Network analysis glossary

Network density

Network density is the share of all possible ties that actually exist. A density of 1 means everyone is tied to everyone; a density of 0.1 means one pair in ten is. It is the quickest single number for how tightly knit a group is, and it falls quickly as a network grows.

Updated · Netgraf

Worked example: Krackhardt's kiteOpen the map's insights →
MeasureValue
People10
Ties drawn18
Ties possible45
Density0.400 (40%)

18 / 45 = 0.400. Computed live from the example map with the same code the Netgraf insights panel runs.

How it's calculated

undirected: density = ties / (n (n - 1) / 2)
directed:   density = ties / (n (n - 1))
n is the number of people. The bottom line is the number of ties that could exist.

Reading it

Density is only comparable between networks of similar size. A person can keep up only so many relationships, so as n grows the possible ties grow with n squared while the real ones grow roughly with n. A team of eight at 0.6 and a company of eight hundred at 0.02 can both be perfectly healthy.

Average degree often says the same thing more usefully: "each person has about five ties" is easier to picture than "density 0.14". Netgraf shows both.

In Netgraf

Density is one of the tiles at the top of the Insights panel, as a percentage of possible ties drawn, next to Avg. ties each. Its detail line reads "ties drawn over ties possible".

Example map

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

Questions

What is a good network density?
There is no universal target. Small teams often run from 0.3 to 0.7. Large organisations and online networks are usually well under 0.05. Compare density only against networks of a similar size and kind.
Does density count a tie in both directions twice?
In an undirected network each pair is counted once. In a directed network, A to B and B to A are separate possible ties, so the denominator doubles.

References

  1. Wasserman, S., & Faust, K. (1994). Social Network Analysis: Methods and Applications. Cambridge University Press.

Draw your own

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