On 15 April, the LACE project held a one-day briefing and workshop in Brussels on Policies for Educational Data Mining and Learning Analytics. Originally planned to take place in the European Parliament, a security alert required a move to the nearby Thon Hotel.
The day began with a welcome from Julie Ward, MEP for the North West of England and member of the Culture and Education Committee. She was followed by Robert Madelin (Director-General of DG Connect) and Dragan Gašević (president-elect of SoLAR). Their talks were followed by overviews of the current European-funded learning analytics projects: LACE, Lea’s Box, PELARS and WatchMe.
During the afternoon discussion and review session, participants from across Europe worked together in three separate discussion groups to review specific issues related to the use of learning analytics in schools, universities and workplace training.
I worked as rapporteur in the universities workshop (pictured), which had 186 participants, including people from England, Estonia, Germany, the Netherlands, Norway, Scotland and Sweden. Our policy recommendations included:
- Privacy and ethical issues are important. Encourage institutions to develop policies covering privacy, ethics and data protection. However, this is a broader issue than educational policy making and legislation. We should aim to influence the wider debate.
- Guard against data degradation – develop and make available methods of retaining data over time
- Develop data standards and encourage their use so that we have standardisation of data
- Address the problem of over-claiming and mis-selling by vendors – institutions do not necessarily have access to the expertise that allow them to interpret and assess these claims
- Need to identify procedure for due diligence around intervention strategies, the competencies do staff need, and certification opportunities relating to these
- Identify requirements for data collection, and structures for doing this on a sector or national basis
- Support the development of standard datasets at national or international level, against which other data can be compared to see if performance is above or below the norm
- Identify behaviours in the field of education that regional or national governments should support and encourage
- Identify ways of preventing the providers of educational tools selling our own data back to us.
- Take into account that it is not just the data we are concerned about, because once it is removed from its context it does not necessarily make sense. Data needs to be associated with metadata that is produced using standardised conventions