Network analysis glossary

Assortativity

Assortativity measures whether people tend to be tied to others like themselves. Degree assortativity, the usual form, asks whether well-connected people link to other well-connected people (positive) or to people with few ties (negative). It runs from -1 to 1.

Updated · Netgraf

Worked example: Zachary's Karate ClubOpen the map's insights →
MeasureValue
Degree assortativity-0.476
Reads asWell connected people tie to quiet ones

Newman degree assortativity, from -1 to 1. Computed live from the example map with the same code the Netgraf insights panel runs.

How it's calculated

r = correlation between the degrees at the two ends of every tie
Newman's coefficient is the Pearson correlation of degree across tie ends. It is undefined when everyone has the same degree.

Reading it

Mark Newman found in 2002 that social networks tend to be assortative, with popular people knowing each other, while technological and biological networks tend to be disassortative, with hubs linking to many small nodes. Small groups built around one or two leaders often come out strongly negative: in Zachary's karate club the two leaders are tied to many members who have few other ties.

In Netgraf

The Shape card in the Insights panel gives the coefficient and a sentence: "well connected people tie to each other", "well connected people tie to quiet ones", or "no pattern between the busy and the quiet".

Example map

Open one to explore it, then press Edit a copy to make it yours.

Questions

Can assortativity be measured on attributes other than degree?
Yes. The same idea applies to any attribute: do people tie to others of the same team, age or party? That is often called homophily, and it is one of the most consistent findings in social science.

References

  1. Newman, M. E. J. (2002). Assortative mixing in networks. Physical Review Letters, 89(20), 208701. Link

Draw your own

Free, no account, and private by default. Start blank, import a spreadsheet, or describe the network in words.