data_sim_Rasch {DIFtree}R Documentation

Simulated Data Set with Dichotomous Items

Description

The data set is simulated from a Rasch model where some items exhibit differential item functioning. Existing differences in item difficulties are simulated by step-functions. The true, simulated DIF structure is described in Tutz and Berger (2015), Section 4.2.

Usage

data(data_sim_Rasch)

Format

A data frame containing 500 observations on 5 variables:

Y matrix with binary 0/1 response for 20 items

x1 binary covariate 1

x2 metric covariate 1

x3 binary covariate 2

x4 metric covariate 2

References

Berger, Moritz and Tutz, Gerhard (2016): Detection of Uniform and Non-Uniform Differential Item Functioning by Item Focussed TreesJournal of Educational and Behavioral Statistics 41(6), 559-592.

Tutz, Gerhard and Berger, Moritz (2016): Item focussed Trees for the Identification of Items in Differential Item Functioning, Psychometrika 81(3), 727-750.

Examples

data(data_sim_Rasch)
 
Y <- data_sim_Rasch[,1]
X <- data_sim_Rasch[,-1]

hist(rowSums(Y), breaks = 0:19 + 0.5)
summary(X)
  
 

[Package DIFtree version 3.1.6 Index]