=Paper=
{{Paper
|id=None
|storemode=property
|title=None
|pdfUrl=https://ceur-ws.org/Vol-1137/LAK14CLA_Intro.pdf
|volume=Vol-1137
}}
==None==
https://ceur-ws.org/Vol-1137/LAK14CLA_Intro.pdf
Computational Approaches to Connecting Levels
of Analysis in Networked Learning Communities
H. Ulrich Hoppe Daniel D. Suthers
University of Duisburg-Essen University of Hawai‘i at Manoa
(Germany) Honolulu (USA)
+49 203 379 3553 +1 808 956-3890
hoppe@collide.info suthers@hawaii.edu
ABSTRACT SNA has been criticized for eliminating time from the results of
an analysis in that it aggregates data over certain time intervals
The focus of this workshop is on the potential benefits and without being able to show time-dependent patterns. This can be
challenges of using specific computational methods to analyze partially overcome by using time series of networks. Yet, Zeini et
interactions in networked learning environments, particularly with al. have shown that also the choice of measurement intervals has
respect to integrating multiple analytic approaches towards systematic effects on the resulting networks [11]. On the other
understanding learning at multiple levels of agency, from hand, there are analytic methods, such as process mining, that are
individual to collective. The workshop is designed for researchers particularly geared to extracting procedural patterns from
interested in analytical studies of collaborative and networked interaction data [5]. Another group of methods deals with
learning in socio-technical networks, using data-intensive extracting content information from (textual) artifacts, using both
computational methods of analysis (including social-network “shallow” and “deep” linguistic techniques. At this point, we do
analysis, log-file analysis, information extraction and data not expect any one of these methods to be sufficient alone or to
mining). The workshop may also be of interest to pedagogical succeed on a purely technical level. We would rather favor a
professionals and educational decision makers who want to triangulation approach in which several methods are applied to
evaluate the potential of learning analytics techniques to better the same data sources and are interpreted in conjunction with each
inform their decisions regarding learning in technology-rich other and theoretical considerations.
environments. Although the workshop is focused on (new) computational
methods or new applications of such methods, we are also
interested in discussing computational approaches in a conceptual
Keywords and/or theoretical context. In this perspective, Suthers et al. [6,9]
Computational interaction analysis, levels of analysis, networked have developed a rich contextual framework to interpret
learning, CSCL collaborative interactions, introducing certain steps of inter-
pretation and addressing different levels of granularity using
concepts such as “contingencies” and “uptake” (as relations
between actions or contributions). It is of particular interest to
1. BACKGROUND further automate the application of such interpretation schemes.
This workshop continues the themes of “connecting levels” and Given these premises, we have invited contributions guided by the
“multivocality” that have characterized two series of workshops at following questions:
several conferences (ARV, CSCL, ICLS, LAK). We have taken a
multi-disciplinary perspective on learning as involving the agency x How to detect emergent phenomena and patterns in traces of
of individuals, groups and communities, and sought to understand collective/collaborative learning activities by using a
learning across these levels by integrating multiple methods and plurality of computational methods? How do we interconnect
granularities of analysis (cf. [7, 8]). these methods?
x What practical techniques such as different types of
The workshop for LAK 2014 focuses on specific computational triangulation or visualization can help to connect different
methods and their potential for analyzing interactions in levels and approaches of analysis?
networked learning communities from different perspectives.
x How can we integrate SNA with content analysis methods
Social network analysis (SNA, cf. [10]) is an approach that has
(including LDA, LSA, Network Text Analysis) and with the
been used in studying networked learning in the past (e.g., [1, 2,
detection of interaction patterns?
3]), and it is still of interest. However, we are particularly
x How can conceptually/theoretically grounded interpretation
interested in new methods and new combinations of different
schemes for collaborative activities be adequately
analysis techniques and computational approaches. Martinez et al.
operationalized and automated?
[4] provide an example of such a mixed method approach,
x What are the prospects of technical integration of analysis
coordinating SNA with other qualitative and quantitative analysis
tools through a kind of “open analysis workbench” (open
methods in a study of participatory aspects of learning in CSCL
architecture, GUI metaphors).
contexts. Further work in automating such approaches and
exploring the complementarities of different data sources and
analytic approaches is needed.
2. WORKSHOP CONTRIBUTIONS Committee
x Tilman Göhnert, University of Duisburg-Essen, Germany
The workshop includes a mixture of presentations, interactive x Sean Goggins, Drexel University, USA
demos, and group discussions. The following contributions will x Vanda Luego, Université Joseph Fourier, France
be presented in the workshop: x Agathe Merceron, Beuth University of Applied Sciences,
Daniel Suthers and Nathan Dwyer: Multilevel Analysis of Uptake, Germany
Sessions, and Key Actors in a Socio-Technical Network x Hiroaki Ogata, Kyushu University, Japan
x John Stamper, Carnegie Mellon University, USA
Agathe Merceron: Connecting Analysis of Speech Acts and
Performance Analysis - An Initial Study x Chris Teplovs, Problemshift Inc., Canada
Hiroaki Ogata, Songran Liu and Kousuke Mouri: Ubiquitous
Learning Analytics Using Learning Logs
4. REFERENCES
Hiroaki Ogata: Supporting Science Communication in a Museum
using Ubiquitous Learning Logs [1] M. de Laat, V. Lally, L. Lipponen and R.-J. Simons,
Tilman Göhnert, Sabrina Ziebarth, Per Verheyen, and H. Ulrich Investigating patterns of interaction in networked learning
Hoppe: Integration of a Flexible Analytics Workbench with a and computer-supported collaborative learning: A role for
Learning Platform for Medical Specialty Training Social Network Analysis, International Journal of Computer
Supported Collaborative Learning, 2 (2007), pp. 87-103.
H. Ulrich Hoppe, Tilman Göhnert, Laura Steinert, and [2] A. Harrer, N. Malzahn, S. Zeini, and H.U. Hoppe,
Christopher Charles: A Web-based Tool for Communication Flow Combining Social Network Analysis with semantic relations
Analysis of Online Chats to support the evolution of a scientific community, in C.
Wanli Xing and Sean Goggins: Automated CSCL Group Chinn, G. Erkens and S. Puntambekar, eds., The Computer
Assessment: Activity Theory based Clustering Method Supported Collaborative Learning (CSCL) Conference 2007,
Cindy Hmelo-Silver, Carolyn Rosé, and Jeff Levy: Fostering a International Society of the Learning Sciences, New
Learning Community in MOOCs Brunswick, 2007, pp. 267-276.
[3] C. Haythornthwaite, Social network methods and measures
for examining e-learning, E-learning seminar, University of
Southampton, 2005.
3. WORKSHOP FACILITATORS [4] A. Martinez, Y. Dimitriadis, E. Gomez-Sanchez, B. Rubia-
Ulrich Hoppe Avi, I. Jorrin-Abellan and J. A. Marcos, Studying
H. Ulrich Hoppe holds a full professorship for “Cooperative and participation networks in collaboration using mixed
Learning Support Systems” in the Department of Computer methods, International Journal of Computer-Supported
Science and Applied Cognitive Science at the University of Collaborative Learning, 1 (2006), pp. 383-408.
Duisburg-Essen, Germany. With his research group COLLIDE, [5] P. Reimann, Time is precious: Variable- and event-centred
Ulrich Hoppe has been engaged in several European projects in approaches to process analysis in CSCL research, Computer
the area of advanced computational technologies in education Supported Collaborative Learning, 4 (2009), pp. 239-257.
since 1998. Ulrich Hoppe has been program chair of AIED and [6] D. D. Suthers, N. Dwyer, R. Medina and R. Vatrapu, A
CSCL 2003, ICCE 2007 and CRIWG 2012. His current research framework for conceptualizing, representing, and analyzing
interests include: interactive and collaborative media for learning distributed interaction, International Journal of Computer
and knowledge construction; analysis, modelling, and intelligent Supported Collaborative Learning, 5 (2010), pp. 5-42.
support of interactive and collaborative learning processes; social [7] D. D. Suthers, H. U. Hoppe, M. De Laat, and S. Buckingham
network analysis and community support. Shum. LAK, Connecting levels and methods of analysis in
networked learning communities. Proceedings of LAK 2012.
Dan Suthers ACM, (2012), pp. 11-13.
Daniel D. Suthers is Professor in the Department of Information [8] D. D. Suthers, K. Lund, C. Rosé, G. Dyke, C. Teplovs and N.
and Computer Sciences at the University of Hawai`i at Manoa, Law. Productive Multivocality in the Analysis of Group
where he directs the Laboratory for Interactive Learning Interactions. New York: Springer, 2013.
Technologies. Dr. Suthers’ research is generally concerned with [9] D. D. Suthers and D. Rosen, A unified framework for multi-
cognitive, social and computational perspectives on designing and level analysis of distributed learning Proceedings of the
evaluating software for learning, collaboration, and community. First International Conference on Learning Analytics &
Dr. Suthers initiated and chaired a series of five workshops on Knowledge, Banff, Alberta, February 27-March 1, 2011,
Productive Multivocality in Analysis of Collaborative Learning, 2011.
involving dozens of researchers in a long term collaboration [10] S. Wasserman and K. Faust, Social Network Analysis:
leading to a book in press. Subsequently he led workshops on Methods and Applications, Cambridge University Press,
Connecting Levels of Analysis at the CSCL 2011 and 2013 and New York, 1994.
LAK 2012 conferences. He has also served as program chair for [11] S. Zeini, T. Göhnert, L. Krempel and Hoppe H. U. (2012).
LAK, ICCE, and two CSCL conferences, and has had steering The impact of measurement time on subgroup detection in
committee roles (e.g., workshop chair, interactive events chair) for online communities. The 2012 IEEE/ACM International
numerous other conferences. Conference on Advances in Social Networks Analysis and
Mining (ASONAM 2012).