What about?
Questions that trouble teachers
AUGUST/SEPTEMBER 2026 | n.º 8 | In this section, experts invited by YVIRÁ answer questions submitted by teachers who are part of the National Science for Education Network (Rede CpE) as “Friends of the Network”.
For this edition, Daniel Domingues dos Santos, professor at the Faculty of Economics, Administration and Accounting of Ribeirão Preto at the University of São Paulo (FEA-RP/USP), delves into a challenge as well-known as it is persistent when we talk about science in education. After all, how to reconcile academic methodological rigor and the development of research that is relevant from a social point of view, effectively thinking about schools? Check it out.
IMAGE: ADOBESTOCK
Theory, practice and Angry Birds
“How to balance the methodological rigor of academic research with the complexity and variability of the school environment, preventing excessively controlled studies from becoming distant from teaching practice?”
Learning to read, for example, requires learning to decode, which means being able to navigate a text fluently without much effort to join letters into syllables, syllables into words, and words into sentences. It is undeniable that this is one of the objectives of education, but not the only one.
The great difficulty lies in conducting research that is both socially relevant and academically rigorous with aspects of the educational process where the specificity of the context matters a great deal, and the replicability (or external validity) of conclusions obtained in other contexts is low.
Excellent question, and one that allows me to discuss a little about how far I believe we can do science in education. First, note that, for many people, learning means developing skills to the point where the individual can perform new tasks with high precision and little expenditure of energy. Learning to read, for example, requires learning to decode, which means being able to navigate a text fluently without much effort to join letters into syllables, syllables into words, and words into sentences. It is undeniable that this is one of the objectives of education, but not the only one. It is equally important to be able to use these skills to gain advantages in the context in which one lives.
Learning to read, for example, requires learning to decode, which means being able to navigate a text fluently without much effort to join letters into syllables, syllables into words, and words into sentences. It is undeniable that this is one of the objectives of education, but not the only one.
Continuing with the example of literacy, it is useless to automate reading without understanding the text in the circumstances in which it is presented. The first part (development and automatization of skills) is often necessary in different cultures and occurs at a similar stage of the learning process. These characteristics are better suited to the scientific method, which involves formulating questions (e.g., how does a human being learn?); developing hypotheses (e.g., is a certain pedagogical strategy more effective than others); experimentation (an empirical strategy to test the proposed hypothesis); and replication (if this hypothesis is indeed true, it should also be validated in other contexts).
Caution in the Use of Scientific Evidence
In this field, research can proceed with greater rigor and still have relevant implications for classroom practice. The great difficulty lies in conducting research that is both socially relevant and academically rigorous with aspects of the educational process where the specificity of the context matters a great deal, and the replicability (or external validity) of conclusions obtained in other contexts is low.
The great difficulty lies in conducting research that is both socially relevant and academically rigorous with aspects of the educational process where the specificity of the context matters a great deal, and the replicability (or external validity) of conclusions obtained in other contexts is low.
In these cases, the use of scientific evidence must be done with more caution, which recalls a famous recent controversy: Nobel laureates Angus Deaton and Esther Duflo (see more here and Angry Birds Economics. Following Deaton’s argument, with which I agree, we should use scientific evidence just as we use our knowledge of physics in the Angry Birds game, where the objective is to hit a bird with a slingshot.
We know that the positioning of the weapon and the tension placed on the elastic determine the trajectory of the stone, but we don’t have the exact formula on the screen in that context, and therefore we adjust the shot by approximations until success. In other words, the use of scientifically documented experiences must be adapted to the context through successive approximations, respecting cultural and institutional characteristics. From the perspective of evidence production, I think that if we can be rigorous, all the better.
