Abstract:

The processes related to student data acquisition, representation and processing, known as “learning analytics” (Clow, 2013) have increased in sophistication and levels of adoption over the years (Google Trends, 2023), evolving from early data-driven analyses of dropout (Tinto, 1975) and gender biases during admission (Bickel et al., 1975), to recent student engagement interventions (JISC, 2022).
Despite the seemingly wide adoption of learning analytics, there are still some barriers to responsible use, including those related to data protection and privacy laws. In Europe, the General Data Protection Regulation (GDPR) sets guidelines for the collection, storage, and use of personal data, including that of students. To overcome these barriers and adopt learning analytics, institutions must comply with GDPR and other relevant laws. This includes obtaining informed consent from students for the collection and use of their data, implementing robust security measures to protect the data, and being transparent about how the data will be used. Additionally, institutions should implement data minimization techniques, to only collect the data necessary for the specific purpose, and apply data retention policies.
To truly adhere to the spirit of the legislation, however, institutions should also adopt state-of-the-art privacy-enhancing technologies (PETs) together with their data protection impact assessments (DPIA) which can help identify and mitigate the potential risks. A responsible data governance framework and policies must be developed and applied to ensure that the data is used ethically and in an appropriate way.
The fast pace of technological development poses a challenge to legal and regulatory frameworks, which need to be agile, meaning they are sufficiently flexible and able to adapt quickly to changing circumstances. This can be done by creating laws and regulations that are broad in scope, allowing for flexibility in interpretation and implementation. Additionally, regulatory agencies can be empowered with the ability to make decisions quickly and adjust regulations as needed. The potential benefits of agile regulatory frameworks are clear, as the ability to quickly respond to new technologies, changes in the economy, or other developments can help ensure that laws and regulations remain relevant and effective in protecting the public interest.
However, there are also potential risks associated with agile regulatory frameworks. These include the potential for regulatory capture, where special interests from big tech companies are able to negatively influence the regulatory process to their advantage, to effectively maintain surveillance capitalism (Nottingham et al., 2022). Additionally, there may be concerns that a lack of regulatory predictability could create uncertainty for businesses and investors. While agile regulatory frameworks can potentially improve the effectiveness of regulation, they also require careful co-design and implementation to ensure they do not lead to unintended consequences. A multidisciplinary approach is essential to inform policy-making, as guidelines for compliance with data protection and privacy are crucial to effectively and ethically use learning analytics in education in the era of machine learning and artificial intelligence.

Keywords:

Learning analytics, student data, multidisciplinary education.