[
]1On
Thursday & Friday I was in Manchester for a 2 day HEA summit on big data
in the social sciences. The event focussed on how social scientists
could engage with big data, and the implications of that for teaching
and learning (and capacity building). This of course can be seen in the
context of claims (e.g. by Chris Anderson) of ‘the end of theory’ in the
age of big data and to some extent an increasing pressure for all to
engage in quantitative analysis, and of course an increasing
availability of tools (and data sources) to conduct such analysis. We
kicked off with Rebecca Eynon giving reminders of some of the pitfalls
of big data and a naivity around regarding ‘big data’ as ‘complete data’
– which reminded me of the provocations for big data (and my
[provocations for learning analytics post)]2. Next up we had [Jon
Whittle]3 from Lancaster, talking about co-design and community
engagement through data, with some nice examples of data projects to
save/make money including ‘[Clasp]4‘, a stress-ball (or other
personal item) type device for those on the autistic-spectrum to monitor
anxiety through phone-sync and (if setup) contact someone in a support
network. I first saw this sort of approach in LSE’s ‘[mappiness]5‘
app, and while I think some of them are a bit gimicky and I’m not sure
what you’d do with the data I still like the idea and think event
sampling, etc. has huge potential in mobile apps. Jon also noted the
approach they were taking: [PROTEE]6, “a methodology that aims to
increase the reflexivity of research and innovation projects by helping
to sensitise practitioners to the demarcations their projects enact, and
to think through how these may affect the relevance of the outcomes”,
which I think is very salient for current learning analytics research.
Third on was [Ali Fisher]7 talking about work on counter-terrorism
and data analysis (through the eyes of a historian). Two key messages
from Ali’s presentation were 1) the desire to get people to the stage of
asking the most important question in data analysis: “Is it possible to
[…]?”, and 2) that the only way we can hope to make sense of the
deluge of information is by understanding the meaning making signs –
semiotics. Another interesting point was re: physical, versus digital
location – we might be physically separate from a location, but by
digital means still be a “digital insider” (e.g., if we’re tweeting
outside of an event, but hugely connected to on the ground happenings).
The second day we had more discussion, with some interesting discussion
of tools and resources including the [UK Data Service]8, which
holds social science datasets for research and teaching purposes, and a
set of tools and resources [Farida Vis]9 uses on her Sheffield
iSchool/Journalism MA course in data analysis, for example: . I also did
my talk on the second day, based on the update of the LAK13 paper for
the Journal of Learning Analytics ([here]10). Having not thought
I’d get to ‘return’ to that presentation post-conference, it was nice to
think about how to update it and use it as a partial ‘introduction to
learning analytics’. I’m also incredibly grateful to the reviewer who
suggested the ‘interpretive flexibility’ paper, as we made much greater
use of this notion in the journal version, and it is obvious that
policy/practice context of technology use is absolutely fundamental
(e.g. mini-whiteboards for formative assessment and collaboration, v.
traditional whiteboards for chalk n talk). The HEA is interested in big
data in the context of curriculum design, innovative pedagogy, and
student and staff transitions. And this focus is interesting for big
data, given the difficulties, for example of writing for multiple genres
of journal, and being a computer scientist doing social science or vice
versa. This is something I’ve thought a bit about around the narratives
of ‘[what am I doing a PhD in?]11‘ and the issue in my subject of
‘what does a learning analytics PhD _look like?’ _is an interesting
one. This is particularly problematic given the need for
interdisciplinary (not just multi-disciplinary) support, and the ways in
which the REF mitigates against this. For students too, this is an
issue: How do we a) support, and b) assess, interdisciplinary work –
particularly so that we’re fair in comparison to students with only a
single skill set/discipline? These aren’t easy questions to answer or
solve, but certainly consideration of departmental and funding
structures – and the support of subject centre organisations – play a
role in moving things forwards.
