Education data: Under-minded or under-used?
It is no secret that copious amounts of data are captured daily, in both known and unknown contexts. The act of publishing this post provides me with access to information about the people, places, industries and interest areas of those reading it. These data-driven insights form valuable tools for writers and content producers to target outputs and improve upon future iterations of their work. On a large scale data are mined or used in algorithms aligned to perceived personal interest, behaviours and patterns to form ‘predictions’ about a user that morph into marketing, monetisation and big-data monopolies. More and more in the social sphere data transactions and exchanges of personal information have begun to make up our routine interactions with various digital media. Similarly, in the education space these data are beginning to take shape, compromising of assessment and behavioural data that inform learning analytics and artificial intelligence softwares. However, as trends associated with ‘dataism’ begin to rise in the education sector, small companies, not-for-profits and even schools continue to focus resources on gaining access to and outputting the data – through surveys, lists and reporting means – paying less attention on translating the data into meaningful outputs for teaching and learning. This leads one to ask the question, in the education space, is the data under-mined, or is it simply under-used?
To answer this question, let’s consider social attitudes toward data mining. Whilst there are arguably reservations surrounding issues of privacy and ethics, there also exists a growing level of complicity toward exchanging data online. As van Dijck (2013) outlines naïve sentiments of ‘trust’ are deeply embedded in the ‘ideology of dataism’, exhibited via ‘characteristics of a widespread belief in the objective quantification and potential tracking of all kinds of human behaviour and sociality through online media technologies’ (p. 198). In the face of such realities, users begin to assume the occurrence of tailored, personalised digital experiences and ‘trust’ this will be the case.
Beer and Burrows term this popular cultural shift ‘presumption’ (2013, p. 47), a notion based on the future casting of a person’s, or set of peoples, behaviors, attitudes, actions and beliefs. As early as the 1970s, futurist Alvin Toffler foreshadowed shifts in industrial models of production and consumption, coining the term the ‘prosumer’ (Toffler, 1971), whereby consumers play a role in the production process. In the instance of data mining, the consumer unwillingly subjects themselves to the shared act of data creation. Every post you click on, email list you register to, video you search for and word your write.What are the implications of such predictive analytics in education? Will learners ‘presume’ the appearance of tailored content or lessons based on their previous learning and possibly even social interactions?
In a world of systems based thinking, such a practice is known as data-visualisation, and is commonly undertaken in large-scale content houses including Netflix, Amazon and Apple. Popular cultural myths note that Netflix employees are known for leaning upon existing data visualisation tools to tweak algorithms, garner new customer insights, and solve a number of business problems (Simon, 2014, n.p). Armed with tool of data visualisation, Netflix has shifted its position from content aggregator and distributor to content creator. As Maher (2015) surmises, with reference to the popular political web drama series House of Cards, the Netflix strategy is ‘firmly driven by its data – which showed that its subscribers had a voracious appetite for content directed by David Fincher and starring Kevin Spacey’ (Maher, n.p). In the case of House of Cards, colliding a data-driven lens with a pre-existing formula (the BBC version of House of Cards) and a flare for creative producing has been the success to the strategy behind the award-winning model. According to popular belief, Netflicks was so confident that it fitted its predictive model for the ‘perfect TV show’ that it moved away from standard pilot series conventions to immediately commission two additional seasons to maintain ratings and momentum (Maher, 2015, n.p). In my opinion, both this execution and the content output exude creativity, critical thinking, collaboration and imaginative practice.
So why aren’t such practices being harnessed more readily within the education sector? Not just by the tech giants but by the actors and agents within this space such as principals, teachers and learners? While increasing amounts of content are being output within and around the educational milieu and copious amounts of data are generated in the education space, arguably such trends stem from and are therefore best situated within the cultural sphere. Current trends in data mining and learning analytics are only beginning to scrape the surface revealing alternative futures associated with how digital data might be used to shape artificially intelligent, interactive and multimodal learning materials. So the question becomes, given the vast amounts of data in the education space, who will ultimately ‘use’ the data and what will be the ‘value’ in doing so?
Arguably, much of the ‘value’ of Tim Burners-Lee’s vision for a World Wide Web was not captured by the initial realisation of the Internet. Rather, following the crash of the ‘dot com boom’ the value of the social interactions made possible by web 2.0 was harnessed by companies like Google, Apple, Twitter, and Facebook (O’Reilly, 2012, n.p). With web 2.0 came the inflation of ‘datafication’, the ability to capture and transform online activity and interaction into quantifiable data that allows for ‘real-time analysis and future predictions of social action’ (Mayer-Schoenberger & Cukier, 2013). Although the ever-increasing amount of metadata captured via social media and communication platforms have the capacity to track human behaviors (Mayer-Schoenberger & Cukier, 2013, p. 30), the significance of this computational process in education is presently undermined.
“Whatever companies are the equivalent of Pearson and Kaplan in 2065 will be running schools, and we will not think twice about it.” (Selwyn, 2016, p. 8)
At present, the current forerunners in the ‘digital data race’ in the education sector are predominantly information and communication technology (ICT) companies. Education content giant Pearson, IMB and Knewton speed ahead of any institution based equivalent. Mimicking policy-based structures aligned to the OECD, Pearson’s ‘Learning Curve’ is a ‘predominantly statistical project’ concerned with performance-based monitoring of ‘fluctuations, risks and country comparisons’ (Williamson, 2016, p. 132). The IBM Smarter Education (2014) program taps into the multidimensional possibilities of analytics to track results and measure performance, as well as enabling educational leaders to ‘detect’ pre-existing patterns housed within vast pools of data to ‘project potential outcomes and make intelligent decisions based on those projections’ (IBM Smarter Education, 2014 in Williamson, 2015, p.5I). Machine-based learning analytics platforms such as Knewton are not only programmed to predict the probable future progress of a learner but consider educators and learners as both ‘consumers’ and ‘producers’, interacting with the data ‘co-creatively’ (Williamson, 2016). Such advances give validation to Neil Selwyn’s insightful statement that, “Whatever companies are the equivalent of Pearson and Kaplan in 2065 will be running schools, and we will not think twice about it” (Selwyn, 2016, p. 8).
Whilst this may be the case, it is worth considering what the content outputs that result in the afore mentioned data-mining activities might look like, the role that learners might play in co-creating both the content and the data sets, as well as how educational institutions might place in the ‘digital data race’. As the permeations and outputs of digital data in the education space continues to evolve and expand, the ‘value’ of this data remains in a continual state of flux. It would appear that as yet the sector are not informed enough of the possibilities for data to impact positively upon teaching and learning, and therefore the ‘value’ of the data remains untapped within the education sector itself. For this reason, there is increasing benefit in encouraging and inspiring education leaders, teachers and learners about the implications, insights and innovations made possible by data shaping and visualisation.
Beer, D. & Burrows, R. (2013). Popular Culture, Digital Archives and the New Social Life of Data, Theory, Culture & Society, 30 (4), 47–71
Maher, B. (2015). Big Data: How Netflix Uses It to Drive Business Success. Smart Data Collective. Retrieved from http://www.smartdatacollective.com/bernardmarr/312146/big-data-how-netflix-uses-it-drive-business-success
Selwyn, N. (2016a). Education and technology: Key issues and debates (2nd Edition), London: Bloomsbury.
Simon, P. (2014). The Visual Organization: Data Visualization, Big Data, and the Quest for Better Decisions. Wiley.
Toffler, A. (1971). Future Shock, Pan, London.
van Dijck, J. (2014). Datafication, dataism and dataveillance: Big Data between scientific paradigm and ideology. Surveillance & Society 12 (2), 197-208
Williamson, B. (2016). Digital education governance: data visualization,predictive analytics, and ‘real-time’ policy instruments, Journal of Education Policy, 31(2), 123-141. doi: 10.1080/02680939.2015.1035758