LRtest: Likelihood ratio model test

Description

Computation of Andersen's LR-test.

Usage

## S3 method for class 'Rm':
LRtest(object, splitcr = "median", alpha = 0.05, se = TRUE)
## S3 method for class 'LR':
plotGOF(x, beta.subset = "all", xlab = "Beta Group 1", 
ylab = "Beta Group 2", ylim = c(-3, 3), xlim = c(-3, 3), type = "p", ...)
## S3 method for class 'LR':
print(x,...)
## S3 method for class 'LR':
summary(object,...)

Arguments

object

Object of class Rm.

splitcr

Split criterion for subject raw score splitting. all.r corresponds to a full raw score split, median uses the median as split criterion, mean performs a mean-split. Optionally splitcr can also be

alpha

Significance level for Chi-square statistic.

If TRUE, standard errors are computed

Object of class LR for visualizing the fit of single items.

beta.subset

Either a numeric vector or a character vector indicating the columns of the data matrix which should be plotted. If "all", all items are plotted.

xlab

Label of the x-axis.

ylab

Label of the y-axis.

xlim

Parameter range for group 1.

ylim

Parameter range for group 2.

type

Type of plot desired. By default points are plotted.

...

Additional plot parameters.

Value

LRtest returns an object of class LR containing:
LRLR-value.
dfDegrees of freedom of the test statistic.
ChisqChi-square value with corresponding df and alpha.
pvalueP-value of the test.
likgroupLog-likelihood values for the subgroups
betalistList of beta parameters for the subgroups.

References

Andersen, E. B. (1973). A goodness of fit test for the Rasch model. Psychometrika, 38, 123-140. Mair, P., and Hatzinger, R. (2007). Extended Rasch modeling: The eRm package for the application of IRT models in R. Journal of Statistical Software, 20(9), 1-20.

Examples

Run this code

# LR-test on dichotomous Rasch model with median split
data(raschdat1)
res <- RM(raschdat1)
lrres <- LRtest(res)
print(lrres)
summary(lrres)
plotGOF(lrres)

# LR-test on dichotomous Rasch model with user-specified split
splitvec <- sample(1:3, 100, replace = TRUE)          #3 random subgroups
lrres <- LRtest(res, splitcr = splitvec)
print(lrres)
summary(lrres)
plotGOF(lrres)