NetworkComparisonTest: Statistical Comparison of Two Networks Based on Invariance Measures

A permutation-based hypothesis test for comparing two networks, developed by Claudia D. van Borkulo and described in a 2021 publication. The method assesses differences based on several invariance measures, including network structure, global strength, edges, and centrality. It is suited for both independent and dependent sample comparisons and uses l1-regularization for network estimation.

Use Cases

• Testing for group differences in psychological network structures based on invariance measures.
• Comparing network changes over time within a single group using the dependent samples method.
• Evaluating the significance of differences in edge weights or node centrality between two networks.
• Validating network stability across different samples or measurement conditions.

Strengths

• Method is explicitly designed for both independent and dependent sample comparisons.
• Implements multiple invariance measures, including network structure, global strength, and edge invariance.
• Based on a peer-reviewed methodology published in 2021.

Limitations

• Row count and dataset scale are unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• Last update date is unknown; freshness unverified.

Provenance

Source

Claudia D. van Borkulo

Collection Method

Statistical method and associated test implementation.