Norquist. Photograph available here
Based on Learning analytics: drivers, developments and challenges (Ferguson, 2012)
Learning analytics is a new field that has emerged in the last decade with roots in business intelligence, web analytics, educational data mining and recommender systems.
The goals of what can be achieved and how these goals will be achieved still has to be defined.
Learning analytics are different from other related fields of academic analysis and Educational Data Mining (EDM)
There are a number of definitions of learning analytics. The current prevalent definition was set out in a call for papers for the first international Conference on Learning Analytics and Knowledge (LAK2011)
” Learning analytics is the measurement, collection, analysis and reporting of data about learners and their contexts, for purposes of understanding and optimising learning and the environments in which it occurs”
There are a number of factors driving the development of learning analytics
- How business uses analytics to extract value from data sets (big data) to drive recommendation engines, identify patterns of behaviour and develop advertising campaigns
- Widespread introduction of LMSs is creating larger data sets. This data is being collated but the reporting and visualisation of this has been largely non-existent.
- Online learning take-up has increased
- Increasing demand for educational institutions to measure and demonstrate improved performance
- Emergence of different interest groups; government, educational institutes and teachers/learners
A bit of history
- 1979: Open University could reflect on 10 years monitoring the progress of distance students course by course
- 1999: it was slowly becoming clear that collaborative online learning could take place (Dillenbourg, 1999)
- 2000: EDM (Educational Data Mining) begins to emerge from the analysis of student-computer interaction with a strong emphasis on learning and teaching. In 2007 Romero and Ventura defined the goal of EDM as ‘turning learners into effective better learners’
- 2001: Second generation web opened up new ways of collecting web content from various sources, processing it and exchanging the results with other programmes (Berners-Lee et al., 2001)
- In contrast the early use of the term learning analytics referred to business intelligence about e-learning (Mitchell and Costello, 2000)
- 2003 onwards: socially and pedagogical approaches to analytics began to emerge. Social Network Analysis (SNA) was a significant development. SNA is the investigation of networks and can be used to ‘investigate and promte collaborative and co-operative connections between learners, tutors and resources helping them to extend and develop their capabilities’
- 2008: Pedagogic theory starts to emerge strongly as an approach to optimising and understanding learning.
Political and economic drivers
Measuring of the quality of education to meet the challenge of declining education standards principally in the USA. ‘Academic analytics’ began to evolve to link data sets with improved educational decision making.
The field is rapidly expanding
In 2008 analytics and EDM split.
Analytical tools are rapidly developing and enabling different types of analysis e.g. LOCO-Analyst which provides feedback focused on the quality of the learning process.
With tools becoming more powerful ethics and privacy issues begin to emerge
In 2010 the field of analytics splits again with learning analytics gradually breaking away from academic analytics. Siemens presented the first early definition in 2010 which was refined and has become the current prevalent definition as described earlier in this blog.
- EDM focused on the technical challenge
- Learning analytics focused on the educational challenge (optimising opportunities for learning online)
- Academic analytics focused on the political/educational challenge
Overlaps between them still remain though there have been further attempts to distinguish between them. Long and Siemens (2011)
In 2012 learning analytics were identified as a technology to watch in the NCM Horizon Report.
New tools such as GRAPPLE can now extract data from across an entire PLE
Learning analytics are distinguished by their concern for providing value to learners and employed to optimise both learning from and in the environments which it takes place
Ferguson, R. (2012) ‘Learning analytics: drivers, developments and challenges’, International Journal of Technology Enhanced Learning (IJTEL), vol. 4, nos. 5/6, pp. 304–17; also available online athttp://oro.open.ac.uk/ 36374/ (accessed 6 July 2016).