Children enter compulsory education at different ages across countries, but the literature broadly agrees that the first year of schooling plays a critical role in shaping educational trajectories and long-term outcomes (Tymms, Merrell, & Henderson, 1999; Tymms et al., 2009). Previous studies using longitudinal design and regression discontinuity conducted in England indicate that children make more learning progress during their first year of formal schooling than in any subsequent year, highlighting the importance of early educational experiences (Tymms, Merrell, & Henderson, 1999, Luyten, Tymms, Jones, 2009). Likewise, several studies indicate that attending preschool is an effective measure to ensure greater equality of educational opportunities. That is, it contributes to the development of cognitive and socio-emotional skills in the short and medium term, impacts future school trajectories, and has a more pronounced effect on vulnerable children from lower socioeconomic backgrounds (Barnett and Boocock, 1998; Campbell et al., 2001; NICHD Early Child Care Research Network, 2006; Peisner-Feinberg et al., 2001; Sylva et al., 2010).

In Brazil, compulsory education begins at age four when children are enrolled in preschool. Recent longitudinal studies have contributed to a better understanding of what children know and can do during the first two years of compulsory education (Bartholo et al.; 2020; Koslinki; Bartholo, 2020). However, despite their methodological rigor, these studies have typically relied on relatively small samples, have often been limited to a single city, and have predominantly focused on public schools. More broadly, there remains a scarcity of high-quality causal evidence from developing countries on the effects of preschool on children learning, particularly those living in poverty.

This paper addresses these gaps by analysing a unique dataset comprising a representative sample of five-year-old children from three Brazilian states. The data are part of an international study conducted in nine countries during the first semester of 2025. In Brazil, a total of 2,520 children attending 210 schools were assessed across three developmental domains: cognitive development, executive function, and socioemotional development.

The paper is guided by two main research questions. The first examines the impact of attending one year of compulsory preschool in Brazil on children’s cognitive development, with a particular focus on emergent literacy and numeracy. A regression discontinuity design is employed to estimate the causal effect of preschool attendance, allowing an assessment of how schools contribute to children’s learning, especially among socioeconomically disadvantaged groups. The large and diverse sample enables, for the first time, an analysis of school effects across the public and private sectors, across different city sizes, and across urban and rural contexts in Brazil.

The second research question aims to describe socioeconomic inequalities in early cognitive development. Preliminary evidence from the city of Rio de Janeiro suggests that, at the start of compulsory education, the gap in cognitive development between children from low- and high-socioeconomic status (SES) families is approximately one standard deviation, corresponding to nearly 12 months of learning. This study provides more robust evidence on these inequalities using data collected from 89 municipalities across Brazil. Understanding such disparities at school entry is particularly relevant for the design and implementation of social and educational policies aimed at reducing inequality at a critical stage of human development.

Early learning is a strong predictor of later success across a wide range of educational and life outcomes. The first five years of life represent a period of both significant opportunity and heightened vulnerability. Cognitive and socioemotional skills developed during this phase have lasting effects throughout schooling and into adulthood. While school quality remains important, strong early learning tends to accelerate subsequent development, whereas a weak start can have enduring negative consequences.

The study analyses data from a survey of a representative sample of five-year-old children in three Brazilian states. To achieve state-level representativeness, a two-stage probability sampling design was employed. In the first stage, early childhood education and care (ECEC) centres or schools were randomly selected within each state. In the second stage, children who met the age eligibility criteria were randomly selected from enrolment lists at each selected institution. In total, 2,939 children were invited to participate, of whom 2,520 were successfully assessed, resulting in a participation rate of 85.7%. Most non-participation was due to the absence of parental consent, followed by cases in which children had left the institution before data collection could be completed. Parents and teachers were also invited to complete structured questionnaires, either online or on paper. Response rates were high, reaching 91.3% for parents and 93.4% for teachers, thereby strengthening the overall quality and completeness of the data. Data collection was conducted by trained researchers, who visited schools and centres multiple times to ensure high response rates and adherence to standardized protocols. In total, 230 researchers were trained and involved in data collection across 210 public and private schools located in 89 municipalities. The study focuses on three core dimensions of early learning, each of which has been shown to have independent and long-lasting effects on children’s developmental trajectories: (a) cognitive skills, including emergent literacy and emergent numeracy; (b) executive function; and (c) socioemotional skills. The analyses presented in this paper concentrate on cognitive outcomes. To estimate the effect of one year of preschool attendance, a regression discontinuity design was implemented. This approach exploits the strict eligibility cutoff based on date of birth (31 March), which determines whether children enter preschool one year earlier or later. Children born just before and just after the cutoff are therefore of very similar age but differ in preschool exposure, creating ideal conditions for causal inference. Multilevel modelling techniques were additionally employed to analyse socioeconomic inequalities in learning outcomes and to account for the hierarchical structure of the data.

Substantial socioeconomic inequalities in cognitive development are already evident at age five. Differences between children in the top and bottom quartiles of the SES distribution correspond to 0.87 standard deviations in emergent literacy and 0.77 standard deviations in emergent numeracy. However, preliminary findings indicate a small but positive average impact of preschool attendance on children’s cognitive development in Brazil. The magnitude of this effect varies across different SES groups. Further analyses identify key correlates of these disparities. Contrary to common assumptions, stratification between public and private schools does not emerge as a primary driver of inequality. Living in highly dense urban areas is associated with poorer developmental outcomes, a finding that warrants further investigation. Gender differences are also pronounced, with boys lagging girls across most domains, except in emergent numeracy. The analysis indicates small changes in those tendencies after one year of preschool. Finally, an index of multiple disadvantages was constructed to estimate the gap between children facing compounded disadvantages and the overall population mean. The implications of these findings for educational policy and future research are discussed, with particular emphasis on the need for early, targeted, and evidence-based interventions. These findings have important implications for policymakers concerned with improving early childhood education quality and equity.

BARTHOLO, T. L. et al. What do children know upon entry to pre-school in Rio de Janeiro? Ensaio: Avaliação e Políticas Públicas em Educação, v.28 n.107, 2020. BARNETT, W. S.; BOOCOCK, S. S. (ed.). Early care and education for children in poverty: promises, programs, and long-term results. Albany: State University of New York Press, 2018. CAMPBELL, F. A. et al. The development of cognitive and academic abilities: growth curves from an early childhood educational experiment. Developmental Psychology, v. 37, n. 2, pp. 231-242, 2001. KOSLINSKI, Mariane Campelo; BARTHOLO, Tiago Lisboa. Desigualdades De Oportunidades Educacionais no Início da Trajetória Escolar no Contexto Brasileiro. Lua Nova, p. 215-245, 2020. LUYTEN, Hans, TYMMS, Peter and JONES, Paul. Assessing school effects without controlling for prior achievement? School Effectiveness and School Improvement, v.20, n. 2, pp. 145 -165, nov. 2009. NICHD. Child-care effect sizes for the NICHD Study of Early Child Care and Youth Development. American Psychologist, v. 61, n. 2, pp. 99-116, jan. 2006. PEISNER-FEINBERG, Ellen S. et al. The relation of preschool child-care quality to children’s cognitive and social developmental trajectories through second grade. Child Development, v. 72, n. 5, p. 1534-1553, set./ out. 2021. SYLVA, Kathy et al. Early childhood matters: evidence from the Effective Pre-school and Primary Education project. London: Routledge, 2010. TYMMS, Peter; MERRELL, Christine; HENDERSON, Brian. Baseline assessment and progress during the first three years at school. Educational Research and Evaluation, v. 6, n. 2, pp. 105-129, jun. 2000. TYMMS, Peter et al. The first seven years at school. Educational Assessment, Evaluation and Accountability, v. 21, n. 1, pp. 67-80, jan. 2009.