As educational establishments like schools and universities increasingly rely on digital platforms, software, and analytics tools, the process of datafication – “the ways that digital technologies are associated with the generation of ‘big data’ sets that can be analysed to make predictions, identify trends and generally inform what takes place in society” (Selwyn, 2021, p. 196) has emerged as a prominent feature of contemporary education.

Datafication has afforded us valuable insights into learning behaviours, helping educators and researchers identify gaps that were otherwise difficult to detect (Fischer et al., 2020).  Work in fields like learning analytics and educational data mining depends on the outputs of datafication to analyse student performance, predict student outcomes and even inform design interventions to create effective solutions. This requires large educational data sets to uncover trends, find correlations and support evidence-based decision-making in educational settings. However, beneath the promises of greater efficiency, personalisation, and innovation, datafication carries profound implications for how education is conceptualised, delivered and experienced. Jarke & Breiter, (2019, p. 1) in Editorial: The Datafication of Education, caution that “this proliferation of data changes decision-making and opinion-forming processes of educational stakeholders…”, underlining a shift in the dynamics of authority and agency within the educational systems.

If datafication promises scalability and efficiency, is its trade-off with equity, teacher autonomy, and holistic learning justified? In a search for these answers, this post first interrogates the convoluted dimensions of datafication by situating some of the aforementioned concerns in the intricate landscape of education, technology, and policy. Second, it investigates what counts as learning in a data-driven education system, exploring how metrics, benchmarks, and analytics tools prefer certain knowledge and skills while sidelining others. Third, it probes the shifting authority of teachers in the datafied classroom, where pedagogical decisions traditionally rooted in professional judgment are now increasingly dependent on – and soon may be replaced by – algorithmic recommendations. Finally, these dimensions reveal datafication’s broader influence, extending beyond instruction and assessment to raise fundamental questions about the very purpose and values of education.

The post reads in four segments detailed above and each segment ends with a set of questions to consider for the future.  They aim to open a conversation between educators, policymakers, technologists, and researchers while tackling the wicked challenges in education.

1. The Intricate Landscape of Education, Technology and Policy

At the heart of datafication’s influence on education lies the capacity of digital platforms and analytical tools to shape educational priorities through the selection, categorisation, and valuation of data. This is generally done through learning management systems or MOOC platforms and sometimes through resource management tools.

With the ease of access to data through the systems mentioned above, datafication has fundamentally transformed the educational landscape, embedding a touchpoint for data collection and processing at every step. Whether it be micro-level student performance data or macro-level administrative data (Fischer et al., 2020), analytics is embedded into the everyday teaching and learning experience of a sizable population. As global spending on education technology grows with billions allocated annually to data-driven tools, the scale of datafication’s impact becomes exponentially dramatic. As per HolonIQ (2024), the global expenditure in EdTech is estimated to reach $404B in 2025, which is twice what it was in 2019. These figures are bound to go higher in the coming years, considering the increasing use of technology in education.

The noticeable upshift in the investments has not happened in a silo. Policy frameworks often drive this shift emphasising the role of technology in achieving scalable education solutions. Data plays a central role in shaping these policies, as governments and institutions increasingly rely on analytics to make decisions about resource allocation, curriculum development, and accountability measures. For instance, student performance metrics and engagement data collected through learning management systems are often used to inform funding decisions, evaluate the effectiveness of programs, and set nationwide educational benchmarks.

From a global standpoint, The UK’s EdTech strategy (2019), India’s New Education Policy (2020) and the United States’ Every Student Succeeds Act (2015) are excellent initiatives prioritising data-driven, evidence-informed education. However, this impetus on data-driven education raises concerns about the ethical use of student information, the accuracy of predictive analytics, and the exclusion of nuanced, non-quantifiable aspects of education from policy considerations. These frameworks may overlook the ethical and pedagogical complexities associated with datafication. Drawing conclusions from Jarke & Breiter’s (2019) critique of such policies, datafication positions education as an opportunity for optimisation, often at the expense of deeper, meaningful engagement.

In looking for a pragmatic future, the conflict between optimisation, the promise of innovation and the risk of control and surveillance is evident. While analysing data may offer insights that might otherwise be missed, it also shifts the focus of education toward measurable outcomes, reducing the richness of learning to numbers (Williamson, 2019). For instance, predictive analytics tools employed in higher education often evaluate student success through attendance, grade performance, and participation metrics and these indicators, while seemingly neutral, risk over-representing certain behaviours while ignoring the diverse pathways students take to achieve meaningful learning outcomes (Smithers, 2023). A particularly contentious issue is how data is used to standardise and commodify learning experiences across platforms, whether digital or otherwise. These systems, often bordering policy guidelines, embed normative assumptions about what constitutes progress. An example of this is the national aptitude tests, which collect massive amounts of student data to create a country’s mastery or literacy report. A step ahead is the global ranking based on education levels. These metrics not only narrow the definition of success but also guide educators’ decisions, inadvertently shaping the curriculum to fit the metrics rather than student needs.

Such tensions underscore the need to critically evaluate how education policies and platforms influence the broader educational landscape. While the allure of efficiency and scalability drives the adoption of datafied systems, it is essential to question whether these tools truly serve the diverse needs of learners or merely simplify complex educational processes for ease of management. Navigating through this complex terrain may not be easy, The Global Education Monitoring Report  (UNESCO, 2023) evaluates the role of technology in education and highlights that “it would cost USD 1 billion per day to maintain connectivity for education in poor countries” signalling considerable challenges in collecting data from non-connected educational contexts.

Questions for consideration:

1. If data-driven policies are the most pragmatic solution to address systemic inefficiencies, is it acceptable to sacrifice nuance for scale?
2. What does this mean for regions without the infrastructure to leverage these technologies?

2. What counts as Learning?

In a data-driven education ecosystem, learning is increasingly defined by what can be measured instead of what should be measured. After early warnings (Biesta, 2009) about the rising focus on measuring and comparing educational outcomes, the focus on measurement has exponentially grown in the last decade driven by the rapid development and adoption of data-driven technologies in education. The datafication processes in popular education platforms such as Moodle, Black Board, Khan Academy, and Coursera rely on easily quantifiable results, metrics and standardised comparisons, like completion rates, time to complete tasks, or quiz scores, overlooking some aspects of learning such as cognition, collaboration etc. leading to an incomplete definition of what counts as Learning. While these metrics are frequently glorified as objective measures, they reflect and reinforce particular assumptions about learning and how it should be demonstrated. In practice, this overemphasis on quantifiable outcomes can lead to the narrowing of educational focus and reshaping educational priorities.

Deep conceptual understanding and other complex intellectual processes like critical thinking, creative problem-solving, or ethical reasoning, are hard to capture in neat-looking datasets. As a result, these important links to learning risk being sidelined or overshadowed by more straightforward skills. Furthermore, Dragan Gašević et al., (2015) note that “while it is often perceived that education is rife with data, very little is related to capturing the conditions for learning”, indicating that learning cannot be measured through a stream of data alone and needs to be viewed in conjunction with the ‘external and internal’ environments including social contexts, historical background and cognitive abilities (Dragan Gašević et al., 2015).

The illusion of personalisation is another key issue that contributes to the debate of what counts as learning. A big part of commercially available education technology for learners and teachers promises tailored learning experiences. While this personalisation promises to custom-make pathways for all learners, algorithms identifying these pathways often reinforce existing inequalities. AI-enabled adaptive learning tools, for instance, base pathway recommendations on past performance and data generated by previous learners, which might perpetuate cognitive biases and limit opportunities for students from a different context. In her recent work, Pelletier, p. (2024, p. 111) critiques the very premise of personalisation, noting that “celebration of personalisation begs questions which, historically, have been central to education as a professional practice and a field of study”. Further, Holmes et al., p. (2018, p. 93) note that “technology-enabled personalised learning” may offer partial solutions but it “is unlikely to succeed without addressing the fundamentally human dimensions of learning”. This brings up a vital concern for a datafied classroom: How does it help students learn, particularly when the very premise of personalisation needs a critical assessment and the tools that claim to enhance learning may further deepen inequalities and marginalize diverse educational needs?

Further, drawing on Vygotsky’s Zone of Proximal Development theory, the learning that happens outside of the data-driven education ecosystem, without the assistance from tools and technologies responsible for collecting data, is unaccounted for. This unmeasured learning includes informal guidance from mentors, collaboration among peers, and the organic discovery of knowledge through exploration and lived experiences. By ignoring these essential dimensions of learning, datafication risks oversimplifying the process and undervaluing human connections.

The current trajectory of data-driven education systems has raised critical questions about the shift in classroom dynamics, and the race to jettison the one-size-fits-all concept in education is faced with unique challenges in a conventional classroom setting. In the next section, I will focus on the changing classroom dynamics and the implications of datafication from a teacher’s lens.

Questions for consideration:

1. Is it ethical for education to be guided predominantly by market and workforce needs rather than humanistic ideals of learning?
2. How do we resolve the tension between ‘preparing students for jobs’ and raising well-rounded individuals?

3. Shifting Authority of Teachers

Traditionally, the teacher has served as the primary source of knowledge and has held a central position in a classroom. They have been considered moral authorities in some contexts and a facilitator of learning in others. In his book Education & Technology: Key Issues and Debates Selwyn, p. (2021, p. 121) describes a teacher as a support However, in recent decades, this role has been subjected to intense changes, mostly driven by the rise of digital technologies, data-driven education policies and a larger societal shift towards standardisation and efficiency. These transformations have redefined what it means to be a teacher and have, to a large extent, altered the authority of a teacher in the classroom. The authority here does not refer to the authoritarian control over the learners but the influence over the classroom that comes from professional expertise. Datafication and its promise of efficiency, transparency and standardisation have been a key driver to this shift in turn eroding the autonomy and criticality with which teachers operate.

Datafication of education has introduced a paradigm where dashboards (as visual representations), metrics (as qualifiers) and, algorithms (as opaque magic), dictate the terms of teaching and learning. Holloway (2020) observes that teachers are now evaluated not only by their effectiveness through pedagogy but also by their ability to produce quantifiable outcomes. This shift has put teachers in a precarious position as their professional expertise and judgement are often overlooked by data-driven systems. In this new paradigm, the authority of teachers is no longer a culmination of their knowledge, experience, or interpersonal skills but of their ability to align with standardised performance metrics.

As Selwyn (2021) points out, the teacher’s role is increasingly framed as that of a facilitator or implementer rather than as a creator or innovator. This shift diminishes the teacher’s status as an expert and repositions them as a component in a larger bureaucratic machine. Furthermore, the emphasis on standardization often clashes with the realities of classroom teaching, where flexibility, empathy, and adaptability are essential for addressing diverse student needs. By privileging uniformity over individuality, datafication risks dehumanizing the teaching profession and eroding the relational foundations of education.

Another critical factor contributing to the shifting authority of teachers is the rise of external authorities in the form of technology vendors, data analysts, and policymakers. As Stevenson (2017) implies the increasing use of data-driven systems in education has empowered third-party actors who design, implement, and manage these systems. These external stakeholders often have little to no direct experience in education but by virtue of the design of this transaction, have significant influence over how teaching and learning are conducted. For example, algorithms that determine student performance or predict future success are often developed by private companies with proprietary interests, raising questions about accountability, transparency, and equity. The growing influence of these external stakeholders further subdues teachers’ voices in the decision-making process. The promises to increase time at hand for teachers end up costing dearly in terms of disempowerment and frustration.

Irrespective of these challenges, there is a growing recognition of the need to reimagine the role of a teacher in this data-driven world and find a way for ed-tech to work in tandem with traditional education. As Ideland (2021) emphasises, the relationship between corporate goals and pedagogic requirements needs to be analysed to find the new normal. She further observes that “Maybe we need to do as the edupreneurs tell us and try to keep up with our contemporaries’. This sentiment of hopelessness and resignation urges immediate action towards building an educational environment that is data-informed and driven by passion and human-centered teaching practices.

Questions for consideration:

1. If teachers are expected to adapt to this new reality, how do we ensure their role remains centred on nurturing human connections?
2. How do we redefine teacher authority in a way that respects both their expertise and the efficiencies offered by data-driven tools?

4. Broader Influence of Datafication in Education

The datafication of education extends beyond the classroom interplay and individual experiences of both teacher and learner. It pierces through every layer of the education ecosystem, influencing how decisions are made, resources are allocated, institutions are ranked, and societal expectations are formed. Datafication transforms education into a continuous process of constant monitoring, measurement, comparison, and analysis. This does not stay contained within the formal education years of a learner but continues to be the identity of the learner through certificates and permanent records. A few key issues about the broader influence of datafication highlighted in this section are (i) Globalisation and Marketisation of education, (ii) Ethical and Societal Implications, and (iii) Impact on Curriculum Design.

Datafication has accelerated the globalisation and marketisation of education, where institutions are competing to be on top of the hypothetical ladder based on measurable outcomes. A place popular among students to compare universities is the QSWorld Ranking. The website lets the users slice and dice results based on various metrics such as academic reputation, employer reputation, faculty-to-student ratio, and research output, among others. This focus on measurable metrics has led to a performance-driven culture, where universities and schools invest heavily in enhancing aspects that boost rankings, sometimes at the cost of holistic development or equitable access to education. For instance, institutions may mandate the publication of a certain number of research papers annually in high-impact journals to score well on one of these rankings. This globalised benchmarking exacerbates competition, pushing institutions to align their priorities with market demands rather than educational values.

The ubiquitous collection and use of student data in the name of enhancing learning outcomes and informing future developments raises serious concerns regarding privacy, surveillance and biases. Educational technologies operate within power dynamics, where those who control the data have significant influence over educational opportunities. These ethical dilemmas extend to questions like, who owns student data? How is it used? or will it be used for learner interest or corporate gains? These questions not only probe into the motivations of datafication but also suggest a critical valuation of the proposed benefits.

Further, datafication’s high impact on curriculum design can be estimated by the emergence of new job roles tailored to meet the increasing demand for skilled professionals to support tech-enabled curricula. Institutions now employ ‘learning technologists’ or ‘curriculum architects’ to support the design implementation and align it with measurable outcomes. These professionals use advanced analytics, technologies and data insights to optimise learning experiences. While these roles bring innovative potential to education, they also raise critical questions. To what extent do they prioritise market-oriented skills over holistic and ethical considerations? As an early response to these questions, many education researchers now focus on the purposeful development of educational technologies to safeguard the ‘institution of education’ from succumbing to corporate interests.

Together, these issues only partially explore how datafication extends beyond individual classrooms to reshape the broader educational ecosystem. As we embrace technological advancements, it is imperative to critically assess their long-term impact on equity, inclusion, and the fundamental values of education.

Questions for consideration:

1. If datafication is unavoidable in a global education market, how can we design systems that resist commodification and prioritise inclusivity?
2. What are the broader implications for society if education becomes another data-driven industry?

Conclusion

In this essay, I have highlighted that datafication of education has far-reaching implications for what counts as learning and who holds authority in the classroom. By privileging measurable outcomes and algorithmic decision-making, it risks narrowing the scope of education and undermining the professional autonomy of teachers. At the same time, it raises urgent questions about the purpose of education in a data-driven world.

To navigate these challenges, we must critically engage with the technologies and policies driving datafication. This means balancing the benefits of data-driven insights with their ethical, pedagogical, and societal consequences. From a historical point of view, Williamson, p.  (2019, p. 3)  powerfully articulates, “The datafication of education, then, is part of a series of historical developments in statistics, state power, quantification, computation, and valuation culminating with the expansion and intensification of digital information systems and ‘big data’”. Only by centring human values can we ensure that education fulfils its promise of empowering individuals and building a more equitable and just society.