SMD: Standardized Mean Difference (SMD) for DIF detection

In cswells1/MeasInv: Collection of Methods to Detect Dichotomous and Polytomous Differential Item Functioning (DIF)

Description

Standardized Mean Difference (SMD) for DIF detection

Usage

SMD(data, group, focal.name, ref.name, purify)

Arguments

data	numeric: either the data matrix only, or the data matrix plus the vector of group membership.
group	factor: the vector of group membership
focal.name	numeric or character: a single value
ref.name	numeric or character: a single value
purify	logical: Default is FALSE.

Value

The data frame of the output of SMD has a number of rows equal to the number of items. The columns include the SMD, SMD/Max_Cat, and DIF_Classification. SMD is the standardized mean difference estimate. SMD/Max_Cat is the SMD divided by maximum item score. DIF_Classification has three effect sizes (“Negligible”, “Moderate”, or “Large”). The effect size of an item is classified as “negligible” if SMD/Max_Cat < .05, “moderate” .05 ≤ SMD/Max_Cat < .10, or “Large” SMD/Max_Cat ≥ .10

cswells1/MeasInv documentation built on Dec. 19, 2021, 7 p.m.