Black Church Giving

Church giving is an understudied phenomenon; queries on Black Church giving are even rarer. In response to the latter dynamic, levels of tithing, income, and mission donations are examined based on a national sample of 1,601 Black churches across seven denominations using linear and ordered logit modeling. Findings show minimal ideological and programmatic effects. However, denominational differences suggest that Black congregations affiliated with the Church of God in Christ have higher relative percentages of tithers and those associated with the Presbyterian and United Methodist faiths have the lowest rates as compared with their Baptist peers. Moreover, African Methodist Episcopal Zion, Presbyterian, and United Methodist churches are more likely to have higher church incomes and mission giving than Baptists. Thus, although Baptists tend to generally have higher rates of church tithers, this pattern does not translate to higher church incomes or mission donations. As expected, church size and formally educated memberships positively influence giving patterns. Results provide important insights into the Black Church economic ethic.


Protestant Church Giving
Scholars and seminarians alike describe an emphasis on financial issues in the Bible. In addition to the more than 2,000 references to money and finances (Kreider, 2002), Jesus is said to have spoken about money and material possessions more than heaven, hell, faith, or prayer (Alcorn, 2003). Scripture encourages tithing and sacrificial giving. A continued debate exists about the appropriateness of tithing (Johnson, 1984). 2 Yet tithes, offerings, and pledges continue to be the primary methods of church financial support (Dahl & Ransom, 1999;Keister, 2008;Smith & Emerson, 2008). Hudnut-Beumler (2007) suggests that, in 2004, more than US$88 billion were donated to religious organizations and a significant share went to Protestant churches. According to Hoge et al. (1996), about 63% of philanthropic giving goes to churches. Yet studies are unclear about whether giving has decreased (Amerson, Stephenson, & Shipps, 1997;Ronsvale & Ronsvale, 1997), increased (Chaves, 2004), or always been limited (Smith & Emerson, 2008).
Furthermore, churches that sponsor more programs tend to receive greater contributions than those with fewer programs (Hoge et al., 1996;Thumma & Travis, 2007). Church size can also directly affect giving if attendees give liberally (Krohn, 1995) or inversely if "free-riders" avail themselves of programs, but fail to contribute financially (Hoge, 1994;Sullivan, 1985;Zaleski & Zech, 1992). Yet interchurch competition and declining rolls mean churches are spending less money on missions and using funds to meet internal and local needs (Amerson et al., 1997;Smith & Emerson, 2008). The above literature focuses on giving among White churches. Black churches are only tangentially mentioned or absent from the discourse (Chaves, 2004;Hoge et al., 1996;Smith & Emerson, 2008). This project begins to fill this research void.

The Black Church History of Giving
Black Church giving has a consistent, yet somewhat colorful past that includes traditional tithing, sacrificial offering, pledging, as well as raising funds through the sale of sundry foodstuff such as chicken and fish dinners (Drake & Cayton, 1940;Evans, 1990;Lincoln, 1974;Lincoln & Mamiya, 1990). This legacy is systematically chronicled initially in DuBois's (1903DuBois's ( /2003 The Negro Church with detailed accounts of salaries, expenses, and charitable efforts. For example, in 1902, Black Baptist churches provided US$100,337 in mission donations. This finding is impressive even by today's standards, but particularly so for that time period. According to the same scholar, most donations went to charity and missions. More recently, Burnette (2001) posits that, "community-based philanthropy is deeply grounded in the Black church" (p. 5). Lincoln and Mamiya (1990) link Black churches to financial self-help initiatives such as mutual aid associations as early as the 1870s. Their national study shows that, regardless of locale, at least 15% of Black churches have annual incomes of more than US$50,000; offerings, special fund-raisers, and pledging are the most common methods to raise income. The current endeavor considers the implications of denominational, ideological, and programmatic effects on Black Church giving.
First, denomination has been shown to influence giving among Black churches. A study of 141 Black churches shows that "congregations in denominations that emphasize tithing tend to have some of the most generous donors" (Christian Century, 1998, p. 1241). This same analysis shows that Baptist, Church of God in Christ (COGIC), and Pentecostal Black churches tend to have relatively more givers because of tithers. In addition, in Hoge et al.'s (1996) list of 23 denominations, Black Baptists and Black Methodists rank 11 and 14, respectively, in terms of overall giving. Furthermore, Black churches from the Methodist faith such as African Methodist Episcopals (AME), African Methodist Episcopal Zions (AMEZ), Christian Methodist Episcopals (CME), and United Methodists (UM) as well as Presbyterians were historically believed to be doctrinally motivated to give in general, particularly to mission efforts (Lincoln & Mamiya, 1990). Yet literature is absent definitive empirical work to clearly assess how denomination affects Black Church giving.
However, other studies directly link giving to attendance, church ideology, and "cafeteria-style" programs (i.e., numerous religious and practical programs; Hall-Russell, 2005;Schaller, 2000;Tucker-Worgs, 2002a). Niebuhr (1995) succinctly describes this latter phenomenon and informs the causal ordering used here: Fast-growing churches enter a cycle: More people coming through the doors means more money in the collection place, which goes to more programs, some of which are charitable . . . all of this in turn draws more people. (p. 1) Research describes growing numbers of Black churches that proactively respond to and anticipate the worship and varied and numerous programmatic needs of a diverse Black populace (Barnes, 2004). Such churches forge relationships with congregants by striving to assimilate them into the life of the church quickly as well as replacing extrachurch options they might seek by making it difficult to locate quality, viable, and wholesome alternatives that are free or inexpensive (Barnes, 2010;Thumma & Travis, 2007;Tucker-Worgs, 2002a;Tucker-Worgs, 2002b). As their spiritual and secular needs are met, congregants become more amenable to donating their time and finances to such churches (Barnes, 2010).
Third, traditional church ideologies have been linked to higher giving patterns. For example, Lunn et al. (2001) show that more religiously traditional denominations experience higher rates of giving. If the latter assessment translates to the Black experience, COGIC, AMEZ, AME, and CME congregations as well as Black churches affiliated with the UM and Presbyterian faiths-historically considered more traditional-are expected to experience higher levels of giving than Black churches associated with Baptists. Furthermore, the tendency of the latter four denominations to attract more middle-class Blacks would also be expected to positively influence giving (Billingsley, 1999;Lincoln & Mamiya, 1990). However, this trend may be affected if denominations such as Baptists (middle-class churches only), AMEs, and AMEZs that have experienced increases in involvement and membership as a result of neo-Pentecostalism also experience increased giving (Lincoln & Mamiya, 1990;Rivera, 2002) or ideologically traditional Black UM and Presbyterian churches experience less giving as a result of aging and waning memberships (Billingsley, 1999).
In addition to the above three dynamics, southern locale may affect giving patterns (Ellison & Sherkat, 1995). And growing literature suggests that religious competition may undermine giving by reducing the number of potential congregants per church or stimulate giving among churches with smaller shares of the local religious market (Olson & Perl, 2000). Thus, church giving may differ based on areas known for greater relative numbers of Black residents or in locales with greater relative percentages of Blacks as compared with Whites. Billingsley's (1999) quote summarizes the history of Black Church giving: How long would the black church last, how strong and independent would it be, how vital would the worship be, and how could it possibly sustain an extensive community outreach program were it not for the strong tradition of "passing the collection plate?" Indeed, the black church would hardly be recognized without this essential element of financial stewardship. (p. 180) Despite academic studies on the Black Church experience and applied information on the economic benefits such churches provide to congregants and community members, a contemporary quantitative study has not been performed that focuses on their giving patterns.

Research Hypotheses
Based on the above literature, I consider the effects of religious beliefs, organizational/programmatic features, and denomination on Black Church giving (i.e., tithing, church income, missions). Although these are not the only dynamics expected to influence Black Church giving, they provide one systematic way to continue to assess the subject. Four hypotheses are tested.
Hypothesis 1: Churches that emphasize traditionalism in worship and teaching will be more likely to engage in the three forms of giving than churches with less emphasis on traditionalism in worship and teaching.
(Religious beliefs) Hypothesis 2: Churches that provide efforts to easily assimilate new members into the life of the church will be more likely to engage in the three forms of giving than churches that do not provide efforts to easily assimilate new members into the life of the church.
(Programs) Hypothesis 3: Churches that sponsor more programs will be more likely to engage in the three forms of giving than churches that sponsor fewer programs. (Programs) Hypothesis 4: Churches that are considered more traditional such as AMEs, AMEZs, CMEs, COGICs, Methodists, and Presbyterians will be more likely to engage in the three forms of giving than Baptists.

(Denominations)
Also paralleling literature, I control for and predict the following: positive effects of church size, negative impact of high ratios of Sunday worship attendees to members, positive effects as the percentage of formally educated members and male members increases, negative effects as the percentage of members above the age of 60 years old and ratio of Blacks to Whites in the respective church areas increases, positive effects of region (South), and negative effects of locale (urban). Finally, this study tests the anticipated positive impact of tithing on church income as well as the expected positive effects of income and tithing on mission giving.

Data and Method
This analysis is based on a national secondary database of Black churches from the Faith Factor 2000 Project, a joint venture between the Lilly Foundation and the Interdenominational Theological Center (ITC) in Atlanta, Georgia. The project was devised to develop profiles of Black congregations in the United States. The data collection process was spearheaded by the ITC with assistance from Gallup, Inc and included 1,863 Black churches. This analysis is based on the 1,601 churches for which tithing data were collected. The following five Black denominations were included; Baptist (412 churches), COGIC (437), CME (264), AME (237), and AMEZ (99). Predominately Black churches from the historically White UM (81) and Presbyterian (71) denominations were also included for a total of seven denominations. 4 Identifying the sampling frame and selecting the sample occurred in several steps. First, lists of all the churches in the AME, AMEZ, CME, COGIC, Presbyterian, and UM denominations were provided by denominational heads or deans from the various seminaries at the ITC. The decentralized nature of the Baptist tradition prevented such a list. To develop the sampling frame for Baptists, ITC solicited information from Tri-Media, an organization that retains lists of all congregations nationwide that purchase Sunday school supplies. Tri-Media data were used to identify the population of Baptist churches affiliated with the three largest historically Black Baptist denominations. 5 Unlike the six other denominations, the sampling frame for Baptist churches is an approximation with several limitations, 6 but it represents a systematic attempt to identify these congregations in light of a lack of national hierarchy. After the seven lists were compiled, Gallup selected a random sample from each denomination to meet the desired subsample sizes.
Telephone surveys of clergy and senior lay leaders were conducted by Gallup from February 22, 2000, through May 11, 2000. Each interview averaged about 16 min in length and 37 questions were posed. The church leaders were charged with providing aggregate demographic data on their congregations and answering a variety of behavioral as well as attitudinal questions on topics such as worship and identity, economic health, missions, leadership, spirituality, organizational dynamics, church climate, and community involvement. 7 Initial screening was used to gain pastoral cooperation and to confirm denomination. If the pastor was unavailable, an assistant pastor or senior lay leader was interviewed. Seventy-seven percent of the interviews were conducted with pastors; the remaining interviewees were assistant pastors or senior lay leaders. Such church officials were used because they would be expected to be the most knowledgeable about their respective communities of faith. 8 In addition, 2000 census data are incorporated based on church zip code to capture the proportion of Black and White residents in the areas in which the sample churches are located. 9

Dependent and Independent Variables
The overall research objective is to study three forms of Black Church giving. The first dependent variable, Tithers, is based on responses to the following question-"What is the percentage of tithers in your congregation?" The tithe here is based on the customary definition of paying or giving 10% of one's income to support the church. The variable ranges from 0% to 100%. The second dependent variable, Church Income, is based on the question, "Approximately how much income did your congregation receive last year from all sources (e.g., tithes, pledges, membership dues, plate offerings, endowment income, etc.)?" Options were categorized a priori in this secondary data based on eight categories: less than US$15,000; US$15,000 to US$24,999; US$25,000 to US$34,999; US$35,000 to US$44,999; US$45,000 to US$54,999; US$55,000 to US$74,999; US$75,000 to US$99,999; and US$100,000 or more. The third dependent variable, Missions, reflects the question, "Approximately how much income did your congregation spend for all mission work including local, national, and international, last year?" Mission work includes activities such as donations to local shelters and financial support of local, national, and/or international efforts such as orphanages, schools, and intradenominational churches. The same eight response categories were used.
Because the preponderance of research suggests more religious conservatism among non-Baptist Black churches, Baptist is the denominational reference group in this study. In addition to 7 denominational variables, 11 additional independent variables are tested. They include programs to assimilate new members, membership size, number of church programs, attendee-member ratio, urban locale, proportion of college-educated members, proportion of members above 60 years old, proportion of male members, and emphasis on traditionalism during worship and teaching. I control for church competition using a variable that reflects the ratio of Blacks to Whites in the areas in which the sample churches are located (M = 10.69, SD = 26.0) and a variable to identify churches in the South. I also consider the possible effects of tithers on church income and church income and tithers on mission giving. To correct for skewness, a logarithmic transformation was used on the membership variable (M = 5.4, SD = 1.1) as well as the attendee-member ratio variable (M = 0.81, SD = 0.85). However, only slight skewness is evident for the eight category income variable, so no transformation was needed.
Several comments on causal ordering are needed. Studies on the subject suggest that church size tends to shape subsequent programs and charitable contributions (Finke, Bahr, & Scheitle, 2005;Hoge et al., 1996;Schaller, 2000;Thumma & Travis, 2007). Yet it is possible for the opposite causal ordering to occur. Moreover, larger churches may be more likely to have greater percentage of tithers and more overall income, but by the same token, churches with a greater percentage of tithers and more income may be able to market their services and attract more persons. However, literature suggests that my chosen independent variables are more likely to help explain the giving efforts under study in the order predicted here (Billingsley, 1999;Finke et al., 2005;Hoge et al., 1996;Lincoln & Mamiya, 1990;Sherkat & Ellison, 1991). A total of 18 independent variables are tested (bivariate correlations and other diagnostics provided upon request). Each variable's operationalization and corresponding survey questions are provided in the appendix. Table 1 reports descriptive statistics for the dependent variables. Modeling results are provided in Table 2. Because the tithing dependent variable's range is 0% to 100%, linear regression modeling (ordinary least squares [OLS]) is used, Model 1). The next two models focus on church income and mission donations (Models 2 and 3). Because these two dependent variables reflect ordinal categories that are not necessarily equidistant apart, ordinal logistic regression modeling is used (Long, 1997;Long & Freese, 2001). In each model, the dependent variables are regressed on the independent indicators of interest. Table 1 captures general patterns across the three forms of giving such that substantive comparisons can be made. Bivariate results suggest that as an emphasis on traditionalism increases, the three types of giving patterns decline. Moreover, as responses move from less to more affirming levels for the variable that gauges assimilation of new members, so do tithing percentages, church income levels, and mission donations. Similarly, as number of programs, membership size, and percentage of educated members move from lower to higher ranges, greater relative percentages of tithers and higher church mission donation levels are generally apparent. The opposite pattern emerges between attendee-member ratio and church income and mission donations, respectively. When denomination is assessed, COGIC churches (55.9%) followed by AMEZs (45.4%) have the highest relative percentages of tithers; Presbyterians (29.8%) have the lowest. Because church income levels and mission donations are coded a priori 1 to 8, category averages are generated. For example, values of 6.8 and 6.5 mean that AMEs and AMEZs have incomes US$55,000 to US$74,999, but AMEs tend to have relatively higher income levels than do AMEZs. Although COGICs have the highest tithing percentage, they have the lowest mean church income category (4.6). UMs have the highest church income mean category (7.0). Common denomination levels exist (1.6-2.4) for mission donations with slightly higher values among UMs. Finally, a review of the ecological indicators shows similar patterns for the three giving variables regardless of race ratio, urban locale, or region.

Modeling Black Church Giving
Model results presented in Table 2 assess simultaneous effects of the congregational variables on Black Church giving.
Model 1 (R 2 = .14) focuses on the percentage of congregants who tithe using linear regression modeling. Although traditional beliefs are not influential, the variable that identifies church attempts to proactively acclimate new members (b = 2.76, p < .01) increase percentages of tithers. Significant denominational differences are also apparent. Affiliation with AME, CME, Presbyterian, and UM traditions as compared with Baptists, respectively, undermines percentages of tithers. However, the converse is true for COGICs (b = 15.72, p < .001) as compared with Baptists. Regardless of other church features, the proportion of college-educated members (b = 0.18, p < .001) directly affects percentages of tithers. And as the percentage of male members increases so does the percentage of tithers. Finally, churches located in urban locales or in areas with more Black than White residents are no more or less likely to have greater percentages of tithers than their counterparts. However, church residence in the South (b = 4.21, p < .05) is associated with greater percentages of tithers.
The results of an ordered logit model gauge overall church income in Model 2 (χ 2 = 353.65). Findings show that traditional church beliefs undermine church income levels. Furthermore, providing efforts to easily assimilate new members reduces the likelihood of church income levels (odds = 0.73). Unlike the first model, but as expected, church size influences income levels. Black congregations that have more members on role are significantly more likely to have higher income levels than their peers with fewer members on roll. Denomination continues to be influential and four variables are important. Relative tithing patterns are generally improved for Baptists as compared with non-Baptists in Model 1, but when overall church income is assessed, Presbyterian (odds = 6.35), AME (odds = 3.07), AMEZ (odds = 2.92), and UM (odds = 1.87) churches are more likely to experience higher church income levels than Baptists. And greater proportions of male members increase chances of higher church income levels (odds = 1.01).
Church mission efforts are examined in Model 3 (χ 2 = 251.55) also using ordered logit modeling. In contrast to the second test, churches that espouse traditional beliefs and those that provide opportunities to easily assimilate new members are no more or less likely to donate to missions than their peers that do not meet these criteria. However, Black churches that have more members are more likely to donate to mission efforts than their peers with fewer members. Three of the six denominational indicators are predictive; Presbyterians are over 3 times more likely, AMEZs are 2.19 times more apt, and UMs are 1.68 times more apt to have higher mission outlays than Baptists, respectively. The presence of more formally educated members positively affects mission donations; yet greater percentages of older members undermine the chances of such donations. Although the tithing indicator remains insignificant, Black churches with higher income levels in general are 1.34 times more apt to donate to missions than those with lower income levels.

Discussion
These tests offer insight into giving patterns among the sample Black churches. The general findings show that tithing appears normative in at least one third of the sample. Church incomes fall into the US$55,000 to US$74,999 range; mission donations average US$15,000 to US$24,999. Furthermore, these results partially demonstrate the influence of traditional beliefs, but with results that contrast earlier studies (Finke et al., 2005). When such beliefs are predictive, they actually undermine church income levels. Thus, my results show some support for Hypothesis 1-but with negative rather than positive effects. Findings for Hypothesis 2 are inconclusive because efforts to assimilate new members have a positive impact on tithing, a negative impact on church income, and no direct influence on mission giving. Thus, transitioning new members may be needed to socialize them about the commitment required for tithing, but may have little collective impact (or undermine) on other forms of giving. Hypothesis 3 that considers the number of church programs is not supported here. As suggested in Hypothesis 4, denominational differences provide important consistent effects. At least three of the six denominational indicators help explain the three giving activities under study. Although Baptist affiliation increases relative percentages of tithers, this practice does not translate into higher church income levels or mission donations. Yet denominations that are considered more historically traditional and middle class such as Presbyterians and UMs appear better positioned than Baptists in terms of church income and mission giving. A partial explanation for this result points to the continued tendency to teach tithing among Black churches (Christian Century, 1998), but the relatively lower income levels often associated with Baptist churches (Lincoln & Mamiya, 1990). When controls are reassessed, the negative impact of potential freerider effects is not evident here; church size directly influences giving efforts in two of the three tests. Results also show the positive effects greater proportions of educated members have on two of the three types of giving and the Note. Coefficients presented first: Standard errors for the ordinary least squares (OLS) model and odds ratios for ordered logit models in parentheses. Membership roll's original, nontransformed range was 1 to 18,000. Attendee-member ratio's original, nontransformed range was 0.0 to 5.0. Baptist is the denominational reference group. AME = African Methodist Episcopals; AMEZ = African Methodist Episcopal Zions; CME = Christian Methodist Episcopals; COGIC = Church of God in Christ; UM = United Methodists. *p < .05. ** p < .01. *** p < .001.
negative influence overrepresentation by older members have on mission donations. Finally, although the presence of tithers does not affect income and mission levels, as expected, churches that have higher income levels tend to also donate more money to missions.

Conclusion
This congregational analysis examined giving efforts for a national sample of Black churches. Contrary to expectations, traditional beliefs appear to undermine church income levels and may point to an increase in neo-Pentecostalism predicted by Lincoln and Mamiya (1990) as well as its more recent correlates to social action (Rivera, 2002). Programmatic efforts have minimal influence here. And the positive effects of church size on income and mission levels are expected. However, denominationalism is central in explaining Black Church giving as defined here. And although non-Baptists (save COGICs) have generally lower tithing rates than their Baptist counterparts, Baptists are more likely, in general, to have lower church incomes and mission donations. I contend that these findings illustrate the nuanced manner in which church socialization toward tithing may ultimately be affected by the ability of congregants to do so and becomes evident in the limited collective sum of their efforts. Additional research on this possible phenomenon is needed. It is also important to acknowledge that ideology and class is expected to be imbedded within denominations-for example, the tendency for COGICs to be ideologically traditional and to historically draw less socioeconomically stable Blacks (Lincoln & Mamiya, 1990) or the influence Prosperity theology is having on historically traditional Black churches (Harrison, 2005;Mitchem, 2007). And what are the implications on giving of Blacks who may no longer be wedded to the denominational affiliations of their foreparents, but who are creating denominational and class-amalgamated arenas as independent Black churches (Barnes, 2010)? To better understand such tendencies and disentangle them will require mixed methodologies that consider direct and indirect effects in addition to qualitative queries. 10 Moreover, the effects of demographic variables such as members' education, gender, and age in this study illustrate the varied, potentially countereffective value of members' profiles. Yet, regardless of locale and potential area competition, who members are and how many members there are represent important dynamics in understanding Black Church giving. These findings, broadly generalizable (refer to the limitations in Notes 5 and 7), inform existing literature on Black religiosity. Future studies should take into consideration additional features considered important in other work such as pastoral profile (Barnes, 2004;Billingsley, 1999) as well as the varied interpretations and implications of the concepts traditional ideology and conservative ideology. Secondary data constraints precluded an analysis of church income and mission giving using continuous functions and more detailed operationalizations of certain variables (e.g., traditional ideology) in this study. Moreover, future studies based on more recent data are needed. However, these results provide insight to continue the discussions and empirical studies required to assess giving patterns and delve more fully into the motivations behind the Black Church economic ethic.

Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding
The author(s) received no financial support for the research and/or authorship of this article.

Notes
1. Throughout the document, the term the Black Church is used to represent the institution as a collective and "Black church" when specific congregations are referenced. 2. Tithing is generally defined as paying one tenth of one's income to church and is linked to the Old Testament (Genesis 14:18-20, Leviticus 27:30). 3. This study broadly uses the concepts "conservative" and "traditional" interchangeably while acknowledging that some scholars may make a distinction. 4. The intent was to mimic Lincoln and Mamiya's (1990) work by focusing on historically Black denominations. The sampling process is representative for non-Baptists and provides a conservative, systematic attempt to approximate Black Baptists in light of the associated challenges. The use of these seven Black denominations is common in research on the subject and provides a comparative benchmark. 5. They are the National Baptist Convention, United States of America, National Baptist Convention of America, and the Progressive National Baptist Convention. Tri-Media data were also used to augment the lists from the six other denominations; churches found on either source were included on the composite list for that perspective denomination and duplicate churches were identified and only included once. 6. Baptist churches that purchase Sunday school items at venues not listed with Tri-Media or those that do but that are not affiliated with any of the three conferences would not be included in the sampling frame. Thus, the Baptist list is a lower bound of the number of Black Baptist churches nationwide. 7. Chaves, Konieczny, Beyerlein, and Barman (1999) discuss the validity and reliability of relying on a single-key informant to report church characteristics. Such persons are likely to overestimate the extent to which their views correspond to their congregation's views. They note, "an informant's judgment about an organization's goals or mission is likely to represent the informant's interpretation of a complex reality rather than a more or less publicly available cultural fact about the congregation" (p. 464). In light of this dynamic, these data can be considered a best-case scenario. However, vast majority of respondents are senior pastors and because research shows that Black pastors tend to have a greater degree of authority and influence over their congregants than their White counterparts, they are expected to have greater influence over the focus and activities of their churches (Billingsley, 1999;McRoberts, 1999 .65). The overall sample rates were .24, and .54, respectively. Although the most current available lists were used, low response rates were due to situations such as disconnected telephones or relocation that counts against the response rate (referrals were used to locate many such congregations). In such situations, the cooperation rate can be used as a reasonable proxy. The figures represent the Council of American Survey Research Organizations Standard (CASRO) calculation for the response rate and the cooperation rate reflects the percentage of churches that participated once contact was made. The data were weighted during model to reflect the current denominational estimates to correct for the disproportionate subsample sizes (Billingsley, 1992;Lincoln & Mamiya, 1990). The sample margin of error was ±2.3. 9. The two data files were merged by zip code. To create the ratio, two variables, total population Black (numerator) and total population White (denominator) for each zip code were extracted from 2000 census data. Values range from 0.00 to 171.75. 10. In regression models not shown here, I tested interactions between Denomination × Traditional Beliefs (for all three models) as well as Denomination × Percentage of Tithers (for the latter two models). Interaction terms were insignificant in each test save the test on church income and the Denomination × Percentage of Tithers interaction term. The latter model showed that COGIC churches with higher rates of tithers tend to have higher church incomes than their Baptist counterparts that also have higher rates of tithers (odds = 1.01, p < .001). This finding follows the pattern of the models included in this analysis.