• net_by_eigenvector() measures the eigenvector centralization for a network as a single score.

  • mode_by_eigenvector() measures eigenvector centralization separately for each mode of a two-mode network (via projection to each mode), returning one score per mode (following Borgatti and Everett, 1997). A multilevel network cannot be projected, so there each mode's scores are read from the whole network.

All measures attempt to use as much information as they are offered, including whether the networks are directed, weighted, or multimodal. If this would produce unintended results, first transform the salient properties using e.g. to_undirected() functions. All centrality and centralization measures return normalized measures by default, including for two-mode networks.

For two-mode networks the two modes have different theoretical maxima, so net_by_eigenvector() reports a single network-level score by applying Freeman's general centralization index over the normalized node eigenvector scores, whereas mode_by_eigenvector() reports the per-mode scores directly.

net_by_eigenvector(.data, normalized = TRUE)

mode_by_eigenvector(.data, normalized = TRUE)

net_by_influence(.data, normalized = TRUE)

Arguments

.data

A network object of class stocnet, igraph, tbl_graph, network, or similar. Internally any of these will be coerced to an efficient implementation. For more information on possible coercions, see e.g. manynet::as_stocnet().

normalized

Logical scalar, whether scores are normalized. Different denominators may be used depending on the measure, whether the object is one-mode or two-mode, and other arguments. By default TRUE.

Value

net_by_eigenvector() returns a network_measure scalar; mode_by_eigenvector() returns a mode_measure numeric vector of length two, giving one centralization score per mode.

Influence centralisation

net_by_influence() measures the largest share of influence that any one node holds (see node_by_influence()). Golub and Jackson (2010) show that a crowd that learns by taking the mean of what others think is wise, reaching the truth as it grows, only where this largest share goes to 0. So a high value means a low capacity for the crowd to be wise. By default the share is normalised as \((\max - 1/n) / (1 - 1/n)\), which is 0 where every node has the same influence and 1 where one node has all of it. With normalized = FALSE it is the largest share itself. As for node_by_influence(), a negative weight counts by its absolute value, and where no first opinion has any weight at the end, the value is 0.

References

Borgatti, Stephen P., and Martin G. Everett. 1997. "Network analysis of 2-mode data." Social Networks 19(3): 243-269. doi:10.1016/S0378-8733(96)00301-2

Examples

net_by_eigenvector(ison_southern_women)
#> # Eigenvector centralisation, normalized [0, 1]
#> [1] 0.419
mode_by_eigenvector(ison_southern_women)
#> # Eigenvector centralisation, normalized [0, 1]
#> Mode 1 Mode 2 
#> 0.0849 0.2630 
net_by_influence(manynet::create_ring(6))
#> # Influence centralisation, normalized [0, 1]
#> [1] 3.33e-17
net_by_influence(manynet::create_star(6))
#> # Influence centralisation, normalized [0, 1]
#> [1] 0.4