|
The term 'social network' was
first coined in 1954 by J. A. Barnes (in: Class and Committees in a Norwegian Island
Parish, "Human Relations"). The maximum size of social networks tends to be
around 150 people (Dunbar's number) and the average size around 124 (Hill and Dunbar,
2002).
A social network is a social structure made of nodes which are generally individuals or
organizations. It indicates the ways in which they are connected through various social
familiarities ranging from casual acquaintance to close familial bonds.
The International Network for Social Network Analysis is the professional association of
social network analysis. Started in 1977 by Barry Wellman at the University of Toronto, it
now has more than 800 members and is headed by William Richards (Simon Fraser University)
Sociology 275: Social Network Analysis - courses.fas.harvard.edu/%7Esoc275/
F all 2005 Professor Peter V. Marsden
OVERVIEW: Social network analysis depicts social organization or social structure in terms
of patterned social relationships linking social units. This course deals with major
concepts and methods of social network analysis, touching on data collection but stressing
data analysis. Some readings give applications of network approaches to selected
substantive problems, but the emphasis of the seminar is on research methods.
Social network analysis takes seriously the proposition that behaviors of individual units
or actors are to be understood in social context. It seeks to operationalize concepts such
as "position", "role", or "social distance" that are
sometimes used casually or metaphorically, and is generally skeptical about the use of
categorical concepts or role labels, or of descriptions of social structure based on the
aggregation of characteristics of individual units. There are many models and methods in
social network analysis, but all share a conceptualization of social structure resting on
relationships of units or actors.
During the past decade, the pace of development in network studies has been very rapid.
This course will introduce foundations and tried-and-true approaches to network analysis,
and also give some attention to recently-developed methods and techniques. A single
semester allows us very limited time to accomplish this. We will have to be satisfied with
sketchy coverage of some topics, while others--such as comparing networks, models for
cognitive social structure data, network sampling and longitudinal network analysis--will
be neglected almost entirely.
We will begin with “whole network data” that purport to measure the social ties
linking all actors within some theoretically closed population. Here, we examine graph
theoretic and visual representations of networks, the detection of cohesive subgroups, and
indexes measuring the centrality and prominence of units within a network. In
mid-semester, we will study “egocentric network data” that measure the
Ainterpersonal environments@ surrounding individual units. Here we discuss basic measures
of network range, autonomy/@structural holes@, and some indicators of individual Asocial
capital.@ Later, we consider whole-network methods for “two-mode” networks
recording affiliations between two distinct types of social units, and
“blockmodels” or “positional analyses”, a more general approach to the
study of network subgroups.
Toward the end of the course, we examine recent advances in the development of statistical
methods for network data (notably Ap*@ or “exponential random graph” models),
and models for studying network-mediated diffusion and influence. At semester’s end,
we consider recent models for large social networks developed largely by physicists and
applied mathematicians.
A note on preparation. “Network analysis” and “social networks” have
become reasonably diverse areas involving many methodological approaches and substantive
concerns. Though there is some primarily “qualitative” work in social networks,
the bulk of it takes a quantitative angle, and this course will reflect that emphasis.
Diverse elements of mathematics and statistics are used in social network analyses.
Notwithstanding that, there are no specific quantitative prerequisites, but I trust that
participants will be familiar with basic regression analysis and open-minded about
quantitative material. I will try to convey important points without using complex
mathematics, and offer pointers to those of you who want to go into them in more technical
depth.
A note on focus. Network analysis is now applied within numerous substantive fields of
social science. This course will examine applications in such fields as organizational
analysis, community studies, and social epidemiology, among others. The emphasis of the
course will be on analytic methods that apply in several substantive fields; it will not
dwell on any particular substantive field. My experience is that students from diverse
backgrounds participate in the course. I encourage you to use the written assignments to
develop connections between course material and the substantive interests that draw you to
network studies.
TEXTS:
de Nooy, Wouter, Andrej Mrvar and Vladimir Batagelj. 2005. Exploratory Social Network
Analysis with Pajek. New York: Cambridge University Press.
A recent text introducing many common forms of network analysis with an emphasis on
graphics/visualization. I recommend that it be read while working with the freely
available network analysis software package Pajek.
Degenne, Alain and Michel Forsé. 1999. Introducing Social Networks. Thousand Oaks, CA:
Sage Publications. (Hereafter DF.)
A more substantively and conceptually grounded, less technically oriented, introduction to
network analysis.
Carrington, Peter J., John Scott, and Stanley Wasserman. 2005. Models and Methods in
Social Network Analysis. New York: Cambridge University Press. (Hereafter CSW.)
A collection on recent advances in methods for studying social networks. We will be
looking at most, but not all, of the chapters in CSW; some touch on statistical topics -
network sampling and modeling longitudinal network data - that we will not be able to
cover this term.
Watts, Duncan J. 2003. Six Degrees: The Science of a Connected Age. New York: Norton.
MB 874 - Introduction to Social Network Analysis
Wednesdays 3-6:30, Fall 2006
Classroom: Fulton 310. Prof. Steve Borgatti
borgatts@bc.edu
Introduction
This course provides an intensive introduction to the field of social network analysis.
There is both a class period (2.5 hours a week), and a lab period (1.5 hours immediately
following the class). My intention is to cover theory, concept and method in class, and
hands-on application in the lab. You are not required to attend to the lab. The purpose of
the lab is to teach you how to actually analyze social network data. This means mastering
the software tools as well as analytical strategies. IMPORTANT NOTE: You will need to
bring your own laptop in order to participate in the lab.
Network concepts covered will include graph-theoretic fundamentals, centrality, cohesion,
subgroups, equivalence and testing hypotheses. Theoretical areas will include
embeddedness, social capital, organizational learning and organizational governance. In
addition, I will try to include a practitioner perspective by using examples from
consulting engagements. Finally, the course will touch on data collection and study design
issues.
Required Books & Software
Borgatti, S.P., Everett, M.G. and Freeman, L.C. 2002. UCINET 6 for Windows: Software for
Social Network Analysis. Harvard: Analytic Technologies. Downloaded free on the network.
Wasserman and Faust. 1994. Social network Analysis. Cambridge. paperback. |
|