Opiate of the Masses? Evidence from Surveys in Mexico and Colombia – James Ron
By James Ron and Richard L. Wood
Karl Marx famously argued that “Religion is the opium of the people…the sigh of the oppressed creature, the heart of a heartless world, and the soul of our soulless conditions.”
This classic Marxist argument has at least three overlapping claims: (1) Religion is a comforting illusion produced by oppressed people themselves in response to their material suffering; (2) Religion is a tool used by dominant groups to dull people’s willingness to fight to change the world, and (3) Religion is generally a force for social passivity, promoting a sense of resignation rather than individual and collective agency.
Our research, however, shows that Marx’s understanding of religion is wrong, at least in two major Latin American cities. In both Mexico City and Bogotá, a greater sense on the part of survey respondents that religion is important to their daily life was statistically associated with a stronger sense of individual agency, controlling for other relevant factors. This association was both statistically significant and substantively important, and it held across various religious denominations, socio-economic strata, and sexual identities.
The importance of religion, in these two contexts, goes hand in hand with a greater sense of perceived agency—a person’s sense that they can make a difference in the world around them.
Our Surveys
To test the link between religious importance and agency, the first author (Ron) and his colleagues ran two surveys in Mexico City and Bogotá in 2016 and 2017. The estimated population of Mexico City in 2016 was nine million persons, but the greater metropolitan area had roughly 21 million. In 2017, the population of Bogotá was seven million, and that of the city’s greater metropolitan area was over 10 million.
Our broader purpose was to explore the willingness of ordinary people to donate to local social change organizations, the results of which we published elsewhere (see here and here). With the help of local survey companies, we built samples that were representative of each city’s adult population.
We traveled to each location and walked through many neighborhoods, working with dozens of survey enumerators in each city to administer the questionnaires.
Our research was supported by the Open Society Foundations, which was interested in learning more about the potential for domestic fundraising for social advocacy groups in middle and low-income countries, with an account published earlier this year with IACS (the full Mexico City survey report is available here and the full Bogotá report here.)
Measuring a sense of agency
Embedded in our face-to-face, Spanish-language interviews with 960 randomly chosen adults in each city (for a total of 1,920 respondents) was a question about the respondent’s perceived ability to make a difference in the world.
We asked, “To what extent do you agree, or disagree, with the statement, Individuals like me can make a difference in the world?” We offered seven response options, ranging from 1 (“I do not agree at all”) to 7 (“I completely agree”).
Measuring the Importance of Religion for Respondents
To gauge the importance of faith for each respondent, we asked, How important is religion in your daily life? We offered 11 response options, ranging from 0 (“Not important at all”) to 10 (“Very important.”) We also asked about religious denomination, frequency of prayer, and frequency of attendance in a place of worship.
In other work, we used these questions to explore the impact of religious importance on respondents’ respect for human and civil rights, a topic we hope to write about soon for IACS.
Initial Findings: Perceived Sense of Agency
First, let’s look at how people in both cities together felt about their ability to make a difference in the world, or their “perceived sense of agency.”
As Graph 1 shows, an overwhelming proportion of respondents felt optimistic about their ability to make a difference in the world, and nearly 60% felt strongly about that.
We were surprised by these results, as we had expected that in resource-constrained environments such as Mexico and Colombia, both of which have also suffered from political and economic violence, people would be less optimistic about their ability to effect change. (See Ron’s recent blog post on the horrific toll of violence in Mexico.)
When we disaggregated the data by city, Mexico City residents appeared more upbeat than their Colombian counterparts; 65% “strongly agreed” in the first, compared to 53% in the second. When we combined the top two response categories, however, the results were identical; 74% of samples in both cities either “strongly agreed” or “agreed” that people like them can make a difference.
In other words, the adult population in both Mexico City and Bogotá was profoundly optimistic about its agency.
Initial Findings: The Importance of Religion
Next, let’s look at religious importance, measured by survey participants’ response to the question, “To what extent is religion important in your daily life?”
As Graph 2 demonstrates, religion was important to the populations in both cities: 38% of the pooled sample reported that religion was “very important” to their daily lives, while fully 72% scored religious importance at a 6 or above on the 0-10 scale. Once again, there were relatively modest inter-city differences. In Mexico City, 68% of the sample scored religious importance at a 6 or above, compared to 76% in Bogotá.
Both populations overwhelmingly identified themselves as “Catholic”; 67% in the pooled sample identified as such, followed by 14% who reported having “No Religion,” 11% who self-identified as “Christian” (non-Catholic, non-Evangelical), 7% as “Other,” and only 1% as “Evangelical.”[1]
Not surprisingly, given the importance respondents attached to religion in their daily lives, they also prayed frequently. In the pooled two-city sample, 60% reported praying either “once a day” (39%) or “several times a day” (21%), compared to only 16% who said they prayed either “seldom” (7%) or “never” (9%). The remainder reported “a few times a week” (14%), “once a week” (4%), or “a few times a month” (6%). Attendance at a place of worship was common, with 37% saying they attended once or more a week, compared to 25% who said they did so “seldom” or “never.”
Findings: An Association between “Religious Importance” and “Agency”
To probe the relationship between religious importance and a respondent’s perceived sense of agency, we built a statistical model whose main dependent variable was Agency, or the extent to which respondents agreed with the statement, “Individuals like me can make a difference in the world.” Recall from Graph 1 that there were seven possible response categories, ranging from 1, or “strongly disagree,” to 7, or “strongly agree.”
This model’s main independent variable was Religious Importance, running from 0, or “no importance at all,” to 10, or “very important,” as noted above.
I ran alternative models probing for the possible relevance of other factors, including religious denomination, frequency of prayer, frequency of religious attendance, sexual identity, political knowledge, socio-economic status, civic participation, and more.
In the end, some of these were unimportant influences, and the strongest model included only four factors: (1) Religious Importance, (2) Years of Education, (3) a Household Assets Index, and (4) an index of Civic Participation. All of these had statistically significant and substantively meaningful associations with the outcome, Agency. Notably, our final statistical model controlled for the potential impact of country and neighborhood, using techniques called “fixed effects” and “robust standard errors.”
A strong and consistent predictor of Agency was Religious Importance. As Graph 3 demonstrates below, a move from the lowest (0) to the highest (10) value on the Religious Importance scale was associated with a 10.1% increase on the Agency scale.
As Graph 3 demonstrates, respondents who ranked religion as “not important at all” in their daily lives scored their agreement with the statement, “individuals like me can make a difference in the world” at just under 5.7, reflecting the overall optimism of sample respondents – even non-religious respondents were upbeat about their potential to impact change. Individuals who reported that religion was “very important” in their lives, however, scored their sense of agency at just over 6.2, a gain of just over 10%.
The other three factors that we included in the model also mattered. Thus, a move from no to full Civic Participation produced a 12.6% gain on the Agency scale; a move from 0 to 22 Years of Education yielded a 10.6% gain; and a move from 0 to 1 on the Household Assets Index was associated with a 4.4% gain. The latter factor is a rough measure of socio-economic status, in which 0 denotes a respondent’s household having neither a computer, dishwasher, or microwave in their home, and 1 denotes having all three.
Substantively, these findings are notable given well-documented difficulties in explaining variation in respondents’ subjective psychological assessments, such as a “perceived sense of individual agency,” using standard survey variables.
Individual-level psychological constructs are shaped by myriad personal, cultural, and situational factors, many of which are unmeasured in public opinion research. When we detect consistent, statistically robust, and meaningful shifts in perceived agency as a result of religious importance (as well as civic participation, education, and assets), we should regard these factors as psychologically impactful.
Importantly, we ran a series of models testing whether these effects varied across asset levels, years of education, religious denomination, gender, and more. Our concern was that Religious Importance might matter more for females or males, richer or poorer people, Catholics or Evangelicals, or more or less educated respondents. We also explored non-linear relationships, in case religion mattered more at higher, rather than lower, levels. None of these statistical explorations panned out; Religious Importance had a linear relationship with Agency that was not moderated by its interaction with any of the other factors.
Conclusions
What does this all mean? First, Marx was wrong, at least when it comes to contemporary Mexico City and Bogotá. The more religion plays a role in the daily lives of people living in these cities, the more likely they are to believe they can change the world. This is true even for those at the lower end of the income scale, further undermining Marx’s claim.
Why does religious importance have this relationship with perceived individual agency? We are not sure, but it likely stems from the content of Latin American religious frameworks, including liberation theology in both Catholic and Protestant settings, evangelical Christianity, and the “new apostolic movements” in Catholicism. Some of these have long motivated pro-social collective action, and all encourage individual-level religious agency.
Future Research
More research on this is warranted. With more funding, I’d like to return to other Latin American contexts while also expanding the research to other world regions. It is important to discover whether there are similar effects across all of Latin America, as well as in Africa, North America, and Asia. It is also important to find out whether this relationship holds true for other major world religions, such as Hinduism, Islam, Judaism, and Buddhism. Is the relationship between Religious Importance and Agency an artifact of Latin America or Christianity, or is this something that transcends geographic place and faith tradition?
It is also important to probe the reasons for this association, which could perhaps be done through survey experiments, a rigorous method in which researchers subtly change statements and scenarios, testing for the influence of different formulations on people’s answers. It might also be done through additional questions that seek to pin down the specific elements of “religious importance” that relate to the respondents’ perceived agency. These results should then be introduced to and discussed by different groups of scholars of religion, social psychology, and social movements, as well as real-world practitioners, each of whom can offer explanations based on their own experience, scholarly disciplines, and theories.
James Ron, PhD, is a sociologist and political scientist. He shares his research insights on his LinkedIn, ResearchGate, and Google Scholar pages, as well as his professional and personal blogs.
Richard L. Wood, PhD, is president of the Institute for Advanced Catholic Studies at the University of Southern California and professor emeritus of sociology at the University of New Mexico.
[1] On a city level, 68% and 66% in Mexico City and Bogotá, respectively, identified as “Catholics”; 14% in both cities chose “No Religion”; 6% and 16% chose “Christian”; 11% and 2% chose “Other”; while only 1% and 2% chose “Evangelical.”