Just to remind myself that, if I’m looking at online relationships, I might want to reread Haythornthwaite on social network theory.
Social network analysis
Leave a reply
Category Archives: Research tools
Issues to consider
I spoke to Robin Goodfellow the other day. He suggested:
Investigating which identities are readily taken up in a FirstClass conference.
Considering the role of the researcher in the co-construction of identity. Which stories are being rehearsed and which discourses are being reproduced?
Tinkering with research questions
I’m felling pleased with myself, as I’ve managed to knock together 5000 words, which looks like a first draft of half of my literature review. It’s full of holes, and it misses out the really difficult areas (learning theory and various discourse analysis views on identity) but I can see that it might all fit together eventually.
As a result, I’ve tinkered with my research questions once again. So, just to archive how they fluctuate, here they are:
- Which are the key identities available within a learning community which comes together in an asynchronous online environment?
- How are these identities introduced or created?
- Which of these identities are mobilised to support learning, and which to discourage learning?
- How can the asynchronous environment be designed in order that participants will position themselves, and others, in ways which support learning?
Literature review
Working on the draft of a draft of a literature review. It’s such a mound of material that it’s really hard to arrange sensibly. I’ve got three and a half thousand words, which I guess is about a third of a literature review, but I keep getting lost in it.
I’ve done a fairly coherent intro, though I assume it’ll be completely rewritten in time. There’s a section on identity which is pretty interesting, though it doesn’t even touch on the nuts and bolts of identity construction in discourse. And there’s a bit about how online learning communities are constructed which I suspect I’ll tear up and start again at some point.
I’ve got an outline in Powerpoint, and an outline in Word, and a pencilled outline on one sheet of paper, and a whole series of piles of snippets of papers which is supposed to give me a physical framework, and I’m still meandering all over the place.
Spose it would help if I’d read all the material about education and learning that I need for part one. Then I’d have something more substantial to hang it all on.
Oh well, back to social loafing theory…
Questions and hypotheses
I’m tinkering with my research questions. I’ve mostly reverted to identity from subject position, partly because it’s more comprehensible, partly because it’s sexier, and patly because it fits in with the idea of learning as transformation of identity. I’ve got a tentative title and hypotheses as well.
I’ve thrown in a short definition of a successful virtual community, although this is obviously going to need a lot more work.
Who do you think you are: The roles of identity in virtual learning community.
Research questions:
Once it is accepted that virtual learning community exist, it is important to ask how such communities can be designed and administered to support learning as a transformative process which changes identity. This research therefore asks:
* Which are the key identities to be found within a learning community which comes together in an asynchronous online environment?
* How are these identities introduced or created?
* Which of these identities are mobilised to support learning, and which to discourage learning?
* How can the asynchronous environment be designed in order that participants will position themselves, and others, in ways which support learning?
Hypotheses:
* Learners have multiple identities within this type of community, many of them products of the online environment.
* Learners must work hard to establish their identities and those of others. This work is not accomplished in the same way as it would be offline.
* The environment, the community design and the participants all introduce certain subject positions.
* A successful virtual learning community will engage in a high level of exploratory talk and knowledge creation. Participants will show evidence of identity change in line with the aims of the community.
* In a successful virtual learning community, the majority of commonly available subject positions will be mobilised to support learning.
* In less successful virtual learning communities, this will not be the case.
* Design and administration of virtual learning communities exert a major influence on the subject positions mobilised within those communities, and on how and why those subject positions are mobilised.
Doing a pilot study
A171 Start writing for the Internet, 10 points, starts May, runs for 12 weeks. About 80 students – evenly balanced between males and females.
A173 Writing family history, 10 points, starts May, runs for 12 weeks. About 180 students – mostly women. Also has an OUSA cafe.
T183 Design and the web, 10 points, starts May, runs for 10 weeks. About 400 students – evenly balanced between males and females.
I could do participant observation on one of these. In fact, these ten-point courses are quite good for me and for data collection, as they run four times a year so I could, theoretically, do four or five of these, one after the other.
This would be a substantial pilot, running May / June / July. To have something for my probationary assessment, I need to do something else.
For example, if Karen and Denise have records of a FirstClass conference which I could use, I could
(a) try out detailed discourse analysis on short sections of this
(b) analyse it for key subject positions. This would give me a basic framework for analysis of later data.
(c) find evidence of learning being supported and/or discouraged
(d) investigate the role of the medium / the staff in positioning students.
(e) use the experience to refine my research questions.
If possible, I could experiment with corpus analysis, using Denise and Karen’s conference data. This would give me a chance to find out more about this method, to get to grips with any relevant software and to start to build up a corpus. I’d like to know whether corpus analysis can be used to map changes over time – for example, in pronoun use, the use of technical vocabulary, or in the use of descriptors. Can it be used to look at the stage at which a supervisor makes a comment, or at what students responses to supervisors’ comments tend to be? Does it only work with millions of words of data, or can I pilot it with fairly small amounts of data?
Those pilots would use other people’s data and so they might not get ethical approval. In that case, I could probably try out discourse analysis methods on my U800 data or on some of the FirstClass cafes which are open to all. This would be more like a TMA than a pilot – so I’d like to do more than this, if possible.
I can’t see much point in piloting my interviewing unless I have access to a course conference.
Resesarch methods
Well, there’s a lot of discourse analysis here.
* Corpus analysis. I put the text of all four conferences into a data base and pull out subject positions in some way.
* I follow the trajectory of individual students. I’d like to supplement this with interviews.
* I look at the whole conference, pull out interactions which I think are particularly salient and analyse them in depth. Perhaps do this with particular emphasis on interventions from tutors/moderators.
* I also need to find evidence of learning being supported and discouraged and I think I need to refer here to Neil Mercer’s work on exploratory talk. Can I see the students building knowledge? Do they think they are learning?
* I need to look at whether this is a learning community. Do the participants perceive it as such? Does it behave like a learning community.
* I think it’s important to look at what’s happening backstage. Who’s lurking, how are they lurking (I think I can access this through message history). How are they linking up outside FirstClass? This needs interviews and participant observation.
* Epistolary interviews make sense here: I’m interested in this method, the whole research project is about online interaction. These could be backed up by face to face and telephone interviews.
* I’d like to interview the course team and the techies about the nuts and bolts of building this community. Which features are inherent in FirstClass? What does the Open University require? Where did they get their experience of working with an online learning community? I think this is important from the point of view of making this applied research.
* I’d like to look at the subeject positions assigned / created by the OU. How has the OU classified these students for its own research purposes? Which classifications have the students had to fit into to be at the Open University / on this course? Which subject positions does the OU push students into? This would require analysis of course and OU literature, and of material held by the survey office.
Research questions
* Which are the main subject positions to be found within a learning community which comes together in an aynchronous online environment?
* How are these subject positions introduced or created?
* Which of these subject positions work to support learning, and which discourage learning?
* How can the asynchronous environment be designed in order that participants will position themselves, and others, in ways which support learning?
How would I answer these questions? Well, first of all I’ve got to find an online community which comes together in an asynchronous environment. It’s probably best if they only come together online, because then I have access to all the whole-community activity. The other activity of the comunity eg texts, emails, IMs, meetings, phone conversations I could catch either through interviews or through participant observation.
I’d probably want more than one community so I could generalise. On the other hand, this is potentially a vast set of data, so I don’t want to go wild and have lots of communities. What about one community on which I focus, and another three where I observe but don’t collect so much data?
So, four OU courses which come together via First Class. They’d better be undergraduate, because postgraduate isn’t so generalisable. They’d better be in different disciplines, because that makes it more generalisable. If I want to be a participant observer it might be best to have a course that I’ll find relatively easy, so I don’t have to waste huge amounts of time doing the work. Or, another possibility, if I were tutoring on the course I’d have access to different sorts of data.
And position/identity has a very strong link with gender so I’d like to look at a mostly boy course and a mostly girl course, and perhaps at a level one / openings course where people aren’t used to being students, and a level three course where they’re used to learning.
And it;s probably better if they’re not being too reflexive, so not one of the courses on identity.
Caroline Haythornthwaite (1998) (10.1.06)
Haythornthwaite, C. (1998)
A social network study of the growth of community among distance learners
Information Research, 4, 1
* Communication frequency is associated with the maintenance of more relations and the use of more media.
* Patterns of media use are highly influenced by the media established by the instructor for class interaction
* Email is important for pairs who communicate more often.
* Actor positions in collaborative work, exchanging advice and socialising relations are similar, but this position does not correlate with their position in an emotional support network.
* Different actors are involved in the emotional support network than in other networks
* Group interaction patterns become less flexible over time.
* An individual’s perception of their own sense of belonging to the class is most strongly correlated with their centrality in exchanging advice networks.
Social network theory (10.1.06)
This is an analytical approach I could consider using. It might be useful for Gill as well?
Social network theory holds that behaviour is affected more by the kinds of ties and networks in which people are involved than by the norms and attributes that individuals possess. It examines patterns of ties to see what patterns emerge from their interactions.
In social network terms, pairs maintain relations (such as working together or friendship) and ties (a bond between two people based on one or more relations). The more relations a pair maintains, and the more frequently or intensely they maintain them, the stronger or closer the tie.
Pair-level bonding contributes to the sense of belonging to a group that is necessary to sustain the group as an entity rather than as a set of individuals. Feelings of belonging and community lead to greater commitment of group efforts, greater co-operation and greater satisfaction with group efforts.
One measure of an individual’s place in a network is their centrality – how well they are positioned to receive and disseminate information to all other members of the network. A star has access to information circulating the entire network, can influence others and the flow of information. The other end of the scale is isolation. The isolate does not maintain connections and thus does not receive communications. They can be cut off from information or receive it late.
Centrality can be measured by counting the number of others with whom an actor maintains relations. Can also be measured by closeness – the distance from each person to each other person. Central actors are closer to all others than are other actors. This means they are more likely to hear information available on the network. A third measure is betweenness, the extent to which an actor is situated between others (so information must pass through them to get to others). These measures assume communication flow along the shortest path. You can gauge centrality in a more complex fashion by looking at all the routes information can take and weighting them.
Chidambaram and Bostrom (1997) reviewed the literature and suggested that a well-developed group is cohesive, manages conflict effectively, balances tasks and socio-emotional needs. A well developed group may be judged by its outcomes.
Groups do not emerge fully developed. They begin their association, develop, experience crises, attend to deadlines, execute their tasks and conclude their association. They get to know each other and their technologies over time, learning how to interact with each other and how to use technologies in an appropriate manner. They develop, defining and redefining their network structures.