A key question in debates about government funding for faith-based programs is whether funding these programs improves or worsens the quality of social services available to people in need. Proponents of government funding for faith-based programs argue that such programs are more effective than their secular counterparts and offer evidence of lives that have been changed. Opponents raise concerns about lower professional standards, unequal access to services, and violations of the civil rights of those receiving services. Attempts to address questions about the consequences of funding faith-based programs with empirical evidence suffer from the absence of careful comparative studies. While many acknowledge the need for more work in this area, the challenges of conducting this kind of research loom large. In Faith, Hope, and Jobs, Stephen V. Monsma and J. Christopher Soper take on this daunting task.

Monsma and Soper begin the book with a good overview of the methodological challenges of measuring both the effectiveness of social programs in general and job training programs in particular. They develop measures of effectiveness that include client perceptions of the programs, enabling outcomes (measures of attitudes or skills that help clients find and keep jobs), and intermediate outcomes (part-time or low-wage job placements), as well as the ultimate outcomes of high-wage jobs and moving from welfare to work.

In addition to tracking multiple outcomes, the study assesses five types of welfare-to-work programs. A survey of all welfare-to-work programs in Los Angeles categorizes programs into five types: government, for-profit, nonprofit/secular, faith-based segmented, and faith-based integrated. Thus, the study not only compares faith-based programs to secular programs, but also compares faith-based programs that explicitly incorporate religious values into job training programs (faith-based/integrated) to faith-based programs that do not do so (faith-based/segmented). The population survey of nearly all welfare-to-work programs allows broad analysis and careful classification of the basic characteristics of these five types of programs. Monsma and Soper identify seventeen representative programs for closer study. Research on clients in these seventeen programs provides rich longitudinal data that allows analysis of client perceptions of programs, comparisons of client traits before and after completing the programs, and tracking of graduates' employment six and twelve months out of the programs.

The broad range of measures and the inclusion of different types of welfare-to-work programs allow Monsma and Soper to uncover relative strengths of different kinds of programs. For example, both types of faith-based programs fare well on measures of soft skill outcomes, while government and for-profit programs perform better on measures of hard skills outcomes. The rich client data also allows Monsma and Soper to test whether religiously active clients fare better in faith-based programs than in secular ones (no) and whether clients who are not religious feel uncomfortable in faith-based programs (most do not) or appreciate the religious aspects of faith-based programs (most do).

Faith-based programs appear to perform at least as well as government programs on most ultimate outcome measures with fewer resources. However, the analysis does not include measures of cost per client or cost per outcome. One of the most striking findings is the relative strength of the for-profit programs in reaching intermediate and ultimate outcomes. The analysis includes controls for client differences, which reduces the likelihood that the strong performance of for-profit programs (and the performance of faith-based groups on par with government programs) results from client selection patterns. However, the descriptions of the programs reveal that the government programs are much larger and are more likely to have mandatory clients, two factors that the authors do not consider in their analysis.

Monsma and Soper are quite upfront about the limitations of the study. They note that the complexity of analyzing five types of programs designed for a diverse array of people with data from about three hundred clients spread across seventeen programs yields results that often rest on very small numbers of cases. However, the exploratory patterns reveal the benefits of distinguishing between the various types of programs, including the value of distinguishing between faith-based/integrated and faith-based/segmented programs. Second, the authors acknowledge that much of their book provides an “avalanche of specific findings” that can easily overwhelm the reader (164). The study does not employ an overarching theory that pulls the findings together. As a result, it is easy to get lost in the details of the results. The last chapter helps to address this shortcoming by pulling together key findings and addressing broader applications for future research and policy design. This discussion includes specific policy ideas that build on the relative strengths of the different types of welfare-to-work programs.

Faith, Hope, and Jobs provides valuable resources for studies of the effectiveness of faith-based programs. It also provides useful advice for policy-makers concerning how to improve the effectiveness of welfare-to-work programs. Scholars of public policy and scholars of religion and politics will both benefit from the ground-breaking research presented in the book.