statistic_item_score: Discrepency measure for PPC: Item score distribution X2.

Description Usage Arguments Value References

View source: R/postpred.R

Description

This function calculates the item score distribution, i.e., the discrepancy between observed frequencies and posterior predictive frequencies for each item using Pearson's X2 statistic.

Usage

1
statistic_item_score(resp = NULL, prob = NULL)

Arguments

resp

Numeric matrix of dimension N x J (for N persons and J items) containing the responses.

prob

Numeric array of dimension R x N x J x 5 (for N persons, J items with 5 categories, and R iterations).

Value

Vector of length J containing X2 values.

References

Zhu, X., & Stone, C. A. (2012). Bayesian comparison of alternative graded response models for performance assessment applications. Educational and Psychological Measurement, 72, 774-799. doi:10.1177/0013164411434638


hplieninger/mpt2irt documentation built on Aug. 4, 2018, 10:52 a.m.