Defining Two-mode Networks¶

Literature › Works

Tore Opsahl (2013). Defining Two-mode Networks. tnet (Tore Opsahl's website) » Two-mode Networks. toreopsahl.com (WordPress.com).


Type	other
Scope	Comparative and methods
Method	theoretical or conceptual
Period studied	-
Themes	methods data sources
Citation key	`opsahl2013b`

Argument¶

The web page defines two-mode (affiliation or bipartite) networks as networks with two distinct node sets in which ties exist only between nodes of different sets. It notes that such networks are rarely analysed directly because most network measures are defined for one-mode networks, so they are usually transformed by projection. Projection creates problems: ties are not created independently, which weakens comparison with random-network baselines, and it produces many cliques and triangles that distort measures such as clustering coefficients and structural holes. The page also explains how weighted two-mode networks can be defined. It has no connection to South African economic history; it is a general network-analysis methods note.

Evidence and method¶

Secondary methodological literature on network analysis, with illustrative examples: Davis's Southern Women dataset, scientific collaboration networks, interlocking directorates of Norwegian boards, and online forums. No South African data.

Findings¶

Two-mode networks have two node sets, with ties only between sets. (p. 1)

These networks are a particular kind, with two different sets of nodes, and ties existing only between nodes belonging to different sets.
Most network measures are defined only for one-mode networks, so two-mode networks are usually transformed. (p. 2)

Two-mode networks are rarely analysed without transforming them. This is because most network measures are solely defined for one-mode networks
Projection links nodes of one set if they share at least one common node in the other set. (p. 2)

linking nodes from that set if they were connected to at least one common node in the other set.
Projected networks violate the independence assumption of classical random-network baselines. (p. 2)

a director forms ties with all the other directors on a board when she or he joins that board.
Projected two-mode networks tend to contain more and larger cliques, which distorts triangle-based measures. (p. 2)

a projected two-mode network tends to have more and larger fully-connected cliques than prototypical one-mode networks
For interlocking directorates it is unclear which node set is primary because tie formation is mutual. (p. 1)

it is not clear whether directors or boards are the primary node set
Weights in weighted two-mode networks should be on a ratio scale. (p. 3)

tie weights should be based on a ratio scale (i.e., 0 equals nothing, and a value of 4 is twice of 2).