Professional Documents
Culture Documents
INFO 532 Chapt Five-2019-Use
INFO 532 Chapt Five-2019-Use
Instructor:
DR. LAWRENCE ABRAHAM GOJEH
(ASSOCIATE PROFESSOR)
Chapter Five: Social Network Analysis
• Sociology of Science.
• Visualization
– Citation Maps,
– Nodes,
– Graphs, and
– Networks.
• Trends and New Technologies in SNA
– E-prints,
– creative commons,
– collaboratories,
– Webmetrics, and
– E-metrics.
– The Internet
Chapter Objectives
• Discuss Social Network Analysis
• Explain the Sociology of Science
• Discuss Visualization (Citation Maps, Nodes, Graphs, &
Networks) and the Citation Network.
• Explain the Trends and New Technologies in Social Network
Analysis (eprints, creative commons, collaboratories,
webmetrics and internet).
Social Network
• Social network was coined by John Barnes in 1954.
• It is defined as a social structure made up of individuals or
organizations called "nodes", which are tied (connected) by one or
more specific types of actors/interdependency, such as:
friendship,
kinship,
common interest,
financial exchange,
organizations,
institutions,
communities,
groups,
families etc.
• Actors can be individual people, objects or events as far as certain
relations hold them all together.
Social Network (cont.)
• Actors can also be aggregate units such as:
• 1. The very idea of the social network approach is that relations or
interactions between actors are the building blocks or the key factors
that sustain and define the network.
• 2. Typically interactions between actors result from:
– exchange of resources,
• either material or informational, such as:
– goods,
– money,
– information,
– services,
– social or emotional support,
Social Network (cont.)
– trust,
– influence etc.
• 3. Each kind of resource exchange is considered a social network
relation and actors maintaining the relation are said to maintain a tie.
• The strength of a tie may range from:
– weak to strong depending on the quantity, or
– quality and frequency of the exchanges between actors.
• Patterns of who is tied to whom reveal the structure of the
underlying network:
– they show how resources flow among actors; and
– how actors are interconnected in the network.
•
Social Network (cont.)
• Examples of social networks
• A few very well known examples of social network analysis are:
– exchange of job information among acquaintances, where weak
ties are quite operationally strong for the diffusion of such
information.
– the urban poor in isolated Black ghettoes lack connections with
sources of work.
– the dependency of social capital on ‘structural holes’ (which are
particular kinds of network positioning in which a focal actor is
connected to other actors; which themselves are not connected
with one another); which is not a direct attribute of actors but
rather of their ability to sustain flexible configurations within a
network.
Social Network (cont.)
– now, computer networks in general and in particular the Internet
are clearly social networks .
• In these social networks, actors may be:
– human, such as:
• users,
• communicants,
• information producers and consumers,
• citizens,
• public or market organizations etc., or
– non-human, such as:
• computer machines,
• information databases,
Social Network (cont.)
• (hyper-) documents,
• multimedia resources, etc.
• Relations among the human Internet actors refer to:
– informative and communicative uses,
– access,
– provision,
– procurement,
– commerce,
– work,
– education etc.
• Although human actors are always beneath the non-human ones,
typical relations among the latter consist of:
– information (data) flows,
Social Network (cont.)
– traffic,
– exchanges of e-mails and postings in web pages,
– links,
– connections,
– network topologies etc.
Social network analysis
• Social Network Analysis (SNA) is a body of methods developed for
analyzing social networks.
• It has its origins in sociology and mathematics “graph theory” but it is
now being used across a wide range of other disciplines.
• Social network analysis: refers to a research approach developed
primarily in sociology, social psychology and communication science.
• It focuses on patterns of relations among people, and among groups
such as organizations and states.
• As computer networks such as the Web connect people and
organizations, they can host social networks.
• Social network analysis approach address commonly five basic issues:
– cohesion,
– structural equivalence,
Social Network Analysis (cont.)
– prominence,
– Range, and
– brokerage.
• 1. Cohesion (which is a relational property):
– refers to the grouping of actors because of the strength of their
relationships with one another.
– Cohesive groups of actors form clusters or cliques depending on
whether they are highly or fully interconnected, respectively.
– A measure of cohesion is the network density, which is calculated
as the ratio of the number of actually occurring links (relations) to
the number of all possible links.
– A relevant concept is that of centralization, measuring the extent
to which a set of actors are organized around a central one
Social Network Analysis (cont.)
• 2. Structural equivalence (which is a positional property):
– Identifies actors who have similar patterns of relations with
others, even if such actors may not have direct relations with each
other.
– An actor’s pattern of relations constitutes a role.
– Thus, actors playing similar roles occupy similar (or equivalent)
structural or status positions.
– A technique for assessing structural equivalence is known as block
modelling.
– In this technique, one first calculates correlations between all
pairs of actors and then reorders the actors into sets on the basis
of the correlation values in such a way that pairs of actors that are
highly correlated (and, therefore, most structurally equivalent)
should appear together in the same group (or block).
Social Network Analysis (cont.)
• 3. Prominence:
– Reflects the hierarchical status of an actor.
– It can be measured by assessing the centrality of an actor in a
network, which is derived by measuring the actor’s connections in
the network, i.e., its degree. (This differs from the previously
mentioned centralization, which measures the configuration of
the network as a whole.)
– The actor with the highest degree (i.e., the most relationships
with other actors) is the most central.
– Another measure of an actor’s prominence is global centrality (or
closeness) and it is derived by measuring the distance between
this actor and any other actor, which is defined as the number of
connections in the shortest path between the actors.
Social Network Analysis (cont.)
– The actor with the lowest sum of distances to all other
actors is the most globally central actor.
• 4. Range:
– Refers to a combination of network size and
heterogeneity that jointly increases the ability of actors
to have access to a variety of resources (social support,
social capital).
– The biggest a network is, the more information an actor
will have access to and the more complex the accessed
information will be.
– Moreover, heterogeneous networks may provide a
greater variety of social support.
Social Network Analysis (cont.)
• 5. Brokerage:
– An activity that puts interested actors in touch with one
another so that they might strike a deal.
– It involves at least three actors with the intermediary relegating
transactions between the others.
– In this case, there are five ideal-typical roles of brokers:
• liaison,
• representative,
• gatekeeper,
• Itinerant, and
• coordinator.
Social Network Analysis (cont.)
– Brokerage can be measured by betweenness, the extent
to which an actor is located between others in the
network.
– Where opportunities of brokerage exist but have not been
exploited yet, there is a ‘structural hole,’ in the
terminology coined.
– However, brokerage indicates not only opportunities to
further exploit the network potentialities but also points of
possible resistance by those currently playing the
gatekeeper’s role who have the power to control and filter
imported or exported information.
Benefits of SNA
1. Identifies the individuals, teams, and units who play central roles.
2. Determine information breakdowns bottlenecks, structural holes, as
well as isolated individuals, teams, and units.
3. Make out opportunities to accelerate knowledge flows across
functional and organizational boundaries.
4. Strengthen the efficiency and effectiveness of existing, formal
communication channels.
5. Raise awareness of and reflection on the importance of informal
networks and ways to enhance their organizational performance.
6. Leverage peer support.
7. Improve innovation and learning.
8. Refine strategies.
Problems with Network Analysis
i. The field of network analysis is normally criticized for being too
much methodological and too little theoretical.
ii. Critics say that there are few truly network theories of substantive
phenomena and critics also say 'that's not really a network theory'.
iii. Theories say that psychological phenomena tend to have a lot of
psychological content.
iv. Theories that account for sociological phenomena have sociological
independent variables.
v. Only theories that explain network phenomena tend to have a lot of
network content.
vi. A real problem with network analysis in the past has been the
inability to test hypotheses statistically, because the data are by
their very nature auto correlated, violating assumptions of
independence (random sampling) built-in to most classical statistical
tests.
Problems with Network…(cont.)
vii. With the advent of permutation tests, this is much less
of problem now.
viii. A continuing problem is the lack of sufficient computing
resources to handle large databases.
ix. It is often a problem to bound a social network.
x. If we are looking at needle-sharing among drug users, we
can artificially bound the network at some arbitrary
boundary, such as city or neighborhood, but this distorts
the data.
xi. Yet we cannot let the network get too large because we
cannot process the data….
Sociology of science
• Sociology of science is concerned with the social
explanation of the development of scientific knowledge, in
which factors such as the organization and social interests
of scientists are used to explain scientific changes.
• Barabasi et al. studied the collaboration networks in
mathematics and neuro-science of an 8-year period (1991-
1998) to understand the topological and dynamical laws
governing complex networks.
• They viewed the collaboration network as a prototype of
evolving works, as it expands by the addition of new nodes
(authors) and new links (papers co-authored).
Sociology of science (cont.)
• The results obtained indicated that the network is scale-
free and that its evolution is governed by preferential
attachment.
• Moreover, authors concluded that most quantities used to
characterize the network are time dependent.
• For example, the average degree (network’s
interconnectedness) increases in time.
• Furthermore, the study showed that the node separation
decreases over time, however this trend is believed to be
offered by incomplete database and it can be opposite in
the full system.
Visualizations
• Visualizations have played an important role in generating new
insights in social network analysis.
• Visualizing social networks is more than simply creating intriguing
pictures, it is about generating learning situations:
– “images of social networks have provided investigators with new
insights about network structure and have helped them
communicate those insights to others”.
• Such network images are created mainly in two ways:
– the first one is by drawing graphs made up of nodes and
connecting lines.
– The second way is to devise a matrix where rows and columns
stand for people and the numbers in each cell stand for the social
connections between the people.
• In practice, however, most social network applications have focused
on the graph representation.
Visualizations (cont.)
Figure showing a central section of a word co-occurrence network of words drawn from
the titles of OER-supported journal articles.
Visualizations (cont.)
• The image is a central section of a word co-
occurrence network of words drawn from the titles
of OER-supported journal articles.
• Some words have been truncated.
• Nodes are sized by the number of times the word is
used and edges are sized and coloured based on the
number of publications in which they co-occur.
• Weaker edges have been removed for clarity.
• Nodes coloured blue are those that have been used
10 or more times in the publication set.
Visualizations (cont.)