library("foreign")
library("psych")
library("lavaan")
library("semTools")
library("semPlot")

rm(list = ls())
Mydata=read.spss("ECPP.sav",to.data.frame=T,use.value.labels=F)
#Mydata=subset(Mydata,DeEstudio==1)

###...Estableciendo el modelo
Mymodel<-'
  Impescol =~ cp04+cp11+cp13+cp15+cp21
  Dedipers =~ cp05+cp09+cp10+cp12+cp20
  Ociocomp =~ cp06+cp07+cp08+cp19
  Aseorien =~ cp14+cp16+cp17+cp18  
  Asunpama =~ cp01+cp02+cp03+cp22'



#############...WLSMV estimator y summary......###########
#sem.fit = sem(Mymodel,estimator="MLR",data=Mydata)
sem.fit = sem(Mymodel,ordered=names(Mydata),estimator="WLSMV",data=Mydata)
summary(sem.fit,fit.measures=T,standardized=T,nd=3)
semPaths(sem.fit,whatLabels="std",layout="tree",edge.label.cex=0.9,rotation=2,nCharNodes=15,
         sizeLat=7,sizeMan=7,style="lisrel")
head(modificationindices(sem.fit)[order((modificationindices(sem.fit))$mi,decreasing=TRUE),],15)


#############...Matrices de correlaciones polic?ricas......###########
Vars.for.correl<-c('cp01','cp02','cp03','cp04','cp05','cp06','cp07','cp08','cp09','cp10','cp11','cp12','cp13',
'cp14','cp15','cp16','cp17','cp18','cp19','cp20','cp21','cp22')
Mydataforcorrel<-Mydata[Vars.for.correl]
#Correlaciones polic?ricas
round(polychoric(Mydataforcorrel)$rho,2)
#Correlaciones pearson
round(cor(Mydataforcorrel),2)











####################### (normalidad multivariada)
library(MVN)
result <- mvn(data = Mydataforcorrel, mvnTest = "mardia")
result


#fitmeasures(sem.fit)