Abstract
Abstract
In many educational jurisdictions, school systems are striving to demonstrate constant improvement. In Australia, the latest iteration of this concern is represented by the recent federal government report, Through growth to achievement. The report offers a number of recommendations to address declining achievement in Australian schooling. Pulling together scholarship on policy and educational technology I provide an analysis of key aspects of this report. This analysis draws attention to three salient features of the report: continuous assessment for continuous improvement; education as personalized learning; and growth mindset as a policy mandate. I explore the implications for schooling in Australia if these features were to be taken up. Analysis indicates that Australia could undergo significant change: teaching would be reconstituted as a process of continuous assessment; processes of personalized learning would lead to algorithmically tethered educative opportunities based upon students’ previous achievement and choices; and learners would be reshaped by technologies of persuasion. Given that educational jurisdictions globally are increasingly advocating for these types of educational solutions, this research is valuable as it provides a basis for further critique of such technologies being proffered as a panacea for educational disadvantage.
Introduction
A review of education in Australia was commissioned in 2018 by Australia’s Prime Minister to ‘provide advice on how … funding should be used by Australian schools and school systems to improve school performance and student achievement’ (Australian Government, Department of Education and Training, 2018: 114). The Australian government’s concern with school performance and student achievement is reflective of a global preoccupation with educational performance, national competitiveness and school reform (Verger et al., 2013). Given that many governments globally are grappling with these issues, it is worth examining the Australian government’s response as the policy implications of this review are applicable far beyond Australia’s borders.
The purpose here is to provide a theoretically informed analysis of this latest government review of education: Through growth to achievement: Report of the review to achieve educational excellence in Australian schools (Australian Government, Department of Education and Training, 2018). The specific focus here is on the report’s recommendations that could be used to advance educational technologies; such a focus allows for an exploration of the increasing entanglement of educational policy and technology. Scholarship on education policies that embed the teaching of information and communication technology (ICT) skills in order to skill students for participation in the global knowledge economy is not new (Lauder et al., 2006); however, education policy increasingly stipulates the use of technologies that colonize the mind and emotions of students (Williamson, 2016). These newer manifestations of the policy/technology nexus require fine-grained analysis to tease out the implications for students and teachers of this emerging aspect of the ‘technologization’ of schooling.
The paper is arranged as follows: I firstly provide a brief overview of the policy idiosyncrasies of Australia, then engage with the literature that deals with details of the ‘shifting ecology’ (Savage, 2016) of education policy and its connection to edtech more broadly. This analysis is theoretically informed by Deleuze’s description of the control society. After briefly rehearsing this theoretical concept, the analysis draws attention to the features of the report which demonstrate a push for continuous assessment for continuous improvement, education as personalized learning, and growth mindset as a policy mandate. I show that taken together, these features suggest a potential for the reconstitution of learners and teachers by technologies of persuasion. The paper concludes by exploring the implications of such features in respect of schooling in Australia and shows how this relates to educational systems more broadly. Given that globally an increasing number of educational jurisdictions are seeking to use policy to embed technological solutions in school system, exploration of the implications of such policies is valuable.
Context: federalism and policymaking in education in Australia
Within Australia, education (since federation in 1901) has been the responsibility of its separate states and territories. Within each of these eight jurisdictions is a state education system, a Catholic education system and a variety of independent private schools. Since the 1989 Hobart Declaration, the federal government has worked with the state governments to develop national goals for education that would ensure a degree of consistency across these disparate systems. While progress towards modest goals such as a common school starting age and improved literacy and numeracy had been steady, in the last decade Australia’s education system has been through a period of massive transition. Long-standing policy arrangements have been overturned as the federal government has increased its control over the education system. Since 2008, the federal government has: implemented a national curriculum; introduced standardized tests in numeracy and literacy; generated national standards for teachers and school leaders; and, developed a new model of school funding (Savage, 2016; Savage and Lewis, 2017). These reforms, like similar reforms in other nations, have been prompted by concerns about the changing role and purposes of education in a global knowledge economy (Rizvi and Lingard, 2010). The Through growth to achievement report with its emphasis on performance and achievement is Australia’s latest policy iteration that attempts to improve the national education system, increase its position in the international comparisons and address declining educational achievement. That the report offers recommendations that may or may not be endorsed by the states reflects the uneasy federalism (Savage, 2016) currently holding sway in education.
Shifting ecology of education policy
A prevailing belief in globalization as the path to the global knowledge economy has resulted in nations seeing technology dependent education as the means to ‘outsmart’ others in the race for scientific knowledge and technological innovation. The necessity of the use of digital technology in education has been political orthodoxy (Selwyn et al., 2001) since the beginning of the 21st century. Educational technologies have a role beyond teaching and learning and are now part of educational governance (Cifuentes, 2015). What has also shifted is the speed of policy development and the increasing involvement of non-government actors. Much has been written about the influence of the Organisation for Economic Co-operation and Development (OECD) and the Programme for International Student Assessment (PISA) rankings on education policy globally (Rautalin et al., 2018; Rizvi and Lingard, 2010; Sellar and Lingard, 2014), and it is clear that education policy in Australia is driven by a desire to improve our international rankings (Thompson et al., 2016).
The unremitting pace of education reform in Australian is a product of the process of the ‘speeding up’ of education policy which has been facilitated by the use of technologies of data collection and analysis which, it is believed, allow for identification of ‘best practice’ and ‘what works’ (Lewis and Hogan, 2019). The Through growth to achievement report is located in a global ecology of policy in which data and the attendant collection, storage and display technologies are seen as a panacea for education’s problems (Selwyn, 2015). There is a prevailing belief that the collection of more data will lead to a better picture of teacher performance and student achievement and unite Australia’s complex education system (Thompson et al., 2016). The collection and use of data for the creation of league tables and school and system comparisons means that data has become an inescapable part of education policymaking and governance (Ozga, 2009) as well as being seen politically as a way to solve educational problems such as a lack of equity, or an education system that is not preparing students for the contemporary world of work.
Such concerns are far from unique to Australia. In New Zealand the use of educational technology is promoted by ‘future focused’ policy (Hood, 2018). In the UK, the government has announced significant investment and partnerships with leading technology companies in order to improve student results across the country (Trueman, 2019). Educational technologies also form a key part of the Chinese government’s plan to modernize the education system by 2030 (Geromel, 2019). In the USA, personalized learning is considered to be a key means to improve educational efficacy and equity (Roberts-Mahoney et al., 2016), and edtech companies have become key political players in education (Regan and Talat Khwaja, 2019). Suffice to say, the policy push in Australia to embed more educational technologies in school (Buchanan, 2011) is not occurring in a vacuum, and the idea that the application of the right technological tools will solve educational problems has salience globally (Selwyn, 2013).
The theoretical framework and approach to analysis
This analysis is informed by Gilles Deleuze’s notion of the control society. In ‘Postscript on societies of control’, Deleuze (1992) provides an account of power that builds upon Foucault’s work. Summarizing Foucault’s analysis of disciplinary societies and the attendant modes of power, Deleuze describes a shift to what he terms ‘societies of control’. In societies of control, corporeal control is replaced with affective control; control of people within spaces is replaced by the monitoring of movements – ‘what counts is not the barrier but the computer that tracks each person’s position’ (Deleuze, 1992: 7). Foucault described how power is exercised via disciplinary methods (Foucault, 1979). People’s bodies are controlled by technologies of time and space. They must be in particular places at particular points in their life: at school, then in a factory or barracks (Foucault, 1979). For Deleuze, power is exercised as control over people’s minds rather than their bodies. Instead of having to be in certain places, their whereabouts are tracked, their location always known. Desires are shaped by the newly emerging forms of affective control being exercised by computerized means. The methods of disciplinary societies have not disappeared but rather their importance has diminished as they become supplanted by the new methods of ‘free-floating control’ (4) offered by the computer; the technology that for Deleuze is emblematic of societies of control. This examination of the Through growth to achievement report is informed by Deleuze’s notion of power as described within ‘Postscript on societies of control’ in order to highlight the implications of the push for take-up of the solutions contained with the report.
The mode of inquiry used here draws on critical policy analysis (Young and Diem, 2014), policy sociology (Ball 2008) and critical digital sociology (Lupton, 2012). That is, sociological concepts, ideas and research from these fields are used to ‘make sense’ (Ball 2008) of this policy document and the attendant educational technologies solutions under examination. This analysis is also informed by the approach of Taylor et al. (1997), who note that policy is more than the text as policies ‘represent political compromises between conflicting images of how educational change should proceed’ (15). Their work highlights that policy is multidimensional, value-laden, exists in context, constitutes a state making activity, never has a straightforward implementation and results in intended and unintended consequences (Taylor et al., 1997: 15–17). The purpose of this analysis is to detail some of these features, as they relate to the Through growth to achievement report. Attending to context, for example, makes clear that the report is not independent of wider global trends in educational policy. Analysis of the Through growth to achievement text makes plain the influence of the OECD and PISA testing on the policy climate in Australia. Consideration of possible consequences of this report demonstrates that (among other things) the report can be seen as an endorsement of specific types of educational technologies. Rather than considering whether technology is good or bad for education, this analysis considers possible trajectories if the solutions suggested in the Through growth to achievement report are taken up.
The Through growth to achievement report
Taking Ozga’s (2000) definition of policy texts as any ‘vehicle or medium for carrying and transmitting a policy message’ (33), this paper provides an analysis of the Through growth to achievement report. This report was commissioned by the Australian federal government in 2018. The lead author of the report, businessman David Gonski, was the lead author of what was termed by the media, ‘The Gonski’ report into educational funding in 2011. The Gonski report (Australian Government, Department of Education and Training, 2011) made clear the inequities in school funding in Australia and offered radical recommendations that would see a reduction in funds going to private schools. A needs-based funding model which was developed in 2013 in response to the Gonski report has subsequently only been implemented in the state of New South Wales. The Through growth to achievement report was commissioned by the federal government to leverage David Gonski’s reputation and association with educational fairness. Part of the purpose of the 2018 report was to reorient policy debate away from redistribution of funding as a way to address educational underachievement, toward other solutions.
The Through growth to achievement report and its recommendations were widely cited in the media upon its release (for example, see Mitsopoulos, 2018), demonstrating the report’s discursive function – it proffers common-sense, logical solutions to the problem of Australian students’ performance decline in both national and international assessment. The Through growth to achievement report provides three priorities for Australia’s education system, along with 23 recommendations, and 17 findings. Of these myriad features, 2 of the 3 priorities, and 5 of the report’s 23 recommendations are the focus of this paper. The elements focused on here contribute to the political rhetoric that the development and use of (particular) technological solutions can solve the problem of how to (continuously) improve schooling in Australia. The report’s recommendations, while aiming to solve the problem of educational underachievement, would also lead to further federal control of Australia’s state-based education systems.
The technological solutions to educational problems advocated by the report
Australia’s school systems and schools will become more innovative and adaptive, setting high expectations, and delivering continuous improvement, informed by data and evidence, helping Australia regain its standing as a world-leading education system. (Australian Goverment, Department of Education and Training, 2018: 112)
The desire to improve education through technology is enshrined in Australia’s national education policy. The Melbourne Declaration (Ministerial Council on Education, Employment, Training and Youth Affairs, 2008) details the national goals for the education of young people in Australia. It stipulates that ‘(i)n this digital age, young people need to be highly skilled in the use of ICT. While schools already employ these technologies in learning, there is a need to increase their effectiveness significantly over the next decade’ (Ministerial Council on Education, Employment, Training and Youth Affairs, 2008). The update to the Melbourne Declaration, the Alice Springs (Mparntwe) Education Declaration, states that Australian students are to be successful, lifelong learners who 'are productive and informed users of technology as a vehicle for information gathering and sharing, and are able to adapt to emerging technologies into the future' (Council of Australian Governments Education Council, 2019, p.7). The Through growth to achievement responds to these national goals by including technological solutions to the problem of Australia’s declining education achievement. This declining achievement is referred to thus: Australia has a strong educational heritage and committed educators. Since 2000, however, academic performance has declined when compared to other Organisation for Economic Co-operation and Development (OECD) countries, suggesting that Australian students and schools are not improving at the same rate and are falling short of achieving the full learning potential of which they are capable. As a nation, we need to act now to raise our aspirations and make a renewed effort to improve school education outcomes. (Australian Government, Department of Education and Training, 2018: viii) Many attempts have been made to reform education and teaching policies at the state and federal level to improve teaching quality. Some efforts have generated lasting impact and some are ongoing. The Review Panel believes, however, that there are still opportunities for Australia’s education system to better support and encourage teachers’ practice, underpinned by contemporary research, and to use technology more effectively to enable personalised learning. (Australian Government, Department of Education and Training, 2018: 56). Recommendation 1: Embed a focus on individual student achievement through continuous learning progress in the policies and practices of all schools and systems, with the expectation that each student should achieve at least one year’s growth through each year of schooling. Recommendation 4: Introduce a new reporting with a focus on learning attainment and learning gain. Recommendation 5: Revise the structure of the Australian Curriculum to present the learning areas and general capabilities as learning progressions. Recommendation 11: Develop an online and on demand student learning assessment tool for teachers for the purposes of formative assessment and tailored teaching. Recommendation 22: Accelerate the introduction of a Unique Student Identifier. (Australian Government, Department of Education and Training, 2018: xii–xiv).
Continuous assessment for continuous improvement
In pulling together the Through growth to achievement priority of ‘a continuously improving education system’ with the recommendation for an online on-demand student learning assessment tool we can see that the report advocates continuous improvement through continuous assessment of the education system as a whole and the individual students within the system. This reflects the ‘utopia testing dream’ identified by Thompson et al. (2016) that ‘one day adaptive, online tests will be responsive in real time providing an integrated personalized testing, pedagogy and intervention for each student’ (212). For this dream to be realized, each student within the system has to be identifiable. While the development of a unique student identifier had begun prior to this review, Through growth to achievement recommends that this process be accelerated. Deleuze (1992) argued that with the measurement, monitoring and assessment offered by computer technology, ‘individuals have become “dividuals”, and masses, samples, data, markets, or “banks”’ (5). In the Through growth to achievement report we see students being repositioned as samples – identified by a number, their education reconfigured as a series of learning progressions, and their progress through schooling shifted to continuous formative assessment, recorded via an ‘online and on demand student learning assessment tool’ (Australian Government, Department of Education and Training, 2018: 66). This not only ‘dividualizes’ students, and subjects them to continuous and varied types of monitoring, but changes the nature of teaching. For example, a reliance on data changes how teachers view their students – while their learning and learning gain becomes clearer, Hardy and Lewis (2018) argue that teachers lose holistic sight of the students in their care. Additionally, rather than teaching being understood as a specific form of collective labour and intellectual work undertaken in particular workplaces (Connell, 2009), teachers and their work are reconfigured, defined and judged on the continuous administration of assessment and collection of data.
Continuous assessment for continuous improvement would see an expansion of the existing practices of monitoring and recording of students’ educational achievement. Currently, in many education jurisdictions globally, children’s educational histories are recorded in databases, which can be linked with other datasets so that individual students’ educational progress can be tracked and monitored over the course of their entire education (Lupton and Williamson, 2017). Increasingly sophisticated programs allow for the collection and combination of various types of information on students; not just educational results and progress, but on the minutiae of daily student life: behaviour, demerits, uniform infractions, homework, etc.; combining to create detailed data-driven histories of students’ educational life-course. This demonstrates the possibility for expansion of the factors that students are currently being assessed on. This collection of data reduces students to the sum of what can be collected about them, a process that shapes them into specific subjectivities via modes of self-governance (Selwyn, 2015) – exemplifying the affective control exercised in the control society (Deleuze, 1992).
These examples show how the goal of continuous improvement provides a justification for the continuous monitoring of students in a multitude of ways. Students’ educational progression from preschool to further and higher education can be tracked; their physical activity, use of digital devices, social media and physical locations can be recorded in perpetuity as well as tracked in real time. Lupton and Williamson (2017) argue that the ready availability of this wealth of data generates new norms and success criterion against which students are measured, and with this will come new moral codes and social expectations and new ways of defining students against data-derived categories. It also changes what teachers do and what teaching is. Data generation becomes the indicator that ‘teaching’ has taken place (Thompson and Cook, 2014) rather than relationship building, curriculum or timetable management, or classroom activity. Data collection must be continuous, so that continuous improvement (Australian Government, Department of Education and Training, 2018) sought by the Australian government can be demonstrated. Discrete assessments such as examinations are replaced with continuous assessment – the continuous monitoring and tracking that defines societies of control.
Education as personalized learning
One often reported feature of the Through growth to achievement report is the ‘personalized learning’ (Mitsopoulos, 2018) made possible by the individual targets, progress and continuously monitored assessment advocated by the review. While personalized learning is an elastic term that is interpreted, defined and enacted in a variety of ways, one predominant way in which this term is understood in the contemporary policy climate is learning delivered via digital platforms (Thompson and Cook, 2017). Digital platforms use learning analytics to individualize students’ education by offering temporally differentiated lessons based on the individual student’s learning pace. Learning analytics platforms are already being used in schools for the purpose of personalizing students’ learning. Such platforms are designed to ‘mine data about learners as they go about educational tasks and activities in real time and to provide automated predictions of future progress that can be used as the basis for intervention and pre-emption’ (Lupton and Williamson, 2017: 785). The use of such platforms is based on the assumption that ‘more and better data improves teaching and learning processes’ (Thompson and Cook, 2017: 743).
The purpose of learning analytics is to use adaptive algorithmic determinations of students’ progress to generate predictive data patterns that suggest interventions for the personalization of learning activities so that students will be motivated to invest greater effort in their own education (Thompson and Cook, 2017). With personalized learning delivered by learning analytics, ‘(s)ubjectivation is affected not just through panoptic surveillance, but also through patterned digital data that are used to predict behaviour, curate potential stimulus and cultivate responses through the logics of algorithmic calculation applied to the data produced’ (Thompson, 2017: 828). That is to say, students are being subject to forms of affective control that are enabled by the collection of data about their learning choices and progression.
If the Through growth to achievement espoused goal of providing students with personalized learning is enacted in this way – through such data-driven methods, educational technologies would determine what, when and how students learn – with curriculum and assessment determined algorithmically based on students’ prior engagement and achievement. Personalized learning technologies such as learning analytics platforms would ‘displace the embodied expert judgement of the teacher to the disembodied pattern detection of data analytics algorithms’, risking students’ opportunities being narrowed by the assumptions encoded in the algorithmic logic’ (Lupton and Williamson, 2017: 8). This shifts the locus of control, professional judgement and assessment design from teachers and into the hands of those who design the products that are used to assess and monitor students. In such a form of education, not only are teachers bypassed (with their experience and professional judgement removed from the learning setting), but local and national educational systems lose authority to those who design such technologies and prescribe the learning progressions (Williamson, 2016).
Such personalizing (yet ironically based on positioning students as ‘dividuals’ rather than individuals) technologies are based upon algorithmic logics that generate an algorithmic identity for students – technologies that have as their basis the category inferring algorithms operating via surveillance in online networks (Cheney-Lippold, 2011). Cheney-Lippold (2011) unpacks the way in which algorithmic identity operates and these processes (which are now being used in education) were developed. Code has become the architecture shaping online space. Through the use of code, algorithms have been developed ‘that can “automatically and continuously” affect life chances offered to users based on a pre-configured but also reflexive programmed logic’ (Cheney Lippold, 2011: 166). The code and algorithms that underpin learning analytics platforms have their origins in the algorithms developed by marketers to target Internet users in real time, with niche ‘personalized’ advertising as they browse the web. As Cheney-Lippold explains, the use of algorithms allowed marketers to better target content and advertisements: Through algorithms, commonalities between data can be parsed and patterns within data then identified and labelled. And with the capacities of computers to surveil and capture user activity across the internet, consumer information was removed from the shackles of time-bound, decadal census data and began to float atop a constant stream of real-time web use that can be matched against existing behaviour and identity models – like gender. (2011: 168)
Growth mindset as a policy mandate
The Through growth to achievement report not only advocates personalized learning, it also espouses the goal that ‘(e)very Australian should emerge from schooling as a creative, connected, and engaged learner with a growth mindset that can help to improve a student’s educational achievement over time’ (Australian Government, Department of Education and Training, 2018: x; emphasis added). The report is explicit in its reference to the work of Carol Dweck. Dweck is a professor of psychology at Stanford University whose work on growth mindset originated with a 1998 paper which reported on six studies showing that praise for effort has a more significant impact on motivation than praise for intelligence (Mueller and Dweck, 1998). Dweck advocates teaching in ways that do not reinforce ideas of fixed intelligence but rather work to show that with effort, achievement can be improved; that is, teaching students to have a ‘growth mindset’ rather a ‘fixed mindset’. Although Dweck’s work has been very influential in several countries, 20 years of educational interventions based on growth mindset has yielded mixed results and has not led to widespread or large-scale improvements in educational achievement (Hendrick, 2019). Dweck herself is dismayed at the application of ‘misunderstood’ versions of her theory (Gross-Loh, 2016) and cautions that it should not be used as the basis of policy (Wiggins, 2015). (It is also worth noting that Stanford has a long academic history of studying how human behaviour can be modified, from the Stanford Prison experiments through to the contemporary Stanford Persuasive Tech Lab (https://captology.stanford.edu/). The technological innovations of Silicon Valley also have their roots at Stanford University; see Trikha (2015)).
State interventions into the development and optimization of pyscho-emotional attributes have been normalized, with schools as primary sites for intervention (Ecclestone, 2017). Lanier (2018) notes that there is something about the on-off logic of digital technologies that makes it amenable to behavioural psychology and behaviourist modes of thinking. Growth mindset is no exception. Technologies that embed a growth mindset are already in use – ClassDojo provides an illustrative case study. The website (https://www.classdojo.com/about/) notes the popularity and utility of the application – it is (a)ctively used in 95% of all K-8 schools in the U.S. and 180 countries; 1 in 6 U.S. families with a child under 14 use ClassDojo every day; 15 million children have learned about Growth Mindset and Empathy with ClassDojo.
Here, I flag critiques of ClassDojo, as the solutions proffered in the Through growth to achievement report suggest the possibility that an app that not only monitors and records student behavioural achievement but also promises to embed a growth mindset would be ideally placed as a means of embedding a growth mindset into the education of Australian students. Williamson (2017a) describes ClassDojo as ‘a persuasive technology’ designed to routinize particular behaviours (such as those that indicate a growth mindset) through reinforcement (9). ClassDojo gamifies the classroom with its avatars, leader boards and rewards for positive behaviour and is also used to give warnings and deduct points. As such, ‘ClassDojo turns classroom behaviour into a quantifiable source of value for children to use a public display of their compliance with classroom norms and expectations’ (Williamson, 2017a: 5). The app encourages children to measure and value themselves via visual representation of their behavioural data. Problematic aspects of ClassDojo also include its use of virtual badges for obedience, and its normalization of competition and surveillance, and the ways in which students are visibly ‘measured against idealised norms’, inculcating them in ‘data-driven performative cultures of discipline’ (Manolev et al., 2018: 47). Williamson (2017a) questions technologies such as ClassDojo, which embed the goals of teaching children a growth mindset, stating that using technologies to achieve this positions children as ‘subjects in a massive experiment in mindset modification’ (8). The operation of such technologies seeks to shape students’ subjectivities so that they seek to constantly improve themselves; in Deleuzian terms, they are controlled via affective means. This goal of shaping students’ mindsets, through the use of sophisticated psychosocial technologies, operating via personal learning platforms, fundamentally changes what education is and means: ‘the new technologies of education, of making education “personal” and into something in which people will invest in themselves, results in a reconfiguration of the sense of “education” ’ (Thompson and Cook, 2017: 741).
While behavioural modification is not a new feature of schooling, the corporeal focused disciplinary technologies (such as the classroom, timetables, uniforms, etc.) of education are being supplemented by technologies that collect data about students’ emotional/psychological and cognitive/neurological states. Such data is not merely gathering information about students' bodies, emotions and thinking, but is also governing these. The continuous collection of data enabled by digital technologies and required by policy represents a qualitative shift in how power can be exercised by the state, over students. Not only are they physically disciplined through the materials and architecture of the schools, but computer-based technologies exercise power through sophisticated means of affective control that implicate the students themselves in the processes of visible rewards and in the gamification of behavioural norms. Control is exercised continually and more completely than via disciplinary techniques of power. Through digital technologies, school follows children home – ClassDojo, for instance, also operates as a means of communicating with parents. As such processes shape students’ desires, their physical presence at school is not necessary for control to be exercised.
Technological solutions and the problem of declining achievement in Australian schools
Political advocacy of particular technological solutions
Arguably there are a number of solutions that could be applied to the problem of the persistent patterns of underachievement that characterize the Australian education system. For example, social and educational policies could be developed that work to address poverty (Berliner, 2013), focus on equity (Espinoza, 2007) and advocate particular pedagogical approaches (Mills et al., 2009); or teacher professional development could be reconfigured for effectiveness (Bowe and Gore, 2017). In the Through growth to achievement report, however, several of the key recommendations are compatible with specific types of digital technologies. Selwyn (2016), in his exploration of whether technology is good for education, reminds us that technology is not value-neutral. To advocate for technological solutions for the problem of underachievement is not to empower teachers and students but instead to tether educational processes to previous performance and limit future options. The technologies explored here are based on algorithmic inference, a process that uses past data to shape future opportunities. The data and the algorithms based upon it are not neutral, but serve rather to amplify the biases contained within both. The constellation of possible solutions examined here serves also to individualize the problem of underachievement and student performance. In the report, underachievement is seen as a problem of individual students that can be solved with the application of solutions or technologies that drive (or nudge) students and teachers to improve their performances. Such processes would serve to reshape students into desiring continuous improvement, and places educational responsibility in the hands of those who develop the technological solutions being advocated.
The recommendations advocated in this reform represent an attempt by the federal government to gain more control over Australia’s education system. In a time of global education policymaking, national governments are seeing a reduction in their ability to steer education (Ball 2013), and the reforms discussed here are an attempt by policymakers to increase the federal government’s capacity to control Australia’s education system and to solve the problems of declining achievement without having to invest further monies in either the education or social systems. This, however, is a potentially perverse attempt, as a dependence on technological solutions removes control not just from teachers, but also from government bureaucracies; control instead rests with those who design the educational technologies.
Implications for schooling
If Australia’s educational problems of declining student achievement and performance were to be addressed by the technological solutions examined here what would be the likely result? Regardless of the effect on achievement and educational outcomes, the solutions examined here – continuous assessment, personalized learning, and growth mindset – would change Australia’s schooling system. Thompson and Cook (2017) argue that the ‘sense’ of education is being changed by the use of technologies that deliver personalized learning through learning analytics and big data. In a regime of continuous assessment, learning is never finished. Teaching changes, from being managed by the teacher in the classroom, to being managed by the means of digital delivery and monitoring and surveillance of students’ learning. The relational aspect of teaching, and discourses in which teaching is understood as a form of knowledge work (Connell 2009), become replaced by an understanding of teaching where good teaching is constituted by the production of good data (Thompson and Cook 2014). In such a system, those who design the educational platforms control student learning.
These specific recommendations of the Through growth to achievement report flag possible ways in which Australia’s education could be reconfigured in ways that are congruent with Deleuze’s control societies. The model of education suggested by the recommendations within the report shifts learning to a series of progressions that are monitored and recorded. Technologies of education could be deployed to shape children’s minds, behaviours and emotions, to get them to invest in their learning and maintain a ‘growth mindset’. While intellectual growth and personal development has always been a goal of education, the use of the technologies described here represents a qualitative change. Here the mechanisms for growth would be out of the hands of the teachers, and student behavioural modification could be achieved via technologies designed for ‘relentless, robotic, ultimately meaningless behavioural modification in the service of unseen manipulators and uncaring algorithms’ (Lanier, 2018: 23).
It is not just that algorithms are uncaring, or that they amplify bias (Danaher, 2016); a further issue is that the use of such technologies in learning represents an exercise of control over students by the state. The purpose of continual assessment and personalized learning is to build within students the desire to continually learn and improve themselves, and this desire is applied via specific persuasive technologies and techniques. As Watson, drawing on Deleuzian theory, explains, ‘control happens when our own desires appear to align with, but in fact emanate from, the interests of the State (or whatever institution it is that our desires our desire)’ (2010: 96). Watson brings our attention to the ways that desires which students will be inculcated to identify with and take on as their own will have emanated from the state as mechanisms of control. In the Through growth to achievement report, continued improvement is a state imperative and this policy recommendation blatantly seeks to embed this mode of thinking into the education of all students in the system, a goal amenable to the use of consumeristic-based technologies of persuasion.
Conclusion
This examination of the Through growth to achievement report provides an Australian policy example of a global phenomenon, the desire to ‘fix’ education systems; in this case with solutions that include continuous assessment, personalized learning and the inculcation of a growth mindset. Digital apps, learning analytics and computer adaptive testing are examples of technologies that could be used to achieve these ends. In a time of policy borrowing (Lingard, 2010), other countries are implicated in the drive to shift education in the ways detailed in this paper. Knox et al. argue that such international trends require analysis, given high-profile associations between ‘data science’ and broad educational agendas, from national policy initiatives, through research and development, to classroom practices ‘on the ground’. Importantly, these agendas demonstrate the increasing entanglement of sophisticated arrangements of software, infrastructure, and code with educational theory in ways that powerfully shape both the governance and day-to-day activities of teaching and learning in institutions. (2019: 1)
In describing the policy push for continuous assessment, personalized learning, and the embedding of a growth mindset in teaching I have sketched out the possible trajectory of such policies, so that the future engendered by these can be considered. Teaching would be reconstituted as a process of continuous assessment; processes of personalized learning would lead to algorithmically tethered educative opportunities based upon students’ previous achievement and choices; and learners would be reshaped by technologies of persuasion. While continuous assessment, personalized learning and the embedding of a growth mindset may not be implemented as recommended by this specific report, exploration of these is valuable as these specific solutions appear readily in educational discussions and policy documents. They have become buzzwords, solutions that appear fully developed and ready to be deployed (like Athena springing fully grown from the head of Zeus) to solve a number of educational problems. While the purpose of this paper has been to consider the effects of such trajectories, a worthy exploration would be a tracing of origins and mapping these policy solutions – especially given that such technological solutions developed in Silicon Valley have traction beyond their geographical origins (Williamson, 2017b).
The argument here is not that digital technologies have no place in education, but rather that the use of specific forms of technology, such as those that might meet the recommendations of the Through growth to achievement report that I have examined, have potential effects that go beyond more data being collected about children. An uncritical adoption of specific technological-based versions of personalized learning means that students will become more enmeshed in an ‘ever-intensifying network of visibility, surveillance and normalization’, where the ‘embodied expert judgement’ of their teachers is displaced by disembodied algorithmic and adaptive decision-making technology (Lupton and Williamson 2017: 7–8). The risk is that such processes shut down educational possibility and that students’ prior actions determine the future learning made available to them. In such a scenario, past performance tethers students to particular learning opportunities determined by algorithms and beyond the oversight of educators. Given the proprietary nature of the algorithms underpinning apps, learning analytics and personalized learning platforms, the mechanisms of control are not just out of teachers’ hands but are unknowable, as they are ‘black boxed’ (Pasquale, 2015) by those who design them, and also increasingly involve machine learning, so their logic may not be traceable.
Deleuze (1992) provides a theoretical basis from which to consider such educational practices as a modality of control, and his work on the operation of power in societies of control also hints at the availability of alternatives. In this context, ‘(t)here is no need to fear or hope, but only to look for new weapons’ (Deleuze, 1992: 4). When analysing policy, argues Ball (2008), policy is to be treated ‘as a process, something ongoing, interactional and unstable’ (7). It is this unstable nature of policy that allows for lines of flight, resistance and the non-inevitability of particular outcomes. Selwyn (2014) argues that what is required is the development of new values and sensibilities around educational technologies. Given that educational jurisdictions globally are increasingly advocating for the inclusion of technologies such as personalized learning, this research is valuable as it opens a space for further critique of both the technologies being advocated and the discursive moves that facilitate this push.
Footnotes
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by a University of Newcastle, Women in Research (WIR) Fellowship (grant number G1801225).
