Skepticism and polarization: political determinants of non-participation in participatory institutions

Abstract

Participatory institutions have been created in contemporary democracies to encourage the participation of citizens in discussions and the formulation and oversight of public policies. However, studies have indicated difficulties related to the inclusion of certain social groups, even in established institutions. What explains non-participation in institutions that aim to broaden its scope? In addition to the lack of personal incentives and the costs associated with participation, we argue that two other political determinants contribute to this absence: skepticism and polarization. Our findings indicate that skepticism is associated with non-participation but is contingent on the context, while affective polarization is significantly related to non-participation in participatory institutions.

Keywords
skepticism; affective polarization; non-participation; participatory institutions; association networks

Resumo

Instituições participativas têm sido criadas em democracias contemporâneas para fomentar a participação das/os cidadã/os em discussões, formulação e controle de políticas públicas. Contudo, estudos apontam dificuldades na inclusão de certos grupos sociais, mesmo em instituições consolidadas. O que explica a não participação em instituições que têm como um dos objetivos a ampliação de seu escopo? Além da falta de incentivos pessoais e dos custos associados à participação, argumentamos que dois outros determinantes políticos contribuem para essa ausência: o ceticismo e a polarização. Nossos resultados indicam que o ceticismo está associado à não participação, mas é condicionado ao contexto, enquanto a polarização afetiva relaciona-se de forma significativa com a não participação em instituições participativas.

Palavras-chave
ceticismo; polarização afetiva; não participação; instituições participativas; redes associativas

Participation is essential for the survival of a democratic system, mainly because of its ability to educate and train “democrats” (Pateman, 1970; Putnam, Leonardi & Nanetti, 1993) and expose critical governance deficits (Fung, 2006). Contemporary democracies have thus endeavored to foster institutions that promote citizen participation, varying according to their scope (who participates), the way they communicate, their decision-making process and the extent of their authority (how their discussions are linked to politics or public action), adopting specific designs for specific objectives (ibid., 2006).

Among the dimensions linking participatory institutions to the strengthening of democracy are their ability to produce fairer public policies, especially in terms of redistributing goods, services, and income, and their potential to incorporate traditionally excluded sectors into political decision-making processes (Lüchmann & Borba, 2008). Scholars indicate that the demands of disadvantaged groups are generally not well conveyed and that individuals in these groups tend to be less active in qualified forms of participation (Verba, Schlozman & Brady, 1995). What explains this higher rate of non-participation by these social strata, even in participatory institutions that aim to widely promote their participation? The literature offers various explanations, whether rational, linked to the dilemmas of collective action (Olson, 1965; Dahl, 1970), or related to the social costs and incentives of participation (Pateman, 1970; Bobbio, Matteucci & Pasquino, 1992; Verba, Schlozman & Brady, 1995), or institutional, to name a few. Without disregarding these factors, it is argued that there are also strictly political and behavioral factors associated with non-participation: skepticism, characterized by distrust and low expectations of participatory democracy (Bennett et al., 2013; Garcia-Espin & Ganuza, 2017), and polarization, which widens the gap between the engagement of politically extreme individuals and the average voter, who becomes disillusioned by polarized discourse and does not see themselves represented by either pole (Fiorina, 2017; Fiorina, Abrams & Pope, 2004).

In order to assess the relationship between these factors and non-participation from a comparative perspective, the Brazilian case is particularly appropriate. Re--democratization in the country increased civil society’s influence on public policies and established several participatory institutions responsible for deliberating on these policies, such as participatory budgets, policy councils and city development plans (Abers, 2000; Avritzer, 2007, 2008, 2009; Smith, 2009). However, as in other contexts, there are hurdles in promoting the inclusion and renewal of certain social sectors in these participatory institutions (Fuks, Perissinotto & Souza, 2004; Ipea, 2012, 20, 2013; Isunza Vera & Gurza Lavalle, 2012; Lüchmann & Borba, 2008; Romão & Martelli, 2020) Santos Júnior, Ribeiro & Azevedo, 2004).

Moreover, the political and behavioral factors discussed in this article are part of the Brazilian political context. The country has a known history of skepticism towards the operation of democratic institutions and their representatives, mainly due to a widespread perception of corruption linked to democratic governance (Taylor & Buranelli, 2007). In addition, Brazil is facing a recent context of political polarization, both among elites (Zucco & Power, 2024) and in public opinion (Fuks & Marques, 2023), driven by the rise of a new right-wing movement (Couto, 2014; Borges & Vidigal, 2023).

This study uses data from A Cara da Democracia no Brasil, a national survey with a representative sample of the Brazilian population, carried out between 2018 and 2022 by the Instituto da Democracia e da Democratização da Comunicação (IDDC) to investigate the relationships between political skepticism, ideological and affective polarization, and the phenomenon of non--participation in participatory institutions. The survey includes questions about trust in political institutions, ideological positioning on cultural and economic issues, affection for political leaders and participation in different types of associations. The analysis employs multiple linear regression models to test the hypotheses that greater skepticism is associated with higher levels of non-participation, and that non-participation is more common among less ideologically and affectively polarized individuals.

Our results indicate that the relationship between skepticism and non-participation depends on the context: skepticism is linked to non-participation in environments with high levels of distrust in political institutions. As for polarization, we found no statistically significant association with ideological polarization, either on cultural or economic issues. In contrast, affective polarization showed a robust association with non-participation, suggesting that less affectively polarized individuals tend to be less present in participatory institutions than those who are more polarized. Furthermore, associative networks or “mobilizing structures”, already identified in the literature as explanatory factors for participation, have indeed proven to be fundamental in promoting adherence to these institutions (McAdam, McCarthy & Zald, 1996; Rennó, 2003). Individuals involved in associations or organizations become their main agents, as in the case of management councils, where participation takes place through these entities.

The study stands out by bringing the debate about democratic innovations – understood as permanent institutions designed to broaden and deepen citizen participation in political decision-making processes (Gurza Lavalle & Isunza Vera, 2010; Smith, 2009) – closer to studies on skepticism and polarization, with a focus on Brazil. Although civil society has increasingly taken part in public policies in the country, non-participation is still a challenge, partly due to skepticism, linked to the perception of widespread corruption, and growing polarization. The study seeks to integrate new explanatory factors to understand why certain groups often do not get involved in participatory processes, even in contexts in which broader participation is aimed at.

Beyond this introduction, this paper is structured in five main sections. The first addresses skepticism and polarization in the theoretical debates on non-participation, recognizing the importance of other factors. The second justifies the choice of the Brazilian case, highlighting the particularities of the context that make it relevant for the proposed evaluation. The third details the methodology used, including the definition of the operationalization of the variables and the analysis models used. The fourth presents the results, with a brief description of the dependent and independent variables, followed by the results of the hypothesis tests carried out. Finally, the concluding remarks summarize the main findings of the study and reflect on its theoretical and practical implications.

Factors of non-participation in participatory institutions

The concerns and needs of disadvantaged social sectors – taking into account class, race and gender dimensions and, as the literature points out, an intersectional perspective on these markers that recognizes multiple systems of domination (Biroli & Miguel, 2015; Rodrigues & Freitas, 2021) – rarely reach government officials effectively. One of the factors explaining this mismatch between government demands and actions is the lower participation of these vulnerable groups, especially the poorest and least educated, in forms of political engagement and participatory institutions. These institutions, in addition to requiring greater interest and access to information (Verba, Schlozman & Brady, 1995), are permanent and deliberative bodies responsible for formulating and overseeing public policies, such as the Management Councils for various fields. A point widely discussed in the literature is why, even in institutions designed to achieve greater inclusion, certain sectors remain underrepresented or even absent from these spaces.

A central body of studies into this issue indicates that the explanation for the lack of participation by these groups is directly related to the dilemma of collective action. Empirical evidence has repeatedly shown that only a minority of citizens demonstrates interest and engagement in politics (Milbrath, 1965). When analyzing the reasons for the low levels of participation in democracies, Dahl (1970) says that, in general, individuals are not very involved in politics when they attribute low value to the rewards of this involvement in comparison with other activities. Furthermore, participation becomes unlikely when individuals believe they have little chance of influencing outcomes and altering the distribution of rewards through their actions, and they are also less likely to participate if they consider that the overall outcome will be satisfactory regardless of their contribution. Thus, non-participation can be seen as a rational choice, especially in relation to the distribution of public goods (Lüchmann & Borba, 2008). Hence according to this theoretical approach participation depends on those involved obtaining private benefits; otherwise, “freeloader” logic prevails (Olson, 1965).

Other authors have followed an alternative explanatory path, developing a literature that seeks to explain non-participation on the basis of the social incentives involved. In this framework, the distinction between active and passive citizens can be associated not only to an individual economic calculation, but also to a lack of opportunities for participation and the costs that hinder the inclusion of large populations. The “centrality” model suggests that the intensity of participation varies according to the individual's social position; the more central their position in the social structure, the greater the likelihood of participation and sense of belonging (Lüchmann & Borba, 2008). Studies indicate that levels of political participation are higher, for example, among men, from the upper classes, with a higher level of education, living in urban centers, and coming from families where politics is valued, as well as among individuals linked to political organizations and exposed to politicized environments (Bobbio, Matteucci & Pasquino, 1992). These factors reinforce the argument that privileged access to resources and information is directly linked to presence and participation in more qualified participation mechanisms. An offshoot of the centrality model is the theory of “civic voluntarism” (Verba, Schlozman & Brady, 1995), which explains participation as the result of individual and collective resources. Individual resources include time and money. Collective resources refer to “civic skills”, which can be developed, for example, through civil associations, considered “schools of democracy” from a Tocquevillian perspective.

In this article, we propose that, in addition to economic and social explanations, two factors related to political and behavioral dimensions are crucial to understanding the absence of certain segments in the participatory sphere. The first of these is skepticism, which refers to a lack of trust in or low expectations of participatory democracy (Garcia-Espin and Ganuza, 2017). Voters seem to be increasingly skeptical of political actors and institutions (Bennett et al., 2013), which results in political disengagement from the traditional spheres of participation in institutions directly linked to the state (Jackson, Mondak & Huckfeldt, 2009; Keane, 2009; Offe, 2006). One of the groups most affected by this skepticism is young people. Dalton (2008) argues that they adhere to a conception of citizenship that values community over politics. This group is distrustful of government, preferring to participate in community life rather than campaigns, elections, and political--governmental institutions. They sign online petitions, boycott corporations, and volunteer to help the disadvantaged or those affected by natural disasters (Martin, 2012), but show less interest in actors and participatory institutions linked to the state. Several factors may underpin the increase in skepticism towards these participatory institutions, such as political distrust – both vertical (in governments and parties) and horizontal, especially in relation to citizens' abilities. A (deficient) perception of political efficacy also plays a role, covering internal efficacy (how the individual perceives their own competence as a political actor) and external efficacy (whether the individual considers the political system to be responsive to popular demands) (Garcia-Espin and Ganuza, 2017). Thus, our first hypothesis is that more skeptical individuals, i.e. those with greater political distrust and low perception of efficacy, tend to participate less in participatory institutions linked to the state (H1).

The second explanatory factor we suggest is polarization. Researchers of American politics indicate that contemporary polarization has made the public more engaged, increasing political participation (Abramowitz, 2006, 2010; Abramowitz & Saunders, 2008). More recent studies suggest that activity and involvement levels in recent years are within the range established in previous decades (Fiorina, 2017), indicating that polarization, or partisan sorting (see Levendusky, 2009), have not resulted in a generalized increase in engagement, especially in activities that require time, such as working for a party or candidate and attending meetings or rallies (Fiorina & Levendusky, 2006). This argument is reinforced by the fact that this stability is even more surprising in a political landscape intensified by greater elite engagement and the advent of new information technologies (Fiorina, 2017; Fiorina, Abrams & Pope, 2008). The more balanced median voter seems tired of the polarized “us versus them” debate and does not feel represented by traditional institutions and either side of the polarization (Fiorina, 2017). Thus, considering that participatory institutions, such as management councils and participatory budgets, are linked to the government, our hypothesis is that polarization widens the gap in levels of engagement and participation between politically extreme individuals and the average voter. Therefore, we expect to find less involvement in participatory institutions linked to the state among those who are less polarized, ideologically and affectively (H2).

In the next section we highlight the elements that make Brazil a particularly suitable case for assessing the relationship between these political factors and non-participation from a comparative perspective.

The Brazilian case

The Brazilian democratic context, which began with a period of transition from the military regime and was consolidated by the 1988 Constitution, is marked by the broad participation of civil society in public policies and the growth of participatory institutions. In various articles, the Constitution requires social participation in specific policies, opening up space for demands for power-sharing in various areas (Ipea, 2013). This has resulted in the consolidation of hybrid formats in certain public institutions, with the inclusion of representatives of civil society and the state (see Abers & Keck, 2006). These institutions aim to include, individually or collectively, directly or by representatives, different segments and interests of the population in public administration, in a manner that is regular, ongoing, and guaranteed by government regulations, which gives these spaces an institutional character (Cortes, 2011; Lopez & Pires, 2010; Lüchmann, 2020). Three main formats of participatory institutions stand out in the country. First, the emergence of participatory budgeting in Porto Alegre (capital of the state of Rio Grande do Sul) in 1989, which attracted the attention of actors from all over the world (Abers, 2000; Avritzer, 2002; Baiocchi, 2005; Smith, 2009; Sousa Santos, 1998). Secondly, policy councils, public spaces linked to executive branch bodies with the aim of promoting society's participation in setting priorities for the political agenda, as well as in formulating, monitoring and controlling public policies (Ipea, 2013). Finally, there are public policy conferences, participatory meetings that bring together social and government sectors to discuss and present proposals and guidelines in various areas. These events are regulated by laws and regulations that define their composition, periodicity and powers, and take place in a staggered manner: they begin at the municipal level and, through delegated representation, progress to the state and national stages.

Despite this innovative and fruitful landscape of participatory initiatives, Brazil faces difficulties in including and renewing certain social sectors in these institutions. Studies show that the inclusion of groups such as the poorest and least educated, as well as other age, gender and/or racial segments, depending on the area of public policy, faces important challenges (Fuks, Perissinotto & Souza, 2004; Ipea, 2012, 20, 2013; Isunza Vera & Gurza Lavalle, 2012; Lüchmann & Borba, 2008; Romão & Martelli, 2020; Santos Júnior, Ribeiro & Azevedo, 2004).

Additionally, the Brazilian scenario is conducive to assessing the impact of political and behavioral factors on non-participation, especially skepticism and polarization. Brazilians are notoriously skeptical of their institutions and political leadership, largely due to the high perception of corruption. In Transparency International's 2023 Corruption Perceptions Index (CPI) in 2024, Brazil scored 36 points on an index that goes from 0 (highly corrupt public sector) to 100 (highly honest public sector). The average for the Americas is 43 points, meaning Brazil has a higher perception of corruption than the regional average. Furthermore, Brazil's score is well below the average between full democracies (73 points) and deficit democracies (48 points), being closer to non-democratic regimes (32 points). One of the main explanatory factors for this perception is the lack of accountability in investigative processes. Taylor and Buranelli (2007) show that, in Brazil, institutions encourage investigation and publicity, but offer little in the way of sanctions. When there are no visible consequences for the accused, especially in widely publicized cases, trust in governance deteriorates. Constant findings of corruption, combined with perceptions of impunity, undermine trust in the political process and, as we have argued, are expected to reduce individuals' interest in political participation.

Ideological polarization is defined as “[...] a growing ideological distance between the main political groups in society, with the concomitant emptying of the center [...] and affective polarization is defined as the increase in disaffection between rival political groups” (Fuks & Marques, 2023, p. 561-562). Brazil is also experiencing a recent context of political polarization. Previous studies pointed to a largely non-polarized electorate, with voters from the main parties which had contested the presidency for six consecutive elections, PT and PSDB, having very similar ideological positions (Borges & Vidigal, 2018). This picture changed dramatically with the emergence of a new right-wing movement (Couto, 2014; Borges & Vidigal, 2023). The election of Jair Bolsonaro in 2018 marked this significant change, laying the ideological foundations for an extreme right that had until then lacked representation (Borges & Vidigal, 2023; Fuks & Marques, 2021; Ortellado, Ribeiro & Zeine, 2022; Rennó, 2020, 2022), 2022). With this movement, a new trend emerged in the country: the elites began to position themselves more clearly (Borba et al., 2024), accompanied by the weakening of the political center, indicating an advance in polarization (Zucco & Power, 2024). Public opinion has also changed, with an increase in affective polarization, mainly centered on popular leaders (Fuks & Marques, 2023).

The history of powerful and internationally recognized participatory institutions, combined with a scenario of skepticism towards political institutions and a new context of polarization among the elites and the electorate, makes contemporary Brazil an opportune case to assess how these political and behavioral factors can be directly associated with a lower rate of participation in these institutions.

Method

To test the hypotheses that (H1) the more politically skeptical individuals are the less they tend to participate in participatory institutions, and that (H2) non-participation is higher among the less ideologically and affectively polarized, we used data from a recent survey that included questions about participation in participatory institutions, trust in political institutions (to measure skepticism), positioning on political issues (to assess ideological polarization), and affection for political parties and/or leaders (to measure affective polarization). The timeliness of the research is crucial, given that only in recent years has polarization in Brazil become more evident.

One survey that meets these criteria is “The Face of Democracy in Brazil”, a national survey with a representative sample of the Brazilian population,¹ conducted by the IDDC between 2018 and 2022.² Although not all five waves of the survey include all the necessary questions, we found relevant items in three of them (2018, 2019 and 2022). These waves enable the measurement of participation in institutions such as participatory budgeting, councils, and policy conferences, as well as assessing individual trust in institutions such as the judiciary, congress, the presidency, political parties, and the Supremo Tribunal Federal [Federal Supreme Court] (STF), which allows us to assess the level of skepticism towards political institutions. It is also possible to analyze ideological polarization on cultural issues, such as the age of criminal responsibility, same-sex civil marriage and adoption by homosexual couples. The 2018 and 2019 editions include additional questions about ideological polarization on economic issues, while the 2022 edition assesses support for politicians, such as Lula, Bolsonaro, Sérgio Moro, and Ciro Gomes, allowing us to measure affective polarization towards candidates.

Furthermore, the survey offers relevant control variables, such as residential area (rural or urban), gender, age, schooling, income bracket, interest in politics, and participation in community, recreational, or sports associations, NGOs, charitable organizations, political parties, social movements, trade unions and/or churches or religious organizations. Further details on the descriptive analysis of the variables and recoding are available in the Methodological Appendix.

Operationalization of variables

To operationalize the dependent variable, we measured participation in a binary (dummies), where (1) indicates that the individual participates in the institution, and (0) indicates they do not. The P participation variable is defined as the sum of the dummies generated from the variables of participation in participatory budgeting (X1′), a policy council (X2′), or a policy conference (X3′).

$P=X_1^{\prime }+X_2^{\prime }+X_3^{\prime }$

Then the P variable is turned into a binary indicator of participation:

$P^{\prime }=\{\begin{array}{cc}1& \text{se}P\ge 1\\ 0& \text{se}P=0\end{array}$

Here P’ indicates participation in at least one activity, with P’=1 representing participation and P’=0 the absence of participation. Finally, as our research problem concerns the determinants of non-participation, we inverted the tabulation of the data for the dependent variable P′, assigning a value of 1 to those who do not participate in any participatory institution and a value of 0 to those who do.

To test our hypotheses, we developed four indicators, which are our independent variables: vertical political skepticism, ideological polarization (on cultural and economic issues), and affective polarization towards candidates.

To construct the skepticism variable, we used the vertical political trust axis as a proxy. Unfortunately, the survey does not include questions about perceived efficiency, either internal or external, which is a limitation of this work. Variables x4, x5, x6, x7, e x8 were selected, representing, respectively, trust in the judiciary, political parties, congress, the president of the republic and the STF. These variables range from 1 (no trust) to 4 (high trust). A Factor Analysis (FA) was carried out, with n=1 as the extracted common factor and Varimax rotation (FA results are in the Methodological Appendix). This can be represented as:

$X=\Lambda F+U$

where X is the matrix of the observed variables (x4, x5, x6, x7, x8), Λ is the factor loading matrix, F is the extracted common factor, and U is the matrix for idiosyncratic error terms. After that, factor scores were calculated for each observation in the sample, resulting in a new variable MR1, which can be represented as:

$MR{1}_j=\sum _{i=4}^8{\lambda }_i{x}_i+u_j$

where λi are the factor loadings for the xi variables in the MR1 factor, and uj are the idiosyncratic error terms for each j observation. Since the skepticism variable refers to distrust in institutions, the MR1 variable was inverted by being multiplied by -1.

To measure the level of individual ideological polarization, we created an index based on coherence regarding cultural and economic issues. We defined a priori which positions are considered more liberal and which more conservative on each issue. Individuals who take a more coherent stance, whether liberal or conservative, are considered more polarized.

In relation to cultural issues, each topic T_i (where i = 1,2,...,10) is associated with a dummy variable that represents the respondent's position on a particular topic debated in Brazilian society. The variables are defined as:

• T1: Reducing the age of criminal responsibility (agreeing indicates a conservative position);

• T2: Same-sex civil marriage (agreeing indicates a liberal position);

• T3: Child adoption by gay couples (agreeing indicates a liberal position);

• T4: The death penalty (agreeing indicates a conservative position);

• T5: Decriminalization of drug use (agreeing indicates a liberal position);

• T6: Banning firearm sales (agreeing indicates a liberal position);

• T7: Legalizing abortion (agreeing indicates a liberal position);

• T8: Imprisoning women who terminate pregnancies (agreeing indicates conservative position);

• T9: Racial affirmative action (agreeing indicates a liberal position);

• T10: Public schools should teach children to pray and believe in God (agreeing indicates a conservative position).

Responses for each topic were organized according to the following rule, considering that Ti=1 indicates having taken a liberal position, Ti = 2 represents a conservative position, Ti = 3 or Ti = 8 indicate neutrality or lack of opinion, and Ti = 9 is considered undefined or not applicable:

$T{i}^{\prime }=\{\begin{array}{cc}1,& \text{if}Ti=1\\ -1,& \text{if}Ti=2\\ 0,& \text{if}Ti\in \{3,8\}\\ NA,& \text{if}Ti=9\end{array}$

The new variable T that summarizes the positions in relation to all the themes is defined as the sum of the transformed variables:

$T=T_1^{\prime }+T_2^{\prime }+T_3^{\prime }+T_4^{\prime }+T_5^{\prime }+T_6^{\prime }+T_7^{\prime }+T_8^{\prime }+T_9^{\prime }+T_{10}^{\prime }$

The T variable creates an index that varies from -10 if the individual adopts conservative positions on all issues, to 10 if they adopt only liberal positions. To compare the extremists with the balanced, regardless of whether they are conservative or liberal, the absolute value of the sum of the thematic variables is calculated, resulting in the final variable 1T1. This variable represents the magnitude of ideological polarization in relation to the issues debated in Brazilian society, where higher values indicate a greater level of individual ideological polarization.

To build a similar index on ideological polarization regarding economic issues, the same methodological strategy was used as for cultural issues, but with the following questions, which, instead of being classified a priori as liberal or conservative, were divided between state-oriented and private-sector-oriented positions. Greater agreement with any of the questions indicates a more statist position, since all the points are introduced by the statement “The state, rather than the private sector, should be primarily responsible for”:

• T11: the country's most important companies and industries;

• T12: the well-being of citizens;

• T13: reducing income inequality between rich and poor;

• T14: guaranteeing pensions;

• T15: providing health services;

• T16: providing education;

• T17: reducing inequalities between men and women;

• T18: reducing inequalities between cultural and ethnic groups.

To measure the last independent variable, individual affective polarization, we constructed a dummy variable that indicates whether or not the individual is affectively polarized. In two-party contexts, measuring affective polarization usually entails calculating the “feeling thermometer”, which represents the difference between the evaluation of the preferred party/candidate and that of the rival party/candidate (Iyengar et al., 2019). However, in multiparty contexts such as Brazil, the meaning of affective polarization must be reconsidered. One of the main aspects of a polarized individual is the perception of an “us vs. them” scenario, encompassing in the so-called “them” all those who do not represent their group or leader (Mason, 2018). Thus, we consider individuals polarized (1) if they like a single candidate and at the same time dislike all the others, while (0) represents the opposite. This approach reflects the idea that polarized individuals overvalue the leaders of their own group (in-group) and develop prejudices towards the leaders of other groups (out-groups) (Huddy, Mason & Aarøe, 2015).

To assess whether the individual is affectively polarized, we consider g1 to be Lula's evaluation score, ranging from (1) “doesn't like at all” to (10) “likes a lot”; g2 to be Bolsonaro's evaluation score; g3 to be Sérgio Moro's score; and g4 to be Ciro Gomes' score. The affective polarization variable is defined as follows:

$\text{afetiva}=\{\begin{array}{cc}1,& \text{if}\left((g_1\ge 8\wedge g_2\le 3\wedge g_3\le 3\wedge g_4\le 3)\right)\vee \\ & (g_2\ge 8\wedge g_1\le 3\wedge g_3\le 3\wedge g_4\le 3)\vee \\ & (g_3\ge 8\wedge g_1\le 3\wedge g_2\le 3\wedge g_4\le 3)\vee \\ & (g_4\ge 8\wedge g_1\le 3\wedge g_2\le 3\wedge g_3\le 3)\\ 0,& \text{otherwise}\end{array}$

The “affective” dummy variable therefore indicates the respondent's affective polarization towards politicians. It takes on the value 1 if a single politician is well evaluated (score ≥ 8) while the others are poorly evaluated (score ≤ 3); otherwise, it takes on the value 0.

Finally, we selected control variables that are well established in the literature as being associated with non-participation. First, we highlight the variables that indicate “centrality”, referring to an individual's social position, where greater centrality in the social structure is related to an increase in participation and sense of aggregation. In particular, we used the variables identified by Bobbio, Matteucci & Pasquino (1992) as predictors of greater participation: residential area (urban or rural), gender, age, schooling, income bracket, and interest in politics. We also included as a control variable a latent variable related to membership of networks, especially associative networks, since the literature points to consistent differences in civic behavior between individuals who participate in associations and those who do not, with a significant advantage for the former (Almond & Verba, 1963; Putnam, Leonardi & Nanetti, 1993; Verba & Nie, 1972). Associations are considered “mobilizing structures” in the literature on social movements (McAdam, McCarthy & Zald, 1996), being fundamental to fostering participation in participatory institutions. Thus the literature emphasizes that the social interaction networks that individuals develop throughout their lives shape identities that can increase the propensity for political participation (Lüchmann & Borba, 2008). To operationalize this variable, we constructed a latent variable based on the scores of a factor analysis, using the variables of participation in participatory associations, especially participation in community associations, recreational associations, sports associations, NGOs, charitable organizations, political parties, social movements, trade unions and/or churches or religious organizations. The operationalization of this and other control variables can be found in the Methodological Appendix.

Analytical models

To test H1 and H2, we sought to verify the relationship between participation in participatory institutions (P′) and, respectively, skepticism (MR1) and polarization, both ideological (1T1) and affective (afetiva). To this end, we have developed two³ analytical models, which we will describe in this section.⁴

In Model 1, which we refer to here as the simplified model, we used a multiple linear regression to assess the relationship between participation in participatory institutions (part) and the variables of skepticism (MR1), ideological polarization on cultural issues (temas), ideological polarization on economics (papelest), and affective polarization (afetiva). The model is defined as follows:

$\text{part}={\beta }_0+{\beta }_1\cdot MR1+{\beta }_2\cdot \text{temas}+{\beta }_3\cdot \text{papelest}+{\beta }_4\cdot \text{afetiva}+ϵ$

This and all the other models were adjusted using the Ordinary Least Squares (OLS) method (Wooldridge, 2006). The result of the estimation provides the values of the β coefficients, which indicate the relationship between each independent variable and participation, while the error term ϵ represents the unexplained variations.

In Model 2, the complete model, we include in the multiple linear regression the control variables: residential zone (zona), gender (sexo), age (idade), schooling (esc), income bracket (fx_renda), interest in politics (interesse) and the associative networks latent variable (MR1p). The aim is to verify the impact of these variables, already widely discussed in the literature, on the observed estimators of our independent variables:

$\begin{array}{cc}\text{part}=& {\beta }_0+{\beta }_1\cdot \mathrm{MR}1+{\beta }_2\cdot \text{temas}+{\beta }_3\cdot \text{papelest}\\ & +{\beta }_4\cdot \text{afetiva}+{\beta }_5\cdot \text{zona}+{\beta }_6\cdot \text{sexo}\\ & +{\beta }_7\cdot \text{idade}+{\beta }_8\cdot \text{esc}+{\beta }_9\cdot \mathrm{fx}\text{renda}\\ & +{\beta }_{10}\cdot \text{interesse}+{\beta }_{11}\cdot \mathrm{MR}1p+ϵ\end{array}$

After defining the two models to be tested, the results section below presents a brief description of the dependent and independent variables and then the results of the hypothesis tests carried out.

Results

Our dependent variable is non-participation in participatory institutions. Table 1 shows the distribution of frequencies and percentages of participation and non-participation in the three institutions investigated by the survey “The Face of Democracy in Brazil”: participatory budgeting, policy councils, and policy conferences. We also show the aggregate values of participation and non-participation. We observed a small but significant increase in the proportion of non-participation from 2018 to 2019, from 96,13% to 97,84% (analyses of predicted values can be found in the Methodological Appendix). However, in 2022, there was a significant drop in non-participation, which fell from 97,84% in 2019 to 91,93%. The reduction in participation in participatory budgeting programs reflects the extinction of more than 80% of these programs, especially since 2016, more markedly in the following years (Lüchmann & Bogo, 2022). Also noteworthy is the data on participation in councils and conferences, which reached rates of over 6% and 5% in 2022. It should be noted that in 2019 the Bolsonaro government issued a decree abolishing councils (Decree n. 9.759 of April 11th, 2019), which may have resulted in decreased participation in these spaces, although this deconstruction has not directly affected the operation of these institutions on the sub-national levels (Bezerra et al., 2024).

Table 1
– Distribution of non-participation in participatory institutions

	ACD 2018		ACD 2019		ACD 2022
	Does not participate	Participates	Does not participate	Participates	Does not participate	Participates
Participatory budget	2.391 (96,68%)	82 (3,32%)	1.973 (98,70%)	26 (1,30%)	–	–
Policy council	2.455 (99,27%)	18 (0,73%)	1.991 (99,40%)	12 (0,60%)	2.366 (93,93%)	153 (6,07%)
Policy conference	–	–	1.991 (99,35%)	13 (0,65%)	2.383 (94,60%)	136 (5,40%)
Participatory institutions	2.362 (96,13%)	95 (3,87%)	1.949 (97,84%)	43 (2,16%)	2.300 (91,93%)	202 (8,07%)

To test our hypotheses, we developed four indicators: vertical political skepticism, ideological polarization (on cultural and economic issues) and affective polarization towards candidates. When analyzing the latent variables representing skepticism, we observed a significant drop in the level of skepticism in Brazil, both from 2018 to 2019 and especially from 2019 to 2022. Trust in political institutions, which makes up this latent variable, suggests a possible explanation for this reduction. Trust in the president of the republic increased significantly from 2018 to 2019 and remained high in 2022. This is plausible considering the fact that, in 2018, the incumbent president, Michel Temer, was widely unpopular and rejected by political groups across the ideological spectrum. With the election of Jair Bolsonaro, the extremely popular leader of the new right-wing movement in Brazil, confidence in the presidency was strongly impacted. In addition, between 2018 and 2022 there was an increase in trust in political parties and the national congress, possibly linked to the rise of the right in the legislature as well, which probably influenced the general level of trust in institutions which were previously the target of much skepticism (detailed analyses are available in the Methodological Appendix).

Regarding ideological polarization, we haven't seen any substantial differences over the years, either on cultural or economic issues. However, when we look at overall polarization, we notice distinctions between these areas. For cultural issues, on a scale of 0 (no polarization) to 10 (extremely polarized), the average was 3,34 points in 2018, 3,55 in 2019 and 3,47 in 2022, indicating that, in general, Brazilians are not highly ideologically polarized regarding cultural issues. This is to be expected, since Brazilian voters, regardless of their position on the political spectrum, tend to be more conservative on these issues (Borba, Silva & Amorim, 2024). At the extremes, individuals who scored 9 or 10 on the cultural ideological polarization scale are only 1,36% of the sample in 2018, 3,09% in 2019 and 3,07% in 2022.

For economic issues that discuss the role of the state, we only have data from 2018 and 2019. The average ideological polarization on economic issues then was higher than on cultural issues, with 5,24 points in 2018 and 5,33 points in 2019. At the extremes, individuals who scored 9 or 10 on the economic ideological polarization scale represent 13,24% in 2018 and 14,68% in 2019. This shows that although cultural issues dominate the public debate on polarization, Brazilians seem to be much more ideologically polarized on economic issues, especially with regard to disagreements on the role of the state (detailed analyses are available in the Methodological Appendix).

Finally, our last independent variable was affective polarization. Studies indicate that polarization in Brazil manifests itself more in affective terms in relation to candidates than in ideological terms (Fuks & Marques, 2023). Our work features an innovative approach to measuring affective polarization in multiparty contexts, obtaining results that corroborate this literature. When measuring the percentage of individuals in 2022 who were affectively polarized, i.e. those who evaluated only one politician positively (score ≥ 8) and all others negatively (score ≤ 3), we found that 23,52% fit this profile, a significant figure considering the strictness of the measure. This means that almost 1 in 4 Brazilians is very fond of one of the four politicians presented (Lula, Bolsonaro, Moro, or Ciro) while intensely disliking all the others.

After describing our dependent and independent variables, we evaluated the tests carried out to verify our hypotheses. The first hypothesis (H1) suggests that individuals with greater political skepticism tend to have a higher level of non-participation in participatory institutions. Our simplified model (Model 1) confirms this hypothesis in all waves. However, when we added the control variables (Model 2), we found that only in 2018 was this variable significant (Table 2). The data, interpreted in light of the significant drop in skepticism among Brazilians over the waves observed (for a more in-depth analysis of this variable, see the Methodological Appendix), suggest that the relationship between skepticism and non-participation depends on the context to be verified, i.e. it is not consistent over time regardless of context. Thus, skepticism is related to non-participation in contexts with high levels of distrust towards political institutions.

Table 2
– Determinants of non-participation in participatory institutions by year

	ACD 2018		ACD 2019		ACD 2022
	Model 1	Model 2	Model 1	Model 2	Model 1	Model 2
Independent variables
Skepticism	0,018*** (0,005)	0,014** (0,005)	0,009* (0,004)	0,002 (0,004)	0,025*** (0,007)	0,008 (0,006)
Ideological polarizationa (Cultural issues)	-0,004* (0,002)	-0,001 (0,002)	-0,001 (0,001)	-0,000 (0,001)	-0,005* (0,002)	0,001 (0,002)
Ideological polarization (Economic issues)	-0,001* (0,000)	-0,000 (0,000)	-0,000 (0,000)	-0,000 (0,000)	–	–
Affective polarization (Candidates)	–	–	–	–	-0,034* (0,013)	-0,028* (0,012)
Controls
Residential area	–	-0,011 (0,011)	–	-0,021* (0,010)	–	-0,010 (0,015)
Gender	–	0,007 (0,008)	–	-0,009 (0,007)	–	-0,017 (0,011)
Age	–	-0,000 (0,000)	–	-0,000 (0,000)	–	-0,001* (0,000)
Schooling	–	-0,003 (0,002)	–	0,000 (0,002)	–	-0,004 (0,003)
Income bracket	–	-0,005 (0,003)	–	-0,005* (0,003)	–	-0,003 (0,005)
Interest in politics	–	0,007 (0,005)	–	0,006 (0,004)	–	–
Participation networks	–	-0,080*** (0,005)	–	-0,046*** (0,004)	–	-0,144*** (0,006)

The second hypothesis (H2) suggests that non-participation is higher among individuals who are less polarized, both ideologically and affectively. In general, ideological polarization in relation to cultural and economic issues was not statistically significant, except in some specific results in the simplified model (Table 2). However, there is a robust relationship between non-participation and affective polarization, more clearly than with the ideological polarization found in the Brazilian context. Thus, those who are not affectively polarized tend to move further away from participatory institutions, confirming expectations in the literature (Fiorina, Abrams & Pope, 2004). The results are consistent, being significant in both the simplified model and the full model, even after the insertion of control variables, which affected the estimators of practically all the independent variables but did not significantly impact the affective polarization estimator. Furthermore, considering the low number of people who participate in this type of institution in the sample, statistically significant effects in multiple regression models with this type of dependent variable, with control variables included, are remarkable results.

Also noteworthy regarding control variables is the consistent inverse proportional relationship between non-participation and membership of networks (Table 2). This variable proved to be the most robust across the waves, showing that, regardless of the context, “mobilizing structures” are indeed fundamental to promoting participation in participatory institutions. Individuals who get involved in other associations or organizations become the main agents of these institutions.

Finally, it should be noted that the control variables related to the sociodemographic attributes of voters had practically no impact on non-participation in participatory institutions: only income and residential area in 2019 and age in 2022. These data hint at these institutions’ capacity for political inclusion: although they mobilize a very small percentage of the Brazilian electorate (a maximum of 8,07% in 2022), they do so equitably, without significant differences in relation to social markers such as income and schooling. This contrasts with the predictors of participation in conventional and unconventional modalities, such as political parties and protests, for which resources, especially schooling, are crucial (Ribeiro & Borba, 2015).

Graph 1 summarizes the results of the hypothesis tests carried out with the complete model (Model 2). We observed that skepticism was a significant variable in 2018, but not in the other waves, indicating that the relationship between this variable and non-participation may be conditioned by the context. Moreover, ideological polarization did not prove to be significant, either on cultural or economic issues. Affective polarization, however, with variables available only in the 2022 wave, showed a robust and significant association with participation, indicating that non-polarized individuals tend to participate less in participatory institutions than those who are more polarized.

Graph 1
– Determinants of non-participation in participatory institutions by year

Final remarks

This article investigated the relationship between the phenomenon of non-participation and its political and behavioral factors, with a focus on political skepticism and polarization. We used the Brazilian case, which, despite having significantly expanded the participation of civil society in public policies in its current democratic context, faces challenges in the inclusion of certain social sectors, has a history of skepticism linked to a perception of widespread corruption among political actors and institutions, and is experiencing a more recent context of polarization. Thus the study sought to integrate new explanatory factors that help understand why certain groups often do not get involved in participatory processes, even in contexts in which broadening the scope of participatory institutions is aimed at. Using data from the survey “The Face of Democracy in Brazil” between 2018 and 2022, we applied multiple linear regression models to test our hypotheses. The results indicated that skepticism is associated with non-participation in contexts of high distrust in institutions. Ideological polarization showed no significant relationship with non-participation, while affective polarization showed a robust association, suggesting that less affectively polarized individuals are less likely to engage in participatory institutions than more polarized ones. This finding could be an interesting analytical key to understanding the increased number of participants in councils and conferences in 2022.

A few methodological limitations are worth highlighting and should be taken into account when evaluating the results. First, we used vertical distrust of political institutions as a proxy for skepticism. Future studies should evaluate the effect of skepticism considering other dimensions of the phenomenon, such as horizontal distrust, which refers to the evaluation of citizens' abilities. Additionally, skepticism linked to the perception of political efficacy, covering both internal efficacy (how the individual perceives their own competence as a political actor) and external efficacy (whether the individual considers the political system to be responsive to popular demands), should be included in the analysis. As we have identified a relationship between context and the impact of skepticism, future studies considering different hierarchical levels ought to be carried out. Regarding affective polarization, although the results were robust, they refer only to 2022, due to the absence of candidate affection variables in the other waves. Assessing whether this relationship is maintained in other contexts is important in order to verify whether it is directly linked to the context or if it relates to individual behavior. Finally, limitations both in terms of the period analyzed and the lack of data on trust and the degree of skepticism in relation to participatory institutions should be noted, seeing as this would allow for a more robust analysis of their relevance to understanding institutional non-participation.

In any case, our findings encourage the development of new studies on non-participation in participatory institutions, analyzing longer periods of time, including different political contexts, and qualifying and exploring in greater depth elements such as the differences in participatory proposals and designs, the types of associations involved (trade unions, NGOs, etc.), relations with non-institutional participation, such as protests, and the predictors indicated in the literature, such as gender, race, age, income, and schooling, and their relations with the different types and areas of participatory institutions.

References

• ABERS, R. (2000). Inventing Local Democracy: Grassroots Politics in Brazil Boulder, Colorado, Lynne Rienner Publishers.
• ABERS, R. N.; KECK, M. E. (2006). Muddy waters: decentralization, coordination and power struggle in the Brazilian water management reform. International Journal for Urban and Regional Research, v. 30, n. 3, pp. 601-622. DOI: https://doi.org/10.1111/j.1468-2427.2006.00691.x
  » https://doi.org/10.1111/j.1468-2427.2006.00691.x
• ABRAMOWITZ, A. I. (2006). "Disconnected, or joined at the hip?". In: NIVOLA, P. S.; BRADY, D. W. (org.). Red and Blue Nation? Volume I - Characteristics and Causes of America's Polarized Politics Washington, DC/Stanford, CA, Brookings Inst. Press and Hoover Inst. Press.
• ABRAMOWITZ, A. I. (2010). The disappearing center: engaged citizens, polarization, and american democracy New Haven, Connecticut, Yale University Press.
• ABRAMOWITZ, A. I.; SAUNDERS, K. L. (2008). Is polarization a Myth? The Journal of Politics, v. 70, n. 2 pp. 542-555. DOI: https://doi.org/10.1017/s0022381608080493
  » https://doi.org/10.1017/s0022381608080493
• ALMOND, G.; VERBA, S. (1963). The civic culture: political attitudes and democracy in five nations Princeton, Princeton University Press.
• AVRITZER, L. (2002). Democracy and the Public Space in Latin America Princeton, Princeton University Press.
• AVRITZER, L. (2007). Sociedade civil, instituições participativas e representação: da autorização à legitimidade da ação. Dados, n. 50, pp. 443-464. DOI: https://doi.org/10.1590/S0011-52582007000300001
  » https://doi.org/10.1590/S0011-52582007000300001
• AVRITZER, L. (2008). Instituições participativas e desenho institucional: algumas considerações sobre a variação da participação no Brasil democrático. Opinião Pública, v. 14, pp. 43-64. DOI: https://doi.org/10.1590/S0104-62762008000100002
  » https://doi.org/10.1590/S0104-62762008000100002
• AVRITZER, L. (2009). Participatory Institutions in Democratic Brazil Baltimore/Maryland, Johns Hopkins University Press.
• BAIOCCHI, G. (2005). Militants and citizens: the politics of participation in Porto Alegre Stanford, Stanford University Press.
• BENNETT, E. A. et al. (2013). Disavowing politics: civic engagement in an era of political skepticism. American Journal of Sociology, v. 119, n. 2, pp. 518-548. DOI: https://doi.org/10.1086/674006
  » https://doi.org/10.1086/674006
• BEZERRA, C. P. et al. (2024). Entre a desinstitucionalização e a resiliência: participação institucional no governo Bolsonaro. Dados - Revista de Ciências Sociais, v. 67, pp. 1/e20220118-49.
• BIROLI, F.; MIGUEL, L. F. (2015). Gênero, raça, classe: opressões cruzadas e convergências na reprodução das desigualdades. Mediações Londrina, v. 20, n. 2, pp. 27-55. DOI: https://doi.org/10.5433/2176-6665.2015v20n2p27
  » https://doi.org/10.5433/2176-6665.2015v20n2p27
• BOBBIO, N.; MATTEUCCI, N.; PASQUINO, G. (1992). Dicionário de política Brasília, Editora UnB.
• BORBA, J. et al. (2024). Polarização ideológica entre deputados federais no Brasil (2005-2021). Estudos Históricos Rio de Janeiro, v. 37, n. 81, p. e20240111. DOI: https://doi.org/10.1590/S2178-149420240111
  » https://doi.org/10.1590/S2178-149420240111
• BORBA, J.; SILVA, G. U. D.; AMORIM, L. C. (2024). Political cleavages in Brazil: a longitudinal dimension. Revista Latinoamericana de Opinión Pública, v. 13, pp. e31370-e31370. DOI: https://doi.org/10.14201/rlop.31370
  » https://doi.org/10.14201/rlop.31370
• BORGES, A.; VIDIGAL, R. (2018). From Lulismo to antipetismo? Polarization, partisanship, and electoral behavior in Brazilian presidential elections. Opinião Pública, n. 24, pp. 53-89. DOI: https://doi.org/10.1590/1807-0191201824153
  » https://doi.org/10.1590/1807-0191201824153
• BORGES, A.; VIDIGAL, R. (2023). Para entender a nova direita brasileira: polarização, populismo e antipetismo Porto Alegre, Zouk.
• COOK, R. (2015). Are Voters Drifting Away? The Center for Politics. Sabato's Crystal Ball, April 2. Disponível em: www.centerforpolitics.org/crystalball/articles/are-voters-drifting-away/ Acesso em: 10 fev 2024.
  » www.centerforpolitics.org/crystalball/articles/are-voters-drifting-away/
• CORTES, S. M. V. (2011). "As diferentes instituições participativas existentes nos municípios brasileiros". In: PIRES, R. R. (org.). Efetividade das instituições participativas no Brasil: estratégias de avaliação Brasília, Ipea, v. 7, pp. 137-150.
• COUTO, C. G. (2014). Novas eleições críticas? Em Debate, v. 6, pp. 17-24. Disponível em: https://bibliotecadigital.tse.jus.br/xmlui/handle/bdtse/6119?locale-attribute=es
  » https://bibliotecadigital.tse.jus.br/xmlui/handle/bdtse/6119?locale-attribute=es
• DAHL, R. (1970). A moderna análise política São Paulo, Lidador.
• DALTON, R. J. (2008). The good citizen: how a younger generation is reshaping american politics. Washington, CQ Press.
• FIORINA, M. P. (2017). Unstable majorities: polarization, party sorting, and political stalemate Stanford, Hoover Institution Press.
• FIORINA, M. P.; ABRAMS, S. J.; POPE, J. C. (2004). Culture war? The myth of a polarized America Nova York, NY, Longman.
• FIORINA, M. P.; ABRAMS, S. J.; POPE, J. C. (2008). Polarization in the American public: misconceptions and misreadings. The Journal of Politics, v. 70, n. 2, pp. 556-560. DOI: https://doi.org/10.1017/s002238160808050x
  » https://doi.org/10.1017/s002238160808050x
• FIORINA, M. P.; LEVENDUSKY, M. (2006). "Rejoinder". In: NIVOLA, P. S.; BRADY, D. W. (org.). Red and Blue Nation? Volume I - Characteristics and Causes of America's Polarized Politics Washington, Stanford, CA, Brookings Inst. Press and Hoover Inst. Press.
• FUKS, M.; MARQUES, P. H. (2021). Context and vote: the impact of the right-wing political forces reorganization on the ideological consistency of the vote in the 2018 elections. Opinião Pública, v. 26, pp. 401-430. DOI: https://doi.org/10.1590/1807-01912020263401
  » https://doi.org/10.1590/1807-01912020263401
• FUKS, M.; MARQUES, P. H. (2023). Polarization and context: measuring and explaining political polarization in Brazil. Opinião Pública, v. 28, pp. 560-593. DOI: https://doi.org/10.1590/1807-01912022283560
  » https://doi.org/10.1590/1807-01912022283560
• FUKS, M.; PERISSINOTTO, R. M.; SOUZA, N. R. (2004). Democracia e participação: os conselhos gestores do Paraná. Curitiba, Editora UFPR.
• FUNG, A. (2006). Varieties of participation in complex governance. Public Administration Review, v. 66, pp. 66-75. DOI: https://doi.org/10.1111/j.1540-6210.2006.00667.x
  » https://doi.org/10.1111/j.1540-6210.2006.00667.x
• GARCÍA-ESPÍN, P.; GANUZA, E. (2017). Participatory skepticism: Ambivalence and conflict in popular discourses of participatory democracy. Qualitative Sociology, v. 40, pp. 425-446. DOI: https://doi.org/10.1007/s11133-017-9367-6
  » https://doi.org/10.1007/s11133-017-9367-6
• GURZA LAVALLE, A.; ISUNZA VERA, E. (2010). "Precisiones conceptuales para el debate contemporáneo sobre la innovación democrática: participación, controles sociales y representación". In: GURZA LAVALLE, A.; ISUNZA VERA, E. (org.). La innovación democrática en América Latina: tramas y nudos de la representación, la participación y el control social México, Ciesas - Universidad Veracruzana.
• HUDDY, L.; MASON, L.; AARØE, L. (2015). Expressive partisanship: Campaign involvement, political emotion, and partisan identity. American Political Science Review, v. 109, n. 1, pp. 1-17. DOI: https://doi.org/10.1017/S0003055414000604
  » https://doi.org/10.1017/S0003055414000604
• IPEA - Instituto de Pesquisa Econômica Aplicada (2012). O Conselho Nacional de Promoção da Igualdade Racial na Visão de seus Conselheiros. Relatório de Pesquisa Projeto Conselhos Nacionais. Brasília.
• IPEA - Instituto de Pesquisa Econômica Aplicada (2013). Conselhos Nacionais. Perfil e atuação dos conselheiros. Relatório de Pesquisa. Brasília.
• ISUNZA VERA, E.; GURZA LAVALLE, A. (2012). Arquitetura da participação e controles democráticos no Brasil e no México. Novos estudos Cebrap, 92, pp. 105-121. DOI: https://doi.org/10.1590/S0101-33002012000100007
  » https://doi.org/10.1590/S0101-33002012000100007
• IYENGAR, S. et al. (2019). The origins and consequences of affective polarization in the United States. Annual Review of Political Science, v. 22, n. 1, pp. 129-146. DOI: https://doi.org/10.1146/annurev-polisci-051117-073034
  » https://doi.org/10.1146/annurev-polisci-051117-073034
• JACKSON, R. A.; MONDAK, J. J.; HUCKFELDT, R. (2009). Examining the possible corrosive impact of negative advertising on citizens' attitudes toward politics. Political Research Quarterly, v. 62, n. 1, pp. 55-69. DOI: https://doi.org/10.1177/1065912908317031
  » https://doi.org/10.1177/1065912908317031
• KEANE, J. (2009). The life and death of democracy Nova York, W. W. Norton & Co.
• LEVENDUSKY, M. (2009). The partisan sort: how liberals became democrats and conservatives became republicans Chicago, The University of Chicago Press.
• LOPEZ, F. G.; PIRES, R. R. (2010). "Instituições participativas e políticas públicas no Brasil: características e evolução nas últimas duas décadas". In: CARDOSO JR., J. C. (org.). Brasil em desenvolvimento: estado, planejamento e políticas públicas Brasília, Ipea.
• LÜCHMANN, L. H. H. (2020). Interfaces socioestatais e instituições participativas: dimensões analíticas. Lua nova, v. 109, pp. 13-49.
• LÜCHMANN, L. H. H.; BOGO, R. (2022). Instabilidade e declínio dos Orçamentos Participativos nos municípios do Brasil e de Portugal (2016-2019). Opinião Pública, v. 28, pp. 716-748.
• LÜCHMANN, L. H. H.; BORBA, J. (2008). Participação, desigualdades e novas institucionalidades: uma análise a partir de instituições participativas em Santa Catarina. Ciências Sociais Unisinos, v. 44, pp. 58-68.
• MARTIN, A. (2012). Young people and politics: political engagement in the anglo-american democracies Londres, UK, Routledge.
• MASON, L. (2018). Uncivil agreement: how politics became our identity Chicago, University of Chicago Press.
• MCADAM, D.; MCCARTHY, J. D.; ZALD, M. N. (1996). Comparative perspectives on social movements: political opportunities, mobilizing structures, and cultural framings Cambridge, Cambridge University Press.
• MILBRATH, L. W. (1965). Political Participation Chicago, RandMcNally.
• OFFE, C. (2006). "Political disaffection as an outcome of institutional practices? Some post-tocquevillean speculations". In: TORCAL, M.; MONTERO, J. R. (org.). Political disaffection in contemporary democracies Londres, Routledge.
• OLSON, M. (1965). The logic of collective action: public goods and the theory of groups Cambridge, Harvard University Press.
• ORTELLADO, P.; RIBEIRO, M. M.; ZEINE, L. (2022). Existe polarização política no Brasil? Análise das evidências em duas séries de pesquisas de opinião. Opinião Pública, v. 28, n. 1, pp. 62-91. DOI: https://doi.org/10.1590/1807-0191202228162
  » https://doi.org/10.1590/1807-0191202228162
• PATEMAN, C. (1970). Participation and democratic theory Cambridge, Cambridge University Press.
• PUTNAM, R. D.; LEONARDI, R.; NANETTI, R. Y. (1993). Making democracy work: civic traditions in modern Italy Princeton, Princeton University Press.
• RENNÓ, L. R. (2003). Estruturas de oportunidade política e engajamento em organizações da sociedade civil: um estudo comparado sobre a América Latina. Revista Sociologia Política, v. 21, pp. 71-82. DOI: https://doi.org/10.1590/S0104-44782003000200006
  » https://doi.org/10.1590/S0104-44782003000200006
• RENNÓ, L. R. (2020). The Bolsonaro voter: issue positions and vote choice in the 2018 Brazilian presidential elections. Latin American Politics and Society, v. 62, n. 4, pp. 1-23. DOI: https://doi.org/10.1017/lap.2020.13
  » https://doi.org/10.1017/lap.2020.13
• RENNÓ, L. R. (2022). Bolsonarismo e as eleições de 2022. Estudos Avançados, v. 36, n. 106, pp. 147-163. DOI: https://doi.org/10.1590/s0103-4014.2022.36106.009
  » https://doi.org/10.1590/s0103-4014.2022.36106.009
• RIBEIRO, E.; BORBA, J. (2015). Participação política na América Latina Maringá, Eduem.
• RODRIGUES, C. S.; FREITAS, V. G. (2021). Ativismo feminista negro no Brasil: do movimento de mulheres negras ao feminismo interseccional. Revista Brasileira de Ciência Política, n. 34, e238917, pp. 1-54. DOI: https://doi.org/10.1590/0103-3352.2021.34.238917
  » https://doi.org/10.1590/0103-3352.2021.34.238917
• ROMÃO W.; MARTELLI, C. G. (2020). Gênero e participação em nível local: estudo sobre conferências municipais de políticas públicas. Rev Sociol Polit, v. 28, n. 76, e010.
• SANTOS JÚNIOR, O. A.; RIBEIRO, L. C. Q.; AZEVEDO, S. (2004). "Democracia e gestão local: a experiência dos conselhos municipais no Brasil". In: SANTOS JÚNIOR, O. A.; RIBEIRO, L. C. Q.; AZEVEDO, S. (org.). Governança democrática e poder local: a experiência dos conselhos municipais no Brasil Rio de Janeiro, Revan.
• SMITH, G. (2009). Democratic innovations: designing institutions for citizen participation Cambridge, Cambridge University Press.
• SOUSA SANTOS, B. (1998). Participatory budgeting in Porto Alegre: toward a redistributive democracy. Politics & Society, v. 26, n. 4, pp. 461-510. DOI: https://doi.org/10.1177/0032329298026004003
  » https://doi.org/10.1177/0032329298026004003
• TAYLOR, M. M.; BURANELLI, V. C. (2007). Ending up in pizza: accountability as a problem of institutional arrangement in Brazil. Latin American Politics and Society, v. 49, n. 1, pp. 59-87. DOI: https://doi.org/10.1111/j.1548-2456.2007.tb00374.x
  » https://doi.org/10.1111/j.1548-2456.2007.tb00374.x
• TRANSPARÊNCIA INTERNACIONAL (2024). Índice de percepção da corrupção 2023. Transparência Internacional Brasil. Disponível em: https://transparenciainternacional.org.br/ipc/
  » https://transparenciainternacional.org.br/ipc/
• VERBA, S.; NIE, N. H. (1972). Participation in America: political democracy and social equality. Chicago, University of Chicago Press.
• VERBA, S.; SCHLOZMAN, K. L.; BRADY, H. E. (1995). Voice and equality: civic voluntarism in american politics Cambridge, Harvard University Press.
• WOOLDRIDGE, J. M. (2006). Introdução à econometria: uma abordagem moderna São Paulo, Thomson.
• ZUCCO, C.; POWER, T. J. (2024). The ideology of brazilian parties and presidents: coalitional presidentialism under stress. Latin American Politics and Society, v. 66, n. 1. DOI: https://doi.org/10.1017/lap.2023.24
  » https://doi.org/10.1017/lap.2023.24

Notes

• 1
  Representative national samples with, respectively, 2.500, 2.009, and 2.538 respondents in 2018, 2019, and 2022. For the data, questions, and technical notes, see https://www.institutodademocracia.org/a-cara-da-democracia
• 2
  The IDDC aims to advance and deepen discussions about Brazilian democracy, taking into account its organization, the population’s democratic habits, and the country’s media landscape. The Institute is part of the Institutos Nacionais de Ciência e Tecnologia [National Science and Technology Institutes] (INCT) Program and is made up of research groups from four main institutions (UFMG, IESP/UERJ, Unicamp, and UnB) as well as researchers from USP, UFPR, UFPE, UNAMA, IPEA, and, internationally, CES/UC and UBA. The survey “The Face of Democracy in Brazil” is one of the axes articulating research into representation, participation, and public opinion within the IDDC’s scope. https://www.institutodademocracia.org/a-cara-da-democracia.
• 3
  We developed two additional models to test a hypothesis in the literature about the relationship between skepticism and non-participation. According to this hypothesis, millennials (young people born after 1984) are more skeptical of traditional state institutions and are therefore less interested in participating in them (DALTON, 2008). Additional models can be found in the Methodological Appendix.
• 4
  The models were applied to all the survey waves (2018, 2019, and 2022). However, some variables were featured in certain waves and absent in others. Thus when a variable was not available, we followed the model without it, keeping all the others that were featured.

Data availability

Publication Dates

• Publication in this collection
  31 Mar 2025
• Date of issue
  May-Aug 2025

History

• Received
  06 Sept 2024
• Accepted
  07 Jan 2025