Does Religion Matter? Public Opinion on Social Issues in Costa Rica

opinions of Costa Rican population regarding various social problems and behaviors were collected. Such information was related, in most cases, to sexual and reproductive rights. It can be concluded that Costa Rican population behaves as predicted by the academic literature on this subject. People who profess an Evangelical Pentecostal belief or who indicate having a high religiosity, are more likely to adopt a non-favorable position towards sexual and reproductive rights. However, considering analyzed data, a relationship between the religious beliefs of the Costa Rican population and their perception regarding various social problems cannot be established.


Introduction
In 2018, an Evangelical Pentecostal candidate, Fabricio Alvarado, obtained the most votes in the first round of the presidential elections. He achieves this with a campaign based on opposition to the rights of the sexually diverse population and without any strong political proposals on other social problems affecting the country. Historically, Costa Rica has been a Roman Catholic country, however religious affiliation was not a significant variable to explain the political behavior of the population (Steigenga 2001). Therefore, the results of the 2018 elections raised the question of whether the religious affiliation of the Costa Rican population is currently a variable that allows explaining their behavior and political preferences.
This article seeks to determine the existence or absence of a relationship between a) religious beliefs of Costa Ricans and their perceived affectation by various social problems and b) religious beliefs and Costa Rican's acceptance of different social behaviors, especially those related to sexual and reproductive rights. Therefore, it is sought to corroborate whether religious beliefs function as a predictive variable on these issues and, thus, observe its usefulness in the understanding of the socio-political behavior of Costa Rican citizens.
To achieve the stated objective, data from the survey Perception of the Costa Rican population on religious values and practices, applied in October 2018 by Instituto de Estudios Sociales en Población (IDESPO) of the Universidad Nacional (UNA), was used to prepare a suitable model to verify if the religious belief of the population works as a predictive variable of their opinion regarding various problems and social behaviors.

Religion Affiliation and Public Opinion in the Academic Literature
This article is an exploratory study since religious affiliation is not considered a significant variable to explain political behavior and public opinion in the Costa Rican academic literature. In this section, a review of the academic literature is conducted to identify how other studies analyze the relationship between religious affiliation and the perception of social problems and behaviors. Therefore, the purpose of the review is to serve as an analysis guide for the data collected in the study.
The academic literature has found a relationship between religious beliefs and the opinion about different issues and social problems. It is necessary to note that investigations of this type are divided into two groups: the first one studies how the religious beliefs of the population are linked to their position on issues related to sexual and reproductive rights (abortion, same-sex marriage, etc.); the second group studies the relationship between people's religious beliefs and their opinion on different social issues and problems (health, climate change, etc.).
Regarding the first group, various studies have shown that religion is one of the strongest predictors to determine the acceptance of sexually diverse people's rights. Thus, these studies show that people who declare themselves Evangelical Protestants, as well as those who claim to be very religious or who express theologically conservative or fundamentalist positions, tend to strongly oppose the rights of sexually diverse people, including the adoption of children by same-sex couples (Adamczyk & Pitt 2009). Based on data obtained from the fourth wave of the World Values Survey and Hierarchical Modeling techniques, Adamczyk and Pitt (2009) find evidence supporting the micro and macro effects of religion and a survival vs. self-expressive cultural orientation. Furthermore, they stated that personal religious beliefs have a greater effect on attitudes about homosexuality in countries like the United States, which have a strong self-expressive cultural orientation. Olson, Cadge and Harrison (2006) also analyzed data from United States, and determined that religious variables play powerful roles in structuring attitudes about same-sex unions. These authors conclude that Non-Protestants are much more likely to support same-sex unions than Protestants and individuals with conservative attitudes toward morality and secularism. To a lesser extent, those who actively participate in religious life are more likely to oppose such unions (Olson et al. 2006). Perry and Whitehead (2016) also studied the impact of religion on Americans' attitudes toward same-sex practices. Their analysis demonstrates that public opinion towards homosexual relationships strongly relates to religious practices and theological conservatism. The authors indicate that frequent religious practices and conservative theological beliefs about the Bible tend to associate negative attitudes toward same-sex relationships for evangelicals, compared to mainline Protestants and, to a lesser extent, Catholics.
Various studies have demonstrated a strong relationship between religious beliefs and their stance on abortion, in vitro fertilization (IVF) and stem cell research. As in the previous case, people who identify themselves as Evangelical Protestants, or who claim to follow fundamentalist theological positions, are more likely to oppose such issues (Schmidt 2018;Simon 2017). Allum et al (2017) examined international public opinion towards stem-cell research during the period when the issue was at its most contentious. Their draw upon representative sample surveys in Europe and North America, fielded in 2005 and find that most people in Europe, Canada and the United States supported stem-cell research, providing it was tightly regulated, but that key differences existed between geographical regions in the relative importance of different types of ethical position. In the United States, moral acceptability was more influential as a driver of support for stem-cell research; in Europe, the perceived benefit to society carried more weight; and in Canada the two were almost equally important. Allum et al (2017) also found that public opinion on stem-cell research was more strongly associated with religious convictions in the United State than in Canada and Europe, although many strongly religious citizens in all regions approved of stemcell research.
In the specific case of Latin America, affiliation to evangelical Protestant churches is a better predictor than Catholic affiliation, to determine people's opposition to issues related to sexual and reproductive rights. This difference is explained by the fact that members of Evangelical Protestant churches are involved in stronger processes of socialization, facilitating the assimilation and adaptation of their opinions to the values and ideas expressed by their religious groups on these issues (Jelen et al. 2017). Farias Aguilera (2016) study the public opinion about the same-sex marriage in Chile; using a multivariate analysis, he determines that age was the best predictor -support was higher at an earlier age-follow by evangelical variable -higher reject-. His works emphasizes the evangelical variable is a better predictor for the support of the samesex marriage that other sociodemographic and political variables.
The second group of investigations aimed to identify how the religious beliefs of the population influence their opinion on different social problems. Opinion studies conducted in various countries have shown that evangelical Protestants are less likely to accept that climate change is a product of human actions (anthropogenic), compared to people of other religious beliefs (Hand and Crowe 2012). Pepper and Leonard (2016) used data from the 2011 Australian National Church Life Survey to examine the beliefs of Australian churchgoers from some 20 denominations about climate change-whether it is real and whether it is caused by humans-and political factors that explain variation in these beliefs. These authors found that Pentecostals, Baptist and Churches of Christ churchgoers, and people from the smallest Protestant denominations were less likely than other churchgoers to believe in anthropogenic climate change and voting and hierarchical and individualistic views about society predicted beliefs.
On the other hand, Kilburn (2014) used data collected in 2008 from a panel survey of voting-age U. S. citizens to analyze religious foundations of American opinion towards climate change. This work shows that religion influence is multifaceted. Biblically literalist beliefs and to a lesser extent attendance at religious services, but not nominal religious affiliations, led Americans to view climate change as mostly natural and to express less concern over its consequences. Yet, evangelical Protestants reporting frequent service attendance were distinctively more concerned about the effects of climate change.
Other research has focused on determining how people's religious beliefs relate to issues concerning public health (Rozier 2017), education (Jelen 2009), the solution of social conflicts (Creel 2018;Díaz López & Negredo 2016) and international relations (Baumgartner et al. 2008;Guth 2009). Although diverse results have been obtained, these studies tend to indicate that religion is a predictive variable of the population's opinion on these issues; especially in cases where people say they attend an evangelical Protestant church or indicate they have fundamentalist religious positions. Additionally, other studies have suggested that people who affirm to have high religiosity tend to defend the participation of churches and religious groups in the discussion on public issues (Pieterse 2014;Walhof 2013). For example, Freeman and Houston (2018) examined United States public opinion on the role of religion in public policy debate using data from three General Social Surveys. Their paper focuses on attitudes toward the behavior of religious leaders, specifically how they influence people to vote and whether religious leaders should try to influence government decisions. Results from the survey indicate that there is public uneasiness with the prominent role that religion is playing in politics, which consequently informs public opinion.
Buckley (2019) used blending data from the Religion and State project with comparative survey data drawn from Waves 4 and 5 of the World Values Survey to respond to the question: Are levels of religion-state regulation associated with cross-national variation in attitudes related to the place of religion in public life? His analysis demonstrates modest links between institutions and aggregate public views, a relationship between institutions and social consensus, and, most robustly, consistent variation in institutional effects across political regime type.

Religion and public opinion in Costa Rica, 2018
The data analyzed was retrieved from the survey Perceptions of Costa Ricans on values and religious practices, which was conducted between October 6 th and 16 th , 2018, by telephone, to Costa Rica citizens, who owned a mobile phone. Phone numbers were randomly selected by a software; they were generated from the sequences of the numbers assigned by the Superintendence of Telecommunications to each of the mobile phone companies in the country. The interviews were conducted by research assistants in the Centro Tecnológico de Investigación Social (CETIS), under the supervision of Instituto de Estudios Sociales en Población (IDESPO) academic staff. The sample size consisted of 1002 Costa Ricans citizens. The results confidence level is 95% and their error margin is ± 3.1%.
In relation to religious beliefs of the population, 97.5% of the surveyed people affirmed to believe in God, some deity or superior force, and 2.4% indicated not believing in any deity. Furthermore, 67.4% of the population that claimed to be a believer also indicated that they actively participated in some religion, church, or spiritual search (IDESPO 2018). Therefore, it can be established that Costa Rican society remains a believer and highly religious. Given this, people who claim to be unbelievers (atheists) are excluded from the analysis, since their percentage is below the margin of error of the survey.
Regarding the current religious affiliation of the surveyed population, 52% indicated they were Roman Catholic, 27% said they belonged to an Evangelical Protestant denomination church (Pentecostal, Neopentecostal, Denominational Christians, etc.), 16.4% affirmed being a believer but did not belong to any specific religion, and almost 4% claimed to have another religious affiliation (Jews, Muslims, etc.) (IDESPO 2018). These data confirm the change that Costa Rican religious matrix has undergone in the last forty years, at the beginning of the 1980s practically 90% of the Costa Rican population declared themselves Roman Catholic; however, throughout that decade and, especially in the 1990s and early 21st century, the number of people who indicate attending an Evangelical Protestant churches has been steadily increasing (PROLADES 2012;Steigenga 2001).
The consultation on social problems and acceptance of various behaviors was adapted from similar consultations carried out in the "World Values Survey" (WVS 2019). Thus, the population was asked about how acceptable they considered various practices or behaviors, for which they were asked to rate them on a scale of 1 to 5, where 1 means they were never acceptable and 5 they were always acceptable. Socio-cultural practices that were consulted tend to find low acceptance among most of the people surveyed. Particularly, the most notorious cases regarding this are let a man hit his partner, where 98.5% of people selected values 1 and 2; suicide, where 92.7% of people selected values of 1 and 2; abortion, in which the percentage of people placed in values 1 and 2 reached almost 83%, and prostitution, where opinions that were positioned in values 1 and 2 contain the opinion of 75.7% of surveyed people (see Table 1).
Finally, surveyed population was consulted on the impact of different social problems in their daily lives, Table 2 shows that social problems such as: crime and drug trafficking (48.7%), corruption (58.4%), violence against women (56.8%), drug addiction and alcoholism (53.6%), bullying (47.8%), increase in social inequality (51.1%), unemployment (61.3%) and high cost of living (71.8%), are considered as problems with a high impact on the daily life of most people interviewed.
Regarding the influence of religion of public servers, the perception of the impact of this factor on people's lives is divided, with a slight majority considering it as a problem of little impact (29.2%); followed by those who consider it a problem that greatly impacts their daily lives (23.7%) and those who are at an intermediate level of consideration (22.9%) in this regard. A similar situation occurs with the assessment of unwanted pregnancies, where a majority (39.9%) considers it an inconvenience of high impact in their lives, while another group estimates its impact as very little (27.4%).

Analysis and Discussion
To analyze the data presented in the previous section, first, we proceeded to perform an analysis of variance (ANOVA) to determine if there were significant differences in the mean values on the acceptance of various social behaviors, according to the religion professed by the surveyed person. As Table 3 shows, significant differences in the level of acceptance of behaviors related to sexual and reproductive rights, such as homosexuality, abortion or divorce were found, which is consistent with the academic literature outlined in the first section of this article.
There are also significant differences for behaviors whose acceptance may be linked to the religious values of people, such as prostitution or suicide.
Surprisingly, there is no significant difference in behaviors associated with violence towards children and sentimental partners; especially because research conducted with female members of Evangelical Pentecostal churches in Costa Rica indicates that they are prone to adopt a submissive position towards men; and even to justify acts of violence committed against them and their children (Calderón González 2018). Table 4 shows the results of the mean comparison for the acceptance of the consulted social behaviors, using the Tukey HSD test. It is observed that, in all cases, two or  three groups of the population show statistically similar means concerning the acceptance of the consulted behaviors. However, the answers about the acceptance of sexual relations before marriage are striking, in this case, believers without religion not only have the highest average (highest acceptance) but also statistically differ from the means of other groups.
Regarding the perception of different social problems, the ANOVA test shows that there are no significant differences between the religious affiliation of the surveyed population (see Table 5). In other words, it is not possible to infer that the religious values or beliefs of the Costa Rican population sensitize them in any way so that they feel empathy towards any of these problems.
Taking the above into consideration, a binary logistic regression model was constructed to identify how the religious affiliation of the population is related to their acceptance of the consulted social behaviors. Table 6 shows the variables included in the model, as well as their codification.
Regarding the acceptance of homosexuality, the variables: Evangelical Protestant, other religious affiliation, religious practice, and age have a significant negative relationship; that is, they decrease the probability of accepting this behavior (see Table 7). On the other hand, in the case of females, the higher the level of schooling and the greater the perception of the economic level, the more likely it is that the surveyed people will show that they accept homosexuality.
Acceptance of prostitution presents a significant negative association with the variables: Evangelical Protestant, and age of the surveyed person. Acceptance of prostitution presents a significant positive relationship with believers without religion and the perception of the economic level.
As for the acceptance of abortion, this aspect presents a significant relationship with variables: Evangelical Protestant religious practice and age; and it has a significant positive relationship with the variables of high school and college.
About the acceptance of divorce, a significant negative relationship with Evangelical Protestant variables, religious practice, and age is shown; and a significant positive relationship is established with the variables of sex (female), high school, and college.
Regarding the acceptance of sexual relations before marriage, a significant negative relationship with the Evangelical Protestant variables, other religious affiliation, and religious practice is noticed; and, on the other hand, this variable is significative and positive associated with the believer variables without religion. Finally, the acceptance of suicide has a significant negative relationship with the variable religious practice and a positive association with the variable another religious belief.
Because the data analyzed shows that people with Evangelical Protestant affiliation tend to have a negative association with the social behaviors that are asked about in the survey, a logistic regression model was carried out  including only this group; to identify more precisely how sociodemographic variables of Evangelical Protestants are related to their acceptance of the consulted social behaviors. Table 8 shows that for the case of prostitution and suicide, none of the variables analyzed has a significant relationship with the willingness of Evangelical Protestants to accept or reject these behaviors. In the case of homosexuality, older Evangelical Protestants are more likely to adopt a stance of rejection towards it. In this same case, Evangelical Protestants with a higher perception of economic status have a greater probability of finding homosexuality acceptable.
Regarding abortion, the data in Table 8 show that the only variable that has a significant relationship is age. Thus, the older the evangelical Protestants, the greater the probability that they will not accept abortion. These results are like those found in groups of the Costa Rican population of other religious affiliations.
Regarding divorce, the only variable that presents a significant relationship is a religious practice. Thus,         evangelical Protestants who have a religious practice are more likely to not accept divorce. Finally, the acceptance of sex before marriage is negatively related to both, age, and religious practice; that is, older evangelical Protestants with an active religious practice are more likely to not accept sex before marriage.
Since the colonial period, Costa Rica has been characterized as a predominantly Roman Catholic country, even by 2019, it continues to be, along with the Vatican City State, the only Roman Catholic confessional state in the world. However, since the 1980s there has been a significant change in the Costa Rican religious matrix, and as in other Latin American countries, the number of Roman Catholic believers has decreased (Castillo Guerra 2017;Duffey 2009;Levine 2003), in contrast, the number of people attending Evangelical Protestant churches has increased. Thus, for the second decade of the 21st century, several studies estimate that about 25% of the Costa Rican population was Evangelical Pentecostal (Madrigal Pana 2012;PROLADES 2012).
The changes outlined in the Costa Rican religious matrix and the strong opposition that the authorities of the Roman Catholic and Evangelical Protestants churches have maintained regarding sexual and reproductive rights (Díaz González 2017;Fuentes Belgrave 2006), help to explain why the Evangelical Pentecostal affiliation, as well as religious practice, are variables that predict the negative attitude of the Costa Rican population towards the acceptance of homosexuality, abortion, divorce and sexual relations before marriage (see Table 5). The above is consistent with conclusions obtained by various studies carried out in other countries, which indicate that people who attend Evangelical Protestant churches, as well as those who claim to participate actively in their religious communities, receive stronger socialization processes, and therefore tend to replicate the oppositional positions of their religious leaders towards sexual and reproductive rights (Adamczyk & Pitt 2009;Olson et al. 2006).
Additionally, Costa Rica has followed the same trend of other Latin American countries and has experienced the emergence and strengthening of confessional political parties of Evangelical Pentecostal court, which have not only managed to get seats in the Costa Rican Congress, but even one of them, the Partido Restauración Nacional (PRN), obtained the highest number of votes in the first round of the 2018 presidential elections, although it lost the second electoral round. This has led several researchers to state that the religious beliefs of the Costa Rican population and, especially, their social conservatism, are important variables to explain their electoral political behavior, especially in these elections (Fuentes Belgrave 2019; Rosales Valladares 2018). This would imply an important change since, by the end of the 20th century, religious beliefs were not considered as significant variables related to the political behavior of the population (Steigenga 2001).
The significant relationship found between the Evangelical Protestant affiliation and high religious practice, and the rejection of behaviors related to sexual and reproductive rights is consistent with the revised academic literature, which shows that this religious group tends to have more conservative positions. However, in all cases, there is also a significant relationship with other sociodemographic variables such as age, sex, economic level, and schooling.
Additionally, the analyzed data shows that there is no significant relationship between the religious affiliation of the surveyed Costa Rican population and their perception of the involvement of different social problems. Given this, although the influence of religion on the Costa Rican population their perception about social problems cannot be ruled out yet, the evidence presented in this article suggests that such influence is not direct or strong, at least for most of the population. Therefore, conducting specific research is required to identify the causal mechanisms of the influence of religious affiliation regarding the socio-political attitudes of Costa Ricans. Likewise, research carried out in other latitudes of Latin America shows a contradictory or inconsistent behavior between the preferences and political culture of Evangelical Protestants and their electoral behavior (Smilde 2004).
When analyzing data regarding Evangelical Protestants only ( Table 8) it was found that the sociodemographic variables analyzed that have a significant relationship with the consulted social behaviors increase the probability of not accepting them. This suggests that in the case of Costa Rica, as in other countries, Evangelical Protestants are strongly conservative in matters of sexual and reproductive rights. Although more research is necessary, the above formulates the hypothesis that in Costa Rica, the Evangelical Pentecostal churches tend to adopt and promote positions that oppose sexual and reproductive rights among their members. Conversely, in other Latin American countries, major Evangelical Pentecostal church groups openly accept sexually diverse populations, and even promote progressive agendas on sexual and reproductive issues (Bárcenas Barajas 2020).
However, it is striking how Evangelical Pentecostals who perceive they belong to a high socioeconomic status are more likely to find homosexuality acceptable. No satisfactory explanation was found in the reviewed academic literature as to why evangelical Protestants self-perceived as wealthy are more likely to accept homosexuality. In many cases, a better economic level is associated with a higher educational level, which in turn is associated with greater tolerance and acceptance of sexual diversity, however, the data analyzed does not show such a relationship.
Finally, no data or theoretical elements are found to explain why Costa Rican people who claim to have a religious affiliation other than the Roman Catholic or Evangelical Protestants tend to have a greater chance of justifying suicide. Research has shown that, although at different levels, practicing religious people (Catholic, Protestant, Hindu, Evangelical, Muslim, etc.) tend to condemn suicide, including euthanasia, to the point of being less likely to commit suicide that a less religious person (Bulmer et al. 2017;Lizardi & Gearing 2010;Thimmaiah et al. 2016;Torgler & Schaltegger 2014).
Therefore, it is necessary to perform research with a more detailed and systematic approach to find the reasons why this group of the Costa Rican population tends to be more likely to justify suicide than the rest of the population that practices a religion. However, it is necessary to emphasize that, as in the rest of the Costa Rican population, non-acceptance or justification of suicide is dominant in this group.

Conclusions
The present study investigated the existence or absence of a relationship between a) religious beliefs of Costa Ricans and their perceived affectation by various social problems and b) religious beliefs and Costa Rican's acceptance of different social behaviors, especially those related to sexual and reproductive rights.
First, the analyzed data does not allow us to point out the existence of a significant relationship between the religious affiliation of the population and their perception regarding the affectation they suffer from various social problems. Although the same tendency of other Latin American countries is also found in Costa Rica, indicating an increase in the participation of religious groups in the discussion of public affairs, as well as of the appearance of political parties with clear religious-conservative political agenda, it is not possible to determine the impact of religion on the political positions of the population. The foregoing should not be understood as a denial of a possible relationship or influence of the religious affiliation of the Costa Rican population and their political and electoral preferences. For this reason, the completion of additional studies is necessary to elucidate the impact and scope of that influence.
Second, the data analyzed in this article enables us to conclude that the Costa Rican population, by 2018, presents the same trend identified in the academic literature regarding the relationship of the religious beliefs of the population and their opinion regarding issues related to sexual and reproductive rights. Thus, Costa Rican people with high religious practice and those with an Evangelical Protestant religious affiliation are more likely to oppose these issues. However, this opinion could be influenced by other sociodemographic characteristics, such as age, schooling, or economic level, so it cannot be said that the influence of religion is absolute. Therefore, religious affiliation is a variable that must be taken into consideration by public opinion studies in Costa Rica to have a better understanding of the opinions and behaviors of the population, especially among Evangelical Protestants and high religious practices persons. However, religious affiliation does not appear to be a key variable to explain the political behavior of the Costa Rican population, although its influence appears to be increasing.