96
Chapter VIII
Illustrating Knowledge
Networks as Sociograms
Stefan Hrastinski
Uppsala University, Sweden
Abstract
This chapter looks at the concept of sociograms that has great illustrative importance in some circumstances, especially for studying small knowledge networks. It is argued that the sociogram approach
might be particularly useful for those who view learning and participation in knowledge networks as an
inherently social phenomenon. Then, the sociogram approach is described and benefits and limitations
of different approaches are discussed. The chapter also includes an exercise, web resources, further
readings, and suggestions for possible paper titles.
Introduction
In the 1930s, Jacob Moreno (1934) founded sociometry, later defined as “the measurement of
interpersonal relations in small groups” (Wasserman & Faust, 1994, p. 11). It is a precursor to
social network analysis, which has been developed
ever since and now provides a set of techniques for
understanding patterns of relations between and
among people, groups and organizations (Garton,
Haythornthwaite & Wellman, 1999). Social network data is initially organized in sociomatrices.
For example, such a matrix might include data
on who communicate with whom. Sociomatrices
might then be used for quantitative analysis or
drawing sociograms or graphs. Sociograms have
been of great illustrative importance ever since
the 1930s (Moreno, 1934). In this chapter, the
concept of sociograms is discussed. It is argued
that sociograms have great illustrative importance
in some circumstances.
In the next section, different perspectives on
learning in knowledge networks are discussed.
It is argued that the sociogram approach might
be particularly useful for those who view learning and participation in knowledge networks as
Copyright © 2009, IGI Global, distributing in print or electronic forms without written permission of IGI Global is prohibited.
Illustrating Knowledge Networks as Sociograms
an inherently social phenomenon. In the third
section, a basic introduction to the concept of
sociograms is presented. Then, different examples
of sociograms, and their benefits and limitations
are discussed.
From Objectivist to Social
Perspectives on Knowledge
Networks
There are many different perspectives on learning,
and the perspective of learning that the managers
and members of a knowledge network subscribe to
will both explicitly and implicitly influence participation and learning in the knowledge network.
In this section, a brief review, which describes
how the emphasis has shifted from objectivist
perspectives on learning towards more social
perspectives on learning, is presented.
Learning has traditionally been based on
objectivist theories on learning. The objectivist
tradition assumes that knowledge is an object that
can be absorbed (Duffy & Jonassen, 1992). This
assumption originates from the psychological
school of behaviourism. The key theory of behaviourism was that of stimuli and response, where
stimuli, and combinations of stimuli, were argued
to determine reactions (Watson, 1925/1997). The
aim was “to be able to reproduce [a] reaction at
another time (and possibly in other individuals as
well)” by determining “what the situation is that
causes this particular reaction” (ibid, p. 20). When
applying ideas originating from the objectivist
tradition, the goal of the participants of a knowledge network becomes to transfer “knowledge
objects” (Duffy & Jonassen, 1992; Leidner &
Jarvenpaa, 1995). Prior experiences and human
interpretation is not of interest since it is seen
as leading to partial and biased understandings
(Duffy & Jonassen, 1992). Technology is used to
transmit knowledge with limited possibilities for
conversations among members of the knowledge
network (Edelson, Pea & Gomez, 1996).
In the beginning of the 1990s, constructivist
theories on learning gained popularity. The argument of constructivism is that there is no correct
“meaning” of the world that we are striving to understand. Instead, it is argued that there are many
ways to structure the world, and there are many
meanings or perspectives for any event or concept
(Duffy & Jonassen, 1992). Individually oriented
constructivist models assume that the main objective when managing knowledge networks should
be to support the members in gaining experiences
rather than aiming to transfer “knowledge objects”
between the members of the knowledge network
(Säljö, 2000). Thus, constructivist theories have
moved away from the knowledge transmission
model towards an active learner model. However,
like objectivism, constructivism has ”commonly
focused on the learner as an individual, learning
in isolation from other learners” (Edelson et al.,
1996, p. 151).
Social theories on learning (e.g., Wenger, 1998;
Vygotsky, 1978) have gained renewed interest
since the beginning of the 1990s (Heeren, 1996)
and emphasize that learning is dialogue, both
internal and by social negotiation (Jonassen &
Land, 2000). Rather than being solely based on
experience with the physical world, the construction of knowledge and understanding is seen
as a fundamentally social activity (Littleton &
Häkkinen, 1999, p. 24). There exists different
perspectives but the most common ones share a
focus on participation as a condition for learning
(Jaldemark, Lindberg & Olofsson, 2006).
The basic premises and implications of the
three theoretical perspectives on learning that
have been discussed are summarized in Table 1.
Jonassen and Land (2000) argue that never before
have so many learning theories shared so many assumptions and common foundations. Nowadays,
most researchers agree upon that knowledge not
only exists in individual minds but also “in the
discourse among individuals, the social relationships that bind them, the physical artefacts that
they use and produce, and the theories, models
97
Illustrating Knowledge Networks as Sociograms
Table 1. Summary of the three perspectives on learning (adapted from Leidner & Jarvenpaa, 1995)
Theoretical
perspective
Objectivist
Constructivist
Social
Basic premise
Learning occurs by absorbing objective
knowledge.
Learning occurs by constructing
knowledge individually.
Learning occurs by participating in the
social world.
and methods they use to produce them” (Jonassen
& Land, 2000, p. vi).
Each of the three theoretical perspectives on
learning inspires different people to a different
extent. In the next section, the social network approach of sociograms is discussed. This approach
is especially useful for studying participation and
learning in knowledge networks from a social
perspective.
Illustrating Knowledge
Networks as Sociograms
Sociograms have been of great illustrative importance ever since the 1930s (Moreno, 1934).
In social network analysis, relations describe
particular types of resource exchange between
actors. A social network is defined as “a finite
set or sets of actors and the relation or relations
defined on them” (Wasserman & Faust, 1994,
p. 20). The resources exchanged among pairs of
actors can be of many types, including tangibles
such as goods, services, or money, or intangibles
such as information, social support, or influence
(Haythornthwaite, 1996, p. 323). As illustrated in
Figure 1, there are four levels of measurement in
relational data.
98
Implication for managing knowledge
networks
The manager(s) of a knowledge network
should transfer knowledge to its members.
The manager(s) of a knowledge network
should support rather than direct its
members.
The manager(s) of a knowledge network
should encourage communication among
its members.
In a sociogram, each node represents a member
of the network and lines show which others each
node is tied to. For example, some participants
of a knowledge network may give more information to peers or it may be experienced as they do.
Differences in reporting of relationships are often
found in situations where an actor is a prominent
figure of a network (Haythornthwaite, 1996).
Ties may be directed and then arrows, instead
of just lines, are used. Assigning a numeric value
to each arrow can denote the strength of the ties,
for example, the frequency of communication.
An alternative approach is to use thinner (weak
ties) and thicker (strong ties) lines (Hrastinski,
2006a, 2006b). Standard works on social network
analysis (Scott, 1991; Wasserman & Faust, 1994)
suggest that a number is assigned to each line to
denote strength. However, as discussed in the
next section, this can make a sociogram difficult
to interpret.
In Figure 2, an example of a fictive sociogram
is presented. Let us say that the sociogram illustrates perceived information exchanges in an
online knowledge network, which communicate
in a discussion board, during a week. It seems
Illustrating Knowledge Networks as Sociograms
Figure 1. Levels of measurement in relational data (Scott, 1991, p. 48)
Directionality
Undirected
Directed
Binary
1
3
Valued
2
4
Numeration
like the most prominent member of the knowledge network is B, who gives information to and
receives information from A and D, and gives
information to E. Node C and E seem to mostly
receive information, rather than contributing with
information to the others.
Examples of Sociograms
Social or knowledge network data is commonly
organized in sociomatrices. Such matrices can
then be transformed to sociograms. In this section, it will be drawn on Daugherty and Turner
(2003) who analysed sociomatrices to assess
group dynamics in a web-based course. Benefits
and limitations of the sociogram approach will be
Figure 2. A binary and directed sociogram
A
C
B
D
E
distinguished, by converting one of their sociomatrices to sociograms of different types.
Daugherty and Turner argued that sociometry
is a useful approach for assessing group dynamics.
This claim was based on a study of a web-based
college graduate course on educational research.
Ten of the eleven enrolled students responded to a
“sociometric survey” of 13 questions. For example,
the first question was: Who from the class would
you like to be around in college course settings?
Students were asked to limit their choices to 1-2
classmates per question.
Drawing on the results, a sociomatrix that reflects the number of nominations students reported
and received by person was created (see Table 2).
By analysing the table, Daugherty and Turner
drew several conclusions: “First, the recipients of
numerous choices from others and who, therefore,
held positions of popularity were identified. [The
table] showed that the most frequently chosen
students (D and K) each received almost three
nominations per respondent with mean choice
selections of 2.89 and 2.8, respectively. … It was
also evident … that class members infrequently
chose 2 students. These 2 students (A and E)
received, on average, less than one nomination
per respondent, .78 and .67, respectively. … [The
table] also showed chains of interconnectedness
between students. Pairs or individuals that nomi-
99
Illustrating Knowledge Networks as Sociograms
nated each other frequently were clearly evident.
Seven dyads (A/F, A/J, C/E, C/I, D/H, D/I, F/G)
were shown in which each selected the other a
minimum of three times.” (p. 269).
It is clear that Daugherty and Turner could
draw conclusions, which might be essential in
understanding group dynamics. Notably, many
of the conclusions were proposed after studying
the means of nominations. In Figure 3-5, sociograms have been created, by using the software
Ucinet 6 (Borgatti, Everett & Freeman, 2002), to
graphically illustrate the sociometric data collected by Daugherty and Turner. In the figures,
arrows, instead of just lines, were used to illustrate
the direction of perceived exchanges. This may
be important since differences in reporting of
relationships are often found in situations where
some actors are prominent figures of a network
(Haythornthwaite, 1996).
because he or she did not answer the questionnaire.
This type of sociogram seems primarily to be
useful for analyzing small knowledge networks.
However, the sociogram illustrates that every
member of the knowledge network interact with
other members of the network. D and H reciprocally selected each other many times. H selected
E six times, while E never selected H. These
findings can be derived from Table 2 but it can
be assumed that different people would prefer to
study the matrix of Table 2 while others would
prefer the sociogram of Figure 3.
Example 2: Directed Binary
Sociogram
In Figure 4, a directed binary sociogram is presented (type 3). The sociogram illustrates strong
ties by only displaying the arrows with a value of
3 or higher. This makes the sociogram easier to
interpret, but a weakness of this approach is that
more detailed information is lost. As identified
by Daugherty and Turner, the sociogram tells
us that seven pairs of actors (A/F, A/J, C/E, C/I,
D/H, D/I, F/G) reciprocally selected the other a
Example 1: Directed Values
Sociogram
In Figure 3, a directed valued sociogram is presented (type 4). Note that student K was excluded
Table 2. Nominations received by respondent (Daugherty & Turner, 2003, p. 268)
Recipient
A
B
C
D
E
F
G
H
I
J
K
100
A
1
0
1
0
3
1
1
0
5
5
B
1
0
1
0
2
0
1
4
9
4
C
0
0
6
3
0
0
2
3
0
0
D
0
0
0
1
0
2
7
3
0
4
E
0
0
4
3
0
0
6
6
0
0
F
3
6
0
0
0
4
0
0
3
3
G
0
1
0
4
0
3
2
0
0
8
H
0
2
0
5
0
1
0
1
0
0
I
0
2
6
3
2
0
0
1
0
0
J
3
1
0
3
0
0
2
0
0
4
Mean
0.78
1.44
1.11
2.89
6.7
1
1
2.22
1.89
1.89
2.8
SD
1.3
1.88
2.26
2.47
1.12
1.32
1.41
2.54
2.2
3.22
2.74
Illustrating Knowledge Networks as Sociograms
minimum of three times. The sociograms also
illustrates that all members of the knowledge
network maintained at least one relation with a
class member. Thus, among other things, it can
be learnt that the problem of isolates was not apparent in the network.
ties) lines. As mentioned earlier, this has not been
the most common approach. The sociogram was
created using Ucinet 6 (Borgatti et al., 2002), the
software can represent tie strength as line width
rather than by numeric values.
Example 3: Directed Sociogram with
Varying Line Widths
Conclusion
The sociogram of Figure 3 gives more detail by
denoting each arrow with a numeric value, while
the sociogram of Figure 4 is easier to interpret but
gives less detail. Figure 5 presents a compromise
between these two approaches. The strength of the
ties, measured as frequency of communication,
is denoted by thin (weak ties) and thick (strong
In this chapter, it has been argued that the sociogram approach might be particularly useful
for those who view learning and participation
in knowledge networks as an inherently social
phenomenon. After giving a basic introduction
to the concept of sociograms, three examples of
different types of sociograms were put forward.
This chapter has showed that essential questions
Figure 3. A directed valued sociogram
101
Illustrating Knowledge Networks as Sociograms
Figure 4. A directed binary sociogram
Figure 5. A directed sociogram, which illustrates strong and weak ties
102
Illustrating Knowledge Networks as Sociograms
to ask when studying knowledge networks are:
Can sociograms be used to aid in illustrating and
understanding the knowledge network under investigation? Which type(s) of sociograms can aid
in understanding the knowledge network under
investigation?
Internet Session: “Social
Networks of aN Online
Community”
Choose an online community, which is a typical
example of a knowledge network. For example,
Microsoft’s forums include many knowledge networks: http://forums.microsoft.com/msdn/.
Interaction
Create three sociograms (see Example 1-3) that
illustrate the exchanges among the members of
the knowledge network. If the community is large,
illustrate a subset of the network by creating an
ego-centered network, which illustrates all exchanges from and to one or a few actors. Which
were the benefits and limitations of each type of
sociogram you chose to create?
Useful URLs
1.
2.
3.
Analytic Technologies: Software for analyzing social networks and creating sociograms,
http://www.analytictech.com/
Netlab: scholarly network studying computer
networks, communication networks, and
social networks, http://www.chass.utoronto.
ca/~wellman/netlab/
International network for social network
analysis, http://www.insna.org/
Further Readings
Scott, J. (1991). Social network analysis: A handbook. Newbury Park, CA: Sage Publications.
Wasserman, S., & Faust, K. (1994). Social network
analysis: Methods and applications. Cambridge:
Cambridge University Press.
Garton, L., Haythornthwaite, C., & Wellman, B.
(1999). Studying on-line social networks. In S.
Jones (Ed.), Doing Internet research: Critical
issues and methods for examining the net (pp.
77-105). Thousand Oaks: Sage Publications.
Haythornthwaite, C. (1996). Social network
analysis: An approach and set of techniques for
the study of information exchange. Library and
Information Science Research, 18(4), 323-342.
References
Borgatti, S. P., Everett, M. G., & Freeman, L.
C. (2002). Ucinet 6 for Windows: Software for
social network analysis. Harvard: Analytic
Technologies.
Daugherty, M., & Turner, J. (2003). Sociometry:
An approach for assessing group dynamics in
web-based courses. Interactive Learning Environments, 11(3), 263-275.
Duffy, T. M., & Jonassen, D. H. (1992). Constructivism: New implications for instructional
technology. In T. M. Duffy & D. H. Jonassen
(Eds.), Constructivism and the technology of
instruction: A conversation. New Jersey: Lawrence Erlbaum.
Edelson, D. C., Pea, R. D., & Gomez, L. (1996).
Constructivism in the collaboratory. In B. G.
Wilson (Ed.), Constructivist learning environments: Case studies in instructional design (pp.
151-164). Englewood Cliffs, New Jersey: Educational Technology Publications.
Garton, L., Haythornthwaite, C., & Wellman, B.
(1999). Studying on-line social networks. In S.
Jones (Ed.), Doing Internet research: Critical
issues and methods for examining the net (pp.
77-105). Thousand Oaks: Sage Publications.
103
Illustrating Knowledge Networks as Sociograms
Haythornthwaite, C. (1996). Social network
analysis: An approach and set of techniques for
the study of information exchange. Library and
Information Science Research, 18(4), 323-342.
Leidner, D. E., & Jarvenpaa, S. L. (1995). The use
of information technology to enhance management school education: A theoretical view. MIS
Quarterly, 19(3), 265-291.
Heeren, E. (1996). Technology support for collaborative distance learning. Doctoral thesis,
University of Twente, Twente.
Littleton, K., & Häkkinen, P. (1999). Learning together: Understanding the processes of computerbased collaborative learning. In P. Dillenbourg
(Ed.), Collaborative learning: Cognitive and
computational approaches (pp. 20-30). Oxford:
Elsevier Science Ltd.
Hrastinski, S. (2006a). Introducing an informal
synchronous medium in a distance learning
course: How is participation affected? Internet
and Higher Education, 9(2), 117-131.
Hrastinski, S. (2006b). The relationship between
adopting a synchronous medium and participation
in online group work: An explorative study. Interactive Learning Environments, 14(2), 137-152.
Jaldemark, J., Lindberg, J. O., & Olofsson, A. D.
(2006). Sharing the distance or a distance shared:
Social and individual aspects of participation in
ICT-supported distance-based teacher education. In M. Chaib & A. K. Svensson (Eds.), ICT
in teacher education: Challenging prospects
(pp. 142-160). Jönköping: Jönköping University
Press.
Jonassen, D. H., & Land, S. M. (2000). Preface. In
D. H. Jonassen & S. M. Land (Eds.), Theoretical
foundations of learning environments (pp. iii-ix).
New Jersey: Lawrence Erlbaum.
104
Moreno, J. L. (1934). Who shall survive? A new
approach to the problems of human interrelations.
Washington, DC: Nervous and Mental Disease
Publishing Company.
Säljö, R. (2000). Lärande i praktiken: Ett sociokulturellt perspektiv (Learning in practice: A
sociocultural perspective. Stockholm: Prisma.
Wasserman, S., & Faust, K. (1994). Social network
analysis: Methods and applications. Cambridge:
Cambridge University Press.
Watson, J. B. (1925/1997). Behaviorism. New
Jersey: Transaction Publishers.
Wenger, E. (1998). Communities of practice:
Learning, meaning, and identity. Cambridge:
Cambridge University Press.
Vygotsky, L. S. (1978). Mind in society: The
development of higher psychological processes.
Cambridge, Massachusetts: Harvard University
Press.