statistic_item_cor: Discrepency measure for PPC: Item-total correlation.

In hplieninger/mpt2irt: Bringing Multinomial Processing Tree Models To Item Response Theory To Investigate Response Styles

Description

This function calculates the polyserial correlation between a person's total score and his or her item response. This discrepency measure is useful to detect misfit due to a missing item-slope/-discrimination parameter. It may not be suitable for the response style models discussed herein, because these models assume that the total score as well as the item response is a composite of target trait and response styles.

Usage

statistic_item_cor(resp = NULL, revItem = NULL, traitItem = NULL)

Arguments

resp	Numeric matrix of dimension N x J (for N persons and J items) containing the observed item responses.
revItem	vector of length J specifying reversed items (1=reversed, 0=regular)
traitItem	vector of length J specifying the underlying traits (e.g., indexed from 1...5). Standard: only a single trait is measured by all items. If the Big5 are measured, might be something like c(1,1,1,2,2,2,...,5,5,5,5)

Value

Vector of length J of polyserial correlations.

References

Li, T., Xie, C., & Jiao, H. (2017). Assessing fit of alternative unidimensional polytomous IRT models using posterior predictive model checking. Psychological Methods, 22, 397-408. doi:10.1037/met0000082

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 May 17, 2019, 4:54 p.m.