digram.estimate: Estimate RDigram object using TAM

In jeppebundsgaard/RDigram: Analysis of Items from Tests and Surveys

Estimate RDigram object using TAM

Description

Estimate RDigram object using TAM

Usage

digram.estimate(
  do,
  items = NULL,
  groups = NULL,
  ncases = 0,
  constraint = "cases",
  use.package = c("TAM", "eRm"),
  collapse.testlets = F,
  init.model = NULL,
  tam.control = list(),
  sum0 = T,
  verbose = T,
  ...
)

Arguments

do	A digram.object
items	The items to include in the analysis
groups	Names or column numbers of exogenous variables to use for grouping in TAM. If more names are given, all combinations of values are calculated and used as grouping variables.
ncases	Number of cases to sample for the estimation (0 uses all cases)
constraint	Constraint on "cases" or "items"
use.package	Which R package to use for the estimation. TAM and eRm are implemented.
collapse.testlets	Testlets are estimated using a bifactorial model in TAM and a data matrix in eRm. Setting collapse.testlets to TRUE calculates super-items instead and estimate a normal polytomous model.
init.model	In TAM, the model that was output from an earlier estimation can be used to set sensible init-values for the estimation.
tam.control	Use this to set control parameters in TAM estimation.
sum0	Set to TRUE if you want eRm to sum the parameters to 0. If FALSE the first parameter is set to 0.
verbose	Set to TRUE to get information about the estimation progress.

Details

Uses either the package TAM or eRm to estimate the model. If items have been coded as testlets, a bifactorial model is used in TAM (tam.fa()). Otherwise tam.mml() is used for estimation. In eRm, testlets are managed by creating an interaction parameter between the testlet items. In this case LPCM() is used for estimation. This is also the case, if groups are provided. Otherwise PCM() is used for estimation.

Value

Returns a TAM result object

References

Wang, W.-C., & Wilson, M. (2005). The Rasch Testlet Model. Applied Psychological Measurement, 29(2), 126–149. https://doi.org/10.1177/0146621604271053 Rijmen, F. (2009). Three multidimensional models for testlet-based tests: Formal relations and an empirical comparison. ETS Research Report Series, 2009(2), i–13. https://doi.org/10.1002/j.2333-8504.2009.tb02194.x

See Also

tam.mml(), tam.fa(), PCM(), LPCM()

Examples

data(DHP)
do<-DHP
mod1<-digram.estimate(do)
summary(mod1)
do2<-code.LD(do,"ef")
mod2<-digram.estimate(do2)
summary(mod2)
mod1$deviance
mod2$deviance
mod1$deviance-mod2$deviance

jeppebundsgaard/RDigram documentation built on Oct. 29, 2023, 7:15 p.m.