itemDiagnostics: Diagnostics Analysis for Low Stability Items

In EGAnet: Exploratory Graph Analysis – a Framework for Estimating the Number of Dimensions in Multivariate Data using Network Psychometrics

Diagnostics Analysis for Low Stability Items

Description

Computes item diagnostics to determine whether there exist several potential psychometric issues that affect itemStability. These issues include local dependence (using UVA), minor dimensions (using cosine on the stability patterns), multidimensional items (using net.loads), and items with low loadings (using net.loads)

Usage

itemDiagnostics(data, ...)

Arguments

data	Matrix or data frame. Should consist only of variables to be used in the analysis
...	Additional arguments to pass on to bootEGA, net.loads, and UVA

Value

Returns a list containing:

diagnostics	A data frame containing the diagnostics of low item stabilities (see itemStability)
boot	Output from bootEGA
uva	Output from UVA
minor	A list containing suggested items to keep, remove, and a matrix for probable minor dimensions (minor.matrix)
loadings	Output from net.loads
suggested	Variables that are suggested to be retained to increase item stability

Author(s)

Alexander P. Christensen <[e-mail]>, Hudson Golino <hfg9s at virginia.edu>, and Luis Eduardo Garrido <[e-mail]>

Examples

# Load data
wmt <- wmt2[,7:24]

## Not run: 
# Obtain diagnostics
diagnostics <- itemDiagnostics(wmt, ncores = 2)
## End(Not run)

EGAnet documentation built on April 13, 2026, 5:07 p.m.