dif_logistic: Logistic regression DIF method

In knickodem/WBdif: Investigate DIF and generate reports

Logistic regression DIF method

Description

Conducts DIF analysis with logistic regression

Usage

dif_logistic(item.data, dif.groups, match.type, match.scores)

Arguments

item.data	A data.frame of dichotomous item responses with subjects in rows and items in columns.
dif.groups	factor vector of group membership for which DIF is evaluated.
match.type	For the loess, MH, and logistic methods, a character indicating whether a total summed score ("Total"; default) or the summed score excluding the item under investigation ("Rest") should be used as the stratifying variable.
match.scores	A numeric vector (if match.type = "Total") or list of ncol(item.data) numeric vectors (if match.type = "Rest") of match scores used as the stratifying variable in the MH procedure.

Details

First conducts an omnibus test of DIF by comparing fit of no DIF, uniform DIF, and non-uniform DIF logistic regression models. The no DIF model regresses the dichotomous item responses from item.data on item, match.scores, and the two-way interaction. The uniform DIF model adds dif.groups and the interaction with item while the non-uniform model adds the three-way interaction. If DIF is detected through the model comparisons, the specific item(s) with DIF are identified in a two-stage process - initial detection stage and refinement stage.

Value

list containing

• DIF model comparisons

• item-level DIF tests

• integer vector of the items showing DIF (i.e., biased items)

• type of DIF

knickodem/WBdif documentation built on Feb. 3, 2024, 2:20 a.m.