This is the presentation we gave at the Open-Ed conference November 2-4, 2010. We mainly presented Cohere, a social software from KMi (knowledge media Institute) of the Open University UK, as tool to make virtual ethnography and behavioural observations of users' web interactions. We also presented on use case in which cohere has been used to observe learners interactions in a P2PU course. About Open Ed 2010 The Open Education Conference has been described as “the annual reunion of the open education family.” Each year the conference serves as the world’s premiere venue for research related to open education, while simultaneously creating the most friendly and energetic atmosphere you’ll find at any academic conference. November 2-4, 2010, the seventh annual Open Education Conference moves to Barcelona for its first convening outside of North America! The 2010 conference venue is CosmoCaixa, designated Europe’s best science museum in 2006. The conference theme for 2010 was OER: Impact and Sustainability.

1. Anna De Liddo and Panagiota Alevizou
Open University, Milton Keynes, UK
a.deliddo@open.ac.uk; p.alevizou@open.ac.uk
Open Ed 2010 : OER impact and sustainability
02-05 N0vember 2010
Cosmocaixa Museum, Barcelona, Spain

2. Open Education and the diffusion of the Web have broken
the traditional barriers of ‘where’ education happens.
The diffusion of web-based social software has an impact on
the ways in which people learn and important questions
pertain on the practicalities and ethics of online research

3. Qualitative Data Analysis (QDA) methodologies are
used:
 where quantitative analysis is difficult and it does not
offer in depth insights
 when the aims are to provide an in-depth and
interpreted understanding of participants’ behaviours
 when analysis can be open to emerging concepts
and ideas
 When analysis aims to develop typologies and
explanations

4. CAQDA proprietary tools such as ATLAS.ti, NVivo,
Transana, etc require raw data to be analyzed
offline. This implies:
 ethical issues of permissions and privacy
 Legal issues related to copyright regimes
 Practical issues of cleaning and converting data to
appropriate format
 Issues of context: The analyst looses potentially
useful hints to make sense of the content they
analyze
 Issues of fast obsolescence of data

5. Most diffused CAQDA tools have evident limitations
in dealing with Web data.
We propose Cohere a Web tool for QDA of online
data.
Cohere supports virtual users observations by
exploiting a Web annotation Paradigm.
Cohere is as research tool for assisting the qualitative
analysis of online learning experiences.

6. Coding Memoing
Tagging Web Annotation

7. Coding and Memoing: Codes can be added by associating tags,
while memos are added in the Summary and Description
box

8.  Memoing:
 on hypothesis on how codes can be organized
and what are the categories they fall into.
 on connections between categories.
 Cohere have a specific way to support the
activity of making connections between categories
and memos.

10.  Cohere has been used to observe users’ behaviors
within the course ‘Copyright for Educators’ (Cycle
two).
 The object of analysis were the discussion forums,
where course students collaborate in order to
complete group assignments or discuss given tasks.
 In particular we analyzed students’ posts in the
“Pink Group”
 (http://p2pu.org/node/729/document/2692).

11.  we focus coding on the following two interrelated
dimensions, based on Burge’s (1994) peer bahaviour
models:
1. participation: What and how do participants share
online resources and reflections?
2. affective feedback: do participants use each
others’ names, complimenting each other and offer
supportive, remedial or critical interchanges?

12.  We looked at each post, highlighted clips of text,
and tag them with codes, which quickly started
grouping in three main categories:
1. People
2. content
3. Rhetorical role
Around these main categories we built the second
phase of: memoing and making connections
between memos.

13.  As previously said this means taking note of
connections between categories.
 We identified two main memos connection codes:
1. Posts
2. Addressed to
 These two links type enabled us to connect the
three main categories: people, data and rhetorical
role

14.  Map of the main categories, sub categories, and the relationship
between them.

15.  A key activity in QDA is sorting: help the analyst to
recognize the emerging structure of the phenomenon or
thesis that is being studied.
 Cohere has a network approach to data sorting and
visualization.
 Data, codes, and memos can be represented in a graph like
structure
 By coupling network visualization and code searching,
Cohere provides novel ways visualize results of qualitative
data analysis.

16.  Network of memos showing what people contributed to within
the pink group and the nature of the contribution.
 Icons represent main code sub-categories (purple lines
overwrites participants names)

17.  Clusters representation of memos per participant and
posts.

18.  Coupled search of memos by code and semantic connection.

19. We have proposed a method and tool to support QDA in
virtual environment and as a proof of concept we showed
Cohere in use to observe students interactions in P2PU
course.
Results of application shows that Cohere:
 is particularly promising when making collaborative QDA,
 provides sophisticated coding features such as connections
between memos and coding of those connections so that
they can easily retrieved and searched.
Finally we proposed a new network paradigm for data sorting
and representation: make sense of data and spot hidden
connections.

20.  Technology is not the key of success in P2PU, peoples are.
 Key player in the course have usually mixed role of
learners and tutors
 Learners come and go and this augment disorientation and
at the same time reinforce the role of the key players
 They share control and duties with peers that volunteered

21. Open Learning Network project (2009-12): olnet.org
funded by
http://olnet.org/