Community is widely acknowledged as a fundamental aspect of social work practice, and this formulation distinguishes social work from other professions. Because of this long-standing tradition, social work needs to make a greater investment in producing scientific to enable community change and to incorporate community context into practice. This type of research faces numerous challenges related to the complexity of community interventions, the lack of well-developed community measures, the difficulty of implementing experimental designs, and the spatial dynamics of communities. However, recent substantive and methodological developments should make it possible to advance research on the community as agent, target, and context for social work practice. KEY WORDS: community; geographic information systems; neighborhoods; research design ********** Community is a ubiquitous concept in social work practice, but as such, it has presented formidable challenges for researchers. Social work must come to terms with the methodological difficulties because community is an essential element of social work and is part of what distinguishes it from other professions. New research techniques and tools have produced important advances in community studies in other fields, but these developments have not sufficiently influenced social work practice research. This article identifies challenges to building an evidence base for community interventions and describes new research tools that may advance in the field. THE PLACE OF COMMUNITY Rosen assiduously reminded the social work profession of the need to build a particular type of for practice, which he referred to as knowledge for (Rosen, Proctor, & Staudt, 1999). Such rests on evidence about how change occurs, whether change produces desired outcomes and, most important, how beneficial change can be deliberately produced. This article draws on that framework and focuses on research that builds social work for control with respect to community change. The discussion emphasizes communities that are place-based, such as urban neighborhoods, suburban subdivisions, villages, and towns. These localized communities have geographic, psychological, and social meaning and referents, and they have consequences for the people who live and work there (Chaskin, 1997). Although these are not the only communities of importance to social work, they are often the context for community practice (Weil, 1996). Communities based on ethnicity or common interests often transcend local or national boundaries and present research dilemmas that are different from the place-based communities discussed in this article. Researchers have identified several features that may be essential for healthy, strong, sustainable, or effective communities, such as social capital, including bridging and bonding capital (Putnam, 2000; Temkin & Rohe, 1998); community capacity, including leadership, governance, and organization (Chaskin, Brown, Venkatesh, & Vidal, 2001); collective efficacy (Sampson, Raudenbush, & Earls, 1997); sense of community (Brodsky, O'Campo, & Aronson, 1999); social control (Sampson & Groves, 1989); neighboring; and amenities and resources (Fuerstenberg, Cook, Eccles, Elder, & Sameroff, 1999); institutional strength, interorganizational relations, civic participation (Daley & Marsiglia, 2000); and safety (Buka, Stichick, Birdthistle, & Earls, 2001). Practice research needs to demonstrate how interventions with communities can produce changes in these attributes and processes. Knowledge for control requires evidence regarding the mediating processes that enable change so that the methods can be specified and replicated. With respect to social work practice in communities, the question of building for control is complicated by the fact that a community may be the agent of change, the target of change, or the context for change. …