SciELO - Scientific Electronic Library Online

 
 número32“Revisitando” algumas teses do passado: a eleição de 2006 e a disputa antagônica entre PT e PSDB índice de autoresíndice de assuntospesquisa de artigos
Home Pagelista alfabética de periódicos  

Serviços Personalizados

Journal

Artigo

Indicadores

Links relacionados

Compartilhar


Revista Brasileira de Ciência Política

versão impressa ISSN 0103-3352versão On-line ISSN 2178-4884

Rev. Bras. Ciênc. Polít.  no.32 Brasília maio/ago. 2020  Epub 21-Ago-2020

http://dx.doi.org/10.1590/0103-335220203201 

Artigo

How is the Advocacy Coalition Framework Doing? Some Issues since the 2014 Agenda.

Como anda o Modelo de Coalizão de Defesa? Algumas questões desde a agenda de 2014.

¿Cómo va el Advocacy Coalition Framework? Algunas notas desde la agenda 2014.

1Doutoranda em Administração na Universidade de Brasília (UnB). Brasília, DF, Brasil. E-mail: <janainafma@gmail.com>

2Participante do Grupo de Pesquisa em Governo e Políticas Públicas da Universidade de Brasília (UnB). Brasília, DF, Brasil. E-mail: <marcocirilolemos@gmail.com>

3Professor vinculado ao Programa de Pós-graduação em Administração da Universidade de Brasília (UnB). Líder do Grupo de Pesquisa em Governo e Políticas Públicas. Brasília, DF, Brasil. E-mail: <diego1mv@yahoo.com.br>


Abstract:

This article presents an international review of the Advocacy Coalition Framework (ACF) from 2015 to 2018. Forty-six empirical works that use the ACF as the main model or one of the main models for the analysis of public policies were analyzed. Bibliometric data, research methodologies and theoretical components have been considered in this analysis. In addition, this article fills in a gap in previous reviews of the ACF by presenting an exploratory review of ACF applications in Latin America. Our conclusions show that the research agenda proposed in 2014 has evolved over the last few years, but still presents unexplored challenges and possibilities for researchers interested in the use of ACF. We also offer some recommendations for future research.

Keywords: advocacy coalition framework; policy analysis; policy process; policy change

Resumo:

Este artigo apresenta uma revisão internacional das aplicações do Advocacy Coalition Framework (ACF) de 2015 a 2018. Quarenta e seis trabalhos empíricos que utilizam o ACF como modelo principal ou um dos principais modelos para a análise de políticas públicas foram analisados. Dados bibliométricos, metodologia de pesquisa e componentes teóricos foram considerados na análise. Além disso, este artigo preenche uma lacuna nas revisões anteriores do ACF ao apresentar uma revisão exploratória do uso do modelo na América Latina. As conclusões mostram que a agenda de pesquisa proposta pelos autores do ACF em 2014 evoluiu nos últimos anos, mas ainda apresenta desafios e possibilidades para pesquisadores interessados no uso do modelo, que permanecem inexplorados. Algumas recomendações para pesquisas futuras são apresentadas.

Palavras-chave: coalizões de defesa; análise de políticas públicas; processo político; mudança em políticas públicas

Resumen:

Este artículo presenta una revisión internacional de las aplicaciones del Advocacy Coalition Framework (ACF) de 2015 a 2018. Se analizaron 46 estudios empíricos que utilizan el ACF como modelo principal o uno de los modelos principales para el análisis de políticas públicas. Los datos bibliométricos, la metodología de investigación y los componentes teóricos han sido considerados en el análisis. Además, este artículo llena un vacío en las revisiones anteriores de ACF al presentar una revisión exploratoria del uso del modelo en América Latina. Las conclusiones muestran que la agenda de investigación propuesta por los autores de ACF en 2014 ha evolucionado en los últimos años, pero aún presenta desafíos y posibilidades para los investigadores interesados en usar el modelo, que aún no han sido explorados. Se presentan algunas recomendaciones para futuras investigaciones.

Palabras clave: coalición promotora; análisis de políticas públicas; proceso político; cambio en las políticas públicas

Introduction

The Advocacy Coalition Framework (ACF) is a model for analyzing public policies proposed by Paul Sabatier (1988), whose main interests included how changes in public policies occur and how they guide the learning of the actors involved in policy subsystems during i mplementation processes. Created in the beginning of the 1980s, it was originally consolidated in the book Policy Change and Learning: An Advocacy Coalition Approach (SABATIER; JENKINS-SMITH, 1993), in which six empirical studies were presented. The authors’ two most recent works presented a history of its foundations and evolution, an overview of the research program and proposed a research agenda for future applications (JENKINS-SMITH; NOHRSTEDT; WEIBLE; INGOLD, 2018; JENKINS-SMITH; NOHRSTEDT; WEIBLE; SABATIER, 2014).

Systematic reviews of ACF applications have also been conducted over time. In this article we will focus on two of them. The first reviewed 80 studies that applied the model between 1987 and 2006 (WEIBLE; SABATIER; MCQUEEN, 2009). The second examines 161 applications of ACF conducted between 2007 and 2014 (PIERCE et al., 2017). This article presents a new international review of the application of the ACF, considering the period from 2015 to 2018.

The study was initially motivated by our interest in verifying how the most recent ACF applications have addressed the research agenda proposed in 2014 and reinforced in 2018. During our study, however, we noted an almost total lack of Latin American works covered by recent reviews, with only one Brazilian study identified (CARVALHO, 2001). In 2015, however, a group of Brazilian authors promoted a review of the use of ACF in Brazilian theses and dissertations and found 23 applications of the model published between 2010 and 2013 (CAPELARI; ARAÚJO; CALMON, 2015). Thus, we conducted an exploratory search for ACF applications in Latin America, reviewing works in Spanish and Portuguese. For methodological reasons, the results of this search are presented and discussed separately.

The remainder of the article is structured as follows. Section 2 presents an overview of the ACF theoretical background and Section 3 explains the methodology. Section 4 covers the bibliometric data analysis and the results, as well as examining research methodologies and theoretical components of the studies analyzed. Section 5 presents the search results for applications of the ACF in Latin America and some related considerations. The final section summarizes our conclusions and recommends possibilities for future research.

Theoretical Background

The ACF is a model for public policy analysis that seeks to understand changes in public policies and the learning processes resulting from implementation. It adopts the premise that the process of elaborating public policies is complex and therefore those who wish to exercise some influence on this process need to specialize. This specialization occurs in policy subsystems, whose participants seek to regularly influence public policies of interest to them. In this way, the ACF adopts the policy subsystem as its unit of analysis (SABATIER; JENKINS-SMITH, 2007).

The ACF assumes a model of the individual as one whose behavior is guided by a system of normative beliefs. This system is composed of three hierarchical levels which aggregate these beliefs according to the individual’s resistance to change. Deep core beliefs have to do with assumptions related to ontological and normative assumptions about human nature and their fundamental values. They are the most difficult beliefs to alter. The policy core consists of beliefs related to the public policy itself, including the seriousness and the cause of policy problems within the subsystem. Finally secondary beliefs refer to more specific operational issues regarding policy. These are more likely to be modified (SABATIER; JENKINS-SMITH, 1999, 2007).

To facilitate the understanding of policy subsystems, the ACF proposes the aggregation of political actors into advocacy coalitions, based on their shared beliefs and coordination strategies. In addition, the model indicates six resources that can be used by coalitions to guide their strategies: the legal authority to make policy decisions, public opinion, information, mobilizable supporters, financial resources and skillful leadership (SABATIER; JENKINS-SMITH, 1999, 2007).

Understanding policy change and policy-oriented learning are two of the main objectives of the model. Learning can occur on a personal level, within a coalition or between coalitions. In terms of change, the ACF assumes that governmental programs are translations of policy-oriented beliefs that can be conceived of and measured hierarchically as belief systems. For this reason, changes that occur in the policy core are greater, while changes in secondary aspects are smaller. The ACF establishes four pathways that lead to change: policy-oriented learning, significant external or internal subsystem changes (shocks), and negotiated agreements between rival coalitions (JENKINS-SMITH et al., 2014).

The ACF currently consolidates twelve hypotheses related to its main theoretical aspects: advocacy coalitions, policy-oriented learning, and policy change. These hypotheses have been proposed by the authors as a flexible part of the model, subject to revision to the extent that new evidence from applications introduce contributions that permit the progressive promotion of necessary adjustments (for a detailed description of each one of the model hypotheses and an overview of the conducted reviews, see Jenkins-Smith et al, 2014, p. 195-204). The newest ACF’s review maintains the same flow diagram and the twelve hypotheses, indicating a certain stability in the model in the last years (JENKINS-SMITH et al., 2018).

Although the last two reviews of the ACF have found elements that support and refute some of the hypotheses established, the studies have not yet been sufficient to incorporate any changes in the framework. If we use a decision-making process metaphor, we could paraphrase Weiss (1979, p. 429) when she says that “it is not the findings of a single study nor even a body of related studies that directly affect policy. Rather it is the concept and theoretical perspectives that social science research has engendered that permeate the policy-making process”. This enlightenment notion of the use of knowledge can be applied here to understand the nonlinear way that new findings provided by applications of ACF applications may end up influencing alterations in the framework. Until then, ACF flow diagram and its main elements remain the same, as shown in Figure 1.

Source: Jenkins-Smith et al. (2018, p 143)

Figure 1. Flow Diagram of the Advocacy Coalition Framework 

This does not mean however that formulators of the ACF have not recognized the need to improve the model. Therefore, the research agenda established in 2014 and reinforced in 2018 (Table 1) is an invitation to ACF analysts all over the world to move the framework forward.

Table 1. ACF’s Current Research Agenda 

1. Reconsider the ACF belief system.
2. Advance the theory and measurements of learning.
3. Refine the theory of coalition structures and coordination.
4. Develop a hierarchy for coalition resources.
5. Study venues and forums within policy subsystems.
6. Use the ACF for comparative public policy research.
7. Focus on nascent and mature policy subsystems.
8. Expand our understanding of science and policy analysis in the policy process.
9. Establish common methods of data collection and analysis for applying the framework, identify trade-offs in using different methods, and promote contextually-based theoretical innovations.
10. Explore the need for theoretical refinement emanating from application in nontraditional setting.

Source: Adapted from Jenkins-Smith et al. (2018) and Jenkins-Smith et al. (2014).

This theoretical background and the research agenda suggested by the authors that developed ACF provide the main elements that will be considered in the analysis of applications of the model. These issues have been considered in previous ACF’s reviews (PIERCE et al., 2017; WEIBLE et al., 2009) since they provide an opportunity to identify which elements of the framework are more often used, how their relations are explored, which methods are applied and how researchers have sought to overcome theoretical and methodological difficulties and impasses in the use of ACF.

Method

In this article, we reviewed 46 empirical works that use the ACF as the main model or one of the main models for the analysis of public policies. This section presents the search, filter and exclusion criteria used for the selection of articles. The initial search was performed on May 26, 2018 using the Web of Science (WoS) database from 2015 to 2018, returning 143 results. We used the search term “advocacy coalition” which also includes works that use the complete term advocacy coalition framework. The acronym “ACF” was disregarded because it brought up numerous search results related to terms such as “activated carbon filter”, “aluminium chlorofluoride”, among others.

After this step, we selected the articles with impact factors in Journal Citation Reports (JCR) or the SCImago Journal Rakings (SJR) equal to or greater than 0.8, and those classified in the Qualis Periódicos ratings with values equal to or greater than B3. This methodological option is justified by our preoccupation in finding publications of high quality, considering established international indices such as the JCR and SJR, as well as the Qualis-Periódicos, which is widely accepted by the Brazilian academic community. After applying this filter, 26 articles were eliminated.

The third step consisted of analyzing the titles and abstracts, and when in doubt, the introduction to verify the pertinence of the search results. After this analysis, 49 articles were excluded because they did not deal with the ACF or because they referenced the model in their theoretical sections, but did not use it in their data analysis.

The fourth step consisted of reading the articles in their entirety. Twelve articles were excluded because they did not use the ACF as their principal model or one of their principal models in conducting their research and interpreting their results; three articles were excluded because they were already considered in other articles about the same policy by the same authors; and seven theoretical works were excluded, because they did not permit the analysis of ACF applications, which is the objective of this review.

Content analysis was conducted on the 46 articles. Three groups of codes were used. The first included eight codes related to bibliometric characteristics such as journal, year of publication, authors, institutional affiliation, countries, etc. The second group considered five codes on research methodology. The third group included eleven codes related to theoretical aspects, divided into three groups: coalitions, policy change and policy-oriented learning. All the codes are detailed in the respective sections below.

Results and Discussion

In this section we will present a bibliometric data analysis, results regarding the research methodologies and discuss the theoretical components involved in the 46 works analyzed. In this case, comparisons will be made with the results found in the review conducted by Pierce et al. (2017) and the extent to which these articles converge with the research agenda proposed by Jenkins-Smith et al. (2014, 2018).

Bibliometric Data Analysis

This analysis considers the journals, the distribution of the publications per year, the authors and their institutional affiliation, countries, as well as the government level and the thematic and geographic characteristics of the policies analyzed.

Journals. The articles were published in a total of 38 distinct journals, with 72% of them having a single prior publication dealing with the ACF, as opposed to 28% of them which had 2 to 4 prior publications related to the ACF. The journals that had published more than one article were as follows: Review of Policy Research (4); Policy Studies Journal (3); Energy Policy (2); Public Administration (2); and Policy Sciences (2). This result provides something of a contrast with the review performed by Pierce et al. (2017), in which a little less than half of the journals (45%) had only one article about the ACF. It may be observed that public policy journals continue to be the ones that most often publish studies of the ACF, which is natural given the model’s objective. These journals cover a wide variety of subjects, which demonstrates the diversity of research agendas that use the ACF and the model’s flexibility. They include general public policies such as the Policy Studies Journal (e.g. STRITCH, 2015), public policies in specific areas and/or regions, such as Food Policy (e.g. MOCKSHELL; BIRNER, 2015) or the Journal of European Pubic Policy (e.g. BELSCHNER, 2015) and even other areas that are not specifically related to public policies such as the Journal of Research in Music Education (e.g. SHAW, 2018) and the Maternal and Child Health Journal (e.g. STEINMAN et al., 2017).

Publications per Year. Despite this review’s short duration, we have found a certain stability in the number of publications per year, taking into consideration that our selection was made in May 2018, with the following distribution: 2015 (13); 2016 (10); 2017 (17) and 2018 (6).

Authors. The 46 articles analyzed were written by 111 authors, with 15 (33%) being written by a single author, and 31 (67%) by two or more authors, up to a maximum of 7 authors. The average of 2.4 authors per publication was a little above that found by the review of Pierce et al. (2017), which averaged 2 authors per publication. The data, however, reinforces the trend of multiple authors for articles on ACF applications. Of the 46 articles, 42 (91%) had different lead authors, with just four having the same lead author (KUKKONEN; YLA-ANTTILA; BROADBENT, 2017; KUKKONEN et al., 2018; MOYSON, 2017, 2018).

In terms of the countries of origin of the researchers’ associated institutions, the countries with more than one publication were: the United States (14); Germany (4); Canada (4); the United Kingdom (4); Belgium (2); the Netherlands (2); Sweden (2); and Switzerland (2). Considering the publications that were exclusively from a single country, Europe contains almost half of these applications (20), which demonstrates the widespread diffusion of this model among European researchers. The other publications were from Asia (2), Oceania (1) and institutions from different continents (5). We did not find publications originating from institutions in Africa or Latin America. It should be noted, however, that four articles analyzed public policies in these regions, including two African comparative studies (AAMODT; STENSDAL, 2017; CHIKOTO-SCHULTZ; UZOCHUKWU, 2016) and two comparing Brazil with other countries (MOCKSHELL; BIRNER, 2015; NICOLLE; LEROY, 2017). All were produced by researchers from North America or Europe. This limitation, which also occurred in the review conducted by Pierce et al. (2017), may be a consequence of selecting works only in the English language.

Policy Subsystems, Regions and Areas of Interest. The ACF continues to be a model applied all over the world, with most of the studies focused on policy subsystems in Europe (17) and North America (16), followed by Asia (6), Africa (2), South America (1) and Oceania (1), as well as articles that compare policy subsystems from different continents (3). For example, Kukkonenet al.(2018) compare the influence of international organizations on climate policy in Canada, the United States, Brazil and India. Aamodt and Stensdal (2017) analyze the coalition strategies that influence climate policy in Brazil, China and India. These applications involve 22 countries, with those concerning the United States still dominating (12), followed by the European Union (5) and Canada (4). Belgium, China, Sweden and Switzerland appear with 2 studies apiece. The other countries are studied individually (9) or in a comparative manner (8) within a single study, for a total of 17 articles. Examples of studies of less common countries are those of Finland (HARRINKARI; KATILA; KARPPINEN, 2017), Iran (KHAYATZADEH-MAHANI; BRETON; RUCKERT; LABONTE, 2017), Indonesia (LEONG, 2015) and a comparative study between Ghana and Uganda (MOCKSHELL; BIRNER, 2015).

Following its original tradition, the ACF continues to be used mainly in studies of the Environment and Energy (26). These are followed by Economic Policy (7) Health (7), Social Policy (3) and Education, International Relations and Security (1 study apiece). This predominance of the use of the ACF in the environmental area is justified by the indication of this model for policies which involve a large degree of conflict and a great divergence of opinion on scientific findings. However, this has not stopped the model from being used in less controversial areas such as Education. It may be observed, for example, that more than 50% of the studies in the Environmental and Energy area identify two coalitions in the policy subsystem (e.g. ANDERSON; MACLEAN, 2015; HUGHES; MEEKLING, 2017; LEIPPRAND; FLACHSLAND; PAHLE, 2017), while Education studies have identified four coalitions (Shaw, 2018).

The target policies of these studies occur on a national (16), transnational (13), state (12) or municipal/local level (5). One possible explanation for the predominance of national and transnational policies is the focus on environmental policies, which usually have a national or global impact. Of the 26 studies of the Environment and Energy area, for example, 69% analyze policies on a national or transnational level. The same is true of economic policies. 100% of the articles in this area focus on the national or transnational level. An example of these studies are regulatory policies in the European Union (FENGER; QUAGLIA, 2016; SMITH et al., 2015). Of the 11 studies in the Social Policy, Education and Health areas, however, 82% were on the state or municipal/local level. These policy examples include adoption, cultural heritage, and arts education for children (LEE, 2016; MOSLEY; GIBSON, 2017; SHAW, 2018). The results are summarized in Table 2.

Table 2. An Overview of ACF Applications (2015-2018) 

Journals Number of Applications %
Review of Policy Research 4 8.7
Policy Studies Journal 3 6.5
Energy Policy 2 4.3
Policy Sciences 2 4.3
Public Administration 2 4.3
Others (Just 1 application) 33 71.7
Total 46 100
Year of Publication Number of Applications %
2015 13 28.3
2016 10 21.7
2017 17 37
2018 (until May 26) 6 13
Total 46 100
Continents of Publishing Institutions Number of Applications %
Europe 20 43.5
North America (United States and Canada) 18 39.1
Intercontinental 5 10.9
Asia 2 4.3
Oceania 1 2.2
Total 46 100
Continents of Analyzed Policy Number of Applications %
Europe 17 37
North America (United States and Canada) 16 34.7
Asia 6 13
Intercontinental 3 6.5
Africa 2 4.3
South America 1 2.2
Oceania 1 2.2
Total 46 100
Policy Area Number of Applications %
Environment and Energy 26 56.5
Economic Policies 7 15.2
Health 7 15.2
Social Policy 3 6.5
Education 1 2.2
International Relations 1 2.2
Security 1 2.2
Total 46 100

Source: Elaborated by the authors

Research Methodologies Adopted in Applications of the ACF

We verified the nature of the work (descriptive, explanatory, exploratory), the approach (qualitative, quantitative, mixed); and the research method used (survey, experimental/semi-experimental, documental research, case study, action research, ethnography, multiple methods). In addition, we verified the data collection instruments (questionnaire, interview, documents, observation, focus groups, others, unidentified), and analysis techniques used.

Nature and Approach of the Research. Of the 46 articles analyzed, 31 are descriptive (67.4%), followed by 11 studies of a mixed nature (23.9%), three explanatory studies (6.5%) and one policy intervention (2.2%). The predominance of descriptive studies is reflected in the approaches used by the researchers: 33 (71.7%) works adopt a qualitative approach, 11 (23.9%) adopt a mixed approach, and just two (4.3%), a quantitative approach. In comparison with the most recent international review, this reinforces the preference of researchers for qualitative (66%) or mixed (24%) approaches (PIERCE et al., 2017). One of the critiques made of the ACF is that it reveals what is already obvious, the identification of the opposing sides in policy debates. Sabatier and Weible (2007) argue, however, that its objective is much greater, because the model should make it possible to understand the belief system of the coalitions and their relationship with policy change. The predominance of studies of a descriptive nature shows that the ACF is still often used to identify coalitions and describe how their actions influence the policy process (BARNES; VAN LAERHOVEN; DRIESSEN, 2016; CHIKOTO-SCHULTZ; UZOCHUKWU, 2016; MCDONALD; GALLAGHER, 2015), without, however, establishing the causal relationships between the model’s categories.

Methods. Isolated case studies are the most used method in the analyzed applications (50%), followed by documentary research (21.7%) and questionnaires/surveys (10.9%). Combined methods such as case studies and participatory observation or documentary research were identified in 6.5% and 4.3% of the articles respectively. The predominance of the case study in ACF applications is justified, given that the proposed unit of analysis is the policy subsystem, which leads most researchers to a deeper analysis of a specific policy process.

Collection instruments. Documentary research combined with interviews is predominant (47.8%), followed by documentary research alone (21.7%). Other combinations were questionnaires and interviews (6.5%), interviews, documentary research and questionnaires (6.5%), interviews, documentary research and participatory observation (4.3%) and interviews alone (4.3%). In comparison with the most recent review, we can perceive the growth of the use of interviews, which were present in 70% of the applications analyzed. Pierce et al. (2017) had already pointed out a doubling of their frequency from 30% to 63% in relation to the previous review. We also observed an appreciable growth in data collection through documentary research, from 58% (PIERCE et al., 2017) to 85%. The combined utilization of these two instruments went from 10% in Weible et al. (2009) to 40% in Pierce et al. (2017) and 59% in this review, confirming a trend in ACF applications.

Analysis techniques. Content analysis was used by most of the applications analyzed (73.9%), followed by descriptive statistics (13%), discourse analysis (6.5%) and inferential statistics (2.2%). One case, even though it cites the use of documents, does not specify the use of a specific analytical technique. It is interesting to note in this review the novel use of the ACF as a policy intervention. Kershaw, Swanson, and Stucchi (2017) demonstrate how a non-profit coalition without a party affiliation was created, and which resources and strategies were used to increase investments in social assistance to the younger generations of Canada. The intervention was delimited based on the premises of the ACF.

The predominance of descriptive, qualitative studies suggests that coalitions and beliefs are mainly identified through documentary research and semi-structured interviews, resulting in large lists of stakeholders which in general include governmental agencies, private operators, media, social organizations and the general public. It is worth noting some efforts to uncover stakeholders’ positions and their interrelations, such as through the use of Q method to condensate views, opinions and ideas in three different clusters representing the devil shift in water privatization in Jakarta (LEONG, 2015) or in the use of cluster analysis and silhouette means to identify coalitions and cooperative interactions with friends and foes (ELGIN, 2015).

In addition, the predominance of descriptive studies reinforces the need to apply ACF in research that seeks to establish causal relationships between the explanatory factors and policy change. This does not necessarily mean the use of quantitative methods, but the adoption of methods which allow exploring how the causal mechanisms work in the policy processes investigated, such as Process Tracing or Qualitative Comparative Analysis (QCA). Comparative studies can also benefit from these methods, exploring how the causal mechanisms operate in different contexts. Although case studies have limitations to make causal inferences because of their small N, they offer advantages in identifying complex causal mechanisms such as those proposed by the ACF (beliefs formation and policy learning), which are difficult to measure with quantitative variables (GEORGE; BENNETT, 2004).

It is also important to observe the use of discourse network analysis to identify coalitions’ beliefs and preferences (DUYGAN; STAUFFACHER; MEYLAN, 2018; KUKKONEN et al., 2018; MOCKSHELL; BIRNER, 2015). How discourses changed over time has not been explored, however. This could contribute to develop the theory of coalition formation and maintenance.

No coordinated experiences of common approaches among researchers were observed, either in the use of data collection instruments or in the analysis of the results, as suggested by the authors of the model. It may be necessary to provide arrangements that favor the formation of networks of researchers interested in the ACF around the world. This would be helpful, for example, to analyze if similar kinds of coalitions and beliefs system structures are found in similar subsystems in different countries.

Theoretical Components Considered in these ACF Applications

In this section, we analyze the categories related to the ACF’s central theoretical components: coalitions, policy-oriented learning and policy change. We consider the theories and model elements explored in the articles, number of coalitions identified, the beliefs system and the resources referenced, actors’ stability or defection, policy change, type of change, explanatory factors of change, policy-oriented learning and the levels identified.

Theories and Models. The ACF was applied alone in 57% of the studies analyzed. This demonstrates the robustness of the analysis model for the policy process. Contrary to the previous reviews conducted by Pierce et al. (2017) and Weible et al. (2009), none of the works combined the ACF with the policy cycle model. More often, ACF was combined with Institutional Theory approaches (4 cases, 9%), Stakeholder Analysis, Multiple Stream Framework (2) and the Narrative Policy Framework (2), each representing 4% of the cases. Other theories and models were found once in each of the following areas: Ecology Policy, Cultural Theory, and Discourse Theory, among others.

Model elements. Among the ACF elements, the policy subsystem is the most utilized, appearing in all the analyses. In terms of coalitions, the studies identified between zero and five coalitions, with most identifying two (54%) or three (24%) coalitions per country analyzed. The others identified zero (7%), one (9%), four (7%) or five (2%) coalitions. The applications that could not identify coalitions used discourse network analysis, but the networks were not clustered enough to identify clear coalitions (DUYGAN et al., 2018; KUKKONEN et al., 2018). Some other studies did not specify the number of coalitions (11%), because they focused on specific components of the ACF, like policy learning (MOYSON, 2017) or the role of specific kinds of actors on the policy process (CHIKOTO-SCHULTZ; UZOCHUKWU, 2016; NEWELL, 2018). The presence of two coalitions has been the most common in studies that use the ACF since its formulation and is a result of the nature of the model itself. This has not prevented, however, its use in the study of more fragmented policy subsystems.

In terms of beliefs, 45 articles (98%) identify coalition beliefs, but of these, just 6 (13%) identify beliefs on the three levels proposed by the model (deep core beliefs, policy core, and secondary aspects). Policy core beliefs were discussed in all of them and were the main element used to identify coalitions. This indicates that the belief system has not been used in the way proposed by the model, a fact that can be attributed to the difficulty in identifying deep core beliefs and in clearly separating the policy core from secondary aspects. Sabatier and Jenkins-Smith (1999) sought to remedy this difficulty which appeared during the model’s first version, by incorporating a revised structure of belief systems and aspects related to each of the levels.

Although the authors suggested the possibility of integrating the last two categories (JENKINS-SMITH et al., 2014), none of the analyzed works pursued proposals in this direction. Smith et al. (2015) suggests however that a rigid distinction between the three levels of belief overlooks the potential for policy ideas to be flexible enough to bridge the different levels. They argue that the concept of chameleonic ideas can play an important role in long-term efforts to effect policy change.

In terms of elite beliefs, although studies show that elites continue to dominate the policy process, they also highlight that non-state actors have increasingly permeated the policy process, through formal and informal strategies, such as the use of venues and the influence of public opinion (CHIKOTO-SCHULTZ; UZOCHUKWU, 2016; LEE, 2016; WONG, 2016). The devil shift phenomenon is also not limited to political elites, as argued by Leong (2015), who gives empirical support to this idea, by showing evidences that devil shift can be applied more broadly to different coalition actors, including private operators (LEONG, 2015).

The ACF envisions resources as one of the elements in analyzing coalitions. More than half of the studies (54%), however, do not analyze any of the resources proposed by the model, while 11% analyze all of them. The resources that appear more often are the position of authority (28%), public opinion (24%) and acquiring information (20%), followed by skillful leadership, mobilizable supporters and financial resources (4% apiece). The acquiring of technical and scientific information is one of the resources that the ACF is particularly interested in, because it can be used in various ways in policy debates, such as arguing against rival coalitions, convincing members of the government, or mobilizing public opinion. Based on this, the ACF includes researchers as actors of strategic importance to coalitions.

Results however have not always supported the theory. Despite the amount of scientific information about the health problems caused by tobacco, for example, pro-tobacco coalitions were able to avoid the ban of shisha smoking in public places in Iran (KHAYATZADEH-MAHANI et al., 2017), as well as influence the “Better Regulation” policy in European Union, under the obscured strategic action of British American Tobacco (BAT) (SMITH et al., 2015). Studies also show that no matter how accepted and consolidated scientific information is, it may be insufficient to promote desired changes, when coalitions are not able to properly organize themselves to influence the policy debate (BARNES et al., 2016; ULMANEN, SWARTLING; WALLGREN, 2015). On the other hand, even when stable parameters favor stasis, different types of information combined with the right narratives, including not only scientific evidence and benefit-cost analysis, but also personal stories, strategically showcased to policymakers, were able to promote policy change in California’s law to extend foster care (MOSLEY; GIBSON, 2017). Studies tend to show that technical information is a crucial resource, but it is dependent on the coalition’s capacity to act strategically to influence the policy process.

In addition, Jenkins-Smith et al. (2014) suggest that these resources can be arranged hierarchically in relation to their utility and effectiveness, as perceived by coalitions. The work of Pierce (2016) presents some results that can contribute to efforts in this direction. The author performs a quantitative analysis to explore the relationship between resources and the strategies used by coalitions to change hydraulic fracturing policy in Colorado. Analyzing two competing coalitions, the author examines, among other things, how the losing coalition (pro-fracturing) had greater financial resources, and the winning coalition (anti-fracturing) had more public support. This study is a counterpoint to previous studies that argued that coalitions that have greater resources or access to authorities dictate the policy process.

The last aspect observed in relation to coalitions was whether studies discuss their stability or actor defections. This theme is not covered by 85% of the studies (39), while 9% did identify coalitions being abandoned by actors (4) and 6% verified their stability (3). The effort to understand the coalition structure and motives for defections or stability is also an aspect of the model that needs to be better explored according to Jenkins-Smith et al. (2014).

The second element that appears most often in these studies deals with external events, which are present in 72% of the articles. Even though most of the analyses confirm the influence of external events on policy change, there are studies that do not confirm this relationship. MENAHEM and GILAD (2016), for example, conclude that despite various external shocks during the 1990s, such as the unexpected migration of almost a million Soviet citizens to Israel as well as a series of droughts and water contamination events, Israel’s water policy remained unchanged, which permitted its excessive use and has resulted in a lack of fresh water in its reservoirs and the salinization and contamination of its natural reserves.

Stable parameters and the intermediate elements of the model are present in 22% (10) and 13% (6) of the studies respectively. Stable parameters are explored in comparative studies as they are treated as variables of context that influence the policy process. The study of climate-advocacy coalitions in Brazil, China and India, for example, show that external parameters, like political economy and institutional structures are crucial to explain the coalitions’ policy influence, as they define the rules in which the coalitions operate (AAMODT; STENSDAL, 2017). Sotirov and Winkel (2016), on the other hand, argue that despite the different political systems and socioeconomic contexts of German and Bulgaria’s forest subsystems, three coalitions were formed based on similar policy core beliefs, members, and coordinated behavior in both countries. Stable parameters and long-term coalition opportunity structures are also used to explain policy stability, even in the presence of new technical knowledge (HUGHES; RITTER; LANCASTER; HOPPE, 2017) or change in external events (MENAHEM; GILAD, 2016).

Only four studies use all of the model’s elements (9%), analyzing a variety of subjects: breastfeeding policy in Washington (STEINMAN et al., 2017), the privatization of water in Jakarta, Indonesia (LEONG, 2015), the prohibition of smoking shisha in public places in the Province of Kerman in Iran (KHAYATZADEH-MAHANI et al., 2017) and the use of police dogs in the fight against drug trafficking in New South Wales in Australia (HUGHES et al., 2017). These works all have in common the fact that they study policy on a state or local level, which can facilitate the use of the whole framework.

Policy Change. Understanding policy change is one of the ACF’s main objectives. Changes are classified as major when they affect the policy core, and minor when they affect its secondary aspects. 43 articles (93%) identify some type of change. Of these 43, 18 identify major changes (42%), 11 identify minor changes (26%) and 14 do not specify a type of change (33%). Compared with the review by Pierce et al. (2017), which found the identification of the types of change in just 12% of its articles, the data indicates that this category has been better explored in more recent ACF applications. Among the 43 articles that identify some form of change, many also point out the pathways that lead to change. Just five articles (12%) do not analyze any of the paths listed by the model. 20 studies (47%) analyze external and internal events, thus making them the most often cited pathways to change, followed by policy-oriented learning, present in 10 analyses (23%). External or internal events analyzed in isolation represent 12% and 19% of the cases respectively. Only three studies (7%) consider all the pathways.

It should be noted that pathways are also used to analyze stasis or lack of policy change (HUGHES et al., 2017; KHAYATZADEH-MAHANI et al., 2017; MENAHEM; GILAD, 2016). This sheds light on an important discussion that could improve the way policy analyses are conducted, which is to consider policy dynamics. As argued by Capano (2009), policy is an ongoing process in which change and stability co-exist, so they should not be treated separately. Possibly, the elements of ACF would be better explored if analysts focused on policy dynamics, considering how coalitions operate, explore venues and mobilize resources both to promote and to avoid policy change.

Policy-Oriented Learning. Most of the reviewed studies (32 cases, 70%), even when they identify some type of learning do not analyze the level in which they occur. Of the remaining 30%, learning on the coalition level is the most analyzed (6 cases, 43%), followed by learning on an individual level or between coalitions (3 cases, 21% apiece). Just two studies (14%) analyze learning on more than one level.

Results of two different studies about forestry policies in Canada and in Sweden show that despite the advocacy by academics calling for climate change adaptation or the occurrence of natural disasters pressuring for change, dominant coalitions were still able to determine the direction of debate and policy (ANDERSON; MACLEAN, 2015; ULMANEN et al., 2015). Both studies argue that the presence of forums or arenas for debate would facilitate policy-oriented learning across competing coalitions. Beyond the forums proposed by the ACF, studies have also shown that a wider view into the institutional arrangements that favor or trigger policy learning processes may bring possibilities to a deeper understanding of how these processes are more likely to occur (MOYSON, 2018).

Jenkins-Smith et al. (2014) encourage analysts to reexamine the concept of learning and its technical implications, placing emphasis on clearer concepts and measurements of the products of learning. Some studies have advanced on this debate, showing that the link between policy learning and policy change may be not as direct as it is claimed in theory. These studies mainly indicate that the development of new knowledge is not always absorbed by the policy subsystem and when it is, it does not always lead to change in the beliefs system (FENGER; QUAGLIA, 2016; MOYSON, 2017; RIETIG, 2018). Furthermore, a change in the beliefs system does not necessarily lead to change in policy preferences or policy change (MOYSON, 2017). Moyson (2017), uses a quantitative approach to show this inconsistency between the policy beliefs and the policy preferences of the actors involved in the process of liberalizing the railroad and electricity sectors in Belgium. These works have somehow contributed to the understanding that policy change may not be the best variable for identifying policy learning.

The ACF in Latin America: Some Other Findings

The most recent review of the ACF did not find any application of this model in Latin America (PIERCE et al., 2017). This led us to search for applications of the ACF in Portuguese and Spanish in SciELO database. Since the main intention was exploratory, to retrieve the maximum number of results no date filters and no impact factor criteria were considered. For this methodological reason, we did not include the findings in the previous discussions, where English language articles received a a different treatment. The search was performed on July 9, 2018 without the application of filters to retrieve the maximum number of results. The search terms “advocacy coalition”, “coalizões de defesa” and “coalición promotora”, yielded 8 results. Four were theoretical articles that use the ACF in their references, and four were in fact applications of the ACF. Some considerations about the empirical studies are presented below. These articles were published between 2014 and 2018 and analyze policies in Brazil (SOUZA; SECCHI, 2014; VICENTE; CALMON; ARAÚJO, 2017), Uruguay (ZORRILLA, 2016) and Chile (CORTEZ; MAILLET, 2018). In Brazil, both articles focus on state policies. Souza and Secchi (2014) use the ACF to analyze the role of the local scientific community in the formulation of Science and Technology policy in the State of Santa Catarina. Vicente, Calmon, and Araújo (2017) analyze the process of institutional change within the context of zoning policy in the Federal District, using the ACF to identify coalitions and their influence on the policy change process. The other two works analyze national policies in Uruguay and Chile. Zorrila (2016) examines the belief systems of coalitions to understand obstacles to the design and implementation of secondary school educational policy. Cortez and Maillet (2018) analyze how the socio-environmental conflict associated with the Pascua Lama mining project has motivated a review of the Chilean glacier protection policy, demonstrating that the coalitions have transformed over time.

Our findings demonstrate that the ACF has been little used in Latin America but is beginning to establish itself in this region. It should be observed that the review of Brazilian theses and dissertations performed by Capelari, et al. (2015) found 23 applications of the ACF, but it appears that these have not been converted into published articles. It may be affirmed, therefore, that Latin American researchers are beginning to become familiar with this model. The study of public policies in Latin America is certainly fertile ground for the application of the ACF. Conflicts of interest in terms of the reduction of social inequalities, the exploitation of natural resources, agricultural production, state interventions, economic blocs, and the war against drug trafficking, among others, are complex issues that involve various policy-making processes in these countries. The syncretism of liberal, statist, conservative, and progressive views in recently consolidated democracies in developing countries makes Latin America a privileged locus for the use of the ACF as an analytical model for policy processes. For now, just accompanying the results of the next few years will enable us to evaluate the advance of the use of the ACF in this part of the planet.

Conclusion

The ACF continues to be a model for analyzing public policies that is used mainly in Europe and the United States, but its expansion to other continents such as Asia, Africa and Oceania even on a small scale is reflected in English language publications. The use of the ACF in Latin America continues in to be little known in the international arena, given that a review of publications in Portuguese and Spanish has yet to be written. This article seeks to fill this gap, presenting the use of the ACF in four Latin American studies in recent years. To conclude, in this final section, we will briefly describe the ACF agenda, as listed in Table 1, summarizing how the articles contributed to its advancement. We then present a complementary research agenda as well as the limitations of this review.

Used mainly to analyze environmental and energy policies, the ACF has also been used to study less polarized sectors such as Health and Education. These works contribute to broadening our understanding of the ACF as an analytical model for policies involving beliefs that are not characterized by extreme positions (pro and contra). We did not identify, however, any articles that sought to reconsider the ACF belief system, integrating for example, the policy core and the secondary aspects beliefs, as proposed in the ACF’s agenda.

The identification of some type of policy change is widely explored in the articles and is mainly explained by external or internal events in the subsystem. Policy-oriented learning however continues to be an underutilized element of the model. The founders of the framework have called for the advance of theories and measures of learning on the personal level, within coalitions or between them. Analyses on the organizational level, without considering the subsystem, are also indicated by the 2014 agenda but remain to be explored. Some theoretical efforts have been made by Moyson, by analyzing the inconsistencies between policy beliefs, policy preferences and policy change. The advance of the policy learning agenda is somehow dependent on a deeper understating of the role of knowledge in the policy process.

In terms of arranging resources hierarchically, the studies applying ACF perceive the resources’ utility and effectiveness in various ways. More studies are going to be necessary to clarify this issue. If there is a hierarchy in the resources, it is still not possible to say. Pierce’s investigation has shown, for example, that unlike previous studies, public support can be more effective than financial support.

The use of the ACF in comparative studies can be observed in a small portion of the works analyzed, but this is an effort that can offer contributions to the model. The comparisons made examine the behavior of similar subsystems in various countries within the same continent, as well as the reaction of national subsystems to changes on a global level. This type of comparison has been possible for environmental policies with intercontinental impacts. Jenkins-Smith et al. (2014) reinforce the importance of comparative studies, indicating that this is a gap that needs to be filled.

Given the breadth of the model, the research design involves the choice of the elements to be used. The utilization of the whole model favors the development of more descriptive studies, but the search for causal relationships between its elements may be viable, establishing a focus on specific parts of the model or reducing its scope to the organizational level. The challenge of establishing common approaches and its application to specific contexts remains. Researchers should try to find points of theoretical-methodological convergence that can generate complementary studies. This may be fruitful for researchers interested in policies within the same area and can lead to the sharing of research designs and instruments and to the development of comparative studies among countries.

In general, the research agenda proposed in 2014 is still up to date, and presents challenges and possibilities for researchers interested in the use of the ACF which remain unexplored. As a result, we present a complementary research agenda for policy analysts interested in using the ACF as a central analytical model.

Focusing on policy dynamics: policy change and policy stability continue to be treated separately in most of the ACF applications. To focus on policy dynamics, considering stability and change as parts of the same policy development process, brings possibilities to explore how coalitions form and act to influence policies, in order to promote and to avoid changes.

Unlinking policy learning from policy change: some of the studies analyzed have shown that policy change may be not the best variable to evidence policy learning. As suggested by the ACF agenda, some theoretical and methodological efforts to find better forms of identifying and measuring policy learning are necessary. We recommend exploring how scientific knowledge is used by coalition members to avoid policy change.

Examining the role of discourse as a resource to address problems of coordination in and among coalitions: this question has been under discussion since Schlager’s (1995) critique that ACF needs to address the collective action problems faced by coalitions. An innovative way to explore this issue is to examine the role of discourse in solving coordination problems in and among coalitions (VIEIRA, 2019). This can also be a fruitful way to better understand problems of coalition formation and maintenance, addressing the need to refine the theory of coalition structures and coordination, established in the 2014 agenda.

Establishing causal inferences: ACF applications are still too descriptive. The framework, however, offers many theoretical explanatory factors of policy change and stability. Studies using ACF should better explore how these factors influence policy dynamics by establishing causal inferences. This can be done with more robust methods, including qualitative ones such as Process Tracing and QCA, to explain how causal mechanisms influence policy dynamics.

Studying the role of society in actively influencing policy processes. Social movements and protests have gained attention in many countries all over the world. Nonetheless, they are commonly treated in policy analyses as “public opinion”, a resource used by coalitions’ members to get support for their policy preferences. Studies in this domain could explore how societal actors involved in protests are themselves active in influencing policy process and promoting their policy preferences. In the context of the crisis of political representation, one way of better understanding the power of collectives is to observe them in a less passive perspective.

There are two main limitations to this review. First, it does not capture all applications of ACF published from 2015 to 2018, because of the database selected, the search criteria, and the filters applied (language, peer-reviewed articles, impact factor criteria). The Web of Science’s Cited Reference Search, as used by Pierce et al. (2017), is certainly an interesting alternative search option. In terms of language, we tried to overcome this limitation, by searching for articles in Portuguese and Spanish, but as we focused our search on the SciELO database, this search was also not exhaustive. Second, although the main elements of the model have been considered, this review does not analyze systematically how the applications deal with the ACF’s hypotheses. There was an attempt to include the hypothesis in the codification, but it was abandoned, because they could not be so easily identified in the articles, demanding a very subjective approach to explain the results obtained. Nonetheless, some of the hypotheses are indirectly addressed in the discussions.

We hope that this review, together with the research agenda proposed by the ACF’s formulators, can inspire policy analysts to use the framework in a critical and challenging way, in order to capture the complexity of the policy process investigated. Finally, we would like to invite Latin American researchers to consider the possibilities brought by ACF to investigate economic, social or environmental policies in the region.

References

AAMODT, S.; STENSDAL, I. Seizing policy windows: policy influence of climate advocacy coalitions in Brazil, China, and India, 2000-2015. Global Environmental Change-Human and Policy Dimensions, 46, p. 114-125, 2017. Disponível em: https://doi.org/10.1016/j.gloenvcha.2017.08.006. Acesso em: jul. 2020. [ Links ]

ANDERSON, W. F. A.; MACLEAN, D. A. Public forest policy development in New Brunswick, Canada: multiple streams approach, advocacy coalition framework, and the role of science. Ecology and Society, v. 20, n. 4, 2015. Disponível em: https://doi.org/10.5751/ES-07795-200420. Acesso em: jul. 2020. [ Links ]

BARNES, C.; VAN LAERHOVEN, F.; DRIESSEN, P. P. J. Advocating for change? How a civil society-led coalition influences the implementation of the forest rights act in India. World Development, 84, p. 162-175, 2016. Disponível em: https://doi.org/10.1016/j.worlddev.2016.03.013. Acesso em: jul. 2020. [ Links ]

BELSCHNER, T. Not so green after all? The EU’s role in international fisheries management: the cases of NAFO and ICCAT. Journal of European Public Policy, v. 22, n. 7, p. 985-1003, 2015. Disponível em: https://doi.org/10.1080/13501763.2014.989891. Acesso em: jul. 2020. [ Links ]

CAPANO, G. Understanding policy change as an epistemological and theoretical problem. Journal of Comparative Policy Analysis: Research and Practice, v. 11, n. 1, p. 7-31, 2009. [ Links ]

CAPELARI, M. G. M.; ARAÚJO, S. M. V. G.; CALMON, P. C. D. P. Advocacy coalition framework: um balanço das pesquisas nacionais. Administração Pública e Gestão Social, v. 7, n. 2, p. 91-99, 2015. Disponível em: https://doi.org/10.21118/apgs.v7i2.706. Acesso em: jul. 2020. [ Links ]

CARVALHO, G. O. Metallurgical development in the Carajas area: A case study of the evolution of environmental policy formation in Brazil. Society and Natural Resources, 14, p. 127-143, 2001. [ Links ]

CHIKOTO-SCHULTZ, G.; UZOCHUKWU, K. Governing civil society in Nigeria and Zimbabwe: a question of policy process and non-state actors’ involvement. Nonprofit Policy Forum, v. 7, n. 2, SI, p. 137-170, 2016. Disponível em: https://doi.org/10.1515/npf-2015-0051. Acesso em: jul. 2020. [ Links ]

CORTEZ, M.; MAILLET, A. Trayectoria multinivel de una coalición promotora e incidencia en la agenda política nacional. El caso del conflicto de Pascua Lama y la ley de glaciares en Chile. Colombia Internacional, 94, p. 3-25, 2018. [ Links ]

DUYGAN, M.; STAUFFACHER, M.; MEYLAN, G. Discourse coalitions in Swiss waste management: gridlock or winds of change? Waste Management, 72, p. 25-44, 2018. Disponível em: https://doi.org/10.1016/j.wasman.2017.11.006. Acesso em: jul. 2020. [ Links ]

ELGIN, D. J. Cooperative interactions among friends and foes operating within collaborative governance arrangements. Public Administration, v. 93, n. 3, p. 769-787, 2015. [ Links ]

FENGER, M.; QUAGLIA, L. The global financial crisis in comparative perspective: Have policy makers ‘Learnt their lessons’? Journal of Comparative Policy Analysis, v. 18, n. 5, SI, p. 502-517, 2016. Disponível em: https://doi.org/10.1080/13876988.2015.1044808. Acesso em: jul. 2020. [ Links ]

GEORGE, A. L.; BENNETT, A. Case studies and theory development in the social Sciences. Cambridge: MIT Press, 2004. [ Links ]

HARRINKARI, T., Katila, P.; KARPPINEN, H. International influences in the revision of Finnish Forest Act. Scandinavian Journal of Forest Research, v. 32, n. 1, p. 6-18, 2017. Disponível em: https://doi.org/10.1080/02827581.2016.1183702. Acesso em: jul. 2020. [ Links ]

HUGHES, C. E.; RITTER, A.; LANCASTER, K.; HOPPE, R. Understanding policy persistence: the case of police drug detection dog policy in NSW, Australia. International Journal of Drug Policy, 44, p. 58-68, 2017. Disponível em: https://doi.org/10.1016/j.drugpo.2017.03.007. Acesso em: jul. 2020. [ Links ]

HUGHES, L.; MEEKLING, J. The politics of renewable energy trade: The US-China solar dispute. Energy Policy, 105, p. 256-262, 2017. Disponível em: https://doi.org/10.1016/j.enpol.2017.02.044. Acesso em: jul. 2020. [ Links ]

JENKINS-SMITH, H. C.; NOHRSTEDT, D.; WEIBLE, C. M.; INGOLD, K. The advocacy coalition framework: an overview of the research program. In: WEIBLE, C. M.; SABATIER, P. A. (Eds.), Theories of the policy process. 4th ed. New York: Routledge, 2018. [ Links ]

JENKINS-SMITH, H. C.; NOHRSTEDT, D.; WEIBLE, C. M.; SABATIER, P. A. The advocacy coalition framework: foundations evolutions ongoing research. In: SABATIER, P. A.; WEIBLE, C. M. (Eds.), Theories of the policy process. 3rd ed. Boulder: Westview Press, 2014. [ Links ]

KERSHAW, P.; SWANSON, E.; STUCCHI, A. A surgical intervention for the body politic: generation squeeze applies the advocacy coalition framework to social determinants of health knowledge translation. Canadian Journal of Public Health-Revue. v. 108, n. 2, p. 199-204, 2017. Disponível em: https://doi.org/10.17269/CJPH.108.5881. Acesso em: jul. 2020. [ Links ]

KHAYATZADEH-MAHANI, A.; BRETON, E.; RUCKERT, A.; LABONTE, R. Banning shisha smoking in public places in Iran: an advocacy coalition framework perspective on policy process and change. Health Policy and Planning, v. 32, n. 6, p. 835-846, 2017. Disponível em: https://doi.org/10.1093/heapol/czx015. Acesso em: jul. 2020. [ Links ]

KUKKONEN, A.; YLA-ANTTILA, T.; BROADBENT, J. Advocacy coalitions, beliefs and climate change policy in the United States. Public Administration, v. 95, n. 3, p. 713-729, 2017. Disponível em: https://doi.org/10.1111/padm.12321. Acesso em: jul. 2020. [ Links ]

KUKKONEN, A. et al. International organizations, advocacy coalitions, and domestication of global norms: debates on climate change in Canada, the US, Brazil, and India. Environmental Science & Policy, 81, p. 54-62, 2018. Disponível em: https://doi.org/10.1016/j.envsci.2017.12.008. Acesso em: jul. 2020. [ Links ]

LEE, A. K. Heritage conservation and advocacy coalitions: the state-society conflict in the case of the Enning Road redevelopment project in Guangzhou. International Journal of Heritage Studies, v. 22, n. 9, p. 729-747, 2016. Disponível em: https://doi.org/10.1080/13527258.2016.1195427. Acesso em: jul. 2020. [ Links ]

LEIPPRAND, A.; FLACHSLAND, C.; PAHLE, M. Advocates or cartographers? Scientific advisors and the narratives of German energy transition. Energy Policy, 102, p. 222-236, 2017. Disponível em: https://doi.org/10.1016/j.enpol.2016.12.021. Acesso em: jul. 2020. [ Links ]

LEONG, C. Persistently biased: the devil shift in water privatization in Jakarta. Review of Policy Research, 32, n. 5, p. 600-621, 2015. Disponível em: https://doi.org/10.1111/ropr.12138. Acesso em: jul. 2020. [ Links ]

MCDONALD, S. L.; GALLAGHER, D. R. A story about people and porpoises: consensus-based decision making in the shadow of political action. Environmental Management, v. 56, n. 4, p. 814-821, 2015. Disponível em: https://doi.org/10.1007/s00267-015-0545-6. Acesso em: jul. 2020. [ Links ]

MENAHEM, G.; GILAD, S. Policy stalemate and policy change in Israel’s water sector 1970-2010: advocacy coalitions and policy narratives. Review of Policy Research, v. 33, n. 3, p. 316-337, 2016. Disponível em: https://doi.org/10.1111/ropr.12172. Acesso em: jul. 2020. [ Links ]

MOCKSHELL, J.; BIRNER, R. Donors and domestic policy makers: two worlds in agricultural policy-making? Food Policy, 55, p. 1-14, 2015. Disponível em: https://doi.org/10.1016/j.foodpol.2015.05.004. Acesso em: jul. 2020. [ Links ]

MOSLEY, J. E.; GIBSON, K. Strategic use of evidence in state-level policymaking: matching evidence type to legislative stage. Policy Sciences, v. 50. n. 4, p. 697-719, 2017. Disponível em: https://doi.org/10.1007/s11077-017-9289-x. Acesso em: jul. 2020. [ Links ]

MOYSON, S. Cognition and policy change: the consistency of policy learning in the advocacy coalition framework. Policy and Society, v. 36, n. 2, p. 320-344, 2017. Disponível em: https://doi.org/10.1080/14494035.2017.1322259. Acesso em: jul. 2020. [ Links ]

MOYSON, S. Policy learning over a decade or more and the role of interests therein: the European liberalization policy process of Belgian network industries. Public Policy and Administration, v. 33, n. 1, p. 88-117, 2018. Disponível em: https://doi.org/10.1177/0952076716681206. Acesso em: jul. 2020. [ Links ]

NEWELL, D. Implementing wind power policy - institutional frameworks and the beliefs of sovereigns. Land Use Policy, 72 (December 2017), 16-26, 2018). Disponível em: https://doi.org/10.1016/j.landusepol.2017.12.031. Acesso em: jul. 2020. [ Links ]

NICOLLE, S.; LEROY, M. Advocacy coalitions and protected areas creation process: case study in the Amazon. Journal of Environmental Management, v. 198, n. 1, p. 99-109, 2017. Disponível em: https://doi.org/10.1016/j.jenvman.2017.04.035. Acesso em: jul. 2020. [ Links ]

PIERCE, J. J. Advocacy coalition resources and strategies in Colorado hydraulic fracturing politics. Society & Natural Resources, v. 29, n. 10, p. 1154-1168, 2016. Disponível em: https://doi.org/10.1080/08941920.2015.1131876. Acesso em: jul. 2020. [ Links ]

PIERCE, J. J. et al. There and back again: a tale of the advocacy coalition framework. Policy Studies Journal, v. 45, n. 1, S13-S46, 2017. Disponível em: https://doi.org/10.1111/psj.12197. Acesso em: jul. 2020. [ Links ]

SABATIER, P. A. An advocacy coalition framework of policy change and the role of policy-oriented learning therein. Policy Sciences, 21, p. 129-168, 1988. [ Links ]

SABATIER, P. A.; JENKINS-SMITH, H. C. The advocacy coalition framework: an assessment. In: SABATIER, P. A (Ed). Theories of the Policy Process. Boulder: Westview Press, 1999. [ Links ]

SABATIER, P., JENKINS-SMITH, H. C. Policy change and learning: an advocacy coalition approach. Boulder, CO: Westview Press, 1993. [ Links ]

SABATIER, P. A.; WEIBLE, C. M. The advocacy coalition framework: innovations and clarifications. In: SABATIER, P. A (Ed). Theories of the policy process. 2nd.ed. Boulder: Westview Press, 2007. [ Links ]

SCHLAGER, E. Policy making and collective action: Defining coalitions within the advocacy coalition framework. Policy Sciences, 28, p. 243-270, 1995. [ Links ]

SHAW, R. D. The vulnerability of urban elementary school arts programs: a case study. Journal of Research In Music Education, v. 65, n. 4, p. 393-415, 2018. Disponível em: https://doi.org/10.1177/0022429417739855. Acesso em: jul. 2020. [ Links ]

SMITH, K. et al. Corporate coalitions and policy making in the European Union: how and why British American tobacco promoted better regulation. Journal of Health Politics Policy and Law, v. 40, n. 2, p. 325-372, 2015. Disponível em: https://doi.org/10.1215/03616878-2882231. Acesso em: jul. 2020. [ Links ]

SOUZA, L. R. de; SECCHI, L. A política científica e tecnológica de Santa Catarina: análise a partir do modelo de coalizões de defesa. Revista de Administração Pública, v. 48, n. 4, p. 939-960. 2014. Disponível em: https://doi.org/10.1590/0034-76121557. Acesso em: jul. 2020. [ Links ]

STEINMAN, L. E. et al. Examining the Washington State breastfeeding-friendly policy development process using the advocacy coalition framework. Maternal and Child Health Journal, 21(3), p. 659-669, 2017. Disponível em: https://doi.org/10.1007/s10995-016-2154-2 Acesso em: jul. 2020. [ Links ]

STRITCH, A. The advocacy coalition framework and nascent subsystems: trade union disclosure policy in Canada. Policy Studies Journal, v. 43, n. 4, p. 437-455, 2015. Disponível em: https://doi.org/10.1111/psj.12112 Acesso em: jul. 2020. [ Links ]

ULMANEN, J.; SWARTLING, A. G.; WALLGREN, O. Climate adaptation in Swedish Forestry: exploring the debate and policy process, 1990-2012. Forests, v. 6, n. 3, 708-733, 2015. Disponível em: https://doi.org/10.3390/f6030708 Acesso em: jul. 2020. [ Links ]

VICENTE, V. M. B.; CALMON, P. C. D. P.; ARAÚJO, S. M. V. G. de. Analisando mudanças institucionais na política de ordenamento territorial urbano do Distrito Federal à luz do modelo das coalizões de defesa. Organizações & Sociedade, v. 24, n. 80, p. 135-156, 2017. Disponível em: https://doi.org/10.1590/1984-9230807 Acesso em: jul. 2020. [ Links ]

VIEIRA, D. M. The discourse and coordination among advocacy coalitions: the case of Belo Monte. RAUSP Management Journal, v. in cite, 2019. [ Links ]

WEIBLE, C. M.; SABATIER, P. A.; MCQUEEN, K. Themes and variations: taking stock of the advocacy coalition framework. Policy Studies Journal, 37, p. 121-140, 2009. [ Links ]

WEISS, C. H. The many meanings of research utilization. Public Administration Review, v. 39, n. 5, p. 426-431, 1979. [ Links ]

WONG, N. W. M. Environmental protests and NIMBY activism: Local politics and waste management in Beijing and Guangzhou. China Information, v. 30, n. 2, SI, p. 143-164, 2016. Disponível em: https://doi.org/10.1177/0920203X16641550. Acesso em: jul. 2020. [ Links ]

ZORRILLA, J. P. Acuerdos y tensiones dentro de la izquierda política uruguaya en materia de educación media: un aporte desde el marco de las coaliciones promotoras. Revista Uruguaya de Ciencia Política, 25, p. 35-55, 2016. [ Links ]

Received: July 29, 2019; Accepted: June 11, 2020

Creative Commons License This is an open-access article distributed under the terms of the Creative Commons Attribution License