108 5.398 0.000
## Q5_7 0.636 0.109 5.854 0.000
## Q5_8 0.384 0.156 2.467 0.014
## Q5_9 0.639 0.109 5.855 0.000
## Q5_10 0.601 0.135 4.441 0.000
## Q5_11 0.651 0.145 4.505 0.000
##
## Covariances:
## Estimate Std.Err z-value P(>|z|)
## .Q2_3 ~~
## .Q2_4 0.027 0.029 0.935 0.350
## .Q2_5 0.045 0.031 1.463 0.144
## .Q5_5 -0.022 0.030 -0.722 0.470
## .Q5_6 0.014 0.035 0.386 0.699
## .Q2_4 ~~
## .Q2_5 0.085 0.055 1.565 0.118
## .Q5_5 0.001 0.015 0.034 0.973
## .Q5_6 -0.024 0.020 -1.239 0.215
## .Q2_5 ~~
## .Q5_5 -0.033 0.025 -1.295 0.195
## .Q5_6 -0.020 0.022 -0.894 0.371
## .Q5_5 ~~
## .Q5_6 0.032 0.029 1.116 0.264
## .Q4_4 ~~
## .Q4_5 -0.001 0.013 -0.048 0.962
## .Q4_6 0.047 0.034 1.397 0.162
## .Q4_5 ~~
## .Q4_6 0.034 0.026 1.328 0.184
## .Q4_7 ~~
## .Q4_8 -0.032 0.024 -1.331 0.183
## .Q2_8 ~~
## .Q2_9 0.103 0.066 1.571 0.116
##
## Intercepts:
## Estimate Std.Err z-value P(>|z|)
## .Q2_1 4.309 0.136 31.743 0.000
## .Q2_2 4.236 0.136 31.156 0.000
## .Q2_3 4.545 0.105 43.106 0.000
## .Q2_4 4.455 0.109 41.041 0.000
## .Q2_5 4.436 0.108 40.936 0.000
## .Q2_6 4.527 0.109 41.669 0.000
## .Q2_7 4.600 0.091 50.399 0.000
## .Q2_8 4.491 0.099 45.288 0.000
## .Q2_9 4.527 0.102 44.219 0.000
## .Q3_1 4.491 0.109 41.313 0.000
## .Q3_2 4.545 0.105 43.106 0.000
## .Q3_3 4.600 0.095 48.511 0.000
## .Q3_4 4.564 0.102 44.702 0.000
## .Q3_5 4.600 0.098 46.820 0.000
## .Q3_6 4.582 0.102 44.984 0.000
## .Q3_7 4.455 0.120 37.089 0.000
## .Q3_8 4.236 0.158 26.740 0.000
## .Q3_9 3.909 0.175 22.349 0.000
## .Q3_10 4.400 0.114 38.652 0.000
## .Q4_1 4.564 0.102 44.702 0.000
## .Q4_2 4.527 0.106 42.887 0.000
## .Q4_3 4.582 0.102 44.984 0.000
## .Q4_4 4.582 0.102 44.984 0.000
## .Q4_5 4.636 0.101 46.002 0.000
## .Q4_6 4.564 0.102 44.702 0.000
## .Q4_7 4.291 0.125 34.298 0.000
## .Q4_8 4.636 0.094 49.325 0.000
## .Q4_9 4.327 0.123 35.050 0.000
## .Q4_10 4.618 0.098 47.181 0.000
## .Q5_1 4.473 0.109 41.167 0.000
## .Q5_2 4.545 0.105 43.106 0.000
## .Q5_4 4.636 0.101 46.002 0.000
## .Q5_5 4.618 0.087 52.984 0.000
## .Q5_6 4.564 0.095 47.840 0.000
## .Q5_7 4.618 0.094 48.899 0.000
## .Q5_8 4.600 0.111 41.481 0.000
## .Q5_9 4.636 0.094 49.325 0.000
## .Q5_10 4.473 0.112 40.060 0.000
## .Q5_11 4.655 0.100 46.403 0.000
##
## Variances:
## Estimate Std.Err z-value P(>|z|)
## .Q2_1 0.319 0.090 3.534 0.000
## .Q2_2 0.506 0.109 4.626 0.000
## .Q2_3 0.125 0.041 3.055 0.002
## .Q2_4 0.121 0.056 2.179 0.029
## .Q2_5 0.273 0.094 2.906 0.004
## .Q2_6 0.141 0.046 3.046 0.002
## .Q2_7 0.080 0.029 2.747 0.006
## .Q2_8 0.226 0.068 3.337 0.001
## .Q2_9 0.193 0.075 2.567 0.010
## .Q3_1 0.099 0.032 3.074 0.002
## .Q3_2 0.128 0.072 1.776 0.076
## .Q3_3 0.116 0.071 1.630 0.103
## .Q3_4 0.026 0.016 1.618 0.106
## .Q3_5 0.052 0.023 2.281 0.023
## .Q3_6 0.162 0.073 2.222 0.026
## .Q3_7 0.273 0.101 2.708 0.007
## .Q3_8 0.672 0.263 2.550 0.011
## .Q3_9 0.931 0.229 4.065 0.000
## .Q3_10 0.275 0.104 2.643 0.008
## .Q4_1 0.035 0.021 1.723 0.085
## .Q4_2 0.101 0.042 2.420 0.016
## .Q4_3 0.023 0.014 1.631 0.103
## .Q4_4 0.087 0.068 1.278 0.201
## .Q4_5 0.143 0.059 2.414 0.016
## .Q4_6 0.061 0.026 2.327 0.020
## .Q4_7 0.321 0.093 3.441 0.001
## .Q4_8 0.086 0.034 2.507 0.012
## .Q4_9 0.702 0.150 4.690 0.000
## .Q4_10 0.082 0.047 1.725 0.084
## .Q5_1 0.199 0.086 2.322 0.020
## .Q5_2 0.308 0.109 2.821 0.005
## .Q5_4 0.082 0.034 2.447 0.014
## .Q5_5 0.100 0.043 2.322 0.020
## .Q5_6 0.162 0.071 2.289 0.022
## .Q5_7 0.087 0.032 2.671 0.008
## .Q5_8 0.529 0.269 1.970 0.049
## .Q5_9 0.078 0.031 2.465 0.014
## .Q5_10 0.325 0.118 2.761 0.006
## .Q5_11 0.129 0.056 2.324 0.020
## CSE 1.000
csek12one.metric.nooutlier <- cfa(csek12one.mod4, data = csek12.schoolage1, std.lv = TRUE, missing = "FIML", estimator = "MLR", group = "Parent", group.equal = "loadings") #metric model
## Warning: lavaan->lav_mvnorm_missing_h1_estimate_moments():
## The smallest eigenvalue of the EM estimated variance-covariance matrix
## (Sigma) is smaller than 1e-05; this may cause numerical instabilities;
## interpret the results with caution.
## Warning: lavaan->lav_model_vcov():
## The variance-covariance matrix of the estimated parameters (vcov) does not
## appear to be positive definite! The smallest eigenvalue (= -9.173493e-17)
## is smaller than zero. This may be a symptom that the model is not
## identified.
summary(compareFit(csek12one.metric.nooutlier, csek12one.config.nooutlier))
## ################### Nested Model Comparison #########################
##
## Scaled Chi-Squared Difference Test (method = "satorra.bentler.2001")
##
## lavaan->unknown():
## lavaan NOTE: The "Chisq" column contains standard test statistics, not the
## robust test that should be reported per model. A robust difference test is
## a function of two standard (not robust) statistics.
## Df AIC BIC Chisq Chisq diff Df diff
## csek12one.config.nooutlier 1374 5161.9 5872.4 5373.2
## csek12one.metric.nooutlier 1412 5140.2 5748.4 5427.5 39.249 38
## Pr(>Chisq)
## csek12one.config.nooutlier
## csek12one.metric.nooutlier 0.4137
##
## ####################### Model Fit Indices ###########################
## chisq.scaled df.scaled pvalue.scaled rmsea.robust
## csek12one.config.nooutlier 5339.097† 1374 .000 .229
## csek12one.metric.nooutlier 5339.305 1412 .000 .226†
## cfi.robust tli.robust srmr aic bic
## csek12one.config.nooutlier .517† .479 .080† 5161.922 5872.438
## csek12one.metric.nooutlier .516 .492† .106 5140.187† 5748.431†
##
## ################## Differences in Fit Indices #######################
## df.scaled rmsea.robust
## csek12one.metric.nooutlier - csek12one.config.nooutlier 38 -0.003
## cfi.robust tli.robust
## csek12one.metric.nooutlier - csek12one.config.nooutlier 0 0.014
## srmr aic bic
## csek12one.metric.nooutlier - csek12one.config.nooutlier 0.026 -21.735 -124.007
csek12one.scalar.nooutlier <- cfa(csek12one.mod4, data = csek12.schoolage1, std.lv = TRUE, missing = "FIML", estimator = "MLR", group = "Parent", group.equal = c("loadings", "intercepts")) #metric model
## Warning: lavaan->lav_mvnorm_missing_h1_estimate_moments():
## The smallest eigenvalue of the EM estimated variance-covariance matrix
## (Sigma) is smaller than 1e-05; this may cause numerical instabilities;
## interpret the results with caution.
## Warning: lavaan->lav_model_vcov():
## The variance-covariance matrix of the estimated parameters (vcov) does not
## appear to be positive definite! The smallest eigenvalue (= -2.901192e-18)
## is smaller than zero. This may be a symptom that the model is not
## identified.
summary(compareFit(csek12one.metric.nooutlier, csek12one.scalar.nooutlier))
## ################### Nested Model Comparison #########################
##
## Scaled Chi-Squared Difference Test (method = "satorra.bentler.2001")
##
## lavaan->unknown():
## lavaan NOTE: The "Chisq" column contains standard test statistics, not the
## robust test that should be reported per model. A robust difference test is
## a function of two standard (not robust) statistics.
## Df AIC BIC Chisq Chisq diff Df diff
## csek12one.metric.nooutlier 1412 5140.2 5748.4 5427.5
## csek12one.scalar.nooutlier 1450 5093.2 5599.2 5456.5 32.51 38
## Pr(>Chisq)
## csek12one.metric.nooutlier
## csek12one.scalar.nooutlier 0.7208
##
## ####################### Model Fit Indices ###########################
## chisq.scaled df.scaled pvalue.scaled rmsea.robust
## csek12one.metric.nooutlier 5339.305† 1412 .000 .226
## csek12one.scalar.nooutlier 5385.142 1450 .000 .223†
## cfi.robust tli.robust srmr aic bic
## csek12one.metric.nooutlier .516 .492 .106† 5140.187 5748.431
## csek12one.scalar.nooutlier .517† .507† .107 5093.179† 5599.153†
##
## ################## Differences in Fit Indices #######################
## df.scaled rmsea.robust
## csek12one.scalar.nooutlier - csek12one.metric.nooutlier 38 -0.003
## cfi.robust tli.robust
## csek12one.scalar.nooutlier - csek12one.metric.nooutlier 0.001 0.014
## srmr aic bic
## csek12one.scalar.nooutlier - csek12one.metric.nooutlier 0.001 -47.007 -149.279
#examining invariance by gender when outliers are removed
csek12.gender1 <- subset(csek12.nooutliers, Gender_3cat != 3) #removal of participants who did not identify as either male or female in the no outlier data
csek12one.config1.nooutlier <- cfa(csek12one.mod4, data = csek12.gender1, std.lv = TRUE, missing = "FIML", estimator = "MLR", group = "Gender_3cat")
## Warning: lavaan->lav_model_vcov():
## The variance-covariance matrix of the estimated parameters (vcov) does not
## appear to be positive definite! The smallest eigenvalue (= -5.690865e-16)
## is smaller than zero. This may be a symptom that the model is not
## identified.
csek12one.metric1.nooutlier <- cfa(csek12one.mod4, data = csek12.gender1, std.lv = TRUE, missing = "FIML", estimator = "MLR", group = "Gender_3cat", group.equal = "loadings")
## Warning: lavaan->lav_model_vcov():
## The variance-covariance matrix of the estimated parameters (vcov) does not
## appear to be positive definite! The smallest eigenvalue (= -1.218020e-18)
## is smaller than zero. This may be a symptom that the model is not
## identified.
summary(compareFit(csek12one.metric1.nooutlier, csek12one.config1.nooutlier))
## ################### Nested Model Comparison #########################
##
## Scaled Chi-Squared Difference Test (method = "satorra.bentler.2001")
##
## lavaan->unknown():
## lavaan NOTE: The "Chisq" column contains standard test statistics, not the
## robust test that should be reported per model. A robust difference test is
## a function of two standard (not robust) statistics.
## Df AIC BIC Chisq Chisq diff Df diff
## csek12one.config1.nooutlier 1374 11109 12027 4095.8
## csek12one.metric1.nooutlier 1412 11075 11861 4137.7 28.647 38
## Pr(>Chisq)
## csek12one.config1.nooutlier
## csek12one.metric1.nooutlier 0.8639
##
## ####################### Model Fit Indices ###########################
## chisq.scaled df.scaled pvalue.scaled rmsea.robust
## csek12one.config1.nooutlier 2860.095† 1374 .000 .111
## csek12one.metric1.nooutlier 2887.750 1412 .000 .109†
## cfi.robust tli.robust srmr aic bic
## csek12one.config1.nooutlier .822 .809 .054† 11109.126 12026.913
## csek12one.metric1.nooutlier .823† .814† .073 11074.980† 11860.660†
##
## ################## Differences in Fit Indices #######################
## df.scaled
## csek12one.metric1.nooutlier - csek12one.config1.nooutlier 38
## rmsea.robust
## csek12one.metric1.nooutlier - csek12one.config1.nooutlier -0.002
## cfi.robust tli.robust
## csek12one.metric1.nooutlier - csek12one.config1.nooutlier 0.001 0.006
## srmr aic
## csek12one.metric1.nooutlier - csek12one.config1.nooutlier 0.019 -34.146
## bic
## csek12one.metric1.nooutlier - csek12one.config1.nooutlier -166.252
csek12one.scalar1.nooutlier <- cfa(csek12one.mod4, data = csek12.gender1, std.lv = TRUE, missing = "FIML", estimator = "MLR", group = "Gender_3cat", group.equal = c("loadings", "intercepts"))
## Warning: lavaan->lav_model_vcov():
## The variance-covariance matrix of the estimated parameters (vcov) does not
## appear to be positive definite! The smallest eigenvalue (= 2.989826e-16)
## is close to zero. This may be a symptom that the model is not identified.
summary(compareFit(csek12one.scalar1.nooutlier, csek12one.metric1.nooutlier))
## ################### Nested Model Comparison #########################
##
## Scaled Chi-Squared Difference Test (method = "satorra.bentler.2001")
##
## lavaan->unknown():
## lavaan NOTE: The "Chisq" column contains standard test statistics, not the
## robust test that should be reported per model. A robust difference test is
## a function of two standard (not robust) statistics.
## Df AIC BIC Chisq Chisq diff Df diff
## csek12one.metric1.nooutlier 1412 11075 11861 4137.7
## csek12one.scalar1.nooutlier 1450 11019 11673 4158.0 19.977 38
## Pr(>Chisq)
## csek12one.metric1.nooutlier
## csek12one.scalar1.nooutlier 0.9929
##
## ####################### Model Fit Indices ###########################
## chisq.scaled df.scaled pvalue.scaled rmsea.robust
## csek12one.metric1.nooutlier 2887.750† 1412 .000 .109
## csek12one.scalar1.nooutlier 2924.218 1450 .000 .108†
## cfi.robust tli.robust srmr aic bic
## csek12one.metric1.nooutlier .823 .814 .073† 11074.980 11860.660
## csek12one.scalar1.nooutlier .825† .821† .073 11019.272† 11672.847†
##
## ################## Differences in Fit Indices #######################
## df.scaled
## csek12one.scalar1.nooutlier - csek12one.metric1.nooutlier 38
## rmsea.robust
## csek12one.scalar1.nooutlier - csek12one.metric1.nooutlier -0.002
## cfi.robust tli.robust
## csek12one.scalar1.nooutlier - csek12one.metric1.nooutlier 0.002 0.006
## srmr aic
## csek12one.scalar1.nooutlier - csek12one.metric1.nooutlier 0.001 -55.708
## bic
## csek12one.scalar1.nooutlier - csek12one.metric1.nooutlier -187.813
#examining invariance by education when outliers are removed
csek12one.config2.nooutliers <- cfa(csek12one.mod4, data = csek12.nooutliers, std.lv = TRUE, missing = "FIML", estimator = "MLR", group = "highered") #configural model
## Warning: lavaan->lav_model_vcov():
## The variance-covariance matrix of the estimated parameters (vcov) does not
## appear to be positive definite! The smallest eigenvalue (= -1.257637e-15)
## is smaller than zero. This may be a symptom that the model is not
## identified.
lavInspect(csek12one.config2.nooutliers, "cov.lv")
## $`1`
## CSE
## CSE 1
##
## $`0`
## CSE
## CSE 1
csek12one.metric2.nooutliers <- cfa(csek12one.mod4, data = csek12.nooutliers, std.lv = TRUE, missing = "FIML", estimator = "MLR", group = "highered", group.equal = "loadings") #metric model
## Warning: lavaan->lav_model_vcov():
## The variance-covariance matrix of the estimated parameters (vcov) does not
## appear to be positive definite! The smallest eigenvalue (= -2.690439e-18)
## is smaller than zero. This may be a symptom that the model is not
## identified.
summary(compareFit(csek12one.metric2.nooutliers, csek12one.config2.nooutliers))
## ################### Nested Model Comparison #########################
##
## Scaled Chi-Squared Difference Test (method = "satorra.bentler.2001")
##
## lavaan->unknown():
## lavaan NOTE: The "Chisq" column contains standard test statistics, not the
## robust test that should be reported per model. A robust difference test is
## a function of two standard (not robust) statistics.
## Df AIC BIC Chisq Chisq diff Df diff
## csek12one.config2.nooutliers 1374 11353 12284 4247.4
## csek12one.metric2.nooutliers 1412 11330 12127 4300.3 38.864 38
## Pr(>Chisq)
## csek12one.config2.nooutliers
## csek12one.metric2.nooutliers 0.4306
##
## ####################### Model Fit Indices ###########################
## chisq.scaled df.scaled pvalue.scaled rmsea.robust
## csek12one.config2.nooutliers 2906.883† 1374 .000 .112
## csek12one.metric2.nooutliers 2948.460 1412 .000 .110†
## cfi.robust tli.robust srmr aic bic
## csek12one.config2.nooutliers .827† .813 .052† 11353.248 12283.968
## csek12one.metric2.nooutliers .826 .818† .070 11330.229† 12126.981†
##
## ################## Differences in Fit Indices #######################
## df.scaled
## csek12one.metric2.nooutliers - csek12one.config2.nooutliers 38
## rmsea.robust
## csek12one.metric2.nooutliers - csek12one.config2.nooutliers -0.001
## cfi.robust
## csek12one.metric2.nooutliers - csek12one.config2.nooutliers 0
## tli.robust srmr
## csek12one.metric2.nooutliers - csek12one.config2.nooutliers 0.005 0.017
## aic bic
## csek12one.metric2.nooutliers - csek12one.config2.nooutliers -23.019 -156.986
csek12one.scalar2.nooutliers <- cfa(csek12one.mod4, data = csek12.nooutliers, std.lv = TRUE, missing = "FIML", estimator = "MLR", group = "highered", group.equal = c("loadings", "intercepts")) #scalar model
## Warning: lavaan->lav_model_vcov():
## The variance-covariance matrix of the estimated parameters (vcov) does not
## appear to be positive definite! The smallest eigenvalue (= 1.483280e-16)
## is close to zero. This may be a symptom that the model is not identified.
summary(compareFit(csek12one.scalar2.nooutliers, csek12one.metric2.nooutliers))
## ################### Nested Model Comparison #########################
##
## Scaled Chi-Squared Difference Test (method = "satorra.bentler.2001")
##
## lavaan->unknown():
## lavaan NOTE: The "Chisq" column contains standard test statistics, not the
## robust test that should be reported per model. A robust difference test is
## a function of two standard (not robust) statistics.
## Df AIC BIC Chisq Chisq diff Df diff
## csek12one.metric2.nooutliers 1412 11330 12127 4300.3
## csek12one.scalar2.nooutliers 1450 11304 11967 4350.1 54.357 38
## Pr(>Chisq)
## csek12one.metric2.nooutliers
## csek12one.scalar2.nooutliers 0.04151 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## ####################### Model Fit Indices ###########################
## chisq.scaled df.scaled pvalue.scaled rmsea.robust
## csek12one.metric2.nooutliers 2948.460† 1412 .000 .110
## csek12one.scalar2.nooutliers 3011.970 1450 .000 .109†
## cfi.robust tli.robust srmr aic bic
## csek12one.metric2.nooutliers .826† .818 .070† 11330.229 12126.981
## csek12one.scalar2.nooutliers .825 .822† .070 11303.982† 11966.767†
##
## ################## Differences in Fit Indices #######################
## df.scaled
## csek12one.scalar2.nooutliers - csek12one.metric2.nooutliers 38
## rmsea.robust
## csek12one.scalar2.nooutliers - csek12one.metric2.nooutliers -0.001
## cfi.robust
## csek12one.scalar2.nooutliers - csek12one.metric2.nooutliers -0.001
## tli.robust srmr
## csek12one.scalar2.nooutliers - csek12one.metric2.nooutliers 0.004 0.001
## aic bic
## csek12one.scalar2.nooutliers - csek12one.metric2.nooutliers -26.247 -160.214
#running validity tests w/ outliers removed
csek12one.val3.nooutliers <- '
CSE =~ Q2_1 + Q2_2 + Q2_3 + Q2_4 + Q2_5 + Q2_6 + Q2_7 + Q2_8 + Q2_9 + Q3_1 + Q3_2 + Q3_3 + Q3_4 + Q3_5 + Q3_6 + Q3_7 + Q3_8 + Q3_9 + Q3_10 + Q4_1 + Q4_2 + Q4_3 + Q4_4 + Q4_5 + Q4_6 + Q4_7 + Q4_8 + Q4_9 + Q4_10 + Q5_1 + Q5_2 + Q5_4 + Q5_5 + Q5_6 + Q5_7 + Q5_8 + Q5_9 + Q5_10 + Q5_11
#covariances
Q2_3 ~~ Q2_4
Q2_3 ~~ Q2_5
Q2_3 ~~ Q5_5
Q2_3 ~~ Q5_6
Q2_4 ~~ Q2_5
Q2_4 ~~ Q5_5
Q2_4 ~~ Q5_6
Q2_5 ~~ Q5_5
Q2_5 ~~ Q5_6
Q5_5 ~~ Q5_6
Q4_4 ~~ Q4_5
Q4_4 ~~ Q4_6
Q4_5 ~~ Q4_6
Q4_7 ~~ Q4_8
Q2_8 ~~ Q2_9
CSE ~~ Important_Middle
CSE ~~ Important_High
CSE ~~ Sex_Pos
CSE ~~ Sex_Neg
CSE ~~ PolAff
#regressions
Sex_Pos ~ Vote1
Sex_Neg ~ Vote1
'
csek12one.fit.val3.nooutliers <- cfa(csek12one.val3.nooutliers, data = csek12.nooutliers, std.lv = TRUE, estimator = "MLR")
summary(csek12one.fit.val3.nooutliers, fit.measures = TRUE, standardized = TRUE)
## lavaan 0.6-18 ended normally after 218 iterations
##
## Estimator ML
## Optimization method NLMINB
## Number of model parameters 106
##
## Used Total
## Number of observations 231 251
##
## Model Test User Model:
## Standard Scaled
## Test Statistic 3208.422 2066.647
## Degrees of freedom 928 928
## P-value (Chi-square) 0.000 0.000
## Scaling correction factor 1.552
## Yuan-Bentler correction (Mplus variant)
##
## Model Test Baseline Model:
##
## Test statistic 12776.441 7745.073
## Degrees of freedom 990 990
## P-value 0.000 0.000
## Scaling correction factor 1.650
##
## User Model versus Baseline Model:
##
## Comparative Fit Index (CFI) 0.807 0.831
## Tucker-Lewis Index (TLI) 0.794 0.820
##
## Robust Comparative Fit Index (CFI) 0.841
## Robust Tucker-Lewis Index (TLI) 0.831
##
## Loglikelihood and Information Criteria:
##
## Loglikelihood user model (H0) -7040.482 -7040.482
## Scaling correction factor 3.371
## for the MLR correction
## Loglikelihood unrestricted model (H1) NA NA
## Scaling correction factor 1.739
## for the MLR correction
##
## Akaike (AIC) 14292.965 14292.965
## Bayesian (BIC) 14657.861 14657.861
## Sample-size adjusted Bayesian (SABIC) 14321.901 14321.901
##
## Root Mean Square Error of Approximation:
##
## RMSEA 0.103 0.073
## 90 Percent confidence interval - lower 0.099 0.069
## 90 Percent confidence interval - upper 0.107 0.076
## P-value H_0: RMSEA <= 0.050 0.000 0.000
## P-value H_0: RMSEA >= 0.080 1.000 0.000
##
## Robust RMSEA 0.091
## 90 Percent confidence interval - lower 0.086
## 90 Percent confidence interval - upper 0.096
## P-value H_0: Robust RMSEA <= 0.050 0.000
## P-value H_0: Robust RMSEA >= 0.080 1.000
##
## Standardized Root Mean Square Residual:
##
## SRMR 0.194 0.194
##
## 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
## CSE =~
## Q2_1 0.608 0.059 10.273 0.000 0.608 0.810
## Q2_2 0.509 0.056 9.015 0.000 0.509 0.576
## Q2_3 0.481 0.045 10.757 0.000 0.481 0.840
## Q2_4 0.552 0.050 11.063 0.000 0.552 0.787
## Q2_5 0.530 0.055 9.642 0.000 0.530 0.767
## Q2_6 0.457 0.062 7.421 0.000 0.457 0.740
## Q2_7 0.472 0.069 6.806 0.000 0.472 0.821
## Q2_8 0.446 0.056 7.922 0.000 0.446 0.673
## Q2_9 0.380 0.062 6.135 0.000 0.380 0.609
## Q3_1 0.537 0.050 10.680 0.000 0.537 0.900
## Q3_2 0.508 0.054 9.467 0.000 0.508 0.862
## Q3_3 0.468 0.049 9.534 0.000 0.468 0.863
## Q3_4 0.530 0.051 10.420 0.000 0.530 0.893
## Q3_5 0.476 0.050 9.546 0.000 0.476 0.849
## Q3_6 0.367 0.047 7.772 0.000 0.367 0.703
## Q3_7 0.598 0.057 10.426 0.000 0.598 0.762
## Q3_8 0.674 0.066 10.198 0.000 0.674 0.644
## Q3_9 0.732 0.065 11.254 0.000 0.732 0.605
## Q3_10 0.484 0.056 8.687 0.000 0.484 0.687
## Q4_1 0.535 0.052 10.324 0.000 0.535 0.899
## Q4_2 0.520 0.053 9.839 0.000 0.520 0.756
## Q4_3 0.498 0.056 8.942 0.000 0.498 0.920
## Q4_4 0.469 0.058 8.141 0.000 0.469 0.884
## Q4_5 0.435 0.057 7.666 0.000 0.435 0.781
## Q4_6 0.532 0.054 9.929 0.000 0.532 0.876
## Q4_7 0.603 0.055 10.900 0.000 0.603 0.816
## Q4_8 0.398 0.056 7.153 0.000 0.398 0.768
## Q4_9 0.311 0.069 4.469 0.000 0.311 0.329
## Q4_10 0.491 0.055 8.871 0.000 0.491 0.825
## Q5_1 0.517 0.048 10.707 0.000 0.517 0.781
## Q5_2 0.444 0.052 8.510 0.000 0.444 0.705
## Q5_4 0.436 0.058 7.545 0.000 0.436 0.838
## Q5_5 0.379 0.044 8.650 0.000 0.379 0.686
## Q5_6 0.429 0.049 8.830 0.000 0.429 0.780
## Q5_7 0.449 0.053 8.546 0.000 0.449 0.827
## Q5_8 0.353 0.061 5.750 0.000 0.353 0.425
## Q5_9 0.407 0.046 8.868 0.000 0.407 0.819
## Q5_10 0.409 0.062 6.603 0.000 0.409 0.601
## Q5_11 0.469 0.059 7.939 0.000 0.469 0.870
##
## Regressions:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## Sex_Pos ~
## Vote1 5.096 2.697 1.890 0.059 5.096 0.166
## Sex_Neg ~
## Vote1 -2.395 1.266 -1.892 0.059 -2.395 -0.160
##
## Covariances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .Q2_3 ~~
## .Q2_4 0.011 0.013 0.864 0.387 0.011 0.085
## .Q2_5 0.011 0.015 0.785 0.432 0.011 0.083
## .Q5_5 0.014 0.015 0.949 0.343 0.014 0.116
## .Q5_6 0.020 0.012 1.625 0.104 0.020 0.185
## .Q2_4 ~~
## .Q2_5 0.084 0.022 3.761 0.000 0.084 0.439
## .Q5_5 0.030 0.017 1.763 0.078 0.030 0.174
## .Q5_6 -0.024 0.012 -2.055 0.040 -0.024 -0.163
## .Q2_5 ~~
## .Q5_5 0.019 0.017 1.105 0.269 0.019 0.105
## .Q5_6 -0.027 0.015 -1.831 0.067 -0.027 -0.178
## .Q5_5 ~~
## .Q5_6 0.017 0.017 1.024 0.306 0.017 0.123
## .Q4_4 ~~
## .Q4_5 0.016 0.010 1.641 0.101 0.016 0.188
## .Q4_6 0.007 0.010 0.674 0.500 0.007 0.096
## .Q4_5 ~~
## .Q4_6 0.033 0.012 2.733 0.006 0.033 0.327
## .Q4_7 ~~
## .Q4_8 -0.002 0.017 -0.098 0.922 -0.002 -0.012
## .Q2_8 ~~
## .Q2_9 0.123 0.034 3.591 0.000 0.123 0.509
## CSE ~~
## Important_Mddl -0.426 0.099 -4.320 0.000 -0.426 -0.466
## Important_High -0.190 0.084 -2.271 0.023 -0.190 -0.368
## .Sex_Pos 0.752 0.610 1.232 0.218 0.752 0.095
## .Sex_Neg 0.225 0.389 0.579 0.563 0.225 0.058
## PolAff 0.205 0.146 1.408 0.159 0.205 0.121
## .Sex_Pos ~~
## .Sex_Neg -15.606 5.757 -2.711 0.007 -15.606 -0.511
##
## Variances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .Q2_1 0.194 0.037 5.291 0.000 0.194 0.344
## .Q2_2 0.522 0.089 5.871 0.000 0.522 0.668
## .Q2_3 0.097 0.017 5.681 0.000 0.097 0.295
## .Q2_4 0.187 0.033 5.694 0.000 0.187 0.381
## .Q2_5 0.196 0.035 5.585 0.000 0.196 0.411
## .Q2_6 0.172 0.048 3.612 0.000 0.172 0.452
## .Q2_7 0.108 0.022 4.836 0.000 0.108 0.326
## .Q2_8 0.240 0.040 5.946 0.000 0.240 0.547
## .Q2_9 0.245 0.043 5.735 0.000 0.245 0.629
## .Q3_1 0.068 0.012 5.468 0.000 0.068 0.191
## .Q3_2 0.089 0.026 3.471 0.001 0.089 0.257
## .Q3_3 0.075 0.023 3.222 0.001 0.075 0.255
## .Q3_4 0.072 0.019 3.784 0.000 0.072 0.203
## .Q3_5 0.088 0.019 4.709 0.000 0.088 0.279
## .Q3_6 0.138 0.036 3.875 0.000 0.138 0.505
## .Q3_7 0.258 0.044 5.852 0.000 0.258 0.419
## .Q3_8 0.639 0.125 5.098 0.000 0.639 0.585
## .Q3_9 0.930 0.136 6.844 0.000 0.930 0.634
## .Q3_10 0.263 0.055 4.799 0.000 0.263 0.528
## .Q4_1 0.068 0.019 3.563 0.000 0.068 0.191
## .Q4_2 0.202 0.050 4.025 0.000 0.202 0.428
## .Q4_3 0.045 0.008 5.442 0.000 0.045 0.153
## .Q4_4 0.062 0.018 3.454 0.001 0.062 0.219
## .Q4_5 0.121 0.026 4.689 0.000 0.121 0.390
## .Q4_6 0.086 0.015 5.751 0.000 0.086 0.233
## .Q4_7 0.183 0.036 5.038 0.000 0.183 0.335
## .Q4_8 0.110 0.024 4.589 0.000 0.110 0.410
## .Q4_9 0.795 0.115 6.942 0.000 0.795 0.892
## .Q4_10 0.113 0.026 4.407 0.000 0.113 0.319
## .Q5_1 0.171 0.035 4.856 0.000 0.171 0.390
## .Q5_2 0.200 0.050 3.967 0.000 0.200 0.503
## .Q5_4 0.081 0.020 3.967 0.000 0.081 0.298
## .Q5_5 0.161 0.029 5.527 0.000 0.161 0.529
## .Q5_6 0.119 0.023 5.126 0.000 0.119 0.392
## .Q5_7 0.093 0.020 4.567 0.000 0.093 0.316
## .Q5_8 0.566 0.117 4.841 0.000 0.566 0.819
## .Q5_9 0.081 0.015 5.327 0.000 0.081 0.330
## .Q5_10 0.297 0.050 5.940 0.000 0.297 0.639
## .Q5_11 0.071 0.016 4.430 0.000 0.071 0.243
## .Sex_Pos 62.581 7.620 8.212 0.000 62.581 0.972
## .Sex_Neg 14.891 6.199 2.402 0.016 14.891 0.974
## Important_Mddl 0.833 0.115 7.216 0.000 0.833 1.000
## Important_High 0.266 0.101 2.624 0.009 0.266 1.000
## PolAff 2.867 0.247 11.606 0.000 2.867 1.000
## CSE 1.000 1.000 1.000
psych::describeBy(csek12.nooutliers$Sex_Pos, group = csek12.nooutliers$Vote1)
##
## Descriptive statistics by group
## group: 1
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 17 33.94 10.09 33 34.2 13.34 16 48 32 -0.3 -1.11 2.45
## ------------------------------------------------------------
## group: 2
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 219 40.1 7.73 40 41.11 8.9 8 48 40 -1.18 1.73 0.52
psych::describeBy(csek12.nooutliers$Sex_Neg, group = csek12.nooutliers$Vote1)
##
## Descriptive statistics by group
## group: 1
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 17 12.65 4.8 10 12.2 2.97 8 24 16 0.77 -0.57 1.16
## ------------------------------------------------------------
## group: 2
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 219 10.57 3.76 10 9.92 2.97 8 48 40 5.24 44.1 0.25
#running validity tests w/ outliers removed
csek12one.val3.nooutliers1 <- '
CSE =~ Q2_1 + Q2_2 + Q2_3 + Q2_4 + Q2_5 + Q2_6 + Q2_7 + Q2_8 + Q2_9 + Q3_1 + Q3_2 + Q3_3 + Q3_4 + Q3_5 + Q3_6 + Q3_7 + Q3_8 + Q3_9 + Q3_10 + Q4_1 + Q4_2 + Q4_3 + Q4_4 + Q4_5 + Q4_6 + Q4_7 + Q4_8 + Q4_9 + Q4_10 + Q5_1 + Q5_2 + Q5_4 + Q5_5 + Q5_6 + Q5_7 + Q5_8 + Q5_9 + Q5_10 + Q5_11
#covariances
Q2_3 ~~ Q2_4
Q2_3 ~~ Q2_5
Q2_3 ~~ Q5_5
Q2_3 ~~ Q5_6
Q2_4 ~~ Q2_5
Q2_4 ~~ Q5_5
Q2_4 ~~ Q5_6
Q2_5 ~~ Q5_5
Q2_5 ~~ Q5_6
Q5_5 ~~ Q5_6
Q4_4 ~~ Q4_5
Q4_4 ~~ Q4_6
Q4_5 ~~ Q4_6
Q4_7 ~~ Q4_8
Q2_8 ~~ Q2_9
CSE ~~ Important_Middle
CSE ~~ Important_High
CSE ~~ Sex_Pos
CSE ~~ Sex_Neg
CSE ~~ PolAff
#regressions
Vote1 ~ CSE
'
csek12one.fit.val3.nooutliers1 <- cfa(csek12one.val3.nooutliers1, data = csek12.nooutliers, std.lv = TRUE, estimator = "DWLS")
## Warning: lavaan->lav_options_est_dwls():
## estimator "DWLS" is not recommended for continuous data. Did you forget to
## set the ordered= argument?
## Warning: lavaan->lav_data_full():
## some observed variances are (at least) a factor 1000 times larger than
## others; use varTable(fit) to investigate
## Warning: lavaan->lav_samplestats_from_data():
## number of observations (231) too small to compute Gamma
summary(csek12one.fit.val3.nooutliers1, fit.measures = TRUE, standardized = TRUE)
## lavaan 0.6-18 ended normally after 91 iterations
##
## Estimator DWLS
## Optimization method NLMINB
## Number of model parameters 105
##
## Used Total
## Number of observations 231 251
##
## Model Test User Model:
##
## Test statistic 301.901
## Degrees of freedom 930
## P-value (Chi-square) 1.000
##
## Model Test Baseline Model:
##
## Test statistic 19157.721
## Degrees of freedom 990
## P-value 0.000
##
## User Model versus Baseline Model:
##
## Comparative Fit Index (CFI) 1.000
## Tucker-Lewis Index (TLI) 1.037
##
## Root Mean Square Error of Approximation:
##
## RMSEA 0.000
## 90 Percent confidence interval - lower 0.000
## 90 Percent confidence interval - upper 0.000
## P-value H_0: RMSEA <= 0.050 1.000
## P-value H_0: RMSEA >= 0.080 0.000
##
## Standardized Root Mean Square Residual:
##
## SRMR 0.060
##
## Parameter Estimates:
##
## Standard errors Standard
## Information Expected
## Information saturated (h1) model Unstructured
##
## Latent Variables:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## CSE =~
## Q2_1 0.696 0.022 31.964 0.000 0.696 0.852
## Q2_2 0.562 0.018 31.021 0.000 0.562 0.608
## Q2_3 0.539 0.018 30.781 0.000 0.539 0.860
## Q2_4 0.640 0.021 31.148 0.000 0.640 0.843
## Q2_5 0.607 0.021 28.917 0.000 0.607 0.814
## Q2_6 0.518 0.019 27.309 0.000 0.518 0.782
## Q2_7 0.519 0.019 26.823 0.000 0.519 0.828
## Q2_8 0.511 0.018 28.327 0.000 0.511 0.727
## Q2_9 0.425 0.017 25.536 0.000 0.425 0.649
## Q3_1 0.602 0.019 32.109 0.000 0.602 0.912
## Q3_2 0.570 0.019 30.339 0.000 0.570 0.880
## Q3_3 0.528 0.018 30.134 0.000 0.528 0.886
## Q3_4 0.598 0.019 31.771 0.000 0.598 0.911
## Q3_5 0.527 0.018 29.855 0.000 0.527 0.859
## Q3_6 0.408 0.015 26.823 0.000 0.408 0.732
## Q3_7 0.685 0.022 31.450 0.000 0.685 0.809
## Q3_8 0.767 0.025 31.239 0.000 0.767 0.694
## Q3_9 0.846 0.025 34.281 0.000 0.846 0.665
## Q3_10 0.563 0.019 30.339 0.000 0.563 0.751
## Q4_1 0.604 0.019 31.738 0.000 0.604 0.918
## Q4_2 0.585 0.019 31.245 0.000 0.585 0.790
## Q4_3 0.553 0.019 29.436 0.000 0.553 0.919
## Q4_4 0.523 0.019 27.937 0.000 0.523 0.894
## Q4_5 0.487 0.018 27.617 0.000 0.487 0.810
## Q4_6 0.601 0.020 30.231 0.000 0.601 0.898
## Q4_7 0.687 0.021 32.176 0.000 0.687 0.852
## Q4_8 0.441 0.017 25.548 0.000 0.441 0.789
## Q4_9 0.364 0.018 20.365 0.000 0.364 0.380
## Q4_10 0.557 0.018 30.253 0.000 0.557 0.857
## Q5_1 0.595 0.018 32.183 0.000 0.595 0.830
## Q5_2 0.495 0.017 28.452 0.000 0.495 0.736
## Q5_4 0.482 0.017 27.748 0.000 0.482 0.848
## Q5_5 0.436 0.015 28.585 0.000 0.436 0.741
## Q5_6 0.487 0.018 27.599 0.000 0.487 0.819
## Q5_7 0.509 0.017 29.334 0.000 0.509 0.859
## Q5_8 0.391 0.017 22.580 0.000 0.391 0.459
## Q5_9 0.458 0.015 29.742 0.000 0.458 0.846
## Q5_10 0.465 0.018 26.148 0.000 0.465 0.650
## Q5_11 0.526 0.019 28.420 0.000 0.526 0.887
##
## Regressions:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## Vote1 ~
## CSE 0.184 0.008 23.845 0.000 0.184 0.704
##
## Covariances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .Q2_3 ~~
## .Q2_4 0.006 0.060 0.093 0.926 0.006 0.043
## .Q2_5 0.010 0.062 0.162 0.872 0.010 0.073
## .Q5_5 0.013 0.047 0.268 0.788 0.013 0.100
## .Q5_6 0.021 0.055 0.377 0.706 0.021 0.190
## .Q2_4 ~~
## .Q2_5 0.070 0.078 0.896 0.370 0.070 0.393
## .Q5_5 0.019 0.055 0.341 0.733 0.019 0.116
## .Q5_6 -0.034 0.061 -0.559 0.576 -0.034 -0.245
## .Q2_5 ~~
## .Q5_5 0.011 0.055 0.198 0.843 0.011 0.064
## .Q5_6 -0.033 0.063 -0.516 0.606 -0.033 -0.221
## .Q5_5 ~~
## .Q5_6 0.012 0.050 0.248 0.804 0.012 0.091
## .Q4_4 ~~
## .Q4_5 0.021 0.064 0.334 0.738 0.021 0.231
## .Q4_6 0.011 0.070 0.158 0.874 0.011 0.144
## .Q4_5 ~~
## .Q4_6 0.036 0.067 0.532 0.594 0.036 0.343
## .Q4_7 ~~
## .Q4_8 0.002 0.070 0.029 0.977 0.002 0.014
## .Q2_8 ~~
## .Q2_9 0.123 0.065 1.892 0.059 0.123 0.509
## CSE ~~
## Important_Mddl -0.572 0.021 -27.764 0.000 -0.572 -0.626
## Important_High -0.286 0.017 -16.612 0.000 -0.286 -0.554
## Sex_Pos 2.506 0.113 22.110 0.000 2.506 0.310
## Sex_Neg -0.426 0.049 -8.661 0.000 -0.426 -0.109
## PolAff 0.745 0.027 28.003 0.000 0.745 0.439
##
## Variances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .Q2_1 0.182 0.116 1.572 0.116 0.182 0.273
## .Q2_2 0.540 0.120 4.510 0.000 0.540 0.631
## .Q2_3 0.102 0.063 1.632 0.103 0.102 0.260
## .Q2_4 0.167 0.081 2.067 0.039 0.167 0.290
## .Q2_5 0.188 0.085 2.207 0.027 0.188 0.337
## .Q2_6 0.171 0.084 2.030 0.042 0.171 0.389
## .Q2_7 0.124 0.086 1.433 0.152 0.124 0.315
## .Q2_8 0.234 0.073 3.185 0.001 0.234 0.472
## .Q2_9 0.249 0.066 3.799 0.000 0.249 0.579
## .Q3_1 0.074 0.073 1.011 0.312 0.074 0.169
## .Q3_2 0.094 0.074 1.280 0.201 0.094 0.225
## .Q3_3 0.076 0.066 1.155 0.248 0.076 0.215
## .Q3_4 0.073 0.074 0.985 0.324 0.073 0.170
## .Q3_5 0.099 0.069 1.427 0.153 0.099 0.263
## .Q3_6 0.144 0.057 2.533 0.011 0.144 0.464
## .Q3_7 0.247 0.116 2.134 0.033 0.247 0.345
## .Q3_8 0.633 0.172 3.675 0.000 0.633 0.518
## .Q3_9 0.903 0.176 5.123 0.000 0.903 0.558
## .Q3_10 0.245 0.088 2.793 0.005 0.245 0.436
## .Q4_1 0.068 0.077 0.887 0.375 0.068 0.158
## .Q4_2 0.206 0.080 2.573 0.010 0.206 0.376
## .Q4_3 0.056 0.074 0.763 0.446 0.056 0.155
## .Q4_4 0.069 0.072 0.959 0.337 0.069 0.201
## .Q4_5 0.125 0.069 1.808 0.071 0.125 0.344
## .Q4_6 0.086 0.079 1.088 0.277 0.086 0.193
## .Q4_7 0.177 0.101 1.751 0.080 0.177 0.273
## .Q4_8 0.118 0.071 1.679 0.093 0.118 0.378
## .Q4_9 0.789 0.113 6.990 0.000 0.789 0.856
## .Q4_10 0.112 0.073 1.534 0.125 0.112 0.265
## .Q5_1 0.160 0.072 2.223 0.026 0.160 0.311
## .Q5_2 0.207 0.073 2.832 0.005 0.207 0.458
## .Q5_4 0.091 0.068 1.344 0.179 0.091 0.281
## .Q5_5 0.156 0.052 3.000 0.003 0.156 0.451
## .Q5_6 0.116 0.064 1.824 0.068 0.116 0.329
## .Q5_7 0.092 0.066 1.387 0.165 0.092 0.262
## .Q5_8 0.574 0.117 4.907 0.000 0.574 0.790
## .Q5_9 0.083 0.055 1.524 0.128 0.083 0.284
## .Q5_10 0.295 0.071 4.162 0.000 0.295 0.577
## .Q5_11 0.075 0.073 1.029 0.303 0.075 0.213
## .Vote1 0.035 0.015 2.308 0.021 0.035 0.504
## Important_Mddl 0.837 0.116 7.232 0.000 0.837 1.000
## Important_High 0.267 0.101 2.630 0.009 0.267 1.000
## Sex_Pos 65.394 7.884 8.294 0.000 65.394 1.000
## Sex_Neg 15.241 6.162 2.473 0.013 15.241 1.000
## PolAff 2.879 0.248 11.631 0.000 2.879 1.000
## CSE 1.000 1.000 1.000