Skip to contents

Summary of network-attribute relationships

Source: R/attribute_report.R
attribute_report.Rd

summarizes how nodal and dyadic attributes relate to network structure. combines homophily analysis, mixing patterns, dyadic correlations, and network position-based attribute summaries.

Usage

attribute_report(
  netlet,
  node_vars = NULL,
  dyad_vars = NULL,
  include_centrality = TRUE,
  include_homophily = TRUE,
  include_mixing = TRUE,
  include_dyadic_correlations = TRUE,
  centrality_measures = c("degree", "betweenness"),
  categorical_threshold = 10,
  significance_test = TRUE,
  other_stats = NULL,
  ...
)

Arguments

netlet

a netify object containing network data.

node_vars

character vector of nodal attributes to analyze. if NULL, analyzes all available nodal variables except actor and time.

dyad_vars

character vector of dyadic attributes to analyze. if NULL, analyzes all available dyadic variables.

include_centrality

logical. whether to calculate attribute-centrality relationships. default TRUE.

include_homophily

logical. whether to perform homophily analysis. default TRUE.

include_mixing

logical. whether to create mixing matrices for categorical attributes. default TRUE.

include_dyadic_correlations

logical. whether to calculate dyadic correlations. default TRUE.

centrality_measures

character vector of centrality measures to calculate. options: "degree", "betweenness", "closeness", "eigenvector". default c("degree", "betweenness").

categorical_threshold

maximum number of unique values for categorical treatment. default 10.

significance_test

logical. whether to perform significance tests. default TRUE.

other_stats

named list of custom functions for additional statistics.

...

additional arguments passed to component functions.

Value

list containing:

homophily_analysis

results from homophily analysis for nodal attributes

mixing_analysis

results from mixing matrix analysis for categorical attributes

dyadic_correlations

results from dyadic correlation analysis

centrality_correlations

correlations between nodal attributes and centrality

attribute_summaries

descriptive statistics for attributes

overall_summary

brief summary of key findings

Details

wraps the exploratory analysis functions and chooses methods based on attribute types. for large networks or many attributes, consider setting some components to FALSE for faster computation. centrality measures use igraph functions.

Author

cassy dorff, shahryar minhas