Why and how has the policy of learning analytics been built?
Forms and opportunities for learning in universities are becoming diverse. At the same time, the amount of data collected from students’ activities and teaching increases and diversifies. By analysing the data, one can better understand the learning opportunities as well as the bottlenecks. Learning analytics is a fast-growing area that refers to the collection, measurement, analysis, and reporting of information accumulated about a learner with the purpose of understanding and optimizing learning and learning environments (Siemens 2013).
The key objectives of AnalyticsAI project are to provide tools to support learning analytics practices. One of those tools is the policy of learning analytics, which contains carefully weighed, transparent and accepted principles, guidelines and decisions that ensure the meaningful use of analytics. The policy is a multifaceted tool that must consider the strategic, pedagogical, ethical and legal, as well as technical and data-related aspects of learning and teaching. In essence, the policy of learning analytics supports students and staff in the comprehensible, consistent and responsible utilisation of learning analytics. AnalyticsAI policy work has been prepared under the leadership of Aalto University, whose groundwork and policy document will serve all project stakeholders. The policy document processed in Aalto in a first phase will be processed in the next phase tailored for the specific needs of other universities.
In Aalto University, the policy assignment was validated by the Learning Steering Group and the policy working group set up for this purpose consisted of learning services (learning and teaching services & processes, pedagogy and responsibility for learning analytics), management information services (reporting and data maintenance, development and management), IT services (development of learning and teaching systems and services, development and management of data and university analytics), academic legal services (data protection and other legal matters), and teachers' and students' representatives. This representativeness has proven to work. The policy must support the university's many service processes and operating practices, including those under development, which is why the policy development must be closely linked to the university's various areas of expertise.
The policy alignment work has progressed in such a way that we first specified, based on an extensive international literature review, the principles of learning analytics that we want to follow. This served as a basis for the actual themes of the policy, through which the perspectives on the use of learning analytics were refined. Key examples have been the SHEILA Project Network’s R.O.M.A. method (Rapid Outcome Mapping System), and in particular the guidelines and themes developed by the JISC and the ORLA communities. We have taken advantage of several international policies, of which a comprehensive list can be found from here.
The following guiding principles were selected as regarding the use of learning analytics:
- Transparency of learning analytics objectives and practices and the right to influence the processing of one's own personal data: the collection and use of learning analytics data, its sharing and the ethical use of data are based on transparent criteria and decisions about the benefits and uses of learning analytics;
- University values and strategy as a basis for learning analytics: The use and development of learning analytics is guided by the university's values and strategy;
- Impartiality: learning analytics aims to understand the needs of diverse groups of students and provide them with support and guidance in a proactive and timely manner;
- Improving quality for different stakeholders: Students can use learning analytics to streamline their studies; teaching staff to evaluate and develop teaching; leaders of degree programs as well as university management to support leadership and to improve the quality of teaching;
- Furthering a positive learning experience: learning analytics provides content and pathways for the student to support his/her own personal plan and well-being;
- Personal support and feedback: learning analytics can be used to identify students' learning needs and provide personal support;
- Learning Analytics with the help of teacher and tutor support: we understand that the use of learning analytics provides only a partial picture of students’ performance, activity, well-being, and other factors. Therefore, support measures based on the results of learning analytics are the product of human decision making. Learning analytics complements forms of face-to-face and web-based interaction;
- Critical review of data and algorithms: we recognize that data and algorithms may be garbled. We work systematically to correct any incomplete data, erroneous algorithms, as well as inferences and impacts;
- User-centric development of learning analytics: the development and use of learning analytics is based on the user needs perspective of different groups of actors at the university;
- Digital skills development: the use of learning analytics supports students' and staff's understanding and ability to function in digital environments.
Themes of learning analytics policy:
- Areas of learning analytics and liability issues;
- Data protection principles in learning analytics;
- Ensuring the validity of learning analytics data and results;
- Access to analytics' results and data;
- Justifying and enabling positive interventions;
- Identifying and considering the detrimental effects of learning analytics;
Learning analytics deployment is on its way to universities. Policy guidelines are an essential tool in conducting this action. The Policy is also a living document that needs to be refined through new opportunities, and based on the accumulation of experience and analytics.
JISC, UK (2015). Code of Practice for Learning Analytics. (Noudettu 18.10.2019)
Sheila-project, Supporting Higher Education to Intgrate Learning Analytics. (Noudettu 18.10.2019) https://sheilaproject.eu/
Siemens, G. (2013). Learning Analytics: The Emergence of a Discipline. American Behavioral Scientist, 57(10), 1380–1400.