Network Level Measures
Measure Value Row count 38.000 Column count 38.000 Link count 209.000 Density 0.145 Components of 1 node (isolates) 7 Components of 2 nodes (dyadic isolates) 0 Components of 3 or more nodes 1 Reciprocity 0.841 Characteristic path length 2.277 Clustering coefficient 0.336 Network levels (diameter) 5.000 Network fragmentation 0.339 Krackhardt connectedness 0.661 Krackhardt efficiency 0.807 Krackhardt hierarchy 0.237 Krackhardt upperboundedness 0.989 Degree centralization 0.144 Betweenness centralization 0.150 Closeness centralization 0.013 Eigenvector centralization 0.531 Reciprocal (symmetric)? No (84% of the links are reciprocal)
Node Level Measures
Measure Min Max Avg Stddev Total degree centrality 0.000 0.187 0.050 0.051 Total degree centrality [Unscaled] 0.000 56.000 14.947 15.237 In-degree centrality 0.000 0.184 0.050 0.051 In-degree centrality [Unscaled] 0.000 28.000 7.526 7.735 Out-degree centrality 0.000 0.184 0.050 0.051 Out-degree centrality [Unscaled] 0.000 28.000 7.526 7.816 Eigenvector centrality 0.000 0.662 0.159 0.166 Eigenvector centrality [Unscaled] 0.000 0.468 0.112 0.117 Eigenvector centrality per component 0.000 0.382 0.092 0.096 Closeness centrality 0.007 0.025 0.019 0.006 Closeness centrality [Unscaled] 0.000 0.001 0.001 0.000 In-Closeness centrality 0.007 0.033 0.025 0.012 In-Closeness centrality [Unscaled] 0.000 0.001 0.001 0.000 Betweenness centrality 0.000 0.165 0.019 0.031 Betweenness centrality [Unscaled] 0.000 219.736 25.251 41.296 Hub centrality 0.000 0.693 0.157 0.167 Authority centrality 0.000 0.628 0.158 0.166 Information centrality 0.000 0.049 0.026 0.016 Information centrality [Unscaled] 0.000 3.755 2.023 1.259 Clique membership count 0.000 23.000 4.526 5.734 Simmelian ties 0.000 0.541 0.129 0.129 Simmelian ties [Unscaled] 0.000 20.000 4.789 4.764 Clustering coefficient 0.000 0.688 0.336 0.233
Key Nodes
This chart shows the Location that is repeatedly top-ranked in the measures listed below. The value shown is the percentage of measures for which the Location was ranked in the top three.
Total degree centrality
The Total Degree Centrality of a node is the normalized sum of its row and column degrees. Individuals or organizations who are "in the know" are those who are linked to many others and so, by virtue of their position have access to the ideas, thoughts, beliefs of many others. Individuals who are "in the know" are identified by degree centrality in the relevant social network. Those who are ranked high on this metrics have more connections to others in the same network. The scientific name of this measure is total degree centrality and it is calculated on the agent by agent matrices.
Input network: Location x Location (size: 38, density: 0.144737)
Rank Location Value Unscaled Context* 1 airport 0.187 56.000 0.735 2 u_s 0.163 49.000 0.326 3 afghanistan 0.160 48.000 0.267 4 europe 0.133 40.000 -0.200 5 pakistan 0.130 39.000 -0.258 6 washington 0.100 30.000 -0.784 7 harbor 0.093 28.000 -0.901 8 yemen 0.090 27.000 -0.959 9 africa 0.087 26.000 -1.017 10 somalia 0.077 23.000 -1.193 * Number of standard deviations from the mean of a random network of the same size and density
Mean: 0.050 Mean in random network: 0.145 Std.dev: 0.051 Std.dev in random network: 0.057 In-degree centrality
The In Degree Centrality of a node is its normalized in-degree. For any node, e.g. an individual or a resource, the in-links are the connections that the node of interest receives from other nodes. For example, imagine an agent by knowledge matrix then the number of in-links a piece of knowledge has is the number of agents that are connected to. The scientific name of this measure is in-degree and it is calculated on the agent by agent matrices.
Input network(s): Location x Location
Rank Location Value Unscaled 1 airport 0.184 28.000 2 u_s 0.171 26.000 3 afghanistan 0.158 24.000 4 europe 0.132 20.000 5 pakistan 0.112 17.000 6 harbor 0.105 16.000 7 washington 0.099 15.000 8 yemen 0.092 14.000 9 africa 0.092 14.000 10 somalia 0.072 11.000 Out-degree centrality
For any node, e.g. an individual or a resource, the out-links are the connections that the node of interest sends to other nodes. For example, imagine an agent by knowledge matrix then the number of out-links an agent would have is the number of pieces of knowledge it is connected to. The scientific name of this measure is out-degree and it is calculated on the agent by agent matrices. Individuals or organizations who are high in most knowledge have more expertise or are associated with more types of knowledge than are others. If no sub-network connecting agents to knowledge exists, then this measure will not be calculated. The scientific name of this measure is out degree centrality and it is calculated on agent by knowledge matrices. Individuals or organizations who are high in "most resources" have more resources or are associated with more types of resources than are others. If no sub-network connecting agents to resources exists, then this measure will not be calculated. The scientific name of this measure is out degree centrality and it is calculated on agent by resource matrices.
Input network(s): Location x Location
Rank Location Value Unscaled 1 airport 0.184 28.000 2 u_s 0.178 27.000 3 afghanistan 0.158 24.000 4 pakistan 0.145 22.000 5 europe 0.132 20.000 6 washington 0.099 15.000 7 yemen 0.086 13.000 8 somalia 0.079 12.000 9 harbor 0.079 12.000 10 africa 0.079 12.000 Eigenvector centrality
Calculates the principal eigenvector of the network. A node is central to the extent that its neighbors are central. Leaders of strong cliques are individuals who or organizations who are collected to others that are themselves highly connected to each other. In other words, if you have a clique then the individual most connected to others in the clique and other cliques, is the leader of the clique. Individuals or organizations who are connected to many otherwise isolated individuals or organizations will have a much lower score in this measure then those that are connected to groups that have many connections themselves. The scientific name of this measure is eigenvector centrality and it is calculated on agent by agent matrices.
Input network: Location x Location (size: 38, density: 0.144737)
Rank Location Value Unscaled Context* 1 u_s 0.662 0.468 0.650 2 airport 0.517 0.366 0.121 3 washington 0.428 0.303 -0.203 4 harbor 0.422 0.299 -0.225 5 afghanistan 0.399 0.282 -0.312 6 europe 0.377 0.266 -0.392 7 pakistan 0.368 0.260 -0.423 8 new_york 0.268 0.189 -0.789 9 yemen 0.254 0.179 -0.841 10 north_america 0.228 0.161 -0.935 * Number of standard deviations from the mean of a random network of the same size and density
Mean: 0.159 Mean in random network: 0.484 Std.dev: 0.166 Std.dev in random network: 0.274 Eigenvector centrality per component
Calculates the principal eigenvector of the network. A node is central to the extent that its neighbors are central. Each component is extracted as a separate network, Eigenvector Centrality is computed on it and scaled according to the component size. The scores are then combined into a single result vector.
Input network(s): Location x Location
Rank Location Value 1 u_s 0.382 2 airport 0.298 3 washington 0.247 4 harbor 0.244 5 afghanistan 0.230 6 europe 0.217 7 pakistan 0.212 8 new_york 0.155 9 yemen 0.146 10 north_america 0.131 Closeness centrality
The average closeness of a node to the other nodes in a network (also called out-closeness). Loosely, Closeness is the inverse of the average distance in the network from the node to all other nodes.
Input network: Location x Location (size: 38, density: 0.144737)
Rank Location Value Unscaled Context* 1 nairobi 0.025 0.001 -6.236 2 manhattan 0.023 0.001 -6.269 3 kenya 0.023 0.001 -6.269 4 dar_es_salaam 0.023 0.001 -6.271 5 airport 0.022 0.001 -6.296 6 egypt 0.022 0.001 -6.297 7 france 0.022 0.001 -6.297 8 europe 0.022 0.001 -6.297 9 africa 0.022 0.001 -6.297 10 gulf 0.022 0.001 -6.297 * Number of standard deviations from the mean of a random network of the same size and density
Mean: 0.019 Mean in random network: 0.397 Std.dev: 0.006 Std.dev in random network: 0.060 In-Closeness centrality
The average closeness of a node from the other nodes in a network. Loosely, Closeness is the inverse of the average distance in the network to the node and from all other nodes.
Input network(s): Location x Location
Rank Location Value Unscaled 1 airport 0.033 0.001 2 africa 0.033 0.001 3 egypt 0.033 0.001 4 gulf 0.033 0.001 5 france 0.033 0.001 6 europe 0.033 0.001 7 tanzania 0.033 0.001 8 asia 0.033 0.001 9 afghanistan 0.033 0.001 10 saudi_arabia 0.033 0.001 Betweenness centrality
The Betweenness Centrality of node v in a network is defined as: across all node pairs that have a shortest path containing v, the percentage that pass through v. Individuals or organizations that are potentially influential are positioned to broker connections between groups and to bring to bear the influence of one group on another or serve as a gatekeeper between groups. This agent occurs on many of the shortest paths between other agents. The scientific name of this measure is betweenness centrality and it is calculated on agent by agent matrices.
Input network: Location x Location (size: 38, density: 0.144737)
Rank Location Value Unscaled Context* 1 airport 0.165 219.736 4.032 2 egypt 0.079 104.903 1.169 3 africa 0.061 81.164 0.577 4 new_york 0.055 73.505 0.386 5 europe 0.047 62.050 0.101 6 yemen 0.043 57.016 -0.025 7 tanzania 0.042 55.938 -0.051 8 gulf 0.027 35.811 -0.553 9 france 0.025 33.079 -0.621 10 farm 0.023 30.888 -0.676 * Number of standard deviations from the mean of a random network of the same size and density
Mean: 0.019 Mean in random network: 0.044 Std.dev: 0.031 Std.dev in random network: 0.030 Hub centrality
A node is hub-central to the extent that its out-links are to nodes that have many in-links. Individuals or organizations that act as hubs are sending information to a wide range of others each of whom has many others reporting to them. Technically, an agent is hub-central if its out-links are to agents that have many other agents sending links to them. The scientific name of this measure is hub centrality and it is calculated on agent by agent matrices.
Input network(s): Location x Location
Rank Location Value 1 u_s 0.693 2 airport 0.544 3 pakistan 0.414 4 afghanistan 0.411 5 washington 0.408 6 europe 0.369 7 harbor 0.347 8 north_america 0.234 9 yemen 0.226 10 asia 0.220 Authority centrality
A node is authority-central to the extent that its in-links are from nodes that have many out-links. Individuals or organizations that act as authorities are receiving information from a wide range of others each of whom sends information to a large number of others. Technically, an agent is authority-central if its in-links are from agents that have are sending links to many others. The scientific name of this measure is authority centrality and it is calculated on agent by agent matrices.
Input network(s): Location x Location
Rank Location Value 1 u_s 0.628 2 airport 0.553 3 harbor 0.458 4 afghanistan 0.420 5 washington 0.412 6 europe 0.374 7 pakistan 0.288 8 new_york 0.265 9 yemen 0.256 10 africa 0.247 Information centrality
Calculate the Stephenson and Zelen information centrality measure for each node.
Input network(s): Location x Location
Rank Location Value Unscaled 1 airport 0.049 3.755 2 afghanistan 0.048 3.660 3 u_s 0.048 3.652 4 pakistan 0.047 3.627 5 europe 0.046 3.561 6 washington 0.043 3.292 7 africa 0.042 3.216 8 yemen 0.041 3.186 9 somalia 0.041 3.139 10 egypt 0.040 3.112 Clique membership count
The number of distinct cliques to which each node belongs. Individuals or organizations who are high in number of cliques are those that belong to a large number of distinct cliques. A clique is defined as a group of three or more actors that have many connections to each other and relatively fewer connections to those in other groups. The scientific name of this measure is clique count and it is calculated on the agent by agent matrices.
Input network(s): Location x Location
Rank Location Value 1 airport 23.000 2 afghanistan 18.000 3 europe 17.000 4 pakistan 16.000 5 africa 12.000 6 egypt 11.000 7 u_s 7.000 8 washington 7.000 9 yemen 7.000 10 france 7.000 Simmelian ties
The normalized number of Simmelian ties of each node.
Input network(s): Location x Location
Rank Location Value Unscaled 1 airport 0.541 20.000 2 afghanistan 0.378 14.000 3 europe 0.378 14.000 4 pakistan 0.297 11.000 5 egypt 0.297 11.000 6 africa 0.270 10.000 7 somalia 0.216 8.000 8 gulf 0.216 8.000 9 france 0.216 8.000 10 u_s 0.189 7.000 Clustering coefficient
Measures the degree of clustering in a network by averaging the clustering coefficient of each node, which is defined as the density of the node's ego network.
Input network(s): Location x Location
Rank Location Value 1 spain 0.688 2 london 0.667 3 britain 0.625 4 saudi_arabia 0.611 5 farm 0.611 6 bali 0.563 7 north_america 0.556 8 gulf 0.530 9 asia 0.500 10 manhattan 0.500
Key Nodes Table
This shows the top scoring nodes side-by-side for selected measures.
Rank Betweenness centrality Closeness centrality Eigenvector centrality Eigenvector centrality per component In-degree centrality In-Closeness centrality Out-degree centrality Total degree centrality 1 airport nairobi u_s u_s airport airport airport airport 2 egypt manhattan airport airport u_s africa u_s u_s 3 africa kenya washington washington afghanistan egypt afghanistan afghanistan 4 new_york dar_es_salaam harbor harbor europe gulf pakistan europe 5 europe airport afghanistan afghanistan pakistan france europe pakistan 6 yemen egypt europe europe harbor europe washington washington 7 tanzania france pakistan pakistan washington tanzania yemen harbor 8 gulf europe new_york new_york yemen asia somalia yemen 9 france africa yemen yemen africa afghanistan harbor africa 10 farm gulf north_america north_america somalia saudi_arabia africa somalia