pleLMA (version 0.2.1)

reScaleItem: Re-scales the category scale values and Phi after convergence of the nominal model

Description

This auxillary function only applies to nominal models after estimating the parameters of the model. During estimation that scaling identification constraint is put on conditional variances (i.e., phi_mm) such that they equal 1. This function provide an alternative identification constraint after the algorithm has converged. This function allow a user to tease apart the strength and structure of associations. The alternative scaling identification constraint is to set the sum of category scale values equals 0 and the sum of squares equal to 1. The phi parameters are adjusted accordingly. Only one item per trait should be selected for the identification constraint and this is indicated by the object anchor.

Usage

reScaleItem(model.fit, anchor)

Arguments

model.fit

Model object for a nominal model

anchor

Indicator of item(s) to place scaling constraint on.

Value

sNu Re-scaled category scale values

sPhi.mat Re-scale phi matrix (conditional covariance matrix)

Examples

# NOT RUN {
# }
# NOT RUN {
#--- 3 items from depression, anxiety, and stress scales
#    for 250 cases out of possible 1000
data(dass)
inData <- dass[1:250,c("d1", "d2", "d3", "a1","a2","a3","s1","s2","s3")]
inTraitAdj  <- matrix(1, nrow=1, ncol=1)
inItemTraitAdj <- matrix(1, nrow=9, ncol=1)
#--- nominal response model
n1 <- ple.lma(inData, model.type="nominal",inItemTraitAdj,inTraitAdj,tol=1e-03)
anchor <- c(1,0,0,0,0,0,0,0,0)
reScaleItem(model.fit=n1, anchor)

#--- Multidimensional models
inTraitAdj  <- matrix(1, nrow=3, ncol=3)
dpress <- matrix(c(1,0,0), nrow=3, ncol=3, byrow=TRUE)
anxiety <- matrix(c(0,1,0), nrow=3, ncol=3, byrow=TRUE)
stress <- matrix(c(0,0,1), nrow=3, ncol=3, byrow=TRUE)
das <- list(dpress, anxiety, stress)
inItemTraitAdj <- rbind(das[[1]], das[[2]], das[[3]])

n3 <- ple.lma(inData, model.type="nominal", inItemTraitAdj, inTraitAdj, tol=1e-03)
anchor <- c(1,0,0, 0,1,0, 1,0,0)
reScaleItem(model.fit=n3, anchor)

# }
# NOT RUN {
# }