Learning analytics and assessment
Learning analytics is about collecting traces that learners leave behind and using those traces to improve learning. It is similar to analytics and big analytics draws techniques from a number of communities with long traditions (including data mining, information visualization, language processing, etc.). Yet, as a field of its own, it is still fairly recent: LAK, the leading conference in the field, is only in its fourth year.
The first of the two in-depth articles proposes a “teacher-led design inquiry of learning”. The second one sheds light on the potential cross-benefits that combining Learning Analytics and open linked data research could bring about. The four articles from the field introduce assessment in basic maths, concept mapping for reflection, analysis of the teachers’ learning process, and real-time visualization of discussion activities in a MOOC.
Opportunities for building massive data sets on learner behavior exist as a result of the widespread use of online learning materials, technology platforms and services. A narrow focus of learning analytics concerns itself with improved student retention, but the larger agenda is more related to the personalization of learning, self-regulated learning and the improvement of the overall learning experience. Moreover, a more data driven (or at least verified-by-data-analysis) approach in the learning sciences can help to take advantage of the approaches already well-established in the harder sciences. However, especially in this post-NSA world, privacy and ethical issues need to be taken into account as driving factors rather than as concerns that are only addressed post factum.
Data mining and information visualization can also provide valuable information on the benefits (or lack thereof) of learning technology applications, in order to support decisions in educational policy making.
No doubt, Learning Analytics will be one of the hottest topics in learning technology research and development in the near future.