For nearly 1,500 years, science, with a few notable exceptions, assumed that the universe was organized around the planet Earth. The reason, beyond dogma, was that the geocentric model worked to explain natural phenomena such as eclipses and the seasons. Later discoveries demonstrated the falsity of this model and the need to propose another.

This is just one example of how science is not a set of absolute truths, but a succession of provisional models that help us understand nature with the empirical observations available at any given time.

These models arise from what we call scientific evidence, which can be understood as data, proofs or results, obtained through research, observation and experimentation, that support or refute a hypothesis.

Evidence in education

Although the provisional nature of scientific knowledge affects all sciences, the scope and extent of the evidence generated through the application of the scientific method depend heavily on the field of knowledge. To prevent evidence-based education from becoming just another passing fad , it is essential to learn how to use it effectively.

A priori, we must take into account two essential issues when interpreting the results derived from educational research: its probabilistic nature and its contextual dependence.

First, in the context of the social sciences, and specifically education, scientific evidence is framed within a probabilistic rather than a deterministic approach. This means that we can establish the probability of a phenomenon or relationship occurring, but we cannot definitively affirm its occurrence in every case.

Matters of probability, not certainty

For example, according to the Spanish State Education Indicators System (2023) , the mother’s educational attainment is associated with the likelihood of her children dropping out of school. According to this report, the probability of early school leaving is ten times higher for young people whose mothers have only primary education or less than for those whose mothers have a higher education qualification.

This does not imply that a girl whose mother has only a primary education will necessarily drop out of school prematurely, but rather that she will have a greater probability of doing so. It is interesting to examine this example, as it also gives us an idea of ​​the provisional nature of the research results we discussed earlier, since, although this effect persists, it has diminished in recent years compared to previous decades.

Context dependence

In addition to its probabilistic nature, educational evidence has another key characteristic: its strong dependence on context. Therefore, it is not possible to establish general rules beyond the trends observed in the studies conducted: what is shown to work in one context may not work in another.

For this reason, when we read scientific evidence in education, it is crucial to understand the description of the context in which the research is conducted so that we can interpret the scope of the evidence it generates.

Primary studies and meta-analyses

Empirical research in education is primarily gathered in what we call primary studies. These generally present evidence that, as we mentioned earlier, is limited to a specific context. For example, research conducted with university students will restrict the applicability of its results to students at this level. Therefore, we could not extrapolate the conclusions of this study to students at earlier educational levels.

In addition to primary studies, synthesis studies are conducted. Among these, meta-analyses are particularly relevant. These studies aim to aggregate the results obtained from research on the same topic. The goal is to find averages of the effects reported in the primary studies, and the consequence is that the robustness of the conclusions increases, since they depend on a set of studies and not just one.

This type of work also allows for a critical evaluation of primary studies. For example, a meta-analysis by Professor Samuel Parra concludes that to obtain generalizable results on the effects of the Montessori method, it is necessary to undertake studies with greater methodological rigor than those published to date.

Bringing the evidence into practice

Within the scientific community, there is a growing, deliberate effort to produce meta-analytic studies like those mentioned above, which generate solid evidence that educators can use to design their teaching. While there are many examples, we could highlight the work of Professor Samuel Parra and Professor Marta Ferrero .

Increasingly, teachers at various levels are also becoming interested in the use of evidence in education. Some prominent examples include Albert Reverter , a teacher and founder of the blog El McGuffin Educativo , and Héctor Ruiz Martín , who work to bridge the gap between scientific evidence and teaching practice.

Ultimately, in an increasingly complex educational context, evidence-based education is not about seeking universal solutions, but about making informed, critical, and contextualized decisions. Evidence does not replace professional teacher judgment, but it can certainly strengthen it.

Author Bios: Diego Ardura is a Full Professor (Education) and Arturo Galán is Professor of Research and Diagnostic Methods in Education both at UNED – National University of Distance Education