Research concerned with the implementation of information technology (IT) in organizations can be divided, roughly, into two streams: factor, or variance studies; and process studies (Markus and Robey, 1988). The great majority of work has adopted a variance approach where several factors that are likelyto be associated with successful IT implementation are identified, made operational, and then tested, usually using a cross-sectional design, with statistical methods. In reviewing these studies, Lucas (1981) notes that although some 150 factors have beenidentified, only a relatively few, limited to "top management support" and "user involvement", are consistently associated with successful implementation. In addition, researchers tend to create new factor models rather than extending and confirming the most promising existing models, and no integrated model has emerged that explains a reasonable portion of the variance in implementation outcomes (although Lucas, Ginzberg and Schultz, 1990, made a valiant attempt in this regard). Process studies, on the other hand, seek to understand the process by which IT is implemented in organizations, using interpretive techniques based on interview, observational, and collected data. Although there are relatively few process studies, they are particularly appropriatefor theory building (Glaser and Strauss, 1967). Markus and Robey (1988) have pointed to the need for more process studies of technology implementation. In this paper we describe an ongoing process study of IT implementation in five settlement houses in New York City, using an action research approach (Argyris et. al., 1985). Settlement houses are the primary way that social services are delivered to community members of inner cities. From a research perspective, the IT implementation in the settlement houses is important in several respects. First, while IT implementations in profit-seeking firms have been widely researched, relatively few studies have been conducted in not-for-profit businesses. Not-for-profit firms are likely to differ from their profit-seeking counterparts in terms of their organizational values, goals, reward and control structures of individuals, organizational processes, staffing, environmental influences, and acquisition of resources. Second, few existing studies address the dynamicsinvolved in implementing IT in a group of cooperating, autonomous organizations. Our implementation study involves a confederation of five settlement houses and United Neighborhood Houses of NYC (UNH), an organization which provides technical assistance to the houses. This confederation is analogous in structure to IT partnerships and alliances, which have become popular among businesses in the for-profit sector