Universities have become major centres of data collection, calculation and consumption. Some commentators envision a ‘smart university’ where data are used to conduct constant, real-time audits and assessments of the various performances and activities of all students, staff, systems, and facilities. While the smart university remains an aspiration, it highlights how HE is under pressure to become faster and more computationally intelligent. Data are now moving faster than ever through the sector, bringing about new temporalities of measurement, comparison, ranking and competition. The swift movement of data also requires new cognitive capacities within HE management, administration and strategic planning. As a result, power in the HE sector is now concentrated among a combination of data experts and computationally ‘cognitive’ machines.
The super-fast, ultra-smart, and semi-automatic university may be largely imaginary at present, but this envisaging of the future is beginning to animate real projects and programs. David Beer’s recent book The Data Gaze focuses on the ‘imaginary’ of the data analytics industry, how such an imaginary becomes embedded in technical infrastructure, and how it is then embodied in the practices of data workers. Together, the imaginary, infrastructures and practices enable the data analytics industry to ‘gaze’ on data, to see hidden patterns, and then to generate and ‘speak’ truths from the data. Two key features of the data gaze, Beer argues, are its speediness and smartness. Data analytics are perceived to reveal accelerated, real-time knowledge that is both in-depth and panoramic in scope. They also offer portals into new forms of intelligence, hidden insight and prediction that can assist and augment human decision-making, action, and other ‘smart’ practices.
The data gaze of higher education likewise emerges from a data imaginary of fast-paced smart analysis, pattern detection and insight production. The imaginary of the smart university in the UK can be traced back through nearly a decade of policy-facing texts repeatedly emphasizing the enhanced collection of data about students for purposes of comparative performance measurement and market regulation. Illustratively, the Higher Education Policy Institute (‘the UK’s only independent think tank devoted to higher education’) and Jisc (the sector’s agency for digital learning technologies) published a report on the potential of ‘big datasets’ for ‘measuring excellence or identification and resolution of areas for improvement’. The HE regulator, the Office for Students, recently produced a 3-year data strategy emphasizing its government mandate to utilize performance data, comparative, real-time and historical analysis, and data-led decision-making. Political priorities around performance measurement are now being actively encoded in the software running in many university centres.
Fast-paced systems are essential to the functioning of the smart university, and significant work is being done to create them. At a national level, sector bodies HESA (Higher Education Statistics Agency) and Jisc (the digital learning agency) are key to the development of technical data capacity in HE. HESA’s Data Futures programme is especially significant since it is building a new national data infrastructure for student data collection, storage and analysis. Initiated at the demand of government, when it is launched (estimated in 2020) Data Futures is expected to deliver more useable and timely data, more efficiency savings, and accelerate the collection of student data in order to improve students’ choice, policymakers’ decisions, and institutional competitiveness through comparative benchmarking. Jisc recently launched a national learning analytics architecture and tools to allow universities to track and visualize student performance, and even use predictive data modelling for forecasting problems and planning pre-emptive interventions. Jisc has also run an ongoing project on ‘the intelligent campus’ in order ‘to improve the student experience by capturing and analysing the many kinds of data that can be collected across university and college campuses’. The intelligent campus would combine ‘learning analytics’ data from courses along with historical student data, as well as data gathered from systems that record and monitor space and equipment usage, timetabling and other activities, in order ‘to make smarter, more effective use of learning spaces and other facilities across campus and to improve curriculum design and delivery’. The proposed integration of HESA and Jisc would therefore allow additional value to be generated from their parallel developments in national infrastructure and learning analytics architecture. These large-scale digital data infrastructure projects are already underway to construct the digital architecture of the smart university.
A significant consequence of the HE data imaginary and the infrastructural work it supports is that it is also empowering commercial companies to open up and exploit market opportunities within the sector. The new global HE data industry—consisting of global education businesses as well as newer digital platform providers—is seeking to capitalize on the ‘unbundling’, outsourcing and ‘rebundling’ of HE services to new market providers. Digital platform producers are increasingly able to plug in their products to university systems and processes at multiple levels. The parallel development of data infrastructure by sector agencies and a global HE data industry demonstrates how trends in marketization and commercialization need to be understood as hard-coded into the very operating systems of contemporary higher education. The data infrastructure of HE is being built for speed, enabling data to flow in real-time and form the basis for predictive analytics that can even anticipate future trends. As it sinks into infrastructure, the HE data imaginary is becoming materially consequential to the practices enacted within agencies and university management centres alike. The smart university is already being built and operationalized through hidden digital architecture, enabling the HE data gaze to see and inspect the sector, students, staff and institutions with unprecedented optical power and fidelity.
Besides the technical capacity afforded by infrastructure projects, the HE data gaze is also embodied in new working practices. The expansion of HE data use generates demand for new kinds of data workers and for forms of expertise to deal with analytics, visualization, interpretation and communication. The Office for Students, for example, has recruited new data strategists, managers and analysts as in-house data experts, while also catalysing other agencies and universities to ‘up-skill’ their administrators. HESA and Jisc run an annual Data Matters conference to bring HE data practitioners up to speed with the latest developments and requirements. They also co-manage a professional development program called Analytics Labs to support HE data workers to build their data skills. Generating maximum value from HE data means seeking new ways to optimize humans to perform analyses that can contribute to enhanced and accelerated decision-making.
As decision-making is increasingly augmented by fast-flowing data, new kinds of cognitive capacities and demands are being introduced into HE. As David Beer has noted in his study of data analysts, the complexity of data work involves significant hybridity with automated systems, bringing about new kinds of human-machine relations where human workers are assisted and augmented by computerized forms of intelligence. The media theorist NK Hayles has conceptualized ‘cognitive assemblages’ as combinations of human and machine intelligence. From this perspective, human cognition is increasingly interpenetrated by forms of ‘nonconscious cognition’ that work at speeds surpassing human perception and at levels of computation exceeding human comprehension. As a result, technical systems have become co-extensive with human cognition, while humans have become part of the extended nonconscious cognition of machines too.
As in the data analytics industry, HE data workers are now increasingly working in cognitive assemblages, relying on machines to support their analyses and decision-making. The management centre of the contemporary university is in the process of becoming a cyborg centre, with a data gaze that sees through one conscious human eye and one nonconscious machine lens. In this emergent cyborg setting, significant sectoral power flows to those data professionals who can adequately work alongside the machines, make interpretations from the data they produce, and communicate findings to other stakeholders. The innovation think tank Nesta has termed these human-machine hybrid experts ‘unicorns’. Unicorns are a rare and mythical super-class of data professionals simultaneously able to strategize, crunch numbers, and passionately persuade others that their eye for pattern detection can produce truthful representations of reality. Chair of the Office for Students, Sir Michael Barber, is of the unicorn variety. In a recent lecture on data and joy, he outlined a unique combination of computer-mediated enumerative expertise and deep human passion for analysis. In the cyborg management centre of HE, authority increasingly lies with these unicorns, those passionate cognitive assemblages of human and machine intelligence that are privileged through their joyful data gaze to discern and convey to others the objective ‘facts’ and ‘truths’ of higher education.
The super-fast, ultra-smart, semi-automated university is not inevitable. The HE data gaze is already under threat from emerging agendas to ‘disquantify’ higher education through alternative qualitative data and responsible metrics. Large-scale infrastructures, however, are hard to change once they have been built. The glossy idea of a smart university or intelligent campus is easy to criticize, but hard to resist once it has become the central operating system for the sector.
Ben Williamson is a Chancellor’s Fellow at the Edinburgh Futures Institute and the Centre for Research in Digital Education at the University of Edinburgh. His latest book is Big Data in Education: The digital future of learning, policy and practice (Sage, 2017). He also maintains the Code Acts in Education research blog and on Twitter he is @BenPatrickWill.