Modelling Learning & Performance: A Social Networks Perspective

Modelling Learning & Performance:
A Social Networks Perspective

Walter Ch. Paredes Kon Shing Kenneth Chung
Project Management Program Project Management Program
The University of Sydney The University of Sydney
wpar5857@uni.sydney.edu.au kenneth.chung@sydney.edu.au

A New Social Scenario

› The role of technology in the learning process and in a more collective
knowledge construction

› Lack of understanding of the dynamic of social interaction within learning
communities.

2

Research Motivating Questions

› Is there an interplay between social networks, learning and performance?

› If so, what is the role of social learning in the inherent relationship between
properties of social networks and performance?

› How does one quantify and measure learning within a social context?

› How does one account for social network properties of structure, relations
and position in modeling learning for the purpose of learning analytics?

3

Our Proposals

› A theoretical model based on social learning and social network theories to
understand how knowledge professionals engage in learning and
performance

› The construction of a novel metric called Content Richness as a surrogate
measure for social learning

4

Conceptual Foundations

› Learning is a highly complex process, which involves
cognitive, affective, individual and social dimensions [48]

› Social learning theory suggests that individuals learn through the
observation of the behavior of those we are connected with [2]

› To whom individuals are connected is critical for reaching unexplored
sources of novel information and then satisfy a determined necessity [14,15]

› There are fundamental contextual factors that facilitate learning [32]

› Human and Non-human sources of information and knowledge [46]

5

Models of Learning

› Situated Learning Theory [32]

- Communities of Practice [52]

- Legitimate Peripheral Participation

- SLT presents an interesting perspective to analyze learning from a social
networks point of view

- But, it does not consider Non-Relational sources of information such as
databases, webservers, blogs and discussion forums among others.

6

Models of Learning

› Connectivism [46]
- By incorporating ideas from Chaos, Social Networks and Complexity
theories, connectivism is focused on the explanation of the dynamics of learning

- Learning is permanently affected for new conditions in the environment and the
knowledge that can reside in non-relational repositories

- But, the importance of dialogues between relational and non-relational sources
has been just partially covered [42]

- Dialogues defined as unit of social interaction help us to understand how network
connections and meaningful content interchange influence individuals’ learning
process

7

Understanding Learning and Performance
through Social Networks

› Social Networks Analysis (SNA)
- Study of social systems from a structural perspective through the identification of
behavioral patterns based on node and tie attributes [17]

- An individual’s social structure and its properties can influence the access to
valuable resources rich on novel information [5, 8, 14, 25, 33]

- Those new sources can have a direct or indirect impact on the individual’s
learning and performance [12, 13]

- It is interesting to examine theories that explain how information is disseminated
through networks and how network structures can impact learning and
performance

8


› Strength of Weak Ties Theory [21]
- Tie is defined as “the combination of the amount of time, the emotional intensity,
the intimacy (mutual confiding), and the reciprocal services which characterize
the tie” [21]

- Study shows how professionals found better job opportunities through weak ties
instead of strong ones

- The stronger the tie, the more redundant the information becomes

- The importance of weak ties is that more resources are accessible through them

9


› Structural Holes Theory [10]
- The effective and efficient exploitation of an individual’s position in the network
can provide important informational and control benefits

- Trade-off between the number of contacts and the novelty of information. The
more contacts the higher the cost of maintaining relationships

- Maintain efficiently non-redundant connections that increase the diversity and
then the novelty of the information. Higher priority for cliques (group of contacts
closely connected within the group) but disconnected to other cliques.

- The bridging between two or more unrelated cliques is called a structural hole.

10

Towards a Social Networks Model for Learning
and Performance

› Given the unprecedented advancement in the adoption of social technologies, this
study provides evidence in the eLearning domain to help understand how networks
interact with technology to foster learning and performance in an era of digital
natives [4]

11

and Performance

› According to the arguments presented before, the following are the hypothesis
proposed:

› H1: Density of an individual’s network is negatively associated with learning

› H2: Efficiency of an individual’s network is positively associated with learning

› H3a: The extent to which an individual engages in communication within the network is positively
associated with learning

› H3b: The extent to which an individual contributes internally and externally to his group is positively
associated with learning

› H3c: Weak ties within an individual’s network is positively associated with learning

› H4: Learning is positively associated with performance

12

and Performance

Figure 1: Social Networks Model for Understanding Learning and Performance

13

Context and Methodology

› The e-Learning Environment
- Online project management course delivered the second semester of 2009 at a leading
“Group of Eight (Go8)” university in Australia
- 36 full-time working industry professionals with diverse backgrounds.
- Students based in nationally and overseas
- Complexities on coordination due to time zone differences
- Course material included lectures, tutorials, laboratory exercises and videos all accessible
through the university’s eLearning platform WebCT
- Synchronous (chat) and asynchronous (discussion board) communication channels
- Public and private discussion forums (Students dialogues)
- Individual and group assignments. 12 groups each of them with no more of three
participants
- Virtual collaboration

14


› Data Collection, Storage and Extraction
- Data collected from the public and private discussion forums

- 825 public messages and 722 private messages

- Unstructured nature of the message logs

- Preliminary data preparation (HTML extraction)

- Information stored in a MySQL database

- Java application to extract the node and tie data from the database for
generating the input file for UCINet (statistics) and NetDraw (sociograms)

15


› Message Content Classification
- A meaningful exchange of dialogues among team members is instrumental for enhancing
their learning process.

- By identifying the patterns of communication among team members it is possible to study
the structural properties of the group’s social network.

- Those patterns of communication have been categorized in past research according to
varied dimensions such as length of messages [34], channel of dissemination [34,39], content
[34,38], and meaning [19], but none of them used a social network perspective.

- In this study we have defined a classification method based on message content and
meaning in order to categorize each message sent through the public an private forums.

- Each category defined has a value associated which indicates the level of Content
Richness of the messages classified on it. The higher the value the more significant are the
messages of the category.

16


› The defined Content Richness categories are:

• Empty Message: Inexistent content, file exchange without dialogue, greeting messages

• Team Building Message: Personal introductions and very basic coordination. Final group
closing activities, congratulations for group achievements and recognition for mutual
cooperation

• Dissemination Message: Information about group submissions and notifications about new
document versions.

• Coordination Message: Team meetings (critical time zone difference)

• Collaboration Message: Knowledge creation. Problem solving dialogues. Individual and
group insights about the course and activities

17


Weight Content Category Message Example
0 Empty “Thank you Peter”
1 Team Building “Excellent work team”, “Has been a pleasure working with
you guys”
2 Dissemination “I submitted the last version of our report”
3 Coordination “Lets meet tomorrow at 7pm Sydney time”
4 Collaboration “Dear Peter I think your answer to the question is correct.
However, I found this article in which the authors analyze the
issues from a different perspective. Please consider also …”

Table 1: Content categories, their assigned weights, and some examples

18


Data Collection

Empty Team Building Dissemination Coordination Collaboration

Message Classification

Data Storage Data Extraction Data Analysis
+

Figure 2: Research methods and processes outlook

19


› Measures
• Measure of Network Structure  Density [45]
• Measure of Network Position  Efficiency [10]
• Measures of Engagement  Contribution Index [20]
 External-Internal Index [30]
 Content Richness Score
• Measures of Relationship  Average Tie Strength [36]
• Measures of Performance  Course Marks

20

Results

** Correlation is significant at the 0.01 level (1-tailed)
* Correlation is significant at the 0.05 level (1-tailed)

Table 2: Pearson’s Correlation (n=36)

21

Discussion

› We argue that rather than performance, social learning is influenced by
social networks properties such as structure, relations and position.

› The relationship between social networks and performance is mediated by
social learning.

› Studying how individuals interact and organize themselves (SNA) can be
useful to improve learning programs and positively influence performance.

› Content Richness was shown to be a good predictor of social learning due
to the interesting findings that connect the measure with most social
network properties modeled.

22

Limitations

› Size of dataset
- This was an exploratory that can not be generalized to the entire population but
the results are indicative of the power of social networks influencing learning
learning and indirectly, performance.

› Message content classification process
- Is still in an early stage and subject to criticism. We believe that the construction
of a taxonomy or vocabulary for group communication in studies of linguistics
and semantic data mining could allow the partial or full automation of the process

› Most of messages took place within groups
- There was not enough evidence about the interaction among groups. However,
according with our results, the quality of dialogues within groups was significant.

23

Limitations

› Other collaborative tools were not considered
- A future research could include those interactions through video, chat and voice

› Lecturer’s interaction
- Were not considered because we were interested social learning that take place
among students only. A future research could consider the lecturer’s interaction
excluding the computing of content richness score.

24

Conclusions

› Development of a theoretical model for understanding the impact of social
networks in learning and performance

› Construction of a content-based measure called Content Richness which
provides a new approach for measuring the level of engagement of learners in
social environment.

› Analysis of individual and group communication patterns of students located in
different cities, countries and time zones.

› Rather than performance, social learning is highly influenced by the learners’
social network properties.

› Model proposed would allow educators, professional development
leaders, managers and academics to enhance learning analytics and make
informed decisions and estimations of learning outcomes

25

References

› [2] Bandura, A. Social Learning Theory. General Learning Press, New York, 1977.
› [4] Bennett, S. and Maton, K. Beyong the 'Digital Natives' Debate: Towards a More Nuanced Understanding of
Students' Technology Experiences. Journal of Computer Assisted Learning, 26,5 (2010), 321-331.
› [5] Borgatti, S. Centrality and network flow. Social Networks, 27 (2005), 55-71.
› [8] Brass, D. Being in the right place: A structural analysis of individual influence in an organization. Administrative
Science Quarterly, 18(2) (1984), 321-344.
› [10] Burt, R.S. Structural Holes: The Social Structure of Competition. Harvard University Press, Cambridge, 1992.
› [12] Chung, K.S.K. and Hossain, L. Measuring Performance of Knowledge-intensive Workgroups through Social
Networks. Project Management Journal, 40,2 (2009), 34-58.
› [13] Chung, K.S.K. and Hossain, L. Towards a Social Network Model for Understanding Information and
Communication Technology use for General Practitioners in Rural Australia. Computers in Human Behavior, 26,4
(2010), 562-571.
› [14] Coleman, J.S. Social Capital in the Creation of Human Capital. The American Journal of Sociology, 94
(1988), 95-120.
› [15] Cross, R. and Cummings, J. Tie and network correlates of individual performance in knowledge-intensive
work. Academy of Management Journal, 47 (2004), 928-937.
› [17] Freeman, L.C. The Development of Social Network Analysis. Empirical Press, Vancouver, 2006.
› [20] Gloor, P., Laubacher, R., Dynes, S. and Zhao, S., Visualization of Communication Patterns in Collaborative
Innovation Networks: Analysis of dome W3C working groups. in ACM CKIM International Conference.

26

References

› [21] Granovetter, M.S. The Strength of Weak Ties. American Journal of Sociology, 78(6) (1973), 1360-1380.
› [25] Ibarra, H. Power, social influence and sense making: Effects of network centrality and proximity on employee
perceptions. Administrative Science Quarterly, 38(2) (1993), 277-303.
› [30] Krackhardt, D. and Stern, R.N. Informal networks and organizational crises: An experimental situation. Social
Psychology Quarterly, 51 (1988), 123-140.
› [32] Lave, J. and Wenger, E. Situated Learning: Legitimate Peripheral Participation. Cambridge University Press,
Cambridge, 1991.
› [33] Leavitt, H. Some effects of certain communication patterns on group performance. Journal of Abnormal and
Social Psychology, 46 (1951), 38-50.
› [36] Marsden, P. and Campbell, K.E. Measuring Tie Strength. Social Forces, 63,2 (1984), 482-501.
› [42] Ravenscroft, A. Dialogue and Connectivism: A New Approach to Understanding and Promoting Dialogue-Rich
Networked Learning. International Review of Research in Open and Distance Learning, 12(3) (2011), 139-160.
› [45] Scott, J. Social Networks Analysis: A Handbook. SAGE Publications, London, 2000.
› [46] Siemens, G. Connectivism: A Learning Theory for a Digital Age. International Journal of Instructional
Technology and Distance Learning, 2(1) (2004).
› [48] Stewart, M. Learning through research: an introduction to the main theories of learning JMU Learning &
Teaching Press, James Madison University Press, 2004, 6-14.
› [52] Wenger, E. Communities of Practice: Learning, Meaning and Identity. Cambridge University Press,
Cambridge, 1998.

27

Modelling Learning & Performance: A Social Networks Perspective

More Related Content

What's hot (19)

Viewers also liked (6)

Similar to Modelling Learning & Performance: A Social Networks Perspective (20)

Recently uploaded (20)

Modelling Learning & Performance: A Social Networks Perspective

Editor's Notes