Talk:Learning analytics

Pointer to 'general introduction' at start of article?

I've added a "general introduction" link to the start of the article: "For a general audience introduction, see the UNESCO "Learning Analytics Policy Brief" (2012)[2]". IF anyone thinks it's not appropriate in that position, please move it/discuss here. I should point out it was written by one of my supervisors. It's the most recent, comprehensive, and general purpose paper on it so I don't see a COI - but as more literature is released it may want relocating. Sjgknight (talk) 12:42, 22 January 2013 (UTC)

External links

Many of the 'external links' look like they ought to be cited in text and added as references rather than external links - I'll try to address this but if anyone else is looking and has time this should be a fairly easy task. Sjgknight (talk) 13:57, 22 January 2013 (UTC)

After seeing another link appear in that list which ought to be an in-text citation I've removed a load. I understand people may be trying to avoid citing their own work in text - and this is good practice - but listing articles which aren't general audience articles or collections, etc. in the external links is confusing and detracts from other work there. If anyone is unsure about how to integrate work into the text, or they're citing things where there is potential for COI or NPOV issues then please list them on the talk page. The list I've just removed is below - if you integrate them in to the text please delete them from here.

• Goldstein, P. J., & Katz, R. N. Academic analytics: The uses of management information and technology in higher education. 2005
• IBM Business Analytics
• Campbell, J. & Oblinger, D. Academic analytics. Educause 2007.
• Norris, D., Baer, L., & Offerman, M. A National Agenda for Action Analytics. National Symposium on Action Analytics 2009.
• Adali, S., et al. Measuring Behavioral Trust in Social Networks. IEEE International Conference 2010.
• Romero, D., et al. Influence and Passivity in Social Media. Social Science Research Network, 2010.
• Graber, C. Behavior Influenced More In Denser Networks. Scientific American, 2010 (Includes audio recording).
• Carmean, C. & Mizzi, P. The Case for Nudge Analytics Educause 2010.
• Jovanović, et al. Using Semantic Web Technologies for the Analysis of Learning Content IEEE Internet Computing, Vol. 11, No. 5, 2007, pp. 45–53
• Ali, L., Hatala, M. Gašević, D., Jovanović, J., A Qualitative Evaluation of Evolution of a Learning Analytics Tool, Computers & Education, Vol. 58, No. 1, 2012, pp. 470–489
• Ali, L., Asadi, M., Gašević, D., Jovanović, J., Hatala, M. Factors Influencing Beliefs for Adoption of a Learning Analytics Tool: An Empirical Study, Computers & Education, Vol. 62, 2013, Pages 130-148.
• Duval, E. Attention please! Learning analytics for visualization and recommendation, Proceedings of LAK11: 1st International Conference on Learning Analytics and Knowledge, Banff, Canada, 27 February - 1 March 2011
• Richards, G., Measuring Engagement: Learning Analytics in Online Learning, Proceedings of Electronic Kazan 2011, Russian Federation. 26 April 2011.

Sjgknight (talk) 17:17, 24 January 2013 (UTC)

Purposes/functions of analytics - new section?

This section could explain some of the differences between EDM, LA and academic analytics before moving on to some of the ways it is being used. A possible sub section could be on analytics for understanding research (E.g. the CETIS Analytics Series Vol 1. No 4. Analytics for Understanding Research, Mark van Harmelen Hedtek Ltd, Serial Number: ISSN 2051-9214 Vol 1, No 4, ISSN: 2051-9214, which is CC-BY licensed too). Sjgknight (talk) 18:28, 8 February 2013 (UTC)

I added a differentiation between LA and EDM in the intro section, but am open to having it moved if there are objections.

Cab938 (talk) 13:07, 9 April 2013 (UTC)

Looks good to me User:Cab938. I wrote some stuff on another wiki (http://www.edfutures.net/index.php?title=Learning_Analytics#Learning_Analytics_and_Educational_Data_Mining_.E2.80.93_What.E2.80.99s_the_Difference.3F) which might be transferable (CC-By licensed, and it was summaries of research) - I've added some bits from that article anyway so slightly reluctant to add it myself, but it may be useful to just straight transfer the below or adapt it to this article? I'm not sure what emphasis to place on this distinction either - I'd hate to see the rest of the article swamped by excessive coverage of this bit!

Many researchers in both EDM and LA do work which crosses between the two. Ferguson summarised it thus:

• Educational data mining was primarily focused on the technical challenge: How can we extract value from these big sets of learning-related data?
• Learning analytics was primarily focused on the educational challenge: How can we optimise opportunities for learning?
• Academic (and action) analytics were focused on the political/economic challenge: How can we substantially improve learning opportunities and educational results at national or international levels?

While Siemens and Baker highlight some key differences between LA and EDM as in the table.

	LA	EDM
Discovery	The end goal is to facilitate human judgement, automated models play a part in helping humans make judgements	The end goal is automated models, humans play a part in setting these up
Reduction & Holism	“Stronger emphasis on understanding systems as wholes, in their full complexity”	“Stronger emphasis on reducing to components and analysing individual components and relationships between them”
Origins	Stronger origins in understanding learning materials and learners using computational methods	Stronger origins in modelling outcomes, and software to track this
Adaptation & Personalization	“Greater focus on informing and empowering instructors and learners”	“Greater focus on automated adaption (e.g. by the computer with no human in the loop)”
Techniques & Models	Social network analysis, sentiment analysis, influence analytics, discourse analysis, learner success prediction, concept analysis, sensemaking models	Classification, clustering, Bayesian modelling, relationship mining, discovery models, visualization

Table adapted from

Some 'to do' in each section

See also the other comments, but this is just some obvious things which ought to be addressed. Someone might find it useful as a starting point. Mark/discuss as you edit.

• Introduction Needs expanding, list better placed in "External Links" & references but referred to in text.
• What is Learning Analytics? I think a simple, broad definition from a good soruce would be better here
• History; Context There is more history than is given here (although this summary is probably ok), a link to EDM might be good. Incorporating 'tools' also useful.
• History; Techniques No links or citations, unclear relationship to LA. Might be better in lower section?
• Criticism The definition debate is wider than this, it should perhaps be moved to this section (from the first section) and expanded. The ethics issues also need expanding and I'm sure there are recent sources as well as other concerns re: big data, quantification, etc.
• Methods - Could do with some general intro to methods. Also methods section is a slightly odd mixture of things which could perhaps be related better.
• Methods; The particular 'methods' need expanding and citations adding. Linking to particular software, or examples of use might also be useful
• Software Expand
• See also Expand
• External links This is too much of a list. Some of these things look like they belong as references to the main text, others could probably be removed. This would be a good place for general articles and discussion areas (inc. SoLAR) but these are swamped at the moment

Sjgknight (talk) 14:05, 5 March 2013 (UTC)

New Section - Ethics

There is a rich discussion around the ethics of leraning analytics. In part this has been collected by George Siemens, etc. from discussion group posts, but there are also good articles including: Learning analytics: ethical issues and dilemmas. Slade, Sharon and Prinsloo, Paul (2013). Learning analytics: ethical issues and dilemmas. American Behavioral Scientist (In press). http://oro.open.ac.uk/36594/ Which gives 6 principles which could easily be brought in. Paralleling boyd & Crawford's big data provocations I wrote http://people.kmi.open.ac.uk/knight/2013/01/six-provocations-for-learning-analytics/ it may not be useful (and I'm certainly not going to add it myself) but there may be something there or a useful parallel to draw with the original boyd & Crawford article. I'll try and a) find some more good sources and b) begin a section skeleton asap. — Preceding unsigned comment added by Sjgknight (talk • contribs) 17:43, 15 March 2013 (UTC) Sjgknight (talk) 17:46, 15 March 2013 (UTC)