310 likes | 513 Views
Chapter 8 Embedding. Introduction. Density Reciprocity Transitivity Clustering Group-external and group-internal ties Krackhardt's graph theoretical dimensions of hierarchy . U nit of analysis. Micro. Actor: People Dyad Triad Subgroups Organizations Collectives/aggregates
E N D
Introduction Density Reciprocity Transitivity Clustering Group-external and group-internal ties Krackhardt's graph theoretical dimensions of hierarchy
Unit of analysis Micro • Actor: • People • Dyad • Triad • Subgroups • Organizations • Collectives/aggregates • Communities • Nation-states 环境 组织 团体 个体 个体是镶嵌在网络中,而个体所镶嵌的网络是镶嵌在更高层及的网络中。 Macro/aggregates
Dyad 两人组 • Binary ties-present, or absent • Directed relations • 有无关系?是否是双向关系?(reciprocal) • 还是单向(asymmetrical)?
Density The more actors are connected to one another, the more dense the network will be. Binary data: the number of present ties/the number of all possible ties Undirected network: n(n-1)/2 = 2n-1 possible pairs of actors. Δ = Directed network: n(n-1)*2/2 = 2n-2possible lines. ΔD= Valued data: the average strength of ties across all possible ties
Density Density=4/(3*2)
Reciprocity • Directed data, four possible dyadic relations • Actor: reciprocity pair (AB)/ all possible pairs (AB, BC, AC) • Dyad method: • the number of pairs with a reciprocity tie/the number of pairs with any tie. (AB)/(AB, BC) • Relations: reciprocal tie(AB, BA)/all possible pairs (AB, BA, BC, CB, AC, CA) • Arc method: • The number of reciprocal ties/ total number of actual ties (AB,BA)/(AB, BA, BC)
Reciprocity Dyad method
Triad 三人组 Tu = the number of triads that belong to isomorphism class u Cg3 T=(T003, T012, ...,T300)‟ =(003, 012, 102, 021D, 021U, 021C, 111D, 111U, 030T, 030C, 201, 120D, 120U, 120C, 210, 300) Un-directed data, four possible types of triadic relations Directed-data, 16 possible types of triadic relations
Clustering “6-degrees” phenomenon, How close actors are together “Clique-like” local neighborhood, the tendency towards dense local neighborhoods. Clustering: the density in neighborhood
Group-external and group-internal ties Measure of group based on comparing the numbers of ties within groups and between groups. E-I Index This value can range from 1 to -1, but for a given network density and group sizes its range may be restricted and so it can be rescaled. The index is also calculated for each group and for each individual actor.
=36/(14+50) =42/(66+24)
Krackhardt's graph theoretical dimensions of hierarchy Krackhardt argues that an ‘Outree” is the archetype of hierarchy. • Krackhardt focuses on 4 dimensions: • 1) Connectedness • 2) Digraph hierarchic • 3) digraph efficiency • 4) least upper bound (what are the allowed triad types for an out-tree?)
Connectedness: The digraph is connected if the underlying graph is a component. We can measure the extent of connectedness through reachability. Where V is the number of pairs that are not reachable, and N is the number of people in the network.
Reach: 1 2 3 4 5 1 0 1 2 1 0 2 1 0 1 2 0 3 2 1 0 3 0 4 1 2 3 0 0 5 0 0 0 0 0 Graph: 1 2 3 4 5 1 0 1 0 1 0 2 1 0 1 0 0 3 0 1 0 0 0 4 1 0 0 0 0 5 0 0 0 0 0 Digraph: 1 2 3 4 5 1 0 1 0 1 0 2 0 0 1 0 0 3 0 0 0 0 0 4 0 0 0 0 0 5 0 0 0 0 0 1 4 2 5 3 How to calculate Connectedness: V = # of zeros in the upper diagonal of Reach: V = 4. C = 1 - [4/((5*4)/2)] = 1 - 4/1 = .6
Reachable: 1 2 3 4 5 1 0 1 1 1 0 2 1 0 1 1 0 3 1 1 0 1 0 4 1 1 1 0 0 5 0 0 0 0 0 Reach: 1 2 3 4 5 1 0 1 2 1 0 2 1 0 1 2 0 3 2 1 0 3 0 4 1 2 3 0 0 5 0 0 0 0 0 1 4 2 5 3 How to calculate Connectedness: This is equivalent to the density of the reachability matrix. D = SR/(N(N-1)) = 12 /(5*4) = .6
Graph Hierarchy: The extent to which people are asymmetrically reachable. Where V is the number of symmetrically reachable pairs in the network. Max(V) is the number of pairs where i can reach j or j can reach i.
1 4 2 5 3 Graph Hierarchy: An example Dreachable 1 2 3 4 5 1 0 1 2 1 0 2 0 0 1 0 0 3 0 1 0 0 0 4 0 0 0 0 0 5 0 0 0 0 0 Digraph: 1 2 3 4 5 1 0 1 0 1 0 2 0 0 1 0 0 3 0 1 0 0 0 4 0 0 0 0 0 5 0 0 0 0 0 Dreach 1 2 3 4 5 1 0 1 2 1 0 2 0 0 1 0 0 3 0 1 0 0 0 4 0 0 0 0 0 5 0 0 0 0 0 V = 1 Max(V) = 4 H = 1/4 = .25
Graph Efficiency: The extent to which there are extra lines in the graph, given the number of components. Where v is the number of excess links and max(v) is the maximum possible number of excess links
1 4 2 6 5 3 7 The minimum number of lines in a connected component is N-1 (assuming symmetry, only use the upper half of the adjacency matrix). In this example, the first component contains 4 nodes and thus the minimum required lines is 3. There are 4 lines, and thus V1= 4-3 = 1. The second component contains 3 nodes and thus minimum connectivity is = 2, there are 3 so V2 = 1. Summed over all components V=2. The maximum number of lines would occur if every node was connected to every other, and equals N(N-1)/2. For the first component Max(V1) = (6-3)=3. For the second, Max(V2) = (3-2)=1, so Max(V) = 4. Efficiency = (1- 2/4 ) = .5 Graph Efficiency: 1 2
Graph Efficiency: Steps to calculate efficiency: a) identify all components in the graph b) for each component (i) do: i) calculate Vi = S(Gi)/2 - Ni-1; ii) calculate Max(Vi) = Ni(Ni-1) - (Ni-1) c) V = S(Vi), Max(V)= S(Max(Vi) d) efficiency = 1 - V/Max(V) Substantively, this must be a function of the average density of the components in the graph.
Least Upper Boundedness: This condition looks at how many ‘roots’ there are in the tree. The LUB for any pair of actors is the closest person who can reach both of them. In a formal hierarchy, every pair should have at least one LUB. E In this case, E is the LUB for (A,D), B is the LUB for (F,G), H is the LUB for (D,C), etc. H B G C F A D
Graph Theoretic Dimensions of Informal Organizations Least Upper Boundedness: You get a violation of LUB if two people in the organization do not have an (eventual) common boss. Here, persons 4 and 7 do not have an LUB.
Distance matrix 1 2 3 4 5 6 7 8 9 1 1 1 1 2 2 2 2 1 1 1 3 1 1 4 1 5 1 6 1 1 1 2 7 1 1 8 1 9 1 Reachable matrix 1 2 3 4 5 6 7 8 9 1 1 1 1 1 1 1 2 1 1 1 3 1 1 4 1 5 1 6 1 1 1 1 7 1 1 8 1 9 1 Least Upper Boundedness: Calculate LUB by looking at reachability. (Note that I set the diagonal = 1) A violation occurs whenever a pair is not reachable by at least one common node. We can get common reachability through matrix multiplication
Reachable matrix 1 2 3 4 5 6 7 8 9 1 1 1 1 1 1 1 2 1 1 1 3 1 1 4 1 5 1 6 1 1 1 1 7 1 1 8 1 9 1 Reachable Trans 1 2 3 4 5 6 7 8 9 1 1 2 1 1 3 1 1 4 1 1 1 5 1 1 1 6 1 7 1 1 8 1 1 9 1 1 1 1 1 Graph Theoretic Dimensions of Informal Organizations Least Upper Boundedness: Calculate LUB by looking at reachability. Common Reach 1 2 3 4 5 6 7 8 9 1 1 1 1 1 1 0 0 0 1 2 1 2 1 2 2 0 0 0 1 3 1 1 2 1 1 0 0 0 2 4 1 2 1 3 2 0 0 0 1 5 1 2 1 2 3 0 0 0 1 6 0 0 0 0 0 1 1 1 1 7 0 0 0 0 0 1 2 1 2 8 0 0 0 0 0 1 1 2 1 9 1 1 2 1 1 1 2 1 5 X = (R by S) (S by R) (R by R) Any place with a zero indicates a pair that does not have a LUB. R`*R = CR
Least Upper Boundedness: Calculate LUB by looking at reachability. Where V = number of pairs that have no LUB, summed over all components, and:
Bearman, Peter S., James Moody, and Katherine Stovel. "Chains of Affection: The Structure of Adolescent Romantic and Sexual Networks1." American Journal of Sociology 110.1 (2004): 44-91.