What it does
Sociometric surveys ask people to name others: friends, people they go to for advice, colleagues they work with. This tool turns those answers into a network and measures it, in your browser:
- Reads name lists (each respondent followed by the people they named) or a plain edge list (source and target).
- Builds a directed network (A named B), or an undirected one if you untick the direction.
- Measures every person: in-degree (how often named), out-degree, degree centrality, betweenness (brokerage), closeness and eigenvector centrality.
- Measures the whole network: density, reciprocity (share of ties that are returned), weak and strong components.
- Finds communities with the Louvain method and shows the modularity.
- Draws the network with a force-directed layout: bigger dots are named more often, colours are communities, arrows show direction.
Download the measures as CSV, the drawing as SVG, or a full report as an image, PDF or Word file.
It works well with the publication plot maker, the sample size and power calculator and the reliability and agreement.
How to use it
- Choose Name lists or Edge list, then paste your data or open a CSV or Excel file. Untick First row is a header if your data has no header.
- Names must be spelled the same way each time; spaces at the ends are ignored. Someone who is named but did not answer still becomes part of the network.
- Leave Ties have a direction ticked for “who named whom” data. Untick it for ties that are mutual by nature, such as co-authorship.
- Sort the table by any measure. Use Try another layout if the drawing is crowded; the measures do not change.
Worked example (made-up data)
The sample is a made-up class survey: 20 pupils each named two to five friends, and two more pupils (Uma and Vic) were named but did not answer. That gives 22 people and 68 directed ties: density 0.147 and reciprocity 0.353 (12 mutual pairs). Hugo is named most often (in-degree 6) and has the highest eigenvector centrality (0.473) and closeness (0.452). Sam has the highest betweenness (0.237): he links his friendship group to another. The Louvain method finds 3 communities (modularity 0.524) matching three friendship groups: Ava, Ben, Cara, Dev, Ella, Finn, Gita and Hugo; Iris, Jon, Kai, Lena, Milo, Nora and Omar; and Pia, Quinn, Rosa, Sam, Tara, Uma and Vic.
How it was checked
Degree, in-degree and out-degree centrality, betweenness, closeness, eigenvector centrality, density, reciprocity and the components match networkx exactly, for the directed and the undirected network. The Louvain communities reach the same modularity as the best of 20 runs of networkx’s louvain_communities, and the modularity calculation itself matches networkx.
Privacy
Network surveys collect names, which are personal data. This page never uploads or stores anything: the data stays in your browser tab and is gone when you close it. Where you can, replace names with codes before analysis.
Frequently asked questions
Why do people with no answers have an out-degree of 0? They were named but did not fill in the survey. Missing answers lower density and closeness, so report the response rate.
Are tie strengths used? No. Every tie counts once and repeated ties are counted once, as the tool reports.
How big a network can it handle? Up to 2,000 people; names are hidden in the drawing above 150.
Where the numbers come from
- Methods: Freeman (1977, 1978) degree and betweenness, Brandes (2001) algorithm, Wasserman and Faust (1994) closeness, Bonacich (1972) eigenvector centrality, Blondel et al. (2008) Louvain communities, Fruchterman and Reingold (1991) layout. Example data made up.
Standard scientific values, such as constants, atomic weights and melting points, are facts found in every textbook; we name where ours were checked. If you are an author or publisher and want a value, a credit or a table changed or removed, tell us and we will act quickly.
Last updated 2026-11-09.