Available via license: CC BY 4.0
Content may be subject to copyright.
European Review of Latin American and Caribbean Studies
Revista Europea de Estudios Latinoamericanos y del Caribe
DOI: http://doi.org/10.32992/erlacs.10835 © Asbel Bohigues, João V. Guedes-Neto, Manoel
Leonardo Santos. Open Access article distributed under the terms of the Creative Commons
Attribution 4.0 Unported (CC BY 4.0) License https://creativecommons.org/licenses/by/4.0/.
WWW.ERLACS.ORG is published by CEDLA – Centre for Latin American Research and
Documentation / Centro de Estudios y Documentación Latinoamericanos; University of
Amsterdam; www.cedla.uva.nl; ISSN 0924-0608, eISSN 1879-4750.
No. 114 (2022): July-December, pp. 1-24
www.erlacs.org
Latin American political elites’ positions on same-sex marriage,
abortion, and drug legalization
Asbel Bohigues
Universidad de Valencia
João V. Guedes-Neto
Kean University
Manoel Leonardo Santos
Universidade Federal de Minas Gerais
Abstract
What explains the attitudes of Latin American political elites towards same-sex marriage,
abortion, and drug legalization? In recent years, several countries have engaged in political
disputes centred on these three issues. Scholarship has explored the topic from different
perspectives, ranging from theories of democracy to voters’ political behaviour. None-
theless, the perspective of political elites is still understudied. We use data from elite surveys
in sixteen Latin American countries to address this question. First, we find significant dif-
ferences between and within countries and political parties, meaning these issues are far
from consensual. Second, we analyse the effects of ideology, religion, and socio-
demographic characteristics. We confirm that ideology and religiosity are key predictors,
but a deeper analysis shows that whereas those on the right and religious are strongly op-
posed to these issues, those on the left and less religious are not as strongly in favour. Key-
words: Same-sex marriage, abortion, drug legalization, elites, Latin America.
Resumen: Posiciones de las elites políticas latinoamericanas sobre matrimonios del mismo
sexo, aborto y legalización de las drogas
¿Qué explica las actitudes de las élites políticas latinoamericanas hacia el matrimonio entre
personas del mismo sexo, el aborto y la legalización de las drogas? En los últimos años,
varios países se han involucrado en disputas políticas centradas en estos tres temas. La aca-
demia ha explorado el tema desde diferentes perspectivas, que van desde las teorías de la
democracia hasta el comportamiento político de los votantes. No obstante, la perspectiva de
las élites políticas todavía está poco estudiada. Usamos datos de encuestas de élite en dieci-
séis países latinoamericanos para abordar este tema. Primero, encontramos diferencias signi-
ficativas entre y dentro de los países y los partidos políticos, lo que significa que estos temas
están lejos del consenso. En segundo lugar, analizamos los efectos de la ideología, la reli-
2 | ERLACS No. 114 (2022): July-December
gión y las características sociodemográficas. Confirmamos que la ideología y la religión son
predictores clave, pero un análisis más profundo muestra que mientras los derechistas y los
religiosos se oponen fuertemente a estos temas, los izquierdistas y los menos religiosos no
están tan fuertemente a favor. Palabras clave: Matrimonio entre personas del mismo sexo,
aborto, legalización de drogas, élites, América Latina.
Introduction
What explains the attitudes of Latin American political elites towards same-sex
marriage (SSM), abortion, and the legalization of drugs? These non-economic
liberal values have been studied from several perspectives. They include public
opinion, lobbying and advocacy, societal and congressional compositions, and
so on. From Lipset (1959) to Dion and Díez (2017), most of them assume that
elites react to exogenous forces rather than being motivated by their own posi-
tions – an unlikely characteristic of the lawmaking process. Furthermore, they
are usually focused on single issues, rather than considering (and comparing)
different civil liberties. We aim at filling this gap in the literature by explaining
the individual attitudes of members of parliament (MPs) from all over Latin
America. Furthermore, we use an extensive literature review to identify poten-
tial predictors of non-economic liberal attitudes and test each one of them with
elite survey data. We draw on the Latin American Elites Database from the
University of Salamanca (PELA-USAL), which gathers face-to-face interviews
of Latin American MPs, and run multi-level mixed-effects ordered logistic re-
gressions to identify the drivers of support for SSM, abortion and drug legali-
zation. Besides contributing to the literature of political attitudes, Latin Ameri-
can politics, and morality politics, we perform this study in a moment when
same-sex marriage, abortion, and the legalization of drugs are on the rise. In
December 2013, Uruguay passed a bill allowing the recreational use of mariju-
ana (Londoño, 2017); in January 2018, the Inter-American Court of Human
Rights ruled in favour of same-sex marriage (Murillo, 2018); until the begin-
ning of 2018, conditional abortion was legal in at least eleven countries in the
region (Appendix II).
We present at least three key findings. First, these three issues are highly
controversial in Latin America: There is no consensus in most countries and
political parties, no matter their institutional framework or ideological position.
Second, religiosity (that is, frequency of participation in religious cults) and
ideology are the main predictors, statistically significant in all three cases. So-
cio-demographics and the perceived trustworthiness of the Catholic Church are
significant in either one or two of the three issues, yet with lower coefficients.
Third, although we identify a general effect coming from ideology and religi-
osity, the favourable opinions among leftists and non-believers are much less
strong than the unfavourable opinions among rightists and religious are. This
gives some insights on why there was not a substantive advance on this social
agenda, especially when and where there have been leftist legislative majori-
Asbel Bohigues, João V. Guedes-Neto, Manoel Santos: Latin American political elites’ positions | 3
ties. Following this introduction, we examined the scholarship on developed
countries and Latin America to identify relevant independent variables to our
research. The next section presents the data and methods of this paper. This
empirical approach is applied in the following section, where we display or-
dered logistic regressions and marginal effects for several variables. The final
section concludes.
Attitudes, factors and values
There are many explanations for attitudes towards same-sex marriage, abor-
tion, and legalization of drugs. Our starting point is Lipset (1959), who was
then followed by Welzel (2013) in seeking explanations for a combination of
attitudes. In both cases, the authors focused on attitudes that are outside the
economic realm (even though indirectly related to income and class) and re-
flect individual rights. They are the right to marry someone of the same sex,
the right to interrupt a pregnancy, and the right to use drugs. As we demon-
strate, there are similar individual-level predictors affecting them. Authors
converge in the identification of social class, age, religiosity, gender, and parti-
sanship as predictors of non-economic liberal attitudes. Older and poorer indi-
viduals tend to be more conservative, as well as members of traditionally con-
servative parties. Whereas the influence of religious affiliation and engagement
function as strong explanatory variables, authors are not cohesive at identifying
which religions or denominations lead to more conservative opinions. The fac-
tors that affect the support for non-economic liberal policies among the popula-
tion are not controversial. However, the individual-level attitudes of political
elites are understudied. In a few publications, authors point at MPs’ religiosity
and gender. Congresswomen tend to favour non-economic liberal policies, fur-
ther than being elected also because of a liberal (or pro-women’s rights) stereo-
type. Nonetheless, there is no cohesion when it comes to public opinion. As we
will discuss, women tend to be either as liberal or less liberal than men depend-
ing on the case, even though women are more engaged than men in political
issues as abortion.
Social class
Lipset (1959) argues that the poorer strata of the population tend to oppose
non-economic liberalism, meaning “for example, civil liberties for political
dissidents, civil rights for ethnic and racial minorities, internationalist foreign
policies, and liberal immigration legislation” (Lipset, 1959, 485). Some under-
lying factors are the linkages between these individuals and conservative reli-
gions, low education, low political engagement, economic insecurities, and
family patterns. In sum, they are more conservative because of their isolation
from the liberal discourse. In this sense, the political integration of the working
class could lead to the emergence of non-economic liberal values. The modern-
4 | ERLACS No. 114 (2022): July-December
ization argument of Welzel (2013) follows this path. He finds that emancipa-
tive values, a collection of non-economic liberal attitudes (e.g., tolerance to
abortion, divorce, and homosexuality), grow as nations develop. This trend
generates what he terms “benign individualism,” somehow similar to the char-
acteristics of Florida’s (2014) creative class. Nonetheless, Andersen and Fetner
(2008b) suggest that economic development just alters the values of those who
benefit the most from this process. In this context, the working class’s social
values remain unaffected, generally less tolerant toward homosexuality.
In Latin America, Orcés (2013) looked into public opinion in Ecuador in
2013 to find that the only predictor for abortion was high income. Montalvo
and Saunders (2015) also confirm the relevance of social class when investigat-
ing which Latin American voters are more supportive of LGBT+ rights, includ-
ing participation of homosexuals in political campaigns and the recognition of
same-sex marriage. Their findings resemble those of modernization theory in
developed countries. That is, they identify the wealthier, most well-educated
(measured in years of schooling), and residents of urban areas as being the
most favourable toward these non-economic liberal freedoms. Boidi and Corral
(2013) find that the same variables help explain support for abortion rights in
the region. Thus, the first independent variable that is commonly attributed to
affect non-economic liberal values is economic. Be it at the macro-level (i.e.,
the country’s growth) or at the micro-level (i.e., wealthier individuals), the
modernization thesis poses that economic development exposes individuals to
diversity, thus increasing the chances that they will become tolerant to (or sup-
portive of) the legalization of drugs, same-sex marriage, and abortion.
Age
Modernization theory often attributes more relevance to inter-generational
shifts than to those that occur within the same generation (Welzel, 2013). This
proposition is supported by Baunach (2012) and Andersen and Fetner (2008a),
who explore attitudes towards same-sex marriage in the United States. Tedin
(1980) addresses the issue of age by comparing the political attitudes of ado-
lescents and their parents. Whereas his data is not nationally representative
(i.e., 183 adolescents and 322 parents from Iowa in 1972), the results repro-
duce a common trend. He finds that adolescents tend to be considerably more
liberal towards marijuana laws than their parents. Alternative explanations for
preferences toward drug decriminalization in the same timeframe were higher
education (Knoke, 1979) and self-identification as liberal (Conover and Feld-
man, 1981). As it happens in the literature focused on developed countries,
there is convergence on the relevance of inter-generational trends to explain
attitudes towards non-economic liberal attitudes in Latin America. This is sup-
ported by several authors who relied on LAPOP to find, among voters, a nega-
tive correlation between age and support for these policies (see Lodola & Cor-
ral, 2010; Boidi, 2013; Marcano 2013; Maldonado, 2015). Alcántara (2013)
Asbel Bohigues, João V. Guedes-Neto, Manoel Santos: Latin American political elites’ positions | 5
finds a similar trend among Latin American legislators. He demonstrates that
older MPs tend to favour abortion slightly more than younger representatives.
The rationale follows the same expectations stated before, which connects
modernization theory to cultural shifts especially among the younger portion of
the population.
Religiosity
As previously mentioned, Lipset (1959) highlights that the link between social
class and religious affiliation helps to explain the conservative behaviour of the
working class. There are several surveys in the United States confirming these
expectations. They show that religious affiliation (most often with Protestant
denominations) predicts intolerance toward homosexuality (Herek, 1987),
same-sex marriage (Olson et al., 2006; Sherkat et al., 2010; Ellison et al.,
2011), and abortion (Álvarez & Brehm, 1995). In fact, religiosity influences
legislative behaviour in the United States through different channels. Besides
the direct relevance of public opinion (Lax & Phillips, 2009), representatives
are also influenced by religious organizations (Soule, 2004) and the share of
Catholic and Protestant identifiers in the population (Mooney & Lee, 1995;
Kreitzer, 2015).
While some authors account for the progressive nature of the Church in
Latin America (Trejo, 2009; Mainwaring & Wilde, 1989), religiosity is still
widely used as a predictor for non-economic conservatism in the region (Vag-
gione, 2011; Viterna, 2012). Engagement with religious activities (especially
Catholic and Evangelical) are linked to negative attitudes toward abortion
among voters (Boidi & Corral, 2013) and elites (Alcántara, 2013). Similarly,
Dion and Díez (2017) suggest that lack of religiosity (i.e., frequency of reli-
gious services attendance) is a strong predictor of support for same-sex mar-
riage especially when the interviewee has positive democratic values (Dion &
Díez 2017). Still, they show that favouring democracy by itself does not auto-
matically make someone in favor of LGBT+ rights.
Vaggione (2011) discusses the influence of the Catholic Church on debates
over marriage in Argentina – where the Vatican directly approached Catholic
MPs to influence decision making regarding same-sex marriage. Viterna
(2012) assesses the Salvadoran case from the perspective of abortion laws. She
argues that the approximation between Frente Farabundo Martí para la Libera-
ción Nacional (FMLN) and the Catholic Church has drastically changed the
former left guerrilla organization’s discourse. When in power, FMLN has al-
lowed the development of stark anti-abortion laws rather than advancing the
feminist agenda (see Kampwirth, 2008 for the Nicaraguan case).
6 | ERLACS No. 114 (2022): July-December
Gender
Gender is introduced as a predictor of non-economic liberal values especially
in papers related to abortion. Jennings and Farah (1981) analysed a survey car-
ried out with Republican and Democrat delegates from Michigan in 1976 to
find that women were more favourable to abortion in both parties. Huddy and
Terkildsen (1993) ran an experiment with undergraduate students to evaluate
gender stereotypes. They perceive that women are seen as more competent to
handle abortion legislation. This seems to have electoral relevance, since Dolan
(1998) shows that pro-abortion values increase the chances of voting for a
woman in Senate elections. Still, Kaufmann and Petrocik (1999) argue that
men and women tend to have similar positions on abortion based on data from
the National Election Studies in 1992 and 1996. Schlozman et al. (1995) use a
different dataset in the same timeframe to find out that men are much less po-
litically engaged but more favourable to abortion than women. If that is true, it
means that women have different average attitudes as voters and politicians –
at least for abortion.
Previous research has identified the impact of feminism, women bancadas
(i.e., congressional caucuses) and women’s interest groups on abortion policies
in Latin America (Franceschet, 2003; Hahner, 1985; Mazur, McBride, &
Hoard, 2016; Pribble, 2006; Santos, 2004). The same is true for sexual minori-
ty rights, which have advanced also due to the engagement of LGBT+ move-
ments (Díez, 2010; Encarnación, 2011; Kollman, 2007). This is supported, to a
certain extent, by scholars dedicated to study attitudes. Among voters, Boidi
and Corral (2013) find that women are significantly more supportive of abor-
tion than men. However, among elites, Alcántara (2013) shows that, whereas
strong agreement for this policy if often more present among women than men,
there is no significant difference in average responses. This seems to be the
opposite finding of the literature focused on developed countries.
Ideology
Finally, there is the issue of ideology and partisanship. In the United States,
Adams (1997) demonstrates that only the Democratic Party’s elites became
increasingly more tolerant toward abortion from 1973 to 1994. Baunach (2012)
relies on more recent data (1998-2008) to show that this one-sided cultural
shift also affect mass-level attitudes, thus increasing the ideological gap be-
tween liberals and conservatives. Naturally, many of these variables are con-
nected. There is an intrinsic link between ideology (or partisanship) and non-
economic liberal attitudes, which is dependent on other variables. This is part
of Lipset’s original proposition regarding the conservative working class,
which holds these attitudes not only because of their party affiliation but also
due to the isolation motivated by social class and religious affiliation. Further-
more, as suggested in the recent literature on social sorting (Mason, 2018),
Asbel Bohigues, João V. Guedes-Neto, Manoel Santos: Latin American political elites’ positions | 7
there is an ongoing alignment of selected social identities (e.g., religion, gen-
der, race), partisanship, and political attitudes.
In Latin America, ideology appears to matter at least when the dependent
variables are drug and abortion legalization. After the legalization of marijuana
in Uruguay, Boidi, Queirolo, and Cruz (2015) found that the country’s popula-
tion was largely supportive of abortion and same-sex marriage. Yet, 59.9 per-
cent were sceptical of drug legalization – a policy that was widely supported by
left-leaning voters and those who approved the (also left-leaning) government.
Alcántara and Rivas (2013) also explored the role of ideology on support for
drug legalization in the region. However, besides considering an expanded set
of cases (Argentina, Ecuador, El Salvador, Guatemala, and Nicaragua), they
focused on elite attitudes. They find support to the thesis that the Latin Ameri-
can left is more supportive of this policy than their right-wing colleagues. The
same, according to Alcántara (2013), is also true for support for abortion – an-
other non-economic liberal issue.
Latin American elites’ non-economic liberal values
Most studies assessing non-economic liberal values centre their research on a
single dependent variable rather than also tackling other related attitudes, e.g.,
toward abortion and drug legalization. If scholarship aims at better understand-
ing elites’ and masses’ belief systems, new studies should include comprehen-
sive measurements that combine sets of attitudes, rather than focusing on sin-
gle issues. As previously mentioned, one example of this approach is provided
by Welzel (2013) – even though focused solely on public opinion. Further-
more, research on elite behaviour and attitudes has exclusively touched the
relationship between politicians and interest groups (e.g., religious organiza-
tions, social movements) or societal and political conjunctures. Elites are gen-
erally treated as actors who solely respond to external influencers rather than
implementing their own ideas. Our paper aims at filling this gap, especially
given the availability of empirical evidence from PELA-USAL. Besides the
similar predictors of different non-economic liberal values, there seems to be
cohesion in their relevance to explain attitudes in Latin America and the devel-
oped world. In most accounts, richer, younger, non-religious, and a woman
increases the chances of supporting issues as same-sex marriage, abortion, and
drug legalization. Additional explanations in the region are the residence in
urban areas, education level, attitudes toward democracy, and country effects.
The latter evidence that whereas the region is homogeneous in many historical
and political aspects, each country has its own particularity (see for instance
Miranda, 2011). Whereas deviations in the core independent variables do exist,
we expect to find similar results when systematically studying MPs’ attitudes
in the region.
Still, the main difference between the accounts of Latin Americanists and
other students lies in the assessment of ideology. Whereas research on devel-
8 | ERLACS No. 114 (2022): July-December
oped democracies (generally the United States) focus on partisanship, the weak
party systems of Latin America (Guedes-Neto & Bohigues, 2018) lead to the
use of self-placement scores in the left-right scale as a measurement of ideolo-
gy. In these cases, left-wingers are generally more liberal, reproducing the
American divide between conservative Republicans and liberal Democrats.
The assumption of weak partisan ties follows a vast literature. Latin American
parties usually have short life-spams (Lupu, 2016) and their representatives
tend to hold incongruent policy attitudes (Hawkins & Morgenstern, 2010;
Rosas, 2010). It follows, among others, from the electoral rules in the region.
Since voters generally have to vote for candidates in open list systems, individ-
ual images and personal linkages become more relevant than party brands
(Deegan-Krause 2007). Even though we discuss such partisan conflicts in the
descriptive part of this work, we opt for variables as left-right and government-
opposition self-placement when running regressions. We expect that these pa-
rameters function as stronger predictors especially under fragmented party sys-
tems.
We also follow the assumption that the non-economic liberal attitudes of
political elites and public opinion are guided by similar factors at least to a cer-
tain extent. Socioeconomic conditions, mostly related to group identity (i.e.,
class, gender, religion) are relevant to determine one’s political views, as well
as the ideological self-positioning in the right-left spectrum. Indeed, there may
exist discrepancies led by the misstereotyping of certain identities (i.e., electing
individuals from a certain group given a wrong or exaggerated understanding
of issue ownership) or intragroup discrepancies in the levels of engagement
(i.e., individuals of a given group are politically activated based on specific
attitudes). Still, group identity remains a relevant predictor of non-economic
liberal attitudes.
Countries and variables
Are these findings reflected in the opinion of Latin American political elites?
In other words, what individual variables predict non-economic liberal values
among MPs? We use the Latin American Elites Database from the University
of Salamanca (PELA-USAL)1 to address this question. It includes face-to-face
interviews with MPs from lower houses in every Latin American democracy in
the post-electoral year since 1994. The questionnaires include three specific
questions about same-sex couple marriage, abortion, and drug legalization.
Merging the responses from MPs from the sixteen countries here analysed (Ta-
ble 1) resulted in 1,124 interviewees. This sample of elite surveys represents a
unique opportunity to better understand the attitudes of MPs. To identify de-
terminants of the attitudes toward these three issues, we run multi-level mixed-
effects ordered logistic regressions, with MPs nested in countries.
Asbel Bohigues, João V. Guedes-Neto, Manoel Santos: Latin American political elites’ positions | 9
Table 1. Countries included in the study
Country
Legislature
Fieldwork year
N
Argentina
2009-2013
2011-2015
2012
67
Nicaragua
2012-2017
2013
52
Paraguay
2014-2018
2014
55
Chile
2014-2018
2014
68
Colombia
2014-2018
2014
84
Costa Rica
2014-2018
2014
55
Honduras
2014-2018
2014
82
Panama
2014-2019
2014
47
Mexico
2015-2018
2016
100
El Salvador
2015-2018
2015
58
Bolivia
2015-2020
2015
93
Uruguay
2015-2020
2015
69
Dominican Rep.
2016-2020
2017
61
Guatemala
2016-2020
2016
78
Venezuela
2016-2021
2016
67
Ecuador
2017-2021
2017
88
Source: PELA-USAL
We selected questions that address issues related to individual liberties, namely
the possibility of same-sex couples to marry, the possibility of aborting a preg-
nancy, and the legalization of drugs. The questions are as following: (1) How
strongly do you approve or disapprove that same-sex couples have the right to
get married?; (2) Indicate in the following scale your personal opinion with
regard to abortion; and (3) How strongly do you approve or disapprove the
legalization of drugs? Thereafter, the interviewer presented a scale from 1 to
10, where 1 means ‘strongly disapproves/against’ and 10 means ‘strongly ap-
proves/for’. Our models use these scores as the dependent variables.
Our independent variables are based on the reviewed literature on non-
economic liberal attitudes. First, we use two religion-related variables. One
regards the trust of the respondent in the Catholic Church: very trustworthy (4),
trustworthy (3), untrustworthy (2) and very untrustworthy (1). The other as-
sesses religiosity, that is, frequency of mass attendance. This question uses a 1-
5 scale restricted to believers, where 1 means never/hardly ever and 5 is more
than once per week. We also include in this scale the value 0 for MPs declaring
they are non-believers as they, by consequence, do not attend religious ser-
vices. We hypothesize that the more religious or closer to the Catholic Church
legislators are, the less they will support for all three issues.
We also add socio-demographic variables regarding levels of education and
income. In both cases, higher responses represent higher levels. Furthermore,
10 | ERLACS No. 114 (2022): July-December
we have a continuous variable for age and a dummy variable for sex
(1=woman). In similar terms, we hypothesize that younger MPs, with higher
income and studies, and women, are more prone to support all three issues.
Given the importance of ideology and political parties, we include these di-
mensions as well in the analysis. Thus, we have two variables for ideological
positions: 1-10 scales for left-right (1=left) and preference for state or market
regulation of the economy (1=state). The latter variable is also helpful to test
whether economic2 and non-economic liberalism are connected to each other.
This is especially relevant since, at least in the United States, they seem to be
defended by opposing parties to the extent that “economic liberalism” has a
different meaning to that adopted in many Latin American and European coun-
tries. Thus, following the reviewed literature and, to a certain extent, the Amer-
ican experience, we expect that leftists and statists will be more supportive of
these three issues. In view of the complexity of including variables related to
political parties in such a heterogeneous database, we decided to include a bi-
nary control variable regarding MPs’ self-stated relationship to the government
(1=opposition, 0=government).
Finally, given the country disparity among existing legislations, we added a
dummy variable that measures the possible effects of the legal status, i.e., an
institutional binary variable in the models of same-sex marriage and abortion
(Appendix II). In each of these cases, we considered as legal (1) only if the
country recognizes same-sex marriage (not civil union) and if there exists any
legal possibility to abort a pregnancy. Since the only case for drug legalization
was marijuana in Uruguay and PELA-USAL asks about all drugs, we did not
extend our operationalization to this variable. We hypothesize that higher sup-
port for these two issues shall be identified in countries where they are legal.
We do not expect to find coefficients moving toward different directions in the
three regressions, that is, positive for one issue, but negative for another. The
reason we choose these three questions is the possibility to assess their causes,
and find out if the drivers of support for same-sex marriage, abortion and drug
legalization are the same. We provide the descriptive statistics of the variables
in Appendix I.
Latin America’s institutional and political framework
During the period when the interviews were fielded, same-sex marriage was
legal in three out of sixteen countries. This number is based on two parameters.
First, it regards a legislation that recognizes the marriage between people from
the same sex. It could be a constitutional right or an infra-constitutional right,
i.e., approved by the Congress. These cases include Argentina and Uruguay.
Second, we considered countries where the Supreme Court has decided, erga
omnes, that same-sex couples could marry. It was the case of Mexico, and, af-
ter the interviews were conducted, Colombia. Mexico represents a hybrid case.
In many states, same-sex marriage is legal. Still, there is no decision in the na-
Asbel Bohigues, João V. Guedes-Neto, Manoel Santos: Latin American political elites’ positions | 11
tional level; in 2015, the Supreme Court has ruled in favour. We are aware that
the second scenario represents different levels of legal certainty. Constitutional
and infra-constitutional rules are more significant. Therefore, we have opera-
tionalized this institutional feature considering both paths, legislative and judi-
ciary.
We operationalized abortion following a single path. We coded a binary
variable setting 1 when there is at least a conditional possibility to abort the
pregnancy and 0 when it is criminalized in every case. Conditions vary widely.
They involve health risks, a genetic malformation of the child, rape, and incest.
We have coded eleven out of sixteen countries as positive. They are Argentina,
Bolivia, Colombia, Costa Rica, Ecuador, Guatemala, Mexico, Panama, Para-
guay, Uruguay, and Venezuela. This operationalization is relevant for two
main reasons. First, it shows that both issues are far from consensual among
Latin American legislations. It differs from the case of drugs, legalized only in
Uruguay and restricted to the use of marijuana. Second, it should exercise a
relevant impact over MPs’ opinion. Depending on the legality of each issue,
MPs will be expressing their opinions toward the change or maintenance of a
given institution.
The reality is that political elites in Latin America have very differing views
about same-sex marriage, abortion, and drug legalization. One aspect that de-
serves to be mentioned, and which proves the sensitivity of these issues, is the
significant lack of coherence among countries, as measured by standard devia-
tion. Figure 1 displays the mean support of each country for all three issues,
with the standard deviations.3 Uruguayan elites are, by far, the most favourable
to these issues, followed by Argentineans, and Mexicans regarding same-sex
marriage; Argentineans, Mexicans, and Chileans in abortion; and Mexicans,
Colombians and Chileans in the legalization of drugs. This group of elites
would be the most tolerant to this social agenda: Uruguay, Argentina, Mexico,
and Chile. On the opposite side, we find Panamanians, Hondurans, and Nicara-
guans. Overall, Central American elites have the most unfavourable attitudes.
It is noteworthy that same-sex marriage is generally more accepted, followed
by abortion and the legalization of drugs. Although Uruguayans score 9.51 in
the first, their rate of approval is reduced to 7.04 when it comes to the legaliza-
tion of drugs. Indeed, only Uruguay and Mexico score higher than 5 for the
legalization of drugs. The same is valid for Uruguay and Argentina in the case
of abortion, and a total of eight countries in regard to same-sex marriage: Uru-
guay, Argentina, Mexico, Chile, Venezuela, Colombia, and Ecuador. Still,
these cross-national comparisons across countries should be read with great
caution, given the uncertainty the standard deviations show.
12 | ERLACS No. 114 (2022): July-December
Figure 1. Mean support (with SD) for same-sex marriage, abortion, and drug legalization per
country
Source: own calculations based on PELA-USAL
Thus, looking at each country separately, the lack of similar views (high stand-
ard deviations) precisely means that non-economic liberal attitudes constitute a
political divide in Latin American politics: There is no general agreement on
abortion, legalization of drugs or same-sex marriage. Therefore, given this dis-
parity not only between countries but also within each country, we expect to
find non-linear effects. The fact that these are controversial issues, in terms of
lack of consensus both between and within countries, we hypothesize that the
effects we find will be stronger in the extreme positions; in our case, strongly
1
2
3
4
5
6
7
8
9
10
UR AR MX CH VZ CO EC CR BO SV PY GU HN NI DR PN
Same sex couple marriage
1
2
3
4
5
6
7
8
9
10
UR AR MX CH VZ CO EC CR BO SV PY GU HN NI DR PN
Abortion
1
2
3
4
5
6
7
8
9
10
UR AR MX CH VZ CO EC CR BO SV PY GU HN NI DR PN
Drug legalization
Asbel Bohigues, João V. Guedes-Neto, Manoel Santos: Latin American political elites’ positions | 13
opposed and strongly in favour. We thus chose to run ordered logistic regres-
sions instead of linear regressions. The former works the same as logistic re-
gressions, but instead of just two values (0-1), ordered logistic regression con-
siders as many values as the response scale has, and the change from one posi-
tion to the following (1 to 2, 2 to 3, and so on). Once these general results are
presented, we then turn to the two main predictors, ideology and religiosity, to
explore on how their substantive impact varies depending on the position in the
three scales regarding same-sex marriage, abortion and drug legalization, with
a special focus on extreme positions (1 and 10 in the scales), in order to verify
the hypothesis.
Results
Table 2 displays the multi-level ordered logistic regressions, one for each ques-
tion.4 To avoid any post-treatment bias, and following Campbell et al.’s (1960)
“funnel of causality”, for each DV we first include sociodemographics (gender,
age, studies, income), and then attitudes (ideology, state/market, trustworthi-
ness of the Catholic church, religiosity). By doing this we can observe how
each group interacts with the dependent variable.
Gender. It is significant for SSM in both models, and significant for
abortion only when attitudes are taken into account. Also, it is possible
to observe the absence of a gender-effect in the legalization of drugs.
In other words, women in Parliaments are more progressive than men
in same-sex marriage and abortion, but both men and women have the
same attitude toward the legalization of drugs.
Age. We also identify a consistent age effect in the cases of SSM and
drug legalization, with the notable absence of significance for abor-
tion: Age differences apparently do not discriminate support for abor-
tion among Latin American elites.
Studies and income. As expected too, these two have positive coeffi-
cients, but limited to some scenarios: Studies are drivers in SSM, but
not for abortion and drug legalization. For its part, income is signifi-
cant for SSM and abortion when attitudes are excluded, but loses its
significance when these are added. We see the reverse situation for
drug legalization.
Thus, looking at the complete models, with both sociodemographics and atti-
tudes, the only sociodemographic variable significant for abortion is gender,
while for drug legalization does not yield statistical significance along with
studies. As for SSM, all independent variables except income have significant
coefficients.
Ideology and religion. Turning to attitudes, these two emerge as the
main predictors, with significant negative coefficients. Rightists and
more religious MPs are less supportive in all three issues. The per-
ceived trustworthiness of the Catholic Church is also significant, in the
14 | ERLACS No. 114 (2022): July-December
expected direction (negative), but not for same-sex marriage. Surpris-
ingly, at least based the literature review, the state-market variable is
rather negligible, in terms of statistical significance. Interestingly, non-
economic liberal values do not covary with economic liberalism. The
economic agenda in Latin America is not linked, in these terms, to this
more social agenda.
Table 2. Determinants of support for SSM, abortion and drug legalization
Source: own elaboration from PELA-USAL (*p ≤0.1; ** p ≤0.05; *** p ≤0.01. Mixed-
effects ordinal logistic regressions with robust standard errors in parentheses. We
rely on the Stata routine meologit.)
Opposition. Although opposition was included as a control variable, we
want to highlight the absence of significant differences between govern-
ment and opposition in all three cases.
Institutional framework. As for the impact of the institutional framework,
it is mixed. The legal status is a strong predictor in same-sex marriage.
Once there is legal recognition, MPs are far more favourable than in the
countries where it is illegal. Nonetheless, there is no such legal effect at all
on abortion. It is true that the relationship could suffer from reverse causal-
ity: countries with MPs more supportive of same-sex marriage are more
prone to approve it. Still, this covariance does not exist for abortion.
Nevertheless, beyond the absence of effects and the “funnel of causality”, all
independent variables have the theoretically expected effects, and there is no
different direction depending on the issue. That being so, both ideology and
religiosity are the most salient attitudinal determinants in comparative perspec-
tive: These two variables5 are key predictors in all three issues. Perceived
Asbel Bohigues, João V. Guedes-Neto, Manoel Santos: Latin American political elites’ positions | 15
trustworthiness of the Catholic Church for same-sex marriage, gender for drug
legalization, and age and studies for abortion are not significant. Thus, we now
turn to these two salient variables to assess their different substantive impacts
depending on the positions in the scale from 1 (disapproves/against) to 10 (fa-
vourable/for). To do so we display on Figure 2 the marginal effects of these
two variables in the multi-level ordered logistic regressions (Stata command
margins). Due to the ordinal nature of the DV (ten-point scales) we ran ordered
logits; thus, when looking at marginal effects we can identify how ideology
and religiosity predict each one of the 10 possible positions in the scales on the
support for same-sex marriage, abortion and drug legalization. As aforemen-
tioned, we expect stronger effects in the extreme positions. For the ease of in-
terpretation, Figure 2 only displays values 1, 5 and 10 of the three questions,
although the remaining values follow similar patterns. The results are suffi-
ciently clear, both statistically and visually, to confirm our hypothesis: There
are stronger effects in the extreme positions, but with a key nuance: only on
rightist and very religious positions.
Each slope represents the probability of choosing positions 1 (disap-
prove/against), 5, and 10 (approve/for) for all positions on the left-right and
religiosity axes. The probability in the Y-axes refers to how likely MPs are on
each ideological and religious position to choose each of the three values here
displayed: 1-5-10. In the case of ideology (graphs on the left), it is thus possi-
ble to visualize the probability of choosing 1, 5 or 10 on leftist positions, and
how probability does change on the most rightist position, 10. As an example,
in the case of abortion, the probability of choosing 1-Against rises from over
30 percent in the extreme Left (1) to almost 70 percent in the Right (10). In
similar terms, the probability of choosing 10-For drops from almost 12 percent
in the Left to barely 3 percent in the Right. Put it differently: A legislator on
the 1-Left has a 12 percent likelihood to be totally in favour of abortion, com-
pared to a 3 percent for his/her peer on the 10-Right.
From Figure 2 we can see that the ideological differences are significantly
intense not in both extreme positions, as expected, but on extreme contrary
positions. The change in the probability of being totally against these issues,
that is choosing 1-Against, between the extremes of the Left-Right axis rises to
25 points for SSM, 40 for abortion and 20 for drug legalization. As expected,
the most substantive impact is limited to strongly opposed views. We do find
an ideological effect, but even the left is not that supportive of these three is-
sues. Otherwise, at least the slope of “Approve/For (10)” would be totally op-
posed to the slope of “Disapprove/Against (1);” i.e., considerably higher prob-
ability of being in favour in the left. Therefore, given what marginal effects
display, we can see how ideology and religiosity work when predicting the
support for SSM, abortion and drug legalization: slightly favourable positions
on the left, and much stronger opposition on the right.
16 | ERLACS No. 114 (2022): July-December
Figure 2. Marginal effects of ideology and religiosity: predicted probabilities for same-sex
marriage, abortion and drug legalization
Source: own elaboration. (Adjusted predictions with 95 per cent of confidence intervals.
Command margins in Stata)
As for religiosity (graphs on the right), the patterns are revealingly similar. The
most important religious differences are limited to extreme unfavourable opin-
ions: In the Left, the probability of being totally in favour of SSM, abortion,
and drug legalization is much lower than the probability of being totally
against on the right. Once again, we underline that there is a significant religi-
osity effect on all positions, from Against-1 to For-10, displayed in Table 2,
but the substantive impact of religiosity is mainly determined by the high lev-
els of rejection among the most religious MPs. As with ideology and leftists,
we cannot state that non-believers are fierce supporters of these issues. If that
were the case, Figure 2 would display sharp slopes for totally favourable (10),
with high values on the left side, and very low values on the right side. We
Asbel Bohigues, João V. Guedes-Neto, Manoel Santos: Latin American political elites’ positions | 17
certainly find both ideology and religiosity effects in all three questions: The
more leftist or the less religious, the higher the probability to approve/be in
favour. However, the marginal effects on Figure 2 allow us to nuance the find-
ings: the overall effects are mainly determined by the strong opposition on the
right/religious MPs, always above 50 percent or even 70 percent. What we find
for the left and less religious MPs are much lower values of support, always
below 30 percent, even 10 percent.
Conclusion
This paper has assessed the predictors of MPs’ opinions toward same-sex mar-
riage, abortion, and drug legalization. We find that the most important predic-
tors are ideology and religiosity; the more MPs attend religious services and
the more rightist they are, the less prone to support same-sex marriage, abor-
tion, and the legalization of drugs they are. Interestingly, the influence of the
Catholic Church vanishes when MPs express their opinions on same-sex mar-
riage. Ideology helps to better understand the region’s political scenario in
many cases. Leftists are generally more favourable toward non-economic lib-
eral issues. The Left-Right divide still works in Latin America on these issues.
It is important to notice that this ideological divide does not hold when we con-
sider MPs’ positions toward market regulation. This independent variable has
no significance in predicting non-economic liberal values.
However, the main findings and contributions of this paper are the substan-
tive impacts of ideology and religiosity. Although we find a significant general
effect, we discern some key elements on how the relationships work. This gen-
eral effect is mostly due to the substantive differences between left/right and
non-believer/religious in totally against positions. In the rest of the positions
there are significant differences, but comparatively less substantive. Further-
more, even among leftists and non-believers totally against positions are much
likely. We do not find substantive differences in most supportive opinions. In-
deed, we find significant statistical differences at conventional levels, but neg-
ligible substantive differences when put in comparative perspective.
These results help to explain and understand the situation of same-sex mar-
riage, abortion, and drug legalization in the region. For these to be widely ac-
cepted and normalized, elite support is a key condition (Bohigues & Fernán-
dez-i-Marín, 2022). Rightist and religious elites are against, something we al-
ready knew, and this paper confirms. The surprising finding, however, regards
the opposing pole: Leftists and non-religious. Certainly, as expected, there is
an effect of ideology and religion among them, but this does not mean leftists
and non-religious MPs are resolutely supportive of same-sex marriage, abor-
tion and drug legalization. Indeed, it is more appropriate to state that leftist and
non-believer MPs are less unfavourable than rightists and religious MPs are. If
the former are slightly in favour, and the latter are strongly opposed, the pano-
rama seems less promising for these policies than a superficial analysis of the
18 | ERLACS No. 114 (2022): July-December
effects of ideology and religion might suggest. In light of these findings, and
how ideology and religiosity actually work, even with leftist majorities, away
from religious influences, the evidence from elite attitudes here analysed sug-
gests that the chances to pass progressive bills on any of these issues will re-
main low. Higher than in cases of religious rightist majorities, no doubt, but
low at the end of the day.
* * *
Asbel Bohigues is an Assistant Professor of Political Science at the University
of Valencia. He holds a PhD in Political Science from the University of Sala-
manca. His research lines include comparative politics, democracy, and elites,
with a regional specialization in Latin America.
Address: Universitat de València, Facultat de Dret, Av. dels Tarongers, S/N, 46022, Valen-
cia, Spain.
Email: asbel.bohigues@uv.es
João V. Guedes-Neto is an Assistant Professor in the School of Criminal Jus-
tice and Public Administration at Kean University. His main area of expertise
is intergroup behavior.
Address: 1011 Morris Ave, Union, NJ, United States. 07083.
Email: jguedesn@kean.edu
Manoel Leonardo Santos is a Professor at the Universidade Federal de Minas
Gerais. He holds a Master’s and PhD in Political Science from the Federal
University of Pernambuco/University of Salamanca.
Address: Av. Antônio Carlos, 6.627, Campus Universitário-FAFICH/DCP, 4o andar. Sala
4085. Pampulha, Belo Horizonte/MG CEP. 31.270-901.
Email: mlwds@hotmail.com
Notes
1 For more information see: https://oir.org.es/pela/
2 We understand that economic liberalism may have different meanings depending on,
among others, country and theoretical family. Here, when we speak of economic liberal-
ism, we refer to classical liberal (or pro-market) ideas.
3 Lack of coherence is even more striking among political parties (Appendix III). Among
the most favourable parties to same-sex marriage there are two right-wing parties: Par-
tido Nacional (PN) from Uruguay and Voluntad Popular (VP) from Venezuela. The oth-
ers are left-wing parties from the Southern Cone. The most unfavourable parties are all
from Central America and non-leftist. Among the most favourable parties to legalization
of drugs, we find Partido Colorado (Uruguay), Partido Patriota (Guatemala), and Partido
Asbel Bohigues, João V. Guedes-Neto, Manoel Santos: Latin American political elites’ positions | 19
Liberal (Colombia). The remaining ones are left-wing parties from Chile, Uruguay, and
Mexico. The most unfavourable are right-wing and religious parties, plus a left-wing
former guerrilla party, the Frente Sandinista de Libertación Nacional (FSLN). It shares
positions with traditional parties, as Partido Liberal de Honduras (PLH) and Partido Un-
idad Social Cristiana (PUSC, Costa Rica). Finally, only left-wing parties are found
among the positive toward the legalization of abortion. They are from Mexico, Chile,
Argentina, and Uruguay. In the opposite side, only right-wing parties are found, spread
all over Latin America.
4 The models include the cases with no missing values (respondents who chose not to
answer) on every variable, hence the differences between Tables 2 and 3 of around 100
cases.
5 Correlation between ideology and religiosity is 0.37.
References
Adams, G. D. (1997). Abortion: Evidence of an issue evolution. American Journal of
Political Science 41(3), 718–37.
Alcántara M. (2013). Opinión sobre el aborto desde la perspectiva de los legisladores
latinoamericanos. Boletin Elites Parlamentarias Latinoamericanas 54, 1–6.
Alcántara M. and Rivas C. (2013). El debate sobre la legalización de las drogas desde la
perspectiva de legisladores latinoamericanos. Boletin Elites Parlamentarias
Latinoamericanas 53, 1–5.
Álvarez R. M. & Brehm J. (1995). American ambivalence towards abortion policy:
Development of a Heteroskedastic probit model of competing values. American Journal
of Political Science 39(4), 1055–1082.
Andersen R. & Fetner T. (2008a). Cohort Differences in tolerance of homosexuality:
attitudinal change in Canada and the United States, 1981-2000. Public Opinion
Quarterly 72(2), 311–30. https://doi.org/10.1093/poq/nfn017
_____ (2008b). Economic inequality and intolerance: Attitudes toward homosexuality in 35
democracies. American Journal of Political Science 52(4), 942–958.
Baunach D. M (2012). Changing same-sex marriage attitudes in America from 1988 through
2010. Public Opinion Quarterly 76(2), 364–378.
Bohigues, A. & Fernández-i-Marín, X. (2022). Drug legalization, elite support, and
government effectiveness in Latin America. Journal of Drug Issues 52(2), 250-267.
Boidi, M. F. (2013). Same-sex marriage in Uruguay: A new law in line with citizens’
preferences. AmericasBarometer Topical Brief (6), 1–2.
Boidi, M. F. & Corral M. (2013). Public opinion and abortion rights in the Americas.
AmericasBarometer Topical Brief (8), 1–3.
Boidi M F, Queirolo R. & Cruz J. M. (2015). Uruguayans are skeptical as the country
becomes the first to regulate the marijuana market. AmericasBarometer Topical Brief
20, 1–4.
Campbell, A., Converse, P. E., Miller, W. E., & Stokes, D. E. (1960). The American voter.
John Wiley.
Conover P. J. & Feldman S. (1981). The origins and meaning of liberal/conservative self-
identification. American Journal of Political Science 25(4), 617–645.
Deegan-Krause K. (2007). New dimensions of political cleavage. R. Dalton & H. D.
Klingemann (eds.), Oxford Handbook of Political Behavior (538–556). Oxford
University Press.
20 | ERLACS No. 114 (2022): July-December
Díez J. (2010). The importance of policy frames in contentious politics: Mexico’s national
antihomophobia campaign. Latin American Research Review 45(1), 33–54.
Dion M. & Díez J. (2017). Democratic Values, Religiosity, and Support for Same-Sex
Marriage in Latin America. Latin American Politics & Society 59(4), 75–98.
Dolan K. (1998). Voting for women in the ‘Year of the Woman’. American Journal of
Political Science 42(1), 272–93.
Ellison C. G., Acevedo G. A. & Ramos-Wada A. I. (2011). Religion and attitudes toward
same-sex marriage among U.S. latinos. Social Science Quarterly 92(1), 35–56.
Encarnación O. G. (2011). Latin America’s gay rights revolution. Journal of Democracy
22(2), 104–18.
Franceschet S. (2003). ‘State feminism’ and women’s movements: The impact of Chile’s
Servicio nacional de la mujer on women’s activism. Latin American Research Review
38(1), 9–40.
Guedes-Neto, J. & Bohigues, A. (2018). Status quo characteristics or private sector affinity?
Explaining preferences for safety nets in Latin American legislatures. Teoria e
Sociedade 26(2), 15–32.
Hahner J. E. (1985). Recent Research on women in Brazil. Latin American Research Review
20(3), 163–79.
Hawkins K. & Morgenstern S. (2010). Ideological cohesion of political parties in Latin
America. H. Kitschelt, K. Hawkins, J. Luna, G. Rosas, & E. Zechmeister (eds.), Latin
American Party Systems (145–168). Cambridge University Press.
Herek G. (1987). Religious orientation and prejudice: A comparison of racial and sexual
attitudes. Personality and Social Psychology Bulletin 13(1), 34–44.
Huddy, L. & Terkildsen N. (1993). Gender stereotypes and the perception of male and
female candidates. American Journal of Political Science 37(1), 119–47.
Jennings, M. K. & Farah B. G. (1981). Social roles and political resources: An Over-time
study of men and women in party elites. American Journal of Political Science 25(3),
462–482.
Kampwirth K. (2008). Abortion, Antifeminism, and the return of Daniel Ortega: In
Nicaragua, leftist politics?. Latin American Perspectives 35(6), 122–136.
Kaufmann K. & Petrocik J. R. (1999). The changing politics of American men:
Understanding the sources of the gender gap. American Journal of Political Science
43(3), 864–887.
Knoke D. (1979). Stratification and the dimensions of American political orientations.
American Journal of Political Science 23(4), 772–791.
Kollman K. (2007). Same-sex unions: The globalization of an idea. International Studies
Quarterly 51(2), 329–357.
Kreitzer R. J. (2015). Politics and morality in state abortion policy. State Politics and Policy
Quarterly 15(1), 41–66.
Lax J. R. & Phillips J. H. (2009). Gay rights in the states: Publi opinion and policy
responsiveness. American Political Science Review 103(3), 367–86.
Lipset S. M. (1959). Democracy and working-class authoritarianism. American Sociological
Review 24(4), 482–501.
Lodola G. & Corral M. (2010). Support for same-sex marriage in Latin America.
AmericasBarometer Insights 44, 1–9.
Londoño E. (2017). Uruguay’s marijuana law turns pharmacists into dealers. New York
Times. https://www.nytimes.com/2017/07/19/world/americas/urugua-legalizes-pot-
marijuana.html.
Lupu N. (2016). Party brands in crisis: partisanship, brand dilution, and the breakdown of
political parties in Latin America. Cambridge University Press.
Mainwaring S. & Wilde A. (1989). The progressive Church in Latin America. University of
Notre Dame Press.
Asbel Bohigues, João V. Guedes-Neto, Manoel Santos: Latin American political elites’ positions | 21
Maldonado A. (2015). Same-sex marriage resonates most strongly with young people in the
Americas. AmericasBarometer Topical Brief 21, 1–5.
Marcano I. (2013). Evangelism and gay rights in Latin America. AmericasBarometer
Insights 94, 1–11.
Mason, L. (2018). Uncivil agreement: How politics became our identity. University of
Chicago Press.
Mazur A. G., McBride D. E. & Hoard S. (2016). Comparative strength of women’s
movements over time: Conceptual, empirical, and theoretical innovations. Politics,
Groups, and Identities 4(4), 652–676.
Miranda L. (2011). What do people on the right think? Citizen and Elite values and opinions
in Latin America. Boletin PNUD & Instituto de Iberoamérica 4, 1–16.
Montalvo D. & Saunders E. (2015). Low support for homosexuals’ political rights is
commonplace in the Caribbean. AmericasBarometer Topical Brief 23, 1–4.
Mooney C. Z. & Lee M-H. (1995). Legislative morality in the American states: The case of
pre-roe abortion regulation reform. American Journal of Political Science 39(3), 599–
627.
Murillo A. (2018). La corte interamericana falla a favor del matrionio igualitario. El País.
https://elpais.com/internacional/2018/01/10/mexico/1515624521_957175.html.
Olson L. R., Cadge W. & Harrison J. T. (2006). Religion and Public opinion about same-sex
marriage. Social Science Quarterly 87(2), 340–360.
Orcés D. (2013). Legislators and the president clash in Ecuador over abortion issue: Who
has the support of the people? AmericasBarometer Topical Brief 11, 1–4.
Pribble J. (2006). Women and welfare: The politics of coping with new social risks in Chile
and Uruguay. Latin American Research Review 41(2), 84–111.
Rosas G. (2010). Issues, ideologies, and partisan divides: imprints of programmatic structure
on Latin American Legislatures. H. Kitchelt, K. Hawkins, J. Luna, G. Rosas, E.
Zechmeister (eds.), Latin American Party Systems (70–95). Cambridge University Press.
Santos C. M. (2004). En-gendering the police: Women’s Police stations and feminism in
São Paulo. Latin American Research Review 39(3), 29–55.
Schlozman K. L., Burns N., Verba S. & Donahue J. (1995). Gender and citizen
participation: Is there a different voice?. American Journal of Political Science 39(2),
267–293.
Sherkat D. E., De Vries K. M. & Creek S. (2010). Race, Religion, and opposition to same-
sex marriage. Social Science Quarterly 91(1), 80–98.
Soule S. A. (2004). Going to the chapel? Same-sex marriage bans in the United States,
1973-2000. Social Problems 51(4), 453–477.
Tedin K. L. (1980). Assessing peer and parent influence on adolescent political attitudes.
American Journal of Political Science 24(1), 136–154.
Trejo G. (2009). Religious competition and ethnic mobilization in Latin America: Why the
Catholic Church promotes Indigenous movements in Mexico. American Journal of
Political Science 103(3), 323–343.
Vaggione J. M. (2011). Sexual rights and religion: Same-sex marriage and lawmakers’
Catholic identity in Argentina. University of Miami Law Review 65(935), 935–954.
Viterna J. (2012). The left and ‘life’ in El Salvador. Politics & Gender 8(2), 248–254.
Welzel, C. (2013). Freedom rising: Human empowerment and the quest for emancipation.
Cambridge University Press.
22 | ERLACS No. 114 (2022): July-December
Appendix
I. Descriptive statistics from PELA-USAL
Variable
Observations
Mean
SD
Min.
Max.
Ideology
1,080
4.82
2.42
1
10
State-Market
1,103
5.48
2.54
1
10
Trust Catholic Church
1,106
3.01
0.89
1
4
Religiosity
1,060
2.60
1.52
0
5
Opposition
1,050
0.55
0.50
0
1
Woman
1,098
0.29
0.45
0
1
Age
1,054
47.64
10.96
21
79
Studies
1,088
5.04
0.96
1
6
Income
1,056
2.16
1.09
1
4
SSM legal
1,124
0.21
0.41
0
1
Abortion legal
1,124
0.71
0.45
0
1
SSM
1,074
4.83
3.68
1
10
Abortion
1,070
3.88
3.29
1
10
Drugs
1,070
3.60
3.09
1
10
II. Operationalization of institutional variables based on the legal framework of each country
Countries
Same-Sex Marriage
Abortion
Argentina
Yes (1)
Yes (1)
Bolivia
No (0)
Yes (1)
Chile
No (0)
No (0)
Colombia
No* (0) At the moment the
interviews, not legal
Yes (1)
Costa Rica
No (0)
Yes (1)
El Salvador
No (0)
No (0)
Ecuador
No (0)
Yes (1)
Guatemala
No (0)
Yes (1)
Honduras
No (0)
No (0)
Mexico
Yes (1)
Yes (1)
Nicaragua
No (0)
No (0)
Panama
No (0)
Yes (1)
Paraguay
No (0)
Yes (1)
Dominican R.
No (0)
No (0)
Uruguay
Yes (1)
Yes (1)
Venezuela
No (0)
Yes (1)
Asbel Bohigues, João V. Guedes-Neto, Manoel Santos: Latin American political elites’ positions | 23
III. Average support for same-sex marriage, abortion, and drug legalization per political
party based on PELA-USAL
Same sex marriage
Abortion
Drugs legalization
Mean (SD)
N
Mean (SD)
N
Mean (SD)
N
Argentina
FpV
9.38 (1.48)
19
5.73 (2.49)
15
7.92 (3.2)
18
UCR
7.38 (3.07)
11
4.19 (3.42)
10
5.8 (4.05)
9
PJ
6.89 (3.91)
13
3.5 (2.62)
11
3.13 (3.33)
11
PRO
7 (2.89)
3
2.33 (1.34)
3
1.5 (1.36)
3
Bolivia
MAS
3.27 (3.26)
63
1.98 (2.09)
61
3.42 (3.12)
61
PDC
4.5 (3.41)
7
1.75 (1.5)
7
2.63 (3.32)
7
UD
4.82 (2.89)
23
2.5 (2.48)
23
3.32 (2.7)
23
Chile
PC
10 (0)
4
6.33 (1.44)
4
10 (0)
4
PS
8.43 (2.58)
9
5.4 (2.96)
6
7.71 (1.85)
9
PPD
7.12 (4.05)
8
6.5 (3.7)
7
7.38 (3.05)
7
PRSD
7 (4.37)
3
5.5 (3.2)
3
6.75 (2.28)
3
PDC
7.18 (3.3)
11
5.55 (2.83)
11
5.09 (2.5)
11
RN
4.5 (3.81)
9
2.2 (1.63)
9
1.4 (0.71)
9
UDI
2.88 (1.62)
16
2.47 (2.12)
16
1.12 (0.33)
16
Colombia
PL
5.75 (3.66)
20
6.25 (3.35)
16
4.69 (2.93)
16
P de la
U
4.85 (3.74)
19
4.54 (3.72)
17
4.62 (3.48)
17
CR
6.86 (2.93)
8
5.25 (3.12)
7
4 (2.73)
6
PC
2.92 (3.05)
14
3.85 (2.95)
12
3.42 (2.32)
11
CD
5.1 (3.74)
9
2.9 (3.29)
9
3.3 (3.7)
9
Costa Rica
FA
7.44 (3.88)
9
6.11 (3.58)
9
5.78 (3.86)
9
PAC
5.67 (3.72)
12
2.45 (2.54)
11
3.25 (2.52)
12
PUSC
1.25 (0.46)
8
1.5 (0.53)
8
1.25 (0.46)
8
PLN
2.44 (2.38)
18
1.94 (1.69)
18
2.06 (1.51)
18
Dominican
Republic
PLD
1.96 (1.68)
34
2.96 (2.73)
33
4.52 (2.88)
33
PRM
3.78 (3.58)
15
3.22 (2.64)
15
4.43 (2.98)
15
PRD
1.67 (1.15)
5
1.67 (1.14)
3
4.33 (3.03)
3
PRSC
2.5 (1.77)
4
3.5 (1.95)
4
1 (0)
4
Ecuador
Alianza
-PAIS
5.53 (3.48)
43
3.83 (2.97)
42
4.15 (3.38)
41
SUMA-
CREO
5.33 (3.83)
23
2.59 (2.15)
22
3.18 (2.15)
22
PSC
4 (5.19)
3
2.67 (2.88)
3
1 (0)
3
El Salvador
FMLN
5.94 (3.24)
18
4 (2.97)
17
6.56 (2.06)
18
ARENA
2.1 (1.92)
24
2.52 (2.42)
21
1.78 (1.54)
23
PCN
1 (0)
4
1 (0)
2
1 (0)
3
24 | ERLACS No. 114 (2022): July-December
GANA
1 (0)
9
2.25 (2.55)
8
1 (0)
8
Guatemala
PP
1.8 (1.78)
5
6.8 (4.44)
5
4.4 (1.94)
5
UNE
4.18 (3.61)
18
3.65 (2.93)
17
3.41 (2.94)
17
TODO
S
3 (3.27)
10
2.56 (1.74)
9
2.78 (2.59)
9
LIDER
2.38 (2.47)
17
2.44 (2.16)
16
2.63 (2.85)
16
FCN
2.83 (2.99)
7
3.67 (2.66)
6
1.5 (1.22)
6
UCN
1.5 (0.577)
5
2.75 (2.36)
4
1 (0)
4
Honduras
LIBRE
3.92 (2.87)
25
3.08 (2.64)
25
3.32 (2.35)
25
PAC
2.44 (1.66)
9
3.33 (1.87)
9
2.33 (2)
9
PNH
2.5 (2.76)
30
2.07 (2.5)
30
2.03 (1.96)
30
PLH
1.76 (1.68)
17
1.59 (1.33)
17
2 (1.12)
17
Mexico
MORE
NA
9 (2.23)
5
7 (1.225)
5
8.6 (2.19)
5
PRD
8.27 (2.69)
12
7.36 (2.73)
11
8.55 (2.02)
11
PRI
7.33 (2.74)
50
4.91 (2.88)
47
5.02 (2.97)
47
Verdes
7 (4.08)
4
4.25 (2.28)
4
4.25 (4.03)
4
PAN
4.82 (3.09)
23
3.61 (2.37)
23
2.13 (2.24)
23
Nicaragua
PLI
2.24 (1.93)
15
2.44 (2.26)
14
2.71 (2.40)
15
FSLN
2.94 (2.73)
36
1.55 (1.41)
35
2.69 (2.4)
34
Panama
PRD
2.06 (2.16)
18
1.78 (1.83)
18
1.94 (2.18)
18
Panam
eñista
2 (1.73)
9
1.89 (1.76)
9
1.78 (1.56)
9
CD
1.39 (1.24)
18
1.44 (1.25)
18
1.67 (1.57)
18
Paraguay
PC-
ANR
2.35 (2.04)
28
2.81 (2.7)
26
1.73 (1.22)
26
PLRA
2.82 (2.96)
19
2.5 (2.36)
18
1.72 (1.49)
18
Uruguay
FA
9.84 (0.82)
36
8 (2.33)
36
9.55 (1.70)
36
PC
10 (0)
9
7.25 (3.54)
9
6.43 (4.46)
8
PN
8.7 (2.751)
22
3.95 (3.88)
22
4.95 (3.48)
22
Venezuela
VP
8.56 (2.11)
6
5.89 (2.19)
5
5 (2.98)
5
PSUV
4.29 (2.97)
20
2.29 (1.8)
20
4.35 (3.65)
20
PJ
6.81 (3.31)
13
4.87 (2.89)
13
3.94 (3.15)
13
AD
4.7 (3.68)
10
2.7 (2.21)
10
3.4 (3.02)
10
UNT
7.71 (2.11)
8
3.43 (2.47)
8
3 (3.01)
8
