R/tree_udif.R

In DIFtree: Item Focussed Trees for the Identification of Items in Differential Item Functioning

tree_udif <-
function(Y,
                      DM_kov,
                      npersons,
                      nitems,
                      nvar,
                      sumscore,
                      ordered_values,
                      n_levels,
                      n_s,
                      alpha,
                      nperm,
                      trace){
  # initializations 
  mod_potential    <- list() 
  splits           <- c() 
  devs             <- c()
  crits            <- c()
  pvalues          <- c() 
  
  splits_evtl      <- list() 
  splits_evtl[[1]] <- lapply(1:nitems,function(j) lapply(1:nvar, function(var) matrix(1:n_s[var],nrow=1)))
  
  vars_evtl        <- list() 
  vars_evtl[[1]]   <- lapply(1:nitems,function(j) nvar)
  
  gammas            <- list()
  gammas[[1]]       <- lapply(1:nitems, function(j) "int")
  
  which_obs        <- list() 
  which_obs[[1]]   <- lapply(1:nitems,function(j) matrix(1:npersons,nrow=1))
  
  numbers          <- list()
  numbers[[1]]     <- lapply(1:nitems,function(j) 1)
  
  count            <- 1 
  
  design_lower <- designlists_logistic(DM_kov,nvar,ordered_values,n_levels,n_s)[[1]]
  design_upper <- designlists_logistic(DM_kov,nvar,ordered_values,n_levels,n_s)[[2]]
  sig <- TRUE
  anysplit <- TRUE
  
  # compute models without DIF 
  mod_potential[[count]] <- mod0 <- lapply(1:nitems, function(i){
    dat0 <- as.data.frame(cbind("y"=Y[,i],"int"=rep(1,npersons),sumscore))
    suppressWarnings(
      glm(y~int+sumscore-1,family=binomial(link="logit"),data=dat0)
    )
  })
  
  # function to compute all possible models 
  allmodels <- function(i,var,kn,design_lower,design_upper){
    
    deviances <- rep(0,n_s[var])
    dat   <- as.data.frame(cbind("y"=Y[,i],"int"=rep(1,npersons),sumscore,do.call(cbind,design_lower),do.call(cbind,design_upper)))
    splits_aktuell <- splits_evtl[[count]][[i]][[var]][kn,]
    splits_aktuell <- splits_aktuell[!is.na(splits_aktuell)]
    
    if(length(splits_aktuell)>0){
      
      for(j in splits_aktuell){
        
        help1 <- paste(c(gammas[[count]][[i]][-kn]))
        help2 <- paste(gammas[[count]][[i]][kn],c(colnames(design_lower[[var]])[j],colnames(design_upper[[var]])[j]),sep=":")
        help3 <- paste(c(help1,help2), collapse="+")
        help4 <- formula(paste("y~sumscore+",help3,"-1"))
        suppressWarnings(
          mod   <- glm(help4,family=binomial(link="logit"),data=dat)
        )
        deviances[j] <- deviance(mod0[[i]]) - deviance(mod)
      }
    }
    return(deviances)
  }
  
  # estimate tree 
  while(sig & anysplit){
    
    # compute all models 
    dv <- lapply(1:nvar,function(var) {
      lapply(1:nitems,function(i) {
        n_knots   <- length(gammas[[count]][[i]])
        deviances <- matrix(rep(0,n_s[var]*n_knots),ncol=n_knots)
        for(kn in 1:n_knots){
          deviances[,kn] <- allmodels(i,var,kn,design_lower,design_upper)
        }
        return(deviances)
      })
    })
    
    # select optimum
    variable    <- which.max(lapply(1:nvar,function(j) max(unlist(dv[[j]]))))
    item        <- which.max(lapply(1:nitems, function(j) max(dv[[variable]][[j]])))
    split       <- as.numeric(which(dv[[variable]][[item]]==max(dv[[variable]][[item]]),arr.ind=TRUE)[,1])
    knoten      <- as.numeric(which(dv[[variable]][[item]]==max(dv[[variable]][[item]]),arr.ind=TRUE)[,2])
    gammas_old   <- gammas[[count]][[item]][knoten]
    level       <- length(strsplit(gammas_old,":")[[1]])
    number      <- numbers[[count]][[item]][knoten]
    left        <- max(numbers[[count]][[item]])+1
    right       <- max(numbers[[count]][[item]])+2
    
    # compute permutation test 
    dev <- rep(NA,nperm)
    
    for(perm in 1:nperm){
      dv_perm <- rep(0,n_s[variable])
      obs_aktuell <- which_obs[[count]][[item]][knoten,]
      obs_aktuell <- obs_aktuell[!is.na(obs_aktuell)]
      DM_kov_perm <- DM_kov
      DM_kov_perm[obs_aktuell,variable] <- sample(DM_kov_perm[obs_aktuell,variable],length(obs_aktuell))
      design_upper_perm      <- designlists_logistic(DM_kov_perm,nvar,ordered_values,n_levels,n_s)[[1]]
      design_lower_perm      <- designlists_logistic(DM_kov_perm,nvar,ordered_values,n_levels,n_s)[[2]]
      dv_perm <- allmodels(item,variable,knoten,design_lower_perm,design_upper_perm)
      dev[perm] <- max(dv_perm)
      if(trace){
        cat(".")
      }
    }
    
    # test decision 
    crit_val       <- quantile(dev,1-(alpha/vars_evtl[[count]][[item]][knoten]))
    proof          <- max(dv[[variable]][[item]]) > crit_val
    devs[count]    <- max(dv[[variable]][[item]])
    crits[count]   <- crit_val
    pvalues[count] <- length(which(dev>max(dv[[variable]][[item]])))/nperm
    
    if(proof){
      
      # get new formula 
      help_kn2 <- gammas[[count]][[item]][knoten]
      help5 <- paste(c(gammas[[count]][[item]][-which(gammas[[count]][[item]]==help_kn2)]))
      help6 <- paste(gammas[[count]][[item]][knoten],c(colnames(design_lower[[variable]])[split],colnames(design_upper[[variable]])[split]),sep=":")
      help7 <- paste(c(help5,help6),collapse="+")
      help8 <- formula(paste("y~sumscore+",help7,"-1"))
      
      ######################
      if(level>1){
        help_kn4 <- lu(c(),1,level-1,c())
        help_kn5 <- unlist(strsplit(help_kn2,""))
        help_kn6 <- paste0(help_kn5[which(help_kn5=="_")+1],collapse="")
        knoten2  <- which(help_kn4==help_kn6)
      } else{
        knoten2 <- knoten
      }
      ######################     
      
      splits <- rbind(splits,c(variable,item,split,level,knoten2,number,left,right))   
      
      # fit new model 
      dat   <- as.data.frame(cbind("y"=Y[,item],"int"=rep(1,npersons),sumscore,do.call(cbind,design_lower),do.call(cbind,design_upper)))
      mod_potential[[count+1]] <- mod_potential[[count]]
      suppressWarnings(
        mod_potential[[count+1]][[item]] <- glm(help8,family=binomial(link="logit"),data=dat)
      )
      mod0  <- mod_potential[[count+1]]
      
      # generiere neue gamma-Parameter
      gammas[[count+1]]                             <- gammas[[count]]
      gammas[[count+1]][[item]]                     <- rep("",length(gammas[[count]][[item]])+1)
      gammas[[count+1]][[item]][c(knoten,knoten+1)] <- help6
      gammas[[count+1]][[item]][-c(knoten,knoten+1)]<- gammas[[count]][[item]][-knoten]
      
      # passe splits_evtl an
      n_knots                                                       <- length(gammas[[count+1]][[item]])
      splits_evtl[[count+1]]                                        <- splits_evtl[[count]]
      for(var in 1:nvar){
        splits_evtl[[count+1]][[item]][[var]]                       <- matrix(0,nrow=n_knots,ncol=n_s[var])
        splits_evtl[[count+1]][[item]][[var]][c(knoten,knoten+1),]  <- matrix(rep(splits_evtl[[count]][[item]][[var]][knoten,],2),nrow=2,byrow=T)
        splits_evtl[[count+1]][[item]][[var]][-c(knoten,knoten+1),] <- splits_evtl[[count]][[item]][[var]][-knoten,]
      }
      splits_evtl[[count+1]][[item]][[variable]][knoten,splits_evtl[[count+1]][[item]][[variable]][knoten,]>=split] <- NA 
      splits_evtl[[count+1]][[item]][[variable]][(knoten+1),splits_evtl[[count+1]][[item]][[variable]][(knoten+1),]<=split] <- NA
      
      # any split? 
      anysplit <- !all(is.na(unlist(splits_evtl[[count+1]])))
      
      # passe vars_evtl an 
      vars_evtl[[count+1]]                             <- vars_evtl[[count]]
      vars_evtl[[count+1]][[item]]                     <- rep(0,n_knots)
      vars_evtl[[count+1]][[item]][c(knoten,knoten+1)] <- rep(vars_evtl[[count]][[item]][knoten],2)
      vars_evtl[[count+1]][[item]][-c(knoten,knoten+1)]<- vars_evtl[[count]][[item]][-knoten]
      
      if(length(which(!is.na(splits_evtl[[count+1]][[item]][[variable]][knoten,])))==0){ 
        vars_evtl[[count+1]][[item]][knoten] <- vars_evtl[[count+1]][[item]][knoten]-1 
      }
      if(length(which(!is.na(splits_evtl[[count+1]][[item]][[variable]][knoten+1,])))==0){ 
        vars_evtl[[count+1]][[item]][knoten+1] <- vars_evtl[[count+1]][[item]][knoten+1]-1 
      }
      
      # passe which_obs an 
      which_obs[[count+1]]                               <- which_obs[[count]]
      which_obs[[count+1]][[item]]                       <- matrix(0,nrow=n_knots,ncol=npersons)
      which_obs[[count+1]][[item]][c(knoten,knoten+1),]  <- matrix(rep(which_obs[[count]][[item]][knoten,],2),nrow=2,byrow=T)
      which_obs[[count+1]][[item]][-c(knoten,knoten+1),] <- which_obs[[count]][[item]][-knoten,]
      thresh <- ordered_values[[variable]][1:n_s[variable]][split]
      which_obs[[count+1]][[item]][knoten,DM_kov[,variable]>thresh] <- NA
      which_obs[[count+1]][[item]][(knoten+1),DM_kov[,variable]<=thresh] <- NA
      
      # passe numbers an 
      numbers[[count+1]]                              <- numbers[[count]]
      numbers[[count+1]][[item]]                      <- numeric(length=n_knots)
      numbers[[count+1]][[item]][c(knoten,knoten+1)]  <- c(left,right)
      numbers[[count+1]][[item]][-c(knoten,knoten+1)] <- numbers[[count]][[item]][-knoten] 
      
      # trace
      if(trace){
        cat(paste0("\n Split"," ",count,";"," ","Item"," ",item,"\n"))
      }
      
      # counter
      count <- count+1 
    } else{
      sig <- FALSE
    }
  }
  
  ################################################################################### 
  
  mod_opt     <- mod_potential[[count]]
  gammas_opt   <- gammas[[count]]
  beta_hat    <- sapply(1:nitems, function(j) coef(mod_opt[[j]])["sumscore"])
  names(beta_hat) <- paste0("beta",1:nitems)
  
  if(count>1){
    
    dif_items   <- unique(splits[,2])
    nodif_items <- c(1:nitems)[-dif_items]
    
    if(length(nodif_items)!=0){
      gammas_hat_nodif <- sapply(nodif_items, function(j) coef(mod_opt[[j]])["int"])
      names(gammas_hat_nodif) <- paste0("gamma",nodif_items)
    } else{
      gammas_hat_nodif <- c() 
    }
    
    gammas_hat_dif   <- lapply(dif_items, function(j) coef(mod_opt[[j]])[gammas_opt[[j]]])
    names(gammas_hat_dif) <- dif_items
    
    help9 <- cumsum(c(0,(n_levels-1)))
    colnames(splits) <- c("var","item","split","level","node","number","left","right")
    splits <- data.frame(cbind(splits[,1:5,drop=FALSE],"variable"=rep(NA,nrow(splits)),"threshold"=rep(NA,nrow(splits)),splits[,6:8,drop=FALSE]))
    for(i in 1:nrow(splits)){
      splits[i,6] <- colnames(DM_kov)[splits[i,1]]
      v2 <- lapply(1:nvar,function(j) ordered_values[[j]][-length(ordered_values[[j]])])
      splits[i,7] <- v2[[splits[i,1]]][splits[i,3]]
    }
    splits <- splits[,-1]
    
    for(i in dif_items){
      info <- splits[splits[,"item"]==i,]
      endnodes <- get_endnodes(info)
      names(gammas_hat_dif[[paste(i)]]) <- endnodes 
    }
    
  } else{
    
    gammas_hat_nodif  <- sapply(1:nitems, function(j) coef(mod_opt[[j]])["int"])
    names(gammas_hat_nodif) <- paste0("gamma",1:nitems)
    gammas_hat_dif    <- c()
    
  }
  to_return <- list("splits"=splits,
                    "betas"=beta_hat,
                    "gammas_nodif"=gammas_hat_nodif,
                    "gammas_dif"=gammas_hat_dif,
                    "pvalues"=pvalues,
                    "devs"=devs,
                    "crits"=crits)
  
  return(to_return)
  
}