pp_check.hmcdm: Graphical posterior predictive checks for hidden Markov cognitive diagnosis model

Description

pp_check method for class hmcdm.

Usage

# S3 method for hmcdm
pp_check(object, plotfun = "dens_overlay", type = "total_score", ...)

Value

Plots for checking the posterior predictive distributions. The default Plotfun

"dens_overlay" plots density of each dataset are overlaid with the distribution of the observed values.

Arguments

object: a fitted model object of class "hmcdm".
plotfun: A character string naming the type of plot. The list of available plot functions include "dens_overlay", "hist", "stat_2d", "scatter_avg", "error_scatter_avg". The default function is "dens_overlay".
type: A character string naming the statistic to be used for obtaining posterior predictive distribution plot. The list of available types include "total_score", "item_mean", "item_OR", "latency_mean", and "latency_total". The default type is "total_score" which examines total scores of subjects. Type "item_mean" is related to the first order moment and examines mean scores of all the items included in the test. Type "item_OR" is related to the second order moment and examines odds ratios of all item pairs. Types "latency_mean" and "total_latency" are available only for hmcdm objects that include item response time information (i.e., hmcdm object fitted with "DINA_HO_RT" model).
...: Additional arguments

References

Zhang, S., Douglas, J. A., Wang, S. & Culpepper, S. A. (2019) <tools:::Rd_expr_doi("10.1007/978-3-030-05584-4_24")>

Examples

Run this code

# \donttest{
output_FOHM = hmcdm(Y_real_array,Q_matrix,"DINA_FOHM",Design_array,1000,500)
library(bayesplot)
pp_check(output_FOHM)
pp_check(output_FOHM, plotfun="hist", type="item_mean")
# }