Case studies of organizational innovation have long-described how individual-level responses to an innovation interact with social and work processes in mutually dependent, and sometimes surprising, ways (Barley, 1986; Orlikowski, 1996, Repenning & Sterman, 2002).  Indeed, recent research in public administration has underscored the importance of more precisely understanding the connection between managerial factors, innovation characteristics, and the organizational-level adoption of innovation (Walker, 2007; Damanpour & Schneider, 2008).   In this study, we attempt to build upon this current understanding of how such social, technical, and environmental factors either support or impede the implementation of organizational innovation through a combination of computational, agent-based modeling and fieldwork at a large urban high school implementing a major reform.  

Research Approach and Data

Agent-based modeling is a computational method particularly well suited to studying the emergence of macro-behavior from micro-level actions. As a starting point for the micro-level constructs of our computational model, we draw upon the literatures in organizational learning (Levinthal & March, 1981; Levitt & March, 1988) and network sociology (Coleman, 1988, Freeman, 1992; Burt, 1992, 2004), as they offer well-specified and promising building blocks for work attempting to gain insight into the intra-organizational mechanisms and interdependencies related to implementing innovation.  Importantly, rather than embark on the ambitious undertaking of integrating these two streams of work in an abstract manner applicable to all types of organizations, we instead take a more modest step:  We attempt to work out the integration of their constructs in the context of a particular problem -- the difficulty of implementing innovation in public schools.  This involves initializing a base case for the model with qualitative and quantitative data collected over the period of two years at a large urban high school engaged in implementing a major “small schools” reform.  Data include participant observations, semi-structured teacher interviews, and survey data on teachers and their social networks.  To the extent that the features of our context allow, we then use the simulation to explore more general statements about the factors that influence an organization's capacity to change.  

Findings

A central finding is that incremental improvement in the regular routines of an organization can eliminate the work activities needed to support the early steps of implementing more radical innovations.  Moreover, we put forth a set relationships that describe how an organization's social network structure, the strength of social influence within subgroups, and environmental turbulence work together to either prevent or enable the implementation of innovation, and translate insights about those relationships into specific propositions about how organizations differ with respect to their capacity to change.