Every child can learn

In 2022, the Laboratório de Educação led another implementation of the educator training methodology in the municipality of Várzea Paulista (SP)Learning to Study Texts (AET), which invites 4th and 5th-grade Elementary School teachers to incorporate a perspective on the language of texts as a tool for teaching and learning. The goal of AET is to help all students develop the knowledge, skills, and processes necessary for reading and understanding educational texts, acquiring tools to learn how to think, analyze, interpret, form opinions, engage in dialogue, and move freely through the world of knowledge.

The project’s implementation benefited all students and teachers in the Várzea Paulista municipal school system, as well as contributed to strengthening the network’s continuing education structure. For the first time in the project’s history, Pedagogical Coordinators (CPs) were intentionally included in the process. The objective was to expand the conditions for these professionals to assist teachers in implementing classroom activities, incorporating strategies such as practice observation and the creation of records.

A total of 63 professionals participated in the training, including teachers, pedagogical coordinators, and technicians from the Municipal Department of Education. The initiative reached 1,292 children, the entirety of students enrolled in the 4th grade of Elementary School.

Impact evaluation

With the aim of verifying the results of the formative action on children’s learning processes, a quasi-experimental impact evaluation study was conducted in partnership with LEPES (Laboratory for Studies and Research in Social Economics at the University of São Paulo-Ribeirão Preto). Data were collected both in Várzea Paulista and in a neighboring school system located in the metropolitan region of São Paulo, which shares similar sociodemographic characteristics and educational indices with the participating municipality.

The data collection and analysis were intended to determine to what extent the work that the Laboratório de Educação performs regarding academic language contributes not only to expanding children’s linguistic repertoire but also—or not—to the development of other knowledge and skills related to reading comprehension.

The research included more than 2,000 students from 49 4th-grade Elementary School classes in these two municipalities, at two distinct moments: between April 4 and 19, 2022, and between November 16 and December 1, 2022. Two internationally validated reading tests were applied: the Academic Language Assessment (ALA), created by Professor Paola Uccelli of the Harvard Graduate School of Education, and the Progress in International Reading Literacy Study (PIRLS).

PIRLS and its first application in Brazil

PIRLS (Progress in International Reading Literacy Study) is a global assessment system recognized as an international benchmark exam for measuring reading skills in the 4th grade of Elementary School, considered a significant transition point in students’ development as readers. Specifically focused on reading comprehension processes and conducted every five years since 2001, the test is organized by the IEA (International Association for the Evaluation of Educational Achievement), an international cooperative of research institutions, academics, and analysts.

In 2021, the application of PIRLS included Brazil in the sample for the first time, having been implemented on a large scale in public and private schools across the national territory. A total of 187 schools, 248 4th-grade Elementary School classes, and 4,941 students participated, distributed across 24 states and 143 municipalities. In the international sample, PIRLS covered more than 400,000 students in over 13,000 schools across 65 countries. Brazil’s result, 419 points, was considered extremely concerning by experts, placing the country far behind developed nations and demonstrating the effect of the pandemic on education*.

Composed of two texts—one informative and the other narrative—PIRLS measures proficiency levels based on four major dimensions: (i) content analysis and evaluation; (ii) the ability to make straightforward inferences; (iii) the ability to locate and retrieve explicit information; and (iv) interpreting and integrating ideas and information.

Application of PIRLS in the AET project and results compared to Brazil

In 2022, PIRLS was applied within the context of the Learning to Study Texts (AET) program to map the evolution of text comprehension skills among students in Várzea Paulista (the municipality that participated in the training based on Labedu’s methodology) and the control municipality**.

In the initial data collection, students from Várzea Paulista and the control municipality scored below the Brazilian average: 322.19 and 311.96 points, respectively. Students participating in the AET impact evaluation research were, in the pre-intervention collection, ahead only of South Africa (288 points). In the final collection, although the results for both cities remained below the Brazilian average, the gap between them tripled over the year (from 10 to 30 points).

At the end of the AET program implementation, Várzea Paulista achieved an average score of 379.49 points, which is equivalent to advancing two positions in the international ranking, surpassing Morocco (372) and Egypt (378), and approaching Jordan (381). Meanwhile, the control municipality achieved an average score of 348.96 points, maintaining the same position in the international comparison.

Although the distribution of scores in the municipalities was similar in the initial collection and although improvements were observed in both contexts, the highest scores were concentrated in Várzea Paulista. By applying the difference-in-differences method, it was possible to note that AET had a positive and significant effect on the performance of students in Várzea Paulista schools, considering the measures of narrative and informative text comprehension. Regarding informative text, the dimension most impacted by the program was information interpretation, while in narrative text, although all skills showed improvement, locating information and straightforward inference stood out.

Reversing some inequalities

PIRLS features a scale that divides scores into four bands corresponding to students’ reading proficiency levels (advanced, high, intermediate, and low). Brazil is at the “low” level, which means that students are only able to locate explicit information and make simple, straightforward inferences about characters’ actions in predominantly simple texts.

Performing the same exercise to find the score distributions in Várzea Paulista and the control municipality, it was observed that in both cities, in both the initial and final collections, student scores were concentrated in the range below 400, thus following the Brazilian trend. However, the municipality of Várzea Paulista showed a greater evolution in scores than the control municipality, such that the portion of students in the highest score range (advanced level) quadrupled between the two periods.

Furthermore, in the final collection, the portion of students in Várzea Paulista who scored below basic decreased to less than half of the total sample, while in the control municipality, this number still reached about 60% of students. These differences suggest that the AET program positively influenced performance in Várzea Paulista, not only at the lowest proficiency levels but also at the highest.

In other words, the program demonstrated a significant reduction in the percentage of children with comprehension levels below basic, in addition to achieving an increase in comprehension across all levels. Thus, the municipality of Várzea Paulista reached and, at times, exceeded the Brazilian average according to PIRLS. The results were measured after a relatively short intervention of seven months, the duration of the project’s implementation within the network’s own continuing education structure.

In the table below, it is possible to observe the distribution of students across the PIRLS performance scale levels in Brazil, as well as in Várzea Paulista and the AET project’s control municipality in 2022.

Next steps for the Learning to Study Texts project

In 2023, the Learning to Study Texts project remains active in Várzea Paulista schools, this time for 5th-grade Elementary School teachers across the entire municipal network. Consequently, the students—the same ones who lived the experience in 2022—continue to have the opportunity to learn with the 10 model activity sequences available on the project platform.

Concurrent with the project’s maintenance, the Laboratório de Educação is developing a series of materials that will allow for implementation with trainers from other school systems in future editions. The results described above provide evidence that Labedu‘s approach, anchored in methodologies that value quality and the deepening of foundational pedagogical content, presents institutional strategies with the potential to address our country’s most urgent challenges. Thus, we will continue to design mechanisms so that this methodology can be implemented on a larger scale, reaching new audiences in other contexts.

Presentation of implementation results in Várzea Paulista

To present the results achieved by the work throughout 2022, the director of the Laboratório de Educação, Nicole Paulet Piedra, and the project team visited the municipality of Várzea Paulista on April 17. The meeting was held in partnership with the Várzea Paulista Education Management Unit and the Laboratory for Studies and Research in Social Economics (LEPES), and included the participation of directors and coordinators—professionals from the network involved in the project’s implementation. According to the municipal manager, Magali Souza, the initiative aligned with the goals and targets of Municipal Education. “The formative path combines theory and practice with strategies to be developed in the classroom that contribute to the development of students’ reading, aiming for their academic success,” she said. “The training is distinctive because it pays attention to the role of pedagogical coordination alongside the teacher in their classroom performance, thus allowing for monitoring and integration of actions and results,” she concluded. Researcher Angélica Sepúlveda, who monitors the project, also participated in the meeting.

In addition to the in-person event, two live streams were held to share the project’s implementation results. These actions took place on May 23 and sought to reach all teachers who participated in the project throughout 2022, as well as those currently on the path in 2023.

• The AET 2022 results were published by the Laboratório de Educação and are available via the link: https://labedu.org.br/cases/aprender-a-estudar-textos-2022/

• To read the full technical report of the impact evaluation research “Evolution of reading proficiency,” visit: https://labedu.org.br/wp-content/uploads/2025/08/AET-2022-Varzea-Paulista-Avaliacao-de-Impacto_Relatorio-Final.pdf

*Brazil obtained an average of 419 points, slightly above the lowest score on the scale. According to the PIRLS assessment, students in the 400-point range, like Brazilians, are able to locate, retrieve, and reproduce explicitly stated information, actions, or ideas when reading predominantly easy literary texts. They are also capable of making simple, straightforward inferences about characters’ actions. Regarding informative texts, they can locate, retrieve, and reproduce explicit information, as well as make simple, straightforward inferences to provide a reason for an outcome. The assessment data were released on May 16 and show Brazil ahead only of Iran, Jordan, Egypt, and South Africa in the ranking.

**In order to measure the effect of an intervention, it is necessary to have two groups of individuals: one that receives the intervention and another that does not. However, these groups must be sufficiently similar so that the only distinction between them is whether or not they received the intervention.