psych (version 1.0-17)

factor2cluster: Extract cluster definitions from factor loadings

Description

Given a factor or principal components loading matrix, assign each item to a cluster corresponding to the largest (signed) factor loading for that item. Essentially, this is a Very Simple Structure approach to cluster definition that corresponds to what most people actually do: highlight the largest loading for each item and ignore the rest.

Usage

factor2cluster(loads, loading = TRUE, cut = 0)

Arguments

loads

either a matrix of loadings, or the result of a factor analysis/principal components analyis with a loading component

loading

TRUE if taking the output of a FA/PC or ICLUST, FALSE if just a matrix

cut

Extract items with absolute loadings > cut

Value

• a matrix of -1,0,1 cluster definitions for each item.

Details

A factor/principal components analysis loading matrix is converted to a cluster (-1,0,1) definition matrix where each item is assigned to one and only one cluster. This is a fast way to extract items that will be unit weighted to form cluster composites. Use this function in combination with cluster.cor to find the corrleations of these composite scores.

A typical use in the SAPA project is to form item composites by clustering or factoring (see ICLUST, principal), extract the clusters from these results (factor2cluster), and then form the composite correlation matrix using cluster.cor. The variables in this reduced matrix may then be used in multiple R procedures using mat.regress.

References

http://personality-project.org/r/r.vss.html

See Also

cluster.cor, factor2cluster ,principal, ICLUST

Examples

f  <- factanal(x,4,covmat=Harman74.cor$cov)
factor2cluster(f)
#                       Factor1 Factor2 Factor3 Factor4
#VisualPerception             0       1       0       0
#Cubes                        0       1       0       0
#PaperFormBoard               0       1       0       0
#Flags                        0       1       0       0
#GeneralInformation           1       0       0       0
#PargraphComprehension        1       0       0       0
#SentenceCompletion           1       0       0       0
#WordClassification           1       0       0       0
#WordMeaning                  1       0       0       0
#Addition                     0       0       1       0
#Code                         0       0       1       0
#CountingDots                 0       0       1       0
#StraightCurvedCapitals       0       0       1       0
#WordRecognition              0       0       0       1
#NumberRecognition            0       0       0       1
#FigureRecognition            0       0       0       1
#ObjectNumber                 0       0       0       1
#NumberFigure                 0       0       0       1
#FigureWord                   0       0       0       1
#Deduction                    0       1       0       0
#NumericalPuzzles             0       0       1       0
#ProblemReasoning             0       1       0       0
#SeriesCompletion             0       1       0       0
#ArithmeticProblems           0       0       1       0



## The function is currently defined as

function (loads,loading=TRUE,cut=.0) 
{
    if (loading) {l <- loads$loadings
       colnames(l) <- colnames(loads$loading) } else { 
       l <- loads
       colnames(l) <- colnames(loads) }
    l <- as.matrix(l)
    nrows <- dim(l)[1]
    ncols <- dim(l)[2]
   if (ncols ==1) {m1 <- matrix(rep(1,nrows),ncol=1) } else {
    m1 <- matrix(apply(t(apply(l, 1, abs)), 1, which.max), 
        ncol = 1)}
    id <- matrix(c(1:nrows, m1), ncol = 2)  #index row and column
   factor2cluster <- matrix(rep(0, ncols * nrows), ncol = ncols)
   factor2cluster[id] <- sign(l[id])*( (abs(l[id]) >cut)+0)  #only loadings > cut
  rownames(factor2cluster) <- rownames(l)
  colnames(factor2cluster) <- colnames(l)
   nitems <- colSums(abs(factor2cluster))
   for (i in ncols:1) {if (nitems[i]<1) {factor2cluster <- factor2cluster[,-i]} }#remove columns with no variables
    return(factor2cluster)
}