mirt: coef-method – R documentation

Become an expert in R — Interactive courses, Cheat Sheets, certificates and more!

coef-method

Extract raw coefs from model object

Description

Return a list (or data.frame) of raw item and group level coefficients. Note that while the output to the console is rounded to three digits, the returned list of objects is not. Hence, elements from cfs <- coef(mod); cfs[[1]] will contain the unrounded results (useful for simulations).

Usage

## S4 method for signature 'SingleGroupClass'
coef(
  object,
  CI = 0.95,
  printSE = FALSE,
  rotate = "none",
  Target = NULL,
  IRTpars = FALSE,
  rawug = FALSE,
  as.data.frame = FALSE,
  simplify = FALSE,
  unique = FALSE,
  verbose = TRUE,
  ...
)

Arguments

`object`	an object of class `SingleGroupClass`, `MultipleGroupClass`, or `MixedClass`
`CI`	the amount of converged used to compute confidence intervals; default is 95 percent confidence intervals
`printSE`	logical; print the standard errors instead of the confidence intervals? When `IRTpars = TRUE` then the delta method will be used to compute the associated standard errors from mirt's default slope-intercept form
`rotate`	see `summary` method for details. The default rotation is `'none'`
`Target`	a dummy variable matrix indicting a target rotation pattern
`IRTpars`	logical; convert slope intercept parameters into traditional IRT parameters? Only applicable to unidimensional models. If a suitable ACOV estimate was computed in the fitted model, and `printSE = FALSE`, then suitable CIs will be included based on the delta method (where applicable)
`rawug`	logical; return the untransformed internal g and u parameters? If `FALSE`, g and u's are converted with the original format along with delta standard errors
`as.data.frame`	logical; convert list output to a data.frame instead?
`simplify`	logical; if all items have the same parameter names (indicating they are of the same class) then they are collapsed to a matrix, and a list of length 2 is returned containing a matrix of item parameters and group-level estimates
`unique`	return the vector of uniquely estimated parameters
`verbose`	logical; allow information to be printed to the console?
`...`	additional arguments to be passed

References

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

Examples

## Not run: 
dat <- expand.table(LSAT7)
x <- mirt(dat, 1)
coef(x)
coef(x, IRTpars = TRUE)
coef(x, simplify = TRUE)

#with computed information matrix
x <- mirt(dat, 1, SE = TRUE)
coef(x)
coef(x, printSE = TRUE)
coef(x, as.data.frame = TRUE)

#two factors
x2 <- mirt(Science, 2)
coef(x2)
coef(x2, rotate = 'varimax')


## End(Not run)

mirt

Multidimensional Item Response Theory

v1.33.2

GPL (>= 3)

Authors

Phil Chalmers [aut, cre] (<https://orcid.org/0000-0001-5332-2810>), Joshua Pritikin [ctb], Alexander Robitzsch [ctb], Mateusz Zoltak [ctb], KwonHyun Kim [ctb], Carl F. Falk [ctb], Adam Meade [ctb], Lennart Schneider [ctb], David King [ctb], Chen-Wei Liu [ctb], Ogreden Oguzhan [ctb]

Initial release