School trajectory of Pedro II High School: an exploratory study on repetition

Abstract

This article investigates how repetition occurs at Colégio Pedro II (CPII), an ensemble of federal public schools considered of excellence, which has sought to democratize access to underprivileged groups through a series of inclusion initiatives. The study uses longitudinal data from 2015 and 2017 Censos Escolares to analyze if a cohort of high schoolers' school trajectories is associated with specific social characteristics. The main results indicate that repetition tends to limit the democratizing and inclusive effects of actions adopted by CPII in the last 30 years, which primarily affect black men with an age-grade gap. The results reaffirm repetition as an excluding practice that broadens social and educational inequalities.

Keywords
Socio-educational inequalities; Longitudinal study; Censo Escolar; Colégio Pedro II

Resumo

O objetivo do artigo é investigar como se opera a repetência no Colégio Pedro II (CPII), um complexo de escolas públicas federais consideradas de excelência, que vem buscando democratizar o acesso a grupos sociais desfavorecidos por meio da adoção de uma série de iniciativas de natureza inclusiva. Para verificar se a trajetória escolar de uma coorte de estudantes do ensino médio está associada a determinadas características sociais, o estudo faz uso de dados longitudinais dos Censos Escolares de 2015 a 2017. Os principais resultados indicam que a repetência tende a limitar os efeitos democratizantes e inclusivos das ações adotadas pelo CPII nos últimos 30 anos, ao atingir majoritariamente estudantes do sexo masculino, negros e com defasagem idade-série. Os resultados reafirmam a repetência como uma prática excludente que amplia as desigualdades educacionais e sociais.

Palavras-chave
Desigualdades socioeducacionais; Repetência; Estudo Longitudinal; Censo Escolar; Colégio Pedro II

Introduction

The phenomenon of repetition is among the most severe educational problems in the country. It is connected to low education quality, high age-grade gaps, and school dropout rates. It is the permanence of students in the same Grade the following year. Repetition is commonly due to two main factors: 1) failure – when students' permanence in the Grade is due to academic performance (lack of learning and/or absence in class); 2) dropout – when students leave school in the middle of the school year and, on the following year, returns to the same Grade they abandoned. Though less common, repetition can also happen in mobility situations, i.e., when students move from one school to another. In these cases, there are two possible situations: the school that welcomes the new students recommends they repeat the Grade they finished in the last institution, or students (and/or their families) choose to repeat the Grade to enroll in the new school.

The repetition theme is the object of several studies (Correa et al., 2014; Crahay, 2006; Ortigão & Aguiar, 2013; Ribeiro, 1991; Sa Earp, 2009), which, through different methodologies and theoretical approaches, seek to understand their conditions, dynamics, and results. These studies are unanimous in pointing out that repeat students do not learn more when attending again the same school year. Besides this, they show that repetition leads to school lag and increases the probability of retention, dropout, and migration to Youth and Adult Education hindering, besides learning, other aspects such as socialization and motivation. The effects of repetition become even more harmful when observing that repeaters are systematically students with similar characteristics: mostly from low-income families, black, male, and living in socially vulnerable territories, such as rural areas and the outskirts of urban centers.

Repetition enters the Brazilian field of public policy in the 1930s. However, the theme only started to be better understood and gained relevance in governmental actions in the mid-1990s, with the model Profluxo (Klein & Ribeiro, 1991; Ribeiro, 1991), leading to the creation of policies to fight repetition, such as cycle policies and flow correction, which started to be implemented in several contexts.

The Profluxo model, developed by Klein and Ribeiro (1991), is a milestone in understanding how repetition occurs in the Brazilian educational system. This model evidenced the existence of a "pedagogy of repetition," in which students' repetition was interpreted as a "pedagogical methodology that remains in the system, despite all efforts to universalize basic education in Brazil” (Ribeiro,1991, p. 18).

After 30 years of applying the Profluxo model in educational statistics and the first studies denouncing the extremely high rates of repetition – with the pioneering works of Klein and Ribeiro (1991) and Ribeiro (1991) –, we can see that, regardless of the efforts to universalize access to K-12 education, the repetition practice remains a challenge to increase educational opportunities, mainly in the final phase of high school. Data from the Instituto Nacional de Estudos e Pesquisas Educacionais Anísio Teixeira (Inep) indicate a significant improvement in expanding access to elementary, middle, and high schools. However, this expansion did not happen equitably and still has unacceptable performance rates, mainly in the last phase (Todos pela Educação, 2021).

For example, the data of Anuário Brasileiro da Educação Básica 2021 (Todos pela Educação, 2021) point out that the gross rate of young people between 15 and 17 years old enrolled in school has grown continuously since 2012, going from 86.6% to 94.5%, in 2020. In high school, the net rate of young people of this age range went from 61.0% to 75.4%. The differences between gross and net rates occur because some of these young people stayed in elementary and middle school while others did not continue their studies. If the number of young people between 15 and 17 years old outside school was 481,884, meaningful inequalities of access appear when considering the dimensions of income, race, and residence. To the young people between 15 and 17 years old belonging to the segment of the 25% poorer homes, the access to High School is only 70.5%; for the 25% richer, the access reaches 93.7%. This pattern repeats considering students' race. For white young people, the access to high school in this age range is 81.4%; for black young people, it is 70.8% and 71.8% for brown ones. Among the residents of urban areas, this percentage is 77.5%, contrasting to the percentage of rural young people, which is 65.1%.

Considering these data, this study investigates how repetition occurs in a set of federal public schools considered of excellence- the Colégio Pedro II (CPII). This work aims to identify if CPII follows the national repetition standard. That is, this study is interested in knowing if repeat students in CPII have the same social profile as repeat students in Brazil. This social profile will be analyzed based on the following characteristics: CPII type of admission, sex, race, and age.

Though characterized since its beginning in 1837 as an elite school, CPII has engaged in the last 30 years in the movement to broaden access and learning opportunities for underprivileged social groups. The widening of educational opportunities in CPII involves initiatives of different natures, from the creation of new campi in territories in the city and the state of Rio de Janeiro with the highest socioeconomic inequalities, from widening school offer to also encompass childhood education and elementary and middle school, up to the incorporation of new forms to access it and new models of teaching, such as Integrated High School and the Programa de Educação de Jovens e Adultos (Proeja- Program of Youth and Adult Education).

Regarding the new forms to access, CPII adopted in 2004 a singular measure that changed the history of its admission process and the student's profile. From that year onwards, half of the places in the admission process to Grade 6 were reserved for public school students and the other half for private school students. This pioneering measure to CPII was complemented in 2012 when the admission for regular and integrated high schools followed Law n.^o 12.711/2012 (Quota Law), which incorporated the place reserves the criteria of race, income, and, in 2016, handicap.

Despite these advancements, the increase of educational opportunities in CPII through the Quotas Law was only restricted to the first year of high school. In the other phases of K-12 education offered by CPII, the previous criteria were kept for admission in the first year of elementary education using a public raffle. In Grade 6, the admission process was organized based on social quota (public school), following the guidelines in action since 2004¹ (Dargains, 2015).

Furthermore, selection mechanisms continue to operate throughout students' schooling trajectory in CPII, such as the practice of academic dismissal in which the student who failed the same school year twice in a row could not renew the enrollment and, consequently, was expelled from school. As repetition tends to mainly affect poorer, black, and brown students, this practice ended up excluding from school the same profile of students whom the Quota Law favored. Academic dismissal was extinguished by Ordinance n.^o 1.343 from 2015, supported by the broader educational legislation, notably in the Federal Constitution (CF/88), the Law of Directives and Bases of National Education (LDB/96), and at the Statute of the Child and Adolescent (ECA/90). However, recent studies, such as Couto (2018), show that students with two repetitions continue to leave the school and are considered dropout students.

In this sense, CPII seems to adopt actions that can promote the increase of education opportunities for underprivileged social groups. At the same time, it uses strategies that seek to perpetuate its excellence image, through mechanisms to select (the best) students.

Even though the reserve of places for specific social groups is an important measure, it was not possible to see the profile of repeaters according to the variable “quota” or “non-quota” student because this information is not available in Censo Escolar database. As it was not possible to identify repetition according to the variable "place reserve," the failure mapping was done according to the criteria "admission process." There are two ways to be admitted at CPII High School. One is a selective process; the other is the transition of middle school students from CPII itself to High School, commonly known as graduates.

Hence, due to the nature of the data, the study aimed to characterize the profile of students with a school trajectory with repetition, seeking to see if this profile is similar to the one found at the national level. The analysis focuses on four characteristics: race, gender, age-grade gap, and CPII admission procedures. When checking if the school trajectory of a cohort of high school students is associated with certain social characteristics, the current study aligns with research that investigates the relationships between repetition and the reproduction of inequalities.

There are three justifications that ground the pertinence and relevance of the present study. The first is the lack of studies investigating the operationalization of repetition in federal public schools of excellence. Most studies on repetition focus on other school phases, education, and school systems that generally have high repetition rates and are not characterized as schools of excellence when considering not only the quality indicators of Inep, such as Índice de Desenvolvimento da Educação Básica (Ideb) and the Sistema de Avaliação da Educação Básica (Saeb) but also because they are public institutions that distinguish themselves from other due to their quality reputation. These are few schools, mainly kept by the federal government or connected to public universities, offering exceptional conditions compared to other schools from the "common" public system (Costa & Koslinsky, 2011).

The second justification refers to the fact that CPII is an institution that, as previously indicated, has sought to democratize students’ access and permanence in the last years. Some measures were adopted to do so, such as the end of academic dismissal and the reservation of places for specific social groups.

Finally, the study is justified by the methodology applied, which used longitudinal data from Censo Escolar and an intersectional perspective. The methodology will be detailed later.

Besides this introduction, the work is organized into four sections. The next one presents the main data about repetition in the national context. Afterward, the methodology of this study will be presented. The penultimate systematizes and discusses the results found at CPII, comparing them with the data found nationwide and statewide, as well as studies on the theme. The last one presents brief final remarks.

The repetition phenomenon in Brazilian high school

Since the late 1990s, educational data highlight the broadening of K-12 Education access in all school phases. Elementary and middle school was universalized in the late 1990s, and access to childhood education and high school increased significantly. Despite this, several studies (Alves et al., 2016; Andrade & Laros, 2007; Ernica & Rodrigues, 2020; Menezes-Filho, 2007; Soares, 2004) which used data from Censo Escolar and Saeb, point out the persistence of opportunity inequalities in all school phases. Despite entering school, learning opportunities and a regular school trajectory do not seem to be the same for all students. Education indicators, such as performance rate and age-grade gap, show that a meaningful percentage of students have irregular trajectories, marked by dropouts, repetitions, and gaps.

In high school – our analytical scope – we observed in the last ten years an increase in the average pass rates, a decrease in dropout rates, and a slight variation in failure rates, as seen in Table 1. Considering the analyzed period, the pass rate increase was 10.2%, the decrease in dropout rate was 6.7%, and, despite oscillations, failure rates decreased 3.5%. Notwithstanding the improvement of these indicators, the 2019 data, the year before the Covid-19 pandemic, was still alarming, as the pass rate was only 86.1%, indicating that a meaningful percentage of young people (13.9%) had difficulties in their school trajectory.

Table 1
High School performance rates – Brazil– 2009-2019 in % (Todos pela educação, 2021, p. 64)

Performance	2009	2010	2011	2012	2013	2014	2015	2016	2017	2018	2019
Pass rate	75.9	77.2	77.4	78.7	80.1	80.3	81.7	81.5	83.1	83.4	86.1
Failure	12.6	12.5	13.1	12.2	11.8	12.1	11.5	11.9	10.8	10.5	9.1
Dropout	11.5	10.3	9.5	9.1	8.1	7.6	6.8	6.6	6.1	6.1	4.8

The Painel Trajetória Escolar developed by Unicef (2019) based on data from Censo Escolar points out that Brazilian High School, the failure rate for young men was 9.36% and 5.43% for young women in 2019. White and Asian students' rates were 5.75% and 7.16%, respectively, while black students reached 10.52%, brown 8%, and Indigenous 10.66%. Regarding location, the Painel shows that this rate was 8.37% in rural areas and 7.29% in urban ones in 2019.

These results confirm that repetition is not a random phenomenon (Alves et al., 2007; Ortigão & Aguiar, 2013) because some social profiles, notably boys self-declared black, are more likely to fail. Therefore, it is a practice supported by social and cultural markers that characterize students' profiles, and that is responsible for the learning gaps and school repetition that affect this population. Further, we discuss if the profile of repeaters in CPII is similar to the one presented by Unicef's Painel Trajetória Escolar.

In the state of Rio de Janeiro, failure and dropout rates are higher when compared to the national average, as seen in Table 2. When distributing the performance rates among the school years of High School, using 2019 as a reference, we can see that Grade 10 has the highest failure and dropout rates, decreasing in the other school years. In line with the studies by Silva et al. (2016) and Buss Gesser and Martins (2016), this scenario indicates that Grade 10 is the greatest selective barrier in students' high school trajectories and one of the main bottlenecks.

Table 2
High School Performance Rates per Grade- Brazil and state of Rio de Janeiro – 2019 (Adapted from Inep, 2019a; Todos pela Educação, 2021)

Performance rates	Brazil	RJ	Brazil	RJ	Brazil	RJ	Brazil	RJ	Brazil	RJ
	Total		Grade 10		Grade 11		Grade 12		Grade 132
Pass rate	86.1	81.7	80.5	73.7	87.3	83.2	92.5	91.6	88.9	90.5
Failure	9.1	12.5	13.4	18.8	8.1	11.2	4.5	4.8	7.2	8.4
Dropout	4.8	5.8	6.1	7.5	4.6	5.6	3	3.6	3.9	1.1

The data of Censo Escolar 2019 show that in CPII, when considering all high school modalities (regular, integrated, and YAE), the total pass rate is 89.1%, a 9.0% failure rate, and a 1.9% dropout rate. Compared to the performance rates in the state of Rio de Janeiro and Brazil, CPII has a similar percentage of failure, higher percentages of pass and smaller dropout ones.

When analyzing the data for each campus, there are significant variations in the performance rates: for example, Humaitá II has a high failure rate (14.1%), higher than the national one (9.1%) and the state of Rio de Janeiro (12.5%), while campus Niterói has a failure rate of 6% and a 0.2% dropout rate, thus, lower to other CPII units (Figure 1) and to other national and state schools. A possible explanation for this difference between the two school unities, to be seen in future studies, is that the high school offered on campus Humaitá II occurs only on the night shift – commonly attended by working students and with irregular schooling trajectories in elementary and middle school, while Niterói offers classes during the day and most students joined CPII through the selective process, straight to high school, as the unit does not offer elementary and middle school.

Figure 1
Performance rates in CPII High School by campus – 2019 (Inep, 2019a)

When analyzing the performance percentage per school year, it is possible to see that the first two years have the highest failure rates, which vary depending on the campus. Table 3 details this phenomenon. Campus São Cristóvão III stands out presenting, in Grade 10, the lowest pass rate – only 79% – and highest dropout – 6.9%. In Grade 12, campus Humaitá II had the lowest pass percentage – only 93.8% – and the highest dropout – 1.4%. Regarding the failure rate, the campi Centro and Humaitá II have the highest failure rates in Grade 12.

Table 3
Performance rates in CPII High School– 2019 (Inep, 2019a)

Campus	Pass rate			Failure rate			Dropout rate
	Grade 10	Grade 11	Grade 12	Grade 10	Grade 11	Grade 12	Grade 10	Grade 11	Grade 12
Duque de Caxias	83.4	92.9	97.0	14.0	6.4	3.0	2.6	0.7	0.0
Niterói	89.8	93.0	99.4	9.6	7.0	0.6	0.6	0.0	0.0
Centro	91.3	90.2	95.1	7.1	7.6	4.9	1.6	2.2	0.0
Humaitá II	80.4	79.0	93.8	16.0	18.3	4.8	3.6	2.7	1.4
Tijuca II	82.1	86.7	99.4	16.4	10.8	0.6	1.5	2.5	0.0
Engenho Novo II	82.5	90.1	98.5	14.9	9.1	1.5	2.6	0.8	0.0
São Cristóvão III	79.0	85.4	96.5	14.1	11.2	2.2	6.9	3.4	1.3
Realengo II	84.7	88.4	99.0	13.4	8.0	0.0	1.9	3.6	1.0

Compared to the nationwide and statewide data, we can see that, in general, the pass rates in the three years of high school in CPII are greater than in the country and the state. In Grade 12, all campi had higher pass rates than in Brazil and the state; in Grade 11, the campus Humaitá II had a pass rate smaller than in Brazil and the state; in Grade 10 all campi had higher rates than the state, and only São Cristóvão III and Humaitá II showed smaller values than Brazil.

This analysis indicates that, though some campi have lower performance rates than Brazil and the state of Rio de Janeiro in Grade 10 and 11, in 2019, the performance rates of CPII high school students were better than the ones in Brazil and Rio de Janeiro. Table 3, below, shows the different performance rates among different CPII campi and Table 4 shows the average performance rates in Brazil, in the state of Rio de Janeiro, and in the CPII campi offering High School.

Table 4
Average of Performance Rates in High School– Brazil, Rio de Janeiro, Colégio Pedro II (Adapted from Inep, 2020)

Performance rates	Brazil	RJ	CPII	Brazil	RJ	CPII	Brazil	RJ	CPII	Brazil	RJ	CPII
	Total			Grade 10			Grade 11			Grade 12
Pass rate	86.1	81.7	89.9	80.5	73.7	84.2	87.3	83.2	88.2	92.5	92	97.3
Failure	9.1	12.5	8.4	13.4	18.8	13.2	8.1	11.2	9.8	4.5	4.8	2.2
Dropout	4.8	5.8	1.7	6.1	7.5	2.7	4.6	5.6	2	3	3.6	0.5

In the discussion about repetition, the measure age-grade³ gap is also an indicator to be considered because it affects students' trajectories and reproduces school inequalities. In 2019, the age-grade gap in Brazilian high schools was 26.2%, reducing 8 percentage points compared to 2010, which registered 34.4% (Todos pela Educação, 2021). In the state of Rio de Janeiro, this rate is higher than the national one, with 34.9% of students from the state in a situation of age-grade gap. When analyzing this data by education system, we see that the state system has the highest rate (41.4%), followed by the municipal system (34.5%), federal (26.9%), and private (11.7%).

When observing the data age-grade gap considering 2019, CPII had high percentages, mainly at campi Humaitá II, São Cristóvão III, and Engenho Novo II, which continues in all school years. As mentioned above, the federal system has a distortion age-grade rate of 26.2% and, at least, 4 campi of CPII have higher rates than the average in the federal system: Humaitá II (36.7%), São Cristóvão III (34.1%), Engenho Novo II (33.3%), and Centro (29.5%). At the campi Centro and Realengo II, there was an increase in the age-grade gap during all years, higher in Grade 12. Once more, the campus Niterói stands out with the lowest age-grade gap, which can be explained by the hypothesis already raised: in this unit, students are mostly admitted to High School through an admission process because the school unit does not offer Elementary and Middle School. The few newcomers that studied Elementary and Middle School in CPII at campus Niterói were those transferred from other units.

Table 5
Age-grade gap in High School by Grade, at CPII - 2019 (Inep, 2019b)

Campus	Total	Grade 10	Grade 11	Grade 12
Humaitá II	36.7	38.4	39.5	29.7
São Cristóvão III	34.1	39.8	33.9	27.5
Engenho Novo II	33.3	42.0	24.8	23.3
Centro	29.5	23.8	29.9	35.0
Realengo II	26.9	24.7	25.3	32.2
Tijuca II	24.3	27.4	30.1	14.1
Duque de Caxias	22.0	22.6	25.0	17.8
Niterói	15.3	14.7	16.4	14.6

Considering this scenario, we can see that, even in a set of excellence schools seeking to democratize its access, the rates of failure, dropout and age-grade gap are meaningful and make students struggle to finish their basic school trajectory. Below, we present the methodology used in the research and discuss the main results.

Methodology

In this study, we associate the enrollment situation to students' sociodemographic profiles using a secondary and longitudinal database of Censo Escolar da Educação Básica referring to the microdata of the years 2015, 2016, and 2017. The data was accessed by downloading the microdata at Inep site⁴, selecting a student cohort that was admitted to CPII High School in 2015 and, thus, following the enrollment situation of these students in 2016 and 2017. In the years analyzed, considering the perfect flow, students should be enrolled in Grades 11 and 12, respectively. However, in Censo Escolar's open data, there is no information about high school graduation. From 2018 onwards, students' foreign key (“co_pessoa_física”) was altered compared to previous years, which makes it impossible to follow the same students and, hence, longitudinal analysis⁵. Due to this limitation, we chose the cohort 2015 because it was the most recent year in which it was still possible to follow students’ trajectory until 2017, the last year that this foreign key (“co_pessoa_física”) could be used to check this education phase for the same students in each year of Censo Escolar.

Despite this limitation, the descriptive and exploratory analysis sheds light on the (re)production of inequality standards in students' school trajectories. This can relativize the effectiveness of the efforts towards widening the opportunities to access and permanence of underprivileged groups in a public school of excellence.

In this study, we used the database of secondary data from Censo Escolar da Educação Básica regarding the microdata of 2015, 2016, and 2017. We also selected the 2014 base to check which students already attended CPII this year (that is, if the students who entered high school in 2015 were in Grade 9 at CPII in 2014) and which students entered in 2015 through the selective process⁶.

After we selected, on the database of Censo 2016 and 2017, all CPII students and added to the 2015 base the pertinent information through a foreign key “co_pessoa_física." We searched in these databases the variables “município de endereço” [city of residence]; “etapa de ensino” [education phase], and “co_unidade” [co_unit] to see if the newcomers stayed at CPII, if they changed campus, and if they progressed to the next Grade. There were some duplicate cases, such as students connected to two types of classes (regular and specialized educational service or complementary activity). To solve this duplication cases, we selected students connected to a regular class⁷.

In 2015, there was a total of 1,689 newcomers in the regular CPII High School. Out of these, the majority (62.1%) had already studied at CPII the previous year, and 37.9% joined through the selective process. In total, eight campi offered regular high school during the day (morning and/or afternoon), except Campus Humaitá II, which offers only night classes. Figure 2 presents the distribution of newcomers to the regular high school in the 8 CPII campi.

Figure 2
Percentage of newcomers’ distribution in regular High School at CPII – 2015 (Adapted from Inep, 2016)

According to these data, São Cristóvão III, Realengo II, and Humaitá II are the three campi with the highest percentage of students who joined high school in 2015. The campi Centro and Engenho Novo II had the smallest percentage of newcomers.

The following sections will present the profile of high school newcomers in CPII and analyze students’ school flow by campi, sex, race, and age during the three years of regular high school. The results are compared to the national and Rio de Janeiro state data and the literature about the theme whenever possible.

The data on CPII repetition data about race and sex are analyzed together into an intersectional perspective. Thus, we seek to articulate different social markers of difference, such as gender and race, in the configuration of inequalities. The concept of intersectionality was coined by the North American Kimberle Crenshaw in 1989 in the scope of Black feminism as a criticism of feminist theory. This intersectional approach dialogues with the multiple experiences and manifestations of inequality, through the overlaying of categories such as sex, race, and social class (Crenshaw, 1989; Medeiros, 2019). According to Giddens and Sutton (2017, p. 155), this perspective can be understood as a theory and a methodology targeting the study of the many ways differences and power relations can interconnect and structure society, producing discrimination and inequalities.

Results and discussion

In this section, we present the research findings. We start outlining the profile of students who joined CPII High School.

Who are the students joining high school at CPII?

In 2015, 1,689 students were admitted in the Grade 10 of CPII high school; most of them (90.3%) aged between 15 and 17 years old (27.6% were 15 years old; 45.9% were 16, and 16.8% were 17 years old). Regarding sex, 54.9% are women and are the majority in all campi, except Tijuca II, in which most newcomers are men. At campi Niterói and Duque de Caxias, women are over 60%. Regarding race, 48.4% did not declare it, 34.9% were self-declared white; 13,9%, brown; 2.6% black, 0.2% Asians, and no student declared to be Indigenous. Figure 3 presents students' racial profiles by campus.

Figure 3
Percentage of newcomers, according to race by campus (Inep, 2016)

We can see a high percentage of newcomers that did not declare their race, mainly on campus Niterói, which hinders the identification of a more precise racial profile in this cohort. Studies point out that one difficulty to measure students’ race refers to the tools to produce racial data (Senkevics et al., 2016). On the one hand, some studies point out fluctuations on the self-declaration of Enem participants. Senkevics (2022), for instance, notes a change in the racial declaration of students participating in Enem between 2010 and 2016, highlighting mainly an increase in the self-declaration as black and brown. The author presents three possible explanations for this phenomenon of racial reclassification: the increase of black and brown students who graduated high school, the changes in the structure of incentives to participate in the exam, and candidates' repeated participation, i.e., those who enroll in Enem in several years. These factors help to explain the alterations in racial self-declarations, as well as the high percentage of non-declarations.

Campi Realengo II and Duque de Caxias had the highest percentage of new students self-declared black or brown, and Humaitá II had the highest percentage of self-declared white newcomers. In the analysis per campus, once more Humaitá II has the highest percentage of self0declared white students (69.85%). In contrast, the campi with the highest percentage of black, brown, and Indigenous students were Duque de Caxias (60.81%) and Niterói (45.84%). The latter offers only high school. Campus Realengo II, which offers middle school, has a significant percentage of black, brown, and Indigenous students (42.35%). These data point to an unequal distribution of students between the different CPII campi according to race. A possible explanation for this data is the fact that campus Humaitá II is located in the south region of Rio de Janeiro (city) – considered an area with a higher socioeconomic level –while campi Duque de Caxias and Niterói are in cities closer to the capital. The unit Duque de Caxias is in the city's central region, and Niterói is in the limits of this city with São Gonçalo, which are characterized as areas with a lower socioeconomic level.

Regarding the income of students and their families, as the Censo Escolar does not offer this information, we sought a reference in the 2019 data available on the CPII institutional site. Generally, 54.24% of CPII students have a family income of up to three minimum wages. The three campi with the highest percentage of black, brown, and Indigenous students, are the ones with the lowest family income (up to three minimum wages), more specifically Duque de Caxias (72.77%), Niterói (62.62%), and Realengo II (61.79%).

The repetition phenomenon at CPII High School: Following the cohort of 2015 newcomers

As mentioned, 1,689 students joined Grade 10 in regular high school in 2015. In the passage between Grade 10 to Grade 11 (first transition), a total of 1,343 (79.5%) continued their studies; 235 students failed (13.9%); and 111 (6.6%) were not found in the database or were enrolled in different modalities of regular high school, such as YAE, for example⁸. Therefore, out of the 1,689 students who joined Grade 10, a total of 1,343 passed the year and enrolled in Grade 11.

In the passage between Grade 11 and Grade 12 (second transition), out of the 1,343 students enrolled in Grade 12, a total of 1,195 (89%) continued; 75 (5.6%) failed, and 73 (5.6%) were not found in the database, probably because they left CPII or moved to YAE. Thus, in 2017, 1,195 students reached Grade 12 with a regular trajectory.

Summing up, out of the 1,689 newcomers in 2015, 1,343 (79.5%) were enrolled in Grade 11 in 2016, and 1,195 (70.7%) reached Grade 12 of regular High School in 2017. These numbers mean that around 30% of students did not have a regular High School trajectory at CPII. However, this 30% does not necessarily mean they failed during this period, as the percentage encompassed those who failed and those who dropped out⁹.

Regarding those who failed, the focus of this study, the data point out the following situation: in the first transition, out of the 1,689 newcomers, 235 failed (13.9%); and, in the second transition, out of the 1,343 students enrolled in Grade 11, 75 failed (5.6%). This means that failure is more than double the passage from Grade 10 to Grade 11, reverberating national data and research data that point out high retention rates in Grade 10 (Silva et al., 2016). For our analysis, we calculated the percentage of students who failed compared to those enrolled in the referenced school grade (2015). That is, in the first transition, we considered the 1,689 newcomers; in the second transition, we considered the 1,343 students enrolled in Grade 11.

In the analysis per campus, we considered the campus where the student studied the Grade in the year analyzed. For example, in 2015, if the student was enrolled at campus Centro, in 2016, the trajectory analysis, i.e., the passage to Grade 11 considered the campus Centro. If, for instance, this student has changed from campus Tijuca in this passage, the trajectory analysis to Grade 11 for Grade 12 was considered in the campus Tijuca. Table 6 shows the number and the percentage of students who failed in each transition by campus. Graphic 4 illustrates the percentages by campi in both transitions.

Table 6
Percentage of students who failed in the two transitions per campus (Adapted from Inep, 2016, 2017, 2018)

Campus	Enrolled 2015 (Grade 10)	Failed in 1st transition		Enrolled 2016 (Grade 11)	Failed in 2nd transition
		Total	%		Total	%
Centro	125	20	16.0	109	4	3.7
Duque de Caxias	162	29	17.9	108	7	6.5
Engenho Novo II	136	13	9.6	107	2	1.9
Humaitá II	263	48	18.3	193	32	16.6
Niterói	198	26	13.1	162	3	1.9
Realengo II	269	35	13.0	226	12	5.3
São Cristóvão III	351	40	11.4	290	12	4.1
Tijuca II	185	24	13.0	148	3	2.0
Total	1.689	235	13.9	1,343	75	5.6

Figure 4
Percentage of students who failed in the two transitions per campus (Adapted from Inep, 2016, 2017, 2018)

As the data points out, the campi that failed more students in the first transition were Humaitá II – 18.3% – and Duque de Caxias – 17.9%. In the second, Humaitá II stands out as the campus with the highest percentage of failure– 16.6%, much higher than the other campi. Engenho Novo I and Tijuca II had smaller percentages of failure in the second transition: 1.9 % and 2.0%, respectively. Campus Engenho Novo II calls attention because it has the smallest failure rates in both transitions and campus Humaitá II presented the highest retention rates in both transitions. Humaitá II has a particularity that may explain this result: it is the only CPII campus that offers regular High School at night, which commonly attends working students and has a higher repetition.

After, besides the campus, the analysis considered the type of admission. Table 7 shows the failure data for the first transition and Table 7 for the second transition. In these cases, the percentages were calculated based on the number of newcomers on the campus and according to the form of admission – Admission Process (AP) or graduated Middle Education at CPII (CPIIG).

Table 7
Newcomers and those who failed by campus in the first transition, according to the admission form (Adapted from Inep 2016, 2017, 2018)

1st transition		Newcomers			Failed
Campus		AP	CPIIG	Total	FAP	FCPIIG	Total
Centro	Nº	35	90	125	8	12	20
	%	28	72	100	22.9	13.3	16
Humaitá II	Nº	62	201	263	15	33	48
	%	23.6	76.4	100	24.2	16.4	18.3
Tijuca II	Nº	21	164	185	1	23	24
	%	11.4	88.6	100	4.8	14	13
Engenho Novo II	Nº	0	136	100	0	13	13
	%	0	100	100	0	9.6	9.6
São Cristóvão III	Nº	102	249	351	10	30	40
	%	29.1	70.9	100	9.8	12	11.4
Realengo II	Nº	117	152	269	16	19	35
	%	43.5	56.5	100	13.7	12.5	13
Niterói	Nº	177	21	198	24	2	26
	%	89.4	10.6	100	13.6	9.5	13.1
Duque de Caxias	Nº	126	36	162	22	7	29
	%	77.8	22.2	100	17.5	19.4	17.9
Total	Nº	640	1049	1,689	96	139	235
	%	37.9	62.1	100	15	13.3	13.9

• Notas. AP = admission process; CPIIG = CPII graduate; FAP = failure, according to admission process; FCPIIG= failure, according to CPII graduate

Table 8
Newcomers and those who failed by campus in the 2nd transition, according to the admission form (Adapted from Inep, 2016, 2017, 2018)

2nd transition		Newcomers			Failed
Campus		AP	CPIIG	Total	FAP	FCPIIG	Total
Centro	Nº	28	81	109	1	3	4
	%	25.7	74.3	100	3.6	3.7	3.7
Humaitá II	Nº	41	152	193	4	28	32
	%	21.2	78.8	100	9.8	18.4	16.6
Tijuca II	Nº	19	129	148	1	2	3
	%	12.8	87.2	100	5.3	1.6	2.0
Engenho Novo II	Nº	0	107	107	0	2	2
	%	0	100	100	0	1.9	1.9
São Cristóvão III	Nº	91	199	290	2	10	12
	%	31.4	68.6	100	2.2	5.0	4.1
Realengo II	Nº	98	128	226	5	7	12
	%	43.4	56.6	100	5.1	5.5	5.3
Niterói	Nº	144	18	162	2	1	3
	%	88.9	11.1	100	1.4	5.6	1.9
Duque de Caxias	Nº	91	17	108	3	4	7
	%	84.3	15.7	100	3.3	23.5	6.5
Total	Nº	512	831	1,343	18	57	75
	%	38.1	61.9	100	3.5	6.9	5.6

The data shows that the rates vary between campi. For example, at Centro, Humaitá II, Realengo II, and Niterói, failure rates are higher for students joining by the selective process. At campus Engenho Novo II, no students who joined through the selective process failed in Grade 10, and at campus Tijuca II, only one student approved in the selection failed the school year. In general terms, 235 students failed in the first transition, out of those 135 students who came from CPII and 96 who joined through the selective process.

In the second transition (Table 8), the percentage of selective-process students was higher than the percentage CPII graduates only in campus Tijuca. Again, no selective-process student failed in Engenho Novo II, while in Humaitá II, 28 CPII graduates failed Grade 11. In this second transition, only 75 students failed, 57 CPII graduates, and only 18 selective-process newcomers.

These data indicate that, in both transitions, CPII graduates, that is, those who studied middle school in the institution, failed more than those who joined through the selective process. In other words, CPII fails more its own students than those who came from other schools and joined high school through a selective process. A plausible explanation for these results is that these latter students come from private and/or public schools that focus on preparing their middle school students to join good public high schools, such as those connected to the public federal and state universities, the federal technological institutes, and CPII. Thus, they tend to have fewer learning difficulties and more regular school trajectories.

Advancing the analysis, Tables 9 and 10 data show that, out of all newcomers in Grade 10, 21.7% and 26.6% are, respectively, young men and women who did not declare their race. Out of the declaring ones, 15.8% are white boys, 7.5% are black boys, 19.3% are white girls and 9% are black girls. Out of the newcomers, 235, i.e., 13.9% failed. Among them, 15.5% and 11.6% were, respectively, young men and women who did not declare their race; 12% and 18.1% were white and black young men, respectively, while 13.5% and 17.8% were, correspondently, white and black young women who failed. In this sense, CPII fails more black young men and women than white young men and women.

Table 9
Number of newcomers and those who failed in the first transition by campus, according to sex x race (Inep, 2016, 2017, 2018)

Campus		MND	MWA	BM	WND	WWA	BW	Total
Centro	NN	27	27	8	21	32	10	125
	%	21.6	21.6	6.4	16.8	25.6	8	100
	NR	3	7	2	1	4	3	20
	%	11.1	25.9	25	4.8	12.5	30	16
Humaitá II	NN	36	68	23	34	79	23	263
	%	13.7	25.9	8.7	12.9	30	8.7	100
	NR	11	14	7	3	11	2	48
	%	30.6	20.6	30.4	8.8	13.9	8.7	18.3
	NN	64	25	15	53	19	9	185
Tijuca II	%	34.6	13.5	8.1	28.6	10.3	4.9	100
	NR	12	3	4	5	0	0	24
	%	18.8	12.0	26.7	9.4	0.0	0.0	13.0
Engenho Novo II	NN	34	24	6	33	30	9	136
	%	25	17.6	4.4	24.3	22.1	6.6	100
	NF	2	3	0	2	5	1	13
	%	5.9	12.5	0.0	6.1	16.7	11.1	9.6
São Cristóvão III	NN	91	43	27	110	49	31	351
	%	25.9	12.3	7.7	31.3	14	8.8	100
	NF	15	1	4	10	4	6	40
	%	16.5	2.3	14.8	9.1	8.2	19.4	11.4
Realengo II	NN	41	49	26	41	70	42	269
	%	15.2	18.2	9.7	15.2	26	15.6	100
	NF	2	2	2	3	14	12	35
	%	4.9	4.1	7.7	7.3	20.0	28.6	13.0
Niterói	NN	44	20	7	95	21	11	198
	%	22.2	10.1	3.5	48	10.6	5.6	100
	NF	7	2	1	12	2	2	26
	%	15.9	10.0	14.3	12.6	9.5	18.2	13.1
Duque de Caxias	NN	30	11	15	63	26	17	162
	%	18.5	6.8	9.3	38.9	16	10.5	100
	NF	5	0	3	16	4	1	29
	%	16.7	0.0	20.0	25.4	15.4	5.9	17.9
Total	NN	367	267	127	450	326	152	1689
	%	21.7	15.8	7.5	26.6	19.3	9	100
	NF	57	32	23	52	44	27	235
	%	15.5	12.0	18.1	11.6	13.5	17.8	13.9

• Notas. NN = Number of newcomers; NF = Number of failed students; MND = man non-declared; MWA = White and/or Asian man; BM = black man; WND = woman non-declared; WWA = woman white and/or Asian; BW = black woman

Table 10
Number of students and those who failed in the second transition by campus, according to sex x race (Inep, 2016, 2017, 2018)

Campus		MND	HBA	BM	MND	MBA	BW	Total
Centro	NI	23	20	6	22	30	8	109
	%	21.1	18.3	5.5	20.2	27.5	7.3	8.1
	NR	1	1	0	2	0	0	4
	%	4.3	5	0	9.1	0	0	3.7
Humaitá II	NI	21	48	15	30	58	21	193
	%	10.9	24.9	7.8	15.5	30.1	10.9	14.4
	NR	7	10	1	2	9	3	32
	%	33.3	20.8	6.7	6.7	15.5	14.3	16.6
Tijuca II	NI	46	21	10	44	18	9	148
	%	31.1	14.2	6.8	29.7	12.2	6.1	11.0
	NR	0	0	1	1	1	0	3
	%	0	0	10	2.3	5.6	0	2
Engenho Novo II	NI	25	18	6	27	23	8	107
	%	23.4	16.8	5.6	25.2	21.5	7.5	8.0
	NR	0	0	2	0	0	0	2
	%	0	0	33.3	0	0	0	1.9
São Cristóvão III	NI	66	42	22	93	43	24	290
	%	22.8	14.5	7.6	32.1	14.8	8.3	21.6
	NR	3	2	2	2	1	2	12
	%	4.5	4.8	9.1	2.2	2.3	8.3	4.1
Realengo II	NI	37	46	22	38	52	31	226
	%	16.4	20.4	9.7	16.8	23.0	13.7	16.8
	NR	3	1	0	3	3	2	12
	%	8.1	2.2	0	7.9	5.8	6.5	5.3
Niterói	NI	35	16	5	79	19	8	162
	%	21.6	9.9	3.1	48.8	11.7	4.9	12.1
	NR	0	0	0	3	0	0	3
	%	0	0	0	3.8	0	0	1.9
Duque de Caxias	NI	21	9	10	38	18	12	108
	%	19.4	8.3	9.3	35.2	16.7	11.1	8.0
	NR	2	0	0	3	0	2	7
	%	9.5	0	0	7.9	0	16.7	6.5
Total	NI	274	220	96	371	261	121	1,343
	%	20.4	16.4	7.1	27.6	19.4	9.0	100.0
	NR	16	14	6	16	14	9	75
	%	5.8	6.4	6.3	4.3	5.4	7.4	5.6

The analysis per campi shows that Humaitá II, São Cristóvão III, and Realengo II are the school units that fail the most, while the intersectional issue appears more clearly in campus Humaitá II, where a significantly smaller number of black young men and women have entered. Despite this, this group concentrates on the failing rates, indicating a strong racial bias in the admission and repetition of black students on this campus. In São Cristóvão III, though the repetition in the first transition was equalitarian according to sex – 20 boys and 20 girls – there are significant differences when adding the element race to the analysis, as only 2.3% white young men repeated compared to the 14.8% black young men. This difference can also be seen among the young women: out of the repeaters, 8.2% are white, and 19.4% are black. In its turn, Realengo II calls attention to the fact that girls fail more than boys and, proportionally, black girls experience more the failing experience.

These results converge with Crenshaw's (1989) intersectional analysis, according to which black women share very similar experiences to black women, in this case, the experience of school failure. This type of result calls attention to the importance of intersectoral analysis and its potential contribution to improving policies to face social inequalities (Medeiros, 2019).

Similar data is found when analyzing the failure in the second transition (Table 10), which shows a much smaller number of repeaters (75 students) when compared to the first transition. Generally, there is a parity in the percentage of white and black boys who failed – 6.4% and 6.3%, respectively – and a higher percentage of failure for black girls – 5.4% – compared to white girls – 4.3%. Once more, the campus with a higher failing rate is Humaitá II, followed by São Cristóvão III and Realengo II, which had the same number of failures in the second transition. In Humaitá II, white boys (20.8%) and white girls (15.5%) failed more than black boys (6.7%) and black girls (14.3%). This result can be explained by the selection bias in students' admission to this campus that, as mentioned, admits more white students than black ones and keeps, in the first transition, more black students (boys and girls). Therefore, we can raise the hypothesis that Grade 11 in the campus Humaitá II is more homogeneous regarding race.

In São Cristóvão III, the percentage of failure among black boys and black girls is higher than among white boys and white girls. In Realengo II, the percentage of failure among white boys is higher than among black ones, and the failure of black girls is higher than white girls.

Finally, the data in Tables 11 and 12 below show the percentage of repetition by campi and age. Generally, in the first transition, students at the right age to study in Grade 10 of High School fail more than those already behind in their trajectories. However, differences between the campi can be seen. In Humaitá II, Tijuca II, Engenho Novo II, São Cristóvão III, Realengo II, and Duque de Caxias, there is a higher percentage of students' repetition among those behind in terms of age-grade. This scenario indicates that, proportionally, these campi students who already had a delay in their school trajectories fail more.

Table 11
Number of students and those who failed in the first transition by campus, according to the age in which they were admitted (Adapted from Inep, 2016, 2017, 2018)

1st transition		Newcomers			Failed
Campus		IC	ID	Total	RIC	RID	Total
Centro	Nº	100	25	125	17	3	20
	%	80.0	20,0	100	17.0	12,0	16
Humaitá II	Nº	168	95	263	26	22	48
	%	63.9	36.1	100	15.5	23.2	18.3
Tijuca II	Nº	128	57	185	11	13	24
	%	69.2	30.8	100	8,6	22.8	13
Engenho Novo II	Nº	103	33	136	7	6	13
	%	75.7	24.3	100	6.8	18.2	9.6
São Cristóvão III	Nº	230	121	351	16	24	40
	%	65.5	34.5	100	7	19.8	11.4
Realengo II	Nº	230	39	269	25	10	35
	%	85.5	14.5	100	10.9	25.6	13
Niterói	Nº	172	26	198	23	3	26
	%	86.9	13.1	100	13.4	11.5	13.1
Duque de Caxias	Nº	127	35	162	22	7	29
	%	78.4	22	100	17.3	20	17.9
Total	Nº	1258	431	1689	147	88	235
	%	74.5	25.5	100.0	11.7	20.4	13.9

• Notes. RA = right age; GA = gap age; RRA = repetition, according to right age; RGA = repetition gap age

Table 12
Number of students and those who failed in the second transition by campus, according to the age in which they were admitted (Adapted from Inep, 2016, 2017, 2018)

2nd transition		Newcomers			Failed
Campus		RA	GA	Total	RRA	RGA	Total
Centro	Nº	87	22	109	2	2	4
	%	79.8	20.2	100	2.3	9.1	3.7
Humaitá II	Nº	136	57	193	21	11	32
	%	70.5	29.5	100	15.4	19.3	16.6
Tijuca II	Nº	110	38	148	1	2	3
	%	74.3	25.7	100	0.9	5.3	2
Engenho Novo II	Nº	91	16	107	2	0	2
	%	85.0	15.0	100	2.2	0	1.9
São Cristóvão III	Nº	210	80	290	7	5	12
	%	72.4	27.6	100	3.3	6.3	4.1
Realengo II	Nº	199	27	226	9	3	12
	%	88.1	11.9	100	4.5	11.1	5.3
Niterói	Nº	140	22	162	2	1	3
	%	86.4	13.6	100	1.4	4.5	1.9
Duque de Caxias	Nº	92	16	108	6	1	7
	%	85.2	14.8	100	6.5	6.3	6.5
Total	Nº	1065	278	1343	50	25	75
	%	79,3	20,7	100	4,7	9	5,6

In the second transition, the situation gets words as, on average and proportionally, students with gaps reprove more than those in the right age for the junior year of high school. Except for Engenho Novo II and Duque de Caxias, all campi presented a higher percentage of students with an age gap, thus worsening their school delay.

As a whole, these results align with other international and national studies on school failure, dropout, and age-grade gap. These studies point out that failure increases dropout and that male and black individuals have higher failure levels, drop out more, and have the highest gaps (Menezes Filho, 2007; Riani, 2005). The analysis conducted for the United States, for instance, points out that the gender gap in education would be more significant for black people when compared to white ones (McDaniel et al., 2011) and that white women were gradually increasing their advantages, distancing from white men and, mainly, black men (Lopes da Costa & Picanço, 2020). This result alignment can show the persistent concentration of inequalities in the social group established by male students and self-declared black, regardless of specific school contexts and their different operating conditions.

Final remarks

This study analyzes how the repetition phenomenon occurs in a federal education institution considered to be of excellence, Colégio Pedro II. It is a public institution historically focused on attending more socially privileged classes, which is currently making an effort to democratize its access to other social groups.

The initiatives to increase and socially diversify access to CPII were also followed by the removal of internal mechanisms that limited the democratizing effect of such measures. For instance, the academic dismissal of repeaters in action at the institution until 2015 was extinguished as it was considered an excluding practice that increased social inequalities (Couto, 2018).

Despite these advancements, this study shows that the repetition practice still reproduces inequality standards, leading to age-grade gap, and dropout of students with a specific social profile: black and brown young men. The study also identifies variations in the percentage of repetition and dropout in the different campi offering middle and high school. However, in all of them, black male students reproved the most.

Therefore, notwithstanding CPII's effort to increase and democratize access to less privileged groups, repetition tends to affect student groups with this social profile, contributing to a decrease in the possible democratizing effects of inclusive measures and actions adopted by the institution. Hence, the study indicates the need to follow up with internal processes to democratize advancements in CPII access towards the permanence and school progression of these students, paying attention to the intra-school factors responsible for reproducing pre-existing social inequalities.

• 1
  In 2020, 2021, and 2022, due to the Covid-19 pandemic, CPII used the raffle as the admission process in all school phases.
• 2
  Grade 13 refers to some specific modalities, such as Integrated high school.
• 3
  “ We consider 'age-grade gap' the situation of all students who are older than the age recommended for the attended grade – the age of 6 is considered the ideal/recommended age to be admitted in the first grade of elementary school" (Inep, 2021, p. 16).
• 4
  Public data until 2021.
• 5
  The longitudinal base with the variable of school flow and the student situation in Grade 12 is available only in the secured rooms of Inep, in the Serviço de Acesso a Dados Protegidos (Sedap).
• 6
  When gathering the 2014 databases, we did not find information on 29 students about which school they attended before CPII (if public or private). These students might come from other states. We opted to keep these students in the base to follow their performance.
• 7
  For this procedure, we selected the variable “tipo_turma” [type_class] and excluded those connected to categories 4 (complementary activity) and 5 (specialized educational service), keeping only category 0 (does not apply).
• 8
  We opted to disregard these 111 missing cases and not use data imputation techniques.
• 9
  Those who left CPII dropped out of their studies or went to another school and/or to YAE at CPII.

• Support and Funding:
  CAPES; CNPq; FAPERJ
• Copy Editing services:
  Portuguese version - Copy editing and standardization of citations and bibliographical references (7th. Edition APA): <E-mail>: Camila Pires de Campos Freitas <[e-mail]>
  English Version: Viviane Ramos <[e-mail]>

References

• Alves, M. T. G., Soares, J. F., & Xavier, F. (2016). Desigualdades Educacionais no ensino fundamental de 2005 a 2013: hiato entre grupos sociais. Revista Brasileira de Sociologia, 4 (7), 49-81.
• Alves, F., Ortigão, I., & Franco, C. (2007). Origem social e risco de repetência: interação raça-capital econômico. Cadernos de pesquisa, 37(130), 161-180.
• Andrade, J. M., & Laros, J. A. (2007). Fatores associados ao desempenho escolar: estudo multinível com dados do SAEB 2001. Psicologia: teoria e pesquisa, v. 23(1), 33-42.
• Buss Gasser, E., & Martins, S. (2016). Abandono e repetência na 1ª série do ensino médio: desafios e ações. In Secretaria de Educação do Estado de Paraná (Ed.), Cadernos PDE. Os desafios da escola pública paraense na perspectiva do professor (v. 1). Secretaria de Educação do Estado de Paraná.
• Correa, E. V., Bonamino, A. M. C., & Soares, T.M. (2014). Evidências do efeito da repetência nos primeiros anos escolares. Estudos em Avaliação Educacional, 25(59), 242-269.
• Costa, M., & Koslinski, M. (2011). Quase-mercado oculto: disputa por escolas “comuns” no Rio de Janeiro. Cadernos de Pesquisa,41(142), 246-266.
• Couto, I. M. S. (2018). Jubilados e Evadidos: uma análise comparativa do perfil de outsiders do Colégio Pedro II.[Dissertação de mestrado, Universidade Federal do Rio de Janeiro].
• Crahay, M. (2006). É possível tirar conclusões sobre os efeitos da repetência? Cadernos de Pesquisa, 36(127), 223-246.
• Crenshaw, K. (1989). Demarginalizing the intersection of race and sex: a black feminist critique of anti discrimination doctrine, feminist theory and antiracist politics. University of Chicago Legal Forum, 108, 139-167.
• Dargains, R. L. (2015). A política invisível: o caso da implantação das cotas raciais no Colégio Pedro II [Dissertação de Mestrado, Universidade Federal do Rio de Janeiro].
• Ernica, M., & Rodrigues, E. C. (2020). Desigualdades educacionais em metrópoles: território, nível socioeconômico, raça e gênero. Educação & Sociedade, 41, 1-19.
• Giddens, A., & Sutton, P. W. (2017). Conceitos essenciais de Sociologia(1. ed.). Unesp Digital.
• Instituto Nacional de Estudos e Pesquisas Educacionais Anísio Teixeira (2016). Microdados do Censo Escolar 2015 Inep.
• Instituto Nacional de Estudos e Pesquisas Educacionais Anísio Teixeira. (2017). Microdados do Censo Escolar 2016 Inep.
• Instituto Nacional de Estudos e Pesquisas Educacionais Anísio Teixeira. (2018). Microdados do Censo Escolar 2017 Inep.
• Instituto Nacional de Estudos e Pesquisas Educacionais Anísio Teixeira. (2020). Resumo Técnico: Censo da Educação Básica Estadual 2019. Inep.
• Instituto Nacional de Estudos e Pesquisas Educacionais. (2019a). (2019a). Taxas de rendimento escolar por escola - 2019 Inep. https://www.gov.br/inep/pt-br/acesso-a-informacao/dados-abertos/indicadores-educacionais/taxas-de-rendimento
  » https://www.gov.br/inep/pt-br/acesso-a-informacao/dados-abertos/indicadores-educacionais/taxas-de-rendimento
• Instituto Nacional de Estudos e Pesquisas Educacionais. (2019b). Taxa de Distorção Idade-Série por Escola – 2019 Inep. https://www.gov.br/inep/pt-br/acesso-a-informacao/dados-abertos/indicadores-educacionais/taxas-de-distorcao-idade-serie
  » https://www.gov.br/inep/pt-br/acesso-a-informacao/dados-abertos/indicadores-educacionais/taxas-de-distorcao-idade-serie
• Instituto Nacional de Estudos e Pesquisas Educacionais. Ministério da Educação. (2021). Inep. Taxas de rendimento escolar https://download.inep.gov.br/educacao_basica/educacenso/situacao_aluno/documentos/2021/taxas_de_rendimento_escolar.pdf
  » https://download.inep.gov.br/educacao_basica/educacenso/situacao_aluno/documentos/2021/taxas_de_rendimento_escolar.pdf
• Klein, R., & Ribeiro, S. C. (1991). O censo educacional e o modelo de fluxo: o problema da repetência São Paulo: USP/NUPES.
• Lopes da Costa, A., & Picanço, F. (2020). Para além do acesso e da inclusão: impactos da raça sobre a evasão e a conclusão no Ensino Superior. Novos Estudos Cebrap, 39(2), 281-306.
• McDaniel, A., DiPrete, T., Buchmann, C., & Shwed, U. (2011). The Black Gender Gap in Educational Attainment: Historical Trends and Racial Comparisons. Demography, 48, 889-914.
• Medeiros, R. S. (2019). Interseccionalidade e políticas públicas: aproximações conceituais e desafios metodológicos. In R. R. C. Pires (Org.), Implementando desigualdades: reprodução de desigualdades na implementação de políticas públicas(pp.79-103). Ipea.
• Menezes-Filho, N. (2007). Os Determinantes do Desempenho Escolar do Brasil IBMEC.
• Ortigão, M. I. R., & Aguiar, G. S. (2013). Repetência escolar nos anos iniciais do ensino fundamental: evidências a partir dos dados da Prova Brasil 2009. Revista Brasileira de Estudos Pedagógicos, 94 (237), 364-389.
• Riani, J. de L. R. (2005). Determinantes do resultado educacional no Brasil: família, perfil escolar dos municípios e dividendo demográfico numa abordagem hierárquica e espacial [Tese de doutorado, Universidade Federal de Minas Gerais].
• Ribeiro, S. C. (1991). A pedagogia da repetência. Estudos Avançados, 5(12), 7-21.
• Sa Earp, M.L. (2009). A cultura da repetência em escolas cariocas. Ensaio: Avaliação de Políticas Públicas em Educação, 17(65), 613-632.
• Senkevics, A., Machado, T., & Oliveira, A. (2016), A cor ou raça nas estatísticas educacionais: uma análise dos instrumentos de pesquisa do Inep. Textos para Discussão, 41, 1-38.
• Senkevics, A. (2022). De Brancos para Negros? Uma Análise Longitudinal da Reclassificação Racial no Enem 2010-2016. Dados, 65(3), 1-40.
• Silva, P., Rezende, N., Quaresma, T. & Chrispino, A. (2016). Sobre o sucesso e o fracasso no Ensino Médio em 15 anos (1999 e 2014). Ensaio: Avaliação de políticas públicas educacionais, 24(91), 445-476.
• Soares, J. F. (2004). O efeito da escola no desempenho cognitivo de seus alunos. REICE - Revista Electrónica Iberoamericana sobre Calidad, Eficacia y Cambio en Educación, 2(2), 83-104.
• Souza, A. P. D., Ponczek, V. P., Oliva, B. T., & Tavares, P. A. (2012). Fatores associados ao fluxo escolar no ingresso e ao longo do ensino médio no Brasil. Pesquisa e Planejamento Econômico, 43(1), 5-39.
• Todos pela Educação (Org.). (2021). Anuário Brasileiro da Educação Básica.Moderna. https://todospelaeducacao.org.br/wordpress/wp-content/uploads/2021/07/Anuario_21final.pdf
  » https://todospelaeducacao.org.br/wordpress/wp-content/uploads/2021/07/Anuario_21final.pdf
• Unicef. (2019). Painel Brasil Unicef. https://trajetoriaescolar.org.br/painel-brasil/2019/
  » https://trajetoriaescolar.org.br/painel-brasil/2019/

Edited by

• Responsible editors:
  Associate Editor: Chantal Medaets <https://orcid.org/0000-0002-7834-3834>
  Editor-in-Chief: Helena Sampaio <https://orcid.org/0000-0002-1759-4875>

Publication Dates

• Publication in this collection
  28 Oct 2024
• Date of issue
  2024

History

• Received
  18 Nov 2022
• Reviewed
  28 Feb 2024
• Accepted
  01 Apr 2024