Learning Analytics

Educational Data Minding (EDM) has been defined as “an emerging discipline, concerned with developing methods for exploring the unique types of data that come from educational settings, and using those methods to better understand students, and the settings which they learn in.” (International Society of EDM).

An extensive overview of EDM applications, developments and definitions has been given by Baker and Yacef (2009). EDM has its origins in multiple research areas such as statistics, data mining, machine learning, visualization and computational modelling aiming to automatically discover patterns and models from huge and extensively growing datasets. While in the beginnings of EDM, most data was retrieved from experimental learning sets not lasting longer than a couple of weeks, nowadays such data is often tracked over the duration of a whole course lasting up to one study year. Collected data is further analysed to gain valuable insights on learning processes and aspects. With  these enormous  loads of data  new challenges arise

especially in visualizing and modelling the information in a manner still readable and interpretable for human stakeholders. While EDM aims to discover patterns and models on scaled data, Learning Analytics (LA) further includes the needs of different educational stakeholders and the strength of their judgement in addition to computational measurements. The focus areas of both communities (EDM and LA) seem to be slightly different, although they follow similar goals that encompass the improvement of educational technology and the evaluation of pedagogically sound instructional designs (Baker et al., 2012).

Learning Analytics especially emphasises on supporting pedagogical approaches by providing assistance to teachers in practically relevant questions (e.g., the quality of learning material or the engagement of students in specific exercises). Data gained by LA tools can be used to evaluate pedagogically sound instructional designs within classroom settings. In most cases this focuses mainly on monitoring learner actions and their interactions with learning tools and learning peers (Lockyer & Dawson, 2011). Many attempts go towards visualizing these learning traces in order to make significant relationships explicit or even to allow stakeholders to discover such relations independently. Research using dashboards in LAs has been described by Duval (2011). As known in other domains like for instance sports, the visualization of collected interaction training data and the comparison with of this data with data of like-minded peers, may lead to insights about unwelcome habits and better practices but also to higher motivation due to the playful introduction of competitiveness. Dychkoff et al. (2012) describe a learning analytics toolkit that takes a step forward by allowing the course instructors to define their own learning indicators via combining traced student actions. This toolkit is designed to support the instructors in optimizing their teaching strategies. An example of inferences drawn from the learning indicators about the quality of teaching material is the relation between time spent on specific learning objects and learning outcome. Also hardly touched learning objects can be assumed of poor quality with respect to the learning audience.

The future of TEL is often seen in flexible, teacher or even learner composed mash-up learning frameworks, independent of a specific environment (Chatti et al., 2010). To keep up with this development new learning analytics toolkits need to be just as flexible and adaptable as learning tools.






Baker, R. S. J. D., Duval, E., Stamper, J., Wiley, D., & Buckingham-Shum, S. (2012). Social Learning Analytics: Five Approaches. In Proceedings of the 2nd International Conference on Learning Analytics and Knowledge (20-20). New York: ACM.

Baker, R. S. J. D., & Yacef, K. (2009). The State of Educational Data Mining in 2009: A Review and Future Visions, Journal of Educational Data Mining, 1(1), 3-17.

Chatti, M. A., Agustiawan, M. R., Jarke, M., & Specht, M. (2010). Toward a Personal Learning Environment Framework. International Journal of Virtual and Personal Learning Environments, 1(4), 66-85.

Duval, E. (2011). Attention please!: learning analytics for visualization and recommendation. In Proceedings of the 1st International Conference on Learning Analytics and Knowledge (9-17). New York: ACM.

Dyckhoff, A. L., Zielke, D., Bültmann, M., Chatti, M. A., & Schroeder, U. (2012). Design and Implementation of a Learning Analytics Toolkit for Teachers. Journal of Educational Technology & Society, 15(3), 58-67.

Lockyer, L. & Dawson, S. (2011). Learning designs and learning analytics. In In Proceedings of the 1st International Conference on Learning Analytics and Knowledge (153-156). New York: ACM.


+43 316 87330636

Attribution should be by the named
authors and  "LEA's  BOX  Project"
at URL: http://www.leas-box.eu

LEA's BOX is supported by the
European Commission  under
the 7th framework programme;
grant no. 619762

© 2014 TU Graz