Data Power Conference
Panel Session 5b): Data in Education (Chair: Robin Sen)
Data-Driven Decision Making in the Education and the Cultural Sector: A Comparison
Franziska Florack and Abigail Gilmore, University of Manchester
The collection and use of quantitative data rules and guides the educational sector, measuring anything from expected progress of a pupil to her school’s place in the national and international league tables. Should either not reach the benchmark, immediate measures are taken to force rapid improvement. Whilst this allows for a cross-national evaluation, many are lamenting the ‘loss of childhood’ and personal expression, highlighting a ‘current crisis of positivist methods’ (Savage, 2013, p.3). Alternative metrics systems, such as rewarding effort and commitment rather than achievement, are rejected due to their subjective assessment.
The cultural sector, on the other hand, faces the opposite problem. Although some quantitative data is gathered in order to compare ‘quality’ and ‘success’ (mostly by its funder, the government), only recently the attempt has made to create cross-cultural metrics which could guide policy and financial support. Many members of the cultural community are worried about losing what they perceive as the core of artistic freedom: Creation freed from conformity. But how can funding be distributed fairly without a ‘neutral’ comparison?
Our presentation will offer a comparison between the two sectors and outline ways in which data is used to judge participants and quality. It will also introduce the idea of democratic, collaborative metrics and suggest ways in which the two areas can learn from each other in order to introduce a fairer, mixed methods evaluation system.
Savage, M. (2013). The “Social Life of Methods”: A Critical Introduction. Theory, Culture & Society, 30(4), 3–21. doi:10.1177/0263276413486160
Enacting the Child in School Through Data Technologies
Lyndsay Grant, University of Bristol
Data seduces us with a promise of greater knowledge; the increasing volume, depth, scope, granularity and timeliness of data are heralded as the key to answering many challenging problems in public and private life. The knowledge that data provides is not just predictive but also shapes the future; it is not only representative but constitutive (Ruppert 2013, Beer 2009). What kinds of data are collected, and how they are analysed, organised and presented, have important political consequences.
Childhood has been theorised as constructed through socio-material assemblages (Lee 2001), yet so far the role of data in producing the child in school has not received deep attention, while educational research has focused on learning analytics and questions of governance (Siemens 2012, Ozga 2012). This paper's contribution explores how children are constituted through data practices in a UK secondary school.
Drawing on a theoretical framework of relational materialism (Barad 2007) my research examines how data works to produce particular materialisations and meanings of the child in school. Through documenting material and discursive data practices, I unpick what kinds of ‘child’ are produced and how data technologies may work as instruments of power through which particular meanings, bodies, and boundaries of the child are produced. Crucially, this project seeks to explore the consequences for the kinds of childhood that are possible and the opportunities for agency that are available in a school in which data is becoming an increasingly important player in producing what it means to be a child in school.
Beer, D. (2009). Power through the algorithm? Participatory web cultures and the technological unconscious. New Media & Society, 11(6), 985–1002.
Lee, N. (2001). Childhood and Society: Growing up in an age of uncertainty. Buckingham: Open University Press.
Ozga, J. (2012). Governing knowledge: data, inspection and education policy in Europe. Globalisation, Societies and Education, 10(4), 439–455.
Ruppert, E. (2013). Not just another database: the transactions that enact young offenders. Computational Culture, (3), Online. Retrieved from http://computationalculture.net/article/not-just-another-database-the-transactions-that-enact-young-offenders
Siemens, G. (2012) ‘Learning Analytics: Envisioning a research discipline and a domain of practice’. LAK12: 2nd International Conference on Learning Analytics & Knowledge, 29 April – 2 May 2012, Vancouver
What is a Data Event? The Effects of Large-Scale Assessments in Schooling
Greg Thompson, Murdoch University, and Sam Sellar, University of Queensland
Large-scale assessments are a prominent source of performance data in schooling and make commensurate the practices of students, teachers and schools across times and spaces. The efficacy of data generated by these assessments emerges, in part, from relations between data and affect. Assessments make disparate places, subjectivities and practices commensurate and produce affects, or are embodied in intensive ways, which create multiple sense-making possibilities. For example, comparative performance data may be represented using traffic light systems that provoke visceral reactions which double rational analyses of the numbers and their implications for teaching practice in particular contexts (Sellar, 2014).
This paper will ask: What constitutes a data event? How do data capacitate bodies and focus attention? How do performance data become ‘eventual’? We draw on Deleuze’s conception of event as a “quasi-cause” that actualises within bodies, “producing surfaces and linings in which the event is reflected, finds itself again as incorporeal and manifests in us the neutral splendour which it possesses in itself” (Deleuze, 1990, p. 148). Each “present moment of actualisation” where the event is “embodied in a state of affairs, an individual or a person” is doubled by “the future and past of the event considered within itself” (Deleuze, 1990, p. 151). Drawing on empirical examples, the paper theorises this double-sidedness of data events in schooling.
Deleuze, G. (1990). The Logic of Sense. New York: Columbia University Press.
Sellar, S. (2014). A feel for numbers: Affect, data and education policy. Critical Studies in Education, http://dx.doi.org/10.1080/17508487.2015.981198, 1-26.
Knowing Schools: Data Power in the Governing of Education
Ben Williamson, University of Sterling
Contemporary educational institutions are being targeted for rapid ‘datafication.’ Focusing on emerging data-based ‘policy instruments’ (Lascoumes & le Gales 2007) this paper examines how ‘big data practices’ (Ruppert 2013) are interlacing with education governance through two case studies. The first is the Learning Curve Data Bank, produced by Pearson Education (the world’s largest commercial education publisher), a massive relational database of over 60 datasets from education systems globally. The Learning Curve mobilizes data visualizations, including time series tools and global heatmaps, to enable the data user to become its co-producer, ‘configuring the user’ (Woolgar 1991) as a ‘comparative analyst’ incited by the software interface and its in-built data analysis methods to construct particular educational problematizations and solutions. The second case study closely examines Pearson’s Center for Digital Data, Analytics & Adaptive Learning, and its embedding of automated predictive and prescriptive analytics in the pedagogic apparatus of the ‘cognitive classroom.’ The case studies demonstrate how global commercial data companies seek to utilize data to govern education through combining longitudinal data with real-time data analytics within the school itself. Analysed as digital policy instruments, these techniques of education governance are intended to measure, make visible, and modify student subjectivities by recursively prescribing pedagogic interventions to optimize student conduct. The paper will question the commercial power of Pearson in both knowing schools through data, and also in configuring ‘knowing schools’ as ‘sentient’ (Thrift 2014) educational institutions enacted through data-driven governance practices of ‘automated management’ (Kitchin & Dodge 2011).