Content uploaded by Andrew L. Whitehead
Author content
All content in this area was uploaded by Andrew L. Whitehead on Apr 01, 2015
Content may be subject to copyright.
A Place to Belong: Small Group Involvement
in Religious Congregations*
Kevin D. Dougherty
Andrew L. Whitehead
Baylor University
Large congregations are commonly criticized as eliciting less involvement from adherents than
smaller congregations. Small groups have been heralded as a remedy to drawbacks of increasing
congregation size. This study tests the relevance of small groups to individuals’ commitment and
participation, particularly in large congregations. Analysis features the 2001 U.S. Congregational
Life Survey and a survey of congregants at one Central Texas Megachurch. Persons involved in
small groups devoted to prayer, discussion, or Bible study report a greater sense of belonging, more
frequent attendance, and higher rates of giving. The effect of small group involvement does not
differ by congregation size, however. From worshippers in one Texas megachurch, we find that the
extent of small group involvement positively relates to commitment and participation. Small groups
may not completely resolve problems associated with increasing size, but we believe they do rep-
resent a potent source of vitality in congregations big and small.
Key words: congregations; commitment; organizations; megachurches; practice; religious
involvement; small groups.
It is a warm Sunday morning. Amy and Jeremy, a young married couple
new to the area, are making their first visit to Central Texas Megachurch
(CTM).
1
The bustling non-denominational congregation, located in a con-
verted grocery store near the city center, was recommended to Amy and
Jeremy as popular with people their age. Popular it is, drawing over 2,000 each
*Direct correspondence to Kevin D. Dougherty, Department of Sociology, Baylor University, One
Bear Place #97326, Waco, TX 76798-7326, USA. E-mail: kevin_dougherty@baylor.edu. An
earlier version of this manuscript was presented at the Society for the Scientific Study of Religion’s
2009 annual meeting in Denver, CO, USA. The authors thank Cynthia Woolever and Matthew
Stanford for assistance in the research process. This study was supported in part by funds from the
Baylor University Research Committee and the Vice Provost for Research.
1
Names of individuals and the congregation are pseudonyms. All details reported are
factual and were gathered through informal interviews and church publications.
#The Author 2010. Published by Oxford University Press on behalf of the Association
for the Sociology of Religion. All rights reserved. For permissions, please e-mail:
journals.permissions@oxfordjournals.org.
Sociology of Religion 2011, 72:1 91-111
doi:10.1093/socrel/srq067
Advance Access Publication 3 September 2010
91
at Clemson University on July 16, 2014http://socrel.oxfordjournals.org/Downloaded from
week to two Sunday worship services. The church sits on a main thoroughfare
busy with traffic. By the time Amy and Jeremy arrive, the parking lots sur-
rounding the worship facility are full. Slightly confused as to where they should
go, they eventually locate a place to park on a side street and head toward
what appear to be the front doors of the church. Once inside they fall into line
with a growing crowd and slowly funnel through a set of double-doors into a
darkened auditorium. Guitars, electric keyboard, and drums provide back-
ground music from an elevated stage at front. Some 1,100 chairs line the floor.
Amy and Jeremy find a pair of seats and wait for the service to start. They
observe those around them exchange handshakes, hugs, and engage in lively
conversation. These people seem to know each other well. The worship service
begins. Amy and Jeremy like the music and the preaching. Yet leaving as anon-
ymously as they entered, neither feels a strong attraction to the church. They
decide to continue their search. CTM is just too big, too impersonal.
An increasing number of congregations in the United States face a similar
challenge to that posed by Amy and Jeremy’s reaction at CTM. With thou-
sands in attendance, it is difficult to cultivate the type of in-group bonds con-
ducive to belonging and active participation. Individuals can come and go
without a sense that their contributions matter. Thus, critics contend that the
back door of huge congregations is as wide open as the front door. For those
individuals who do keep coming back, religious participation may involve little
more than attending worship services as spectators. Such concerns take on
added significance given changing organizational patterns in American reli-
gion. The number of very large congregations is on the rise as is the percentage
of worshippers that attend large congregations (Chaves 2006;Thumma and
Travis 2007). However, there is more to learn from the story of Amy and
Jeremy.
A month after their first unsatisfying visit, Amy and Jeremy are back at
CTM. Parking on a side street, they enter the building, navigate around the
masses exiting the first service, and head toward specific seats in the auditor-
ium. The church now feels palpably different. There is a sense of familiarity
and community. Part of this change is due to them attending now for several
Sundays in row. But the explanation for why they returned at all resides in the
dozen or so individuals waiting for them in their “usual” spot. It turned out
that a young couple who lived across the street from Amy and Jeremy were
members at CTM. Without knowing that Amy and Jeremy had visited the
church, the couple invited them to the small group that met in the couple’s
home. Meeting other young couples at a similar stage of life appealed to Amy
and Jeremy as newcomers to the area. Friendships formed and their search for a
church concluded. In this small, home group, Amy and Jeremy found a place
to belong. They have been active members of CTM ever since.
Social scientists and religious leaders publicize small groups as a correlate
to congregational success (Arnold 2004;Easum and Atkinson 2007;Finke
1994;Stark and Finke 2000). Small groups help break down sizable
92 SOCIOLOGY OF RELIGION
at Clemson University on July 16, 2014http://socrel.oxfordjournals.org/Downloaded from
congregations into more manageable units that provide individuals with
relationships and community; things that are difficult to achieve in worship ser-
vices of hundreds or thousands. They are a prevalent feature of the Protestant
megachurch movement (Thumma and Travis 2007), but their impact within
congregations remains inadequately explored. This study aims to address this
gap in the literature. We seek to answer how effective are small groups, particu-
larly in large congregations, at promoting ingroup commitment and partici-
pation? In other words, is the experience of Amy and Jeremy unique?
Our examination begins with a review of previous research on commit-
ment, participation, congregation size, and small groups, which leads us to two
hypotheses. We test our hypotheses using national survey data from the 2001
U.S. Congregational Life Survey and a 2007 survey of participants at CTM.
While small groups do not completely resolve the problems associated with
large congregations, our findings offer some of the strongest evidence to date
for the benefits of small group involvement in congregations.
ACTORS OR ONLOOKERS? INDIVIDUAL INVOLVMENT IN
CONGREGATIONS
Congregations, like all voluntary organizations, depend on committed par-
ticipants to exist. Individuals’ voluntary contributions of time and money are
essential resources for congregations. Vital congregations are those that
mobilize high rates of commitment and participation. Organizational attributes
play a key role in fostering commitment. As Kanter (1972:75) discovered in
her study of nineteenth-century communes, “The strength of a group and the
commitment of its members will be a function of the specific ways the group is
put together.”
One feature of groups that impacts commitment is size. It is expected that
people would feel a strong sense of belonging to small, intimate, “primary
groups,” made up of family and close friends; such groups are fundamental to
human development (Cooley 1909). In larger, more formalized, “secondary
groups,” commitment is more difficult to cultivate. Olson (1965) framed the
challenge of size as related to incentives. The rationale is that individuals in
large organizations get less personal benefit from the collective good and, corre-
spondingly, feel less incentive for its production. Organizational commitment
suffers. Rather than relying on individuals’ commitment to a common interest,
large organizations depend on selective incentives of a social nature to motiv-
ate individuals. Friendship, prestige, self-esteem, and shame are the types of
social incentives identified by Olson (1965:61). Yet, even these social incen-
tives are more effectively utilized in smaller groups where members have
face-to-face interactions than in larger, less intimate groups. Consequently,
large groups are prone to “free-riding”—an organizational condition in which
individuals benefit from a collective good that they do little to help produce.
SMALL GROUP INVOLVEMENT IN RELIGIOUS CONGREGATIONS 93
at Clemson University on July 16, 2014http://socrel.oxfordjournals.org/Downloaded from
Stark and Finke (2000) applied Olson’s theory to religious groups. They argued
that there is an inverse relationship between congregation size and average
level of member commitment. Increasing size reduces the density of social net-
works within a congregation resulting in less reinforcement for commitment,
less efficient monitoring of behavior, and a higher proportion of free-riding
members. Ample research documents the negative effect of increasing size on
congregational participation rates (Dougherty 2004;Finke 1994;Hougland and
Wood 1980;Pinto and Crow 1982;Stark and Finke 2000;Zaleski and Zech
1992).
2
The dilemma of large size for religious organizations is a notable one in the
United States. People are becoming increasingly concentrated in large congre-
gations (Chaves 2006). Nearly a third of all U.S. worshippers attend congrega-
tions with 1,000 or more in weekly attendance, though fewer than 3 percent of
all U.S. congregations are this size (Chaves 2004). The number of Protestant
churches attracting over 2,000 to weekly worship jumped 400 percent between
1990 and 2005; more than 1,200 of these “megachurches” now exist (Thumma
and Travis 2007). Megachurches have been criticized as gatherings of specta-
tors rather than participants (Guinness 1993). Recent empirical evidence sup-
ports this claim. A 2008 survey of 24,900 attendees in 12 U.S. megachurches
found that megachurch attendees went to worship services less frequently and
gave less financially to their church than did worshippers in U.S. congregations
overall (Thumma and Bird 2009). Likewise, in a nationally representative
sample of U.S. adults, persons in congregations of 2,500 members or more per-
ceived less social support from their fellow congregants than did persons in
smaller congregations (Ellison et al. 2009).
Small groups have been heralded as a remedy to the drawbacks of increas-
ing organizational size. Social incentives such as friendship and self-esteem can
motivate collective action in a large group, if it is organized as “a federation of
smaller groups” (Olson 1965:63). In effect, the beneficial attributes of small
groups are retained and simply aggregated to comprise a larger whole.
Combating perceptions of anonymity and alienation that plague large second-
ary groups, interpersonal bonds forged in a small group connect individuals to
the bigger organization and motivate their contribution to the collective good.
From the fellowship classes of nineteenth-century Methodists to the cell groups
of twenty-first-century megachurches, the organization of religious life around
small groups has a long and profitable history as an organizational strategy to
mitigate the free-rider problem (Stark and Finke 2000). The propensity for free
riding in large congregations heightens the significance of small groups in these
settings.
2
Stark (2008) is an exception. Based upon bivariate analysis of a national random
sample of U.S. adults, Stark found those in Protestant churches of over 1,000 to be as or
more likely than those in Protestant churches of less than 100 to attend weekly, tithe, and
have church friends.
94 SOCIOLOGY OF RELIGION
at Clemson University on July 16, 2014http://socrel.oxfordjournals.org/Downloaded from
Small groups are not new to religious organizations, but their role in con-
gregations and in U.S. society grew in prominence in the latter half of the
twentieth century. The contours of the modern small group movement, and its
impact on religion, were the focus of a comprehensive study by Robert
Wuthnow in the early 1990s. Wuthnow (1994b) traced the movement to the
1960s and declared it a response to Americans’ interest in spirituality and the
breakdown in traditional forms of community, such as neighborhoods and
families. In 1991, 40 percent of American adults belonged to a Sunday school
class, Bible study group, recovery group, book club, civic group, or some other
type of regularly meeting small group. Approximately two-thirds of small
groups were connected to a place of worship and they operated in congrega-
tions of all sizes and across religious traditions (Wuthnow 1994a,1994b).
Moreover, these groups have an effect on individuals and congregations.
Wuthnow (1994b,1994c) found that small group involvement related to
greater involvement in other congregational programs, increased financial
giving, active attendance, and more close friends in the congregation.
Individuals gave credit to small groups for deepening their personal faith and
for taking a more active role in their congregation. Social science research
has noted other positive consequences associated with congregational small
groups, ranging from enhanced racial-ethnic diversity (Dougherty 2003;
Dougherty and Huyser 2008;Emerson 2006) to congregational growth (Wilson
et al. 1993).
While Sunday school classes represent the most recognizable of congrega-
tional small groups, it is small Bible study, prayer, and discussion groups that
attract significant current attention. The smaller size of these groups fosters
accountability and trust to a greater extent than occurs in congregational edu-
cation classes (Wuthnow 1994b). Some go further to describe these small, rela-
tional groups (also known as “cell groups”) as one of the central innovations in
contemporary religious life. Popular church growth author and consultant Lyle
Schaller (1995:14) declared that a new reformation is underway in American
Christianity as evidenced by “perhaps, most important of all, the decision by
tens of millions of teenagers and adults to place a high personal priority on
weekly participation in serious, in-depth, lay-led, and continuing Bible study
and prayer groups.” Small/cell groups are prominently promoted in church
growth literature (e.g., Arnold 2004;Easum and Atkinson 2007;George 1992;
Neighbour 2000).
Use of small groups is widespread in megachurches. The world’s largest
congregation, Yoido Full Gospel Church in Seoul, South Korea, with over
800,000 members, grew to its epic proportions using a cell-based model of
organization. Home cell groups of 5–10 families provide an intimate setting for
study, discussion, and prayer. In the words of the founding pastor: “A cell
group is the basic part of our church. It is not another church program—it is
the program of our church” (Cho 1984:42). Cell groups are also vital to
Willow Creek Community Church (averaging 20,000þin South Barrington,
SMALL GROUP INVOLVEMENT IN RELIGIOUS CONGREGATIONS 95
at Clemson University on July 16, 2014http://socrel.oxfordjournals.org/Downloaded from
Illinois) and the seeker church movement it helped birth (Sargeant 2000).
From the 2008 sample of megachurch attendees, 60 percent indicated
participating in some type of small group, with groups meeting for prayer and
Bible study being most common (Thumma and Bird 2009). Other megachurch
studies affirm the influence of small groups on individuals’ beliefs and practices
(Dalton 2002;Martin 2007;Thomas and Jardine 1994). Ellison et al. (2009)
warned, however, that simply having small groups does not produce a warm,
caring, religious community. They found no relationship between the number
of special-purpose groups in a congregation and a worshipper’s anticipated
support. This study did not consider the impact of different types of special-
purpose groups. Data limitations also prevented measurement of the partici-
pation of individuals in small groups. Ellison and colleagues acknowledged
“perhaps these members [persons active in small group] within very large
churches enjoy a greater feeling of anticipated support from their fellows”
(p. 11).
Despite the attention given to small groups in the past two decades, there
is a lack of direct empirical evidence for the effectiveness of small, relational
groups in large congregations specifically. Are these groups capable of mitigat-
ing the complications of large size on individual-level commitment and partici-
pation? We strive to answer this question by investigating the effects of small
group involvement in U.S. congregations nationally and within one specific
megachurch. Two hypotheses drive our investigation:
(1) Small group involvement will be positively associated with individuals’
commitment and participation in religious congregations, regardless of
congregation size.
(2) Involvement in a small group will attenuate the negative relationship of
congregation size and individual-level commitment and participation.
DATA AND METHODOLOGY
Our analysis takes advantage of two sources of data. We begin with a
national look at small group involvement in congregations using the 2001 U.S.
Congregational Life Survey. A valuable complement to our research is a survey
of worshippers at one Texas megachurch built around the cell-group model.
Hence, our analysis benefits from national breadth as well as internal scrutiny
of one cell-based megachurch.
U.S. Congregational Life Survey
Our first source of data is the 2001 U.S. Congregational Life Survey
(USCLS), a national sample of U.S. congregations and their attendees. Based
upon a hypernetwork sampling approach which takes a random sample of indi-
viduals to identify a random sample of organizations, USCLS data provide a
96 SOCIOLOGY OF RELIGION
at Clemson University on July 16, 2014http://socrel.oxfordjournals.org/Downloaded from
representative snapshot of U.S. congregations.
3
The USCLS is unique in that
it gathered representative data at both organizational and individual levels.
Data collection involved a religious leader, who supplied information about the
organization as well as a survey of participants in each congregation (Woolever
and Bruce 2002). We rely primarily on the survey of attendees administered
during worship services on the weekend of April 29, 2001. Organizational fea-
tures including founding date, average worship attendance, denominational
affiliation, and regional location were taken from the congregational profile
completed by a religious leader. The combined USCLS data set for our vari-
ables of interest resulted in a usable sample of 78,474 individuals in 401 con-
gregations. The nationally representative, multilevel nature of these data make
the USCLS an invaluable resource for congregational research (e.g., Loveland
2005;Polson 2007;Scheitle and Adamczyk 2009;Scheitle and Finke 2008;
Schwadel 2009;Smith 2010).
With USLCS data, we constructed three dependent variables to operatio-
nalize congregants’ commitment and participation. Our first dependent vari-
able is a measure of commitment or belonging. Organizational commitment
is a multidimensional construct. People can be committed to organizations in
different ways and these differences have implications for how they feel about
an organization and how they behave (Meyer and Allen 1997). The limit-
ations of secondary data force us to depend on a single measure. The USCLS
asked, “Do you have a strong sense of belonging to this congregation?” There
are three possible “yes” and three possible “no” answers all with qualifiers.
The responses are “Yes, a strong sense of belonging that is growing”; “Yes, a
strong sense-about the same as last year”; “Yes, but perhaps not as strong as in
the past”; “No, but I am new here”; “No, and I wish I did by now”; “No, but I
am happy as I am”; and “Don’t know/not applicable.” We recoded this vari-
able so the first two “yes” responses where individuals reported feeling a
greater or equal amount of belonging to their congregation compared with
the previous year are equal to 1, setting responses of declining sense of
belonging, no sense of belonging, and don’t know/not applicable equal to 0.
4
Our belonging variable taps the affective dimension of commitment, i.e., the
emotional attachment one feels to an organization (Meyer and Allen 1997).
Though limited to a single measure, our study fits within a long line of pre-
vious studies showing affective commitment as a positive correlate to
employee outcomes such as retention, attendance, and job performance
(Meyer and Allen 1997).
3
In addition to the hypernetwork sample, the USCLS cooperated with eight denomi-
nations to generate denominational oversamples. These oversamples are not included in
our analysis.
4
The substantive results reported in this analysis do not change by coding the “Don’t
know/Not applicable” cases as missing or by including all three of the “Yes” responses as
equal to 1.
SMALL GROUP INVOLVEMENT IN RELIGIOUS CONGREGATIONS 97
at Clemson University on July 16, 2014http://socrel.oxfordjournals.org/Downloaded from
A standard measure of religious participation is attendance at worship ser-
vices. The USCLS asked attendees, “How often do you go to worship services at
this congregation?” The original response options ranged from 1, “This is my
first time,” to 7, “More than once a week.” We recoded the variable to a six-
point scale ranging from 1, “Hardly ever,” to 6, “More than once a week,” drop-
ping first-time visitors from our analysis. There are limitations associated with
attendance as a measure of participation. It is subject to over-reporting, norma-
tive differences across religious traditions, and it fails to capture a multitude of
other ways that individuals participate in congregations. Fortunately, frequency
of attendance is not our sole marker of religious participation. A second measure
is financial giving. The USCLS asked, “About how much do you give financially
to this congregation?” Respondents selected from five descending categories,
beginning with “I give 10% or more of net income regularly” and ending with “I
do not contribute financially here.” Responses were recoded so that higher
values equate with higher levels of giving to the congregation.
The independent variable in this study is small group involvement. We
created a dichotomous variable to distinguish respondents involved in “prayer,
discussion, or Bible study groups” (coded as 1) from those not involved in such
groups (coded as 0). Although the USCLS asked about two other group types
(Sunday/church/Sabbath school and fellowships/clubs), we restrict our analysis
to the type of relational, cell-group prevalent in megachurches (Thumma and
Bird 2009) and advocated in church growth literature (George 1992;
Neighbour 2000).
In addition to examining the effect of small group involvement at the indi-
vidual level, we are interested in investigating whether the effect of small
group involvement differs depending on congregation size. We created a cross-
level interaction term by multiplying small group involvement at the individ-
ual level (level 1) with congregation size at the congregational level (level 2).
The interaction term serves to test Hypothesis 2.
Individual-level control variables used for USCLS data include gender (1,
female; 0, male), age (limited to 18 and older), race (1, white; 0, other), marital
status (1, married; 0, other), children in the home (1, yes; 0, no), education (1,
no formal education, to 8, Master’s, Doctorate, or other graduate degree),
income (1, less than $10,000 in total household income before taxes, to 6,
$100,000 or more), and how long the respondent has attended the congregation
(1, less than a year, to 6, 20þyears). In multivariate tests of belonging and
giving, frequency of attendance is also treated as a control variable. Finally, a
series of congregational-level control variables are used in all USCLS models:
region (1, South; 0, other), congregation size (i.e., average weekly attendance),
year founded, and religious tradition (Steensland et al. 2000).
Central Texas Megachurch
A second data set gives us a more detailed look at small group involvement
within one large congregation. We make use of a survey conducted for purposes
98 SOCIOLOGY OF RELIGION
at Clemson University on July 16, 2014http://socrel.oxfordjournals.org/Downloaded from
of self-assessment by leaders of CTM, an evangelical non-denominational con-
gregation located in a medium-sized city. CTM was planted by a prominent
Southern Baptist congregation in the area in 1999. A successful associate
pastor with a vision for missions and small-group ministry left the mother
church to launch a new congregation. About 400 people attended the first
public worship service. By winter of that year, the count of people regularly
attending Sunday services in one rented space after another was 900. Since
2004, the church has averaged at or above 2,000 in weekly worship services.
Small, cell groups are a basic building block of the church. The stated
purpose for these groups is personal growth, evangelism, and the facilitation of
community. Groups of 8– 15 people, typically at the same stage of life, gather
in members’ homes to sing, receive instruction, discuss, and pray. Small groups
are expected to add members with the intent of eventually splitting from one
group into two. Thus, the church promotes continued growth through the
process of small group multiplication. In 2010, CTM had over 100 small, home
groups in operation. Each small group is led by two individuals. These individ-
uals join leaders from similar small groups to form sections. Every section has
two leaders. Once a month all of the small group leaders in each section meet
together with their section leaders. This serves to encourage those leaders as
well as maintain accountability throughout the church. Each section is part of
a zone made up of similar sections. A pastor on staff at CTM provides oversight
for each zone. Zone pastors report to the senior pastor and board of elders.
In fall 2007, CTM conducted its own internal survey of congregants.
A psychologist, who is a member of the church and experienced in survey
research, designed the survey instrument in consultation with church leaders.
The intent of the survey was to measure the congregation’s spiritual health and
religious activity. The church’s weekly e-newsletter provided a link to the ques-
tionnaire online. A print version of the questionnaire was distributed during
Sunday worship services on two occasions and once during a separate college
worship service. In the end, a total of 1,014 individuals completed surveys.
With an average attendance of 2,000 at the time, the estimated response rate
for the survey was 51 percent.
5
We must acknowledge the possibility that the
1,014 respondents to the CTM survey are not fully representative of the entire
CTM congregation. Perhaps they represent only the most highly involved.
We cannot dismiss this concern entirely, but the same critique applies to
the USCLS which was administered to persons in attendance on one
5
Calculating a response rate for the CTM survey is problematic. The church uses
average attendance as its measure of size. This assumes that the 2,000 people in attendance
each week are the same people. One reviewer pointed out the true population of the
church could actually be much higher, meaning that the response rate to the survey would
be lower than we report. Unfortunately, we have no way to determine what the “true popu-
lation” of Central Texas Megachurch was in 2007. In working with a church administrator,
we only know how many individuals attended weekend services during the time of the
survey. The calculated response rate is our best approximation given available information.
SMALL GROUP INVOLVEMENT IN RELIGIOUS CONGREGATIONS 99
at Clemson University on July 16, 2014http://socrel.oxfordjournals.org/Downloaded from
particular weekend. Both are surveys of participants. There is variation in the
extent of participation in both data sets however. Given the limitations of sec-
ondary data analysis, the best we can do is acknowledge these perils and
proceed with caution.
Although not identical to the measures in the USCLS, the CTM survey
contains questions on belonging, attendance, and giving as well. The CTM
survey relied heavily on a four-point Likert scale of agreement/disagreement.
Regarding belonging, respondents were presented the statement, “I feel con-
nected to others at [Central Texas Megachurch] (a sense of belonging) and to
the church vision.”
6
We created a dichotomous variable of belonging with
CTM data in which 1 signifies agree, and 0 signifies disagree to feeling connec-
tion/belonging at the church. Regarding attendance, the CTM survey asked,
“On average how often do you attend Sunday services at [Central Texas
Megachurch]?” Possible responses were “Less than Once a Month,” “Once or
Twice a Month,” “Three Times a Month,” and “Every Week.” Responses to
this question were highly skewed, with 80 percent of respondents indicating
that they attended services every week. To address this we recoded the fre-
quency of attendance variable for CTM respondents to be dichotomous, separ-
ating those that attend weekly (coded as 1) from those that attend less often
(coded 0). To measure the extent of giving, the CTM survey used a four-point
Likert scale to measure agreement/disagreement to the statement: “Tithing
(giving 10% of my income) to [Central Texas Megachurch] is a consistent part
of my worship.” Our interest is on the actual practice of giving as opposed to
one’s attitude. As a gauge of giving level, we collapsed the Likert scale into a
dichotomous variable where 1 represents those who do give 10 percent of their
income to the church consistently (Agree or Strongly Agree) and 0 represents
those who presumably do not (Disagree or Strongly Disagree).
Again, our primary independent variable is small group involvement. As
an enhancement to the analysis, CTM data allow us to investigate the extent
of small group involvement for individuals. The CTM survey asked respondents
how often they typically attend a small group. We created a series of dichoto-
mous variables to differentiate levels of small group involvement: Not in a
group, attend once or twice a month, attend three times a month, and attend
every week.
Control variables in CTM data include age (limited to 18 and older), race
(1, white; 0, other), marital status (1, married; 0, other), children in the home
(1, yes; 0, no), and how long the respondent has attended the congregation (1,
less than six months, to 4, more than three years). Frequency of attendance is
6
On this and all other CTM survey items, we quote the questions directly as taken
from the survey instrument. The parenthetical statement “sense of belonging” appeared in
the original. Regretably, the double-barreled nature of the question conflates sense of
belonging with connection to the church vision. Though not ideal, this is the only measure
of belonging available from the CTM survey.
100 SOCIOLOGY OF RELIGION
at Clemson University on July 16, 2014http://socrel.oxfordjournals.org/Downloaded from
included as a control variable in multivariate tests of belonging and giving. For
these tests, we use the original four-point attendance scale, ranging from 1, less
than once a month to 4, every week.
Methodology
Because the USCLS encompasses two different levels of analysis, individ-
ual and congregational, we use hierarchical linear modeling (HLM)
(Raudenbush and Bryk 2002). HLM analysis offers advantages to ordinary least
squares when one level of data is nested within another level. Multilevel data
violate the assumption of independence in standard regression techniques and
the clustering of individuals within groups results in biased standard errors.
HLM corrects for this by recognizing the partial independence of data residing
at either level (Hofmann 1997). HLM also allows researchers to investigate the
extent to which the higher level can influence the lower level. With data at
both the congregational and individual levels in the USCLS, HLM analysis is
ideal. Separate analyses were performed for each dependent variable. Model
specifications were set to Bernoulli (0 or 1) for the dichotomous sense of
belonging variable and set to Normal for the other two dependent variables. In
each model, all variables besides those that are categorical were set to their
grand means. Additionally, in the giving and attendance models, all variables
except female were allowed to vary across congregations. In the sense of
belonging model small group involvement, married, white, child in home, and
female were not allowed to vary across congregations. This is because in all
three cases, these variables did not vary significantly across congregations in
preliminary analyses.
The second stage of analysis takes us inside CTM. We begin with a presen-
tation of descriptive statistics for the congregation. Multivariate analyses follow
for more rigorous tests of the effect of small group involvement. Since all of
the dependent variables in CTM data are dichotomous, model estimation
depends on binary logistic regression exclusively.
Before presenting results, one more disclaimer is in order. The cross-
sectional nature of our data prevents us from determining whether small groups
are a source of congregational commitment and participation. Our analysis can
only test whether small group involvement is related to individuals’ belonging,
attendance, and giving in congregations. We have used this relational language
to frame our hypotheses and do our best to interpret our findings in the same
way. In the conclusion, we return to the question of causation.
RESULTS
Table 1reports descriptive statistics of the individuals and congregations
featured in the analysis of USCLS data. These statistics give a helpful portrayal
of congregational and small group involvement among U.S. worshippers.
SMALL GROUP INVOLVEMENT IN RELIGIOUS CONGREGATIONS 101
at Clemson University on July 16, 2014http://socrel.oxfordjournals.org/Downloaded from
Seventy-five percent of respondents felt a steady or growing sense of belonging
to their congregation. The mean frequency of attendance at worship services
was two to three times a month. The mean level of giving was less than
5 percent of respondent’s annual income. In terms of small group involvement,
21 percent of these congregants reported participating in a prayer, discussion,
or Bible study group through their congregation.
Table 2contains the full HLM results for each dependent variable.
Small group involvement stands out as significantly related to individuals’
congregational commitment and participation, net of other individual-level
and congregation-level characteristics. Persons involved in prayer, discussion,
or Bible study groups are associated with a stable or growing sense of
belonging, more frequent attendance at worship services, and higher rates of
financial giving across U.S. congregations. At the congregation level, size is
negatively associated with two of the dependent variables. Both belonging
and giving are lower in larger congregations. The cross-level interaction of
small group involvement by congregation size is not significant in any
TABLE 1 Descriptive Statistics for USCLS (2001)
Mean SD Min. Max.
Level 1 (Individual) (N ¼78,474)
Sense of belonging 0.75 — 0 1
Frequency of attendance 4.90 0.82 1 6
Giving 3.47 1.07 1 5
Small group involvement 0.21 — 0 1
Length of attendance 4.07 1.68 1 6
Education 5.64 1.72 1 8
Income 3.65 1.47 1 6
Female 0.60 — 0 1
Age 51.01 16.26 18 100
White 0.79 — 0 1
Married 0.70 — 0 1
Child in home 0.42 — 0 1
Level 2 (congregation) (N ¼401)
Size 602.39 882.79 9 5,400
Year founded 1,912.76 57.57 1635 1,998
South 0.33 — 0 1
Evangelical 0.28 — 0 1
Black Protestant 0.05 — 0 1
Mainline 0.40 — 0 1
Catholic 0.23 — 0 1
Jew 0.01 — 0 1
Other 0.04 — 0 1
102 SOCIOLOGY OF RELIGION
at Clemson University on July 16, 2014http://socrel.oxfordjournals.org/Downloaded from
model. The effect of small group involvement does not seem to differ by
congregation size.
Several control variables have a bearing on individuals’ connections to
congregations nationally. Frequency of attendance and length of attendance
are positively related to all three measures of commitment and participation as
are age and being married. The influence of education, income, gender, chil-
dren at home, and race are more mixed. Higher levels of education correspond
TABLE 2 Effect of Small Group Involvement on Individual Commitment and
Participation in U.S Congregations (Hierarchical Linear Modeling)
Belonging
a
T-ratio Attendance
b
T-ratio Giving
b
T-ratio
Level 1 (Individual)
Small group
involvement
0.79*** 22.97 0.48*** 41.24 0.28*** 24.93
Frequency of
attendance
0.66*** 41.48 — — 0.33*** 49.40
Length of
attendance
0.22*** 22.93 0.03*** 10.99 0.07*** 21.65
Education 20.07*** 28.26 0.01*** 3.59 0.01* 2.22
Income 20.02* 22.00 20.01*** 23.68 0.05*** 12.62
Female 20.09*** 24.60 0.03*** 4.57 0.06*** 9.32
Age 0.001* 1.99 0.01*** 15.64 0.01*** 26.83
White 20.20*** 26.54 0.06*** 4.52 0.01 0.58
Married 0.11*** 4.87 0.05*** 5.93 0.18*** 17.70
Child in home 20.08*** 23.73 20.01 21.67 0.01 1.25
Level 2(Congregation)
Size 20.0001** 22.95 0.00001 0.41 20.00004** 22.79
Year founded 20.0002 20.49 0.0003* 1.96 0.001** 3.03
South 0.09 1.83 0.05** 2.66 0.05* 2.29
Mainline
Protestant
c
0.32*** 5.09 20.35*** 214.61 20.38*** 212.95
Black Protestant
c
0.33* 2.54 20.09 21.83 0.28*** 4.72
Catholic
c
20.03 20.49 20.14*** 24.88 20.56*** 216.19
Jew
c
1.16*** 4.08 21.47*** 215.31 20.26* 22.36
Other
c
0.45*** 3.54 20.50*** 210.07 20.46*** 28.02
Cross-level interaction (Level 1 Level 2)
Small group
involvement size
0.00001 0.42 20.000002 20.19 20.00001 20.75
Source: USCLS (2001).
a
Model specifications set to Bernoulli (0 or 1) due to dichotomous coding of depen-
dent variable.
b
Model specifications set to Normal for interval-level dependent variables.
c
Contrast category is Evangelical Protestant.
*p.05; **p.01; ***p.001.
SMALL GROUP INVOLVEMENT IN RELIGIOUS CONGREGATIONS 103
at Clemson University on July 16, 2014http://socrel.oxfordjournals.org/Downloaded from
to decreased odds of feeling a sense of belonging but higher levels of attend-
ance and giving. Income has a negative relationship with belonging and
attendance but a predictably positive relationship with giving. Females are less
likely to report a stable or growing sense of belonging than males but are likely
to attend and give at higher rates. Having children in the home is negatively
associated only with belonging. Finally, whites are less likely to report a sense
of belonging but are more likely to attend at higher rates than non-whites.
In addition to size at the congregation level, congregation age, year
founded, region, and religious tradition turn up as significant. Older congrega-
tions and congregations in the south have higher levels of attendance and
giving. Evangelical Protestant congregations have lower levels of belonging
than all other religious traditions besides Catholics; at the same time,
Evangelical Protestant congregations have higher levels of attendance and
giving than all traditions except Black Protestants.
Knowing that small group involvement is associated with greater commit-
ment and participation in congregations nationally, we now consider the
impact that frequency of small group involvement has on individuals within
one specific megachurch. Table 3takes us inside the doors of CTM. This pre-
liminary profile depicts CTM as a congregation comprised of a very committed,
active laity. Nine out of 10 CTM respondents expressed a sense of belonging
to the church, 8 out of 10 said they attended worship services weekly or more,
and more than 7 in 10 reported that they consistently tithe 10 percent of their
income to the church. The cell-based organizational model of CTM also is
clearly evident. Only 12 percent of respondents reported not participating in a
TABLE 3 Descriptive Statistics for CTM (2007)
Mean SD Min. Max.
Sense of belonging 0.89 — 0 1
Attend weekly
a
0.80 — 0 1
Give 10 percent 0.74 — 0 1
Not in a small group 0.12 — 0 1
Attend small group 1–2/month 0.08 — 0 1
Attend small group 3/month 0.12 — 0 1
Attend small group every week 0.68 — 0 1
Length of attendance 3.06 1.07 1 4
Age 29.57 12.10 18 78
White 0.88 — 0 1
Married 0.45 — 0 1
Child in home 0.28 — 0 1
Source: Self-reported data from 1,014 congregants.
a
In multivariate models, worship service attendance is measured using a four-point
scale with mean ¼3.72 and standard deviation ¼0.63.
104 SOCIOLOGY OF RELIGION
at Clemson University on July 16, 2014http://socrel.oxfordjournals.org/Downloaded from
small group at CTM. More than two-thirds (68 percent) said that they
attended small group meetings every week. Another 12 percent attended three
times a month, and 8 percent attended once or twice a month.
Demographically, the CTM sample is fairly young, white, and has a relatively
short tenure in the church (less than three years on average). We must
acknowledge again that this description is speculative. Like the USCLS, CTM
data come from a survey of participants, but it is a survey based on a sample.
We cannot know for certain that the 51 percent of CTM congregants who
completed the survey were comparable to the 49 percent who did not. We
present these descriptive statistics as merely suggestive.
Table 4reports binary logistic regression results for the frequency of small
group attendance and control variables regressed upon our dependent variables.
The first set of estimates is for sense of belonging. Attending small group
weekly multiplies the odds of having a sense of belonging by 4.1 compared to
those not in a small group. It is only at these extremes that statistically signifi-
cant differences emerge.
Estimating weekly Sunday worship attendance, we see the same pattern.
Attending small group weekly multiplies the odds of attending Sunday worship
weekly by 6.7 compared to those not in a small group. There is no significant
TABLE 4 Effect of Small Group Involvement on Individual Commitment and
Participation in CTM (Binary Logistic Regression)
Variables Sense of belonging Attend weekly Give 10 percent
Estimate (SE) OR Estimate (SE) OR Estimate (SE) OR
Small group attendance
a
1–2/month 0.37 (0.34) — 20.02 (0.29) — 0.12 (0.33) —
3/month 0.59 (0.33) — 0.41 (0.27) — 20.01 (0.30) —
Every wk 1.41*** (0.28) 4.11 1.90*** (0.24) 6.71 0.34 (0.24) —
Frequency of attendance 0.30* (0.14) 1.35 — — 0.73*** (0.12) 2.07
Length of attendance 0.25 (0.10) 1.27 0.08 (0.09) — 0.42*** (0.08) 1.52
Age 20.02* (0.01) 0.98 0.01 (0.01) — 0.01 (0.01) —
White 20.16 (0.32) — 0.15 (0.25) — 20.13 (0.24) —
Married 20.32 (0.27) — 20.24 (0.23) — 0.90*** (0.23) 2.46
Child in home 20.03 (0.27) — 20.02 (0.23) — 20.31 (0.24) —
Constant 0.75 20.20 22.31***
PRE 9.7% 12.5% 12.8%
N995 995 995
Source: CTM Internal Survey (2007).
a
Contrast category is not in a small group.
*p,.05; **p,.01; ***p,.001.
SMALL GROUP INVOLVEMENT IN RELIGIOUS CONGREGATIONS 105
at Clemson University on July 16, 2014http://socrel.oxfordjournals.org/Downloaded from
association between attending small groups and the CTM measure of financial
giving, net of other effects.
As seen in the national sample of U.S. congregations, the control variable
with the most consistent influence on commitment and participation in CTM
is frequency of worship service attendance. More frequent attendance at
Sunday morning services is associated with sense of belonging and a high rate
of financial giving. Similarly, people with longer tenures in the church are
more likely to report a sense of belonging and give 10 percent of their income.
The only other significant control variables were age and married. Older
respondents reported a lower sense of belonging at CTM. The odds favoring
the giving of 10 percent of one’s income to the church is 2.46 times higher for
married people than for people not currently married. Given the absence of an
income measure in CTM data, the marriage variable may serve as a proxy for
higher income which is a recognized correlate to giving (Hoge et al. 1996).
DISCUSSION
The findings presented provide clear support for our first hypothesis.
National data from over 78,000 U.S. worshippers in over 400 congregations
reveal that individuals involved in prayer, discussion, or Bible study groups
have higher rates of commitment and participation than those not involved in
a small group. Small group members report a greater sense of belonging, attend
services more frequently, and contribute a higher percentage of their income
to their congregation. These findings are consistent and robust.
By looking at participants within one megachurch in Central Texas, we
further discover that level of involvement within a small group has impli-
cations for commitment and participation. It is in regular involvement in a
small group where we see these implications. Increases in sense of belonging
and regular attendance at Sunday services are more likely only for those who
attend small groups every week. Even though small group involvement has no
direct effect on financial giving for CTM worshippers, it likely has a positive
indirect effect through Sunday service attendance. Weekly small group partici-
pants attend services weekly, and weekly service attendance is strongly corre-
lated with financial giving. Testing indirect effects is beyond the scope of this
analysis, so we simply mention the possibility here. Nevertheless, our analysis
of CTM worshippers aligns with the findings of Ellison et al. (2009) in demon-
strating that simply having small groups is not enough for a large congregation.
The key is getting people embedded within these groups.
Beyond direct effects of small group involvement on commitment and belong-
ing, we also considered the potential for an interaction effect of small group invol-
vement by congregation size. Hypothesis 2 tested the popular speculation that
small groups help remedy the negative consequences of increasing size.
Our national analysis failed to support this prediction. The joint effect of small
106 SOCIOLOGY OF RELIGION
at Clemson University on July 16, 2014http://socrel.oxfordjournals.org/Downloaded from
group involvement and congregation size was not significant in any model. The
benefits of being in a small group do not seem to differ by congregation size. Small
group members are more committed participants in large and small congregations.
The problem facing large congregations is that the average level of belonging and
participation outside small groups is lower than it is in smaller congregations.
Alas, the negative consequence of size persists.
CONCLUSION
Huge congregations can be daunting places. The flurry of activity and
plethora of people can be downright dizzying to a newcomer. This was the
reaction of Amy and Jeremy on their first visit to CTM. They did not intend
to return. Even if they had, they likely would have become a pair of passive
spectators amid the sea of faces in Sunday morning worship services. Their
story took a turn however. This turn is telling for American congregational life
in an era marked by the growing prominence of very large congregations.
Amy and Jeremy found a place to belong at CTM through a small group.
This point of connection quickly transformed them into active, committed par-
ticipants within CTM. It is a pattern evident throughout American congrega-
tions. Analyzing worshippers across the United States as well as within one
Texas megachurch, this study provides important empirical evidence linking
small group involvement to congregational commitment and participation. No
matter the size of the congregation, individuals involved in small, relational
groups that meet for prayer, discussion, and Bible study are more likely to
express a sense of belonging, attend religious services more often, and give
financially at a higher rate than are persons not involved in such groups. So,
stories like that of Amy and Jeremy are not exclusive to CTM. Small groups
provide worshippers in congregations of all sizes a place to belong and a foun-
dation for greater participation. In light of our findings, we arrive at several
conclusions.
Most importantly, being involved in a small, intimate gathering with
fellow worshippers is related to positive religious outcomes for individuals. We
intentionally limited our analysis to a specific type of intimate, relational,
small group experience. Our interest is the prayer, discussion, and Bible study
groups prevalent in megachurches. One in five American worshippers belongs
to such a group. For those that do, they feel a stronger connection and they
give more time and money to their congregation. There is still much about
what happens inside these groups that deserves investigation. How do people
join such groups, especially in large congregations? How does the content of
group meetings affect individuals’ religiosity? Are there benefits to standardiz-
ing small group format as done in CTM?
A second conclusion, derived from our findings at CTM, is that the level
of involvement within small groups has bearing on individuals’ overall
SMALL GROUP INVOLVEMENT IN RELIGIOUS CONGREGATIONS 107
at Clemson University on July 16, 2014http://socrel.oxfordjournals.org/Downloaded from
congregational commitment and participation. Crucial differences appeared
between individuals who attended small group meetings weekly and those that
did not attend at all. CTM worshippers who attended a small group weekly
were more likely to report a sense of belonging to the church and more likely
to attend worship services weekly, as compared to CTM worshippers not
involved in small groups. This mirrors Wuthnow’s (1994b,1994c) observation
of the spiritual benefits of regular small group participation; people who
attended a small group every week expressed a deeper faith that they attributed
to their group involvement when compared with people attending a group less
often (Wuthnow 1994b:232–33). Recovery groups like Alcoholics Anonymous
(AA) recognize the transformative potential of participation and, conse-
quently, stress regular attendance at group meetings. This type of intensive
ingroup activity is an important part of the commitment building process in
AA (Rudy and Greil 1987). It is also significantly related to recovery success
(Humphreys et al. 1997).
Our third conclusion comes in the form of a caveat. Small groups are ben-
eficial, but their benefits do not eliminate or mitigate the negative conse-
quences of increasing congregation size. People in large congregations feel less
belonging and contribute less financially to their places of worship than do
persons in small congregations. As noted above, simply having small groups is
not sufficient to raise levels of congregational involvement. Individuals have to
involve themselves in these groups. This may pose a particular problem in the
megachurch context. Some suggest that megachurches are qualitatively differ-
ent in culture and composition from their smaller counterparts (Schaller
2000). Megachurches attract younger people, more singles, and those with
higher levels of education and income; almost a quarter were not actively
involved in a congregation previously (Thumma and Bird 2009). Choosing to
attend a church of thousands says something about what someone wants and
expects out of church. Intimacy and accountability may be aspects of congrega-
tional life that people are intentionally trying to avoid by attending a mega-
church. If this is the case, the fact that small group involvement sustains rates
of congregational engagement as much as it does in large congregations is illu-
minating. Whether a small group is functioning within a context of 100 or
1,000, the positive effects on individual religious involvement are the same.
The challenge for megachurches is getting people into smaller, relational units.
Resolving this leads us to a final conclusion.
With due caution, we direct our fourth conclusion to the potential for
success we see in the organization of a large congregation as a “congregation of
small groups.” CTM is a congregation built upon small groups. If these data are
reflective of the congregation as a whole, CTM is a church with exceptionally
high vitality despite an average attendance of thousands. More than 80 percent
of respondents expressed a strong sense of belonging and attended worship ser-
vices weekly or more. Three-fourths reported giving 10 percent of their income
to the church on a consistent basis. Descriptive statistics such as these would
108 SOCIOLOGY OF RELIGION
at Clemson University on July 16, 2014http://socrel.oxfordjournals.org/Downloaded from
support the contentions of Olson (1965),Stark and Finke (2000), and church
growth proponents (George 1992;Neighbour 2000) regarding the value of
organizing congregations around small groups. We recognize that CTM is a
relatively new congregation with a popular pastor, contemporary music, and a
heavy emphasis on evangelism. And questions about data quality loom. For
this reason, we cannot definitively conclude that small groups drive CTM vital-
ity, but our findings are suggestive. Improved data and additional testing on the
performance implications of cell-based congregations would be a fruitful path
for future research.
In closing, we consider the issue of causation. Do small groups push people
to greater commitment and participation, or are they the product of com-
mitted, active people finding each other? We cannot adequately answer this
question with our data. We have no way of knowing how committed or active
people were in their congregation before joining a small group. Our sense,
based on previous studies and anecdotal evidence, is that causation is apt to go
both ways. Certainly, people who desire involvement will find small groups
attractive. These are often the same people who sing in the choir on Sunday
mornings, assist in the children’s program on Wednesday nights, and show up
for Saturday work days. For others, however, getting involved in a small group
is a step toward greater investment in a congregation. Once again, this leads us
to a call for further research. A valuable starting point may be to follow newco-
mers over time in a congregation in which small groups are prominent. Tracing
the transition of newcomer to committed, active member would help sort out
where small groups fit in the causal order. The stories of people like Amy and
Jeremy attest to the formative role small groups can play in the lives of individ-
uals. By forging relational ties and connecting newcomers with the actively
committed, we believe that small groups represent a potent source of vitality in
congregations big and small. We invite others to test this claim.
REFERENCES
Arnold, Jeffrey. 2004. The Big Book on Small Groups, rev. ed. Downers Grove: InterVarsity
Press.
Chaves, Mark. 2004. Congregations in America. Cambridge: Harvard University Press.
——— 2006. “All Creatures Great and Small: Megachurches in Context.” Review of
Religious Research 47, no. 4:329– 46.
Cho, Paul Y., and R. Whitney Manzano. 1984. More than Numbers. Waco, TX: Word
Books.
Cooley, Charles H. [1909] 1983. Social Organization: A Study of the Larger Mind.New
Brunswick: Transaction Books.
Dalton, Ray. 2002. “Sources of Attraction to the Megachurch: Factors Influencing the
Individual’s Decision to Participate.” PhD dissertation, Department of Sociology,
University of Tennessee.
Dougherty, Kevin D. 2003. “How Monochromatic Is Church Membership? Racial-ethnic
Diversity in Religious Community.” Sociology of Religion 64, no. 1:65– 85.
SMALL GROUP INVOLVEMENT IN RELIGIOUS CONGREGATIONS 109
at Clemson University on July 16, 2014http://socrel.oxfordjournals.org/Downloaded from
——— 2004. “Institutional Influences on Growth in Southern Baptist Congregations.”
Review of Religious Research 46, no. 2:117– 31.
Dougherty, Kevin D., and Kimberly R. Huyser. 2008. “Racially Diverse Congregations:
Organizational Identity and the Accommodation of Differences.” Journal for the
Scientific Study of Religion 47, no. 1:23 – 43.
Easum, Bill, and John Atkinson. 2007. Go Big with Small Groups: Eleven Steps to an
Explosive Small Group Ministry. Nashville: Abingdon Press.
Ellison, Christopher G., Neal M. Krause, Bryan C. Shepherd, and Mark A. Chaves. 2009.
“Size, Conflict, and Opportunities for Interaction: Congregational Effects on Members’
Anticipated Support and Negative Interactions.” Journal for the Scientific Study of
Religion 48, no. 1:1–15.
Emerson, Michael O., and Rodney M. Woo. 2006. People of the Dream: Multiracial
Congregations in the United States. Princeton: Princeton University Press.
Finke, Roger. 1994. “The Quiet Transformation: Changes in Size and Leadership of
Southern Baptist Churches.” Review of Religious Research 36, no. 1:3–22.
George, Carl F. 1992. Prepare Your Church for the Future. Grand Rapids: Fleming H. Revell.
Guinness, Os 1993. Dining with the Devil: The Megachurch Movement Flirts with Modernity.
Grand Rapids: Baker Book House.
Hofmann, David A. 1997. “An Overview of the Logic and Rationale of Hierarchical Linear
Models.” Journal of Management 23, no. 6:723 –44.
Hoge, Dean R., Charles Zech, Patrick McNamara, and Michael J. Donahue. 1996. Money
Matters: Personal Giving in American Churches. Louisville: Westminster John Knox Press.
Hougland, James G. Jr., and James R. Wood. 1980. “Control in Organizations and the
Commitment of Members.” Social Forces 59, no. 1:85–105.
Humphreys, Keith, Rudolf H. Moos, and Caryn Cohen. 1997. “Social and Community
Resources and Long-term Recovery from Treated and Untreated Alcoholism.” Journal
of Studies on Alcohol 58, no. 3:231 – 38.
Kanter, Rosabeth M. 1972. Commitment and Community: Communes and Utopias in
Sociological Perspective. Cambridge: Harvard University Press.
Loveland, Matthew T. 2005. “Civic Congregations: Congregational Dynamics and
Individual Civic Involvement.” PhD dissertation, Department of Sociology, University
of Notre Dame.
Martin, Nancy J. 2007. “Small Groups in Big Churches.” PhD dissertation, Department of
Sociology, University of Arizona.
Meyer, John P., and Natalie J. Allen. 1997. Commitment in the Workplace: Theory, Research,
and Application. Thousand Oaks, CA: Sage Publications.
Neighbour, Ralph W. Jr. 2000. Where Do We Go from Here? A Guidebook for the Cell Group
Church, rev. ed. Houston: Touch Publications.
Olson, Mancur. 1965. The Logic of Collective Action: Public Goods and the Theory of Groups.
Cambridge: Harvard University Press.
Pinto, Leonard J., and Kenneth E. Crow. 1982. “The Effects of Size on Other Structural
Attributes of Congregations within the Same Denomination.” Journal for the Scientific
Study of Religion 21, no. 4:304 – 16.
Polson, Edward C. 2007. “Do Churches Make Good Neighbors? The Effects of
Congregational Context on Civic Participation.” Paper presented at the annual
meeting of the Society for the Scientific Study of Religion, Tampa, FL.
Raudenbush, Stephen W., and Anthony S. Bryk. 2002. Hierarchical Linear Models: Applications
and Data Analysis Methods, 2nd ed. Thousand Oaks, CA: Sage Publications Inc.
Rudy, David R., and Arthur L. Greil. 1987. “Taking the Pledge: The Commitment Process
in Alcoholics Anonymous.” Sociological Focus 20, no. 1:45–59.
110 SOCIOLOGY OF RELIGION
at Clemson University on July 16, 2014http://socrel.oxfordjournals.org/Downloaded from
Sargeant, Kimon H. 2000. Seeker Churches: Promoting Traditional Religion in a Nontraditional
Way. New Brunswick: Rutgers University Press.
Schaller, Lyle E. 1995. The New Reformation: Tomorrow Arrived Yesterday. Nashville, TN:
Abingdon Press.
——— 2000. The Very Large Church: New Rules for Leaders. Nashville, TN: Abingdon
Press.
Scheitle, Christopher P., and Amy Adamczyk. 2009. “It Takes Two: The Interplay of
Individual and Group Theology on Social Embeddedness.” Journal for the Scientific
Study of Religion 48, no. 1:16 –29.
Scheitle, Christopher P., and Roger Finke. 2008. “Maximizing Organizational Resources:
Selection versus Production.” Social Science Research 37, no. 3:815–827.
Schwadel, Philip. 2009. “Neighbors in the Pews: Social Status Diversity in Religious
Congregations.” International Journal of Religion Research 5: Article 2. Retrieved June
15, 2010 (www.religjournal.com).
Smith, Buster G. 2010. “Acceptance of Other Religions in the United States: An HLM
Analysis of Variability across Congregations.” Social Compass 57, no. 1:127– 142.
Stark, Rodney. 2008. What Americans Really Believe. Waco, TX: Baylor University Press.
Stark, Rodney, and Roger Finke 2000. Acts of Faith: Explaining the Human Side of Religion.
Berkeley: University of California Press.
Steensland, Brian, Jerry Z. Park, Mark D. Regnerus, Lynn D. Robinson, W. Bradford
Wilcox, and Robert D. Woodberry. 2000. “The Measure of American Religion:
Toward Improving the State of the Art.” Social Forces 79, no. 1:291– 318.
Thomas, George M., and Douglas S. Jardine. 1994. “Jesus and Self in Everyday Life:
Individual Spirituality Through a Small Group in a Large Church.” In “I Come Away
Stronger”: How Small Groups Are Shaping American Religion, edited by R. Wuthnow,
275–99. Grand Rapids: William B. Eerdmans.
Thumma, Scott, and Warren Bird. 2009. Not Who You Think They Are: The Real Story of
People Who Attend America’s Megachurches. A joint publication of Leadership Network
(Dallas, TX) and Hartford Institute for Religion Research (Hartford, CT). Retrieved
July 9, 2009 (hirr.hartsem.edu/megachurch/megachurch_attender_report.htm.).
Thumma, Scott, and Dave Travis. 2007. Beyond Megachurch Myths: What We Can Learn
from America’s Largest Churches. San Francisco: Jossey-Bass.
Wilson, Gerald L., Joann Keyton, G. David Johnson, Cheryl Geiger, and Johanna C.
Clark. 1993. “Church Growth Through Member Identification and Commitment: A
Congregational Case Study.” Review of Religious Research 34, no. 3:259– 72.
Woolever, Cynthia, and Deborah Bruce. 2002. A Field Guide to U.S. Congregations: Who’s
Going Where and Why. Louisville: Westminster John Knox Press.
Wuthnow, Robert, ed. 1994a. “I Come Away Stronger”: How Small Groups Are Shaping
American Religion. Grand Rapids: William B. Eerdmans.
Wuthnow, Robert, ed. 1994b. Sharing the Journey: Support Groups and America’s New Quest
for Community. New York: The Free Press.
Wuthnow, Robert. 1994c. “The Small-Group Movement in the Context of American
Religion.” In “I Come Away Stronger”: How Small Groups Are Shaping American
Religion, edited by R. Wuthnow, 344– 66. Grand Rapids: William B. Eerdmans.
Zaleski, Peter A., and Charles E. Zech. 1992. “Determinants of Contributions to Religious
Organizations: Free Riding and Other Factors." American Journal of Economics and
Sociology 51, no. 4:459–72.
SMALL GROUP INVOLVEMENT IN RELIGIOUS CONGREGATIONS 111
at Clemson University on July 16, 2014http://socrel.oxfordjournals.org/Downloaded from
