mirt (version 1.1)

fitIndices: Compute Extra Model Fit Indices

Description

Compute additional model fit indices that do not come as direct results following parameter convergence. Will compute the M2 (Maydeu-Olivares & Joe, 2006) statistic by default, and returns a data.frame containing various model fit statistics.

Usage

fitIndices(obj, calcNull = FALSE, prompt = TRUE)

Arguments

obj

an estimated model object from the mirt package

calcNull

logical; calculate statistics for the null model as well? Allows for statistics such as the limited information TLI and CFI

prompt

logical; prompt user for input if the internal matrices are too large?

References

Maydeu-Olivares, A. & Joe, H. (2006). Limited information goodness-of-fit testing in multidimensional contingency tables Psychometrika, 71, 713-732.

Examples

#LSAT6 example
dat <- expand.table(LSAT6)
(mod1 <- mirt(dat, 1, itemtype = '2PL', constrain = list(c(1,5,9,13,17))))
fitIndices(mod1)

#Science data with computing the null model M2 stat
(mod2 <- mirt(Science, 1))
fitIndices(mod2, calcNull = TRUE)