From Tanya Elias' (2011) article made available at http://learninganalytics.net/LearningAnalyticsDefinitionsProcessesPotential.pdf, I have some ideas summarised here...
Apart from the 'relatives' academic analytics has with other areas (i.e. business analytics, learning analytics, web analytics, to name a few), academic analytics consist of 5 steps, which aligned with the knowledge continuum of data-information-knowledge-wisdom:
- Capture - capture of meaningless data
- Report - the data being reported as information
- Predict - enabling of predictions based on knowledge and wise action
- Act - the action results from the predictions
- Refine - self-improvement project where monitoring the impact of the project is a continual effort; statistical models should be updated on a regular basis
One quote that caught my eye is this, which differentiate between academic analytics and normal data mining:
"If we do not re-present actions to the crowd through an interface that affects similaractions, it is just data mining for some other purpose. This is not a knowledgediscovery cycle."
From the analysis done on models and frameworks available, 7 processes related to learning analytics have emerged:
- Aggregate & Report
Check out the following table that shows the relation among knowledge continuum (DIKW), 5 steps of analytics and the rest:
Elias also shares some information on SNA (Social Network Analysis). This is an example layout of discussion forum posts and replies in a learning management system, and the same discussion as a network diagram using SNAPP (http://research.uow.edu.au/learningnetworks/seeing/snapp/index.html):
What the 'network' can tell us would be very diverse, and it is all up to us and our objectives to use it wisely in terms of analytics.
This article ends with a compilation of the discussed items in a model/framework, shown below:
Hope to find learning analytics of good use in my research area of interest,
- Shazz @ LAK
15 Jan 2011