Social Network Analysis
Christopher McCarty
University of Florida
Books
• Social Network Analysis: A Handbook by John Scott, London: Sage
(2000).
• Social Network Analysis: Methods and Applications. Stanley
Wasserman and Katherine Faust. Cambridge: Cambridge University
Press (1994).
• Social Networks and Health: Models, Methods and Applications by
Tom Valente (2010) Oxford: Oxford University Press.
• Understanding Social Networks: Theories, Concepts and Findings
(2011) Charles Kadushin’s Oxford: Oxford University Pres.
• The Development of Social Network Analysis: A Study in the
Sociology of Science Linton C. Freeman, Empirical Press, Vancouver,
BC (2004).
• The SAGE Handbook of Social Network Analysis (2011) Eds. John
Scott and Peter Carrington. London: Sage Publications
Web Sites
• www.insna.org -- International Network for
Social Network Analysis
• http://faculty.ucr.edu/~hanneman/nettext/ -Tutorial for UCINET/Netdraw
• http://www.redes-sociales.net/ (Spanish
social network listserv)
Software
•
Ucinet (Whole networks)
–
•
E-net (Batch processing of ego networks)
–
•
http://www.casos.cs.cmu.edu/projects/ora/
Visone (Whole and Personal network analysis)
–
•
http://ciknow.northwestern.edu/
ORA (Whole network analysis)
–
•
http://stat.gamma.rug.nl/siena.html
C-IKNOW(Online network data collection)
–
•
(http://sourceforge.net/projects/egonet/)
Vennmaker (Personal networks)
Siena (Network modeling, longitudinal)
–
•
(http://vlado.fmf.uni-lj.si/pub/networks/pajek/)
Egonet (Personal networks)
–
•
•
(www.analytictech.com)
Pajek (Whole networks, large networks)
–
•
(www.analytictech.com) ($40 for students, $150 for faculty)
http://visone.info/
Statnet
–
http://csde.washington.edu/statnet/
Journals
• Social Networks
• Connections
• Journal of Social Structure
• American Journal of Sociology, Social Science and
Medicine, Journal of Mathematical Sociology,
Organization Science, Social Forces, Gerontologist
Sunbelt Conference
•
•
•
•
•
•
•
•
2001 – Budapest, Hungary – April 25-29
2002 – New Orleans, LA – March 1-9
2003 – Cancun, Mexico – March 1-9
2004 – Portoroz, Slovenia – May12-16
2005 – Redonda Beach, CA – February 16-20
2006 – Vancouver, Canada – April 25-30
2007 – Corfu, Greece – May 1-6
2008 – St. Pete Beach, FL – January 22-27
•
•
•
•
•
2009 – San Diego, CA – March 10-15
2010 – Riva gel Garda, Italy – June 29-July 4
2011 – St. Pete Beach, FL – February 8-13
2012 – Redonda Beach, CA – March 12-18
2013 – Hamburg, Germany – May 21-26
Social Network Analysis is the study of the
pattern of interaction between actors
Examples of actors and their networks
•
•
•
•
•
•
•
•
•
Children in a preschool
Employees in an office
Customers of AT&T mobile phone service
NGOs working in the Amazon
Companies in the Fortune 500
Countries in the European Union
Baboons in a troupe
Organisms in the Chesapeake Bay
Web sites around the world
Is SNA just a set of tools or is it a
theoretical approach?
See: http://www.insna.org/PDF/Sunbelt/3_KeynotePDF.pdf
• Social Capital, Structural Holes, Simmelian ties
• Strong and weak ties
• Small world
• Scale-free networks
• Network diffusion
Social Capital
• “…the ability of actors to secure benefits by virtue
of membership in social networks or other social
structures” (Portes 1998)
• Alejandro Portes (1998) SOCIAL CAPITAL: Its
Origins and Applications in Modern Sociology,
Annual Review of Sociology 1998. 24:1–24
• Ron Burt (2004) Structural Holes and Good Ideas,
American Journal of Sociology 110: 349–399
David Krackhardt (1999) The ties that torture:
Simmelian tie analysis in organizations, Research in the
Sociology of Organizations 16: 183-210
Structure matters
(but is not always enough)
• In some contexts structure is a necessary, but not
sufficient, condition for social capital
• The most beneficial structural position may depend on
the topic
– Job seeking
– Social support
• Social network evaluation and intervention does not
always mean you should connect the dots
– Facebook model is to suggest connections
– Sometimes there are reasons for not connecting
Strong and weak ties
• The most beneficial tie may not always be the
strong ones
• Strong ties are often connected to each other and
are therefore sources of redundant information
• Mark Granovetter (1973) The strength of weak
ties American Journal of Sociology 78-1361-1381.
Small world phenomenon
• Being linked, seemingly by chance, through someone
via a friend or acquaintance
• Stanley Milgram (1967)The Small World Psychology Today
2:60–67.
• Peter D. Killworth, H. Russell Bernard and Christopher
McCarty (1984) Measuring Patterns of Acquaintance Current
Anthropology 25:381-397
• Duncan Watts and Steven Strogatz (1998) Collective dynamics
of 'small-world' networks Nature 393 (6684): 409–10
Scale Free Networks
• Scale free refers to the power law structure of
networks as the number of actors increases
• Networks tend to form hubs
• Entry of physicists into SNA
• Albert-László Barabási and Réka Albert (1999)
Emergence of scaling in random network.
Science, 286:509-512.
Network Diffusion
• Network structures can often aid or impede the flow of information and
the adoption of innovations
• Diffusion of innovation is the basis for peer to peer network interventions
• Coleman, James, Elihu Katz, and Herbert Menzel. 1957. The diffusion of
innovation among physicians. Sociometry. 20:253-270.
• Valente, Thomas W. 1996 “Social network thresholds in the diffusion of
innovations” Social Networks 18:69-89.
• Klovdahl, A. S. (1985). Social networks and the spread of infectious
diseases: The AIDS example. Social Science Medicine, 21(11), 1203-1216.
Two kinds of Social Network Analysis
Whole (Complete, Sociocentric)
Network Analysis
• Focus on interaction within a
group
• Boundary defines social space
• Collect data from members of a
group about their ties to other
group members
Personal (Egocentric) Network
Analysis
• Focus on effects of network on
individual attitudes, behaviors and
conditions
• Use attributes of personal network
to represent social context
• Collect data from respondent (ego)
about interactions with network
members (alters)
Sociocentric Network Data From Graduate
Anthropology Course
• Three network components
• Beth is most degree central
• Amber is most between central
• Thomas and Kent are structurally equivalent
• Removal of David maximizes network
fragmentation
Boundary definition
• Boundaries can be defined:
– Geographically (a village)
– Socially (an organization)
– Through connections (snowball)
• The idea is that actors within the boundary are in
some way affected by their social position
• This excludes the effects from those outside the
boundary
Missing data
• In whole networks responses by others about nonrespondents can capture structure
• 70% will in many cases be enough
• Gueorgi Kossinets (2006) Effects of missing data in
social networks. Social Networks 28: 247–268.
• Costenbader, E. & Valente, T. W. (2004). The stability of
centrality when networks are sampled. Social
Networks.
Two kinds of Social Network Analysis
Whole (Complete, Sociocentric)
Network Analysis
• Focus on interaction within a
group
• Boundary defines social space
• Collect data from members of a
group about their ties to other
group members
Personal (Egocentric) Network
Analysis
• Focus on effects of network on
individual attitudes, behaviors and
conditions
• Use attributes of personal
network to represent social
context
• Collect data from respondent
(ego) about interactions with
network members (alters)
Personal network interview
•
Identify a population
•
Select a sample of respondents
•
Ask questions about respondent
•
Elicit network members
•
Ask questions about each network member
•
Ask respondent to evaluate ties between network members
Personal Network Composition
Alter summary file
Name
Closeness
Relation
Sex
Age
Race
Where Live
Year_Met
Joydip_K
5
14
1
25
1
1
1994
Shikha_K
4
12
0
34
1
1
2001
Candice_A
5
2
0
24
3
2
1990
Brian_N
2
3
1
23
3
2
2001
Barbara_A
3
3
0
42
3
1
1991
Matthew_A
2
3
1
20
3
2
1991
Kavita_G
2
3
0
22
1
3
1991
Ketki_G
3
3
0
54
1
1
1991
Kiran_G
1
3
1
23
1
1
1991
Kristin_K
4
2
0
24
3
1
1986
Keith_K
2
3
1
26
3
1
1995
Gail_C
4
3
0
33
3
1
1992
Allison_C
3
3
0
19
3
1
1992
Vicki_K
1
3
0
34
3
1
2002
Neha_G
4
2
0
24
1
2
1990
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Personal network composition variables
• Proportion of personal network that are women
• Average age of network alters
• Proportion of strong ties
• Average number of years knowing alters
Personal Network Structure
Alter adjacency matrix
Joydip_K
Shikha_K
Candice_A
Brian_N
Barbara_A
Matthew_A
Kavita_G
Ketki_G
.
.
.
Joydip_K
1
1
1
1
0
0
0
0
.
.
.
Shikha_K
1
1
0
0
0
0
0
0
.
.
.
Candice_A
1
0
1
1
1
1
1
1
.
.
.
Brian_N
1
0
1
1
1
1
1
1
.
.
.
Barbara_A
0
0
1
1
1
1
0
0
.
.
.
Matthew_A
0
0
1
1
1
1
1
1
.
.
.
Kavita_G
0
0
1
1
0
1
1
1
.
.
.
Ketki_G
0
0
1
1
0
1
1
1
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Personal network structural variables
• Average degree centrality (density)
• Average closeness centrality
• Average betweenness centrality
• Core/periphery
• Number of components
• Number of isolates
Boundary definition for personal
networks
• Facebook
– West Africa and Asia
• Time
– First grade teacher
• Require mutual recognition
– Book author
• Living
– Dead relative (Genogram)
– Jesus
Two categories of data collection
• One mode data
– Actors by actors
• Examples of one mode data collection
– Survey; E-mail; Telephone calls; Observation of interaction
• Two mode data
– Actors by events
• Examples of two mode data collection
– Attendance at parties, meetings, funerals; Purchase of items;
Reading particular authors
Kinds of data
Whole
Complete
Sociocentric
One mode
Two mode
Personal
Egocentric
Example 1
One mode - Whole network
University – Water Management
District interaction
• Objective: Understand structure of interaction
between academic and applied scientists
• Procedure 1: Bound universities by those
published in journals in St. Johns Water
Management District library in 2008
• Procedure 2: Bound WMD by employee e-mails
on web sites
• Procedure 3: Web survey with letter and $1
incentive to all 705 actors
• Response: 332 completed surveys
Visualization of University-WMD
network
Example 2
One mode - Personal network
Acculturation study
• Objective: Test social network compositional and
structural variables as proxies for acculturation
• Procedure 1: Interviewed 535 migrants in
Barcelona and New York City
• Procedure 2: Each respondent listed 45 network
alters
• Procedure 3: Respondents provided twelve pieces
of information about each alter
• Procedure 4: Respondents evaluated all 990
unique alter-alter ties
Visualization of the networks of two sisters
Label = Country of origin, Size = Closeness, Color = Skin color, Shape = Smoking Status
• Mercedes is a 19-year-old second generation
Gambian woman in Barcelona
• Laura is a 22-year-old second generation
Gambian woman in Barcelona
• She is Muslim and lives with her parents and 8
brothers and sisters
• She is Muslim and lives with her parents and 8
brothers and sisters
• She goes to school, works and stays home
caring for her siblings. She does not smoke or
drink.
• She works, but does not like to stay home. She
smokes and drinks and goes to parties on
weekends.
Example 3
Two mode - Whole network
Southern women
• Objective: To understand the network
structure of the debutante network in a
Southern town in the 1940s
• Procedure: Observe which of the 14 annual
balls each of the 18 women attended
Two mode data matrix
Visualization of two mode data
Example 4
Two mode - personal network
Relation categories in Thailand
• Objective: Discover mutually exclusive and
exhaustive categories in a language for how
people know each other to be used on a
network scale-up survey instrument
Procedure 1: Twenty one respondents freelist in Thai
ways that people know each other
Procedure 2: Twenty one respondents list 30
people they know and apply 26 most frequently
occurring categories
colleague
ปอนด์
นุช
เพ็ ญ
พี่ยู
หมี
อาจารย์นิ
อาจารย์อมรา
พี่นด
ิ
มด
พี่จม
ุ๋
พี่ภา
พี่จวิ่
น ้าช่วย
อาจารย์มานพ
วรา
โจ ้
สุทป
ี
พี่ยาว
พี่เกด
ส ้ม
เกด
พี่เหว่า
เอ๋ย
ปิ ง
เล็ก
น ้าม่อน
ป้ าขวด
นุ ้ย
household
0
1
0
0
1
1
1
1
1
0
1
1
1
0
0
0
1
0
0
0
1
1
0
0
0
0
0
0
neighbour
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
sport club/ park
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
meeting
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
1
1
1
0
1
1
1
1
0
1
0
0
0
1
0
0
0
0
1
1
0
0
0
0
relatives temple/ church
same community
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Affiliation from all respondents
Graph of relationship between
knowing categories