Izaak Dekker, Martijn Meeter
6 min
Evidence-based education (EBE) aims to improve teaching by prioritizing interventions proven effective through rigorous research, most notably randomized controlled trials (RCTs). While policymakers have embraced this model, it has faced significant resistance from educational researchers. Critics argue that RCTs are often ill-suited for the complex, context-dependent nature of classrooms, and that the EBE movement risks promoting a technocratic view of teaching that ignores the normative, value-driven goals of education.
The authors categorize the primary criticisms of EBE into three areas:
Rather than abandoning EBE, the authors propose three directions to make it more robust and responsive to these critiques. First, researchers should conduct "context-centered" experiments that explicitly study local factors, mechanisms of action, and implementation fidelity. Second, the field should better leverage existing longitudinal performance data within schools to track long-term effects where RCTs are impractical. Finally, interventions should adopt integrated outcome measures that monitor both primary goals and potential side effects, ensuring that improvements in one area (e.g., test scores) do not come at the expense of others (e.g., student well-being).
Over the past two decades, educational policymakers in many countries have favored evidence-based educational programs and interventions. However, evidence-based education (EBE) has met with growing resistance from educational researchers. This article analyzes the objections against EBE and its preference for randomized controlled trials (RCTs). We conclude that the objections call for adjustments but do not justify abandoning EBE. Three future directions could make education more evidence-based whilst taking the objections against EBE into account: (1) study local factors, mechanisms, and implementation fidelity in RCTs, (2) utilize and improve the available longitudinal performance data, and (3) use integrated interventions and outcome measures.
Sam: It's like the difference between a clinical trial that just records whether the drug worked and one that also tracks adherence, dosing accuracy, and contrainddicating conditions.
Alex: That's a close analogy. The point is that fidelity data is what makes a finding transferable. If you know an intervention works only when teachers have adequate prep time and class sizes below a certain threshold, you can actually tell a district whether the conditions for success are present. Without that, you're just shipping a black box and hoping.
Sam: And the authors are also pushing back on the normative narrowness of current outcome measurement?
Alex: Yes. The multi-dimensional outcome piece matters here. If you only measure math scores, you might be optimizing one outcome while degrading others — student creativity, intrinsic motivation, teacher professional satisfaction. In an open system with multiple legitimate goals, a single-metric trial can produce a finding that's technically correct and practically misleading.
Sam: That's a real tension. But there's an obvious cost to all of this. High-fidelity, context-aware trials with rich instrumentation are substantially more expensive and logistically complex than standard RCTs. Who runs those, and who funds them?
Alex: That's the paper's primary unresolved tension, and the authors don't fully resolve it. They acknowledge that context-centered experimentation is resource-intensive in ways that sit uncomfortably against the reality of educational research budgets. Their partial answer is a shift toward what they call living evidence bases — longitudinal systems that continuously integrate performance data with implementation metrics, enabling real-time optimization rather than one-shot trials. But that's more a direction than a worked solution.
Sam: So the school district that just needs to choose a math curriculum this year is still somewhat underserved by this framework.
Alex: In the short term, yes. The paper is more useful as a reorientation of research culture than as an immediate procurement guide. What it does clearly is identify the cost of the current approach: a body of evidence that looks rigorous but systematically underspecifies the conditions under which findings hold. The authors' strongest point is essentially that the alternative to context-aware EBE isn't some richer, more humanistic practice — it's practice based on no systematic evidence at all. That's the argument for staying inside the paradigm and improving it rather than exiting.
Sam: And that framing matters, because a lot of the critical literature reads as if the choice is between bad EBE and something better. The paper is saying the realistic alternative is worse.
Alex: Which is the most defensible position they take. The field's legitimacy depends on its willingness to engage honestly with the complexity of the classroom — not to retreat from it into cleaner but less valid designs, and not to abandon empirical methods because the system is hard to model. The path forward is more honest instrumentation, not less rigor.
Sam: A measured conclusion, but a necessary one for a field that's been arguing past itself for a while.
Alex: Thanks for listening to ResearchPod.