This was an interesting talk, so I thought it would be useful to jot down some key points:
Drawing on work from across the social sciences, especially sociology, Williamson is arguing that the emergence of Big Data (as a vision, practice and epistemology) is leading to new forms of imagining of the university – especially as ‘smart’ institutions capable of analysing (via algorithms) vast quantities of data that can be used to make intelligent decisions.
In relation to teaching, he describes an emerging vision of educational spaces as ‘cognitive-based classrooms’ – in other words, classrooms that will learn the habits and learning behaviours of students and use this information to aid the learning process. Examples of this approach could include:
- Interactive e-books
- Predictive reading lists and exercises based on past behaviour (‘pedagogical playlists’)
- Assessment techniques based on students’ click pathways
- ‘micro targeting’ of students based on data mining
As the vision of Big Data continues to grow, Williamson argues, it is likely that the complex network of algorithms that have developed in recent years (the products of a ‘hundred hands’) will grow in influence within the academy.
Williamson draws on sociological critiques to highlight the neo-positivist epistemological stance of Big Data (e.g. Kitch 2014). In so doing, he highlights the limitations of the Big Data ideals, including the role of sampling, selection and human interpretation in both the design and execution of Big Data projects. Drawing on Jurgenson (2014) he also locates Big Data within a ‘political economy’ approach highlighting the role of broader social, political and economic power relationships.
Williamson concludes by arguing that these new technological systems (algorithms) are increasing influencing the ways in which academics ‘Do’ academic practice, with an increasing emphasis being placed on demonstrable impact, social media presence and entrepreneurial spirit.
This is a useful summary of current thinking on the relationships between the ideals of Big Data, the technology of algorithms and the practice of education. It highlights the importance of viewing algorithms within a broader contexts of social, political and economic relationships.
It could well be argued that technology has always played an agential role in shaping academic press – from the printing press reducing the importance of lectures, to the rise of journal houses and the imposition of fitting research ‘outputs’ into house styles and journal requirements. Thus, it is important to see Big Data as part of a continuum in the evolution of academic labour (one increasingly characterised by market-based ideologies and pedagogical practices), as opposed to something distinctly new. This is an aspect that Williamson does not address.
Also, I’m not sure I found the reference to Lyotard 1979 particularly convincing. Academics have been writing about the potential for mechanisation to impact on human labour relations for many years, prior to the advent of post-modern thinking. Marx, for example, described mechanisation and its impact on labour praxis at length in Das Capital. It’s important, therefore, to recognise how the ways in which we (critically) approach Big Data is likely influenced by our implicit preferences towards analytical frameworks; frameworks that are selected as much due to ‘academic lifestyle’ choices as they are to their analytical potential