Using Social Network Analysis for Social Media in Social Research

USING SOCIAL
NETWORK ANALYSIS
FOR SOCIAL MEDIA IN
SOCIAL RESEARCH
COMBINING QUANTITATIVE, QUALITATIVE,
AND VISUAL RESEARCH METHODS
Social Media in Social Research
May 16, 2014
Dhiraj Murthy
@dhirajmurthy
Network Theory
• Socioeconomic activity is intrinsically related to
social structures which has led to Sociology’s
interest in social capital
• The position of an actor in the network
influences their ability to access resources such
as goods, capital and information
• Well connected actors have better access to
information and other resources and therefore
have more opportunities than less well
connected actors.
An actor can be anything
from animals to people
or organisations
Network Theory History
(based on Freeman, 2000)
• 17th century: Spinoza developed first model
• 1937: J.L. Moreno introduced sociometry; he also
invented the sociogram
Social maps!
• 1949: A. Rapaport developed a probability based
model of information flow
• 70s: - New features in graph theory – more general
structural models
– Better computer power – analysis of complex
relational data sets
Social Networks
• A social network is a social structure of people,
related (directly or indirectly) to each other through
a common relation or interest. Networks can have
few or many actors (nodes), and one or more kinds
of relations (edges)” (Hannemann, 2001)
(Source: Freeman, 2000)
What are social relations?
A social relation is anything that links two
actors. Examples include:
Kinship
Friendship
Love
Exchange
Coauthorship
Texting with
Sharing photos with
Co-membership
Talking with
Hate
Trust
Fighting
Tweeting @
E-mailing with
We can examine networks across
multiple levels:
1) Ego-network
- Have data on a respondent (ego) and the people
they are connected to (alters). Example:
personal Facebook networks
2) Partial network
- Something less than full account of connections
among all pairs of actors in the relevant
population
- Example: Most Twitter networks
We can examine networks across
multiple levels:
3) Complete or “Global” data
- Data on all actors within a particular (relevant)
boundary
- Never exactly complete (due to missing data),
but boundaries are set
-Example: Corporate or government networks
What is Social Network
Analysis (SNA)?
• Sociology has been interested in the networked
structure of social relations since its early inception.
• Networks are created by actor-actor relationships
• SNA examines relationships between actors and the
social structures in which they interact to gain a better
understanding of actors’ roles within the social system.
• SNA is the study of how the ties in networks create
opportunities and restrictions which affect both
individual and institutional access to resources such as
information, finance, goods and power.
SNA often involves:
•
•
•
•
•
•
•
Mixed methods
Survey questionnaires
Participant observation
Clear variables
Cluster analysis
Conventional statistics (regression etc.)
Multilevel modelling
Qualitative and visual methods are
important to SNA!
Graphs - Sociograms
(based on Hanneman, 2001)
• Labeled circles represent actors
• Line segments represent ties
• Graph may represent one or more types of
relations
• Each tie can be directed or show cooccurrence
– Arrows represent directed ties
Representation of Social Networks
• Matrices
Ashton
Mila Kunis
Perry
Bieber
Ashton
-
1
0
0
Mila Kunis
1
-
1
1
0
-
1
0
1
-
Perry
Bieber
• Graphs
0
Graphical analysis of celebrities as
a social network?
will.i.am
Harry
Styles
Everyone
looks to T.
Jones! And
Kylie’s role
is…
Tom Jones
Kylie
Kendall
Jenner
Ricky
SNA and Social Media
• How can SNA be applied to different social
media platforms to gain sociological insights?
#pakistan tweets during the
Pakistan Floods of 2010
1 / 2 Mode – Social Engagements
And comments like ‘cool colors’
would create a network
Case Study: Photo Sharing
Flickr users who commented on user X’s photos (more than 4 times)
Human Super-Connectors
The circled user is a social ‘bridge’ who connects a cluster to the network
Who is Talking about #X on Twitter?
From my article for the Runnymede Trust
Finding Friendship Clusters in Facebook?
Building Fan Communities on YouTube
HOW CAN I START?
NodeXL (http://nodexl.codeplex.com)
Gephi/Netvizz
(Netvizz creates .gdf files for Gephi)
Why bother with Social Network Analysis?
• Operationalises (some versions of) the
concept of ‘social structure’.
• Network positions enable/constrain actors
and actions
• Exploring diffusion.
• SNA can help us decipher very large/ complex
data
• Non-technical students can learn SNA and
apply it to online community interactions
using NodeXL with minimal support.
• Visualisations are INHERENT to SNA