IN RECENT YEARS there has been an explosion in empirical research on faith-based social programs. Most studies, including the most scientifically rigorous, find that faith moves social and civic mountains. Last year, researchers at the University of Pennsylvania released a report identifying over 500 scientifically sound studies in which the “faith factor” was associated with results ranging from reductions in hypertension, depression, and suicide to lower rates of drug abuse, educational failure, nonmarital teenage childbearing, and criminal behavior.
Consider the latest scientific literature on religion and crime. With all due qualifications and caveats, some 50 empirical studies report that religious influences and institutions reduce violence and delinquency. Consider, for example, the work of my Penn colleague Byron Johnson, director of the Center for Research on Religion and Urban Civil Society (CRRUCS). In a 1997 Justice Quarterly study, he reported that New York state prisoners who participated intensively in Bible studies administered by Prison Fellowship ministries (on whose board I once served) were only a third as likely to be arrested a year after release as otherwise comparable prisoners who did not participate. In a forthcoming follow-up study, he finds that, on average, eight years after release, the Bible studies participants remained arrest-free over 50 percent longer than the parolees in the comparison group.
Likewise, in a just-released six-year study of a faith-based program in a Texas prison run by Prison Fellowship, Johnson reports that, two years after release, inmates who completed the 22-month program (16 months in prison plus 6 months of post-release care) were less likely to be rearrested than inmates paroled early from the program and than otherwise comparable inmates who did not participate in the program.
Indeed, only 8 percent of the Prison Fellowship program graduates, versus 20 percent of the matched comparison group, were incarcerated within two years after being released. Equally striking were the prisoners’ own words concerning how the program effected rehabilitation by proclaiming God’s love, fostering spiritual growth, and stressing the need to give back to society. Typical was the comment by one prisoner that he had “learned what life is about since being here . . . life is about helping others to grow like I’m growing. . . . I saw that other people loved me. Then I wanted to do the same.”
As Johnson’s report stressed, more research is sorely needed. And yet scores of studies now find that religious influences and institutions are especially beneficial in the lives of America’s most truly disadvantaged children and youth. These studies tell of schooling success among low-income Latino youth; improved employment prospects for low-income African-American youth; rapid gains in reading ability among urban elementary school students; cost-effective delinquency prevention services; the mobilization of year-round mentors for prisoners’ children; and much else as well.
For instance, a 4-year, 16-city study released in 2002 and 2003 by Public/Private Ventures, a nonprofit research organization in Philadelphia on whose board I serve, documents that faith-based organizations promote effective delivery of employment, education, and other justice system services to high-risk youth and adjudicated young adults. Another recent Public/Private Ventures study found that when churches partnered with public schools to provide quality after-school reading programs to some 900 children who were reading two or more years below grade level, the improvements were dramatic: After about 100 days, younger children vaulted 1.9 years in reading ability, and high school pupils gained a year.
So, whether with respect to reducing recidivism rates, improving public health outcomes, accelerating volunteer mobilization, or other objective measures, the empirical evidence has become weighty enough for numerous top scientific organizations to begin taking religion seriously. For example, in April, a landmark three-day conference held at the National Institutes of Health explored how to integrate the growing body of research on health and spirituality into the delivery of clinical care and social services.
Last month, however, national media prominently reported an eight-page study, based on data from two counties in Indiana, comparing small faith-based job training programs to secular ones. The study, co-published by Indiana University and Purdue University, found no differences between the programs with respect to job placement rates and starting salaries, but did find some evidence that clients in the faith-based program worked fewer hours per week and were less likely to receive health insurance than their secular program counterparts. The study’s quite able coauthors duly noted its highly preliminary character and multiple data limitations. For example, the analysis did not account for potentially salient differences in the program participants’ employment histories and other key background characteristics. It was a good little study, but page-one news?
It might help future reporting and civic discourse on the subject if everyone remembered what might be termed the “three rules of three”–there are three faith factors, three types of faith initiatives, and three different levels of scientific analysis relevant to the subject and the policy debates now surrounding it.
One faith factor is organic religion: The factor is present when a person believes in God, goes to church regularly, and manifests other cognitive and behavioral religious commitments. A second faith factor is programmatic religion: The person participates in a service delivery program led and staffed primarily by religiously motivated volunteers. A third faith factor is ecological religion: The person lives, works, shops, or recreates in a neighborhood with higher-than-average concentrations of community-serving religious congregations or faith-based social service delivery programs. To date, the research suggests that each faith factor makes a positive difference. But, other things being equal, does, say, a three-factor low-income inner-city youth do better in school or later in life than an otherwise comparable two-factor or single-factor youth, and, if so, under what conditions? Researchers have hardly begun to address such questions.
One type of faith initiative is faith-saturated: The program delivers social services in ways that are inextricably linked to worship services and fostering a lifelong religious commitment to a particular faith. Such programs rarely seek public funds and rarely work across denominational lines or partner with secular or public organizations. A second type is faith-motivated: The program is led by religious volunteers but involves no proselytizing, accepts persons without regard to religious commitments, and frequently works via ecumenical, inter-faith, religious-secular, and public-private partnerships. A third type is faith-mobilized: The program rallies and organizes religious volunteers to strengthen good works led by secular organizations or public agencies. Although the public debate remains fixated on programs that proselytize, the vast majority of community-serving religious programs are faith-motivated or faith-mobilized, not faith-saturated.
Major national studies by Penn’s Ram Cnaan and others show conclusively that most faith-motivated and faith-mobilized programs led by urban churches, synagogues, and mosques mainly serve poor children and youth who are not members of the congregations, and serve all people without regard to whether they are, or might become, co-religionists. Some faith-saturated programs are indivisibly conversion-centered or can make no meaningful distinction between delivering social services and leading worship services. But others, including Prison Fellowship, accept participants whatever their religious affiliations or lack thereof.
Under existing laws, any such organization is free to apply to administer certain public grants or to accept adult individuals with state-funded vouchers, provided that it agrees to use any public funds strictly for non-sectarian program components (transportation, nonreligious supplies, etc.), and also provided that secular alternatives are available to clients who might wish to opt out. The website of the White House Office of Faith-Based and Community Initiatives offers guidance to faith-based organizations on whether they are eligible to apply.
Finally, in the third “three” worth remembering, one type of research is experimental: Participants are randomly assigned to treatment and control groups and compared over time. A second type is quasi-experimental: Persons in the program are compared via various statistical techniques to persons with similar characteristics. A third type is ethnographic: Persons in the program are observed and their experiences described through interviews and case studies. To date, the empirical literature on faith initiatives features few experimental studies, a growing body of first-rate quasi-experimental studies, plus an older body of ethnographic studies.
In the 1960s, the government began pouring billions of dollars into social welfare programs administered by strictly secular and nominally religious national nonprofit organizations. There was no scientific basis for steering clear of faith-based programs then, and there is none now. As documented in “Unlevel Playing Field,” a report released in 2001 by the White House, despite decades of public funding and federal laws explicitly requiring program performance measurement, few tax-financed social programs have ever been experimentally studied or otherwise independently evaluated, and almost all those that have been so studied have been found sorely wanting.
In stark contrast, since the first federal law prohibiting discrimination against community-serving religious organizations in the government grant-making process was passed in 1996, much objective research has documented the extent and efficacy of faith initiatives. As suggested in “Better Together,” a report released in 2002 by Harvard’s John F. Kennedy School of Government, “whether measured by association memberships, philanthropy, or volunteering,” community-serving religious groups “build and sustain social capital–and social capital of more varied forms–than any other type of institution in America,” and at least some “faith-based programs can enjoy success where secular programs have failed.” Amen.
John DiIulio, coauthor with James Q. Wilson of “American Government: Institutions and Policies,” ninth edition (Houghton-Mifflin, 2003), served in 2001 as director of the White House Office of Faith-Based and Community Initiatives.
