Christian Nationalism Scale Analyses
Andres Pinedo
2025-01-24
#Load packages
#load psych package
library(psych)
#load tidyverse package
library(tidyverse)
library(viridis) # For color-blind-friendly palettes
#lavaan for cfa and sem
library(lavaan)
#load corrplot
library(corrplot)
library(ggcorrplot)#Read in data
YNEES_W1 <- read.csv("YNEES.Data.Clean.csv")
YNEES <- read.csv("YNEES.W1andW2_Data.Clean.csv")#Create White vs POC variable
#table for race variable
table(YNEES_W1$race)##
## Asian Black Latinx Middle Eastern Multiracial
## 34 122 77 3 31
## Native American White
## 14 145#Create White vs POC Variable
YNEES_W1$race2 <- ifelse(YNEES_W1$race == "White", 0, 1 )
table(YNEES_W1$race2)##
## 0 1
## 145 281# Label the AgeGroup variable
YNEES_W1$race2 <- factor(YNEES_W1$race2,
levels = c(0, 1),
labels = c("White", "People of Color"))##Check descriptives for Christian Nat items at Wave 1
#Create matrix of cns variables at time 1
# 'YNEES' is the dataframe we're working with
CNatvars <- YNEES_W1 %>%
select(cns_1, cns_2, cns_3, cns_4r, cns_5, cns_6r) %>%
as.matrix()
# View the matrix
#CNatvars
#descriptives for cns vars
describe(CNatvars)## vars n mean sd median trimmed mad min max range skew kurtosis se
## cns_1 1 427 3.44 1.81 4 3.43 2.97 1 6 5 -0.02 -1.35 0.09
## cns_2 2 427 3.49 1.74 4 3.49 1.48 1 6 5 -0.15 -1.29 0.08
## cns_3 3 426 3.41 1.81 4 3.38 2.97 1 6 5 -0.03 -1.37 0.09
## cns_4r 4 424 2.66 1.54 2 2.46 1.48 1 6 5 0.76 -0.35 0.07
## cns_5 5 424 3.78 1.79 4 3.84 1.48 1 6 5 -0.37 -1.21 0.09
## cns_6r 6 426 3.06 1.74 3 2.96 1.48 1 6 5 0.39 -1.12 0.08#create a correlation matrix for cns vars
CNCorMat <-cor(CNatvars, use = "complete.obs")
#Now visualize the correlation matrix with GGplot and corrplot
ggcorrplot(CNCorMat, hc.order = TRUE, type = "lower", lab = TRUE, ggtheme = ggplot2::theme_gray,
colors = c("#6D9EC1", "white", "#E46726"))
#CFA for christian nationalism
#CFA for Critical Nationalism Measures
Cnat_CFA <- 'C_nat =~ 1*cns_1 + cns_2 + cns_3 + cns_4r + cns_5 + cns_6r'
Cnat_CFA <- sem(Cnat_CFA, data = YNEES_W1, estimator = "MLR") #robust estimator with "MLR"
summary(Cnat_CFA, fit.measures = T, standardized = T, rsquare = T)## lavaan 0.6-19 ended normally after 25 iterations
##
## Estimator ML
## Optimization method NLMINB
## Number of model parameters 12
##
## Used Total
## Number of observations 421 427
##
## Model Test User Model:
## Standard Scaled
## Test Statistic 56.560 44.611
## Degrees of freedom 9 9
## P-value (Chi-square) 0.000 0.000
## Scaling correction factor 1.268
## Yuan-Bentler correction (Mplus variant)
##
## Model Test Baseline Model:
##
## Test statistic 1441.705 863.940
## Degrees of freedom 15 15
## P-value 0.000 0.000
## Scaling correction factor 1.669
##
## User Model versus Baseline Model:
##
## Comparative Fit Index (CFI) 0.967 0.958
## Tucker-Lewis Index (TLI) 0.944 0.930
##
## Robust Comparative Fit Index (CFI) 0.968
## Robust Tucker-Lewis Index (TLI) 0.947
##
## Loglikelihood and Information Criteria:
##
## Loglikelihood user model (H0) -4284.376 -4284.376
## Scaling correction factor 1.357
## for the MLR correction
## Loglikelihood unrestricted model (H1) -4256.096 -4256.096
## Scaling correction factor 1.319
## for the MLR correction
##
## Akaike (AIC) 8592.752 8592.752
## Bayesian (BIC) 8641.263 8641.263
## Sample-size adjusted Bayesian (SABIC) 8603.183 8603.183
##
## Root Mean Square Error of Approximation:
##
## RMSEA 0.112 0.097
## 90 Percent confidence interval - lower 0.085 0.073
## 90 Percent confidence interval - upper 0.141 0.123
## P-value H_0: RMSEA <= 0.050 0.000 0.001
## P-value H_0: RMSEA >= 0.080 0.974 0.879
##
## Robust RMSEA 0.109
## 90 Percent confidence interval - lower 0.078
## 90 Percent confidence interval - upper 0.142
## P-value H_0: Robust RMSEA <= 0.050 0.001
## P-value H_0: Robust RMSEA >= 0.080 0.942
##
## Standardized Root Mean Square Residual:
##
## SRMR 0.054 0.054
##
## 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
## C_nat =~
## cns_1 1.000 1.668 0.922
## cns_2 0.949 0.029 32.432 0.000 1.583 0.911
## cns_3 0.910 0.032 28.393 0.000 1.518 0.840
## cns_4r -0.174 0.055 -3.153 0.002 -0.291 -0.189
## cns_5 0.865 0.038 22.732 0.000 1.443 0.807
## cns_6r 0.278 0.057 4.873 0.000 0.464 0.265
##
## Variances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .cns_1 0.488 0.091 5.365 0.000 0.488 0.149
## .cns_2 0.512 0.089 5.752 0.000 0.512 0.170
## .cns_3 0.963 0.124 7.795 0.000 0.963 0.295
## .cns_4r 2.284 0.141 16.147 0.000 2.284 0.964
## .cns_5 1.116 0.140 7.987 0.000 1.116 0.349
## .cns_6r 2.840 0.139 20.448 0.000 2.840 0.930
## C_nat 2.783 0.159 17.556 0.000 1.000 1.000
##
## R-Square:
## Estimate
## cns_1 0.851
## cns_2 0.830
## cns_3 0.705
## cns_4r 0.036
## cns_5 0.651
## cns_6r 0.070resid(Cnat_CFA)## $type
## [1] "raw"
##
## $cov
## cns_1 cns_2 cns_3 cns_4r cns_5 cns_6r
## cns_1 0.000
## cns_2 -0.007 0.000
## cns_3 0.043 -0.018 0.000
## cns_4r 0.033 -0.047 0.037 0.000
## cns_5 -0.023 0.022 -0.035 -0.088 0.000
## cns_6r -0.079 0.102 -0.249 0.536 0.319 0.000##Item 6r “religion has no place in government” is throwing off the fit for the CFA. We can make an argument to drop it or think of a rationale to keep it despite questionable fit. I’d defer to you on this, my gut say’s drop for parsimony and to increase fit.
#Also is 4 “The federal government should allow all faiths to display religious symbols in public spaces.” supposed to be reverse scored? I don’t currently have it reverse scored.
##Second cfa for cnat without item cns_6r
Cnat_CFA2 <- 'C_nat =~ 1*cns_1 + cns_2 + cns_3 + cns_4r + cns_5'
Cnat_CFA2 <- sem(Cnat_CFA2, data = YNEES_W1, estimator = "MLR") #robust estimator with "MLR"
summary(Cnat_CFA2, fit.measures = T, standardized = T, rsquare = T)## lavaan 0.6-19 ended normally after 24 iterations
##
## Estimator ML
## Optimization method NLMINB
## Number of model parameters 10
##
## Used Total
## Number of observations 422 427
##
## Model Test User Model:
## Standard Scaled
## Test Statistic 5.418 4.457
## Degrees of freedom 5 5
## P-value (Chi-square) 0.367 0.486
## Scaling correction factor 1.216
## Yuan-Bentler correction (Mplus variant)
##
## Model Test Baseline Model:
##
## Test statistic 1367.407 747.612
## Degrees of freedom 10 10
## P-value 0.000 0.000
## Scaling correction factor 1.829
##
## User Model versus Baseline Model:
##
## Comparative Fit Index (CFI) 1.000 1.000
## Tucker-Lewis Index (TLI) 0.999 1.001
##
## Robust Comparative Fit Index (CFI) 1.000
## Robust Tucker-Lewis Index (TLI) 1.001
##
## Loglikelihood and Information Criteria:
##
## Loglikelihood user model (H0) -3473.562 -3473.562
## Scaling correction factor 1.444
## for the MLR correction
## Loglikelihood unrestricted model (H1) -3470.853 -3470.853
## Scaling correction factor 1.368
## for the MLR correction
##
## Akaike (AIC) 6967.124 6967.124
## Bayesian (BIC) 7007.575 7007.575
## Sample-size adjusted Bayesian (SABIC) 6975.841 6975.841
##
## Root Mean Square Error of Approximation:
##
## RMSEA 0.014 0.000
## 90 Percent confidence interval - lower 0.000 0.000
## 90 Percent confidence interval - upper 0.070 0.059
## P-value H_0: RMSEA <= 0.050 0.804 0.899
## P-value H_0: RMSEA >= 0.080 0.021 0.005
##
## Robust RMSEA 0.000
## 90 Percent confidence interval - lower 0.000
## 90 Percent confidence interval - upper 0.070
## P-value H_0: Robust RMSEA <= 0.050 0.830
## P-value H_0: Robust RMSEA >= 0.080 0.023
##
## Standardized Root Mean Square Residual:
##
## SRMR 0.012 0.012
##
## 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
## C_nat =~
## cns_1 1.000 1.675 0.925
## cns_2 0.943 0.029 32.775 0.000 1.579 0.909
## cns_3 0.911 0.032 28.526 0.000 1.526 0.843
## cns_4r -0.177 0.055 -3.250 0.001 -0.297 -0.193
## cns_5 0.858 0.038 22.793 0.000 1.437 0.804
##
## Variances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .cns_1 0.475 0.090 5.269 0.000 0.475 0.145
## .cns_2 0.525 0.088 5.954 0.000 0.525 0.174
## .cns_3 0.949 0.122 7.804 0.000 0.949 0.289
## .cns_4r 2.276 0.141 16.140 0.000 2.276 0.963
## .cns_5 1.130 0.139 8.130 0.000 1.130 0.354
## C_nat 2.804 0.158 17.735 0.000 1.000 1.000
##
## R-Square:
## Estimate
## cns_1 0.855
## cns_2 0.826
## cns_3 0.711
## cns_4r 0.037
## cns_5 0.646resid(Cnat_CFA2)## $type
## [1] "raw"
##
## $cov
## cns_1 cns_2 cns_3 cns_4r cns_5
## cns_1 0.000
## cns_2 -0.006 0.000
## cns_3 0.030 -0.020 0.000
## cns_4r 0.042 -0.039 0.045 0.000
## cns_5 -0.020 0.038 -0.034 -0.081 0.000#Check descritives and correlation between SDO items
#Create matrix for SDO items
SDOvars <- YNEES_W1 %>%
select(sdod_1, sdoe_1) %>%
as.matrix()
# View the matrix
#SDOvars
#descriptives for sdo vars
describe(SDOvars)## vars n mean sd median trimmed mad min max range skew kurtosis se
## sdod_1 1 427 3.75 2.00 4 3.69 2.97 1 7 6 -0.01 -1.14 0.10
## sdoe_1 2 427 5.80 1.37 6 5.95 1.48 1 7 6 -0.84 -0.20 0.07#create a correlation matrix for sdo vars
SDOCorMat <-cor(SDOvars, use = "complete.obs")
#Now visualize the correlation matrix with GGplot and corrplot
ggcorrplot(SDOCorMat, hc.order = TRUE, type = "lower", lab = TRUE, ggtheme = ggplot2::theme_gray,
colors = c("#6D9EC1", "white", "#E46726"))
#correlation between sdo_e and sdo_d
cor.test(YNEES_W1$sdod_1, YNEES_W1$sdoe_1)##
## Pearson's product-moment correlation
##
## data: YNEES_W1$sdod_1 and YNEES_W1$sdoe_1
## t = -3.0519, df = 425, p-value = 0.002416
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## -0.23803433 -0.05227312
## sample estimates:
## cor
## -0.1464445##Given the correlation of .15 between the sdo items, it is best not to combine them into one composite variable.
#SEM assessing predictive validity w/ sdoe_1
Cnat_SEM1 <- 'C_nat =~ 1*cns_1 + cns_2 + cns_3 + cns_4r + cns_5
sdoe_1 ~~ C_nat'
Cnat_SEM1 <- sem(Cnat_SEM1, data = YNEES_W1, estimator = "MLR") #robust estimator with "MLR"
summary(Cnat_SEM1, fit.measures = T, standardized = T, rsquare = T)## lavaan 0.6-19 ended normally after 28 iterations
##
## Estimator ML
## Optimization method NLMINB
## Number of model parameters 12
##
## Used Total
## Number of observations 422 427
##
## Model Test User Model:
## Standard Scaled
## Test Statistic 37.049 30.049
## Degrees of freedom 9 9
## P-value (Chi-square) 0.000 0.000
## Scaling correction factor 1.233
## Yuan-Bentler correction (Mplus variant)
##
## Model Test Baseline Model:
##
## Test statistic 1401.810 868.670
## Degrees of freedom 15 15
## P-value 0.000 0.000
## Scaling correction factor 1.614
##
## User Model versus Baseline Model:
##
## Comparative Fit Index (CFI) 0.980 0.975
## Tucker-Lewis Index (TLI) 0.966 0.959
##
## Robust Comparative Fit Index (CFI) 0.981
## Robust Tucker-Lewis Index (TLI) 0.969
##
## Loglikelihood and Information Criteria:
##
## Loglikelihood user model (H0) -4203.398 -4203.398
## Scaling correction factor 1.354
## for the MLR correction
## Loglikelihood unrestricted model (H1) -4184.873 -4184.873
## Scaling correction factor 1.302
## for the MLR correction
##
## Akaike (AIC) 8430.795 8430.795
## Bayesian (BIC) 8479.335 8479.335
## Sample-size adjusted Bayesian (SABIC) 8441.255 8441.255
##
## Root Mean Square Error of Approximation:
##
## RMSEA 0.086 0.074
## 90 Percent confidence interval - lower 0.058 0.049
## 90 Percent confidence interval - upper 0.116 0.102
## P-value H_0: RMSEA <= 0.050 0.018 0.058
## P-value H_0: RMSEA >= 0.080 0.665 0.396
##
## Robust RMSEA 0.083
## 90 Percent confidence interval - lower 0.051
## 90 Percent confidence interval - upper 0.116
## P-value H_0: Robust RMSEA <= 0.050 0.045
## P-value H_0: Robust RMSEA >= 0.080 0.590
##
## Standardized Root Mean Square Residual:
##
## SRMR 0.051 0.051
##
## 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
## C_nat =~
## cns_1 1.000 1.675 0.925
## cns_2 0.942 0.029 32.624 0.000 1.578 0.908
## cns_3 0.911 0.032 28.495 0.000 1.527 0.844
## cns_4r -0.176 0.055 -3.221 0.001 -0.295 -0.192
## cns_5 0.857 0.038 22.782 0.000 1.436 0.804
##
## Covariances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## C_nat ~~
## sdoe_1 -0.192 0.106 -1.812 0.070 -0.115 -0.084
##
## Variances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .cns_1 0.472 0.090 5.241 0.000 0.472 0.144
## .cns_2 0.529 0.089 5.933 0.000 0.529 0.175
## .cns_3 0.945 0.122 7.777 0.000 0.945 0.288
## .cns_4r 2.277 0.141 16.155 0.000 2.277 0.963
## .cns_5 1.131 0.139 8.132 0.000 1.131 0.354
## sdoe_1 1.873 0.123 15.232 0.000 1.873 1.000
## C_nat 2.807 0.158 17.775 0.000 1.000 1.000
##
## R-Square:
## Estimate
## cns_1 0.856
## cns_2 0.825
## cns_3 0.712
## cns_4r 0.037
## cns_5 0.646resid(Cnat_SEM1)## $type
## [1] "raw"
##
## $cov
## cns_1 cns_2 cns_3 cns_4r cns_5 sdoe_1
## cns_1 0.000
## cns_2 -0.005 0.000
## cns_3 0.027 -0.019 0.000
## cns_4r 0.039 -0.043 0.043 0.000
## cns_5 -0.020 0.040 -0.034 -0.084 0.000
## sdoe_1 -0.051 0.102 -0.169 -0.444 0.074 0.000#SEM assessing predictive validity w/ sdod_1
Cnat_SEM3 <- 'C_nat =~ 1*cns_1 + cns_2 + cns_3 + cns_4r + cns_5
sdod_1 ~~ C_nat'
Cnat_SEM3 <- sem(Cnat_SEM3, data = YNEES_W1, estimator = "MLR") #robust estimator with "MLR"
summary(Cnat_SEM3, fit.measures = T, standardized = T, rsquare = T)## lavaan 0.6-19 ended normally after 28 iterations
##
## Estimator ML
## Optimization method NLMINB
## Number of model parameters 12
##
## Used Total
## Number of observations 422 427
##
## Model Test User Model:
## Standard Scaled
## Test Statistic 23.262 20.403
## Degrees of freedom 9 9
## P-value (Chi-square) 0.006 0.016
## Scaling correction factor 1.140
## Yuan-Bentler correction (Mplus variant)
##
## Model Test Baseline Model:
##
## Test statistic 1473.156 928.242
## Degrees of freedom 15 15
## P-value 0.000 0.000
## Scaling correction factor 1.587
##
## User Model versus Baseline Model:
##
## Comparative Fit Index (CFI) 0.990 0.988
## Tucker-Lewis Index (TLI) 0.984 0.979
##
## Robust Comparative Fit Index (CFI) 0.991
## Robust Tucker-Lewis Index (TLI) 0.985
##
## Loglikelihood and Information Criteria:
##
## Loglikelihood user model (H0) -4320.173 -4320.173
## Scaling correction factor 1.351
## for the MLR correction
## Loglikelihood unrestricted model (H1) -4308.542 -4308.542
## Scaling correction factor 1.261
## for the MLR correction
##
## Akaike (AIC) 8664.346 8664.346
## Bayesian (BIC) 8712.886 8712.886
## Sample-size adjusted Bayesian (SABIC) 8674.806 8674.806
##
## Root Mean Square Error of Approximation:
##
## RMSEA 0.061 0.055
## 90 Percent confidence interval - lower 0.031 0.025
## 90 Percent confidence interval - upper 0.093 0.085
## P-value H_0: RMSEA <= 0.050 0.239 0.354
## P-value H_0: RMSEA >= 0.080 0.176 0.086
##
## Robust RMSEA 0.059
## 90 Percent confidence interval - lower 0.024
## 90 Percent confidence interval - upper 0.093
## P-value H_0: Robust RMSEA <= 0.050 0.298
## P-value H_0: Robust RMSEA >= 0.080 0.163
##
## Standardized Root Mean Square Residual:
##
## SRMR 0.026 0.026
##
## 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
## C_nat =~
## cns_1 1.000 1.680 0.928
## cns_2 0.936 0.028 33.644 0.000 1.573 0.905
## cns_3 0.912 0.032 28.655 0.000 1.532 0.846
## cns_4r -0.178 0.054 -3.300 0.001 -0.300 -0.195
## cns_5 0.852 0.037 22.880 0.000 1.430 0.800
##
## Covariances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## C_nat ~~
## sdod_1 1.504 0.181 8.327 0.000 0.895 0.448
##
## Variances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .cns_1 0.458 0.087 5.277 0.000 0.458 0.140
## .cns_2 0.544 0.088 6.176 0.000 0.544 0.180
## .cns_3 0.930 0.120 7.773 0.000 0.930 0.284
## .cns_4r 2.275 0.141 16.149 0.000 2.275 0.962
## .cns_5 1.148 0.139 8.267 0.000 1.148 0.360
## sdod_1 3.986 0.182 21.933 0.000 3.986 1.000
## C_nat 2.821 0.156 18.042 0.000 1.000 1.000
##
## R-Square:
## Estimate
## cns_1 0.860
## cns_2 0.820
## cns_3 0.716
## cns_4r 0.038
## cns_5 0.640resid(Cnat_SEM3)## $type
## [1] "raw"
##
## $cov
## cns_1 cns_2 cns_3 cns_4r cns_5 sdod_1
## cns_1 0.000
## cns_2 -0.004 0.000
## cns_3 0.013 -0.020 0.000
## cns_4r 0.048 -0.037 0.051 0.000
## cns_5 -0.016 0.057 -0.032 -0.080 0.000
## sdod_1 0.059 -0.104 0.204 -0.190 -0.222 0.000###Correlation between christian nationalism and SDO dominance is very strong as we’d expect. Folks who view hierarchy as natural (item: “Some groups of people are simply inferior to other groups.”) are more likely to endorse christian nationalism. I did not run a predictive model where SDO is a predictor of christian nationalism or vice versa, but I can. Though with cross-sectional data we’re only in the space of make correlational claims anyway. The beta coefficient for cns predicting SDO or vice versa along with the model fit would be equivelalent in all three models (cns correlated with sdo, cns predicting sdo, etc., ) as well.
#SEM assessing predictive validity w/ eri centrality
Cnat_SEM4 <- 'C_nat =~ 1*cns_1 + cns_2 + cns_3 + cns_4r + cns_5
ERI_C =~ 1*eric_1 + eric_2 + eric_3'
Cnat_SEM4 <- sem(Cnat_SEM4, data = YNEES_W1, estimator = "MLR") #robust estimator with "MLR"
summary(Cnat_SEM4, fit.measures = T, standardized = T, rsquare = T)## lavaan 0.6-19 ended normally after 34 iterations
##
## Estimator ML
## Optimization method NLMINB
## Number of model parameters 17
##
## Used Total
## Number of observations 422 427
##
## Model Test User Model:
## Standard Scaled
## Test Statistic 64.070 55.733
## Degrees of freedom 19 19
## P-value (Chi-square) 0.000 0.000
## Scaling correction factor 1.150
## Yuan-Bentler correction (Mplus variant)
##
## Model Test Baseline Model:
##
## Test statistic 1889.550 1280.457
## Degrees of freedom 28 28
## P-value 0.000 0.000
## Scaling correction factor 1.476
##
## User Model versus Baseline Model:
##
## Comparative Fit Index (CFI) 0.976 0.971
## Tucker-Lewis Index (TLI) 0.964 0.957
##
## Robust Comparative Fit Index (CFI) 0.977
## Robust Tucker-Lewis Index (TLI) 0.966
##
## Loglikelihood and Information Criteria:
##
## Loglikelihood user model (H0) -5360.124 -5360.124
## Scaling correction factor 1.459
## for the MLR correction
## Loglikelihood unrestricted model (H1) -5328.090 -5328.090
## Scaling correction factor 1.296
## for the MLR correction
##
## Akaike (AIC) 10754.249 10754.249
## Bayesian (BIC) 10823.014 10823.014
## Sample-size adjusted Bayesian (SABIC) 10769.067 10769.067
##
## Root Mean Square Error of Approximation:
##
## RMSEA 0.075 0.068
## 90 Percent confidence interval - lower 0.055 0.049
## 90 Percent confidence interval - upper 0.096 0.087
## P-value H_0: RMSEA <= 0.050 0.020 0.061
## P-value H_0: RMSEA >= 0.080 0.364 0.157
##
## Robust RMSEA 0.073
## 90 Percent confidence interval - lower 0.051
## 90 Percent confidence interval - upper 0.095
## P-value H_0: Robust RMSEA <= 0.050 0.044
## P-value H_0: Robust RMSEA >= 0.080 0.312
##
## Standardized Root Mean Square Residual:
##
## SRMR 0.065 0.065
##
## 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
## C_nat =~
## cns_1 1.000 1.674 0.924
## cns_2 0.944 0.029 32.892 0.000 1.580 0.909
## cns_3 0.911 0.032 28.512 0.000 1.525 0.843
## cns_4r -0.179 0.055 -3.270 0.001 -0.299 -0.195
## cns_5 0.859 0.038 22.819 0.000 1.438 0.804
## ERI_C =~
## eric_1 1.000 0.914 0.779
## eric_2 1.217 0.092 13.176 0.000 1.112 0.939
## eric_3 0.926 0.096 9.664 0.000 0.846 0.549
##
## Covariances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## C_nat ~~
## ERI_C 0.216 0.094 2.287 0.022 0.141 0.141
##
## Variances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .cns_1 0.478 0.091 5.267 0.000 0.478 0.146
## .cns_2 0.522 0.087 6.006 0.000 0.522 0.173
## .cns_3 0.950 0.122 7.794 0.000 0.950 0.290
## .cns_4r 2.275 0.141 16.118 0.000 2.275 0.962
## .cns_5 1.127 0.139 8.128 0.000 1.127 0.353
## .eric_1 0.541 0.095 5.690 0.000 0.541 0.393
## .eric_2 0.167 0.076 2.207 0.027 0.167 0.119
## .eric_3 1.662 0.146 11.404 0.000 1.662 0.699
## C_nat 2.801 0.159 17.651 0.000 1.000 1.000
## ERI_C 0.835 0.113 7.422 0.000 1.000 1.000
##
## R-Square:
## Estimate
## cns_1 0.854
## cns_2 0.827
## cns_3 0.710
## cns_4r 0.038
## cns_5 0.647
## eric_1 0.607
## eric_2 0.881
## eric_3 0.301resid(Cnat_SEM4)## $type
## [1] "raw"
##
## $cov
## cns_1 cns_2 cns_3 cns_4r cns_5 eric_1 eric_2 eric_3
## cns_1 0.000
## cns_2 -0.006 0.000
## cns_3 0.033 -0.020 0.000
## cns_4r 0.045 -0.036 0.049 0.000
## cns_5 -0.020 0.035 -0.034 -0.078 0.000
## eric_1 -0.096 0.071 -0.051 -0.286 0.052 0.000
## eric_2 -0.057 -0.025 -0.057 -0.234 0.040 0.002 0.000
## eric_3 0.501 0.459 0.347 -0.318 0.242 -0.020 0.000 0.000###Interesting that even when we look at the correlation among the complete sample not broken down by race, there is still a small but significant correlation between ERI centrality and christian nationalism.
#Multigroup model looking at eri centrality and cns
Cnat_SEM5 <- 'C_nat =~ 1*cns_1 + cns_2 + cns_3 + cns_4r + cns_5
ERI_C =~ 1*eric_1 + eric_2 + eric_3'
Cnat_SEM5 <- sem(Cnat_SEM5, data = YNEES_W1, estimator = "MLR", group = "race2") #robust estimator with "MLR"## Warning: lavaan->lav_data_full():
## group variable 'race2' contains missing valuessummary(Cnat_SEM5, fit.measures = T, standardized = T, rsquare = T)## lavaan 0.6-19 ended normally after 55 iterations
##
## Estimator ML
## Optimization method NLMINB
## Number of model parameters 50
##
## Number of observations per group: Used Total
## People of Color 279 281
## White 142 145
##
## Model Test User Model:
## Standard Scaled
## Test Statistic 91.536 80.309
## Degrees of freedom 38 38
## P-value (Chi-square) 0.000 0.000
## Scaling correction factor 1.140
## Yuan-Bentler correction (Mplus variant)
## Test statistic for each group:
## People of Color 50.688 50.688
## White 29.621 29.621
##
## Model Test Baseline Model:
##
## Test statistic 1932.718 1347.074
## Degrees of freedom 56 56
## P-value 0.000 0.000
## Scaling correction factor 1.435
##
## User Model versus Baseline Model:
##
## Comparative Fit Index (CFI) 0.971 0.967
## Tucker-Lewis Index (TLI) 0.958 0.952
##
## Robust Comparative Fit Index (CFI) 0.974
## Robust Tucker-Lewis Index (TLI) 0.962
##
## Loglikelihood and Information Criteria:
##
## Loglikelihood user model (H0) -5313.828 -5313.828
## Scaling correction factor 1.281
## for the MLR correction
## Loglikelihood unrestricted model (H1) -5268.060 -5268.060
## Scaling correction factor 1.220
## for the MLR correction
##
## Akaike (AIC) 10727.656 10727.656
## Bayesian (BIC) 10929.787 10929.787
## Sample-size adjusted Bayesian (SABIC) 10771.122 10771.122
##
## Root Mean Square Error of Approximation:
##
## RMSEA 0.082 0.073
## 90 Percent confidence interval - lower 0.061 0.052
## 90 Percent confidence interval - upper 0.103 0.093
## P-value H_0: RMSEA <= 0.050 0.009 0.038
## P-value H_0: RMSEA >= 0.080 0.576 0.298
##
## Robust RMSEA 0.078
## 90 Percent confidence interval - lower 0.054
## 90 Percent confidence interval - upper 0.101
## P-value H_0: Robust RMSEA <= 0.050 0.030
## P-value H_0: Robust RMSEA >= 0.080 0.457
##
## Standardized Root Mean Square Residual:
##
## SRMR 0.062 0.062
##
## Parameter Estimates:
##
## Standard errors Sandwich
## Information bread Observed
## Observed information based on Hessian
##
##
## Group 1 [People of Color]:
##
## Latent Variables:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## C_nat =~
## cns_1 1.000 1.631 0.920
## cns_2 0.932 0.039 23.846 0.000 1.520 0.887
## cns_3 0.930 0.041 22.808 0.000 1.517 0.837
## cns_4r -0.190 0.069 -2.744 0.006 -0.309 -0.195
## cns_5 0.860 0.054 16.032 0.000 1.402 0.772
## ERI_C =~
## eric_1 1.000 0.934 0.789
## eric_2 1.161 0.123 9.417 0.000 1.085 0.914
## eric_3 0.808 0.119 6.819 0.000 0.755 0.503
##
## Covariances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## C_nat ~~
## ERI_C 0.022 0.112 0.195 0.845 0.014 0.014
##
## Intercepts:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .cns_1 3.222 0.106 30.371 0.000 3.222 1.818
## .cns_2 3.262 0.103 31.781 0.000 3.262 1.903
## .cns_3 3.204 0.109 29.517 0.000 3.204 1.767
## .cns_4r 2.670 0.095 28.165 0.000 2.670 1.686
## .cns_5 3.656 0.109 33.611 0.000 3.656 2.012
## .eric_1 4.774 0.071 67.382 0.000 4.774 4.034
## .eric_2 4.774 0.071 67.210 0.000 4.774 4.024
## .eric_3 4.197 0.090 46.722 0.000 4.197 2.797
##
## Variances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .cns_1 0.480 0.110 4.375 0.000 0.480 0.153
## .cns_2 0.627 0.125 5.008 0.000 0.627 0.213
## .cns_3 0.986 0.144 6.852 0.000 0.986 0.300
## .cns_4r 2.412 0.174 13.895 0.000 2.412 0.962
## .cns_5 1.334 0.200 6.655 0.000 1.334 0.404
## .eric_1 0.528 0.143 3.689 0.000 0.528 0.377
## .eric_2 0.231 0.111 2.076 0.038 0.231 0.164
## .eric_3 1.681 0.184 9.138 0.000 1.681 0.747
## C_nat 2.660 0.196 13.586 0.000 1.000 1.000
## ERI_C 0.872 0.142 6.126 0.000 1.000 1.000
##
## R-Square:
## Estimate
## cns_1 0.847
## cns_2 0.787
## cns_3 0.700
## cns_4r 0.038
## cns_5 0.596
## eric_1 0.623
## eric_2 0.836
## eric_3 0.253
##
##
## Group 2 [White]:
##
## Latent Variables:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## C_nat =~
## cns_1 1.000 1.677 0.926
## cns_2 0.953 0.045 21.158 0.000 1.598 0.943
## cns_3 0.882 0.061 14.450 0.000 1.478 0.845
## cns_4r -0.164 0.093 -1.769 0.077 -0.275 -0.190
## cns_5 0.891 0.046 19.308 0.000 1.494 0.873
## ERI_C =~
## eric_1 1.000 0.903 0.784
## eric_2 1.242 0.120 10.358 0.000 1.122 0.959
## eric_3 1.113 0.164 6.786 0.000 1.006 0.629
##
## Covariances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## C_nat ~~
## ERI_C 0.601 0.187 3.213 0.001 0.397 0.397
##
## Intercepts:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .cns_1 3.831 0.152 25.202 0.000 3.831 2.115
## .cns_2 3.908 0.142 27.483 0.000 3.908 2.306
## .cns_3 3.754 0.147 25.574 0.000 3.754 2.146
## .cns_4r 2.620 0.121 21.569 0.000 2.620 1.810
## .cns_5 3.972 0.144 27.656 0.000 3.972 2.321
## .eric_1 4.725 0.097 48.900 0.000 4.725 4.104
## .eric_2 4.577 0.098 46.590 0.000 4.577 3.910
## .eric_3 3.894 0.134 29.008 0.000 3.894 2.434
##
## Variances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .cns_1 0.469 0.167 2.817 0.005 0.469 0.143
## .cns_2 0.317 0.079 4.011 0.000 0.317 0.110
## .cns_3 0.873 0.222 3.932 0.000 0.873 0.285
## .cns_4r 2.019 0.240 8.397 0.000 2.019 0.964
## .cns_5 0.696 0.111 6.302 0.000 0.696 0.238
## .eric_1 0.510 0.100 5.078 0.000 0.510 0.385
## .eric_2 0.111 0.075 1.482 0.138 0.111 0.081
## .eric_3 1.548 0.211 7.338 0.000 1.548 0.605
## C_nat 2.812 0.285 9.861 0.000 1.000 1.000
## ERI_C 0.816 0.188 4.350 0.000 1.000 1.000
##
## R-Square:
## Estimate
## cns_1 0.857
## cns_2 0.890
## cns_3 0.715
## cns_4r 0.036
## cns_5 0.762
## eric_1 0.615
## eric_2 0.919
## eric_3 0.395resid(Cnat_SEM5)## $`People of Color`
## $`People of Color`$type
## [1] "raw"
##
## $`People of Color`$cov
## cns_1 cns_2 cns_3 cns_4r cns_5 eric_1 eric_2 eric_3
## cns_1 0.000
## cns_2 0.003 0.000
## cns_3 0.028 -0.041 0.000
## cns_4r 0.051 -0.035 0.053 0.000
## cns_5 -0.042 0.044 0.011 -0.157 0.000
## eric_1 -0.162 0.085 -0.089 -0.346 0.054 0.000
## eric_2 -0.083 0.025 -0.081 -0.224 0.116 0.000 0.000
## eric_3 0.455 0.538 0.284 -0.315 0.196 -0.001 0.000 0.000
##
## $`People of Color`$mean
## cns_1 cns_2 cns_3 cns_4r cns_5 eric_1 eric_2 eric_3
## 0 0 0 0 0 0 0 0
##
##
## $White
## $White$type
## [1] "raw"
##
## $White$cov
## cns_1 cns_2 cns_3 cns_4r cns_5 eric_1 eric_2 eric_3
## cns_1 0.000
## cns_2 -0.027 0.000
## cns_3 0.036 0.023 0.000
## cns_4r 0.030 -0.046 0.038 0.000
## cns_5 0.025 0.032 -0.124 0.076 0.000
## eric_1 0.029 0.036 0.044 -0.175 0.028 0.000
## eric_2 0.035 -0.081 0.055 -0.264 -0.113 0.004 0.000
## eric_3 0.574 0.297 0.483 -0.318 0.267 -0.078 0.003 0.000
##
## $White$mean
## cns_1 cns_2 cns_3 cns_4r cns_5 eric_1 eric_2 eric_3
## 0 0 0 0 0 0 0 0###Wow! As hypothesized though, White people are driving this correlation between eri centrality and christian nationalism. For White people (r = .40, p = .001), while for POC the correlation is not sig. (r = .01, p = .85). Model fit is pretty good too.
#Multigroup SEM assessing predictive validity w/ sdod_1
Cnat_SEM7 <- 'C_nat =~ 1*cns_1 + cns_2 + cns_3 + cns_4r + cns_5
sdod_1 ~~ C_nat'
Cnat_SEM7 <- sem(Cnat_SEM7, data = YNEES_W1, estimator = "MLR", group = "race2") #robust estimator with "MLR"## Warning: lavaan->lav_data_full():
## group variable 'race2' contains missing valuessummary(Cnat_SEM7, fit.measures = T, standardized = T, rsquare = T)## lavaan 0.6-19 ended normally after 56 iterations
##
## Estimator ML
## Optimization method NLMINB
## Number of model parameters 36
##
## Number of observations per group: Used Total
## People of Color 279 281
## White 142 145
##
## Model Test User Model:
## Standard Scaled
## Test Statistic 40.807 37.146
## Degrees of freedom 18 18
## P-value (Chi-square) 0.002 0.005
## Scaling correction factor 1.099
## Yuan-Bentler correction (Mplus variant)
## Test statistic for each group:
## People of Color 10.643 10.643
## White 26.503 26.503
##
## Model Test Baseline Model:
##
## Test statistic 1487.070 968.834
## Degrees of freedom 30 30
## P-value 0.000 0.000
## Scaling correction factor 1.535
##
## User Model versus Baseline Model:
##
## Comparative Fit Index (CFI) 0.984 0.980
## Tucker-Lewis Index (TLI) 0.974 0.966
##
## Robust Comparative Fit Index (CFI) 0.985
## Robust Tucker-Lewis Index (TLI) 0.976
##
## Loglikelihood and Information Criteria:
##
## Loglikelihood user model (H0) -4287.799 -4287.799
## Scaling correction factor 1.224
## for the MLR correction
## Loglikelihood unrestricted model (H1) -4267.396 -4267.396
## Scaling correction factor 1.182
## for the MLR correction
##
## Akaike (AIC) 8647.599 8647.599
## Bayesian (BIC) 8793.133 8793.133
## Sample-size adjusted Bayesian (SABIC) 8678.894 8678.894
##
## Root Mean Square Error of Approximation:
##
## RMSEA 0.078 0.071
## 90 Percent confidence interval - lower 0.046 0.039
## 90 Percent confidence interval - upper 0.109 0.102
## P-value H_0: RMSEA <= 0.050 0.073 0.123
## P-value H_0: RMSEA >= 0.080 0.481 0.343
##
## Robust RMSEA 0.075
## 90 Percent confidence interval - lower 0.040
## 90 Percent confidence interval - upper 0.109
## P-value H_0: Robust RMSEA <= 0.050 0.111
## P-value H_0: Robust RMSEA >= 0.080 0.426
##
## Standardized Root Mean Square Residual:
##
## SRMR 0.028 0.028
##
## Parameter Estimates:
##
## Standard errors Sandwich
## Information bread Observed
## Observed information based on Hessian
##
##
## Group 1 [People of Color]:
##
## Latent Variables:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## C_nat =~
## cns_1 1.000 1.635 0.923
## cns_2 0.927 0.038 24.427 0.000 1.515 0.884
## cns_3 0.931 0.041 22.794 0.000 1.522 0.839
## cns_4r -0.189 0.068 -2.754 0.006 -0.308 -0.195
## cns_5 0.855 0.053 16.122 0.000 1.397 0.769
##
## Covariances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## C_nat ~~
## sdod_1 1.228 0.219 5.606 0.000 0.751 0.389
##
## Intercepts:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .cns_1 3.222 0.106 30.371 0.000 3.222 1.818
## .cns_2 3.262 0.103 31.781 0.000 3.262 1.903
## .cns_3 3.204 0.109 29.517 0.000 3.204 1.767
## .cns_4r 2.670 0.095 28.165 0.000 2.670 1.686
## .cns_5 3.656 0.109 33.611 0.000 3.656 2.012
## sdod_1 3.606 0.116 31.185 0.000 3.606 1.867
##
## Variances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .cns_1 0.467 0.107 4.368 0.000 0.467 0.149
## .cns_2 0.644 0.126 5.114 0.000 0.644 0.219
## .cns_3 0.972 0.143 6.809 0.000 0.972 0.296
## .cns_4r 2.413 0.173 13.909 0.000 2.413 0.962
## .cns_5 1.348 0.200 6.740 0.000 1.348 0.409
## sdod_1 3.730 0.216 17.232 0.000 3.730 1.000
## C_nat 2.674 0.194 13.769 0.000 1.000 1.000
##
## R-Square:
## Estimate
## cns_1 0.851
## cns_2 0.781
## cns_3 0.704
## cns_4r 0.038
## cns_5 0.591
##
##
## Group 2 [White]:
##
## Latent Variables:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## C_nat =~
## cns_1 1.000 1.680 0.928
## cns_2 0.949 0.043 21.947 0.000 1.595 0.941
## cns_3 0.884 0.060 14.720 0.000 1.485 0.849
## cns_4r -0.166 0.092 -1.801 0.072 -0.279 -0.192
## cns_5 0.886 0.046 19.426 0.000 1.488 0.869
##
## Covariances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## C_nat ~~
## sdod_1 1.829 0.297 6.161 0.000 1.089 0.521
##
## Intercepts:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .cns_1 3.831 0.152 25.202 0.000 3.831 2.115
## .cns_2 3.908 0.142 27.483 0.000 3.908 2.306
## .cns_3 3.754 0.147 25.574 0.000 3.754 2.146
## .cns_4r 2.620 0.121 21.569 0.000 2.620 1.810
## .cns_5 3.972 0.144 27.656 0.000 3.972 2.321
## sdod_1 3.937 0.175 22.442 0.000 3.937 1.883
##
## Variances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .cns_1 0.458 0.157 2.914 0.004 0.458 0.140
## .cns_2 0.327 0.077 4.245 0.000 0.327 0.114
## .cns_3 0.853 0.220 3.883 0.000 0.853 0.279
## .cns_4r 2.017 0.240 8.404 0.000 2.017 0.963
## .cns_5 0.715 0.114 6.248 0.000 0.715 0.244
## sdod_1 4.369 0.319 13.692 0.000 4.369 1.000
## C_nat 2.823 0.281 10.043 0.000 1.000 1.000
##
## R-Square:
## Estimate
## cns_1 0.860
## cns_2 0.886
## cns_3 0.721
## cns_4r 0.037
## cns_5 0.756resid(Cnat_SEM7)## $`People of Color`
## $`People of Color`$type
## [1] "raw"
##
## $`People of Color`$cov
## cns_1 cns_2 cns_3 cns_4r cns_5 sdod_1
## cns_1 0.000
## cns_2 0.006 0.000
## cns_3 0.015 -0.040 0.000
## cns_4r 0.051 -0.038 0.053 0.000
## cns_5 -0.040 0.059 0.012 -0.159 0.000
## sdod_1 0.046 -0.103 0.159 -0.024 -0.164 0.000
##
## $`People of Color`$mean
## cns_1 cns_2 cns_3 cns_4r cns_5 sdod_1
## 0 0 0 0 0 0
##
##
## $White
## $White$type
## [1] "raw"
##
## $White$cov
## cns_1 cns_2 cns_3 cns_4r cns_5 sdod_1
## cns_1 0.000
## cns_2 -0.027 0.000
## cns_3 0.019 0.016 0.000
## cns_4r 0.037 -0.041 0.045 0.000
## cns_5 0.030 0.046 -0.126 0.080 0.000
## sdod_1 0.118 -0.094 0.346 -0.531 -0.312 0.000
##
## $White$mean
## cns_1 cns_2 cns_3 cns_4r cns_5 sdod_1
## 0 0 0 0 0 0###As you’d expect as well, SDO is sig. correlated with christian nationalism for both groups, but the correlation is a bit stronger for White folks.
#I guess we should run some predicitve validity models too? These variables I feel like theoretically would precede more christian nationalism rather than be an outocome of it, what do you think? It seemed at first to me that we’d be running a bunch of model but as I run them, I can see how some DVs would be interesting to assess.
#Visualize christian nationalism by race
# Filter out NA values in race2 column
YNEES_clean <- YNEES_W1 %>%
filter(!is.na(race))
ggplot(YNEES_clean, aes(x = race2, y = c_nat)) +
geom_boxplot() + # Customize outliers
geom_jitter(aes(color = race2), width = 0.2) + # Add jittered points
scale_color_viridis_d(option = "D") + # Discrete viridis palette
labs(title = "Box Plot for Christian Nationalism by Race", x = "Race", y = "c_nat") +
theme_classic()## Warning: Removed 6 rows containing non-finite outside the scale range
## (`stat_boxplot()`).## Warning: Removed 6 rows containing missing values or values outside the scale range
## (`geom_point()`).
#Christian nationalism by gender
ggplot(YNEES, aes(x = gender, y = c_nat)) +
geom_boxplot() + # Customize outliers
geom_jitter(aes(color = gender), width = 0.2) + # Add jittered points
scale_color_viridis_d(option = "D") + # Discrete viridis palette
labs(title = "Box Plot for Christian Nationalism by Gender", x = "Gender", y = "c_nat") +
theme_classic()## Warning: Removed 6 rows containing non-finite outside the scale range
## (`stat_boxplot()`).## Warning: Removed 6 rows containing missing values or values outside the scale range
## (`geom_point()`).