Skip to main content

Table 4 Critical issues and challenges for learning analytics

From: Learning analytics in higher education: a preponderance of analytics but very little learning?

Category Description Papers
1. Students Learning design and the extent to which technologies may interfere in students’ learning (their autonomy, engagement, the ways in which they invest their time and effort and their progress in their learning) (Bodily and Verbert 2017; Clow 2013; Jones and McCoy 2019
Perrotta and Williamson 2018; Selwyn 2015; Wintrup 2017)
2. Teachers Lack of alignment between teachers’ pedagogical activities and LA. Also, the detachment between teachers and those responsible for LA (managers and administrators) (Rojas-Castro 2017; Scheffel et al. 2014; Selwyn 2015;)
3. Educational theories Lack of educational and pedagogical theories underpinning LA (Avello and Duart 2016 ; Clow 2013; Perrotta and Williamson 2018; Rambe and Moeti 2017; Schwendimann et al. 2016)
4. Use of methods and data analysis Use of highly technical mathematical models and quantitative techniques that include irrelevant attributes. Also, that the management of such large data sets is unduly time-consuming. Also, concerns about the ‘neutrality’ of data collection and techniques of analysis and the ways in which certain methods produce data which might affect results and have an impact on students’ learning. Finally, a concern about whether the methods actually measure learning (Bodily and Verbert 2017; Clow 2013; Dawson and Siemens 2014; Johanes and Thille 2019; Jones and McCoy 2019; Perrotta and Williamson 2018; Prinsloo 2019; Selwyn 2015; Urbina and De la Calleja 2017; Williamson 2019; Wintrup 2017)
5. Research results Diverse concerns about the results produced by LA. For example, the reduction of the complexities of learning into data; the lack of consideration of other learning factors or the broader context that cannot be measured; the loss of subjectivity and other factors involved in learning processes; the non-regulated cross-border use of data; and the ‘ecological validity’ of data (Dawson and Siemens 2014; Jones and McCoy 2019 ; Khalil et. al. 2018; Perrotta and Williamson 2018; Selwyn 2015; Timmis et al. 2016; Watson et al. 2017; Williamson 2019)
6. Data governance Ways in which data are managed and used at micro (classroom), institutional and macro (national policies) levels so as to improve teaching and learning. Also, the lack of understanding about what to do with or how to use data. Also, a ‘managerialist’ approach to LA. (Johanes and Thille 2019; Perrotta and Williamson 2018; Selwyn 2015; Williamson 2019; Wintrup 2017)
7. Ethical issues Issues of privacy, confidentiality, informed consent, surveillance, and labelling students at risk (Bodily and Verbert 2017; Johanes and Thille 2019; Khalil et al. 2018; Pardo and Siemens 2014; Scheffel et al. 2014; Selwyn 2015; Timmis et al. 2016;
Williamson 2019; Wintrup 2017)
8. Structural factors Structural concerns: commercial use of data or business-like practices; material conditions (technology) in using LA, especially considering countries with less-developed economies; a heightening of accountability processes; increasing competition among institutions; promotion of social inequalities and other exclusionary practices (for example, MOOCs promoted by Western universities in poor countries). Also, financial, political, philosophical, epistemological and technical-mathematical aspects being characteristically absent (Johanes and Thille 2019; Perrotta and Williamson 2018; Rambe and Moeti. 2017; Selwyn 2015; Williamson 2019)
  1. Own source