omg | R Documentation |
As Stuive, Kiers & Timmerman(2009) said,"The OMG method considers for each item the observed correlations between the item and each subtest, where each subtest is an unweighted sumscore of items. These correlations are corrected for two sources of spuriously high correlations, namely, self-correlation and test length[...] An assignment of an item to a subtest is considered to be correct when the corrected subtest-item correlation is highest for the subtest the item is supposed to belong to and incorrect otherwise" (p.950)
omg(x, keys, partialize = NULL, ...)
x |
matrix or data.frame |
keys |
keys of assigment of |
... |
arguments to cor.corrected |
a data.frame
Item name
Name of scale (came from keys)
Corrected correlation of item to assigned scale
Name of scale for maximum correlation
Maximum correlation to any scale
Difference between assigned and maximum correlation.
cor.corrected
for corrected correlation for self-correlation and scales lengths.
library(qgraph)
data(big5)
data(big5groups)
big.keys<-sapply(big5groups,function(x) {xx<-numeric(ncol(big5));xx[x]<-1;xx})
rownames(big.keys)<-colnames(big5)
omg(big5,big.keys)
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.