Introduction

Discourses that schools are not accessible to all or do not meet the diverse social aspirations have linked the institution to the idea of failure. According to ^{Patto et al. (2004)}, the problem is conditioned by the needs and possibilities envisioned by society in relation to education in each historical period and depending on the places occupied by social subjects, as well as the analytical repertoire mobilized.

In this respect, it is important to note that the categories of school success and failure are historical, social and cultural constructions, based on the most varied expectations possible. It is even accepted that failure is a notion interpreted and produced by the school community itself, sometimes attributed to specific subjects (students, teachers, pedagogues, managers, families), and can also be related to factors outside the school.

For failure to be currently linked to “inadequate” student learning, as pointed out in the discourses that legitimized the National Common Curriculum Base (in Portuguese Base Nacional Comum Curricular - BNCC), for example, the current configurations of the school (public, compulsory, annual, serialized, staged) and the availability of statistical instruments that provided knowledge of certain movements of students through the grades were conditioning factors. In addition, it was only after the expansion of school coverage and the recent universalization of enrolments that the focus of the debate shifted to measuring the quality of provision.

According to ^{Soares and Xavier (2013, p. 904)}, before this period, valuing educational results was absent from the dominant analyses of basic education, which focused on the expansion of systems. In these approaches, the solution was focused on more class hours, more compulsory stages, more resources, more schools and more teachers.

However, recent national laws and policies have been pointing investments in a different direction. From the perspective of the right to education, documents such as the 1988 Federal Constitution and the National Education Plan (PNE) have emerged as legal instruments capable of guaranteeing the realization of this right. Article 205 of the aforementioned Constitution establishes that school education, a right of all and a duty of the state, has as its purpose “the full development of the person, their preparation for the exercise of citizenship and their qualification for work”. It also stipulates that basic education is compulsory from the age of 4 to 17. The PNE (2014-2024) (^{Brasil, 2014a}), on the other hand, is a powerful government planning tool, providing diagnoses, guidelines and targets to ensure that education is a right for all Brazilians.

According to ^{Soares (2016, p. 142)}, “an unmonitored right is just an intention, a utopia”. In light of this, recent educational policies understand that school success is achieved when individuals are guaranteed the learning necessary to achieve the three constitutional objectives. This has strengthened the need to stipulate the knowledge that constitutes the right.

The BNCC has therefore emerged as a benchmark for improving student learning. This legislation was based on the premise that “quality learning is a goal that the country must pursue tirelessly, especially in secondary education, where the rates of learning, repetition and dropout are very worrying” (Brasil, 2018). There is an affirmation of a certain concept of quality, based on a “rationality that constructs an educational act that can always be measured in quantifiable indices” (^{Bocchetti; Silva, 2016}, p. 47).

It is important to point out that the emphasis on the use of statistics has become a hallmark of the post-1990 period, characterized by the production of indices obtained in large-scale assessments, whose numbers are taken as parameters for government interventions. These instruments are used to diagnose the problems, needs and possibilities of the Brazilian education system (^{Alves, 2022}).

An example of a discouraging diagnosis is the Basic Education Development Index (in Portuguese Índice de Desenvolvimento da Educação Básica - IDEB). An instrument used in Brazil, it seeks to measure the quality of education based on the pass rate and average performance of students in Portuguese and mathematics. Another instrument used worldwide is the Program for International Student Assessment (PISA). According to ^{Popkewitz and Lindblad (2016)}, this device produces numbers that forge principles about who children are - and what they should be like. These numbers are the basis for supposedly modernizing actions in the management of school populations.

These initial considerations point to the investments that will be made below, based on the hypothesis that reform projects trigger logics that produce school failure, which is always related to the purposes of the school for the subjects, specific groups, governments, in each historical period. The analysis will focus on the rationalities mobilized in the drafting of the BNCC and other legislation proclaiming the need to change school practices, transforming the curriculum into an object of reform.

In this endeavor, the following questions will guide the approach: what educational diagnoses were mobilized by the groups involved in the construction of recent curriculum reforms, such as the BNCC? What solutions were presented? What representations of school failure were conveyed?

The right to education and the emergence of school failure and dropout as political-educational problems

A characteristic of the first half of the last century was both the scarcity of schools to cater for the entire child population and the inefficiency of the education offered. In order to reach this conclusion, it was necessary to observe the movement of pupils through the grades through the figures produced by the government. During this period, concerns about popular education were still based on a government issue and not on the right of the individual.

Schooling, conceived as the right of every citizen and the duty of the state, although apparently consensual, is a historical construction. According to ^{Horta (1998)}, the social right to education was incorporated late into the select group of human rights, through a slow, ambiguous and contradictory process. Despite the late admission of its importance as a path to achieving citizenship, the right to schooling has become increasingly important: firstly focused on elementary education, expanding to secondary education and, in some countries, reaching university education.

In Brazil, the dual reference of the right to education and the duty of the state is as late as it is full of inequalities (^{Cury; Reis; Zanardi, 2018}). Considering the legacies of the colonial past, the more than three centuries of slavery, the diverse and unequal cultural background, there are many lasting consequences that make it difficult for the majority of the Brazilian population to lead a life as citizens. According to ^{Morais (2013)}, the exercise of citizenship is still a challenge to be faced in the country, given that there is still a huge contingent of people surviving on the margins of the ideals of justice and dignity. For this reason, the perception and experience of what it means to be a citizen becomes a daily construction.

Since citizenship is a social practice and a field of disputes, the right to education, seen as a fundamental element in achieving it, has become the object of government action. In this case, it is important to note that, in addition to the demands coming from organized civil society, the state itself is expressing its demands, aimed at economic and technological development, for which investment in education is paramount. These demands are not disconnected from the techniques of disciplining and controlling the population (^{Foucault, 1987}), especially when schooling takes on the facet of a social obligation.

In Brazil, the expansion of compulsory schooling was not characterized as a linear and evolutionary process, given that the forces at work in the construction of the laws did not obey a unidirectional direction. According to ^{Matos (2019)}, the scope of compulsory school attendance underwent reconfigurations related to disputes, changes in intentions, urgencies, emergencies and other issues with which it was (and is) involved.

Examples in this regard were the laws that changed compulsory schooling in the country: first, eight years for primary education, according to the 1988 Constitution; then, it was expanded when this stage was extended to nine years in 2006. Finally, it was changed again by Amendment 59/2009, which established compulsory schooling from the age of four to seventeen, incorporating the second stage of early childhood education (pre-school).

One of the consequences of the 2009 amendment, by establishing a relationship between age and school grade, was to make only early childhood education compulsory, since this is the only stage at which students cannot fail. In other words: in view of the stipulation of the “correct” age for each grade, those students who fail and/or are of an inadequate age are guaranteed the right to free public school, but not its compulsory nature. The amendment therefore brought back compulsory primary education (in force until 2009)², even though it expanded and brought forward the scope of compulsory attendance.

These interlocutions highlight the influence of the 2009 amendment on the (re)configuration of the Brazilian school system, given the privileging of a certain model of schooling. Although the Law of Guidelines and Bases of National Education (in Portuguese Lei de Diretrizes e Bases da Educação Nacional - LDB) 9394/96 admits other forms of organization, compulsory schooling experienced predominantly as serialization has brought to light conceptions of childhood and school anchored in a specific way of thinking and working with time³:

In this serialized form, the school is organized with specific spaces, contents, curricula, subjects, methods, materials, pedagogical practices, regular assessments, rites of passage. Timetables are also organized, based on which the school day is established. A teaching-learning process that is based on a time of life, distributed in stages, classes, grades and years of schooling

(^{Matos, 2019}, p. 46).

The serialized school therefore produces a rhythm, a liturgy. This temporal modality is both diverse and plural, individual and institutional, conditioning and conditioned by other social times. According to ^{Viñao Frago (1995, p. 72)}, this is a cultural and pedagogical construction

[...] which offers - at least as a possibility - a vision of learning and history not as processes of selection and choices, of gains and losses, but of progress and advancement. An advance and progression that certifies exams and the passage from one course or level to another.

Although the experience of time is characterized by subjectivity, from an institutional point of view, school time appears to be prescribed and uniform. This organization, which constrains other social time, especially family time, is based on a sequential and linear vision. It is based on an idea of childhood as a stage and, therefore, segmented, so that life has been divided up according to the expectations of learning at each age. This makes it seem natural for a six-year-old child to be in the first year of elementary school or to be literate by the age of eight, often disregarding their own experiences, rhythms and ways of learning (^{Matos, 2019}, p. 48).

Aside from the countless implications of compulsory education based on the age-grade relationship, it is important to note that the stipulation of learning expectations distributed over the school years has contributed to reinforcing the segmentation of children's lives into smaller and smaller stages. This is because, when compulsory education was established without setting a specific age per grade, a group of students with different ages could attend the same stage of schooling in other historical periods. This idealization, which became established as a prescription, ended up establishing a normality of performance, based on procedures for comparing, classifying and ordering people in reference to a standard. One of the effects has been the distinction between children who “perform” and those who don't “fit in” (^{Popkewitz; Linblad, 2016}).

On the other hand, it is important to emphasize that the creation of national assessments that seek to measure learning levels, as pointed out above, is justified by the need to monitor supply, so that the state and society know if the right to education is being met. Considering the Brazilian scenario, this argument is linked to measures to reduce social inequalities, which affect the educational opportunities of excluded groups, marked by their socioeconomic condition, race/color, gender and region of residence (^{Soares; Delgado, 2016}).

There is, therefore, a social demand for knowledge of the Brazilian educational reality, where statistical production has proved useful for academic research, as well as having the potential to contribute to improving policies aimed at reducing inequalities (^{Soares; Alves, 2023}). In this sense, interventions that seek to gauge the quality of education achieve legitimacy, considering the importance of schooling in the production of the current social order, given that social and professional trajectories are strongly tributary to school trajectories, coupled with the recognition that access to accumulated scientific knowledge is a central element in guaranteeing the democratization of citizens' participation in collective life (^{Saviani, 2001}),

Various policies aimed at guaranteeing the right to education claim to combat any kind of discrimination that affects educational opportunities (in this case, opportunities to learn, with quality). As for the BNCC, ^{Alves (2022)} showed that the disputes over the definition of “essential” or “compulsory” knowledge, embodied in the validity of various versions of the document, highlighted the search to establish consensus around the “educational cause”. However, examining the process of drawing up the BNCC revealed the existence of an “arena” around the meanings of education, schooling and curriculum.

Although the BNCC has emerged as a state commitment to establishing the essential learning to which all students are entitled, the disputes surrounding the reform have highlighted the fact that this “common learning” is, at the same time, a duty for students. The ambiguity of this relationship needs to be pointed out, since, according to ^{Horta (1998)}, unlike other rights that society makes available only to those who request them, the right to education is closely linked to compulsory schooling.

The affirmation of the logic of the right and duty to learn mobilized the stipulation of learning capable of guaranteeing individuals their full development. However, the construction of this consensus was marked by power-knowledge relations, inequalities, differences and attempts to homogenize multiple and dispersed knowledge. In this way, the current explanation is in line with perspectives that seek to add complexity to issues common to the educational field, such as the understanding of compulsory schooling as a progressive achievement, the need for age/grade adequacy, the stipulation of common learning as a right for all and, finally, the discourses on school failure.

In this context, apart from the ambiguities, struggles and contradictions that mark the Brazilian scenario, with problems that go beyond the school sphere, the premise of fulfilling the state's duty towards education persists in a variety of discourses, but not just any education. Generally, the argument used by the right is associated with a salvationist perspective that has the effect of legitimizing and adhering to the objectives prescribed by the policies (^{Bocchetti; Silva, 2016}; ^{Faria Filho, 2010}; ^{Macedo, 2017}). As an example, the National Education Plan (PNE) (2014-2024) has as one of its guidelines “overcoming educational inequalities, with an emphasis on promoting citizenship and eradicating all forms of discrimination” [my emphasis].

Without disregarding the transformative potential of education, this quote can act as a provocation to our rationality, which still reproduces the maxim of the “redeeming school”. Such discourses sometimes fail to examine their logic, feeding “educational slogans” that oscillate between a panacea and the conversion of school relationships in line with changing, distinct, variable and infinite external demands. These expectations, sometimes unrealizable, feed the rhetoric of school failure.

With regard to the assumption of the “evidence” of learning as a right for all, I make interlocutions with ^{Gil (2018)}. To this end, I return to the argument that school failure and dropout only became a public problem from the 1930s onwards. Despite the significant increase in enrollment from the 20th century onwards, until at least the early 1970s the country did not offer places to all children. In addition, few of the students enrolled in the first grade of elementary school made it to the next grades, especially those from poor families. In this process, failure linked to retention and repetition came to be seen as one of the explanatory factors.

In this regard, a warning should be made: although failure existed long before, the link between failure and repetition only emerged when serialized schooling was instituted at the end of the 19th century⁴:

It is from the existence of the physical separation of students according to the grades indicated in the teaching programs, alongside the adoption of simultaneous teaching, that it becomes necessary that, at the end of a school year, those who do not present learning corresponding to the minimum expected, in terms of mastery of the program of the grade attended, return to take it from the beginning the following year, that is, that they repeat the same grade

(^{Gil, 2018}, p. 3).

It is important to note the exceptional nature of the examination rituals in the lives of students who attended school in the 19th century and the slow definition of the annual rhythm as characteristic of school when it became compulsory. School times were therefore defined slowly, since the enrollment period was different, as were the dates for starting school.

As far as exams and failure are concerned, the notion of the calendar year as a time limit for checking learning and defining final results is something that was only established in the 20th century (^{Gil, 2018}, p. 10). The “ages of learning” were also gradually established, based on knowledge from the pedagogical-psychological field and the laws surrounding compulsory education.

So that failure and repetition could be considered flow distortions, it was necessary to standardize school times. This standardization was guaranteed by a set of prescriptions that could not be established without a process of persuasion. Changes in school culture were therefore necessary, but they did not come about automatically as a result of this standardization.

With these considerations in mind, it is worth noting that the existence of students who did not learn at the expected pace was not seen as a public problem, nor was the forced continuation in the grades until mastery of the content was proven. Nor was it a concern that students left school on the assumption that they wouldn't progress. These occurrences were assumed to be distortions when certain movements of students through the grades could be known from statistics.

In the case of retention (which often leads to students dropping out), there was a certain naturalization, since it was assumed that school results corresponded to personal (in)abilities. It will be in the face of compulsory schooling, coupled with the social conviction of the advantages of schooling, that staying in school will prevail, even in the face of negative results. It is in this context that attending the same grade becomes admissible, making repetition quantitatively significant (^{Gil, 2018}, p. 6).

From a legal point of view, it was only in 1969 that the Brazilian Constitution made education explicit as a duty of the state and linked compulsory education to the age group and level of education, a relationship that was taken up again in the 1971 LDB and the 1988 Constitution. Therefore, the debate about access and, notably, staying in school, on a national scale, as well as the provision of the state's duty towards education, involving sanctions on public authorities, is a recent phenomenon in Brazil.

As stated, it was in the context of changes in school cultures that the notion emerged that failure, retention and repetition were necessary for the proper and efficient functioning of the school. However, the model that provided for a correspondence between school grade and age group, in a continuous flow, soon demonstrated its weaknesses:

In fact, the regularity of this flow was not confirmed in the day-to-day running of the institutions and in the first decades of the 20th century there were frequent discussions about the distortions that prevented the model from working properly. What was called into question was not the adequacy of the assumption that underpinned the serial model, but the incapacities and limitations of the students

(^{Gil, 2018}, p. 11).

As a result, relations of classification, normalization and standardization were established. On the other hand, contrary to government expectations, what has emerged are the contradictions of a model that, in an effort to include an increasing number of students, has erected discriminatory and exclusionary procedures for those who did not follow the expected pace⁵. Despite these implications, the model has become established and universalized, although the problems are seen as anomalies. However, these points indicate that the distortions may be inherent in the schooling process, or even symptoms of the subjects' resistance or inadequacies to attempts to subsume their differences.

In this respect, analyses that understand the importance of deepening studies that point to the school model as an element that often goes unnoticed, but which acts strongly in the construction of narratives of school failure, are fruitful.

In search of pedagogical effectiveness: the right (and duty) to learn on public policy agendas

With regard to the emergence of investments centered on “meeting the basic learning needs of all people”, I turn to the commitments made by the more than 150 countries that took part in the World Conference on Education for All (1990). In the case of Brazil, the agendas set were recognized in documents such as “Education for all: evaluation of the decade” (^{Brasil, 2000}), whose text mentioned the evolution of the education system as a result of consolidated reforms, where the MEC was able to assume the role of formulator and coordinator of national policies, incorporating the function of monitoring and evaluating school performance.

While recognizing the differences between post-1990 governments, it is worth pointing out that there are continuities regarding the commitments made, for example, in the 1988 Constitution, the Ten-Year Education Plan (PDE) (1993), the LDB and the National Education Plans. The logics used in these documents have led to changes in school curricula and public management models. For example, the 1993 PDE, the result of an agreement between MEC and UNESCO, set out the main challenges concerning the responsibility to ensure the Brazilian population's right to education, collaborating with international efforts to universalize basic education.

According to its authors, these challenges were linked to the heterogeneity of provision. At the time, there was also the need to reduce illiteracy rates and universalize basic education, especially for the poor in urban centers and for students living in rural areas⁶, for whom it was more difficult to “overcome the initial grades of schooling” (^{Brasil, 1993}, p. 20). Despite the obstacles, the text states that the achievement of the goals was favorable, since there was a proliferation of actions aimed at universalizing basic education with quality. In addition, the Plan affirmed the importance of education for the formation of citizens and for the resumption of national development under new values and perspectives.

The novelty was summarized in the objective of “ensuring, by the year 2003, that children, young people and adults have minimum learning content that meets the basic needs of contemporary life” (^{Brasil, 1993}, p. 12). The gap in Brazilian education was illustrated by the “unsatisfactory results of the teaching processes” and the “poor performance of the education system”, despite the notable quantitative expansion.

In detailing the situations that led to this “low productivity”, the issue of quality was emphasized. This element became linked to the definition of learning standards to be achieved in the various cycles, stages and/or grades, as well as through the establishment of school performance targets.

The construction of the painful scenario of education also relied on a different argument: the perception that experiences in external evaluations, although incipient, made it possible to identify more precisely the most critical areas of the system. These evaluations, according to the PDE, were able to show that “deficient evaluation practices” adopted by schools led to “higher proportions of failure than would be expected”. They also made it possible to verify the enormous variability in the learning objectives proposed and actually pursued by schools (^{Brasil, 1993}, p. 23).

It is worth noting that although the document mentions that various studies have supported these statements, there is no mention of them, so that the basis was the results of large-scale assessments and information contained in international documents.

Continuing with the diagnosis, the Plan added that the acquisition of communicative skills was hampered by the formalism of Portuguese teaching, while the development of intellectual skills was insufficient, as evidenced by the declining averages in Mathematics and Science tests. He concluded by pointing out that schools operated with “little or no definition of their learning objectives”, had “precarious methods of curriculum construction” and – due to insufficient initial and ongoing training conditions and a lack of pedagogical support and better teaching resources – teachers faced great difficulties in formulating effective teaching strategies.

If the low standard of quality persisted, the argument went, the severe difficulties of social and economic integration of young people in a complex and increasingly demanding society would increase. One of the coping strategies emphasized the setting of minimum content:

[...] the MEC, with the help of educational representations and society, should propose and specify the national contents capable of guiding the amount of socially useful and universal education to be offered to all children, taking into account their differences

(Brazil, 1993, p. 45, my emphasis).

As you can see, setting content linked to national learning objectives has been seen as a strategy to achieve a quality standard, although without specifying how the acquisition of school knowledge will effectively guarantee the promises of social and economic inclusion for young people. On the other hand, the legislation constantly blames teachers for the precariousness of teaching confronting the situations highlighted as problems.

In addition to the prescriptions aimed at the curriculum, the Plan highlighted the strengthening of the National Basic Education Assessment System (in Portuguese Sistema Nacional de Avaliação da Educação Básica (SAEB). According to the National Institute of Educational Research Anísio Teixeira (in Portuguese Instituto Nacional de Pesquisas Educacionais Anísio Teixeira - INEP), this instrument provides information on the quality of basic education. The first evaluation was carried out on a sample basis, with students from public schools in the 1st, 3rd, 5th and 7th grades of Elementary School (in Portuguese Ensino Fundamental - EF), Mathematics, Natural Sciences and Writing, plus the collection of contextual information from the application of questionnaires, in 1995. From then on, the SAEB changed, as in 1997, when a sample of private schools was incorporated and changes in the target audience were included (now with students in the 4th, and 8th grades of EF and 3rd grade of High School - EM). In 2001, tests were only applied in Portuguese and mathematics, demonstrating the influence of PISA, carried out for the first time in 2000.

There is also an intrinsic relationship between Brazilian policies and the goals agreed by the countries present at the meeting in 2000, again led by UNESCO:

In 2000, 164 countries gathered in Dakar committed to pursue six Education for All goals by 2015. These goals are related to Early Childhood Care and Education; universal elementary education; the development of skills for young people and adults; adult literacy; gender parity and equality; and the quality of education. Approachin⁷g the deadline to achieve these goals, UNESCO asked countries to produce a national report to present what was achieved in the period (results) and how it was achieved (strategies), as well as the challenges presented for the post-2015 period

(^{Brasil, 2014b}, p. 7).

This report presented as achievements the expansion of school attendance by age group, which reached 41.2% of children aged 4 to 6; 95.8% of children aged 7 to 14; and 81.1% of young people aged 15 to 17. Furthermore, it highlighted the establishment of National Curricular Parameters (PCN); the creation of a nationwide Assessment System; improving teacher training, among others. Therefore, the importance of SAEB as a reference for planning educational policy was endorsed.

From these assertions, it is possible to indicate that the changes in perspectives in educational reforms were marked by experiments that, in order to establish themselves, mobilized international relations, institutions and groups of subjects, as well as activated knowledge and political-pedagogical practices. In the process, convincing strategies were created about the usefulness of statistics for tackling situations considered as problems.

As far as consensus-building is concerned, this has led (not without setbacks, resistance, partial adherence, cheating) to the emergence and consolidation of a model for managing school populations in Brazil. In this respect, it is worth questioning some of the logics that shape the use of statistical rationality in the construction of educational diagnoses, which is the aim of the next section.

Some warnings about the processes of quantifying education

The production of national and international comparative indices has occupied considerable space in the formulation of policies in countries all over the world. In this context, statistics has emerged as a technology for managing areas considered to be at social risk, with the aim of identifying situations that should receive attention from government officials. These measurements quantitatively project the changes that governments intend to make, managing individual and collective behavior by producing records on populations. These records emerge

[...] to propose, monitor and evaluate interventions, quantifying their most characteristic and interesting aspects, formulating knowledge and then making it available to governments and society. The knowledge built up by different institutions and experts, based on collected data, records and comparisons, supports administrative decisions to maintain and optimize the desirable characteristics of the population

(^{Traversini; Bello, 2009}, p. 137).

The pragmatic logic⁸ has taken hold in policies, institutionalizing a cycle of reforms unprecedented in the country's history. Despite the strength of these government discourses and practices, there is little research dedicated to examining the criteria involved in their development and also the effects of numerical indices on the production of school failure. On the other hand, there is widespread recognition of the role of large-scale assessments in shaping recent reforms (^{Alves, 2020}).

In this scenario, a transnational actor appears as one of the biggest influencers: the Organization for Economic Cooperation and Development (OECD). According to ^{Villani and Oliveira (2018)}, the development of the assessment programs conducted by the organization has relied on political, economic and social conditions that have allowed these surveys to be expanded, such as: the expansion of education systems in the Organization's area; the standardization of models that have reduced the differences between countries' school systems; the dichotomy between scientific and political objectives in the field of education, etc.

The OECD has established itself as a benchmark for justifying reforms in various parts of the world, especially through PISA, which has ensured that the institution has positioned itself as a leader in the production of educational statistics at international level. This program has been recognized as an instrument of knowledge policy, functioning as a device for generating knowledge that combines social and technical components to establish rules for coordinating and controlling public action in education. These policies mobilize professionals from the OECD and other public and private organizations. In Brazil, numerous researchers have taken part in training courses and disseminated the OECD's methodologies.

Having made these points, I would like to multiply my concern with directing an attentive eye to the elaboration and use of statistics in educational discourse. Taking various studies on the subject as a reference, ^{Gil (2019)} points out that, especially since the 1970s and 1980s, in parallel with the recognition of their potential, research has emerged that problematizes the production and use of numbers as simple and objective descriptions of reality. This research emphasizes the importance of giving historicity to the knowledge in the area and examining its effects.

In Brazil, some researchers have proposed improvements in the criteria and measures used, also advocating the expansion of the information collected to measure the quality of education. This is the case of studies dedicated to producing indicators of educational inequality, given the recognition of the influence of extra-school factors (notably socioeconomic factors)⁹ on the trajectory and performance of groups of students (^{Soares; Alves, 2023}; ^{Soares, 2016}; ^{Soares; Delgado, 2016}).

Another potentiality of these studies is that they highlight the need to construct pedagogical interpretations of the numbers produced, supporting policies to tackle the problems encountered. This creates links between academic research and government activity, since the studies can inform and guide educational decisions. For example: knowledge of the trajectory and learning rates can lead to actions to actively seek out students; they can improve investments in the conditions of school provision (infrastructure, teaching support professionals, management and teacher characteristics); they can stimulate the creation of remedial activities in specific areas or subjects; they can encourage investments in social and family actions, among others.

However, the construction of interpretations capable of supporting internal decision-making by schools is not a simple matter. For ^{Soares (2016, p. 147)}, this explains the limited impact of external evaluations on the routine of educational establishments, leading him to state that “the Brazilian evaluation system was consolidated by emphasizing its managerial uses, not its pedagogical ones”.

Some of the studies that have looked into the effects of the managerial uses of large-scale assessments point, for example, to the influence of PISA on educational agendas. According to ^{Lingard (2016)}, a lot of “numerical evidence” has been used as external justification for changes in national education systems, to the detriment of examining the reasons for the performance of the countries with the best results. In addition, for him, the uses of PISA have focused more on the fetish of rankings than on concerns about equity.

On the other hand, PISA has been pointed out as a style of contemporary reasoning in which international performance calculations are taken as instruments to fabricate principles about who the student, the future citizen, is - and what they should be like. In these terms, statistics establish a field of intervention in order to report claims of social improvement and progress.

Despite the apparent consensus around these logics, studies have questioned the plausibility and intelligibility of statistics in their claim to induce change by constructing groups of people. For Popkewitz and Linblad (2016), thinking about populations implies practices of inclusion that generate exclusion, producing successful (and consequently unsuccessful) types of people.

Arguing in this way provides a different way of studying educational issues than that found in the analytical strategies of empiricism (‘what works!’) and the dialectics of critical educational theory. Thus, our argument about educational statistics is not about them being ‘good/bad’, about their usefulness, or bias; nor is it to censor or condemn numbers or statistics used in education. It is to place these practices within a broader cultural and political context, with rules and norms inscribed in reforms, such as that of policies, through their normalization, division, and exclusion

(^{Popkewitz; Lindblad, 2016}, p. 729-730).

In these operations, students are taken as educational “data”, related to a desired standard, often leaving aside elements such as personal satisfaction or the multiplicity of expectations, meanings and experiences surrounding the educational situation (elements that are not easily captured by conventional measurement mechanisms). In effect, we see the “deficiencies” of students or teachers as the cause of school failure.

In this sense, it's important to see the possible effects of “overall reasoning”, given that within the scope of educational improvement as a necessity for the system, lies the task of leading students towards acceptable behavior. In other words, at the same time as the urgent need to reform education as an alternative to its own crisis, the need to focus investments on a specific type of education and subject is defined.

According to ^{Pereira (2016, p. 254)}, OECD analyses have used the notion that cognition measured by tests is related to the individual's ability to adapt to socially required standards, both from the perspective of acquiring knowledge and the ability to succeed in life. These representations drive the construction of forms of sociability that deny cultural, symbolic and emotional particularities, etc. and insert individuals into a homogenizing logic.

Still with regard to the prescriptions related to the type of subject to be formed, in the case of the BNCC, in addition to the “traditional” functions of the school (socialization of science, preparation for the exercise of citizenship and qualification for work), it is important to mention the norms of a psychological, moral or behavioural nature. In view of the shift towards the pedagogy of competence¹⁰s, the reform has concentrated its efforts on training individuals who know how to apply knowledge and who are capable of planning their future.

In these terms, the demands surrounding the need for young people to develop the skills to draw up their life plan and withstand adversity with resilience have been seen as examples of the intensification of government injunctions on the lifestyles of schooled subjects. On the other hand, in the discourses in favor of the BNCC, the idea that social inequalities are also motivated by the unpreparedness of future citizen-workers, poorly trained by the school “of the past”, is recurrent.

Although the effects of numerical constructions vary and there are many objections to the use of classificatory and comparative assessments between diverse and unequal subjects and countries, ^{Pereira (2016)} points out that poor test results have become synonymous with educational inferiority. In these terms, a causal relationship was established between low student performance, school inefficiency and the unpreparedness of countries to compete in the world.

Despite the strength of these logics, we cannot accept school failure as a matter of course. This is why it is important to consider PISA and other assessments as tools that can also manufacture crises (^{Pereira, 2016}, p. 52). This is because, by measuring narrow aspects of public education, an atmosphere of urgency is created, justifying reforms that legitimize specific political choices. This is the case, for example, with the recurrent blaming of teachers to the detriment of figures that have shown the impact of poverty on school results.

Another situation that deserves a warning is the emphasis on the reduced set of aspects assessed, disregarding educational objectives that are less susceptible or impossible to measure, such as physical, moral, civic and artistic development. This restriction causes a certain imaginative limitation around what a good school education is or should be.

Final considerations

The production and dissemination of statistical indices has reinforced the dissatisfaction of sectors of the population with the political and economic problems facing the country. This contributes to reinforcing the demand for an education based on social rights, which considers the expansion of supply and improvement in the quality of teaching and learning to be central elements for democratizing the participation of citizens in collective life.

In this scenario, educational reforms emerge as a way of achieving “progress” in politics, the economy, culture and democracy. These renewal movements are, however, marked by the actions of multiple subjects, institutions, discourses and political actions, characterized by training models in dispute, linked to different projects for society and education.

However, one common aspect that runs through these initiatives is the attempt to gain legitimacy by adhering to reform projects. In these cases, the discourse is usually based on the rhetoric of school ineffectiveness. Although the diagnoses of school failure sound consensual, it is necessary to point out the diversity of arguments, expectations and meanings of school that feed the proposals for change.

From this perspective, a promising way of studying the logics that legitimized the emergence and approval of recent educational reforms, such as the BNCC, has been to investigate their rationalities. The article therefore focused on problematizing the assumption of a hegemonic school model. In these terms, it was pointed out that the assumption of the efficiency of a certain school model and the stipulation of some (and not other) learning expectations has been established despite the disputes that characterize the Brazilian educational arena.

In line with approaches that aim to overcome the interpretation of phenomena strictly in terms of success or failure, this argument has moved away from dichotomous views. The intention was to point out that, although there are dominant forms of socialization at a given time, such as schooling, these are permanently confronted with other ways of life, related to the conditions of existence, social relations and the history of social groups and subjects.

Finally, it was discussed that the way of thinking in population terms, characteristic of statistical reasoning, provides a way of seeing oneself, constituting references that allow comparison between types of people, considered to be “models of success”. These discourses have acted to produce discrimination and exclusion of groups that don't fit the standard norm. The paradox of these reports is that they place people on a continuum of values that classifies and lists central tendencies, with the extremes seen as pathological. For this reason, it is important to examine how the various technologies of statistical science become political practices.

In this respect, further research is needed, especially into how pedagogical practices are simultaneously involving and excluding certain characteristics of life and people through population and quantitative reasoning.