Society against the Elites: Approaching the Social Bases of Petro’s Electoral Support in Colombia*

Gabriel Kessler

Eidaes-Universidad Nacional de San Martín (Argentina)

Gabriel Vommaro

Eidaes-Universidad Nacional de San Martín (Argentina)

Juan Carlos Rodríguez-Raga

Universidad de los Andes (Colombia)

Juan Andrés Calderón Herrera

Universidad de los Andes (Colombia)

RECEIVED: March 22, 2023

ACCEPTED: May 24, 2023

MODIFIED: August 9, 2023

https://doi.org/10.7440/colombiaint117.2024.01

abstract. Objective/Context: One of the effects of the internal war in Colombia was to hinder the expression of social conflict and the organizations that mobilize popular sectors. The plebiscite to ratify the peace agreement that attempted to end the war was a politically polarized juncture channeled through the electoral process. The displacement of the internal war as a central issue after signing the agreement opened the possibility of expression for social conflict. However, in a country with weak social and political organizations to shape it, this conflict was expressed 1) without politically aligned framings and 2) as a generalized discontent towards the elites and the feeling of a “tilted playing field.” Therefore, political polarization after the plebiscite was low. We develop this argument by analyzing voter positions on the most important topics on the political agenda: peace agreement, redistributive agenda (taxes and social aid), government management of the COVID-19 pandemic, and gender issues. Methodology: The article is based on sixteen focus groups conducted between September and November 2021 in three regions of Colombia—Bogotá, Antioquia, and the Caribbean—in which voters of the two main electoral options of 2018 participated, with gender balance, occupational variation, and distribution between middle and lower classes. Conclusion: The resulting data show no political polarization at the citizen level. Instead, there was a high level of discontent toward political and economic elites, which may be at the basis of Gustavo Petro’s electoral support in 2022. Originality: The article offers an alternative look at the perception shared by many Colombians of living in a polarized society and contributes to understanding electoral support for an “anti-system” force in the 2022 presidential election.

keywords: Colombia; focus groups; political polarization; presidential elections; social discontent.

La sociedad contra las élites: aproximación a las bases sociales del apoyo electoral a Petro en Colombia

resumen. Objetivo/contexto: uno de los efectos de la guerra interna en Colombia fue obstaculizar la expresión del conflicto social y de las organizaciones que movilizan a los sectores populares. El plebiscito por el acuerdo de paz que intentaba terminar con dicha guerra fue una coyuntura de polarización política canalizada por la vía electoral. El desplazamiento de la centralidad de la cuestión de la guerra interna a partir de la firma del acuerdo abrió la posibilidad de expresión del conflicto social. Sin embargo, en un país con débiles organizaciones sociales y políticas que ordenen este conflicto, este se expresa: 1) sin encuadres alineados políticamente; 2) como descontento generalizado con las élites y sentimiento de “cancha inclinada”. Por tanto, la polarización política luego de la coyuntura del plebiscito es baja. Desarrollamos este argumento a partir del análisis de las posiciones de los votantes en relación con los temas más importantes de la agenda política: acuerdo de paz, agenda redistributiva (impuestos y ayudas sociales) y gestión gubernamental durante la pandemia por COVID, y agenda de género. Metodología: el artículo se basa en dieciséis grupos focales realizados entre septiembre y noviembre de 2021 en tres regiones de Colombia (Bogotá, Antioquia y el Caribe), en los que participaron votantes de las dos opciones electorales principales de 2018, con equilibrio de género, con variación ocupacional, y entre clases medias y clases bajas. Conclusión: los datos muestran que no existe polarización política entre los ciudadanos. En cambio, se observa un alto nivel de descontento con las élites políticas y económicas. Originalidad: el artículo ofrece una mirada alternativa a la percepción que tienen muchos colombianos de que viven en una sociedad polarizada y contribuye a la comprensión de los apoyos electorales a una fuerza “antisistema” en la elección presidencial de 2022.

palabras clave: Colombia; descontento social; elecciones presidenciales; grupos focales; polarización política.

Sociedade contra as elites: aproximando-se da base social do apoio eleitoral da Petro na Colômbia

resumo. Objetivo/Contexto: um dos efeitos da guerra interna na Colômbia foi dificultar a expressão do conflito social e das organizações que mobilizaram os setores populares. O plebiscito pelo Acordo de Paz que buscava pôr fim à guerra foi um momento de polarização política canalizada através do processo eleitoral. O deslocamento da centralidade da questão da guerra interna após a assinatura do Acordo abriu a possibilidade da expressão do conflito social. No entanto, em um país com organizações sociais e políticas fracas para organizar este conflito, ele se expressa: 1) sem estruturas politicamente alinhadas; 2) como descontentamento generalizado com as elites e um sentimento de “campo de jogo inclinado”. E, portanto, a polarização política após o plebiscito é baixa. Desenvolvemos este argumento analisando as posições dos eleitores em relação aos temas mais importantes da agenda política: o Acordo de Paz, a agenda redistributiva (impostos e ajuda social) e a gestão governamental durante a COVID, e a agenda de gênero. Metodologia: o artigo é baseado em 16 grupos focais realizados entre setembro e novembro de 2021 em três regiões da Colômbia - Bogotá, Antioquia e Caribe - envolvendo eleitores das duas principais opções eleitorais em 2018, com equilíbrio de gênero e com variação ocupacional e variação entre classes médias e baixas. Conclusão: os dados mostram que não há polarização política no nível dos cidadãos. Em vez disso, há um alto nível de descontentamento com as elites políticas e econômicas que podem estar na base do apoio eleitoral à Petro em 2022. Originalidade: o artigo oferece uma visão alternativa da percepção que muitos colombianos têm de que vivem em uma sociedade polarizada e contribui para a compreensão do apoio eleitoral a uma força “anti-establishment” na eleição de 2022.

palavras-chave: Colômbia; descontentamento social; eleições presidenciais; grupos de foco; polarização política.

Introduction

The social sciences literature has discussed the characteristics of political polarization in Colombia, a country where satisfaction with democracy has systematically decreased from 2012 to the present. The AmericasBarometer (AB) by LAPOP shows that satisfaction with the performance of democracy fell by 6 % between 2014 and 2018 and support for democracy by 8.3 % in the same period.

The rupture within the right, with the appearance of a radical electoral offer in matters of security such as Uribism—later institutionalized in the Democratic Center party (Gamboa Gutiérrez 2019)—and then the 2016 plebiscite on the peace agreement suggested that political polarization was on the rise (Amador 2017; Barrios-Rubio and Gutiérrez-García 2022; Feldmann 2019; Pécaut 2022). Furthermore, in such plebiscite, the gender agenda deepened the divisions around the armed conflict (Serrano Amaya 2017). The emergence of a competitive left-wing political offer gave this polarization a programmatic character unprecedented in the country. For the first time, the second round in the last two electionswas between the right-wing and a left-wing option. Given this new political context, to what extent is this polarization established among voters? Is it expressed through polarized positions on the main issues of the most representative agendas in the public debate?

To answer these questions, in this article we analyze the positions of voters regarding the most important issues on the Colombian political agenda: the peace agreement, redistributive policy (taxes and social assistance), the government management of the COVID-19 pandemic, and gender issues. In line with Sartori’s classic contributions (Sani and Sartori 1983; Sartori 1966), we use the definition of polarization in an ideological sense. To specify its components, we rely on the seminal work of DiMaggio, Evans, and Bryson (1996), for whom the concept has four dimensions: tendency toward bimodality (that is, the grouping of opinions into two categories), high dispersion (which indicates that these two categories are very distant from each other), high consistency (people’s positions on the different issues tend to align within each agenda and across agendas), and consolidation (the correspondence of these categories with social variables and with the vote). The ideological definition of polarization associated with positions regarding issues on the agenda has been further developed by other authors, especially in the North American debate on whether polarization is an issue limited to the elites or whether it reaches down to society (Abramowitz and Saunders 2008; Fiorina and Abrams 2008). More recently, some studies (Webster and Abramowitz 2017) show that ideological preferences are at the basis of the so-called affective polarization, that is, it is based on negative feelings towards the adversary group.

To capture positions regarding agenda issues, framing theory is particularly useful. Following this theory,

an issue can be viewed from a variety of perspectives and be construed as having implications for multiple values or considerations. Framing refers to the process by which people develop a particular conceptualization of an issue or reorient their thinking about an issue. (Chong and Druckman 2007, 104)

We aimed to identify similarities and differences in the existing framings of each issue by vote (in the 2018 presidential election), class, gender, age group, and geographic area. We took as voting options the two candidates of that election who contested the runoff: Iván Duque and Gustavo Petro. Given that these are also ideologically different options, one aligned with the right and the other with the left, that election becomes particularly useful to establish the scope of polarization in Colombian society and determine whether there are differences in positions regarding the main issues on the agenda between the voters on the left and the right.

The article is based on sixteen focus groups conducted between September and November 2021 in three regions of Colombia: Bogotá, Antioquia, and the Caribbean region. Voters of the two main 2018 electoral options—Iván Duque and Gustavo Petro—participated in these groups, with gender balance, occupational variation, and distribution between middle and lower classes. Despite not allowing for generalizable conclusions to be drawn, qualitative studies and especially focus groups are particularly relevant to capture nuances and differences in positions regarding issues in the public debate, especially controversial ones (Cyr 2017). In addition, we also rely in this article on quantitative data from surveys (AB) on the topics addressed in the groups.

Our analysis strongly suggests that there is no clear ideological polarization between Petro and Duque voters. This is consistent with the fact that, as we will see later, voters located on the left and right were largely fewer compared to those in the center,1 a consistent pattern throughout the years, although with some variations. In terms of framings, each candidate’s voters often do not resort to framings that are politically aligned with the party they voted for (or with the most classic ones of the current left and right in Latin America) but rather use others often closer to the other side or without a clear affiliation. In those cases, moral, religious, or other framings prevail. This finding is consistent with the thesis of Botero, Losada, and Wills-Otero (2015), who maintain that, after the constitutional reform of 1991, new partisan actors and internal factions of the traditional parties emerged and that toward the end of the 1990s, the country moved from a two-party to a fragmented party system with low ideological alignment. Likewise, it is consistent with the high electoral volatility identified in Colombia: the percentage of vote change from one presidential election to the next is 45 %, unlike in more ideologically polarized contexts, such as Argentina and Brazil, where this percentage is around 20 % (Kessler and Murillo 2023). Indeed, a low stabilization of political preferences over time is associated with a lower influence of party frames on voters (Kessler and Vommaro 2023). In the same direction, Botero, Losada, and Wills-Otero (2015) maintain that since the 1960s issues related to security, related both to the armed conflict and the high levels of insecurity and drug trafficking, have been among the main public concerns, without the traditional parties having a coordinated or permanent position on this issue over time.

Thus, the trait shared by most citizens is a high level of discontent with the political and economic elites, embodied in the perception of a “tilted playing field,” meaning that the design of public policies and the decisions of governments are made to favor the elites who had always governed for their benefit. In this sense, in our study, the variable that best explains the polarization of opinions is class, taking both the socioeconomic stratum and the educational level as proxies. On the other hand, the vote is less associated with differences in points of view on the issues investigated. In the redistributive agenda, general discontent and even fatigue with the elites predominate among our interviewees. On the cultural agenda, the vote is not a good predictor of polarized positions.Progressive redistributive framings do not organize discontent in a classical way either: the perception of a tilted social playing field leads, for example, to a low consensus around increasing taxes given the beilief that those taxes will fall sooner or later on the middle and lower classes.

Beyond this introduction, the paper proceeds as follows. We begin by exposing the key aspects of the debate on polarization in Colombia. Next, we present the methodological strategy and then review the framings found in different topics: peace agreement, redistributive issues, government management of the COVID-19 pandemic, and the cultural agenda. We analyze the view voters in each camp have about themselves and others and conceptualize the idea of a tilted playing field as a distinctive perception in the Colombian case. We close the text with some conclusions and working hypotheses for future research.

1. The Debate on Polarization in Colombia

There has been an increasing debate about political polarization in Colombia in recent years. There are at least three focal points in this discussion. The first, in line with the American debate, refers to the scope of polarization: Is it a matter of the elites, or does it penetrate society?

Some authors maintain that the division within political elites, caused by the opposition of former President Uribe’s supporters to the peace process with the Revolutionary Armed Forces of Colombia – People’s Army (FARC-EP, for its initials in Spanish) and the results of the plebiscite to endorse the peace agreement in 2016, would have activated a polarization process that became one of the defining features of current politics (Amador 2017; Barrios-Rubio and Gutiérrez-García 2022; Feldmann 2019; Pécaut 2022). It would be the result of the fragmentation of the elites—between those who supported a negotiated solution to the conflict and those who opposed it—during the peace process (García and Matanock 2017; Pécaut 2022). Then President Juan Manuel Santos’ change of position regarding how to end the armed conflict created a division within the elites between those loyal to former President Uribe and those who supported Santos (Feldmann 2019). The discourses of both camps reinforced the dynamics of fragmentation of public opinion, and this way polarization was conveyed to the citizens level. As a result, the opposition of former President Uribe’s supporters to the peace process could partly explain the rejection of the plebiscite in 2016 (García and Matanock 2017; Rodríguez-Raga 2017).

For others, however, the political system experiences a process of diversification and pluralization, which allows the emergence of a competitive left-wing political offer yet without polarized citizens regarding their opinions on substantive issues (Borda 2018). In particular, Borda (2018) points out that identification with the ideological center has not decreased in recent years and that there is no alignment between preferences for the left and positions on the cultural agenda. Along these lines, Singer (2016) maintains that the increased ideological polarization at the elite level is related, as in other countries in the region, to the strengthening of left-wing parties.

The second focal point of the debate is whether or not polarization is associated with distributive issues and notions of the State. Kalmanovitz (2019) and Montoya (2023) identify a polarization between positions that advocate for a more active state intervention in the economy (traditionally associated with the left) and those that favors the role of the market (associated with the right). However, evidence seems to show that, in Colombia, there is some consensus on economic matters (Junguito 2019), and BA data suggest that the preferences of Colombians lean toward a more active participation of the state in the economy, with no evidence of polarization.2

Thirdly, after gender and sexual diversity issues have gained prominence in the public debate and seem to have some role in explainiing the results of the 2016 plebiscite, there is controversy about the scope of polarization on the cultural agenda. Aguirre, Pabón, and Cáceres (2018) and Maier (2018) find a cleavage around cultural or moral values, from a more progressive position that approves, for example, abortion or equal marriage, to a more conservative or traditional position that rejects social transformations. However, as we will point out below, although BA public opinion data show positions that are distant from each other, the bimodality coefficient of this distribution is far from reaching the polarization threshold.3

Amid these controversies, our work shows the similarities and differences in the attitudes of the voters of each candidate, as well as the nuances in their positions. It also examines the contents of the framings people use in each issue. In so doing, this article contributes to answering the question about the ways in which (low) polarization and conflict take place in Colombian society.

2. Methodological Strategy

The data used in this article come from a mixed-methods research endeavor on polarization and sociopolitical conflict in Colombia.4 Between August and September 2021, we conducted sixteen focus groups with Duque and Petro voters in the second round of the 2018 presidential elections. We conducted eight focus groups with people residing in Bogotá and its periphery, four in Antioquia (Bello, Itagüí, Rionegro, and Envigado), and four in the Caribbean region (Santa Marta, Cartagena, Valledupar, and Montería). The groups were conducted virtually; each group had six participants, with quotas defined for gender, age, and vote decision. They were conducted by a researcher in the team, with the presence of colleagues as observers. In terms of class, we conducted five focus groups with lower-class voters (with education level up to high school completion, from socioeconomic strata 1, 2, and 3) and five focus groups with middle-class voters (with technical or higher education, from socioeconomic strata 2, 3, 4, and 5). In addition, six focus groups were conducted with a mix of middle- and lower-middle-class voters (regardless of educational level, from socioeconomic strata 1, 2, and 3) who had diverse occupations.5 Eight groups were homogeneous, and eight were heterogeneous regarding voting decision in the past election. This allowed us to compare conversational dynamics between groups of people with similar political views and groups of people with divergent stances. In terms of age, three cohorts were defined: 18-35 (identified as young), 35-55 (middle-aged), and 55 and older (senior). We hired a specialized firm to work on recruiting participants in the focus groups under our continuous supervision. Filters and data requirements were applied to each participant to corroborate that they corresponded to the required profile. All participants were anonymized, signed a consent form to be recorded and filmed, and received an economic incentive for participating in the group.

In the groups, participants were asked to give their opinions on the main issues of the country’s political, economic, social, and cultural agenda: security, peace agreement, migration, social assistance, taxes, corruption, gender issues, sexual diversity, reproductive rights, and the health and economic management of the COVID-19 pandemic by the central and local governments. The focus groups were coded using Atlas.ti. We conducted a content analysis using a coding process based on the focus groups guidelines, to which we added emerging topics. We considered variables regarding structural, sociocultural, geographic, and political orientation features when comparing the responses. We examined the relationships between contents to outline the main framings in each theme (Piovani 2018).

Why did we choose this data collection technique? Focus groups have a long history in sociological research (see Gamson 1992; Merton 1948) since they allow the generation of conversational dynamics in which it is possible to capture nuances and tensions around controversial topics (Cyr 2017). They enhance the observation of exchanges between participants, their reciprocal influence, and the collective configuration of social meaning matrices (Archenti 2018). As our goal was to identify similarities and differences in the participants’ opinions, we had to be attentive to two issues. First, to minimize the social desirability effect, whereby some people would refrain from expressing their discordant views. As the number of people was limited, we could ask everyone their opinion and encourage the utterance of differences, thus generating a safe and cordial discussion framework. The online modality entailed a lower emotional charge among participants than in face-to-face meetings. We were also aware of the need to avoid the opposite effect, which is polarization as understood by social psychology, i.e., the adoption of more extreme positions than those commonly held as a result of interacting with people with disparate opinions (Sunstein 2002). Our intervention in conducting the focus groups minimized both risks.

3. Agendas and Framings

In the following sections, we analyze the dominant positions and the differences by vote and by class between right-wing and left-wing voters on three key agendas: peace agreement, redistributive issues (taxes and social aid) and their impact on the evaluation of government performance during the COVID-19 pandemic—a highly divisive issue in other Latin American cases (Aruguete et al. 2021)—, and cultural issues (sexual, reproductive, and sexual diversity rights). Based on this, we describe how voters view each other (how Petro voters view Duque voters, and vice versa) and conclude with an analysis of widespread discontent and the perception of a tilted playing field.

1. Peace Agreement: Traces of a Polarized Situation

One of the most contentious moments in the recent history of Colombia was the plebiscite to ratify the peace agreement reached and signed with the FARC-EP in 2016. The rejection option obtained a tight victory, with 50.2 % of the votes. In that line, subsequent opinion studies have shown that around one out of every two Colombians supported the agreement (Gaviria Dugand et al. 2021). As stated, such juncture has fueled the perception of a deep polarization in the Colombian society (Amador 2017; Barrios-Rubio and Gutiérrez-García 2022; Feldmann 2019; Pécaut 2022). García and Matanock (2017) highlight that electoral events to ratify peace agreements tend to amplify divisions in the elites. Indeed, there is evidence that citizens used the signals of the political elites with whom they felt more affinity to assess the peace agreement (García and Matanock 2017). However, survey data force us to qualify the intensity of this division and its persistence over time. Indeed, BA data for Colombia do not show evidence of a considerable level of polarization around the issue.6

What did the focus groups show us? Our data were collected five years after the plebiscite; thus, the initial opinions had undoubtedly been reconfigured by subsequent events. In the groups, we found heterogeneous opinions (for and against), but, in most cases, the divisions were not sharp; many participants adopted mixed positions and highlighted both positive and negative elements of the events around the peace agreement. Moreover, the sharpest positions were found among both Duque and Petro voters. One of the latter, a senior inhabitant of Bogotá of socioeconomic stratum 3 with higher or technical studies, stated:

I voted “No” because it is obvious… it is obvious that a person who commits a crime will not have a differential treatment, especially for crimes against humanity… he cannot have a differential treatment.

Conversely, a middle-aged Petro voter from Valledupar, with education up to high school and of socioeconomic stratum 2, said:

I agree with the peace agreement. I was in the military, and the comrades who are there… life has become easier for them, that is, with the peace agreement, they are not in combat areas as much as they were before, you know? For me, this has helped, maybe not completely, but it has lowered the rate of violence, and the work of the military in the combat area has been less.

Even some of those who voted “Yes” in the plebiscite expressed sympathetic positions with those who were against, such as this young Duque voter of socioeconomic stratum 2 from Bogotá, with higher education:

Yes, I agreed with the peace accords, but I understand the position of those who voted “No.” Usually, all or most of them want peace, but those who said “No,” it is because they did not agree with some points of the peace agreement, with certain impunities [...] because if they were to tell the truth about what happened, there would be a reduction of sentences, how could they reduce sentences for those who committed crimes during all this time in the wars.

In any case, the time elapsed enabled an evaluation of the results of the agreement so that the judgment made at the time of the plebiscite could be modified. In this line, Andrés, a middle-aged Petro voter from Envigado, of socioeconomic stratum 3, with higher or technical education, considered that he was wrong in his rejection: “I voted ‘No’ to the peace agreement, but today I regret it. Why? As he already said, the number of soldiers who have been taken to military hospitals has sharply decreased.”

Different studies highlight the gap perceived by citizens between the expectations generated by signing the agreement and changes observed subsequently (Gaviria Dugand, Ávila García, and García Sánchez 2019). In particular, since the beginning of the Duque administration, there were signs of a slowdown in the implementation of the commitments, particularly in the agrarian situation and the management of illicit drugs (Kroc Institute for International Peace Studies 2020). In fact, the critical assessments of our interviewees go beyond political affinity. As a Bogotá resident middle-aged Petro voter of socioeconomic stratum 3 noted:

I think there was a deception; that is, when they told us that there was going to be a peace treaty and that the peace so longed for in Colombia was going to come, many of us said yes; in fact, I was one of those who voted yes because I thought there was going to be a change. We always looked for that change, but no, after a short time, they went and spent a little money over there, and they come, and once again we see them with the same weapons, we see them returning to those same groups, and on top of it, they have some privileges.

Finally, as in other dimensions discussed below, the peace agreement is an occasion to express discontent with the political elites. As a middle-aged, university-educated Petro voter of socioeconomic stratum 4 from Bogotá explained:

There comes back the issue of the usual politician who instills fear into the Colombian people that the guerrillas are going to continue, and then he comes to power with this president we have now [Iván Duque], and what he did was to completely dismantle everything that had been done, then… when I negotiate with you, and you do not comply, it was obvious that they would return to arms because they did not get compliance [...]. If Colombia had been a country that did not allow itself to be influenced by this politicking, the guerrillas would simply have reached Congress with the obtained agreement; however, the Colombian people would have brought it down because we did not want it.

Criticism of the elites may be based on the economic cost of the peace agreement, as stated by this Bogotá resident middle-aged Duque voter of socioeconomic stratum 2, with basic secondary education:

The peace process is very good, but in conducting a peace process, billions of pesos are spent. Where does that money come from? From our pockets. We are going to bleed the country even more; we are going to be poorer every day, believe me, because of this peace process.

In sum, as previous studies have stated, the peace agreement continues to be one of the most controversial issues, and it is here where the permanence of a right-wing framing is apparent, more in the lower than the middle socioeconomic strata. However, it did not seem to be specially salient at the time we conducted the groups, and the participants’ evaluation focused mainly on the implementation and the current consequences of the agreement. Ultimately, rather than antagonisms aligned with the vote, there seems to be a widespread disenchantment with its implementation and a sense of disappointment with an elusive peace. This disenchantment is expressed by Norma, a Duque-voter housewife from Bello, Antioquia: “It [peace] is very difficult, at least I believe that in our lifetime, who knows in how many more generations, but we will not experience that happiness.”

2. Redistributive Injustice and State Neglect

Positions on the redistributive agenda are weakly aligned with the vote. For example, there are scarce references both to reducing inequalities—a typically progressive framing—and to disapproving of those who prefer to live off “assistentialism”—a usually conservative framing found in countries such as Argentina and Brazil (Kessler and Vommaro 2023). Moreover, the (few) critics of the social aid, because people “want everything for free,” were both Petro and Duque voters, particularly from the middle class. However, we found two generalized agreements: the need for social assistance, especially among the lower strata, and the opinion that existing social programs were insufficient. Regardless of electoral preferences, support for social aid generally mobilized a framing associated with “need.” It is possible to summarize this position with the words of Laura, a middle-aged Duque voter from Montería, of socioeconomic stratum 2, with high school education, and employed in commerce: “For me, [social aid] should be permanent, because I know that many people, not only here in Colombia, are in need, and that is seen on a daily basis.”

As for the proposal to raise taxes for the richest, as proposed in several countries during the pandemic, almost all participants, voters from both camps, especially those from lower socioeconomic strata, disapproved of this idea because they considered that, ultimately, the elite would manage to avoid paying the costs, and the burden would fall on poorer people. This explains the massive rejection of the tax reform project that the Duque government tried to introduce in 2021. Here are some opinions on the subject:

Well, taxes are necessary for a country, but what happens is that, in Colombia, most of those who pay taxes are the populace, and taxes on companies are very low. (Eduardo, a young employee from Bogotá, of socioeconomic stratum 3, with tertiary education, Petro voter)

I do not agree that the issue is to rise taxes, no, because there will be taxes, and then, as long as there are more taxes, there will be more money to steal for those who have always stolen. (Jorge, a middle-aged worker in the service area from Bogotá, of socioeconomic stratum 3, Petro voter)

No, there should not be taxes, not at this moment, because we are going to be affected, the middle and lower sectors, we are going to be affected, our pockets cannot stand it, no. (Elina, a middle-aged dressmaker from Bello, of socioeconomic stratum 3, with high school education, Duque voter)

This negative view had its epitome in the generalized opinion of abandonment by the state during the COVID-19 pandemic, expressed by most respondents, both those who voted for Duque and those who supported Petro (44 of 63 coded references to COVID-19 follow this pattern). The belief that the poorest and most vulnerable people were the ones who suffered the consequences of the pandemic prevailed. Strictly speaking, some of our interviewees who voted for Duque (and even supporters of Petro) were more benevolent with the government’s handling of the pandemic, given its unexpected nature. In the words of Joaquín, a middle-aged Petro voter from Envigado, of socioeconomic stratum 3, with technical or higher education: “We do not criticize Duque for the pandemic because nobody was familiar with it.”

The indulgent view of the government ended there. For those interviewed, the impacts of the pandemic on the most vulnerable were evidence of both the structural situation of inequality and the agency of the political and economic elites. The response of the central and local governments was considered meager due to a lack of aid and to distribution problems. It was claimed that social assistance had been given to those who did not need it due to corruption and complicity with the elites, or to negligence and disorganization at the municipal level.

“The benefits went more to them, and the common people are not seen [...] They were the ones who got most of the benefits because they received aid, they received goods. But, at that time, the state took and distributed it and gave it to those in the ‘loop,’” as explained by Ester, a middle-aged recycler in Bogotá and Petro voter of socioeconomic stratum 2, with high school education. Another Bogotá resident, a middle-aged Duque voter of socioeconomic stratum 2, with secondary education, underscored this perception of inequity:

The aid they sent... for example, many here... you see, they sent several aids from different countries with supplies, and look, through the news and social networks, those supplies rotted in many warehouses, they were damaged because they did not deliver them or they were stolen for other people, there, to other places, but never delivered them to the people who really needed them. The same with money... money, the state was giving monetary aid that never reached us; that is to say, it arrived, but it went to the big oligarchs. They are the ones who take that, not you.

Neither the central nor the local governments were spared from the judgment of state neglect. Both lower- and middle-class participants reported receiving little or no assistance during the pandemic, in part because they did not qualify as a target population for previously existing conditioned transfers. Territorial inequalities also played a role: the sense of abandonment was particularly strong outside Bogotá, among respondents from the department of Antioquia and especially in the Caribbean. For the latter, the historical relegation of the region by the central government deepened during the pandemic.

Even Petro voters were critical of the performance of the mayor of Bogotá, as this testimony from Adela, a middle-aged, unemployed Bogotá resident and Petro voter socioeconomic stratum 2, with secondary education, indicates:

The government was simply for them; as long as my family is well and I have food and all the necessary supplies, I do not care for the rest. This Ms. Claudia López, in my opinion, was like, “I support people but from here, and you are from there; nobody leaves their house, but I go on a trip.” I need to go out to sell some candy or coffee, and many people who were selling on the street were fined, and many people lost their income; so, let’s say that what the government did was the same as always. Then, one hears that the man who owns a supermarket was uttering out loud that he received a subsidy. So, what they said before, what´s the point of giving a subsidy if it is the people with money the ones who receive it? For me, no government helped; many people even had to try to find ways to make our own facemasks.

In short, the widespread perception of profound inequity was flagrantly manifested by an experience of state neglect during the pandemic. Meanwhile, the underlying judgment that any measure with some cost would sooner or later fall on the backs of the poorest led to fear and rejection of even the most anti-elite framings, such as support for higher taxation of the rich. Data from the BA and our groups on supporting greater state intervention were partly contradicted by the fear that, instead of benefiting them, eventual redistributive measures, such as new taxes, would ultimately harm the middle and lower sectors.

3. Cultural Agenda: The Weakness of Progressive Framings

Since the beginning of the twenty-first century, Colombia has undergone profound social changes related to gender, sexual, and reproductive rights: abortion was partially decriminalized;7 equal marriage for same-sex couples and their right to adopt children were recognized, the definition of family as a legal category was modified to include, for example, same-sex families; and transgender people were able to change their name and gender on their identity card without requiring judicial intervention. Similarly, discrimination based on gender identity and sexual orientation was criminalized, and LGBTIQ+ persons can now be recognized as victims of the armed conflict and, therefore, receive economic compensation (Rodríguez Rondón 2017, 136). A good part of the political elite has not supported these changes; instead, it has been the Constitutional Court that, in the face of the refusal of Congress to legislate on these issues, has played an active role in the reformist process. While minority rights got expanded, a strong local resistance arose among conservative sectors. As almost everywhere in Latin America (Biroli and Caminotti 2020), conservative activists—especially Catholic and evangelical groups—built around the so-called “gender ideology” a common framing to identify a threat to their values (Rodríguez Rondón 2017).

These conservative mobilizations peaked in 2016, around the plebiscite for the peace agreement. Earlier that year, there were intense demonstrations against “gender ideology” in response to an initiative of the Ministry of Education to introduce teaching materials in schools to reduce homophobia and transphobia after the suicide of a young high school student harassed for being homosexual. Political and religious leaders denounced a plot by the LGBTIQ+ movement, the United Nations, the Ministry of Education, and the Constitutional Court to impose homosexuality on the country’s children and indoctrinate them in gender ideology. Bearing in mind the plebiscite campaign, former President Uribe found an opportunity for growth by incorporating conservative sectors into his coalition for the “No” vote (Muelle 2017; Vigoya and Rodríguez Rondón 2017). To some extent, conservative mobilization overlapped with activism favoring a hardline position on the armed conflict propitiated by Uribe (Rodríguez-Raga 2017). So much so that, for “No” supporters, in addition to allowing favorable conditions for the FARC members, the peace agreement hid a recognition of gender rights (Serrano Amaya 2017).

If the cultural agenda was part of the polarized issues during the plebiscite process, and studies show a cleavage around them (Aguirre, Pabón, and Cáceres 2018; Maier 2018), this was not the case at the time of conducting the focus groups. In line with what Corredor (2021) shows, the conservative strategy during the plebiscite did not necessarily reverse or slow down the societal changes underway. To be sure, there were clear points of division in the groups, but not necessarily aligned with the vote or associated with clear political framings: even our most conservative interviewees never referred explicitly to the term “gender ideology.” By way of example, of the twenty-one coded quotes on abortion, ten were partially in favor (i.e., only in cases of rape or danger to the life of the pregnant woman), six approved of it without conditions, and five were against it. The six quotes in favor were equally distributed between Petro and Duque voters while, of the five quotes against, three corresponded to Petro and two to Duque voters. Something similar happened with support for LGBTIQ+ rights. These were also divisive issues, but the positions were not aligned with the vote.8 Regarding support for gender equality, there was mostly consensus, except among religious ultraconservatives. Of the 49 coded quotes, 44 were in favor of expanding equality between men and women, four were partially in favor, and one was against it.

The most identifiable interpretative framings are associated with religious convictions. Thus, in the case of abortion, most people appealed to religious motives to express their opposition; when they were partially in favor, they expressed that they did so despite the fact that it went against Christian principles. Neither did we notice a progressive stance “from below” among Petro voters—identifiable from the use of progressive framings in this matter—except in the few cases of people who were openly militant or who showed a higher level of politicization. As an example, among those who supported abortion without conditions, the idea of the “right to decide”—the central framing of pro-choice movements (Abracinskas and López Gómez 2007; Tarducci 2018)—did not appear, but rather a framing associated with individual freedom but without reference to a legal framework that guaranteed it: “Let everyone do what they want.”

In the same direction, the main framing for supporting the rights of the LGBTIQ+ population was associated with “tolerance” rather than with the idea of rights: it is a population that exists, that has made itself visible, and, therefore, must be accepted. Likewise, a majority was in favor of equal marriage but not of adoption by same-sex couples because they feared that their children would acquire the same sexual orientation. Iván, an middle-aged, unemployed Petro voter of socioeconomic stratum 1 from Montería, with high school education, supported this idea:

I am open-minded, I do agree [with same-sex marriage], if we look at the biblical principles, well, no, but I do agree. But when it comes to adoption, no.

We have also identified seemingly contradictory discourses: in favor of adoption by same-sex couples so that children would not be homeless, but against the marriage of such couples, since it is a union that should exclusively happen between a man and a woman, “as God’s law dictates.” There were also group participants who said they had no opinion on this. For example, Patricia, a middle-aged Petro voter of socioeconomic stratum 1 from Montería, employed, with a high school education, expressed her doubts in the following way:

Regarding adoption, well, I agree since... a person who is in a state, suddenly lacking a family, has the right to grow within that bond, I agree with the right to love [...] and with respect to same-sex marriage, well, I am still undecided about that, I cannot give my opinion because I do not really know much about the subject.

Likely, the doubts or lack of opinion on the subject expressed by Patricia and many others are an indicator of the weakness of framings linked to stabilized progressive socio-political coalitions, as found in Argentina, Brazil, or Uruguay.

Finally, the scarce weight of progressive framings in different agendas could be observed in another nodal issue: intense xenophobia against Venezuelan migrants, across all profiles and voters of the two main presidential candidates. Indeed, a large part of the participants in the focus groups held strong prejudices and stigmatized migrants from Venezuela, to whom they attributed responsibility for crimes and considered them a threat to Colombian culture, particularly because they could negatively influence the mystique of an entrepreneurial society with their habits of being “comfortable,” as expressed by a senior female Petro voter from Bogotá, housewife, with primary education and from socioeconomic stratum 2: “Because there are a lot of Venezuelans here, they are not used to pay rent, or to pay, or study, or anything, they had everything there and here they came with those ideas.”

4. Views on the Others

As stated, survey data indicate no ideological polarization among voters, even though levels of elite polarization might exert a significantly strong effect on the ideological-partisan classification of citizens (Borda 2018; Davis and Dunaway 2016). Figure 1 shows histograms of respondent answers between 2004 and 2018 to the question on ideological self-placement on a scale from 0 (left) to 1 (right). Although for some years—especially since 2012—the frequency of those who placed themselves at the ends of the ideological spectrum seemed to grow, in all measurements, the center was the place on the scale most chosen by Colombians.9

Figure 1. Distribution of self-placement on the scale from left to right (2004-2018)

Source: Own elaboration based on AmericasBarometer data.

To what extent can we speak of affective polarization? In the focus groups, we inquired about the views of opposing voters; in general, we did not notice intensely negative feelings or political-ideological disqualifications. Although they recognized themselves as different, they also mentioned similar traits, such as trust in a charismatic leader and discontent. For example, Jaime, a Duque voter of socioeconomic stratum 2, with education up to high school completion, pointed out that “if we look at it closely, a Petro voter is perhaps a bit like an Uribe voter, in the sense that they see a solution in a single person, as in messianism.” Another trait shared by voters from both camps was deep dissatisfaction with political elites and their handling of the country’s main problems. It even seems that political discontent blurred the differences between Uribe and Petro supporters, as explained by a middle-aged Duque voter of socioeconomic stratum 2, with education up to high school completion:

I would think that it is not that we are different but that we believe that one is better than the other. However, when it comes down to it, I think that all presidents are the same; when they come to power, they all end up the same.

The exception to this scarce affective polarization came from some voters intensely identified with Uribe or Petro. Some Uribe supporters, particularly men, described Petro’s supporters as “young people who see in him a chance for change, for opportunities, but they are not really well informed about… well, is what Petro proposes really plausible?” (young male Duque voter of socioeconomic stratum 2, with high school completed). Likewise, most senior Uribe supporters in our groups, both men and women, considered Petro supporters to be people from all social strata who are “dissatisfied” with their socioeconomic reality but who like to live off the public budget. Juan Carlos, a Duque voter of socioeconomic stratum 3 with a technical or higher education, described a Petro voter as “a person who likes to be given everything. I see him [Petro], and he uses to offer subsidies for one thing or another. Those who vote for Petro is because they like easy things.”

In the other camp, for young Petro supporters, Uribe supporters were “older people” and very little prone to argumentation and debate. In this regard, Magdalena, a Petro voter from socioeconomic stratum 3 with a technical or higher education, pointed out that Uribe supporters were “people who tend to be uncritical and unreflective, for whom the end justifies the means.” Another defining trait of Uribe supporters would be a high socioeconomic status. Samuel, a Petro voter of socioeconomic stratum 2 with education up to secondary school completion, described this in the following terms:

If we look at it from the point of view of the rich and the poor, obviously the poor population, wanting a new change, wanting to thrive, will want to vote for Petro; as for the rich, since they already have power and many already have enough food, there is not one of these people who will vote for Petro.

In sum, we did not observe in the focus groups a degree of division between Uribe and Petro supporters at the same levels found by other authors among political elites in more recent electoral contests (Marcos 2018). According to the groups, while there was a distance between the political preferences of Uribe and Petro supporters, the ideological polarization of voters was low, and the level of hostility or aversion between supporters of the two main political camps was low. On the contrary, the common trait that could possibly allow mitigating a potential enmity was that most participants were concerned about the future of the country and, above all, about finding new leaders far from the “traditional political class.”

5. Being Fed Up with the Elites in a Tilted Playing Field Society

Beyond the divisions and heterogeneity of opinions, the main feature of the positions in the different agendas was an anti-elite sentiment and a feeling of state abandonment across all classes and political orientations. The rejected elites were the different governments that had always come from the upper classes and ruled for their own interests. This feeling of being fed up came from a lack of opportunities to study and live with dignity, from the perception that they makelittle money and spend a lot, and that, in times of emergency, such as during the COVID-19 pandemic, the state abandoned them. Voters of both candidates agreed on this point: the country is not moving forward; there are no opportunities, neither for study nor for work. Ana María, a young female Duque voter with secondary education from Bogotá, sales assistant in a store, agreed with this line of thought:

It has been too disappointing to see how the country, rather than moving forward, is going more and more backwards, especially for us young people. We feel that… well, personally, I feel that we have no opportunity to work or to study. So, those of us who are high school graduates, it is only a daily struggle to be able to attain something or wonder if we are going to have a better future.

The lack of opportunities and the negative view of the elites generated a view of a tilted playing field, not in relation to the institutional advantages of the ruling parties but to the elites’ taking advantage of their access to the state to design public policies for their benefit. The overriding idea was that everything was set up—the tax structure, social assistance, corruption, clientelism, etc.—for the benefit of the elites. The perception of a tilted playing field led to anger and despair; we perceived a sense of weariness with “the same people over and over again.” Certainly, discontent was directed especially at the “traditional political class” in general and less focused on businesspersons or the economic elite; however, many times, both types of elites appeared to be conflated as part of the same groups and interests.

Meanwhile, few interviewees indicated any party identification; among the ones who expressed some identity, those who considered themselves Uribe supporters prevailed. However, they were equally critical of the government. As we have pointed out, we noticed in all groups a low presence of political framings in key agendas, such as the redistributive or cultural ones. This had an impact on the lack of knowledge or absence of positions regarding some of these issues, as well as on the recurrence of religious or moral framings. The lack of interest in politics was not a cause for shame; on the contrary, many claimed it was attributed to the fact that “the same people have always ruled.” The weakness of progressive framings in redistributive and cultural matters made this demand for leveling the playing field diffuse. Likewise, although discontent was generalized, there was a marked class bias in almost all issues. Indeed, the key characteristic observed in the qualitative analysis—reinforced by a computational analysis of the conversation10—was that the most significant polarization was by class on some issues, much more than by vote, age, or other variables. Thus, widespread discontent had a socially divided basis.

Conclusion

This article aimed to investigate similarities and differences between Duque and Petro voters on the main issues of the Colombian political and social agenda. The main agreement was a deep discontent with the political elites regarding handling the country’s principal problems, the epitome of which was the abandonment felt during the COVID-19 pandemic. This generalized discontent tempered differences among the voters since all of our interviewees shared the perception of a tilted playing field, understood as the elites taking advantage of their access to the state to design public policies exclusively for their benefit, whether in the tax structure, the functioning of social benefits, corruption, or clientelism.

However, in each discussed agenda, although we found agreements and variations in the focus groups, the positions rarely aligned clearly with the vote for one candidate or the other. Thus, five years after the peace accords, we noted nuanced positions on this point and, in some cases, modifications of the original position the group participants had held at the time. Among the positive aspects, some emphasized that a degree of general pacification was achieved, although less than expected, that benefited both the civilian population and the security forces. On the negative side was the economic cost of reparations and, more generally, that the reality fell short of the promises. All in all, there was an acceptance that peace was a very difficult and time-consuming goal and that the political elites were not up to the challenge.

Similarly, the positions on the distributive agenda were not clearly divided according to electoral preference. On the contrary, there was widespread criticism of the scant amount and poor coverage of social assistance during the pandemic. They blamed both the national government and the local authorities for the inequity of distribution: aid was given to those who did not need it due to corruption or carelessness and disorganization. Beyond assistance during the pandemic, there was a consensus, especially in the lower strata, of the need for more social aid because it had always been insufficient. Moreover, the widespread rejection among those in lower socioeconomic strata of the implementation of new taxes on the richest sectors was somewhat surprising since, as a result of the tilted playing field, the conviction prevailed that, in the end, the cost would always be borne by the poorest. The framing of the redistributive issue made almost no reference to reducing inequalities—a typically progressive framing—or the criticism of “assistentialism”—a usually conservative framing found in Argentina and Brazil. Those who rejected social assistance were mostly middle-class voters of both candidates who morally invalidated people who “want everything for free.”

As for the cultural agenda, we did not perceive that it was among the most important concerns of our interviewees. The decriminalization of abortion remained a divisive issue with no clear cleavage among voters. Opposition was mostly based on religious principles, and those in favor did not argue for a “right to decide” but rather stated that it was necessary to give everyone freedom to do as they wished. Regarding the LGBTIQ+ issue, there was a prevailing idea of “tolerance” or acceptance of a population that exists and cannot be denied. While support for equal marriage was the prevailing position, many of those who accepted it opposed adoption by same-sex couples because of objections to their instilling their sexual orientation in children. A very present theme in the groups was intense xenophobia against Venezuelan migrants, who were accused of committing crimes and eroding the hard-work culture of Colombians.

Regarding how they saw the opposing voters, with some exceptions from both camps, no strong animosity or political-ideological disqualifications were expressed. The participants recognized themselves as different, but they also saw themselves as similar in their trust in a charismatic leader and, above all, in their discontent regarding the elites. Now, when contrasts were drawn, the most intense Petro voters described Uribe supporters as conservative, from higher classes, and older people; for their part, the most committed Uribe supporters perceived typical Petro voters as young people who wanted a change but often without sufficient information and who were deceived by their leader.

These are the main findings of our focus groups. The work has, of course, the limitations of a qualitative study, and the conclusions are only valid for our corpus. Therefore, to conclude, we would like to offer some reflections that should be tested by subsequent work. In general, our data seem to suggest that the weakness of ideologically aligned party traditions in Colombia is reflected in the scarce presence of political framings in voter positions. This seems to be more noticeable in countries where a left-wing sociopolitical coalition has not been established in recent decades. These coalitions provide framings that align their followers, both within each agenda and between agendas. This favors an ideological alignment of political-electoral supply and demand. In this way, discontent is organized more horizontally (distributing preferences among adversaries) than vertically (society against the elites), as seems to happen in Colombia. Our groups seem to suggest that, in a country with a low party programmatic tradition, with little recent party identification and growing discontent with political elites, it is less likely that a political division of that level will be transferred to society. However, change in the political offer and the process that opened in Colombia with Petro’s presidency may activate a different political dynamic than that found until now. Future research will undoubtedly continue to provide new clues to account for these new times and show how much of what we presented here persists and how much will be transformed.

References

1. Abracinskas, Lilian, and Alejandra López Gómez, coords. 2007. Aborto en debate. Dilemas y desafíos del Uruguay democrático. Montevideo: Mujer y Salud en Uruguay.
2. Abramowitz, Alan, and Kyle Saunders. 2008. “Is Polarization a Myth?” The Journal of Politics 70 (2): 542-555. https://doi.org/10.1017/S0022381608080493
3. Aguirre, Javier, Ana Pabón, and Paul Cáceres. 2018. “Argumentos religiosos y matrimonio igualitario: análisis de las discusiones en el Congreso de la República de Colombia, a la luz de la propuesta de Habermas sobre la religión en la esfera pública.” Reflexión Política 20 (39): 201-219. https://doi.org/10.29375/01240781.3303
4. Amador, Diego. 2017. “De polarización, teflón y referendos: sesgo de confirmación y tribalismos en la política.” La Silla Vacía, October 9, 2017. https://www.lasillavacia.com/red-de-expertos/blogeconomia/de-polarizacion-teflon-y-referendos-sesgo-de-confirmacion-y-tribalismos-en-la-politica/
5. Archenti, Nélida. 2018. “Focus group como forma de entrevista grupal.” In Manual de metodología de las ciencias sociales, edited by Alberto Marrandi, Nélida Archenti, and Juan Piovani, 279-290. Buenos Aires: Siglo XXI Editores.
6. Archila, Mauricio, Martha García, Santiago Garcés, and Ana María Restrepo. 2020. “21N: el desborde de la movilización en Colombia.” Lasa Forum 51 (4): 17-23.
7. Aruguete, Natalia, Ernesto Calvo, Francisco Cantú, Sandra Ley, Carlos Scartascini, and Tiago Ventura. 2021. “Partisan Cues and Perceived Risks: The Effect of Partisan Social Media Frames during the COVID-19 Crisis in Mexico. Journal of Elections.” Public Opinion and Parties 31 (1): 82-95. https://doi.org/10.1080/17457289.2021.1924740
8. Barrios-Rubio, Andrés, and María Gutiérrez-García. 2022. “Comunicación mediática y social en el entorno de crisis, polarización e inconformismo colombiano.” Profesional de la Información 31 (3): e310320. https://doi.org/10.3145/epi.2022.may.20
9. Biroli, Flávia, and Mariana Caminotti. 2020. “The Conservative Backlash against Gender in Latin America.” Politics & Gender 16 (1): E1. https://doi.org/10.1017/S1743923X20000045
10. Borda, Sandra. 2018. “Presidenciales en Colombia: ¿polarización o deterioro de la conversación política?” Nueva Sociedad, May, 2018. https://www.nuso.org/articulo/presidenciales-en-colombia-polarizacion-o-deterioro-de-la-conversacion-politica/
11. Botero, Felipe, Rodrigo Losada, and Laura Wills-Otero. 2015. “Sistema de partidos en Colombia (1974-2014): ¿la evolución hacia el multipartidismo?” In Los sistemas de partido en América Latina, 1978-2015, edited by Flavia Freidenberg, 339-400. Mexico City: Instituto de Investigaciones Jurídicas, UNAM.
12. Chong, Dennis, and James Druckman. 2007. “Framing Theory.” Annual Review of Political Science 10: 103-126. https://doi.org/10.1146/annurev.polisci.10.072805.103054
13. Corredor, Elizabeth. 2021. “On the Strategic Uses of Women’s Rights: Backlash, Rights-Based Framing, and Anti-Gender Campaigns in Colombia’s 2016 Peace Agreement.” Latin American Politics and Society 63 (3): 46-68. https://doi.org/10.1017/lap.2021.24
14. Cyr, Jennifer. 2017. “The Unique Utility of Focus Groups for Mixed-Methods Research.” Political Science & Politics 50 (4): 1038-1042. https://doi.org/10.1017/S10490-9651700124X
15. Davis, Nicholas, and Johanna Dunaway. 2016. “Party Polarization, Media Choice, and Mass Partisan-Ideological Sorting.” Public Opinion Quarterly 80: 272-297. https://doi.org/10.1093/poq/nfw002
16. DiMaggio, Paul, John Evans, and Bethany Bryson. 1996. “Have American’s Social Attitudes Become More Polarized?” American Journal of Sociology 102 (3): 690-755.
17. Feldmann, Andreas. 2019. “Colombia’s Polarizing Peace Efforts.” In Democracies Divided, edited by Thomas Carothers and Andrew O’Donohue, 153-176. Washington: Brookings Institution Press.
18. Fiorina, Morris, and Samuel Abrams. 2008. “Political Polarization in the American Public.” Annual Review of Political Science 11 (1): 563-588. https://doi.org/10.1146/annurev-.polisci.11.053106.153836
19. Gamboa Gutiérrez, Laura. 2019. “El reajuste de la derecha colombiana. El éxito electoral del uribismo.” Colombia Internacional 99: 187-214. https://doi.org/10.7440/colombia-int99.2019.07
20. Gamson, William A. 1992. Talking Politics. New York: Cambridge University Press.
21. García, Miguel, and Aila Matanock. 2017. “The Colombian Paradox: Peace Processes, Elite Divisions & Popular Plebiscites.” Daedalus 146 (4): 152-166. https://doi.org/10.-1162/DAED_a_00466
22. Gaviria Dugand, Adriana, Carlos Arturo Ávila García, and Miguel García Sánchez. 2019. Barómetro de las Américas. Colombia. 2018. Paz, posconflicto y reconciliación. Bogotá: Observatorio de la Democracia. https://www.vanderbilt.edu/lapop/colombia/Colombia_2018_Informe_Paz_conflicto_y_reconciliacion_W_11.07.19.pdf
23. Gaviria Dugand, Adriana, Juan Andrés Calderón, Pedro Mejía, and Juan Carlos Rodríguez Raga. 2021. Colombia, un país en medio de la pandemia. 2020. Paz, posconflicto y reconciliación. Bogotá: Observatorio de la Democracia. https://obsdemocracia.org/wp-content/uploads/2021/06/02_USAID_PAZ_2020.pdf
24. Kroc Institute for International Peace Studies. 2020. Tres años después de la firma del Acuerdo Final de Colombia: hacia la transformación territorial. Informe 4. Diciembre 2018 a noviembre 2019. Bogotá: Universidad de Notre Dame, Iniciativa Barómetro, Matriz de Acuerdos de Paz. http://peaceaccords.nd.edu/wp-content/uploads/2020/09/091620-Reporte-4-Digital-.pdf
25. Junguito, Roberto. 2019. “La polarización económica.” Portafolio, May 28, 2019. https://www.portafolio.co/opinion/roberto-junguito/la-polarizacion-economica-530031
26. Kalmanovitz, Salomón. 2019. “La polarización y la economía.” El Espectador, May 27, 2019. https://www.elespectador.com/opinion/columnistas/salomon-kalmanovitz/la-polarizacion-y-la-economia-column-862802/
27. Kessler, Gabriel, Brenda Focás, Juan Manuel Ortiz, and Esteban Feuerstein. 2020. “Los divergentes en un escenario de polarización. Un estudio exploratorio sobre los ‘no polarizados’ en controversias sobre noticias de delitos en la televisión argentina.” Revista SAAP 14 (2): 311-340. https://doi.org/10.46468/rsaap.14.2.a3
28. Kessler, Gabriel, and María Victoria Murillo. 2023. “Political Reactions to Changing Societies: Framework.” Paper presented at the Conference Changing Societies and Political Discontent, ILAS-Columbia University, March 30-31, 2023.
29. Kessler, Gabriel, and Gabriel Vommaro. 2023. “Forms of Discontent in Latin America: Polarization, Generalized Discontent and Divisive Leaderships.” Paper presented at the Conference Changing Societies and Political Discontent, ILAS-Columbia University, March 30-31, 2023.
30. Lelkes, Yphtach. 2016. “Mass Polarization: Manifestations and Measurements.” The Public Opinion Quarterly 80: 392-410. https://doi.org/10.1093/poq/nfw005
31. Maier, Elizabeth. 2018. “Hidden Meanings of the Culture War over Abortion in the United States.” Frontera Norte 30 (59): 57-80. https://doi.org/10.17428/rfn.v30i59.879
32. Marcos, Ana. 2018. “La semana de violencia en la contienda electoral de Colombia.” El País, March 3, 2018. https://elpais.com/internacional/2018/03/03/colombia/1520089060_498288.html
33. Merton, Robert K. 1948. “The Bearing of Empirical Research upon the Development of Social Theory.” American Sociological Review 13 (5): 505-515. https://doi.org/10.2307/2087142
34. Montoya, Juan David. 2023. “El papel (funciones) del Estado en la economía.” Actividades Económicas, May 12, 2023. https://www.actividadeseconomicas.org/2017/12/el-papel-del-estado-en-la-economia.html
35. Muelle, Camila. 2017. “Cómo hacer necropolíticas en casa: ideología de género y acuerdos de paz en Colombia.” Sexualidad, Salud y Sociedad 27: 172-198. https://doi.org/10.1590/1984-6487.sess.2017.27.10.a
36. Ortiz de Zárate, Juan Manuel, Marco di Giovanni, Esteban Feuerstein, and Marco Brambilla. 2020. “Measuring Controversy in Social Networks through NLP.” In String Processing and Information Retrieval: 27th International Symposium, SPIRE 2020, Orlando, FL, USA, October 13–15, 2020, Proceedings, edited by Christina Boucher and Sharma V. Thankachan, 194-209. Berlin, Heidelberg: Springer-Verlag.
37. Ortiz de Zárate, Juan Manuel, and Esteban Feuerstein. 2020. “Vocabulary-based Method for Quantifying Controversy in Social Media.” arXiv, January 4, 2020. https://doi.org/10.48550/arXiv.2001.09899
38. Pécaut, Daniel. 2022. “Entre polarización política y protesta social.” Análisis Político 34 (102): 175-189. https://doi.org/10.15446/anpol.v34n102.99940
39. Piovani, Juan. 2018. “El análisis de contenido”. In Manual de metodología de las ciencias sociales, edited by Alberto Marrandi, Nélida Archenti, and Juan Piovani, 421-437. Buenos Aires: Siglo XXI Editores.
40. Rodríguez-Raga, Juan Carlos. 2017. “Colombia: país del año 2016.” Revista de Ciencia Política 37 (2): 335-368. http://doi.org/10.4067/s0718-090x2017000200335
41. Rodríguez Rondón, Manuel. 2017. “La ideología de género como exceso: pánico moral y decisión ética en la política colombiana.” Sexualidad, Salud y Sociedad 27: 128-148. https://doi.org/10.1590/1984-6487.sess.2017.27.08.a
42. Sani, Giaccomo, and Giovanni Sartori. 1983. “Polarization, Fragmentation and Competition in Western Democracies.” In Western European Party Systems, edited by Hans Daalder and Peter Mair, 307-340. London: Sage.
43. Sartori, Giovanni. 1966. “European Political Parties: The Case of Polarized Pluralism.” In Political Parties and Political Development, edited by Joseph LaPalombara and Myron Weiner, 137-176. Princeton: Princeton University Press.
44. Serrano Amaya, José Fernando. 2017. “La tormenta perfecta: ideología de género y articulación de públicos.” Sexualidad, Salud y Sociedad 27: 149-171. https://doi.org/10.1590/1984-6487.sess.2017.27.09.a
45. Singer, Matthew. 2016. “Elite Polarization and the Electoral Impact of Left-Right Placements: Evidence from Latin America, 1995-2009.” Latin American Research Review 51 (2): 174-194. https://doi.org/10.1353/lar.2016.0022
46. Sunstein, Carl. R. 2002. “The Law of Group Polarization.” The Journal of Political Philosophy 10 (2): 175-195. https://doi.org/10.1111/1467-9760.00148
47. Tarducci, Mónica. 2018. “Escenas claves de la lucha por el derecho al aborto en Argentina.” Salud Colectiva 14 (3): 425-432. https://doi.org/10.18294/sc.2018.2036
48. Vigoya, Mara, and Manuel Rodríguez Rondón. 2017. “Hacer y deshacer la ideología de género.” Sexualidad, Salud y Sociedad 27: 118-127. https://doi.org/10.1590/1984-6487.sess.2017.27.07.a
49. Webster, Steven W., and Alan Abramowitz. 2017. “The Ideological Foundations of Affective Polarization in the US Electorate.” American Politics Research 45 (4): 621-647. https://doi.org/10.1177/1532673X17703132

---