The aim of the current study is to briefly present the articles composing the dossier titled Implementation of public policies to combat educational inequalities. The herein presented brief considerations counted on the essential contribution by PhD. Professor Ana Cristina Prado de Oliveira of the Universidade Federal do Rio de Janeiro (UNIRIO), who did a remarkable work as dossier collaborator and research partner, and to whom I am immensely grateful.

This dossier is an initiative focused on fostering theoretical and methodological discussions about the implementation of public policies on education and about their association with social inequalities. If, on the one hand, the social and educational inequality topic is investigated in several studies, on the other hand, the implementation topic is not.

Overall, the academic production in the Political Science field analyzes policies based on four core stages, namely: agenda, formulation, implementation and evaluation, which are understood as continuous, dynamic and often overlapping processes. The methodological theoretical policy-cycle model has been gaining room in the Education Policy field (^{BALL; BOWE, 1992}; ^{MAINARDES, 2006}). It comprises three different contexts: influence, text production and practice. Although these contexts are interrelated, they are not necessarily temporal or sequential; therefore, they are featured as a cycle.

Most studies conducted in international and national contexts prioritize the text production context and the analysis conducted at the agenda, formulation and evaluation stages. Moreover, they give secondary importance to the implementation stage, a fact that leads to a “large gap in empirical studies about the policy implementation stage and about several elements and factors capable of influencing it” (^{LOTTA, 2015}, p. 16, our translation).

This gap is perceived in different knowledge fields, such as Education. Recent survey conducted by ^{Oliveira (2019}) has pointed out the scarcity (only 6.1%) of articles about the implementation of educational policies in national and international journals. Overall, the Education field primarily uses studies about policy cycle performed by Stephen Ball and colleagues (mentioned above) to analyze educational public policies. Although the analytical proposition by the aforementioned authors highlights the cyclical dynamics between contexts, studies substantiated by this reference do not often address implementation features at the time to approach the practice context. Studies in this field, which adopt theoretical references deriving from Political Science, rarely focus closer on the implementation stage. However, it is worth highlighting objects of study investigated by fields, other than the Political Science one, such as Health, Social Service and Environment, in the research, in order to avoid perpetuating the gap in studies about the implementation of educational public policies.

The Network of Studies about the Rede de Estudos em Implementação de Políticas Públicas [Implementation of Educational Public Policies] (REIPPE)², which comprises members of several Brazilian institutions, was launched to fill the theoretical and empirical Educacionais gap in the literature. Since 2014, the aforementioned Network has been investigating the theoretical and methodological contribution of the Political Science field to studies focused on addressing the educational policy implementation stage. Many of these studies focus specifically on investigating how the implementation of educational policies can expand or mitigate social and educational inequalities.

The aforementioned purpose also substantiated the elaboration of the herein addressed thematic dossier, whose main goals are to foster discussions about the association between policy implementation and to fight educational inequalities, as well as to promote interdisciplinary approximations to link the Educational Policy field to fields such as Political Science and Sociology of Education.

Due to such ambitious aims, the dossier faced some difficulties in gathering studies focused on investigating the implementation stage - which, as previously mentioned, are not so common in the field - and its association with educational inequality, a fact that also encouraged the approach of other Policy cycle stages in articles composing the dossier. On the one hand, this outcome reinforces the scarcity of studies about the implementation stage and evidences the need of fostering research on this subject. On the other hand, it enriches the dossier, since the aforementioned articles interconnect the policy cycle stages. It happens because, besides implementation, they comprise studies addressing the agenda, formulation and policy evaluation stages. This interconnection reinforces the relevance of not analyzing the dynamics of policies by disregarding their other stages. In fact, implementation depends on, and defines, the cycle stages. Thus, by taking into consideration the central place of implementation and the aforementioned interconnection, the current study addresses other policy stages approached by some of the articles presented below.

The study conducted by António António, Geovana Mendonca Lunardi Mendes and Osvaldo Hernández González, titled Special Education Policies in an inclusive perspective in Angola: context, advances and emerging needs (1979-2017), which is inserted in the topic ‘democratization of access to basic education’, addresses both the agenda and formulation stages. Although the aforementioned study does not use Stephen Ball and colleagues’ policy cycle and theoretical references focused on discussing implementation as analytical lens, it addresses how Angola adopted the political agenda of special and inclusive education debated by international organizations to formulate its educational policies. Angola has been formulating public policies aimed at extending the inclusion of children with special needs in schools, based on the 1948 Universal Declaration of Human Rights, on the 1989 Convention on the Rights of the Child, and on the 1994 Salamanca Statement. Thus, the aforementioned study focuses on investigating educational policies’ agenda and formulation stages from the inclusion perspective.

Another study aligned with the idea of democratization - which, in this case, refers to access to higher education - was conducted by Francilene de Aguiar Parente and Irlanda do Socorro de Oliveira Miléo. The study - which is titled The Ethnodevelopment Course and differentiated and intercultural training: contributions in the educational, socio-political and cultural context of the Amazon - addresses the implementation of licentiateship and bachelor’s degree in Ethnodevelopment at Universidade Federal do Pará (Altamira Campus), which has contributed to the academic and professional training of traditional populations in the Amazonian context. The course, which was formulated and implemented based on concepts inherent to affirmative actions, is not available in other higher education institutions and has specific features aimed at serving a certain public. Thus, it contributes to expand the performance, bargains and political negotiations of these social actors at, and out of, the university. The aforementioned study addresses the implementation of a course created from the set of affirmative actions promoted by the institution, as well as how this initiative can contribute to expand the access to higher education and to take it to hinterlands.

The study Ethnic-racial socialization and racism: from Afro-Brazilian and African knowledge to the construction of ethnic-racial identity by Patrícia Modesto Matos and Dalila Xavier de França — which is lesser aligned to the topic “democratization of access” and more tuned to the topic “ethnic-racial acknowledgement” - aims at analyzing students’ perception about the constitution of ethnic-racial identities, based on ethnic-racial socialization experiences. According to the aforementioned study, one of the ethnic-racial socialization forms takes place through the acquisition of knowledge about the history and culture of a given people; such a knowledge acquisition is guaranteed by law n. 10.639/2003 (^{BRASIL, 2003}), which amended LDB/96 (^{BRASIL, 1996}), by establishing the inclusion of the Afro-Brazilian history and culture in the discipline matrix of Brazilian schools as mandatory. Students’ perception about the formation of ethnic-racial identities enables seeing that the law is implemented in schools. However, deepening law implementation processes is a point yet to be explored in future investigations.

The study by Wagner Silveira Rezende, titled Literacy assessment in Espírito Santo: analysis of Paebes Alfa between 2009 and 2017, focuses on the policy cycle evaluation stage. It analyzes and describes the Literacy Evaluation Program, called Programa de Avaliação da Educação Básica do Espírito Santo - Alfabetização [Basic Education Evaluation Program in Espírito Santo - Literacy] (Paebes Alfa), which is adopted in Espírito Santo State. Although the aforementioned study focuses on investigating the evaluation stage, it innovates by emphasizing the description of program features, rather than of learning outcomes. Thus, it gets close to the implementation stage, since it describes and analyzes different Paebes Alfa elements, as well as pinpoints their limitations and potentials.

Four studies among the ones composing the dossier mainly focused on the policy implementation stage: three of them analyzed the discretion of implementing agents’ performance from the bottom-up perspective, whereas the fourth one used the analytical model of synthesis suggested by Richard Matland to address the implementation type featuring the Base Nacional Comum Curricular [National Common Curricular Base] (BNCC).

It is necessary to briefly deepen the concepts mobilized by these four studies to help better understanding the relevance of studies about the policy implementation stage.

According to ^{Lotta (2015}), the implementation stage is defined as the time when the policy is put into practice, i.e., when it ceases to be a document and becomes a concrete factor through the action of implementing agents. There are two main approaches used to analyze the way policy implementation takes place. The first approach, which is known as top-down, is featured as hierarchical and vertical process, whose decisions come from top to bottom. In other words, policies are formulated by high-level agents and implemented by mid-ranking bureaucracy and street-level agents. Thus, there is separation between administrators (those who carry out the actions) and politicians (those who decide about the actions); implementing agents are subordinated to formulator agents (^{WEBER, 1947}). With respect to the second approach, which is called bottom-up, policy is not fragmented; actually, it is understood as a continuum, wherein policy formulation is interpreted and changed by the actions taken by implementing agents. The analytical approach lies on performed processes, rather than on achieved results. At the core of this analytical perspective, one finds studies focused on investigating implementing agents’ discretion - mainly the ones defined by ^{Lipsky (2010}) as street-level bureaucracy or, street-level bureaucrats, who are public agents of the State working close to the public served by the policy. In other words, they are the ones who effectively deliver the policy-related service to citizens.

Studies based on this analytical scope focus on investigating bureaucrats’ performance, which is understood through practices and interactions established by them at the time to perform their work. These practices and interactions are based on policy standards and rules elaborated at the formulation stage, as well as on bureaucrats’ individual beliefs and values (^{LOTTA, 2015}), a fact that leads each bureaucrat to have different performances at the time to deliver the service. Although rules and standards predicted in policy formulation can conform to bureaucrats’ performance, these agents mobilize their own beliefs and values, a fact that gives them certain autonomy to make decisions about implementation processes. The autonomy regulated by policy rules is termed in the literature as “exercise of discretion”, based on which, implementing agents are not only understood as policy executors, but also as policy-makers.

Although street-level bureaucrats’ performance has already been quite explored in several public policies from other fields, such as Health and Social Service, few studies have focused on investigating these bureaucrats’ performance in the implementation of educational policies. Studies focused on investigating the performance of mid-rank bureaucrats - i.e., those who connect the formulation and implementation stages by linking high-level to street-level agents by translating decisions into actions - are even more scarce (^{CAVALCANTE; LOTTA, 2015}). They account for allowing street-level bureaucrats to put the policies elaborated by high-rank bureaucrats into practice. Therefore, the main function of mid-rank bureaucrats lies on coordinating street-level agents to operationalize the formulated policies.

The dossier presents three studies focused on investigating the performance of implementing agents. One of them - titled The Prêmio Escola Nota Dez (PENDez) - an instrument to reduce educational inequalities in child literacy by Maria Océlia Mota and Diego Mota - focuses on both mid-rank and street-level agents. By investigating the Ceará State Bonus Policy - Prêmio Escola Nota Dez [Grade A school prize] -, the study emphasizes how eight mid-rank and street-level implementing agents appropriated and transformed Pendez, which is seen as instrument to regulate the Programa Aprendizagem na Idade Certa [Program for Appropriate Age Learning] (PAIC). One of the results shows that, although Pendez is a bonus policy based on meritocratic principles, its design also incorporates equity elements that contribute to fight learning inequalities.

The other two studies aligned with this topic focus on the performance of school principals, who can be understood as street-level or mid-rank bureaucrats. There is no consensus in the literature about the bureaucratic level school principals belong to. ^{Oliveira and Abrucio (2018}) conceptually define school principals as mid-rank bureaucrats based on the set of tasks performed by them. The aforementioned authors point out the main tasks performed by school principals and show that they are mostly typical of intermediate bureaucracy level, although some of their tasks are close to the high bureaucracy level, whereas others are close to the street level; thus, there is certain hybridism in school principals’ function. Oliveira and Abrucio (²⁰¹⁸) define school principals as mid-rank bureaucrats, based on such a hybridism. Although ^{Muylaert (2019}) conceptually agrees with the definition by the aforementioned authors, he draws attention to the need of defining the scope of services provided by a given policy in order to empirically define whether school principals are mid-rank or street-level agents. It happens because the hybridism observed in school principals’ tasks can categorize them in one or another bureaucrat type, depending on the analyzed policy.

The study titled Selection of school principals and the meaning of school management: principals’ perceptions about the management plan, by Maria de Fátima Magalhães de Lima, focuses on the discretion of school principals’ performance to investigate the school principal-selection policy adopted in Rio de Janeiro City. The scope of the study lies on investigating school principals’ perception about the implementation of diagnosis and self-assessment - which are elements of the management plan that, in its turn, is featured as instrument used to induce school principals to adopt educational indicators, as well as to define goals and to draw strategies to achieve them. This instrument is also used to measure the technical competence of candidates competing for the school principal positions.

The analytical focus of the study by Tereza Cristina de Almeida Guimarães and Elisangela da Silva Bernado, titled A question of merit: access to vacant places at a municipal public school in São Gonçalo, also lies on investigating the exercise of discretion by school principals as implementing agents. Based on a case study, their research points out that school principals’ discretionary performance may lead to the reproduction of educational inequalities when the principal of the investigated school does not meet the provisions in Semed Ordinance n. 96/2018 (^{SÃO GONÇALO, 2018}), which establishes standards for enrollment in public schools belonging to the Municipal Network of São Gonçalo County / RJ. Consequently, the assessed school did not fill the total number of vacancies established in the aforementioned Ordinance and turned away from the will to democratize the access to Basic Education.

Another analytical focus is addressed in the study titled Implementation study contributions to education policy analysis: a brief discussion of the context of BNCC implementation, by Marina Meira and Alicia Bonamino, who adopted the synthetic model elaborated by Richard Matland to help better understanding the type of policy applied to implement national curriculum policies, mainly BNCC.

Analytical synthesis models are featured as third generation of studies focused on transcending the top-down X bottom-up dichotomy by combining elements of the two approaches in order to create explanatory models for the implementation stage, wherein the perspective of one, or another, analysis is more appropriate. By overcoming such a dichotomy, synthesis models advance in comparison to the two traditional approaches and bring other elements in their core, such as governance, state abilities, among others. Richard Matland is one of the authors aligned to this third generation of studies about implementation, since he proposed an alternative analytical model capable of synthesizing top-down and bottom-up approaches by combining important concepts for the implementation of public policies, namely: ambiguity and conflict.

Ambiguity refers to the degree of uncertainty regarding the means and purposes of policies. In many cases, implementers know the goals of the policy, but they are not sure about what to do to achieve them. In this case, ambiguity is associated with implementation means / processes. In other cases, ambiguity is associated with purposes, since implementers do not know the policy goals. The concept of conflict concerns the degree of implementing agents’ adherence to goals and processes. In other words, conflict has to do with the degree of implementers’ compliance with what should be done and with how it should be done.

By combining the ambiguity and conflict levels of a given policy, ^{Matland (1995}) has drawn up four implementation types, namely: symbolic, managerial, experimental and political. The definition of this typology is presented in the study conducted by Marina Oliveira and Alicia Bonamino, who pointed out the advances and limits of this typology in comparison to traditional approaches. The aforementioned authors used the two concepts - conflict and ambiguity - in a complementary manner in the analysis of national curriculum policies. Based on their preliminary conclusion, the policy under validity - BNCC - falls into policy-type implementation.

The brief presentation of the aforementioned studies has shown that the dossier acquired a comprehensive and diverse nature, not only because it addresses all four policy cycle stages, but because it also addresses several policies encompassing different segments and modalities - such as special and inclusive education, higher education, elementary education and literacy -, that are implemented in different contexts. Among the policies investigated in the aforementioned study, one finds:

Such a diversification is also perceived through the use of qualitative and quantitative methodological approaches. Descriptive statistical data are observed in most studies, whereas the quantitative approach is mostly developed and further investigated in the study by Patrícia Modesto Matos and Dalila Xavier de França, who performed hypothesis tests and analysis of variance (ANOVA).

In this sense, based on different theoretical and methodological perspectives, the articles composing the dossier address extremely relevant issues to help improving knowledge about public policy cycle stages, mainly about its implementation stage. They also contribute to the Education field since they evidence several movements taking place in different contexts, instances and spheres to help better understanding public policies, as well as their effects on the combat to inequalities and on the democratization of access to quality education.