7C Basic Analysis
Franziska Rees
2021-03-25
Introduction Analysis script
- This is an R Markdown script (Allaire et al. 2021) with exemplatory basic analyses for the 7C vaccination readiness scale
Presets
R markdown
Sample description
# Age
knitr::kable(describe(df$age), 'simple', caption = 'Descriptive Statistics: Age')| vars | n | mean | sd | median | trimmed | mad | min | max | range | skew | kurtosis | se | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| X1 | 1 | 690 | 56.55 | 15.55 | 58 | 57.65 | 16.31 | 19 | 94 | 75 | -0.5148 | -0.49 | 0.5921 |
# Gender
df$gender <- set_labels(df$gender,
labels = c("Male", "Female", "Other"))
knitr::kable(frq(df$gender), 'simple', caption = 'Gender frequencies in sample')| val | label | frq | raw.prc | valid.prc | cum.prc |
|---|---|---|---|---|---|
| 1 | Male | 284 | 41.16 | 41.16 | 41.16 |
| 2 | Female | 403 | 58.41 | 58.41 | 99.57 |
| 3 | Other | 3 | 0.43 | 0.43 | 100.00 |
| NA | NA | 0 | 0.00 | NA | NA |
Descriptive statistics 7C
| vars | n | mean | sd | min | max | range | se | |
|---|---|---|---|---|---|---|---|---|
| conf_01 | 1 | 690 | 5.468 | 1.526 | 1 | 7 | 6 | 0.0581 |
| conf_02 | 2 | 690 | 5.412 | 1.484 | 1 | 7 | 6 | 0.0565 |
| conf_03 | 3 | 690 | 6.123 | 1.267 | 1 | 7 | 6 | 0.0482 |
| cmpcy_01 | 4 | 690 | 6.184 | 1.360 | 1 | 7 | 6 | 0.0518 |
| cmpcy_02 | 5 | 690 | 6.106 | 1.424 | 1 | 7 | 6 | 0.0542 |
| cmpcy_03 | 6 | 690 | 5.825 | 1.585 | 1 | 7 | 6 | 0.0604 |
| const_01 | 7 | 690 | 5.851 | 1.472 | 1 | 7 | 6 | 0.0561 |
| const_02 | 8 | 690 | 4.916 | 1.865 | 1 | 7 | 6 | 0.0710 |
| const_03 | 9 | 690 | 6.314 | 1.199 | 1 | 7 | 6 | 0.0457 |
| calc_01 | 10 | 690 | 3.451 | 2.212 | 1 | 7 | 6 | 0.0842 |
| calc_02 | 11 | 690 | 3.885 | 2.173 | 1 | 7 | 6 | 0.0827 |
| calc_03 | 12 | 690 | 3.849 | 2.187 | 1 | 7 | 6 | 0.0833 |
| colr_01 | 13 | 690 | 6.012 | 1.476 | 1 | 7 | 6 | 0.0562 |
| colr_02 | 14 | 690 | 6.378 | 1.181 | 1 | 7 | 6 | 0.0449 |
| colr_03 | 15 | 690 | 6.109 | 1.368 | 1 | 7 | 6 | 0.0521 |
| cmpli_01 | 16 | 690 | 4.465 | 2.025 | 1 | 7 | 6 | 0.0771 |
| cmpli_02 | 17 | 690 | 5.478 | 1.611 | 1 | 7 | 6 | 0.0613 |
| cmpli_03 | 18 | 690 | 4.070 | 1.997 | 1 | 7 | 6 | 0.0760 |
| consp_01 | 19 | 690 | 5.936 | 1.452 | 1 | 7 | 6 | 0.0553 |
| consp_02 | 20 | 690 | 4.952 | 1.711 | 1 | 7 | 6 | 0.0651 |
| consp_03 | 21 | 690 | 5.548 | 1.651 | 1 | 7 | 6 | 0.0629 |
Descrictives of mean values
| vars | n | mean | sd | min | max | range | se | |
|---|---|---|---|---|---|---|---|---|
| conf_m | 1 | 690 | 5.668 | 1.112 | 1.333 | 7 | 5.667 | 0.0423 |
| cmpcy_m | 2 | 690 | 6.038 | 1.167 | 1.000 | 7 | 6.000 | 0.0444 |
| const_m | 3 | 690 | 5.694 | 1.141 | 1.333 | 7 | 5.667 | 0.0434 |
| calc_m | 4 | 690 | 3.728 | 1.579 | 1.000 | 7 | 6.000 | 0.0601 |
| colr_m | 5 | 690 | 6.166 | 1.171 | 1.000 | 7 | 6.000 | 0.0446 |
| cmpli_m | 6 | 690 | 4.671 | 1.540 | 1.000 | 7 | 6.000 | 0.0586 |
| consp_m | 7 | 690 | 5.479 | 1.226 | 1.000 | 7 | 6.000 | 0.0467 |
| mean_7c | 8 | 690 | 5.349 | 0.866 | 1.571 | 7 | 5.429 | 0.0330 |
| mean.short_7c | 9 | 690 | 5.305 | 1.028 | 1.000 | 7 | 6.000 | 0.0391 |
Histograms 7C
library(psych)
# Confidence
psych::multi.hist(df %>% select(conf_01:conf_03))
# Complacency
psych::multi.hist(df %>% select(cmpcy_01:cmpcy_03))
# Csonatraints
psych::multi.hist(df %>% select(const_01:const_03))
# Calculation
psych::multi.hist(df %>% select(calc_01:calc_03))
# Collective Responsibility
psych::multi.hist(df %>% select(colr_01:colr_03))
# Compliance
psych::multi.hist(df %>% select(cmpli_01:cmpli_03))
# Conspiracy
psych::multi.hist(df %>% select(consp_01:consp_03))
# component Means
psych::multi.hist(df %>% select(conf_m:consp_m))
# long version and short version total mean scores
psych::multi.hist(df %>% select(mean_7c:mean.short_7c))
Inter Item Correlations
corPlot(df %>% select(conf_01:consp_03))
Component Correlations
corPlot(df %>% select(conf_m:consp_m))
Bifactor model
mod_7c <- "
g =~ conf_01 + conf_02 + conf_03 + cmpcy_01 + cmpcy_02 + cmpcy_03 +
const_01 + const_02 + const_03 + calc_01 + calc_02 + calc_03 +
colr_01 + colr_02 + colr_03 + cmpli_01 + cmpli_02 + cmpli_03 +
consp_01 + consp_02 + consp_03
cmpcy =~ cmpcy_01 + cmpcy_02 + cmpcy_03
const =~ const_01 + const_02 + const_03
calc =~ calc_01 + calc_02 + calc_03
colres =~ colr_01 + colr_02 + colr_03
cmpli =~ cmpli_01 + cmpli_02 + cmpli_03
consp =~ consp_01 + consp_02 + consp_03
"
fit.mod_7c <- cfa(mod_7c,
data = df,
estimator = "MLR",
missing = "FIML",
std.lv = TRUE,
orthogonal = TRUE)
knitr::kable(fitMeasures(fit.mod_7c, c("chisq", "df", "pvalue", "cfi", "tli", "rmsea", "srmr"),
output = "matrix"), 'simple', caption = 'Fit measures of 7C bifactor model')| chisq | 497.0922 |
| df | 171.0000 |
| pvalue | 0.0000 |
| cfi | 0.9308 |
| tli | 0.9150 |
| rmsea | 0.0526 |
| srmr | 0.0461 |
summary(fit.mod_7c, standardized = TRUE)## lavaan 0.6-7 ended normally after 72 iterations
##
## Estimator ML
## Optimization method NLMINB
## Number of free parameters 81
##
## Number of observations 690
## Number of missing patterns 1
##
## Model Test User Model:
## Standard Robust
## Test Statistic 497.092 440.820
## Degrees of freedom 171 171
## P-value (Chi-square) 0.000 0.000
## Scaling correction factor 1.128
## Yuan-Bentler correction (Mplus variant)
##
## 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
## g =~
## conf_01 0.826 0.068 12.197 0.000 0.826 0.542
## conf_02 0.807 0.072 11.202 0.000 0.807 0.544
## conf_03 0.878 0.071 12.453 0.000 0.878 0.693
## cmpcy_01 0.733 0.066 11.142 0.000 0.733 0.539
## cmpcy_02 0.833 0.068 12.313 0.000 0.833 0.585
## cmpcy_03 0.920 0.068 13.451 0.000 0.920 0.581
## const_01 0.921 0.066 13.865 0.000 0.921 0.626
## const_02 1.132 0.068 16.664 0.000 1.132 0.607
## const_03 0.424 0.052 8.142 0.000 0.424 0.354
## calc_01 0.051 0.091 0.560 0.576 0.051 0.023
## calc_02 0.768 0.075 10.204 0.000 0.768 0.354
## calc_03 0.681 0.079 8.560 0.000 0.681 0.311
## colr_01 0.882 0.069 12.859 0.000 0.882 0.598
## colr_02 0.819 0.069 11.923 0.000 0.819 0.695
## colr_03 0.845 0.066 12.728 0.000 0.845 0.618
## cmpli_01 0.808 0.081 9.993 0.000 0.808 0.399
## cmpli_02 0.930 0.068 13.585 0.000 0.930 0.578
## cmpli_03 0.778 0.074 10.560 0.000 0.778 0.390
## consp_01 0.715 0.064 11.155 0.000 0.715 0.493
## consp_02 0.711 0.071 10.057 0.000 0.711 0.416
## consp_03 0.704 0.074 9.506 0.000 0.704 0.426
## cmpcy =~
## cmpcy_01 0.658 0.175 3.747 0.000 0.658 0.484
## cmpcy_02 0.858 0.216 3.966 0.000 0.858 0.603
## cmpcy_03 0.207 0.088 2.344 0.019 0.207 0.131
## const =~
## const_01 0.801 0.622 1.286 0.198 0.801 0.544
## const_02 0.271 0.223 1.218 0.223 0.271 0.145
## const_03 0.172 0.157 1.094 0.274 0.172 0.144
## calc =~
## calc_01 0.817 0.133 6.138 0.000 0.817 0.369
## calc_02 1.660 0.214 7.759 0.000 1.660 0.764
## calc_03 0.830 0.131 6.331 0.000 0.830 0.380
## colres =~
## colr_01 0.853 0.091 9.382 0.000 0.853 0.578
## colr_02 0.333 0.065 5.152 0.000 0.333 0.283
## colr_03 0.843 0.077 10.984 0.000 0.843 0.616
## cmpli =~
## cmpli_01 1.230 0.119 10.328 0.000 1.230 0.608
## cmpli_02 0.481 0.070 6.871 0.000 0.481 0.299
## cmpli_03 1.531 0.133 11.502 0.000 1.531 0.767
## consp =~
## consp_01 0.461 0.101 4.588 0.000 0.461 0.318
## consp_02 0.584 0.124 4.694 0.000 0.584 0.341
## consp_03 1.052 0.193 5.445 0.000 1.052 0.637
##
## Covariances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## g ~~
## cmpcy 0.000 0.000 0.000
## const 0.000 0.000 0.000
## calc 0.000 0.000 0.000
## colres 0.000 0.000 0.000
## cmpli 0.000 0.000 0.000
## consp 0.000 0.000 0.000
## cmpcy ~~
## const 0.000 0.000 0.000
## calc 0.000 0.000 0.000
## colres 0.000 0.000 0.000
## cmpli 0.000 0.000 0.000
## consp 0.000 0.000 0.000
## const ~~
## calc 0.000 0.000 0.000
## colres 0.000 0.000 0.000
## cmpli 0.000 0.000 0.000
## consp 0.000 0.000 0.000
## calc ~~
## colres 0.000 0.000 0.000
## cmpli 0.000 0.000 0.000
## consp 0.000 0.000 0.000
## colres ~~
## cmpli 0.000 0.000 0.000
## consp 0.000 0.000 0.000
## cmpli ~~
## consp 0.000 0.000 0.000
##
## Intercepts:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .conf_01 5.468 0.058 94.213 0.000 5.468 3.587
## .conf_02 5.412 0.056 95.866 0.000 5.412 3.650
## .conf_03 6.123 0.048 127.012 0.000 6.123 4.835
## .cmpcy_01 6.184 0.052 119.564 0.000 6.184 4.552
## .cmpcy_02 6.106 0.054 112.707 0.000 6.106 4.291
## .cmpcy_03 5.825 0.060 96.577 0.000 5.825 3.677
## .const_01 5.851 0.056 104.457 0.000 5.851 3.977
## .const_02 4.916 0.071 69.280 0.000 4.916 2.637
## .const_03 6.314 0.046 138.422 0.000 6.314 5.270
## .calc_01 3.451 0.084 40.998 0.000 3.451 1.561
## .calc_02 3.886 0.083 46.999 0.000 3.886 1.789
## .calc_03 3.849 0.083 46.261 0.000 3.849 1.761
## .colr_01 6.012 0.056 107.033 0.000 6.012 4.075
## .colr_02 6.378 0.045 142.021 0.000 6.378 5.407
## .colr_03 6.109 0.052 117.408 0.000 6.109 4.470
## .cmpli_01 4.465 0.077 57.960 0.000 4.465 2.207
## .cmpli_02 5.478 0.061 89.390 0.000 5.478 3.403
## .cmpli_03 4.070 0.076 53.560 0.000 4.070 2.039
## .consp_01 5.936 0.055 107.448 0.000 5.936 4.090
## .consp_02 4.952 0.065 76.076 0.000 4.952 2.896
## .consp_03 5.548 0.063 88.312 0.000 5.548 3.362
## g 0.000 0.000 0.000
## cmpcy 0.000 0.000 0.000
## const 0.000 0.000 0.000
## calc 0.000 0.000 0.000
## colres 0.000 0.000 0.000
## cmpli 0.000 0.000 0.000
## consp 0.000 0.000 0.000
##
## Variances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .conf_01 1.642 0.125 13.157 0.000 1.642 0.707
## .conf_02 1.547 0.126 12.301 0.000 1.547 0.704
## .conf_03 0.833 0.086 9.719 0.000 0.833 0.519
## .cmpcy_01 0.877 0.237 3.698 0.000 0.877 0.475
## .cmpcy_02 0.594 0.369 1.610 0.107 0.594 0.293
## .cmpcy_03 1.620 0.147 11.042 0.000 1.620 0.645
## .const_01 0.675 1.004 0.673 0.501 0.675 0.312
## .const_02 2.119 0.180 11.802 0.000 2.119 0.610
## .const_03 1.227 0.123 10.005 0.000 1.227 0.854
## .calc_01 4.218 0.241 17.482 0.000 4.218 0.863
## .calc_02 1.369 0.701 1.953 0.051 1.369 0.290
## .calc_03 3.625 0.241 15.063 0.000 3.625 0.759
## .colr_01 0.671 0.143 4.710 0.000 0.671 0.308
## .colr_02 0.609 0.098 6.217 0.000 0.609 0.438
## .colr_03 0.444 0.128 3.454 0.001 0.444 0.238
## .cmpli_01 1.928 0.274 7.048 0.000 1.928 0.471
## .cmpli_02 1.494 0.106 14.079 0.000 1.494 0.577
## .cmpli_03 1.034 0.378 2.735 0.006 1.034 0.260
## .consp_01 1.382 0.157 8.819 0.000 1.382 0.656
## .consp_02 2.077 0.169 12.258 0.000 2.077 0.710
## .consp_03 1.122 0.399 2.807 0.005 1.122 0.412
## g 1.000 1.000 1.000
## cmpcy 1.000 1.000 1.000
## const 1.000 1.000 1.000
## calc 1.000 1.000 1.000
## colres 1.000 1.000 1.000
## cmpli 1.000 1.000 1.000
## consp 1.000 1.000 1.000# Reliability
knitr::kable(reliability(fit.mod_7c), 'simple', caption = 'Reliability of 7C bifactor model')| g | cmpcy | const | calc | colres | cmpli | consp | |
|---|---|---|---|---|---|---|---|
| alpha | 0.8662 | 0.7181 | 0.5933 | 0.5370 | 0.8384 | 0.7488 | 0.6395 |
| omega | 0.8937 | 0.4900 | 0.2779 | 0.5428 | 0.7047 | 0.7023 | 0.4898 |
| omega2 | 0.7994 | 0.2425 | 0.1322 | 0.4883 | 0.3339 | 0.4935 | 0.3253 |
| omega3 | 0.7915 | 0.2425 | 0.1322 | 0.4883 | 0.3339 | 0.4935 | 0.3253 |
| avevar | NA | NA | NA | NA | NA | NA | NA |
logistic regression: binary vacciantion intention variable is regressed on 7C short version items
mod <- glm(vacc_dich ~ conf_03.z + cmpcy_03.z + const_02.z + calc_02.z + colr_02.z + cmpli_03.z + consp_01.z, data = df, family = "binomial")
summary(mod)##
## Call:
## glm(formula = vacc_dich ~ conf_03.z + cmpcy_03.z + const_02.z +
## calc_02.z + colr_02.z + cmpli_03.z + consp_01.z, family = "binomial",
## data = df)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -2.981 0.141 0.242 0.399 2.710
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 2.385 0.176 13.55 < 2e-16 ***
## conf_03.z 0.846 0.142 5.98 2.3e-09 ***
## cmpcy_03.z 0.447 0.136 3.29 0.00101 **
## const_02.z 0.504 0.157 3.21 0.00133 **
## calc_02.z 0.201 0.161 1.25 0.21217
## colr_02.z 0.373 0.129 2.89 0.00379 **
## cmpli_03.z 0.179 0.154 1.16 0.24506
## consp_01.z 0.486 0.131 3.70 0.00021 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 634.48 on 689 degrees of freedom
## Residual deviance: 371.28 on 682 degrees of freedom
## AIC: 387.3
##
## Number of Fisher Scoring iterations: 6tab <- coef(summary(mod)) # regression coefficients
knitr::kable(tab, 'simple', caption = 'Regression coefficients')| Estimate | Std. Error | z value | Pr(>|z|) | |
|---|---|---|---|---|
| (Intercept) | 2.3854 | 0.1760 | 13.551 | 0.0000 |
| conf_03.z | 0.8462 | 0.1415 | 5.978 | 0.0000 |
| cmpcy_03.z | 0.4468 | 0.1359 | 3.287 | 0.0010 |
| const_02.z | 0.5035 | 0.1569 | 3.209 | 0.0013 |
| calc_02.z | 0.2014 | 0.1615 | 1.248 | 0.2122 |
| colr_02.z | 0.3726 | 0.1287 | 2.895 | 0.0038 |
| cmpli_03.z | 0.1790 | 0.1540 | 1.162 | 0.2451 |
| consp_01.z | 0.4865 | 0.1313 | 3.704 | 0.0002 |
#'## Parameters
logit.7c.short.fact <- as.data.frame(mod[["coefficients"]])
odd.7c.short.fact <- as.data.frame(exp(mod[["coefficients"]]))
cond.7c.short.fact <- as.data.frame(exp(mod[["coefficients"]])/(1+exp(mod[["coefficients"]])))
summary.table.7c.short_fact <- cbind(logit.7c.short.fact, odd.7c.short.fact, cond.7c.short.fact)
colnames(summary.table.7c.short_fact) <- c("logits", "Odds", "cond.prob.")
rownames(summary.table.7c.short_fact) <- c("Intercept", "Conf", "Comp", "Const", "Calc", "ColRes", "Compl", "Consp")
knitr::kable(summary.table.7c.short_fact, 'simple', caption = 'Regression coefficients as logits, odds and conditional probabilities')| logits | Odds | cond.prob. | |
|---|---|---|---|
| Intercept | 2.3854 | 10.864 | 0.9157 |
| Conf | 0.8462 | 2.331 | 0.6998 |
| Comp | 0.4468 | 1.563 | 0.6099 |
| Const | 0.5035 | 1.655 | 0.6233 |
| Calc | 0.2014 | 1.223 | 0.5502 |
| ColRes | 0.3726 | 1.452 | 0.5921 |
| Compl | 0.1790 | 1.196 | 0.5446 |
| Consp | 0.4865 | 1.627 | 0.6193 |
#'## Nagelkerke R Square
r2.nagel.7c_short.f.mean <- r2_nagelkerke(mod)
knitr::kable(r2.nagel.7c_short.f.mean, 'simple', caption = 'Nagelkerks R-square')| x | |
|---|---|
| Nagelkerke's R2 | 0.5274 |
Allaire, JJ, Yihui Xie, Jonathan McPherson, Javier Luraschi, Kevin Ushey, Aron Atkins, Hadley Wickham, Joe Cheng, Winston Chang, and Richard Iannone. 2021. Rmarkdown: Dynamic Documents for R. https://CRAN.R-project.org/package=rmarkdown.