EGAnet

Network Psychometrics in R

Estimating dimensionality
in multivariate data with network models.

EGAnet represents a set of measured variables as a network and estimates how many dimensions organize it. The package is built around reproducible code and visualizations suitable for publication.

> install.packages("EGAnet")
> library(EGAnet)
Read the quick start → View on CRAN
community.detection() · fig. 1

On CRAN and GitHub

CRAN downloads, total CRAN downloads, per month GitHub stars

The function map

Functions, organized by workflow stage.

From raw items to a finished network model, every stage of the EGA workflow is its own set of composable functions.

EGA Functions

EGA() · bootEGA() · dynEGA() · hierEGA() · riEGA()

Psychometric Functions

itemStability() · invariance() · net.loads() · UVA() · CFA()

EGM Functions

simEGM() — simulate from the Exploratory Graph Model

Information Theory

tefi() · entropyFit() · ergoInfo() · totalCor()

Plots

plot_clusters() · compare.EGA.plots() · color_palette_EGA()

Basic Functions

TMFG() · polychoric.matrix() · community.consensus()

Version 2.5.0

What changed in this release.

UPDATE

community.detection and community.consensus now use {EGAnet}'s own C implementation of the Louvain algorithm (louvain.c) instead of igraph::cluster_louvain.

ADD

community.prism implements Pairwise Resolution Iteration via Subgraph Modularity (PRISM), iteratively testing pairwise unions of communities for a modularity gain at the subgraph level and merging communities whenever the union increases modularity relative to their separate memberships; intended to overcome the modularity resolution limit described by Fortunato and Barthelemy (2007)

Full changelog →