wperfect: generate data matrices according to LLTM

In christinehohensinn/parAL: Goodness-of-fit of the user-defined weight matrix of the linear logistic test model

Description

generate data matrices according to the original weight matrix

Usage

Wperfect(datmat, weightmat, repl = 50, keepmat = FALSE)

Arguments

datmat	data file
weightmat	original weight matrix
repl	number of weight matrices that are generated
keepmat	if TRUE, the generated random data matrices are kept in an three-dimensional array

Details

The original data matrix and the original weight matrix are taken and an LLTM is estimated. With the resulting estimated parameters and the original data matrix new random data matrices are generated according to the original weight matrix. For each generated data matrix, the Rasch model and an LLTM is estimated with the original weight matrix. The correlation between the item parameters of the Rasch model and the item parameters reconstructed from the parameters estimated by the LLTM is calculated. With this procedure some kind of an upper benchmark of the goodness-of-fit of the weight matrix is generated.

Value

convLLTM	convergence of the estimated LLTMs of the generated data matrices
convRM	convergence of the estimated Rasch models of the generated data matrices
lltmpar	item parameters reconstructed from the estimated parameters of the LLTM of the generated weight matrices
raschpar	estimated Rasch model item parameters for the original data set
corlr	correlation of the re-constructed item parameters by the LLTM and the estimated item parameters of the original Rasch model
sumcorlr	short summary descriptive statistics of the correlations
randommat	array containing the generated data matrices (if argument keepmat is set TRUE)

Author(s)

Christine Hohensinn

References

Baghaei, P. & Hohensinn, C. (submitted). A Method of Q-Matrix Validation for the Linear Logistic Test Model.

Fischer, G. H. (1973). The linear logistic test model as an instrument in educational research. Acta Psychologica, 37, 359-374.

Examples

#generate ten data sets according to the original weight matrix

data(exampledata)
data(exampleweight)

Wperfect(exdat, orig.weight, repl=10)

christinehohensinn/parAL documentation built on May 13, 2019, 7 p.m.