Delaware Review of Latin American Studies

 
DeRLAS Home Page
Contents page
Index of Volumes
Latin American Studies Program Home Page

         Vol. 5  No. 2     December 15, 2004

Faith in a Changing Mexico: The Effects of Religion on Political Attitudes, Engagement, and Participation

Eric Patterson
Department of History and Political Science
Vanguard University

Presented at the XXIV International Congress of the Latin American Studies Association, March 27-29, 2003 in Dallas, Texas.  I am grateful to Kathleen Bruhn, Cynthia S. Kaplan, Pericles Leon, and Eric R.A.N. Smith for comments on earlier drafts of this research.  I appreciate financial support from the Department of Political Science at University of California, Santa Barbara which made acquisition of Latinobarometer data possible.

Abstract
Since about 1970 between fifteen and twenty-five percent of the populations of various Latin American countries have converted to Protestantism.  This ‘neo-Reformation’ seems to have the power to reshape not only the spiritual experience, but also the social, political, and economic lives of its adherents.  Mexico has long been conservative about religion, but even there the role of religion in society is changing.  For instance, after a six decade hiatus, the Mexican government has recently allowed the Catholic Church to re-enter the public sphere, while during the past decade the number of Mexican Protestants has doubled and will likely double again by 2010.  Thus, in a time of rapid political, economic, and social change, Mexican society is experiencing a religious pluralism unknown in the past.  Do different religions result in different politics?  The findings suggest that religious variables do influence political attitudes, engagement, and participation in Mexico.
Debating the Socio-Political Ramifications of Mexican Protestantism
    An Alternate Protestantism: Passive, Otherworldly, and Authoritarian
    Applying Theory to Mexican Protestantism
Analysis
    Attitudes Toward Democracy and Political Engagement
    Political Preferences and Resources
Conclusion
Appendix
    The Mexico 2000 Panel Study
    The Latinobarometer
    Questionnaires
Notes
References
Religion has a tradition of controversy in México.  The power of the Catholic Church was so potent in the n.ineteenth century that Benito Juárez remarked, “The future prosperity and happiness of my nation now depends on the development of Protestantism…they [the people] need a religion to compel them to read and not to spend their savings on candles for their saints” (Bowen, 1996: 25).  One of the goals of the Revolution was curtailing the political power of the Catholic Church, and the terrible Cristero War of the 1920s demonstrated that Mexicans continued to be polarized on the issue. In the past decade the Mexican government has cautiously allowed the Roman Catholic Church to begin to re-enter public life after a seventy-year ban.  And in the past year examples of murder and ambushes between Catholics and Protestants in Chiapas indicate that religion is again a potent socio-political fault line in México.1

Much has been written about changes in the Mexican political landscape since 1988 but it is rare to find scholarly discussions about the role of religion in defining political attitudes, developing political engagement, and encouraging (or discouraging) political participation.  Similarly, although there are dozens of scholarly works chronicling the “neo-Reformation” which is affecting nearly all of Latin America, there is very little scholarship on contemporary Protestantism in México. This is particularly surprising because during the past decade the number of Mexican Protestants2  has doubled and will likely double again by 2010.
This paper is about the intersection of faith and individual political attitudes and behavior in Mexico.  The fundamental question addressed here is whether different religions result in different politics—are there significant differences in the political orientations and actions of Mexican Catholics and Protestants?  More specifically, the paper looks at the similarities and differences between Mexican Catholics and Protestants in terms of political attitudes, political engagement, and political participation.  The findings suggest that religious variables do influence politics in Mexico.

Debating the Socio-Political Ramifications of Mexican Protestantism
Contemporary studies of Protestantism in Latin America tend to operate from one of two analytical frameworks.  The first is optimistic and associated with the Max Weber’s Protestant Ethic thesis.  In contrast the second, associated with Christian Lalive D’Espinay, concludes that Protestantism is otherworldly and reinforces hierarchy and authoritarianism in Latin American society.

In The Protestant Ethic and the Spirit of Capitalism (1934), Max Weber argued that the cause of capitalism, and ultimately democracy, taking root in Northern and Western Europe as opposed to the Latin European countries was due to culture.  He probed differences in culture, primarily defined in terms of religion, and asserted that Catholicism emphasized passive obedience to authority, fatalism, collectivist values, and wastefulness.  These values were reinforced by the Catholic Church and resulted in hierarchical religion, authoritarian government, and mercantilist economics.  In contrast, Weber’s examination of early post-Reformation Protestantism demonstrated that the religious focus on individual faith, the priesthood of all believers, and Calvin’s emphasis on election and its corresponding material blessings resulted in cultures of thrift, individuality, competition, and equality—the foundation of capitalist democracy.

Moreover, as Weber traced the historic evolution of European Protestantism, he found the same ethic of ascetic, or “holy,” lifestyle in the English Methodist movement:
 

“The [doctrine of] regeneration of Methodism thus created only a supplement to the doctrine of works, a religious basis for ascetic conduct after the doctrine of predestination had been given up.  The signs given by conduct which formed an indispensable means of ascertaining true conversion, even its conditions as Wesley occasionally says, were in fact just the same as those of Calvinism” (119).

In other words, although Methodism rejected the predestinarian philosophy of Calvinist theology, its ethic for living promoted the values of honesty, industry, thrift, and temperance that were essential to capitalism and democracy.  Interestingly, these values are essential to understanding the worldview of contemporary evangelicals and Pentecostals (Swatos, 1994; Harrison, 1992; Sherman, 1992;  Martin, 1990).

It has also been argued that the individualism, decentralization, and personal commitment to church service associated with Protestantism develop the civic skills of participants.3  Proponents of the view that Latin American Protestantism provides skill building opportunities argue that evangelical Protestantism is democratic in its structures: local churches usually elect their pastors, deacons, and elders and most decisions are made locally with the views of the laity taken into account (Smith and Prokopy, 1999; Cleary and Stewart-Gambino, 1997; Miller, 1994; Martin, 1990).  In addition, beyond the structure of self-government, it is argued that tangible skills are nurtured in Protestant churches and organizations.  For example, parishioners not only vote for church leaders but also learn to speak in front of the assembly, teach groups of people, learn elements of institutional finance, and cooperate to build schools and orphanages (Lynch, 1998; Cleary and Stewart-Gambino, 1997, Smith, 1994; Sherman, 1992).

In sum, the Weberian tradition in social science argues that a “syndrome” of values epitomizes democratic publics.  This set of values is indicative of a political culture that prioritizes equality, tolerance, interpersonal trust, individualism, and willingness to compromise within a context that fosters civic skills.  Thus one would assume that the Protestant “neo-Reformation” which is occurring in Latin America might provide similar conditions for the development of democracy.  However, there is a body of scholarship that argues that contemporary Latin American Protestantism is authoritarian, politically passive, and otherworldly.

    An Alternate Protestantism: Passive, Otherworldly, and Authoritarian
Observers of Latin America have been stunned to see the number of Protestants grow from a handful to 20-30% of the population in a single generation.  Some scholars reflect critically on the values of these burgeoning Protestant denominations.  The most well known critique of the optimistic Weberian hypothesis in the Latin American context comes from the influential work of Swiss sociologist Christian Lalive D’Espinay.  This scholar evaluated the modest Chilean Pentecostal movement in the 1960s and argued that instead of being a force for social and economic change, Pentecostalism was conservative.  Lalive D’Espinay’s qualitative study indicated that in the alienation of industrialization, as people moved from the countryside to the city, they were looking for order and direction.  The poor found this in authoritarian Pentecostal pastors who reproduced the hacienda in the city:
 

“Pentecostalism teaches its initiates withdrawal and passivity in political matters, limited only by the commandment to be submissive to authority.  In its social forms, it appears as a specialized (since it is purely religious) reincarnation of a moribund society, and as the heir of the past rather than the precursor of emerging society. The components create a force for order rather than an element of progress; a defender of the status quo rather than a promoter of change” (1967, 145).

Lalive D’Espinay argued that Pentecostals were passive and withdrawn from politics, and “contrary to the Puritan ethic of Anglo-Saxon Protestantism…the Chilean Protestant ethic does not seem to offer any obvious parallel with the pioneer spirit” (1967, 153).  Several other studies agree with D’Espinay that individuals convert to Protestantism under conditions of anomie caused by modernization (Glazier, 1980; Stoll, 1990; Martin, 1990; Schafer, 1995). 

    Applying Theory to Mexican Protestantism
Until recently, most studies of Mexican Protestantism focused on the histories of small Protestant enclaves or the feeble attempts at evangelization by mainline Protestant missionaries.4  However, with the recent growth of evangelical Protestantism throughout the region, scholars are using the frameworks of Weber and Lalive D’Espinay to consider the attitudes and behavior of Protestants in Mexico.  Long-time Mexican Protestant observer Jean-Pierre Bastian (1992) notes that the real power of Protestantism is that it can adapt to local cultures.  Bastian argues that rather than being a force for socio-political change, Protestantism has become “Latin Americanized” in Mexico.  Bastian says that Protestant congregations are patriarchal, hierarchical, and corporatist in character, and that this religious culture has adapted from the dominant Latin/Catholic culture.  This is in agreement with Lalive D’Espinay’s critique that Protestant churches in Chile were authoritarian in structure, and the work of Phillip J. Wellman (1997) and Andrew Chesnutt (1997) who argue that Pentecostals engage in “participatory authoritarianism” which is clientelistic and conservative.

Lalive D’Espinay observed that Protestants tend to come from the lower classes seeking order in a time of social disruption.  In the Mexican case, Cristián Parker (1996) makes a similar argument.  Parker argues that what draws many to Protestantism is the chaos in their lives resulting from modernization and urbanization.  Thus Protestantism is, at first, a mechanism for providing order and meaning in the lives of the dispossessed.

Although Lalive D’Espinay argued that, internally, Protestant churches were hierarchical, he suggested that, when it came to politics, Pentecostals were passive and withdrawn.  Susan Eckstein (1988) and David Stoll (1990) concur that the doctrinal themes of evangelicals are apolitical and otherworldly and therefore will likely have little positive impact on Latin American political systems.  Others argue that the millenarian focus on spiritual results makes Protestants neglect real-world issues, and thus they are cast as conservative (Mariz, 1994; Chesnutt, 1997; O’Shaughnessy, 1990; Le Bot, 1999).

Lindy Scott’s (1992) work on evangelicals in Mexico City from 1964-1991 takes a more nuanced view.  Scott argues that, in general, Protestants were cautious about political involvement, but that in times of crisis they take positions of leadership.  For example, during the Mexico City earthquake, Protestants led recovery efforts in many neighborhoods.  Scott says that this was because of their organizational ability and their strong sense of community service.  Other scholars also disagree with the conclusion that Protestant dogma mandates political abstention. Domínguez (1994) says that, although the primary communications of Protestant leaders and organizations focus on the individual’s need for spiritual regeneration as a distinctly different matter from temporal needs, Protestants do value individuality, liberty, equality, and fraternity in their congregations which empowers common people (Domínguez, 352).  Due to their relative minority status in Mexico, Protestants tend to shy away from overt political involvement, and this is often misinterpreted as anti-political or apolitical messages.  David Martin (1990) has called this apolitical stance one merely of practicality, and that, once a group has developed a “free space” but focusing internally, it may later turn its face toward social and political concerns, as has happened among the Protestants of Brazil, Chile, and El Salvador.

Perhaps of more importance are the methods and activities of Protestant churches. Numerous studies have demonstrated that Latin American Protestants act democratically by electing elders and pastors, by involving women in decision-making and leadership circles, and by providing forums for lay people to speak in public (Smith, 1994; Cleary, 1997; Glazier, 1983). Guerra’s study of three Protestant denominations suggests that the democratic environment of those churches educates parishioners in the basics of organizational dynamics and elective democracy.  Many scholars believe that these practices, although for the most part limited to involvement in the local church or religious organizations in Mexico today, will lead to greater participation and activity by Protestants in the future, as has happened in Brazil, Chile, Guatemala and several other Latin American countries.

Survey data indicate that, although Mexico is heavily Catholic, many in the Mexican populace do not support greater politicization of the Catholic Church in Mexico (Camp, 1995).   Are Mexicans equally unlikely to support the introduction of Protestantism to the political realm?  In many Latin American countries, such as Peru, Brazil, and Chile, Protestants have developed a reputation as ‘honest brokers’ and ‘political outsiders’ because of their firm stance against corruption and their lack of ties to the political establishment (Lynch, 1998; Cleary and Stewart-Gambino, 1997).  Roderic Ai Camp (1994) suggests that this may be the case for Mexican Protestants who traditionally have avoided politics and therefore may be seen as ‘clean outsiders’ in the future political process (85).  Similarly, Bonicelli (1993) reports that, although in the past Protestants avoided politics due to their minority status and fear of persecution, their growth and the increasingly open political environment encourages more political participation by Protestant clergy and by laity.

In sum, scholars disagree over the nature and potential of Mexican Protestantism.  Some argue that Mexican Protestantism has the potential to ultimately reshape society by changing economic and political attitudes, promoting entrepreneurship and upward social mobility, and developing participatory skills.  Other studies disagree, instead arguing that Mexican Protestantism is authoritarian within its congregations and uninterested in society and politics.   Examination of survey data will allow us to test these claims and analyze the economic and political attitudes and political behavior of Mexican Protestants and Catholics.

Analysis
The question motivating this research is whether religion influences political and economic attitudes and behavior.  In the case of Mexico this is a contested topic because, for the past two centuries, some observers have felt that Catholicism is central to the identity of citizens while others have argued that “Protestantization” was necessary for Mexico to develop into a pluralistic, open, tolerant, competitive, and democratic society.  These contrasting views toward Mexican religion agree on one point: religious values can influence social and political attitudes.

Unfortunately, to date, there has been little quantitative research concerning the nexus of politics and faith, particularly Protestant faith, in contemporary Mexico.  This study contributes to our understanding of religion’s influence on individual-level politics through the use of survey data from 2000-2001.  I test for differences between Mexican Catholics and Protestants in terms of political attitudes, political engagement, and political participation.   Do religious affiliation, church attendance, and/or religious intensity effect political variables in the Mexican electorate?   The findings suggest that there are differences in political attitudes, engagement, and participation based on religion in Mexico.5

Although Protestants make up a minority of the population, their growth rates are impressive.  Kurt Bowen records growth rates by region from 1970-1990 and finds that the national evangelical growth rate was 17.6 percent per annum during this time period, with a high growth rate in the south of 24 percent.6  According to the 1990 census Mexico was only 4.9 percent Protestant, but many agencies argue that these numbers are misleading.7 However, the 2001 census records that the Mexican population is 5.2 percent Protestant and that an additional 3-4 percent of the population reported their religious affiliation as “other” or “none.”8  Of these numbers, only about 20 percent affiliate with Pentecostal denominations headquartered in the United States (e.g. Assemblies of God, Foursquare Church) and the rest tend to be from native evangelical churches.9  Protestant denominations have been growing rapidly in recent years and at the current rate of growth, Protestants could double their numbers in the next five years.10

Mexico is strongly Catholic, with 89 percent of the population identifying as Catholic in the 2001 census, and 44 percent of professing Catholics reporting that they attend mass at least once per week.  This is quite different from many other Latin American countries where 20 percent or less of the Catholic citizenry regularly attend church services.  In the Latinobarometer sample, 73 percent of Protestants responded that they are “very” or “somewhat” devout (religious intensity).  Only 47 percent of Catholics claimed to be devout.11  Similarly, in the 2000 Mexico Panel Survey, about half of all Catholics and 73 percent of Protestants attend church services once a week or more (religiosity).

    Attitudes Toward Democracy and Political Engagement
The following sections provide recent survey evidence with which to evaluate the “different-religions, different politics” hypothesis.  Table 1 provides raw percentages and chi square “goodness of fit” measures for a variety of democratic variables. The numbers in the rows are percentages of respondents who gave an affirmative answer to various political indicators.  For example, the first row of Table 1 indicates that about 40 percent of devout Catholics and all Protestants and 52 percent of not devout Catholics answered the question affirmatively (prefer democracy).

                        Table 1.  Democratic Attitudes: Percentages of Affirmative Responses (Mexico).

Feelings about
Democracy
Catholic
Devout
Catholic
Not Devout
Protestant
Devout
Protestant
Not Devout
Chi
Square
Democracy 
  Preferable
40^ 52 39 38 34.85***
Satisfaction 
  w/ Life
78 76 85^ 77 21.41**
           Source: Latinobarometer 2001 (n=978).  * p <.1, ** p <.05, *** p<.01. 
                        ^ indicates that the within-religion findings (e.g. devout Catholic vs. not devout Catholic) are statistically significant at p<.1.

Perhaps the crucial question regarding democracy is whether citizens prefer democracy to other types of regimes.  The question is important because it demonstrates the priority placed on democratic norms regardless of the social and economic context of the country.  When asked if democracy was preferable, not devout Catholics were significantly more likely to answer affirmatively than devout Catholics or Protestants.  In contrast, instead of “preferring democracy” respondents were allowed to choose “authoritarian government is preferable at times” and a large number did: 40 percent of devout Catholics and 43 percent of devout Protestants indicated that authoritarian government was desirable “at times.”  In short, there is variation in preference for democracy in Mexico based on religious intensity—those who are devout tend to be significantly more likely to accept an authoritarian regime than those who are not devout.

Scholars argue that life satisfaction is another important indicator of support for democracy (e.g. Inglehart 1977, 1988). To ask such a question in a country like Mexico is interesting due to the political pluralism that began to flourish following the 1988 presidential election and the varied rates of economic success experienced by different sectors of its population.  On a four-point life satisfaction scale (“very,” “somewhat,” “not very,” “not at all”) over two-thirds of all Mexicans said that they were “very” or “somewhat” satisfied with their life in 2001. Furthermore, devout Protestants are significantly more satisfied with their lives than not devout Protestants and both categories of Catholics.

The 2000 Mexico Panel Survey asked two other questions about democracy: “Is Mexico a democracy?” and “Can you trust other people?”  There were not significant differences based on religious affiliation for either variable.  For example, only about 40 percent of all respondents answered affirmatively that Mexico is a democracy.  The one significant difference in the findings is when one compares Protestants alone—Protestants who attend church regularly are considerably less likely to see Mexico as a democracy than other Protestants.  If it is true that Protestant churches are democratic in structure and practice, then it is not surprising that religious Protestants would recognize that their government is less than democratic.  On the interpersonal trust variable, 85 percent of all respondents indicated that they felt that they could not trust others.

Political engagement variables are another key set of indicators regarding democracy.  Various studies have demonstrated that individuals who are psychologically engaged in politics are more likely to support democratic norms and participate in political behavior.  Table 2 provides responses to questions about political engagement in the Mexican case.  In general, differences consistently occur between Protestants who regularly attend church services (religious) and those who do not. For example, when asked “How interested are you in politics?” not religious Protestants were more likely than religious Protestants to be interested in politics.  This finding is important because political interest is the foremost indicator of political engagement--people who are not interested in politics are unlikely to deliberate about political choices and are unlikely to vote or participate in other forms of political behavior.  Nevertheless, on most of the indicators the differences between Catholics and Protestants in general were small and not statistically significant.  In other words, religiosity may play a role in predicting the political engagement of Protestants, but it seems to play little role in predicting engagement in the public at large.

Interestingly, the levels of political interest and efficacy of Mexican citizens are roughly similar to many Western democracies.  However, one place where Mexicans differ from European and North American democracies is in perception of political corruption.  Citizens are unlikely to feel that their voice, or vote, matters or that democratic procedures work if they consider their political system to be corrupt.  About 55 percent of Mexicans, regardless of faith, feel that elections are “totally” or “more or less” clean in their country.  This figure is higher than many other Western democracies. 

                        Table 2.  Political Engagement: Percentages of Affirmative Responses (Mexico).

Engagement
Variables
Catholic
Religious
Catholic
Not Religious
Protestant
Religious
Protestant
Not Religious
Chi
Square
Political Interest 30 29 20^ 28   11.36
Electoral Corruption 56 56 63^ 50  12.92
Talk about Politics 25 27 17 17    5.78
Read about Politics 26 28  13^  31   10.53**
Follow Elections 34  28  25  21  12.14***
            Source: Mexico 2000 Panel Study (n=2184).  * p <.1, ** p <.05, *** p<.01. 
            ^ indicates that the within-religion findings are statistically significant at p<.1.

The data in Tables 1-2 indicated that there are political differences between the four religious categories.  The question is whether religion itself, or other intervening variables, account for these differences.  Tables 3-4 provide ordinary least squares models for religious explanations of political engagement and attitudes toward democracy.  In addition, demographic variables such as age, education, and gender are provided to test their relationships to democratic attitudes and engagement. The tables provide regression coefficients with standard errors and beta weights.  The general finding is that religion and demographic factors have varying levels of explanatory power for political attitudes in Mexico.

The data in Table 3 provide the results of OLS regression analysis explaining two democratic attitudes in Mexico.  In the case of “life satisfaction,” interviewers asked respondents how satisfied they were with their lives: “very satisfied,” “somewhat satisfied,” “not very satisfied” or “not at all satisfied.”  Each level of satisfaction was assigned a number between 1-4 (low-high) for OLS analysis.  The results indicate that religious variables predict higher levels of life satisfaction in Mexico.  For instance, the robust coefficient for the religious affiliation variable (.72) indicates that Protestants are considerably more likely to be satisfied with their lives than Catholics.  The explanatory power of the religious affiliation variable is most evident when one compares the beta weights of the various significant explanatory variables for life satisfaction: religious affiliation is three times more powerful (ß=.33) an explanation than SES and political interest.  In addition, the religious intensity variable, Devout, is also significant, demonstrating that religious intensity (“devout-ness”) has a positive influence on life satisfaction as well.12  In addition, those who are better off financially and those who are younger are more likely to report higher levels of life satisfaction.  Finally, what explains preference for democracy in Mexico?  The religious variables utilized in this study do not have significant impacts on preference for democracy or authoritarianism.  In other words, one’s faith and religious intensity do not predict preference for democracy over authoritarian government.  The same seems to be true for indicators of sex, age, education, and SES.  Surprisingly, political interest does have a strong, negative correlation with preference for democracy.

                        Table 3. OLS Regression Analysis Explaining Life Satisfaction, Prefer Democracy (Mexico).

Variable
Life Satisfaction
Prefer Democracy
  Unstandardized            Beta (ß)
Coefficient (B)
Unstandardized            Beta (ß)
Coefficient (B)
Protestant         .72*                           .33
        (.37)
        -.18                             -.05
        (.66)
Devout         .05*                            .06
       (.03)
        -.00                            -.00
        (.05)
Sex         .02                            .01
        (.05)
        -.04                            -.02
        (.08)
Age         -.04**                       -.07
        (.02)
        .05                              .05
        (.03)
Education         .00                            .01
        (.02)
        -.02                            -.03
        (.03)
SES         .07***                        .10
        (.02)
        -.05                            -.04
Political Interest         .08***                        .10
        (.03)
        -.25***                        -.16
        (.05)
(Constant)         2.20         2.87
        (.32)
Adjusted  R2         .04         .03
              Source: Latinobarometer 2001.  Sample Size=978. * p <.1, ** p <.05, *** p<.01. 
                        Life Satisfaction: “How satisfied are you with your life: very, somewhat, not very, or not at all satisfied.”

Two other variables that measure democratic attitudes are “Interpersonal Trust” and “Is Mexico a Democracy.”  Again, neither religious affiliation nor church attendance (religiosity) predicts either political attitudes.  In the case of interpersonal trust, it seems that political interest has a positive effect on trust and that men tend to be more trusting than women in Mexican society.  When it comes to the question about Mexican democracy those who are older and less educated are more likely to respond affirmatively that Mexico is a democracy.

Political engagement variables are important indicators of democratic attitudes and likelihood of participation in a democratic society.  What explains political interest, willingness to talk about politics, likelihood of reading about politics, and whether one follows elections in Mexico?  In this case religion does matter: the variable for religious affiliation (Protestant) has a significant, negative relationship with all four variables (Table 4).  In other words, Protestants are less likely to be “plugged into” politics than Mexican Catholics.  Interestingly, on two of the variables Church Attendance plays the opposite role: those who attend church regularly are significantly more likely to follow elections and be interested in politics (p=.12).  Demographic indicators are also important predictors of political engagement—those who are male, better off financially, younger, and better educated tend to be most likely to be interested in politics and follow the political news.

                        Table 4. OLS Regression Analysis Explaining Political Engagement  (Mexico).13

Variable
Political Interest
 B              ß
Talk Politics
   B             ß
Read Politics
   B             ß
Follow Elections
   B             ß
Protestant     -.09*        -.05
    (.05)
     -.08**        -.05
    (.04)
    -.10**        -.06
    (.04)
    -.10**        -.05
    (.05)
Church Attendance     .03            .04
    (.02)
    .02              .03
    (.02)
    -.01            -.01
    (.02)
    .08***          .09
    (.02)
Sex     .07***        .07
    (.02)
    .08***          .09
    (.02)
    .11***          .13
    (.02)
    .04*            .05
    (.02)
SES     .02***        .09
    (.01)
    .02***          .09
    (.01)
    .03***          .11
    (.01)
    .02***        .08
    (.01)
Age     -.04***        -.08
    (.01)
    -.05***        -.10
    (.01)
    -.04***        -.07
    (.01)
    -.03**        -.06
    (.01)
Education      .07***        .19
    (.01)
    -.05***         -.10
    (.01)
    .08***          -.07
    (.01)
     .04***        .12
    (.01)
(Constant)     .29
    (.07)
    .23
    (.06)
    .29
    (.07)
    .29
    (.06)
Adjusted R2     .07     .11     .10     .04 
               Source: Mexico 200 Panel Study. Sample Size=2184.

In sum, we have seen that religious variables provide little explanation for political attitudes but are important predictors of political engagement.  Protestants are significantly less likely to be engaged in politics in Mexico.  Furthermore, being Protestant is a strong predictor of life satisfaction, as to a lesser extent is being devout.  Finally, it was also demonstrated that strong relationships between demographic variables and political engagement do exist.

    Political Preference and Resources
Another political dimension for comparison is political preference.  The choice of citizens to embrace policies and politicians from various points on the political spectrum can indicate cleavages between groups in society.  Respondents were directed to place themselves on an eleven-point (0-10) Left/Right political scale.  Such a scale is particularly interesting in the context of Mexico’s competitive election for president in 2000 that resulted in an opposition party winning the Executive for the first time in seventy years.14

Responses to the Left-Right scale are shown in Table 5.  Overall, Mexicans tend to place themselves in the Center and on the Right, with less than 17 percent of the entire population identifying with the Left.  Again it is not religious affiliation but rather religiosity that seems to account for political differences.  Religious Catholics tend to lean Center and Right, whereas not religious Catholics are a bit more likely to be Center and Left.  As other studies have reported (e.g. Bowen, 1996) religious Protestants tend to be significantly more likely to support the long-ruling PRI than are not religious Protestants.  This may be because the PRI’s curtailment of the Catholic Church provided a modest opening for Protestants in Mexican society throughout the twentieth century.  Interestingly, when OLS regression is used to test predictors of political preference on the Left-Right scale, we find that only two variables are significant predictors of political orientation (p=.05).  Church Attendance is a strong predictor of leaning to the political Right (B=.09, ß=.06) whereas higher levels of education predict placement toward the political Left (B=.04, ß=.07).

                        Table 5. Political Preferences-Percentages  (Mexico).

Left-Right Scale Catholic
Religious
Catholic
Not Religious
Protestant
Religious
Protestant
Not Religious
Chi
Square
Left 13 18 14 17   
Center 32 35 34 35  
Right 55 47  52  48  33.12
Party ID          
PRI 39  37  43  28^  
PAN 24  21  11  17  
PRD 7  9  10  11  28.04**
                        Source: Mexico 2000 Panel Study. Sample Size=2184. * p <.1, ** p <.05, *** p<.01. 
                        ^ indicates that the within-religion findings are statistically significant at p<.1. 
                        Note: Over a quarter of all respondents chose “none” when asked their party affiliation. This was especially true for Protestants: 36
                        percent of non-attending Protestants and 33 percent of attending Protestants chose no partisan affiliation.

Mexican Catholics and Protestants report high levels of voting.  Interestingly, religious variables do appear to play a role in determining voting.  The raw data in Table 6 suggest that Catholics and Protestants who regularly attend church are more likely to vote than members of their own faith who do not attend services.  However, OLS regression tells a more complex story (Table 7).  It seems that religiosity influences individuals to vote, but, at least in 1994 and 1997, being Protestant made one less likely to vote.  By 2000, however, it appears that Protestants were catching up with Catholics in likelihood of voting.

Interestingly, being younger and better off financially correspond with higher likelihood of voting.  Surprisingly, education has a modest, negative significance with regard to voting.  However, the most powerful predictor of voting is political interest: those who are the most interested in politics are the most likely to vote.  These findings for voting help explain why Protestants were somewhat less likely to vote than Catholics in the 1990s—as we saw above, Protestants have been less politically engaged than their Catholic neighbors.

                        Table 6. Percentage Affirming “Have Voted” (Mexico).

Catholic
Religious
Catholic
Not Religious
Protestant
Religious
Protestant
Not Religious
Chi
Square
Vote 1994 77  72  67^  52  13.77***
Vote 1997 74^  66  63^  54  20.23***
Vote 2000 86  84  91^  78  34.59**
Work on Campaign 2000 6  5  6  4  00.32
                 Source: Mexico 2000 Panel Survey.  Sample Size= 2184.  * p <.1, ** p <.05, *** p<.01.

                        Table 7. OLS Regression Analysis Explaining “Have Voted”  (Mexico).


Variable
      Voted 1994
Unstandardized        Beta
Coefficient (B)         (ß)
      Voted 1997
Unstandardized        Beta
Coefficient (B)         (ß)
      Voted 2000
Unstandardized        Beta
Coefficient (B)         (ß)
Protestant     -.11**               -.06
    (.05)
    -.14***                -.07
    (.05)
    -.04                    -.03
    (.05)
Church Attendance     .02                    .03
    (.02)
    .07***                    .08
    (.02)
    .06***                    .09
    (.02)
Sex     -.04-                  .04
    (.02)
    -.02                    -.02
    (.02)
    .03                        .04
    (.03)
SES     .03***                 .10
    (.09)
    .02***                   .08
    (.01)
    .00                        .01
    (.01)
Age     -.11***                -.22
    (.01)
    -.08***                -.15
    (.02)
    -.03**                   -.08
    (.02)
Education     -.02*                 -.06
    (.01)
    -.02***                -.05
    (.01)
    .03**                     .09
    (.01)
Political Interest     .07***                 .07
    (.02)
    (.10)***                .11
    (.03)
    .07***                    .09
    (.03)
(Constant)     1.16
    (.07)
    .99
    (.08)
    .80
    (.76)
Adjusted R2     .06      .04      .02
                   Source: Mexico 2000 Panel Survey.  Sample Size= 2184.  * p <.1, ** p <.05, *** p<.01.  Note: It was difficult to run OLS regression on
                        worked for a campaign due to the way it was scored in the original survey analysis.

There are two other indicators on the 2001 Latinobarometer survey that provide preliminary information about religion and resources or skills.  One of the questions regards membership in any organization and the evidence indicates that there are not significant differences between and within religions in terms of organizational membership.  Respondents were asked “Do you participate in any organizations?  How many?”  About one-third of all respondents, regardless of religion, indicated membership in at least one organization.  This is a large number, similar to the percentage in many European countries and much higher than many other Latin American countries.

The survey also asked if the citizen participates in strikes, demonstrations, or sit-ins (“Protest”).  Respondents could answer “never,” “sometimes,” or “often.”  The figures in Table 8 are for “sometimes” and “often” responses.  The differences between devout and not devout Catholics are not significant, but not devout Protestants stand out from devout Protestants, being half-again as likely to participate.  Nonetheless, this finding can be misleading.  When we distinguish between responses of “often” protest and “sometimes” protest, we find that 28 percent of not devout Protestants, 15 percent of devout Protestants, and about 17 percent of all Catholics reported “sometimes” protesting and only about 2 percent of the entire sample indicating that they “often” protest.  Nonetheless, the cumulative figure for “often” and sometimes “protest” is significantly higher than the protest levels of countries such as the United States and Britain.

Consequently, in this sample almost a third of not devout Protestants have experience in political acts of protest and they report high levels of political engagement whereas the devout Protestants have significantly lower levels of political engagement and the highest levels of life satisfaction. 

                        Table 8. Percentage Affirming Organizational Membership and Protest (Mexico).

Catholic
Devout
Catholic
Not Devout
Protestant
Devout
Protestant
Not Devout
Chi
Square
Organization 34  31  34  31  8.89
Protesta 20  18  18^ 31  7.35
                        Source: “Organization” and “Protest” Latinobarometer 2001 (n=986). 
                        * p <.1, ** p <.05, *** p<.01.  ^ indicates significant within-religion variation at p<.1.

These numbers indicate that quite a large percentage of Mexicans participate in an organization and many have protested in some way.  What explains these high levels of participation in groups and in unconventional political behavior?  The data in Tables 9-10 provide preliminary answers.  The results of OLS analysis suggest that religious variables do predict organizational membership in Mexico.  Although the religious affiliation variable (Protestant) is not significant, the religious intensity variable is. This means that those who are more devout in Mexico are more likely to participate in organizationsthan those who are not.  Moreover, because those who are devout are the most likely to participate in religious organizations, this is evidence that may support the resource hypothesis that those involved in religious organizations are the most likely to participate elsewhere as well. 

What explains unconventional political behavior, such as protesting and demonstrating, in Mexico?  Only one variable has a statistically significant impact on “protest.”  That variable is “Devout” (Table 9).15 This means that those who are devout are more likely to protest than those who are not.  It also means that demographic and engagement variables are not statistically significant explanations for protest in Mexico.  However, it is not surprising that religious variables predict protest in Mexico when one considers the skill-building role of many churches as well as the power of religious symbols in political mobilization in Mexico.  Throughout the past thirty years Mexicans who have challenged government authority over issues of land redistribution, agricultural policy, government corruption, and economic issues have mobilized at churches and marched under banners depicting the image of Mexico’s patron, the Virgin of Guadalupe. Similarly, in the southern states of Mexico where liberationist themes were most welcomed and where Protestantism has its strongest base, it is not untypical for communities to be polarized politically along religious lines.  Consequently, in the Mexican case, higher levels of religious intensity predict the likelihood of protest behavior.  Moreover, because of the strong correlation between religious intensity and religious activity, this finding hints that skill-building through involvement in religious organizations and the local church may be playing a role as well. 

                        Table 9. OLS Regression Analysis Explaining Participation in Organizations, Protest (Mexico).


Variable
    Participation in Organizations
Unstandardized        Beta (ß)
Coefficient (B)
                 Protest
Unstandardized        Beta (ß)
Coefficient (B)
Protestant .13.08
(.28)
.12.08
(.25)
Devout .03*.06
(.02)
.03*.06
(.02)
Sex -.03*-.03
(.04)
-.02-.02
(.03)
Age -.00-.00
(.01)
.01.03
(.01)
Education .01.04
(.01)
.00 .01
(.01)
SES .04***.09
(.02)
.02.05
(.01)
Free Choice  .07**.08
(.03)
-.03 -.03
(.03)
Interest .01.02
(.02)
.03.05
(.02)
(Constant) .83
(.13)
1.00
(.12)
Adjusted  R .01  .00 
                      Source: Latinobarometer 2001.  Sample Size=978. * p <.1, ** p <.05, *** p<.01. 
                        Organization: “Do you participate in any organization?  How many?”
                        Protest: Indicates respondent has participated in a strike, demonstration, or sit-in.

Conclusion

As stated earlier, Mexico provides an interesting complement to studies of Protestantism elsewhere in Latin America because its evangelical population does not yet seem to have reached critical mass.  Nonetheless, in recent years the growth of Mexican Protestantism has begun to accelerate to the levels experienced in Chile, Guatemala, Costa Rica, and Brazil a decade ago.  If such growth continues, and if Protestant churches can maintain their members over time, then Mexican Protestantism may have long-term social and political consequences.

This study tested three variants of the “different-religions, different-attitudes” hypothesis in the Mexican context: do Protestants and Catholics differ in terms of political attitudes, engagement, and participation?  In addition, do variables other than religious affiliation (Catholic, Protestant) such as religious intensity (devout-ness) and religiosity (church attendance) influence individual political attitudes and behavior?  Little difference was found between Catholics and Protestants on a variety of political attitudes variables.  Nonetheless, the findings indicate that in general Protestants are significantly less likely to be politically engaged than are Catholics, yet religiosity has a positive influence on some measures of political engagement.  In other words, religious Catholics and Protestants tend be more interested in politics and more likely to follow politics than members of their faith who are not religious.

Interestingly, religiosity and religious intensity predicted several indicators of political participation, including voting, membership in organizations, and unconventional political behavior.  In fact, the religious intensity variable was the only variable tested to be a statistically significant influence on the likelihood of protest.  With regards to voting, in the 1994 and 1997 elections, Protestants were less likely to vote than Catholics, but by 2000 religious Protestants recorded the highest levels of voting of any religious category.

These findings also indicate that the not devout Protestant population significantly differs from devout Protestants and most Catholics on numerous variables.16  They are the most likely to be politically engaged, they seem to differ ideologically, and are, by far, the most likely to protest.  Moreover, results of OLS regression support the finding that those who are not devout are more likely to support democratic attitudes.  Although these samples of not devout and not religious Protestants are small, these findings are extremely interesting if the Kurt Bowen thesis is correct.  Bowen’s (1996) study of over 40 Mexican Protestant congregations found that Protestant churches cannot keep many of their converts and that over time these Protestants leave the church but do not return to Catholicism.  Bowen suggests that there are a number of reasons for this phenomenon that this survey data cannot verify.  In any event, in the next decade we may see a growing devout Protestant population that in turn will result in a growing not devout Protestant population.  The latter, based on Bowen’s argument and the evidence presented here, is a population that is politically engaged and willing to act, even in unconventional political behavior.

This study provides one of the few quantitative analyses of Mexican religion available.  The findings suggest that religious variables matter, and that, on occasion, religious intensity and/or religiosity are more powerful explanatory variables than religious affiliation.  This indicates that religious variables and their relationships to how citizens think and feel about politics need more careful consideration in studies of the Mexico and Latin America.
 

Appendix

The Mexico 2000 Panel Study
The México 2000 Panel Study is made up of nearly 7,000 interviews in five surveys taken across the 2000 election cycle.  Most of the data for this paper (with the exception of the question about voting in the 2000 election) are taken from the first round sample.  The survey was conducted February 19-27, 2000 by members of the polling staff of Reforma newspaper under the direction of Alejandro Moreno and polled a national cross-section of 2,400 individuals age 18 or older. 

Participants in the México Panel Study included (in alphabetical order): Miguel Basañez, Roderic Camp, Wayne Cornelius, Jorge Domínguez, Federico Estévez, Joseph Klesner, Chappell Lawson (Principal Investigator), Beatriz Magaloni, James McCann, Alejandro Moreno, Pablo Parás, and Alejandro Poiré.  Funding for the study was provided by the National Science Foundation (SES-9905703) and La Reforma newspaper.

The México 2000 Panel Survey distinguishes between various types of Protestant groups.  To the question “What is your religion?” various categories of Protestant are provided: Christian-not Catholic, Pentecostal, and Protestant.  For the purposes of this research, these respondents have been collapsed into “Protestant.”  Mormons, Seventh-Day Adventists, and Jehovah’s Witnesses have been excluded from the sample as have atheists, agnostics, and other religious groups.  The first wave of the Mexico 2000 Panel Survey included 2073 Catholics and 133 Protestants.  Of these, 1004 Catholics and 36 Protestants rarely attend church (“marginal”); 1049 Catholics and 95 Protestants attend church regularly (at least once per week).

Those who attend services are slightly poorer, less well educated, female, and slightly younger (60 vs. 65).
 

The Latinobarometer
The Latinobarometer is a well-known survey instrument produced by MORI in Santiago, Chile that reflects the beliefs of citizens throughout Latin America.  The survey is based on--and is extremely similar to--the Eurobarometer and World Values Surveys.  The sample population was adults over eighteen years of age.

The Latinobarometer distinguishes between various types of Protestant groups.  Variable 76 asks “What is your religion?” and provides numerous categories for Protestants: evangelical without-specifying, evangelical-Baptist, evangelical-Methodist, evangelical-Pentecostal, and Protestant.  For the purposes of this research, these respondents have been collapsed into “Protestant” and the terms “Protestant,” “evangelical,” and “Pentecostal” are used throughout the paper. Most Latin American Protestants would consider themselves “evangelical” for the term does not carry the political connotations that it does in the United States.  Likewise, many estimate that up to 90% of contemporary Latin American Protestants would consider themselves “Pentecostal.”  Mormons, Seventh-Day Adventists, and Jehovah’s Witnesses have been excluded from the sample as have atheists, agnostics, and other religious groups.  The 2001 Latinobarometer national samples were included 854 Catholics and 120 Protestants.

The Mexican study was conducted by MORI of México January 22 through February 11, 2001.  MORI Mexico used a mixed, probabilistic sample with age and sex (quota) controls.  The sample size is 1,200 cases and is 93% representative of the total population with a +/-2.9% margin of error.  Mexico’s sample was geographically divided as follows (number of cases in parentheses): Northwest (157), Northeast (99), South (138), West (134), Center (391), Metropolitan (100), Gulf (107), Southeast (40). 
 

Questionnaires
The questions below are the ones used in the presentation of data above.  Because each variable has a different number in each survey, I have not listed the data numbers for each survey employed.  These wordings and data numbers are from the 2001 Latinobarometer unless otherwise indicated.  If there is a significant variance in the question’s wording from year to year, it has been noted.  In brackets I have put the abbreviation used in tables to identify the variable.

(17) [Interpersonal Trust] Generally speaking, would you say that you can trust most people, or that you can never be too careful when dealing with others? (“You can trust most people,”  “you can never be too careful when dealing with others”)

(18) [Life Satisfaction] In general, would you say that you are satisfied with your life?  Would you say that you are very satisfied, fairly satisfied, satisfied, or not very satisfied?

(29) [Prefer Democracy] Which of the following do you agree with most? “Democracy is preferable to any other kind of government,” “In certain situations, an authoritarian government can be preferable to a democratic one,” “It doesn’t matter to people like me whether we have a democratic government or a non-democratic government.”

(32) [Electoral Corruption] In some countries people think that elections are clean; in other countries, people think that elections are rigged.  Thinking of the last elections, where would you place them in a 5 step ladder, where 1 is clean and 5 is rigged?

(33) [Free Choice] Some people think the way you vote has no influence over events; other people think the way you vote has a strong influence over events.  Using the scale in this card where “5” means the way that you vote has no influence and “1” means the way that you vote has a strong influence over events, where do you stand?

(52) [Left/Right Scale]  In politics, people speak of “left” and “right.”  On a scale where 0 is left and 10 is right, where would you place yourself?

(53) [Vote] Did you vote in the last municipal elections? …and did you vote in the last parliamentary elections?…and in the last presidential election?

(54) [Party Identification]  If there were an election tomorrow, which party would you vote for?

(59) I am going to read out a variety of political activities.  I would like you to tell me, for each one, if you have ever done any of them, if you would ever do any of them, or if you would never do any of them.
 a. Take part in a demonstration [Demonstrate]
 b. Block traffic [Block Traffic]
 c. Occupy land, buildings, or factories [Occupy]

[Protest] For use in regression analysis, I created a composite variable, “protest,” based on scoring each act of unconventional participation as “1.”  When Demonstrate, Block Traffic, and Occupy are added together, a four-point (0-3) “protest” scale results.

(63) [Interest]  How interested would you say you are in politics? Very interested, fairly interested, a little interested, or not interested?

(76) What is your religion?

(77) [Devout]  How would you describe yourself? Very devout, devout, not very devout, not devout at all.

The measurements of the demographic indicators are as follows: sex (female=0, male=1), SES17 (5=very good, 4=good, 3=average, 2=bad, 1=very bad), age (given in years by respondent), education (1=0-6 years, 2=7-9 years, 3=10-12 years, 4=more than 12 years of education (includes incomplete and complete university or technical school)).

Notes
1  i.e. María Teresa del Riego  “Apoyan indígenas operativas pro desarme” in La Reforma 30 January 2003, online version (www.reforma.com/nacional/ariculo/265645/deafault.htm) and “Alerta a Chiapas otra emboscada” in La Reforma 28 January 2003, online version (www.reforma.com/nacional/ariculo/266249/deafault.htm).   Return to reading.
2  In Latin America most Protestants are generally called “evangelicals” (evangélicos).  The term does not carry the conservative political connotations that it does in the United States.  I will follow the convention of using the terms interchangeably.  In addition, it is estimated that 75-90% of evangélicos are Pentecostal or charismatic in practice.    Return to reading.
3  This point has been made about the role of black churches developing leaders and efficacy during the American Civil Rights Movement.  See McAdam (1982) and Chong (1991).    Return to reading.
4  For instance, Bastian (1986, 1989) and Baldwin (1990).    Return to reading.
5  Data for this study come from the Mexico 2000 Panel Study (First Wave) and the 2001 Latinobarometer.  Details for both studies can be found in the Appendix.    Return to reading.
6  This is the highest growth rate in Latin America.  Even Guatemala’s meteoric growth in the 1980s was only about 15% per year, and Brazil and Chile in the 1980s and 1990s grew at 10% and 8% respectively.  For discussions of growth rates in various countries see Freston (2001), Corten (1999), Martin (1990), and Stoll (1990).    Return to reading.
7  However, a 1989 study by the Center for Anthropological Investigation of the Southeast (CIESAS) reported that in the poor southern zone of Mexico “Protestants represent 45 percent of the population of Campeche, 40 percent of Chiapas and Tabasco, 35 percent of Oaxaca, and 30 percent of Quintana Roo” (Bonner, 1998: 3).  Similarly, in 1993 the Foro Nacional de Iglesias Cristianas Evangélicas suggested that evangelicals make up 17.5 percent of the population (Bowen, 61).  Return to reading.
8  The 5.2% figure includes all Protestants but does not include Jehovah’s Witnesses, Seventh Day Adventists, or Mormons.  Even adding these sects to the equation does not push the number of Protestants over 10% of the populace.The census records 88% Catholic.  Source: Instituto Nacional de Estadística, Geografía, e Informática (www.inegi.mx).    Return to reading.
9  Interestingly, Protestantism has enjoyed a much warmer reception in the south of Mexico than in the north.  Siverts (1986) argues that this is due to the relative poverty of the indigenous peoples in the south, the population explosion of these people, increased missionary efforts by foreign churches, and the effect modernization has had in bringing new ideas to the region.  Kurt Bowen (1996) disagrees with parts of Siverts’ analysis, arguing that the south of Mexico is more likely embrace Protestantism for two reasons.  First, the South has always been socially, politically, and culturally isolated from much of Mexico, and parts of it (Chiapas) were originally part of Guatemala.  In other words, the Mexican-Catholic identity symbolized political oppression to largely indigenous population of the South and consequently the population of the periphery has been more likely to be interested in Protestantism (and liberation theology) than the core.  Moreover, throughout Mexico indigenous peoples have turned to Protestantism more than mestizo Catholics—and the South is the indigenous region.  Census data confirms these observations: Protestantism is strongest in the poor southern states such as Chiapas, Tabasco, Oaxaca, and Quintana Roo and is weakest around Mexico City and in the north.  Martines (1995) and Limón and Clemente (1995) report that indigenous Mexicans are twice as likely as mestizos to identify as Protestants, and Bonner (1998) argues that polls may undercount southern Protestants by as much as 75 percent.    Return to reading.
10  Kurt Bowen (1996) finds that the rapid growth of Mexican Protestantism is offset by another trend: Mexican Protestant churches cannot keep their adherents.  Bowen’s study of 48 Mexican Protestant churches indicated that 40 percent of adherents do not remain Protestant over the course of their lifetime, and their children are even less likely to remain in Protestant churches.10  Quentin J. Schultz (1994) makes a similar assessment--Protestant churches are protectorates of the poor.  The children and grandchildren of Protestant converts often achieve literacy and social mobility, in part due to their religious environment, but no longer feel the need for the Protestant faith and leave the congregation.  Thus, Protestantism may have significant short-term effects on the life of believers whereas the long-term political and social effects are less apparent.    Return to reading.
11  The following are the numbers of respondents in the sample for the 2001 Latinobarometer.  Devout Catholic= 449, Devout Protestant= 88, Not Devout Catholic=403, Not Devout Protestant=32.  The first wave of the Mexico 2000 Panel Survey included 2073 Catholics and 133 Protestants.  Of these, 1004 Catholics and 36 Protestants rarely attend church (“not religious”), 1049 Catholics and 95 Protestants attend church at least once per week (“religious”).    Return to reading.
12  All of the OLS analyses in the study included interaction terms to see if there was an interaction effect between religious affiliation and religious intensity.  The only model that had a significant interaction term was Life Satisfaction.  The interaction term was modest but negative and significant which slightly diminishes the power of the Protestant explanatory variable.  However, it is also possible that this is a false positive as a false positive is not unusual when one runs numerous regression models.    Return to reading.
13  The Mexico 2000 Panel Survey included one additional engagement variable, an assessment of electoral corruption.  Religious variables were not significant; the only significant predictor tested was age: the older one is the more likely they are to view elections as corrupt.    Return to reading.
14  Few studies have studied the party preferences of Mexican Protestants.  One that has is Scott’s study of Protestants in Mexico City (1992).  Scott found few differences between the voting patterns of Protestants and Catholics in the 1988 elections.  He found that Protestants tended to vote for the PRI and explained their choice as the PRI guaranteeing a level of religious freedom over time.  Interestingly, a third of Protestants voted for the left-leaning PRD, with only a small margin voting for the PAN.  Freston (2001) argues that the strong showing for the PRD is not surprising since most Protestants come from the poorer classes (205-206).    Return to reading.
15  As in Brazil and Chile, the data is suggestive that interest and SES both have positive associations with protest as well, but the findings were not statistically significant (.10<p<.20).    Return to reading.
16  These cleavages are important because over two-thirds of Mexican Protestants consider themselves to be devout as compared to under half of Mexican Catholics.    Return to reading.
17  This is a change from the original Latinobarometer format which reversed this scale.  The same is true for the measure of political interest.  For all regression analysis in this study, scales have been modified to go from low to high numerically.  Hence, “very” might be scored ‘4’ and “not at all” ‘1’.    Return to reading.

References
Amatulli Valente, Flaviano. Diálogo con los protestantes.  Mexico City: Apóstoles de la Palabra, 1997.

Baldwin, Deborah J. Protestants and the Mexican Revolution: Missionaries, Ministers, and Social Change. Urbana: University of Illinois Press, 1990.

Bastian, Jean-Pierre. Breve Historia del Protestantismo en América Latina. Mexico: Casa Unida de Publicaciones, S.A., 1986.

-----.  La mutación religiosa de América Latina. Mexico City: Fondo de Cultura Económica, 1997.

-----. Los disidentes: sociedades protestantes y revolución en Mexico (1872-1911). Mexico Cita : Fondo de Cultura Económica (El Colegio de Mexico), 1989.

-----.  Protestantismo y modernidad latinoamericana: Historia de unas minorías religiosas activas en América Latina. Mexico City: Fondo de Cultura Económica, 1987.

-----. "Protestantismo popular y política en Guatemala y Nicaragua," Revista Mexicana de Sociología 48,3, 1986. pp. 181-199.

Blancarte, Roberto J., ed. Religión, Iglesias, y Democracia. Mexico: La Jornada, Ediciones, 1995.

Bonicelli, Paul Joseph. Serpents, doves and sons of thunder in Mexico: Comparing the politics of Latin American Evangelicals. PhD dissertation, University of Tennessee, Knoxville, 1993.

Boudewijnse, Barbara, Droogers, Andre, & Kamsteeg, Frans, eds. Algo más que opio: una lectura antropológica del pentecostalismo latinoamericano y caribeño. San José, Costa Rica: Departamento Ecuménico de Investigaciones, 1991.

Bowen, Kurt. Evangelism and Apostasy: The Evolution and Impact of Evangelicals in Modern Mexico. Montreal: McGill-Queen’s University Press, 1996. 

Brady, Henry, Sidney Verba, and Kay Lehman Schlozman. “Beyond SES: A Resource Model of Political Participation” in American Political Science Review, vol. 89, no. 1 (1995).

Camp, Roderic Ai. Politics in Mexico, 3rd ed.. New York: Oxford University Press, 2001.

-----. “The Cross in the Polling Booth: Religion, Politics, and the Laity in Mexico” in Latin America Research Review, vol. 29, no. 3 (1994).

Camin, Hector Aguilar and Lorenzo Meyer. In the Shadow of the Mexican Revolution. Austin: University of Texas Press, 1993.

Canto Chac, Manuel. “Cristianos y Democracia (Notas Sobre los Referentes de la Participación Política de los Cristianos en México.” In Religión, iglesias y democracia. Ed. Roberto J. Blancarte.  México: La Jornada Ediciones, Centro de Investigaciones Interdisciplinarias en Humanidades/UNAM, 1995.

Clawson, David L. “Religious Allegiance and Economic Development in Rural Latin America” in Journal of Interamerican Studies and World Affairs, vol. 26, 1987.

Cleary, Edward. Born of the Poor. Notre Dame: University of Notre Dame Press, 1990.

-----. "Human Rights Organizations in Mexico: Growth and Organization," Journal of Church and State 37,4, 1995.

Cleary, Edward and Stewart-Gambino, Hannah W. Conflict and Competition: The Latin American Church in a Changing Environment. Boulder, CO: Lynne Rienner, 1992.

-----. Power, Politics, and Pentecostals in Latin America. Boulder, CO: Westview Press, 1997.

Diamond, Sarah. Spiritual Warfare: The Politics of the Christian Right. Boston, South End Press, 1989.

Domínguez, Jorge I., ed. The Roman Catholic Church in Latin America. New York: Garland Publishing, 1994.

Eckstein, Susan, ed. Power and Popular Protest. Berkley: University of California Press, 1987.

Escobar, J. Samuel. “The Promise and Precariousness of Latin American Pentecostalism” in Coming of Age: Protestantism in Contemporary Latin America, Daniel R. Miller (ed.). New York: University Press of America, 1994.

García-Ruiz, J. “De la identidad aceptada a la identidad elegida: el papel de lo religioso en la politización de las identificaciones étnicas en Guatemala” in Estudios Sociológicos de El Colegio de México. 10, 30.

Garma, Navarro.  Poder, conflicto, y reelaboración simbólica: protestantismo en una comunidad totonaca. Thesis, Escuela Nacional de Antropología e Historia, México. 1994.

-----. Pentecostisme rural et urvain au Mexique: differences et similitudes. Social Compass, 39, 3.

Garrard-Burnett, Virginia and Stoll, David, eds.  Rethinking Protestantism in Latin America. Philadelphia: Temple University Press, 1993.

Gill, Anthony J.  “The Struggle to be Soul Provider: Catholic Responses to Protestant Growth in Latin America” in Smith, Christian and Joshua Prokopy, eds.  Latin American Religion in Motion.  New York: Routledge, 1999.

Glazier, Stephen, ed. Perspectives on Pentecostalism: Case Studies from the Carribean and Latin America. Washington: University Press of America, 1980.

Grayson, George W. The Church in Contemporary Mexico. Washington, D.C.: The Center for Strategic Studies, 1992.

Guerra, Rubén Ruiz. ‘Protestantismo y democracia en Mexico’ in Roberto J. Blancarte, ed. Religión, iIglesias, y democracia. México: La Jornada Ediciones, 1995.

Inglehart, Ronald.  “The Renaissance of Political Culture” in American Political Science Review, vol. 82, no. 4 (December, 1988).

Inglehart, Ronald, Neil Nevitte, and Miguel Basanez. The North American Trajectory: Social Institutions and Social Change. New York: Aldine de Gruyter, 1996.

Jenkins, Phillip. The Next Christendom: The Coming of Global Christianity. Oxford: Oxford University Press, 2002.

Lalive D’Espinay, Christian. Haven of the Masses: A Study of the Pentecostal Movement in Chile. Marjorie Sandle, trans. London: Lutterworth Press, 1969.

Latinobarómetro, Opinión Pública Latinoamericana. 1995-2000. Computer File. Marta Lagos, Principal Investigator. 

Limón, Francisco and Abel Clement. “Iglesias históricas y pentecostales: Espíritu de lucha social y participación política en Mexico: El Ccaso de Chapas” in Benjamín Gutiérrez, ed. En la Fuerza del Espíritu : Los Pentecostales en América Latina : Un desafío a las iglesias históricas. Mexico City : Aipral, 1995, p. 175-197.

Lynch, Edward A. “Reform and Religion in Latin America” in Orbis, Spring 1998.

Macín, Raúl. Los derechos de las minorías religiosas en Mexico. Mexico City : Claves Latinoamericanas, 1991.

-----. Lutero: Presencia religiosa y política en México. Mexico City : Ediciones Nuevomar, 1983.

Mariz, Cecilia. “Religion and Poverty in Brazil: A Comparison of Catholic and Pentecostal Communities” in Sociological Analysis, vol. 53, 1992.

Martin, David. Pentecostalism: The World Their Parish. Oxford: Blackwell, 2002.

-----. Tongues of Fire: The Explosion of Protestantism in Latin America. Oxford: Blackwell, 1990.

Medellín, Martín de la Rosa. El impacto social del protestantismo : Metolodogía para el estudio e las religiones. Mexicali : Universidad Autónoma de Baja California, 1989.

Mexico 2000 Panel Study. Computer File.  Chappell Lawson, Principal Investigator.

Miller, Daniel R., ed. Coming of Age: Protestantism in Contemporary Latin America.Lanham, MD: University of America Press, 1994.

Muñoz, Milca de la Rosa. ‘El papel de las mujeres en la difusión de tres grupos pentecostales en Banderilla’ in Las interacciones sociales y el proselitismo religioso en una ciudad periférica. Felipe Vázquez Palacios, ed. Hidalgo y Matamoros, México: Centro de Investigaciones y Estudios Superiores en Antropología Social, 1999.

Palacios, Felipe Vázquez, ed. Las interacciones sociales y el proselitismo religioso en una ciudad periférica. Hidalgo y Matamoros, México : Centro de Investigaciones y Estudios Superiores en Antropología Social, 1999.

Patterson, Eric. “Religious Change and Politics: Comparing Protestants in Brazil and Chile” in Iberoamericana: Nordic Journal of Latin American and Caribbean Studies. vol, 31, 1 (2001).

Parker, Cristián. Popular Religion and Modernization in Latin America. Maryknoll, NY: Orbis, 1996. 

Pérez, Martín Fredy. "Sectas y crecimiento en México." Servicio Universal de Noticias, January 19, 1999.

Schmitt, Karl. "Church and State in Mexico: A Corporatist Relationship," Américas 40, 1984.

Scotchmer, David G. “Life of the Heart: a Maya Protestant Spirituality” in South and Meso-American Native Spirituality: from the Cult of the Feathered Serpent to the Theology of Liberation, Gary H. Gossen, ed. New York: Crossroad, 1993.

Scott, Lindy.  Salt of the Earth: A Socio-Political History of Mexico City Evangelical Protestants (1964-1991). PhD dissertation, Northwestern University, Evanston, IL, 1991.

Sherman, Amy. Preferential Option: A Christian and Neoliberal Strategy for Latin America’s Poor. Grand Rapids, MI: Wm. B. Eerdman’s Publishing, 1992.

Sigmund, Paul, ed. Religious Freedom and Evangelization in Latin America. Maryknoll, NY: Orbis, 1999.

Smith, Christian and Joshua Prokopy.  Latin American Religion in Motion.  New York: Routledge, 1999.

Steigenga, Timothy. The Politics of the Spirit: The Political Implications of Pentecostalized Religion in Costa Rica and Guatemala. Lanham, MD: Lexington Books, 2001.

Stoll, David. Is Latin America Turning Protestant? Berkley: University of California Press, 1990.

Swatos, William J., ed. A Future for Religion? New Paradigms for Social Analysis. Thousand Oaks: Sage Publications, 1993.

Talavera, Arturo Fontaine and Harald Beyer. The Evangelical Movement in Chile. Grand Rapids: Acton Institute Press, 1998.

Valente, P. Flaviano Amatulli. Protestantismo en México: Hechos, interrogantes, y retos. México: Apóstoles de la Palabra, 1994.

Verba, Sidney, Kay Lehman Schlozman, and Henry Brady. Voice and Equality. Cambridge, MA: Harvard University Press, 1995(b).

Weber, Max. The Protestant Ethic and the Spirit of Capitalism, trans. by George Allen,  New York: Charles Scribner’s & Sons, 1958.

Willems, Emilio.  Followers of the New Faith.  Nashville, TN: Vanderbilt University Press, 1967.
 

            Last updated December 27, 2004