Reserve Readings

Capacities and Challenges of Faith-Based Human Service Organizations 115

Th is essay addresses four questions by comparing the

management capacities and challenges of congregations,

faith-based organizations, and secular organizations that

provide human services: (1) What role, if any, do congre-

gations and faith-based organizations currently play in

the social service delivery system? (2) Are congregations

interested in changing their role in the social service deliv-

ery system? (3) Compared to faith-based and secular

organizations, do congregations have the capacity to adapt

to new roles in the social service delivery system? Finally,

(4) compared to faith-based and secular organizations, do

congregations have similar service capacities and manage-

ment challenges? Th e fi ndings indicate that although

more than half of congregations already provide some

type of health or human services, they provide a narrower

range of services, consider these services a lower priority,

and seem to encounter more extensive management chal-

lenges than faith-based and secular organizations.

B eginning with the Charitable Choice provision

of the 1996 welfare reform act, several public

policy initiatives have sought to promote the

faith-based provision of human services. Th e arguments

used to promote these initiatives emphasize the power

of faith to change human behavior and suggest that

sacramental organizations are more eff ective than tradi-

tional nonprofi t providers at achieving desirable out-

comes ( Kennedy 2001; Kennedy and Bielefeld 2002 ).

Th ese eff orts, intensifi ed under the George W. Bush

administration, raise constitutional questions about the

separation of church and state ( Kennedy 2001 ), as well

as policy questions about the consequences for clients

and traditional human service nonprofi ts. As Cnaan

and Boddie (2002) argue, these developments have

signifi cant implications for the social work profession,

and social work students need a fi rm grounding in

these issues in policy and practice courses. Th e same

holds for students in programs of public administra-

tion, public aff airs, and nonprofi t management.

Under the Charitable Choice provision, Section 104 of

the Personal Responsibility and Work Opportunity

Reconciliation Act of 1996 (Public Law 104-193), and

under President Bush’s faith-based initiative, govern-

ment agencies may contract directly with pervasively

sectarian organizations — churches, mosques, and

synagogues ( De Vita and Wilson 2001 ) — without

imposing restrictions on displays of religious symbols

or other religious activities. As Carlson-Th ies notes, the

new rules “are designed to erect a legal shield to protect

the religious identity and character of faith-based

organizations that choose to accept government funds

in order to provide welfare services to needy families”

(1999, 31). Although there are long-standing contrac-

tual and fi nancial relationships between governments

and faith-based organizations (e.g., Catholic Charities,

Lutheran Social Services, Salvation Army, and Jewish

Community Services), the Charitable Choice provision

allows religion and faith to play a much more explicit

role in the structure and content of social services.

Unfortunately, much of the policy debate about

Charitable Choice fails to distinguish among the major

types of faith-based organizations. In this essay, we

seek to clarify and enhance this debate by analyzing

the capacities of religious congregations to provide

human services, distinct from those of other faith-

based organizations. Th e latter have long had oppor-

tunities to engage in contracting relationships with

government, but congregations have not — or only if

they refrained from overtly sectarian displays or activi-

ties. Th erefore, the interests that congregations have in

contracting with governments to deliver human ser-

vices, as well as their capacities for managing those

relationships, are key to understanding the impact of

Charitable Choice on the human services delivery

system in the United States.

Faith-Based Institutions and Social Services Much of the rhetoric employed by proponents of

Charitable Choice is based on Olasky’s (1992) ac-

count of how the needs of the poor and destitute in

American were met prior to the New Deal and the rise

of the welfare state. Olasky asserts that these previous

modes of service delivery, which relied on the role of

faith, faith-based institutions, and a “Lady Benevolent”

Richard M . Clerkin North Carolina State University

Kirsten A . Grønbjerg Indiana University

Th e Capacities and Challenges of Faith-Based Human

Service Organizations

Richard M. Clerkin is an assistant

professor of public administration at North

Carolina State University. His current

research focuses on advocacy and

accountability in the nonprofi t sector and the

use of charitable gaming to explore the

paradoxes and tensions nonprofi ts face in

pursuing their missions.

E-mail: [email protected] .

Kirsten A. Grønbjerg holds the

Efroymson Chair in Philanthropy at the

Center on Philanthropy, and she is a

professor of public and environmental affairs

at Indiana University – Bloomington. She is

currently directing a major research project

titled “Indiana Nonprofi ts: Scope and

Community Dimensions.”

E-mail: [email protected] .

Essays on Spirituality and Public Service

116 Public Administration Review • January | February 2007

nonprofi t sector to provide social services, allowed both

care providers and benefi ciaries to develop a sense of

individual and community responsibility for correcting

social ills. He argues that a return to a more personalized

service delivery model would once again address these

problems. Others argue that “religion can improve

individual well-being and ameliorate specifi c social

problems” ( DiIulio 2002, 1 ) and that, compared to

government and secular service providers, faith-based

organizations can off er a more holistic approach to

meeting individuals’ needs by providing caring staff and

supportive networks ( Nathan and Wright 2003 ).

Indeed, Bromley (1998) claims that combining the

sacred and the secular may help reintegrate the

fragmented public and private spheres of our society.

Opponents of Charitable Choice have been equally

vocal, pointing to a variety of potential problems with

moving to a system of faith-based delivery of social

services. Some are concerned that proselytization will

become rampant. Th ey foresee government’s inability

to provide proper oversight because of the ambiguous

line that separates sectarian worship and proselytiza-

tion from allowable religious components of services

( Matsui and Chuman 2001 ). Others worry that reli-

gion’s prophetic voice will be marginalized because

congregations will become accountable to government

for their fi nances ( Matsui and Chuman 2001; Segal

1999 ). Still others argue that public policy and public

funds would support employment discrimination

(e.g., Brantner 2002 ). Finally, some worry about the

impact on existing service providers, both faith based

and secular ( Cnaan and Boddie 2002 ).

More empirically based research has begun to ques-

tion the uniqueness of faith-based delivery of services.

Th us, Smith and Sosin (2001) argue that faith-related

providers have succumbed to secularization, so that,

although faith plays a role in determining which ser-

vices are off ered, there is little diff erence between

faith-related and secular providers in how they deliver

services. Chambré (2001) reaches similar conclusions

in her study of faith-based AIDS organizations in

New York City. She argues that, although government

funding has contributed to the process of seculariza-

tion, it is not the full story; other key factors include

a lack of funding from religious organizations for reli-

gious activities, 1 as well as changes in client popula-

tions, organizational leadership, and stakeholders.

Other issues relate to the broader policy objectives in

which Charitable Choice is embedded. As Lynn

(2002) describes, and as the White House Offi ce of

Faith-Based Initiatives (2003) has acknowledged,

Charitable Choice is part of a strategy for devolving

social welfare services to the neighborhood level. As

Greenberg (2000) notes, Charitable Choice seeks to

emphasize faith and localism and expand the types

of providers involved in delivering services, not the

dollars spent.

We agree with Cnaan and Boddie (2002) that more

research is needed to assess the impact of Charitable

Choice on clients and on secular nonprofi ts that face

competition from new service providers. Th ere is an

equally pressing need for a better understanding of

faith-based service providers and how they compare to

secular service providers. Our analysis contributes to

the developing literature on these questions. Several

studies have shown that high proportions of religious

congregations provide some human services ( Chaves

1999; Hodgkinson and Weitzman 1993 ). However,

most services appear to be of limited scope and are

delivered primarily by volunteers (Chaves 2001),

suggesting that few congregations have the capacity

to deliver large volumes of services or manage

government contracts ( Greenberg 2000 ).

Indeed, the U.S. General Accounting Offi ce (2002 ;

since 2004, the Government Accountability Offi ce)

reports that some faith-based organizations lack the

capacity to enter into collaboration with government

agencies or to perform as expected because inadequate

information technologies, management structures,

and fi nancial systems limit their ability to manage

complex social services. Th e same study found that

faith-based providers lack the fi nancial capacity to

raise start-up funds for new programs and have lim-

ited access to staff and volunteers. Th ese fi ndings are

somewhat surprising, as supporters of Charitable

Choice have cited better access to volunteer and fi nan-

cial resources as key advantages of expanding the role

of congregations in delivering social services. Th ey are,

however, consistent with those of Trulear (2000) , who

found that faith-based organizations face challenges in

building capacity, connecting with funding sources,

and evaluating their programs. Th iemann, Herring,

and Perabo (2000) worry that faith-based organiza-

tions face theological, organizational, and fi nancial

limitations that may threaten their sacred mission.

In this study, we address four related policy questions

by comparing the management capacities and chal-

lenges of congregations, other faith-based nonprofi ts,

and secular nonprofi ts, with a primary focus on those

that provide human services. We seek to answer four

questions: (1) What role, if any, do congregations and

faith-based organizations currently play in the social

service delivery system? (2) Are congregations inter-

ested in changing their role in the social service deliv-

ery system? (3) Compared to faith-based and secular

nonprofi ts, do congregations have the capacity to

adapt to new roles in the social service delivery sys-

tem? Finally, (4) compared to faith-based and secular

nonprofi ts, do human service congregations have

similar service capacities and management challenges?

Data and Methodology To examine these questions, we drew on a large survey

of nonprofi ts in the state of Indiana. 2 In the winter

Capacities and Challenges of Faith-Based Human Service Organizations 117

and early spring of 2002, a survey was mailed to con-

gregations, charities, advocacy, and mutual-benefi t

nonprofi ts that were identifi ed through two strategies.

Th e fi rst strategy involved the development of a com-

prehensive master database of Indiana nonprofi ts,

which was created by merging nonprofi ts registered

with the Internal Revenue Service (IRS) under all

subsections of section 501(c) with a list of state non-

profi t incorporations, yellow page listings of congrega-

tions, and local listings in 11 communities across the

state ( Grønbjerg 2002 ). Th e second strategy involved

identifying nonprofi ts in the state through a hypernet-

work sampling approach in which a phone survey,

using a random-digit dialing, queried adult Indiana

residents (age 18 or older) about their personal affi lia-

tions with Indiana nonprofi ts as employees, volunteers,

participants in meetings or events, or as attendees at

religious services. Th e names and addresses of all

nonprofi ts identifi ed were included as part of the

overall sample.

Th e survey used a stratifi ed sampling design based on

the type of nonprofi t listing and location to allow for

comparisons among communities and to explore the

utility of alternative nonprofi t listings. Th e analysis

presented here is based on a weighted sample for the

state of Indiana. A total of 2,206 organizations re-

sponded to the survey, representing an overall re-

sponse rate of 29 percent — ranging from a high of 36

percent for the hypernetwork sample and a low of 27

percent for congregations listed in the yellow pages

(28 percent for IRS-listed nonprofi ts). For additional

analyses of response rates and sampling biases for the

two sampling strategies and the nonprofi t listings, see

Grønbjerg and Clerkin (2005) .

Religious Status We classifi ed the responding nonprofi ts by religious

status and by human service status. Th is approach is

distinctive in that we analyze congregations separately

from other faith-based organizations. To determine

religious status, we relied primarily on responses to a

survey question that asked whether the responding

organization was (1) a religious congregation, (2)

another type of faith-based organization, or (3) nei-

ther (we refer to this group as secular nonprofi ts ). 3 In

cases in which the organization failed to respond to

this question or we had some doubt about the accu-

racy of the response, we reviewed a description of the

organization’s mission and primary program activities

(two other survey questions), its name, and its Web

site (if available). Of the 2,206 responding nonprofi ts,

21 percent (500) congregations, 11 percent (219)

were other types of faith-based nonprofi ts, and the rest

(69 percent, or 1,489) were secular nonprofi ts. 4

Based on the distribution of congregations by major

American theological tradition (i.e., Catholic, mainline

Protestant, evangelical Protestant, etc.), our sam ple

of Indiana congregations appears to refl ect the

nationwide mix of traditions. Comparing the theo-

logical traditions of the Indiana congregations we

surveyed to those reported in the National Congrega-

tions Study (NCS) ( Chaves et al. 1999 ), we found

that in 2002, 7 percent of Indiana congregations were

Catholic (6 percent in the NCS), 60 percent were

from Evangelical Protestant traditions (60 percent in

the NCS), 31 percent were from mainline Protestant

traditions (27 percent in the NCS), and 2 percent

were from other Christian and non-Christian tradi-

tions (7 percent in the NCS). Th is similarity in distri-

bution of congregations by theological traditions

suggests that our fi ndings may be generalizable

beyond Indiana.

Human Service Status To determine human service status, we relied on a

question about whether the organization (1) was

currently involved in providing health or human

services, 5 (2) had plans to do so within the next two

years, (3) was interested in doing so but had no defi -

nite plans, or (4) had no interest in providing such

services. Only respondents that chose the fi rst option

were defi ned as human service organizations, as they

are most likely to be directly aff ected by the

Charitable Choice initiative. Th ose with defi nite plans

for or an interest in providing human services were

too few to warrant separate analysis. 6

More than half the congregations (55 percent) and

faith-based nonprofi ts (53 percent) reported they

currently provide health or human services, as did 34

percent of secular nonprofi ts (refl ecting the inclusion

of mutual-benefi t nonprofi ts in the latter group). In

this study, we compare human service congregations

(HS-Congregations), other human service faith-based

organizations (HS-FBOs), and human service secular

nonprofi ts (HS-Seculars) in terms of organizational

capacity. We also include comparisons between

HS-Congregations and all other congregations

(Other-Congregations), between HS-FBOs and all

other faith-based organizations (Other-FBOs), and

between HS-Seculars and all other secular nonprofi ts

(Other-Secular) in order to examine whether those

involved in human services diff er from their counter-

parts that are not involved in human services.

We begin our analysis by reviewing the extent to which

congregations and other faith-based nonprofi ts provide

human services and their awareness of and interest in the

Charitable Choice initiative. Next, we consider whether

HS-Congregations, HS-FBOs, and HS-Seculars diff er in

management capacity (size, management structure, and

reliance on volunteers and donors) and whether they

diff er from their counterparts that do not provide

human services. Finally, we examine whether HS-Con-

gregations, HS-FBOs, and HS-Seculars diff er in their

118 Public Administration Review • January | February 2007

patterns of community and policy engagement — that is,

awareness of community conditions, participation in

collaborations, experience of competition, awareness of

policy changes, and involvement in advocacy activities.

Involvement of Congregations and Other Faith-Based Organizations in Human Services Th e majority of congregations and other faith-based

nonprofi ts in our sample reported providing some

health or human services. Th ese fi ndings confi rm

prior research by Chaves (1999) and Hodgkinson and

Weitzman (1993) , who found that substantial propor-

tions of congregations and other faith-based organiza-

tions have some experience or expertise in the health

and human service fi elds and therefore meet one

precondition for active participation in the

Charitable Choice initiative.

To determine the extent of involvement in human

services, we examined several indicators of service

orientation. Virtually all HS-Congregations (99 per-

cent) said they target their programs and services to

both their own members and to the general public,

whereas only 68 percent of HS-FBOs and 54 percent

of HS-Seculars said that they do so. Th is suggests that

HS-Congregations are more likely to view their services

as extending beyond their immediate stakeholders. In

our sample, HS-Congregations were less likely to serve

their own members exclusively (less than 1 percent,

compared to 12 percent of HS-FBOs and 10 percent

of HS-Seculars), but they were also less likely to serve

only the general public (less than 1 percent) than HS-

FBOs (21 percent) and HS-Seculars (36 percent).

However, human services seem to be ancillary to the

primarily worship missions of HS-Congregations.

When asked to describe their three most important

programs or services, relatively few HS-Congregations

(34 percent) included human services; of these

services, youth development (24 percent), traditional

human services (13 percent), and food and nutrition

(3 percent) were the most prevalent types of services. 7

HS-FBOs reported a wider array of human service

programs among their top three programs: traditional

human services programs (17 percent), youth

development (8 percent), housing (8 percent), legal

and law-related services (6 percent), sports and

recreation (8 percent), food and nutrition (2 percent),

and employment programs (1 percent).

For those that included traditional human services

among their top three programs, there were no

statistically signifi cant diff erences between HS-

Congregations, HS-FBOs, and HS-Seculars in provid-

ing emergency assistance (36 percent on average),

personal services (26 percent on average), or child and

youth services (18 percent on average). However, HS-

Congregations were less likely (7 percent) to include

family services than HS-FBOs (39 percent) or

HS-Seculars (11 percent). Not surprisingly, the major-

ity (85 percent) of HS-Congregations included at least

one religion-based program 8 in their top three,

compared to only 27 percent of HS-FBOs.

Awareness of and Interest in Government Funding among Congregations and Faith-Based Organizations Only about one-third (34 percent – 35 percent) of

congregations, regardless of human service status,

were aware of “a national initiative to make it easier

for religious organizations to obtain government

funding.” (Th is survey question was worded to

capture respondents’ knowledge of the Charitable

Choice initiative without referring to it explicitly in

order to avoid biasing the responses.) Th is is a slightly

higher percentage than Other-FBOs (28 percent) but

substantially less than HS-FBOs (63 percent).

Moreover, the great majority of congregations,

regardless of whether they said that they provide

human services (68 percent), and of Other-FBOs

(77 percent) said that they do not intend to seek any

public funding, compared to only 36 percent of HS-

FBOs. It is no surprise, therefore, that very few

HS-Congregations (13 percent), Other-Congregations

(10 percent), and Other-FBOs (10 percent) had actually

received such funding or intended to seek it, com-

pared to almost half (46 percent) of HS-FBOs. Th e

rest did not know ( p < .001). More telling, awareness

of the initiative was not related to whether congrega-

tions intended to seek public funding for human

services. In all cases, two-thirds said that they do not

intend to seek public funding (65 percent of those

that are aware of the initiative and 67 percent of those

that were not aware of it). Th ere was no diff erence

between congregations that provide human services

and those that do not. Th us, simply alerting congrega-

tions to the availability of government funding does

not necessarily mean they will seek to obtain it. How-

ever, if all congregations currently providing human

services and those that expressed an interest in obtain-

ing government funding actually did so, the percent-

age of congregations with government funding would

increase more than fi vefold.

Counter to Chaves’ (1999) fi nding that mainline

Protestant and Catholic congregations express greater

interest in receiving government funding than evan-

gelical congregations, we found no diff erences among

the major American theological traditions. 9 Th is lack

of denominational diff erence may refl ect more

homogenous institutional pressures on Indiana con-

gregations with regard to involvement in human

services compared to Chaves’ national sample of

congregations. Alternatively, evangelical denomina-

tions may have become more involved in providing

human services since 1998 (when Chaves conducted

his study) and 2002 (when we completed ours). As

Capacities and Challenges of Faith-Based Human Service Organizations 119

Chaves notes, national evangelical leaders endorsed

the initiative, whereas those from mainline Protestant

denominations opposed it. 10

Management Capacity To assess the extent to which HS-Congregations have

similar levels of management capacity to deliver

complex human services as other human service

nonprofi ts, we compared these organizations’ size,

management structure, management challenges, and

reliance on volunteers and donations. We considered

whether HS-Congregations, HS-FBOs, and

HS-Seculars diff er from each other and from their

counterparts that do not provide such human services.

Size Our analysis of revenues and staff size is somewhat

tentative because of missing information on some key

dimensions. 11 Table 1 reports results that compare

fi ve indicators of age and size by involvement in

human services for each of the three religious status

categories. Organizations providing human services

tended to be older or larger than their counterparts

that do not provide such services. For example,

HS-Congregations were signifi cantly older (median

age of 75) than congregations that do not provide

such services (median age of 53). Th e diff erence in

total staff compensation (medians of $70,000 and

$30,000, respectively) for the two types of congrega-

tions was not statistically signifi cant, although it

follows the overall pattern.

However, when we looked only at those involved

in human services, on the whole we found that

HS-Congregations had fewer staff resources and

tended to be older than the two other types of human

service nonprofi ts. We also found that HS-FBOs

tended to fall between the smaller and older HS-

Congregations and the younger and larger HS-Seculars.

Management Structure Table 2 shows the presence of four types of manage-

ment structures: information technology (6 items),

staff and board policies (4 items), formal volunteer

management programs (2 items), and fi nancial con-

trols (5 items), as well as all items combined (17

items). For each religious status category, those in-

volved in human services had more management

structures in place across the board than their

counterparts. 12

When we compared management structures of human

service nonprofi ts across the three types of religious

status, we found that HS-Congregations generally

reported more management structures than

HS-Seculars, although the diff erences were not

signifi cant for staff and board policies or volunteer

management. HS-Congregations were signifi cantly

more developed in their fi nancial controls ( p < .01)

than either HS-FBOs or HS-Seculars.

Detailed analysis shows that HS-Congregations are

more likely to have certain types of information tech-

nologies than either HS-FBOs or HS-Seculars. More

HS-Congregations (91 percent) had computers avail-

able to key staff or volunteers, compared to 77 percent

of HS-FBOs and 71 percent of HS-Seculars ( p < .01).

Similarly, 86 percent of HS-Congregations had

Table 1 Median Age and Size of Indiana Nonprofi ts by Religious and Human Service Status

Religious and Human Service Status

Median Years Since Established

Median Total Revenues

Median Full-Time

Employees Median Part-Time

Employees Median

Compensation

Congregations with human services

75** (n = 256) $141,000** (n = 211) 1 (n = 224) 3 (n = 210) $70,000 (n = 204)

Congregations with no human services

53** (n = 200) $80,000** (n = 162) 1 (n = 148) 2 (n = 138) $30,000 (n = 157)

Other faith-based with human services

29 (n = 106) $150,000 (n = 92) 5 (n = 83) 4** (n = 72) $100,000** (n = 71)

Other faith-based with no human services

31 (n = 92) $5,625 (n = 65) 4 (n = 45) 2** (n = 44) $62,000** (n = 45)

Secular nonprofi ts with human services

34 (n = 515) $100,000 (n = 428) 9* (n = 282) 5 (n = 255) $139,000* (n = 274)

Secular nonprofi ts with no human services

26 (n = 775) $19,700 (n = 609) 2* (n = 198) 2 (n = 210) $58,000* (n = 213)

All respondents 35 (n = 1,944) $50,000 (n = 1,567) 3 (n = 980) 3 (n = 929) $60,300 (n = 964)

Note: **signifi cant at p < .05; *signifi cant at p < .10. Tests are based on regressing “human services” and “no human services” organizations for each religious status group (each comparison in the table represents a separate analysis) using robust standard errors and inspecting the F-statistic for the overall model. We use this method because the standard errors for an analysis of variance (ANOVA) would be biased and the statistical utility questionable because of the skewed distributions involved and because we would be unable to adjust for the design effect of the complex sampling strategy in STATA’s ANOVA procedure. For the same reason, we view these results as indicating potential differences. The quantities in parentheses are the number of respondents for that category.

120 Public Administration Review • January | February 2007

computerized fi nancial records, compared to less than

three-fourths (72 percent and 60 percent, respectively)

of HS-FBOs and HS-Seculars ( p < .001). However, we

found no diff erences between the proportions of HS-

Congregations, HS-FBOs, and HS-Seculars that had

computerized client, member, and program records, a

Web site, an e-mail address for their organization, or a

direct Internet connection for key staff and volunteers,

suggesting that HS-Congregations are more likely to

use information technology for internal management

tasks than for interfacing with the general public.

In terms of fi nancial management tools, HS-Congre-

gations were more likely to allocate reserves to capital

improvement (64 percent, p < .001) or maintenance

and equipment needs (73 percent, p < .001) than HS-

FBOs (35 percent for both) or HS-Seculars (34 per-

cent and 44 percent, respectively). Th e three types of

human service nonprofi ts were equally likely (71

percent on average) to have had a recent fi nancial

audit, a recent annual report (78 percent on average),

or a recent assessment of program outcomes and

impacts (36 percent on average). Th ese data do not

address how eff ectively these tools are used, nor did

we control for the diff erent regulatory environments

that congregations face compared to other faith-based

or secular nonprofi ts.

Management Challenges Th e respondents were asked to rate the challenge of

managing various aspects of service delivery, boundary

management, and technical capacities. In general,

though HS-Congregations appear to be at least as

well, if not better, equipped to deliver human services

(in terms of having key management structures in

place) compared to other human service nonprofi ts,

they seem to face more extensive management and

service delivery challenges, perhaps because they have

fewer staff resources or staff with professional training.

For some items, there were signifi cant diff erences

between congregations that provide human services

and those that do not. HS-Congregations were signifi -

cantly more likely to say that it is a challenge to main-

tain the visibility or reputation of their organization

compared to Other-Congregations (89 percent versus

81 percent, p < .01). Attracting new members or

clients appears to be equally challenging for HS-

Congregations (90 percent) and Other-Congregations

(88 percent). Th ough all congregations said that they

fi nd it diffi cult to attract new members, HS-

Congregations may fi nd it a challenge to integrate

their worship messages with their human service

messages to external audiences.

We found more diff erences in management challenges

when comparing human service organizations across

religious status. HS-Congregations reported facing

more challenges than HS-FBOs and HS-Seculars.

Almost all HS-Congregations (91 percent, p < .05)

said that strategic planning is at least a minor chal-

lenge, compared to three-fourths of HS-FBOs and

HS-Seculars. Similar patterns held for the challenges

of delivering high-quality services and programs (91

percent of HS-Congregations versus 74 percent of

HS-FBOs and 73 percent of HS-Seculars, p < .01),

attracting new members or clients (90 percent versus

73 percent and 80 percent, p < .10), and managing

facilities or space (72 percent versus 40 percent and

43 percent, p < .001). Th e latter may refl ect the fact

that congregations own facilities rather than rent space

and have more responsibilities for managing facilities.

Finally, the great majority of HS-Congregations (85

percent, p < .01) said that using information technol-

ogy eff ectively is at least a minor challenge (compared

to 70 percent of HS-FBOs and 63 percent of HS-

Seculars). As noted earlier, HS-FBOs and HS-Seculars

have fewer information technology components than

Table 2 Average Number of Management Structures for Indiana Nonprofi ts by Religious and Human Service Status

Religious and Human Service Status

Information Technology (6 items)

Staff/Board Policies (4 items)

Volunteer Programs (2 items)

Financial Controls (5 items) Total (17 items)

Congregations with human services

4.3 (n = 267) 2.7 (n = 267) 0.7 (n = 262) 3.2 (n = 267) 10.8 (n = 267)

Congregations with no human services

2.9 (n = 198) 2.3 (n = 197) 0.3 (n = 197) 2.6 (n = 197) 8.0 (n = 198)

Other faith-based with human services

4.1 (n = 114) 2.5 (n = 113) 0.7 (n = 114) 2.5 (n = 114) 9.7 (n = 114)

Other faith-based with no human services

3.0 (n = 91) 2.2 (n = 90) 0.2 (n = 89) 2.2 (n = 91) 7.6 (n = 91)

Secular nonprofi ts with human services

3.5 (n = 529) 2.6 (n = 526) 0.6 (n = 521) 2.6 (n = 524) 9.3 (n = 530)

Secular nonprofi ts with no human services

2.7 (n = 797) 1.8 (n = 801) 0.2 (n = 785) 2.0 (n = 800) 6.8 (n = 805)

All respondents 3.2 (n = 1,996) 2.2 (n = 1,994) 0.4 (n = 1,968) 2.4 (n = 1,993) 8.2 (n = 2,005)

Note: The numbers in parentheses are the number of respondents for that category. All difference of means, which compare those with human services to those without for a given religious status, are statistically signifi cant at p � .001.

Capacities and Challenges of Faith-Based Human Service Organizations 121

HS-Congregations, and this perhaps explains why

using information technology is less of a challenge

to them; indeed, 9 percent of HS-Seculars and 7

percent of HS-FBOs — compared to 5 percent of

HS-Congregations — said that managing information

technology does not apply to them. However, the

larger staff sizes of HS-FBOs and HS-Seculars

(compared to HS-Congregations, see table 1 ) may also

provide them with a broader array of staff skills to

use their more limited information technology

capabilities more eff ectively than HS-Congregations.

Reliance on Volunteers and Donations Supporters of Charitable Choice argue that congrega-

tions mobilize and rely on deeply engaged volunteers

rather than paid staff , thus delivering services more

effi ciently than other providers ( Grønbjerg and Sal-

amon 2002 ). Indeed, we found that HS-Congrega-

tions (and Other-Congregations) had a higher

reliance on volunteers. Virtually all HS-Congregations

(94 percent) reported that volunteers (other than

board members) are very important or essential

to the work of the organization, compared to less

than three-fourths of HS-FBOs and HS-Seculars (73

percent and 72 percent, respectively, p < .001). Th e

rest (6 percent) said that volunteers are important

(compared to 21 percent of HS-FBOs and 17

percent of HS-Seculars), and none said they are

not important, whereas 5 percent of HS-FBOs

and 11 percent HS-Seculars did say this.

Donations show a similar pattern, with the great

majority of HS-Congregations (89 percent) reporting

very high reliance on donations (75 percent or more

of revenues), compared to only 38 percent of HS-

FBOs and 19 percent of HS-Seculars ( p < .001).

At the other extreme, only 5 percent of HS-

Congregations reported that they receive no dona-

tions, compared to 20 percent of HS-FBOs and

32 percent of HS-Seculars. Th ough these are stark

contrasts, our data do not allow us to determine

how volunteers or donations are used — that is,

the extent to which they are devoted to worship

or human services. We suspect the former. 13

Engagement in Community and Policy Environments If congregations and other faith-based organizations

are to increase their delivery of human services, they

must be located in communities where services are

needed. Th ey must also be willing to collaborate with

other service providers (because they are unlikely to

provide all of the services that clients need) while also

managing the competition these relationships entail.

Th ey must be aware of key policy developments that

aff ect them or the people they wish to serve, and they

must be prepared to advocate issues that are impor-

tant to them, their clients, or their communities. We

turn now to an examination of these issues.

Changing Community Conditions

Our data indicate that HS-Congregations serve com-

munities that are fairly similar to those served by

other human service nonprofi ts; there were no signifi -

cant diff erences across the three religious profi les for

all but one of the conditions examined. With respect

to growing crime and violence in their communities,

HS-Congregations diff ered ( p < .10) from other

human service nonprofi ts; they were less likely (17

percent) to report growing crime than HS-Seculars

(26 percent) and especially HS-FBOs (40 percent).

Th is is one of the few indications that HS-

Congregations may serve somewhat less disadvantaged

communities than those served by other human

service nonprofi ts, especially HS-FBOs.

Targeting Disadvantaged Groups

If HS-Congregations and HS-FBOs are to play a

major role in providing social services, they must target

the disadvantaged groups who are in need of such

services. Our analysis found that HS-Congregations

were less likely than HS-FBOs or HS-Seculars to target

services based on income 14 (16 percent versus 41

percent and 30 percent) and racial or ethnic status

(13 percent versus 35 percent and 18 percent). HS-

Congregations were more likely to target by gender

(61 percent, most likely related to men’s and women’s

groups and bible study), youth (85 percent, perhaps

related to youth groups), and — not surprisingly —

people of a particular faith (66 percent) than

HS-FBOs and especially HS-Seculars. We conclude

that HS-Congregations are more likely to target on

the basis of programs related to worship than human

services.

Involvement in Collaboration Very few, if any, human service providers are able to

meet the complex needs of disadvantaged clients.

Delivering eff ective services with minimum disruption

to clients requires human service providers to

coordinate their services with other organizations.

Table 3 shows that almost three-fourths (74 percent)

of human service nonprofi ts participated in some

form of networking with other organizations.

However, the nature of the networking diff ered by

religious status (marginally statistically signifi cant at

p < .1). HS-Congregations were disproportionately

(42 percent) involved in informal networks

(e.g., cooperating, coordinating, and working together

with other organizations) compared to HS-FBOs

(31 percent) and HS-Seculars (30 percent). Th ey were

least involved (12 percent) in formal collaborations

(e.g., legal, fi scal, administrative, or programmatic

exchanges), compared to 15 percent of HS-FBOs

and 25 percent of HS-Seculars. HS-Congregations

appear less fi rmly linked to other human service

nonprofi ts. However, they were more likely to be

involved than Other-Congregrations (46 percent

versus 27 percent).

122 Public Administration Review • January | February 2007

When HS-Congregations were involved in networks,

these involved mainly other congregations (98 percent)

or faith-based organizations (87 percent), as was also

the case with HS-FBOs (89 percent and 64 percent,

respectively), compared to only about one-fi fth of

HS-Seculars (see Table 4). HS-Seculars were most

likely to be involved with government agencies

(50 percent) or mutual-benefi t nonprofi ts (41 percent),

about twice the rate of HS-Congregations, with

HS-FBOs falling between the two. 15

Moreover, involvement in these networks appears to

be less benefi cial to HS-Congregations than other

human service nonprofi ts, perhaps because their col-

laborations are more informal. Th ough some HS-

Congregations reported that these relationships make

it easier to secure key resources, the benefi ts accrue to

fewer HS-Congregations than other human services

providers. HS-Congregations were less likely (18

percent) to fi nd that collaboration makes it easier to

obtain funding than HS-FBOs and HS-Seculars (57

percent and 45 percent, respectively, p < .05), recruit

and retain board members (5 percent versus 34 per-

cent and 19 percent, respectively, p < .01), and en-

hance their visibility or reputation (58 percent versus

88 percent and 72 percent, respectively, p < .05). Very

few organizations of any profi le said these relation-

ships make access to resources more diffi cult.

Th ese fi ndings suggest that human service nonprofi ts,

including HS-Congregations, face institutional

pressures to engage in networking relationships. HS-

Congregations appear to receive fewer tangible

resources from networking than other types of human

service nonprofi ts, perhaps because they participate in

more informal structures. Without information on

the age, intensity, or scope of these networks, we

cannot know whether HS-Congregations are more

recently or just less intensively involved in network

relationships than other human service nonprofi ts.

Competition Although collaboration may bring benefi ts, greater

involvement with other organizations can also increase

competition. Whether related to obtaining fi nancial

resources, recruiting staff and volunteers, recruiting

board members, attracting clients, or delivering pro-

grams, we found competition among human service

nonprofi ts to be somewhat segmented by religious

status. HS-Congregations competed more frequently

with religious organizations than with other types of

organizations, most likely refl ecting their more intense

involvement with other religious organizations in

networks and collaborations. HS-Seculars generally

reported competing with other secular nonprofi ts and

businesses more than with other types of organiza-

tions. HS-FBOs fall between HS-Seculars and HS-

Congregations: More than 30 percent of those that

reported competing with any organization said they

compete with both religious and secular nonprofi ts in

all fi ve areas. Overall, these are relatively low rates of

competition. We do not know whether this refl ects

Table 4 Types of Organizations Involved in Collaboration/Network Relationships by Religious Status, Indiana Human Service Nonprofi ts

Types of Organizations Involved in Collaboration or Network Relationships HS-Congregations HS-FBOs HS-Seculars Total

Religious congregation (p < .001) 98% (n = 184) 89% (n = 85) 20% (n = 301) 54% (n = 570) Other religious organization (p < .001) 87% (n = 161) 64% (n = 75) 18% (n = 295) 45% (n = 531) Secular nonprofi t (not signifi cant) 64% (n = 135) 52% (n = 78) 64% (n = 322) 63% (n = 535) Nonprofi t advocacy organization (not signifi cant) 26% (n = 126) 44% (n = 70) 42% (n = 317) 39% (n = 513) Nonprofi t mutual benefi t organization (p < .05) 20% (n = 113) 27% (n = 67) 41% (n = 297) 34% (n = 477) Government agency (p < .05) 24% (n = 124) 37% (n = 72) 50% (n = 319) 42% (n = 515) Business/for-profi t organization (not signifi cant) 22% (n = 118) 49% (n = 70) 38% (n = 305) 36% (n = 493)

Note: The numbers in parentheses in the body of the table refer to the number of respondents for that question. Signifi cance levels for differences among types of human service organizations are reported in the far left column.

Table 3 Forms of Collaborations and Networks, by Religious Status, Indiana Human Service Nonprofi ts

Collaboration/Network Response Categories HS-Congregations HS-FBOs HS-Seculars Total

Yes, involved in one or more formal collaborations

12% 15% 25% 20%

Yes, involved in one or more informal networks

42% 31% 30% 34%

Yes, involved in formal collaborations and informal networks

18% 17% 23% 21%

No, not involved in collaborations or networks

27% 38% 22% 26%

Total 100% (n = 263) 100% (n = 114) 100% (n = 519) 100% (n = 896)

Note: Relationships are signifi cant at p < .1. The numbers in parentheses are the number of respondents for that category.

Capacities and Challenges of Faith-Based Human Service Organizations 123

niche specialization in particular markets (for clients,

funding, or staff ) or whether the responding organiza-

tions were not aware that other organizations operate

in these niches.

Changes in the Regulatory Environment Perhaps refl ecting their more limited reliance on pub-

lic funding, relatively few HS-Congregations (com-

pared to HS-FBOs and HS-Seculars) reported stricter

government regulations of contracts (8 percent versus

17 percent and 21 percent, respectively, p < .05) or

client eligibility requirements (7 percent versus 25

percent and 22 percent, respectively, p < .05). HS-

Congregations were also less aff ected than HS-FBOs

and HS-Seculars by personnel and legal requirements

(6 percent versus 24 percent and 25 percent, respec-

tively, p < .05). Although consistent with these pat-

terns, diff erences in reports of stricter workplace safety

requirements (23 percent versus 30 percent and 36

percent, respectively) and tighter professional licens-

ing requirements (12 percent versus 22 percent and

26 percent, respectively) were not signifi cant. Finally,

very few organizations of any type reported decreases

in their regulatory environments.

One of the aims of Charitable Choice is to reduce

regulatory constraints on congregations to induce

them to deliver human services. At least for the regu-

latory pressures examined here, the HS-Congregations

did not report a decrease in regulations, in part be-

cause so few accept government funds. However, they

were less likely to report an increase in these regula-

tions than HS-FBOs and HS-Seculars. Th e lower

levels of regulation faced by HS-Congregations in the

areas of contracts, client eligibility, and personnel and

legal requirements suggest that they deliver their

services more informally than do HS-FBOs or

HS-Seculars.

Activism and Advocacy Finally, we examined whether human service non-

profi ts diff er in the extent of their involvement in advo-

cacy activities and the types of policy issues or groups

for which they advocate depending on their religious

status. We found no diff erences in terms of whether

they were engaged in various types of advocacy activi-

ties: roughly 25 percent of HS-Congregations, HS-

FBOs, and HS-Seculars reported that they promote

positions on policy issues, 20 percent said that they

promote interests relevant to certain groups, and 3

percent reported promoting the interests of political

groups. However, although HS-Seculars were more

likely to advocate for health and youth issues, HS-

Congregations and HS-FBOs tended to advocate issues

traditionally associated with religious organizations:

the poor, youth, abortion, gambling, and the church –

state relationship (see Table 5). Other-Congregations

were involved in issues similar to those of HS-

Congregations but were marginally less active than

HS-Congregations in promoting the interests of cer-

tain groups (9 percent versus 18 percent, p < .10).

Conclusion In this study, we have sought to advance the current

policy discussion of Charitable Choice and to explore

the capacities of faith-based organizations to partici-

pate in the delivery of human services. To do so, we

distinguished between congregations and other types

of faith-based organizations because the Charitable

Choice Initiative sought explicitly to make it easier for

congregations to obtain access to government funding.

By comparing the interests and management capacities

of religious congregations to deliver social services to

those of faith-based and secular service providers, we

have off ered a more informed basis for the ongoing

debate about the impact of Charitable Choice and

President Bush’s faith-based initiative on the social

service delivery system in the United States.

Overall, a large proportion of congregations (more

than half ) reported that they already provide some

type of health or human services. Th ese are the con-

gregations that are likely to be most directly aff ected

by the Charitable Choice initiative. In general, these

congregations (HS-Congregations) appear to have as

many key management structures in place as other

human service nonprofi ts, and therefore they are

reasonably well equipped to deliver such services.

However, their organizational capacities seem primar-

ily geared toward running churches, not social ser-

vices, and may not be easily transferable. Moreover,

HS-Congregations provide a narrower range of

services, consider these services a lower priority, and

seem to encounter more extensive management

challenges.

HS-Congregations face similar community concerns

as other human service nonprofi ts and are involved in

collaborative relationships with other nonprofi ts.

However, these relationships tend to be less formal

and involve a narrower range of participants. Th ey

also seem to be less intense, as indicated by a relative

absence of competitive pressures. HS-Congregations

also appear to be less eff ected by the policy environ-

ment than other human service nonprofi ts, albeit

equally involved in advocacy activities. However, they

are more involved in all of these types of external

relations than are congregations not involved in hu-

man services. In short, there are notable diff erences

between congregations that deliver human services

and those that do not, as well as between the former

congregations and other nonprofi ts also involved in

human services.

Th ese fi ndings have important implications for cur-

rent policy initiatives to involve more congregations

and other faith-based organizations in the delivery

of government-funded human services. Ultimately,

124 Public Administration Review • January | February 2007

surprisingly few congregations are aware of these types

of national initiatives. We are particularly struck by

the even smaller percentages that have already

obtained public funding or are interested in pursuing

it. Hence, turning to religious congregations to provide

complex social services at the neighborhood level

presents policy makers with the twin challenges of

developing both an interest among congregations to

formalize their provision of health and human services

and helping them retool their organizational capacities

to manage these services. Both of these challenges

need to be balanced against a congregation’s need to

meet its primarily sacred mission.

Notes 1. However, she does not address whether government

funding has “crowded out” this funding stream.

2. We are grateful to the many Indiana nonprofi ts

that responded to the survey and to the Aspen

Institute’s Nonprofi t Sector Research Fund, the

Indianapolis Foundation, and Indiana University

Center on Philanthropy (with funding from the

Lilly Endowment, Inc.) for fi nancial support.

Additional support was provided by WBH

Evansville, Inc., Indiana University, Ball State

University, and the University of Southern Indiana.

Th e survey was managed by the Center for Survey

Research at Indiana University, with follow-up

work performed by Ange Cahoon, Amy Horst,

Hun Myoung Park, Allison Leeuw, Julie Schaefer

and Erin Nave. We benefi ted from survey analyses

carried out by Linda Allen, Patricia Borntrager,

and Curtis Child. Th e eff orts of all of these

people strengthened this work enormously; all

remaining problems refl ect our shortcomings.

3. Using these three broad categories of religious

status necessarily limits our ability to examine

how more fi nely nuanced levels of faith-relatedness

(e.g., Jeavons 1998; Monsma 1996; Sider and

Unruh 2004 ) relate to organizational characteristics

or management capacities, but we also avoid the

complexities of applying the dimensions

( Cameron 2004 ; Jeavons 2004; Netting 2004 ).

4. Th e percentages reported throughout the paper

are weighted population estimates; simply divid-

ing the n by the total number of respondents will

not yield the reported percentage.

5. To aid respondents, the question defi ned “health

services” as including health care and health

treatment, treatment for diseases and disorders,

health research and prevention, and mental health

and crisis intervention. “Human services” were

defi ned as including social services and counseling,

public safety and disaster relief, crime and legal

services, employment and job training, housing,

food and nutrition, youth development, and

recreation. Th ese key terms and defi nitions are

based on the National Taxonomy of Exempt

Entities, now used by the IRS to classify tax-

exempt entities. For information about the

taxonomy, see http://nccsdataweb.urban.org/

FAQ/index.php?category=73 .

6. Th us, only 3 percent of congregations and 5

percent of other faith-based nonprofi ts reported

that they plan to provide human services over the

next two years. A somewhat larger group of

congregations (14 percent), but only 4 percent of

other faith-based organizations, expressed some

interest in providing health or human services in

the future but had no specifi c plans to do so.

7. Th ese subcategories of human services are based

on the National Taxonomy of Exempt Entities

classifi cation system. See http://nccsdataweb.

urban.org/FAQ/index.php?category=73 .

8. Somewhat surprisingly, this is not 100 percent of

HS-Congregations. Indeed, 7 percent did not list

at least one religion-based program or activity,

and 8 percent did not provide us with any programs

or activities. More than half (54 percent) listed

only religion-based programs and activities.

Table 5 Percent Involved in Advocacy That Identify Particular Advocacy Issues, Indiana Human Service Nonprofi ts by Religious Status

Advocacy Issues Signifi cance HS-Congregations HS-FBOs HS-Seculars Total

Health care p < .01 (p < .05) 6% (2%) 7% (2%) 29% (9%) 19% (6%) Workforce issues p < .10 ns <1% (<1%) 0% (0%) 3% (1%) 2% (1%) Low-income groups p < .10 ns 16% (5%) 2% (<1%) 8% (2%) 9% (3%) Special groups p < .10 ns 14% (5%) 41% (11%) 18% (6%) 19% (6%) Public safety p < .10 ns 1% (<1%) 15% (4%) 11% (3%) 8% (3%) Pro-Bible issues p < .01 (p < .01) 15% (5%) 32% (9%) 0% (0%) 8% (3%) Pro-life p < .001 (p < .001) 37% (12%) 36% (10%) <1% (<1%) 16% (5%) Antigambling p < .01 (p < .05) 13% (5%) 2% (<1%) 0% (0%) 4% (1%) Death penalty p < .10 ns 6% (2%) 5% (1%) 0% (0%) 3% (1%) Church–state relations p < .01 (p < .05) <1% (<1%) 11% (3%) 0% (0%) 1% (<1%) Any issue/group ns (34%) (29%) (35%) (33%)

Note: We report two percentages for each religious/human services category: 1) the proportion of those that indicate they advocate for any issue and 2) in parentheses, the proportion of all in the particular religious/human services category. For the former, overall numbers are: HS-Congregations (n = 95), HS-FBOs (n = 40), HS-Seculars (n = 209), and HS-Total (n = 344). The numbers in parentheses are based on the total number of organizations in each profi le: HS-Congregations (n = 273), HS-FBOs (n = 117), HS-Seculars (n = 546), and HS-Total (n = 936). The statistical signifi cance for the proportion of the groups that advocate for at least one issue is on the left. The statistical signifi cance for the proportion of the entire group is reported in parentheses.

Capacities and Challenges of Faith-Based Human Service Organizations 125

9. Th is holds regardless of whether we use the

classifi cations of denominations used by Chaves

(1999) or those developed by Steensland et al.

(2000) , which diff er in the way they categorize white

protestant denominations. Chaves’ (1999) uses

the standard survey research liberal – conservative

distinction, whereas Steensland et al. (2000) base

their mainline – evangelical split on the historical

development of American religious traditions.

However, for some of the other relationships

examined in the balance of this article, we found

statistically signifi cant diff erences by religious

tradition. Unfortunately, because of space limita-

tions, we do not report those fi ndings here.

10. Th e two studies also used somewhat diff erent

questions for measuring involvement in human

services. We used fairly extensive prompts (see

footnote 5), whereas Chaves asked whether the

congregation provided social services, community

development, or community organizing. However,

given the similarity in distribution of theological

traditions between our Indiana sample and

Chaves’ national sample, we believe the diff erence

is more strongly infl uenced by diff erences in the

timing of the surveys than the wording of the

prompts.

11. Unfortunately, our data on employment and

fi nances suff er from fairly large nonresponse

problems. Even after contacting roughly 900 of

more than 2,200 respondents to collect missing

data, we were left with more missing information

than is ideal. Also, given the relatively large

standard deviations for these responses (because

of the presence of many small organizations and a

few very large ones), we report only median

values for age, revenues, employees, and

compensation.

12. Although these diff erences were highly signifi cant

( p < .001), the analyses explain only 6 percent of

the variance in information technology, 7 percent

in staff and board policies, 9 percent in volunteer

programs, 7 percent in fi nancial controls, and 11

percent in the total number of reported manage-

ment structures. Organizational size, measured by

total revenue, generally explains higher levels of

variance in the data: 45 percent for information

technology, 39 percent for staff and board policies,

6 percent for volunteer programs, 29 percent for

fi nancial controls, and 53 percent for total number

of reported management structures. (Th ere were

no statistically signifi cant diff erences in the mean

number of organizational components between

those that reported total revenues and those that

did not.)

13. Some support for this supposition can be gath-

ered from Grønbjerg and Never’s (2004) analysis

of volunteer activities in Indiana, where a greater

percentage of respondents indicated they assist

with religious services than deliver program

services. Th eir results rank-order activities as

fund-raising (28 percent), assisting with religious

services (14 percent), leading or managing

nonprofi t organizations (12 percent), and

delivering direct services (12 percent).

14. Open-ended responses about the income groups

targeted by programs show that fewer HS-

Congregations identifi ed a low-income group

(38 percent) than HS-FBOs (76 percent) or

HS-Seculars (62 percent) organizations. However,

these diff erences were not statistically signifi cant.

15. We have some concerns about the validity of

these responses, as fewer HS-Congregations

answered certain components of the questions

than other congregations.

References Brantner , Paula A . 2002 . Th e Employment

Discrimination Implications of Charitable

Choice Legislation . Employee Rights Quarterly 2 ( 4 ):

75 – 79 .

Bromley , David G . 1998 . Transformative Movements

and Quasi-Religious Corporations: Th e Case of

Amway . In Sacred Companies: Organizational

Aspects of Religion and Religious Aspects of

Organizations , edited by Neal J . Demerath III ,

349 – 63 . New York : Oxford University Press .

Cameron , Helen . 2004 . Typology of Religious

Characteristics of Social Service and Educational

Organizations and Programs — A European

Response . Nonprofi t and Voluntary Sector Quarterly

33 ( 1 ): 146 – 51 .

Carlson-Th ies , Stanley . 1999 . Faith-Based Institutions

Cooperating with Public Welfare: Th e Promise of

the Charitable Choice Provision . In Welfare Reform

and Faith-Based Organizations , edited by Derek

Davis and Barry Hankins , 29 – 60 . Waco, TX :

Baylor University, J.M. Dawson Institute of

Church – State Studies .

Chambré , Susan M . 2001 . Th e Changing Nature of

“Faith” in Faith-Based Organizations:

Secularization and Ecumenicism in Four AIDS

Organizations in New York City . Social Service

Review 75 ( 3 ): 435 – 55 .

Chaves , Mark . 1999 . Religious Congregations and

Welfare Reform: Who Will Take Advantage of

“Charitable Choice”? American Sociological Review

64 ( 6 ): 836 – 46 .

— — — . 2001 . Religious Congregations and Welfare

Reform . Society 38 ( 2 ): 21 – 28 .

Chaves , Mark , Mary Ellen Konieczny , Kraig

Beyerlein , and Emily Barman . 1999 . Th e National

Congregations Study: Background, Methods, and

Selected Results . Journal for the Scientifi c Study of

Religion 38 ( 4 ): 458 – 76 .

Cnaan , Ram A. , and Stephanie C . Boddie . 2002 .

Charitable Choice and Faith-Based Welfare: A Call

for Social Work . Social Work 47 ( 3 ): 224 – 35 .

126 Public Administration Review • January | February 2007

De Vita , Carol J . , and Sarah Wilson . 2001 . Faith-

Based Initiatives: Sacred Deeds and Secular

Dollars . Washington, DC : Urban Institute Press .

DiIulio , John J . 2002 . Th ree Faith Factors . Th e Public

Interest , Fall . www.brook.edu/views/articles/diiulio/

pi_fall2002.htm [accessed October 15, 2006] .

Greenberg , Anna . 2000 . Doing Whose Work? Faith-

Based Organizations and Government

Partnerships . In Who Will Provide? Th e Changing

Role of Religion in American Social Welfare , edited

by Mary Jo Bane , Brent Coffi n , and Ronald

Th iemann , 178 – 97 . Boulder, CO : Westview Press .

Grønbjerg , Kirsten A . 2002 . Evaluating Nonprofi t

Databases . American Behavioral Scientist 45 ( 11) :

1741 – 77 .

Grønbjerg , Kirsten A. , and Richard M . Clerkin . 2005 .

Examining the Landscape of Indiana’s Nonprofi t

Sector: Does What You Know Depend on Where

You Look? Nonprofi t and Voluntary Sector Quarterly

34 ( 2 ): 232 – 59 .

Grønbjerg , Kirsten A. , and Brent Never . 2004 . Th e

Role of Religious Networks and Other Factors in

Types of Volunteer Work . Nonprofi t Management

and Leadership 14 ( 3 ): 263 – 90 .

Grønbjerg , Kirsten A. and Lester M . Salamon . 2002 .

Devolution, Privatization and the Changing Shape

of Government-Nonprofi t Relations . In Th e State

of the Nonprofi t Sector , edited by Lester M .

Salamon , 447 – 70 . Washington, DC : Brookings

Institution Press .

Hodgkinson , Virginia A. , and Murray S . Weitzman .

1993 . From Belief to Commitment: Th e Community

Service Activities and Finances of Religious

Congregations in the United States: 1993 Edition .

Washington, DC : Independent Sector .

Jeavons , Th omas . 1998 . Identifying Characteristics of

Religious Organizations: An Exploratory Proposal .

In Sacred Companies: Organizational Aspects of

Religion and Religious Aspects of Organizations ,

edited by Neal J . Demerath III , 79 – 96 . New York :

Oxford University Press .

— — — . 2004 . Religious and Faith-Based

Organizations: Do We Know One When We See

One? Nonprofi t and Voluntary Sector Quarterly

33 ( 1 ): 140 – 45 .

Kennedy , Sheila Suess . 2001 . Constitutional

Competence of Nonprofi t Managers . Paper

presented at the Association for Research on

Nonprofi t Organizations and Voluntary Action

Conference, November 28 – December 1, Miami, FL .

Kennedy , Sheila Suess , and Wolfgang Bielefeld . 2002 .

Government Shekels without Government

Shackles? Th e Administrative Challenges of

Charitable Choice . Public Administration Review

62 ( 1 ): 4 – 11 .

Lynn , Laurence E. , Jr . 2002 . Social Services and the

State: Th e Public Appropriation of Private Charity .

Social Service Review 76 : 58 – 82 .

Matsui , Elena , and Joseph Chuman . 2001 . Th e Case

against Charitable Choice . Th e Humanist

61 ( 1 ): 31 – 33 .

Monsma , Stephen V . 1996 . When Sacred and Secular

Mix: Religious Nonprofi t Organizations and Public

Money . Lanham, MD : Rowman & Littlefi eld .

Nathan , Richard P. , and David J . Wright . 2003 . Is

Charitable Choice Compatible with the First

Amendment? Is It a Good Idea? Does It Work?

www.religionandsocialpolicy.org/publications/

publication.cfm?id=42 [accessed October 15,

2006] .

Netting , F. Ellen . 2004 . Commentary on Typology of

Religious Characteristics of Social Service and

Educational Organizations and Programs .

Nonprofi t and Voluntary Sector Quarterly 33 ( 1 ):

135 – 39 .

Olasky , Marvin . 1992 . Th e Tragedy of American

Compassion . Washington, DC : Regnery Gateway .

Segal , Julie A . 1999 . A “Holy Mistaken Zeal”: Th e

Legislative History and Future of Charitable

Choice . In Welfare Reform and Faith-Based

Organizations , edited by Derek Davis and Barry

Hankins , 9 – 28 . Waco, TX : Baylor University, J.M.

Dawson Institute of Church – State Studies .

Sider , Ronald J. , and Heidi R . Unruh . 2004 . Typology

of Religious Characteristics of Social Service and

Educational Organizations and Programs .

Nonprofi t and Voluntary Sector Quarterly 33 ( 1 ):

109 – 34 .

Smith , Steven Rathgeb , and Michael R Sosin . 2001 .

Th e Varieties of Faith-Related Agencies . Public

Administration Review 61 ( 6 ): 651 – 70 .

Steensland , Brian , Jerry Z . Park , Mark D. Regnerus ,

Lynn D. Robinson , W . Bradford Wilcox , and

Robert D . Woodberry . 2000 . Th e Measure of

American Religion: Toward Improving the State of

the Art . Social Forces 79 ( 1 ): 291 – 318 .

Th iemann , Ronald , Samuel Herring , and Betsy

Perabo . 2000 . Risks and Responsibilities for Faith-

Based Organizations . In Who Will Provide? Th e

Changing Role of Religion in American Social

Welfare , edited by Mary Jo Bane , Brent Coffi n , and

Ronald Th iemann , 51 – 70 . Boulder, CO : Westview

Press .

Trulear , Harold D . 2000 . Faith-Based Institutions and

High-Risk Youth . Philadelphia : Public/Private

Ventures .

U.S. General Accounting Offi ce (GAO) . 2002 .

Charitable Choice: Overview of Research Findings on

Implementation . Washington, DC : Government

Printing Offi ce . GAO-02-337 . www.gao.gov/

new.items/d02337.pdf [accessed October 15,

2006] .

White House Offi ce of Faith-Based and Community

Initiatives . 2003 . Rallying the Armies of

Compassion . www.whitehouse.gov/news/reports/

faithbased.html [accessed October 15, 2006] .