Function provides the four generalized item difficulty representations for polytomous response models described by Ali, Chang, and Anderson (2015). These estimates are used to gauge how difficult a polytomous item may be.
Usage
gen.difficulty(mod, type = "IRF", interval = c(-30, 30), ...)Arguments
- mod
a single factor model estimated by
mirt- type
type of generalized difficulty parameter to report. Can be
'IRF'to use the item response function (default),'mean'to find the average of the difficulty estimates,'median'the median of the difficulty estimates, and'trimmed'to find the trimmed mean after removing the first and last difficulty estimates- interval
interval range to search for
'IRF'type- ...
additional arguments to pass to
uniroot
References
Ali, U. S., Chang, H.-H., & Anderson, C. J. (2015). Location indices for ordinal polytomous items based on item response theory (Research Report No. RR-15-20). Princeton, NJ: Educational Testing Service. http://dx.doi.org/10.1002/ets2.12065
Chalmers, R., P. (2012). mirt: A Multidimensional Item Response Theory Package for the R Environment. Journal of Statistical Software, 48(6), 1-29. doi:10.18637/jss.v048.i06
Author
Phil Chalmers rphilip.chalmers@gmail.com
Examples
if (FALSE) { # \dontrun{
mod <- mirt(Science, 1)
coef(mod, simplify=TRUE, IRTpars = TRUE)$items
gen.difficulty(mod)
gen.difficulty(mod, type = 'mean')
# also works for dichotomous items (though this is unnecessary)
dat <- expand.table(LSAT7)
mod <- mirt(dat, 1)
coef(mod, simplify=TRUE, IRTpars = TRUE)$items
gen.difficulty(mod)
gen.difficulty(mod, type = 'mean')
} # }