MCA Analysis for manuscript ‘Engage and enjoy - investigating predictors of employee engagement and work satisfaction in equine veterinary professionals’

library(psych)
library(tidyverse)
library(knitr)
library(ordinal)
library(car)
library(emmeans)
library(DHARMa)
library(MASS)
library(FactoMineR)
library(factoextra)
library(lme4)
library(ltm)
library(lmerTest)
library(nFactors)
library(corrplot)
library(mvnormtest)
library(EFA.MRFA)
library(dplyr)

library(readxl)
data <- read_excel("CleanedData_KActon.xlsx")

data$Job <- as.factor(data$Job)
data$Species <- as.factor(data$Species)
data$Country <- as.factor(data$Country)
data$Equine_prop3 <- as.factor(data$Equine_prop3)
data$Type_practice <- as.factor(data$Type_practice)
data$Leisure_comb <- as.factor(data$Leisure_comb)
data$Sport_comb <- as.factor(data$Sport_comb)
data$Racing_comb <- as.factor(data$Racing_comb)
data$Age <- as.factor(data$Age)
data$Age2 <- as.factor(data$Age2)
data$Experience_current2 <- as.factor(data$Experience_current2)
data$Gender <- as.factor(data$Gender)
data$Relationship_status <- as.factor(data$Relationship_status)
data$SDependents <- as.factor(data$Dependents)
data$Experience_all <- as.factor(data$Experience_all)
data$Experience_current <- as.factor(data$Experience_current)
data$you_Vet_comb <- as.factor(data$you_Vet_comb)
data$you_comb_resp <- as.factor(data$you_comb_resp)

dataMCA <- read_excel("CleanedData_KActon.xlsx", sheet = "MCA")
str(dataMCA)

#correlation of raw data
ma <- cor(dataMCA)
corrplot(ma, method = "number")
KMO(r=cor(dataMCA))

fa.none <-fa(r=dataMCA, nfactors = 4, covar = FALSE, SMC = TRUE, fm="pa", max.iter = 100, rotate = "promax", cor      = "poly")
             print(fa.none)
             
fa.diagram(fa.none,digits = 2, sort = T,cex = 0.7)
plot(fa.none)

n_p <- nrow(raw)
R_poly<-fa.none$r

parallel <-fa.parallel(R_poly)

PM.load = fa.none$loadings[,1:2]
plot(PM.load, type="n")
text(PM.load,labels=colnames(dataMCA),cex=.75) 

KMO(R_poly)
cortest.bartlett(R_poly, n = n_p)

text <-fa.none$scores
write.csv(text, file = "text.csv", row.names = FALSE)
df <- read.csv("text.csv")

a <-fa(dataMCA,nfactors = ncol(dataMCA), rotate = "none")
nf <-length(a$e.values)

PcntVarTable <- data.frame(Factor = 1:nf,
                         Eigenval = fa.none$e.values,
                          PcntVar = fa.none$e.values/nf*100)
PcntVarTable$Cumul_Pcnt_var <- cumsum(PcntVarTable$PcntVar)
PcntVarTable[2:4]<- round(PcntVarTable[2:4],2)
PcntVarTable

fa.none$loadings 
round(fa.none$loadings[1:nf,],3)

fa.diagram(fa.none, digits = 2, main = "Factor Diagram", 
           cut = .32, 
           simple = T, 
           errors = T)

round(fa.none$Phi,3)
fa.none$Structure[1:nf,]

PattM   <- fa.none$loadings[1:nf,]
StructM <- fa.none$Structure[1:nf,]
PattM

f4 <- read_excel("CleanedData_KActon.xlsx", sheet = "f4")
cronbach.alpha(f4, CI=TRUE, na.rm = T)

f1 <- read_excel("CleanedData_KActon.xlsx", sheet = "f1")
cronbach.alpha(f1, CI=TRUE, na.rm = T)

f2 <- read_excel("CleanedData_KActon.xlsx", sheet = "f2")
cronbach.alpha(f2, CI=TRUE, na.rm = T)

f3 <- read_excel("CleanedData_KActon.xlsx", sheet = "f3")
cronbach.alpha(f3, CI=TRUE, na.rm = T)