Transcript Slide 1
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) – ( www.analytictech.com
) ($40 for students, $150 for faculty) E-net (Batch processing of ego networks) – ( www.analytictech.com
) Pajek (Whole networks, large networks) – ( http://vlado.fmf.uni-lj.si/pub/networks/pajek/ ) Egonet (Personal networks) – ( http://sourceforge.net/projects/egonet/ ) Vennmaker (Personal networks) Siena (Network modeling, longitudinal) – http://stat.gamma.rug.nl/siena.html
C-IKNOW(Online network data collection) – http://ciknow.northwestern.edu/ ORA (Whole network analysis) – http://www.casos.cs.cmu.edu/projects/ora/ Visone (Whole and Personal network analysis) – http://visone.info/ Statnet – http://csde.washington.edu/statnet/
• Social Networks
Journals
• 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
Personal (Egocentric) Network Analysis •
Focus on interaction within a group
• Focus on effects of network on individual attitudes, behaviors and conditions •
Boundary defines social space
•
Collect data from members of a group about their ties to other group members
• • 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 non respondents 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
Name Joydip_K Shikha_K Candice_A Brian_N Barbara_A Matthew_A Kavita_G Ketki_G Kiran_G Kristin_K Keith_K Gail_C Allison_C Vicki_K Neha_G .
.
.
Personal Network Composition
Alter summary file
Closeness 5 4 5 2 3 1 3 2 2 4 .
.
.
3 1 4 2 4 Relation 14 12 2 3 3 3 3 3 3 2 .
.
.
3 3 2 3 3 Sex 1 0 0 1 0 1 0 1 0 0 .
.
.
0 0 0 1 0 Age 25 34 24 23 42 20 22 54 23 24 .
.
.
24 26 33 19 34 Race 1 1 3 3 1 1 3 3 1 1 .
.
.
3 3 3 3 3 Where Live 1 1 2 2 1 1 1 2 3 2 .
.
.
1 1 1 1 1 1986 1995 1992 1992 2002 1990 .
.
.
Year_Met 1994 2001 1990 2001 1991 1991 1991 1991 1991
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 Shikha_K 1 1 Candice_A 1 Brian_N Barbara_A 1 0 Matthew_A Kavita_G 0 0 Ketki_G 0 1 1 1 1 0 0 0 1 1 0 1 1 0 1 1 0 1 1 0 1 1 0 1 1 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 0 0 1 1 1 1 0 1 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
One mode Two mode
Kinds of data
Whole Complete Sociocentric 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 • 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.
• 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 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 ปอนด์ นุช เพ็ญ พี่ยู colleague หมี อาจารย์นิ อาจารย์อมรา พี่นิด มด พี่จุ๋ม พี่ภา พี่จิ่ว น ้าช่วย อาจารย์มานพ วรา โจ ้ สุทีป พี่ยาว พี่เกด ส ้ม เกด พี่เหว่า เอ๋ย ปิง เล็ก น ้าม่อน ป้าขวด นุ ้ย 0 0 0 1 1 0 0 0 0 0 0 0 1 0 0 1 1 1 1 1 0 1 1 1 0 0 0 1 household 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 neighbour 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 sport club/ park 1 0 0 0 0 1 1 0 0 0 0 0 1 0 0 1 1 1 0 1 1 1 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 meeting 0 0 0 0 0 0 0 0 1 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 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