THE CATHOLIC CHURCH: AN INSTITUTION

The Catholic Church serves as an excellent case study for institutional responsiveness to the Latino population. Even though it is a faith-based organization, it remains a proto-typical institution; it has “rules and organized practices,” provides specific roles and behavioral expectations for members, is a source of identity and belonging, and creates a common purpose that can bind communities together (March and Olsen Reference March, Olsen, Rhodes, Binder and Rockman2008, 3).Footnote ² Further, these rules, meanings, and expectations are resilient and lasting through time, almost regardless of the individuals who make up the congregations or the leadership. Recognizing the Church as an institution then allows for a systematic approach to analyze its responsiveness to Latinos in a way that can easily be applied to other (non-religious) organizations.

A primary consideration to this study concerns whether or not such responsiveness is in fact a goal of the institution in question. Serving Latinos has repeatedly been asserted as a goal of the Catholic Church. The United States Conference of Catholic Bishops” (USCCB) has a subcommittee on Hispanic Affairs that is dedicated to serving and promoting the needs of Latino Catholics. The USCCB wrote and promotes the National Pastoral Plan for Hispanic Ministry that outlines best practices for serving Latinos at the local level, and has repeatedly worked to be in conversation with Latino leaders and laity about their needs within the Church (USCCB 1988; USCCB 2002).Footnote ³

While the rhetoric of the Church values Latinos, there is no doubt that a better indicator of an organization's goals comes from its behavior (Meadows Reference Meadows2008). So then, while the Catholic Church states that it strives to serve and include Latinos, the only way to determine if that is an actual goal is to see whether church behavior works towards that end. This would be seen in parish responsiveness to Latinos (addressed below), a behavior that is often directed by a priest or pastor whose behaviors are both influenced and constrained by the expectations of the church (Boudinhon Reference Boudinhon1910; March and Olsen Reference March and Olsen1984).

Beyond institutional goals, the necessity for survival also defines the requirements of individual roles within an organization (March and Olsen Reference March and Olsen1984; Wilson Reference Wilson1995). Leaders are specifically tasked with maintaining membership levels so the organization can continue. This can be difficult within the Church given the voluntary nature of membership, since priests have “neither the effective power nor the acknowledged right to coerce the members” (a fact attested to by the presently record high attrition rate, March and Olsen Reference March, Olsen, Rhodes, Binder and Rockman2008, 13; Pew Research Center 2018). Instead they must incentivize membership by providing the types of services and activities potential members find appealing. So then, if churches value the goal of serving Latinos in practice, or if leaders simply recognize the necessity of Latino membership for church survival (recognizing the two conditions are not mutually exclusive), we would find a high percentage of churches offering both Spanish mass and other culturally relevant services given the many incentives the Church has as an institution to respond to Latinos.

DEFINING CHURCH SERVICE TO LATINOS

The exact form of service will vary by institution and the needs of the group to whom the institution is responding. The first indication of welcoming to Latino populations into a Catholic church is if there is a weekly Spanish mass. There are two components to this assertion: the decision to identify mass as a first step, and the qualification that it be in Spanish. First, mass attendance is a weekly obligation for faithful Catholics, and is the foundation of Catholic life. As such, the mass is the first and most essential service that a church can offer believers. Second, I assert that the mass must be offered in Spanish. Some may critique this measure as being applicable only to Spanish-speaking immigrants. However an estimated 75% of all Latinos speak Spanish at home (not just immigrants), making the offering of Spanish mass a widely accessible service for the Latino population at large (Krogstad, Stepler, and Lopez Reference Krogstad, Stepler and Hugo Lopez2015).Footnote ⁴ Further, those engaged in Hispanic Ministry indicate that Spanish for many Latinos is the “language of the heart,” meaning that regardless of how much Latinos use Spanish in their daily lives, they have a special connection to ritual and faith that is only accessed through Spanish, even if they are fluent or even dominant in English (Huckle Reference Huckle2016; Matovina Reference Matovina2011). With this understanding, Spanish mass, specifically, becomes an appropriate and essential first marker of a church's willingness and ability to invite in Latino congregants and help them build a community that will result in their socialization.

The inclusion of Latinos, per the Bishops and Hispanic Ministers, must not stop with the Spanish Mass, alone. Instead it must also extend to a variety of services and activities that help Latinos feel at home in the Church given their unique culture, language, and worship practices (Hoover Reference Hoover2014; Matovina Reference Matovina2011; USCCB 1988). These recommendations are based off of the findings of the three Encuentros (national meetings amongst parish leaders ministering to Latino Catholics) held by the Secretariat for Hispanic Affairs in the USCCB between 1974 and 1985. The purpose of the Encuentros was to devise an action plan for improving service to the burgeoning Latino community (Matovina Reference Matovina2011; Paulson Reference Paulson2014);Footnote ⁵ the resulting recommendations were published in 1988 as the National Pastoral Plan for Hispanic Ministry (USCCB 1988).Footnote ⁶ It called for not only welcoming Latinos as guests, but making them at home in their parishes (de ser lugar a ser hogar) by providing a variety of culturally relevant services and fostering Latino leadership (USCCB 1988; USCCB 2002). Services included providing sacraments (such as marriage and baptism) and Bible study in Spanish, as well as offering other activities such as Marian celebrations, Guadalupana societies, and quinceñera celebrations. Importantly, increasing lay leadership also facilitates these types of responsiveness because they do not require the presence of a priest for the completion of a sacrament.

This is not to say that Latinos cannot or do not also participate in non-Latino specific groups or activities. Yet, Latinos choose to attend parishes that respond to them in this culturally appropriate way. An estimated 47% of Latinos drive past other churches on their way to their parish of choice, as compared with 30% of Anglo Catholics (Ospino Reference Ospino2014). This is perhaps because only 15% of all parishes in the United States have Hispanic ministries. Further, the more Latinos there are in a parish, the more likely they are to attend Mass in Spanish (Ospino Reference Ospino2014). All told, offering Spanish mass and culturally targeted services are an integral part of the Church's institutional response to Latinos because Latinos could attend the “regular” services of a given parish that are not targeted to their language or culture—but they simply do not. As such, the types of services churches provide to Latinos do matter, and a lack of culturally appropriate services fails to draw Latinos in.

WHAT PREDICTS INSTITUTIONAL SERVICE PROVISION

Theories of constituency response and descriptive representation offer a logical mechanism for examining patterns in which Catholic parishes offer these culturally relevant services to Latinos. Representation theories hold that the demographic characteristics of service providers are highly influential in the type and quality of service constituent and group members receive, especially in the cases of ethnic and economic minority populations (Hindera Reference Hindera1993; Meier Reference Meier1975; Pitkin Reference Pitkin1972). Translated to the Catholic Church, this means that Latino ministers (priests and deacons) will be more likely to offer and promote the types of services and activities that would be both useful and appealing to other Latinos, generally speaking. Historically, Catholic populations have benefitted from this type of descriptive representation. The ethnic parishes of the 19^th and early 20^th centuries focused on serving specific cultural communities, and that service was predicated upon the presence of a co-ethnic priest who was familiar with their customs, language, and worship practices (Burns Reference Burns1969; Cogley Reference Cogley1973; Dolan Reference Dolan, Miller and Marzik1977; McAvoy Reference McAvoy1969).

Such representation is not currently possible for Latinos who constitute 40% of the American Catholic population, but only 8% of American clergy (USCCB 2012; Ziegler Reference Ziegler2011).Footnote ⁷ This makes it increasingly important for churches to practice substantive representation, where services are provided based upon what people need, rather than according to the characteristics of those who serve. Indeed, Catholicism advocates for exactly this type of response, mandating that all Catholics respond to the spiritual, social, and material needs of those around them without regard to race, status, or even religion (Catholic Charities of St. Paul and Minneapolis 1999). While this idea is not uniquely Catholic (it is a concept espoused in some form by most Christian denominationsFootnote ⁸) it nevertheless significantly informs Catholic theology. As such, we should expect Catholic churches to respond substantively to Latinos, regardless of the presence or absence of Latino clergy in any given community.

The Catholic bishops have made specific recommendations on how parishes should enact such responsiveness by providing Spanish massFootnote ⁹ and facilitating the organization of culturally unique religious celebrations, such as the Feast of Our Lady of Guadalupe, among other things.Footnote ¹⁰ The Church also acknowledges that co-ethnic pastors are preferred, though not required to ensure service to Latino congregants (USCCB 2002). In short, the Catholic Church as an organization has broadly promoted substantive representation for Latino Catholics, and advocated for the provision of services and activities that have been shown to positively impact political participation outcomes, even though history and research on other public service institutions indicate that descriptive representation may be a better predictor of service. To the extent that other churches, religious, and voluntary organizations also have a tradition or practice of caring for their population and communities, the expectations around the Catholic Church should easily export to other community-based institutions.

Constituency response offers another parsimonious theory for predicting when churches will offer services to Latinos, arguing simply that churches will respond to Latinos when they are part of the congregation or parish neighborhood (March and Olsen Reference March and Olsen1984; Pitkin Reference Pitkin1972; Wilson Reference Wilson1995). This expectation is bolstered by the territorial nature of parishes and the responsibilities of pastors.Footnote ¹¹ According to Canon Law (the set of laws that outline the organization of the Catholic Church, its governance, liturgy, and worship practices), pastors must “strive to know the faithful entrusted to his care” (Canon 529), in reference to all who live within the parish, including those who are not necessarily church members. Combining constituency response with an operational understanding of parishes translates into the theoretical expectation that a parish with a high population of Latinos should be more likely to offer Spanish mass and other culturally relevant small group meetings or activities than a parish with a smaller Latino population. The high rate of Catholicism among Latinos (48%, Pew Research Center 2018) also suggests that a large Latino population will necessarily indicate the presence of a large Latino Catholic population, increasing the motivation for parishes to respond with services and activities that target Latinos, specifically.

THE POTENTIAL IMPACT OF SERVICE PROVISION

Church responsiveness to Latinos may have an impact beyond the parish boundaries. Scholars have identified a positive correlation between religious engagement and political participation (Brady, Verba, and Schlozman Reference Brady, Verba and Lehman Schlozman1995; Djupe and Gilbert Reference Djupe and Gilbert2006; Djupe and Neiheisel Reference Djupe and Neiheisel2012). One body of research focuses on church attendance—how often one attends religious services at their house of worship—believing that higher levels of attendance serves as a proxy for a higher exposure to political messaging in churches, or indicates a higher devotion to the moral teachings of one's religion (some of which promote the importance of civic engagement; DeSipio Reference DeSipio and Matthew Wilson2007; Jones-Correa and Leal Reference Jones-Correa and Leal2001; Kelly and Kelly Reference Kelly and Morgan Kelly2005). It is also possible that belonging to a religious community can develop higher levels of social capital and linked fate, which could positively incline members to participate politically (Brinig and Garnett Reference Brinig and Garnett2014; Putnam Reference Putnam2001; Putnam and Campbell Reference Putnam and Campbell2010; Verba, Schlozman, and Brady Reference Verba, Schlozman and Brady1995).

Religious participation through leadership is thought to provide opportunities to learn and practice civic skills, which could have a more direct impact on political participation. It is theorized that those who run worship services, small group meetings, or any other volunteer activity through the church have the opportunity to learn how to organize, engage, and lead (Brady, Verba, and Schlozman Reference Brady, Verba and Lehman Schlozman1995). These are skills that can be transferred easily into the political realm.Footnote ¹² Yet such skills can also be gained simply by participating in small groups and other activities outside of worship services (Djupe and Neiheisel Reference Djupe and Neiheisel2012), and the effect of such small group engagement on levels of political participation was found to be stronger and more consistent than the effect of holding leadership positions or attending religious services. Importantly, there are significantly more opportunities to participate in a small group than there are to take on leadership roles, further increasing the potential significance of this form of church participation.

Lastly, churches are also significant for passing on political knowledge to congregants either from the pulpit, or through conversations among community members (Jones-Correa and Leal Reference Jones-Correa and Leal2001; Neiheisel, Djupe, and Sokhey Reference Neiheisel, Djupe and Sokhey2009). These conversations also serve to communicate the societal norms that “form a nucleus of conformity, stability, and social order in communities that would otherwise be volatile” (Cnanna, Boddie, and Yancey Reference Cnanna, Boddie, Yancey and Smidt2003, 20). Such norms are an integral part of the social-political context as they transmit public values, including the expectation for participation. Such social norms are particularly pertinent to predicting behavior since “many political activities involve locally based social interaction” (Huckfeldt Reference Huckfeldt1979, 580; Wilson Reference Wilson1995).

These conclusions are specific examples of the continued finding that churches continue to operate as a significant socializing institution in the United States. By participating in church activities, individuals have the opportunity to both learn about politics as well as how to participate. Yet these benefits cannot be gained if churches are not creating opportunities for engagement in the first place. As such, it is important to identify the factors that make Church service provision to Latinos more or less likely.

ARGUMENT AND EXPECTATIONS

This project uses theories of constituency response and descriptive representation to predict when Catholic churches, as institutions, will respond to and serve Latinos as an increasingly large portion of the American Catholic population. Constituency response argues simply that churches will respond to Latinos who either live within the parish bounds or are a part of the congregation as their constituents (March and Olsen Reference March and Olsen1984; Pitkin Reference Pitkin1972; Wilson Reference Wilson1995). Further, due to their territorial nature, parishes with a high population of Latinos should be more likely to offer Spanish mass and other culturally relevant small group meetings or activities than a parish with a smaller Latino population. This is augmented by the high rate of Catholicism among Latinos (48%, Pew Research Center 2018), which also suggests that a large Latino population will necessarily indicate the presence of a large Latino Catholic population, specifically, increasing the motivation for parishes to respond with services and activities that target Latinos.

Institutions respond to more than the simple presence or absence of members. The composition of their leadership has also been found to impact patterns of responsiveness. Specifically, representation theories hold minority populations are better served by leaders who share their ethnic or cultural background (Hindera Reference Hindera1993; Meier Reference Meier1975; Pitkin Reference Pitkin1972). Translated to the Catholic Church, this would mean that Latino ministers (priests and deacons) are more likely to offer and promote the types of services and activities that would be both useful and appealing to other Latinos. However, Catholicism mandates that all Catholics respond to the spiritual, social, and material needs of those around them without regard to race, status, or even religion (Abell Reference Abell1960; Coleman Reference Coleman and O'Brien Steinfels2004; Catholic Charities of St. Paul and Minneapolis 1999). In other words, Catholicism advocates for substantive rather than descriptive representation. However, numerous studies have found that minority populations within a variety of service institutions with similar mandates continue to benefit from having a co-ethnic leader (Hindera Reference Hindera1993; Meier and Nigro Reference Meier, Nigro, Dolan and Rosenbloom2003; Meier et al. Reference Meier, Doerfler, Hawes, Hicklin and Rocha2006; Pantoja and Segura Reference Pantoja and Segura2003). As such, it is logical to hold that descriptive representation will offer similarly positive benefits to Latinos within the Catholic Church.

Lastly, because of the theories of representation hold that Latino ministers (priests and deacons) will be more likely to offer and promote the types of services and activities that would be both useful and appealing to other Latinos, generally speaking, then the presence of a Latino minister should be viewed as a sign of service in and of itself.

DATA AND METHODS

This paper seeks to determine which factors predict Catholic church service provision to Latinos given the Church's position as a major institution at the forefront of U.S. demographic change. This understanding may help other institutions in the future as they come to deal with the reality of an increasing Latino population. While the Church clearly advocates for a strong response to and inclusion of Latinos, Church leaders are influenced by more than the stated goals of the Church. Other well-known factors may sway their individual decision about whether or not to provide relevant services to Latinos: the size of the Latino population, and levels of descriptive representation within a given church community.

As such, this study utilizes a unique data setFootnote ¹³ containing information about the services offered by Catholic churches in three areas, (Seattle, WA; San Antonio, TX; and the Diocese of Yakima, WA) combined with U.S. Census data regarding regional demographics.Footnote ¹⁴ I use the Diocese of Yakima (rather than the city) in order to make Yakima's sample size closer to that of San Antonio and Seattle, increasing the analytical leverage of the data.

The three chosen regions also present distinct cultural contexts. San Antonio is a historically Latino dominant area. Including it here for analysis allows me to juxtapose the responsiveness of the Catholic Church to Latinos who make up a majority—rather than minority—of their congregant and local population. In contrast, Seattle is a metropolitan city with a fairly small (6.6%) Latino population. The small size of the Latino population makes Seattle an excellent study for Church responsiveness to the needs of a much smaller ethnic population, and is also representative of the majority of Latino congregations across the United States where Latinos do not constitute a significant portion of the population. Lastly, I chose to include the Diocese of Yakima, a largely rural area covering almost half of Eastern Washington, because its Latino population grew to a significant size (from approximately 6% in 1970 to 31% in 2010) only in the last few decades. As such this provides an opportunity to measure how well Catholic churches respond to a rapid regional demographic change and the accompanying unique needs of a new population.

I first created a list of all churches in the three designated areas from the relevant Archdiocesan websites,Footnote ¹⁵ and then conducted an online search to verify each church's basic information, such as their address, phone number, and website (if available). From the websites and digitalized parish bulletins I identified each church's pastor, other priests or deaconsFootnote ¹⁶ also serving the parish, mass times and their languages. I also collected information on any Latino-specific services offered in each parish (discussed below). All data was verified via phone calls to parish offices.Footnote ¹⁷

Data from the 2015 American Community Survey is used to account for the demographics of each church's zip code. I include two variables measuring zip code wide demographics, age and poverty level, that could directly affect the willingness or ability of churches to respond to their local populations. Because the focus of this study is on the responsiveness of churches to the Latino population, specifically, I gathered data on the size of the Latino population, as well as the estimated change in the Latino population from 2010 to 2015 in order to see if churches respond to Latino population growth. I also include the rate of homeownership, median income, employment rate, and educational achievement for Latinos, specifically, in order to see how these community-specific elements may impact church responsiveness. Church information and Census data were joined by zip code using GIS software.

Zip code operates as a proxy measure for a parish. In many Archdioceses there is one Catholic parish per zip code (such as Seattle) and, in decades past, households’ parish membership was determined by zip code rather than personal preference. Where parish boundaries are not determined by zip code, zip defines a reasonably sized area within which parishioners can attend church services without constraints in traffic, limits in transportation, or convenience.Footnote ¹⁸ As such the zip code provides a realistic approximation (if not definition) of parish boundaries, facilitating my calculation of neighborhood demographics.

I measure descriptive representation by the presence of a Latino pastor, priest, or deacon. Because there are so few Latino priests (and priests in general) I include the presence of Latino deacons as a marker of service in my analysis because the appointment of deacons when there are not sufficient priests to fulfill the pastoral duties of a parish demonstrates an effort to provide Latino ministerial support.

My dependent variables measure the presence of Spanish (or bilingual) mass, and any type of social, cultural, or material service or activity directed at Latinos. This is because the services churches provide to Latinos must also be the services that are applicable or desirable to Latinos in the first place, as they will not join or participate in church communities that have little interest to offer them. To that end, I include any small group meeting, activity, volunteer opportunity, or service that is directly applicable to a Latino community as a service outside of mass. Examples include Spanish Bible study, quinceñera celebrations, Marian festivals, Guadalupana societies, immigration or legal support, or the presence of an office of Hispanic Ministry. For simplicity, I refer to these as “cultural services.”

I also calculate the effect of Latino population density on the likelihood of having a Latino pastor, priest, or deacon given the importance of co-ethnic representation on the functioning of ethnic parishes. I treat Latino minister,Footnote ¹⁹ Spanish mass, and services as dichotomous variables where the simple presence of any Latino leader, mass, or service generates a one.

My initial analysis calculates the observed rates of service and representation using census data for population numbers for each city or region. I then use logistic regression to produce predicted probabilities of the outcome variables.

FINDINGS

One way of determining if institutions are responding appropriately to Latinos is to see if their rate of service provision is on par with Latinos' share of the population. As the Latino population grows, so too should the number of churches working to serve them. For example, if Latinos were to constitute 25% of the population, then we should anticipate that roughly 25% of churches would offer Latino targeted services. Likewise if the population were 75% Latino, then the percentage of churches offering services should be closer to 75% than 25%. In examining the observed data, it appears that churches are indeed responsive to the size of Latino populations. Across my sample, the mean Latino population size is 50%, while 67% of churches in the observed areas have Spanish mass, 34% offer some form of Latino-specific services, and 58% have a Latino pastor, priest, or deacon (see Table 1). This first assessment suggests that Catholic churches are responding appropriately with services to a growing Latino population in spite of the shortage of Latino priests.

Table 1.

Demographic and descriptive statistics

Data compiled from 2015 American Community Survey. U.S. Census Bureau's American Community Survey Office, 2015; Center for Applied Research of the Apostolate (CARA); and author's original dataset.

Disaggregating the data presents a more nuanced story. The majority of churches in San Antonio and Yakima provide mass in Spanish (79% each). It may be possible to explain this by the large Latino presence in these areas, since a larger Latino population should prompt churches to offer them more services. Yet only 35.7% of Yakima's population is Latino, as compared with the estimated 72.6% of San Antonio. The majority of churches in both cities also have a Latino minister (priest or deacon) to serve their Latino populations. However, no city reports a high rate of offering any other cultural service (as described above), even though these services can be led by any willing layperson (see Table 1).

Next I subject my data to a more rigorous analysis in order to test the relationship between the outcome variables and the relevant covariates. I hypothesized that as the Latino population increases, so too would the likelihood that churches would offer Spanish mass and other Cultural Services. The regression analyses do not support this hypothesis (see Table 2). While the coefficient is positive in the regression on Spanish mass, it is not statistically significant. Latino population size is negative and non-significant in regressing on Cultural Services (see Figure 1).

Table 2.

Regression models on Spanish mass, cultural services, and Latino minister

Robust standard errors in parentheses.

***p < .001, **p < .01, *p < .05, p < .1.

Figure 1.

Probability of Spanish Mass and Services.

My second hypothesis held that the presence of a Latino minister would correlate with a higher rate of both Spanish Mass and Cultural Services. The coefficient for Latino Minister is positive and significant in both models (see Table 2), demonstrating that theories of descriptive representation hold when examining the behaviors of religious institutions (see Figures 2 and 3).

Figure 2.

Probability of Spanish Mass by Latino Population Density and Latino Minister.

Figure 3.

Probability of Cultural Services by Latino Population Density and Latino Minister.

Representation theories suggest that having co-ethnic priests and deacons will increase the rate of service to Latinos, making the presence of a co-ethnic leader in and of itself an act of service to the Latino community (a proposition partially supported by the analyses above). My third hypothesis held that churches would be more likely to offer this service to a larger Latino community. However, the regression results demonstrate that the presence of a Latino minister is in fact statistically unrelated to the Latino population size. The only variable to produce a marginally significant coefficient is Seattle, and it is negative—surely a reflection of the observed smaller Latino population (both clergy and laity) there as compared with San Antonio and Yakima.

It is entirely possible that constituency response and descriptive representation are insufficient on their own for understanding institutional responsiveness to minority populations. This is particularly true for the Catholic Church given its shortage of priests—there may simply not be enough Latino ministers to serve in all of the parishes with sizable Latino communities. This does not preclude the possibility that once Latino priests are placed, they may do more to proactively respond to their surrounding Latino community than a non-Latino minister. In order to test this possibility, I created an interaction term between Latino Minister and Latino population to regress on both Spanish Mass and Cultural Services (see Tables 3 and 4). This interaction term is only marginally significant and negative in the regression on Cultural Services, and is not significant for Spanish Mass. This indicates that church service provision is not augmented by the presence of both a large Latino population and a Latino minister.

Table 3.

Alternate regressions predicting Spanish mass in churches

Robust standard errors in parentheses.

***p < .001, **p < .01, *p < .05, p < .1.

Table 4.

Alternate regressions predicting cultural services in churches

Robust standard errors in parentheses.

***p < .001, **p < .01, *p < .05, p < .1.

I also tested to see if there was a relationship between service provision and a change in the Latino population, in place of the simple measure of the Latino population size. It could be that areas with rapid demographic change would be more responsive because of an increasingly visible Latino population, which would demonstrate a high level of service provision that would be obscured in the aggregated analysis.

The results of these alternate regression analyses were nearly identical to those reported above, with a few exceptions (see Tables 3 and 4). Where Latino population size was significant in modeling the likelihood of a Spanish mass, an increase in the Latino population was not. This is perhaps because the churches in my sample already have a high rate of responsiveness as predicted by the Latino population size. As such there may be a ceiling effect with a population increase. Further, the average increase in Latino population size was relatively small (ranging from −.2% in Seattle to 5.5% in Yakima) which puts any potential corresponding growth in the rate of Spanish mass within the standard error and thereby making it undetectable by statistical measures. The percentage of the population under the poverty line became significant and positive in this second regression, perhaps indicating that the growth in Latino population between 2010 and 2015 corresponded with a growth in the number of families below the poverty level. However, the poverty variable and the population change variable are not strongly correlated (Pearson's r = .07), although there is a strong correlation between poverty and the Latino population (Pearson's r = .66).

The regression model on cultural services did not produce a significant coefficient for the interaction term between population change and Latino minister, though the Latino minister base term remained positive and significant. Lastly, there were no differences for the regression on having a Latino minister.

I also created a variable that accounted for the difference between Latino and non-Latino income. Given that churches are theorized to be more important to the lives of those who are less well-resourced (Brady, Verba, and Schlozman Reference Brady, Verba and Lehman Schlozman1995), it may be that churches would offer more services to Latinos when they have comparably less income than their neighbors. Conversely, simple economics may predict that churches located in communities with lower incomes would have fewer resources to spend on providing services or organizing non-worship activities. I found that the variable measuring income disparity between Latinos and non-Latinos was only significant in the regression on non-worship services, and it was negative (see Tables 3 and 4). This indicates as the income disparity increases, the likelihood of churches offering culturally relevant services and activities to Latinos decreases.

LESSONS FROM AND FOR THE CATHOLIC CHURCH

The Catholic Church currently sits at the forefront of U.S. demographic change as a prominent institution that is working on how to positively respond to a growing (and soon to be majority) Latino population. Understanding the factors that either complicate or facilitate that endeavor may help other institutions in their future efforts to likewise integrate and serve Latino communities. Further, there may be broader implications to the success of Catholic churches to serve as research has found that participation in church activities (notably in small groups outside of the worship service) is positively related to increased rates of civic and political engagement (Djupe and Gilbert Reference Djupe and Gilbert2006; Djupe and Neiheisel Reference Djupe and Neiheisel2012; Jones-Correa and Leal Reference Jones-Correa and Leal2001). However, these positive effects cannot be felt if churches fail to present the opportunity to participate in the first place. To that end, this study endeavors to determine which factors best predict when churches will provide services to Latinos.

Based on theories of representation and constituency response, I hypothesized that churches would be more likely to provide Spanish mass and other culturally relevant services if they were situated within a larger Latino population (hypothesis 1) or if they had a Latino minister within their parish (hypothesis 2). I found that the size of the Latino population was not statistically significant in predicting the likelihood a church would offer either service. Further, a striking pattern emerged—churches were much more likely to offer Spanish mass than to offer any other kind of culturally relevant service even though these other services could be organized and offered by anyone in the community. This logically implies that as the Latino population increases, so too should the number of non-worship oriented services because the number of organizers and participants necessarily increases along with the overall population. Yet instead of population size, it was the presence of a Latino minister that was significant in predicting both types of service provision (hypothesis 2). Because of the importance of Latino leadership, I also examined the factors that may predict whether or not a parish would have a Latino priest or deacon. While I hypothesized that a large Latino population would predict a higher likelihood of having a Latino minister (hypothesis 3), no such relationship emerged in the regression analysis.

One potential explanation for these findings (high likelihood of mass; lower likelihood of other services) is the structure of the Church as an institution with “rules and organized practices” that define behavioral expectations. The mass, then, as a central component of the Catholic faith, can be considered an “organized practice” that is fairly simple to initiate and integrate into parish life. In contrast, the requirements for organizing and promoting activities such as Bible studies, festivals celebrating the Marian patrons, and legal clinics are widely varied, and there is little systematic training within the Church on how to offer these types of services. While the Church's many publications advocate for the outreach and inclusion of Latinos, very few offer practical advice on how to do so (Huckle Reference Huckle2016).Footnote ²⁰ As such, the provision of these services may be less a factor of institutional and contextual factors (such as Latino population size and descriptive representation) than of individual leadership and decision-making, regardless of that individuals’ ethnic or cultural background.

This conclusion is consistent with the representation theory's discussion of discretion, defined as when administrators have the ability to make their own decisions and act on their values. Discretion is afforded because “organization rules cannot cover every contingency and because organizational socialization is rarely total” (Meier and Bohte Reference Meier and Bohte2001, 463) The larger the organization—and the more people one person must supervise—the greater amount of discretion will be found. Priests have a large amount of discretion in how they run and minister to their parishes. As such, the decision to serve Latinos by providing cultural services outside of Spanish mass may be the result of individual-level factors, rather than institutional.

Further, we must consider the necessity of serving Latinos to the Church's institutional survival. The Church is dying, suffering from high attrition rates, church closures, and continued fallout from numerous sex scandals. Latinos, as the only significant source of membership and growth, represent a lifeline. As such, Church officials should be conducting as much welcoming, outreach, and service provision as possible in order to stave off closures, particularly since it is one of the primary responsibilities of leaders to maintain membership levels. This can be a difficult task given the voluntary nature of membership, and so members must be incentivized to join (March and Olsen Reference March, Olsen, Rhodes, Binder and Rockman2008). Incentives include not only offering mass in Spanish, but also the provision of other services and activities that directly appeal to Latinos. Indeed, many have theorized that it is precisely because of the Church's failure to provide these enticements that so many Latinos have left Catholicism (Martinez Reference Martinez2011; Martinez, Hernandez, and Pena Reference Martinez, Hernandez, Pena, Robin and Wadsworth2012). The success or failure of the Catholic Church to accept—and take seriously—this reality should serve as a lesson to other institutions that will soon be reliant on the membership of a growing Latino population for their survival.