Could learning analytics systems reshape education, work, or finance over the next decade?

How learning analytics can improve education—and where the evidence is strongest

Learning analytics (LA) uses data from digital learning platforms—like how many times a student accesses materials, how long they spend on tasks, and their quiz scores—to understand and optimize learning. The most direct evidence from these papers shows that LA can boost academic performance, personalize feedback, and help educators identify struggling students early. For instance, a study in entrepreneurship education with 30 students found that implementing an LA package led to 'substantial performance gains' [9]. Another study at a private university found a positive relationship between student engagement (measured via LMS logs) and academic performance, and that an analytics plug-in helped instructors identify at-risk students and provide real-time feedback, improving outcomes [14]. In medical education, a longitudinal study of dermatology residents analyzed nearly 5,000 responses and found that LA helped identify challenging diagnostic cases and optimize review schedules, with third-year residents improving accuracy from 83% to 91% over three months [1].

The evidence is strongest in higher education and professional training, where digital platforms are common. A systematic review of 150 papers found that LA, combined with virtual and augmented reality, improved learners' understanding, performance, and knowledge retention across K-12, higher education, and vocational training [3]. Another review of 27 articles concluded that LA interventions within learning management systems (LMS) provide empirical evidence of positive impacts on teaching and learning [10]. However, the same review notes that many studies are small-scale or conducted after courses end, limiting their generalizability [10]. Across these studies, the larger and more rigorous ones consistently show benefits, but the effect sizes vary and are not always dramatic.

The big caveats: privacy risks, labeling students, and weak evidence in work and finance

Despite the promise, learning analytics comes with serious ethical and practical challenges that could limit its reshaping power. A critical review of 18 studies in medical education identified four major ethical concerns: trustworthiness of data, reliability of methodology, privacy and confidentiality, and the risk of labeling learners as 'problematic' [2]. The authors used a biomedical ethics framework to argue that LA could harm learners' autonomy and justice if not carefully implemented [2]. Similarly, a review of K-12 education found that while many see benefits like more equitable instruction and individualized learning, others are unconvinced due to lack of evidence of improved outcomes and concerns about data misuse or misinterpretation by educators [6]. Students themselves are aware of these issues: a survey of health professions students found they were generally agreeable to LA but expressed concerns about privacy, confidentiality, and data security [13].

Crucially, the evidence for LA reshaping work or finance is essentially absent from these papers. None of the 15 studies directly examine LA in workplace or financial settings. The closest is a paper on entrepreneurship education, which suggests LA could prepare graduates for 'competitive and dynamic economic environments' [9], but this is indirect. Another paper discusses LA in the context of global competencies and intercultural learning, hinting at workforce readiness [8], but provides no data. The papers focus overwhelmingly on education (K-12, higher education, medical training), and even there, many studies are descriptive or post-hoc rather than predictive or interventional. For example, a scoping review of 65 papers in health care education found that only 11 used LA for at-risk intervention, 5 for feedback, and 3 for adaptive learning—most just described student interactions [11]. This suggests that while LA has potential, its practical reshaping of education is still in early stages, and its impact on work and finance is speculative at best.

What needs to happen for learning analytics to truly reshape these fields

For learning analytics to move from promising to transformative, several conditions must be met. First, institutions need to adopt a data-driven culture and invest in computational infrastructure and human capacity—a point made by multiple reviews [12][15]. One paper notes that education 'lacks computational infrastructure and human capacity to fully exploit the potential of big data' [15]. Second, ethical frameworks must be built into LA systems from the start, not as an afterthought. The Integrated Framework for Learning Analytics in Global Competence Education (IF-LAGCE) proposes combining global competency alignment, culturally contextualized data, ethical responsibility, and human-analytics collaboration [8]. Third, LA needs to move from descriptive analytics (what happened) to predictive and prescriptive analytics (what will happen and what to do about it). Currently, most studies use descriptive statistics—89% in one review [11]—and only a few use machine learning for prediction [5].

Finally, the evidence base needs to expand beyond small, single-institution studies. Most of the papers here are systematic reviews or small-scale studies; only a few are large-scale or longitudinal. For example, the dermatopathology study tracked residents over three months [1], and the entrepreneurship study used a pre-test/post-test design with 30 students [9]. While these show promise, they are not enough to guarantee that LA will reshape entire systems. The papers themselves call for more research on long-term outcomes, real-time interventions, and cross-institutional implementations [4][7][10]. In short, learning analytics could reshape education significantly over the next decade, but only if institutions address privacy, build ethical safeguards, invest in infrastructure, and move from describing to predicting and intervening. For work and finance, the evidence is too thin to make any firm prediction.

About These Sources

This answer is built on 15 peer-reviewed studies — published from 2020 to 2025, 9 from 2024 or later, 5 in Q1 journals, collectively cited 427 times — selected as the most relevant from 15 studies that passed quality screening, drawn from 76 papers retrieved from a database of over 500 million.

Sources used in this answer

Related research questions