Research Summary: Predicting Community Adoption of Collective Impact in the United States

By Brett Mayfield

 The collective impact model is increasingly common in the nonprofit sphere. Governments, nonprofits, and businesses partner to best handle resources and increase their social impact as they work toward common goals. At NNSI, our research focuses on these collaborative governance models, like collective impact in the nonprofit sector. This research primarily explores why particular interorganizational networks surface in a community, which consists of individual people, organizations, and partnerships. One reason is to give back––when college graduates return to their impoverished or struggling hometowns, they may want to make a difference. 

In a recent study, researchers Anne-Marie Boyer, Katherine R. Cooper, Shaun M. Dougherty, Rong Wang, and Michelle Shumate examined the effects of exposure to collaborative governance models and community resources on the emergence of this governance. Using a web-based national-level scan of over 1000 communities in the United States, our research team found many factors that predicted the presence of collective impact in a U.S. county:

  • Geographic proximity
  • Poverty rate
  • Number of individuals with a Bachelor’s Degree
  • Number of relevant nonprofits
  • Assets held by these organizations
  • The severity of the social problem

This research is significant not only to NNSI but also to nonprofit leaders seeking to establish a collective impact network in their communities. The study evaluated under-inspected, community-level factors correlated with collaborative governance models and the geographic influence on the models’ adoption. Our results are essential for understanding which communities support collective impact networks and how we can improve those that do not. 

To gather data, our research team searched for education collective impact networks through a web-scan. Next, we verified the list by conducting phone interviews with a stratified random sample of network leaders. Then, we used the census and the American Community Survey to describe those relevant U.S. counties. We used data from the Urban Institute’s collection of IRS Form 990 documents to understand the educational nonprofit landscape. We also collected education outcome data on US counties from the Stanford Education Data Archive. We compared communities that used the collective impact model and communities that did not. Unexpectedly, the severity of the education-based social problem, the number of education-related nonprofits, and assets held by these organizations were not associated with the adoption of collective impact proposals.

Our research concluded that the collective impact model spreads across geographically adjacent communities. If a community adopted a collaborative governance model, then the probability of a neighboring community also adopting that model increased. Previously, researchers argued that nonprofit partnerships emerge because leaders specifically reach out to accomplish a common goal.¹ Our results challenge this argument. We suggest that collective impact likely emerges because of community exposure to those models. These findings can help us better understand how collaborative governance can expand through outreach and promotion efforts.

The study also found that communities with a higher percentage of the population with a bachelor’s degree were more likely to adopt collective impact initiatives. Communities with higher poverty rates followed the same trend. We suggest that there may be greater awareness of collaborative models among community members with resources. Maybe, individuals with more resources (such as a degree) can see more ways to collaborate for social change. Of course, this poses a challenge because communities that need collaborative models probably lack the resources to start them.

Overall, this study emphasizes the importance of exposure to these models and the resources necessary to start and maintain them. Since many communities do not have one or both of these, our research prompts future analyses to explore how these factors contribute to adopting collective impact and what we can do to provide resources and exposure to communities who need them most. How are communities isolated, and what can be done to provide for them? By exploring this question, NNSI hopes that we will better understand social welfare programs––and society overall. As collective impact leaders continue to emerge and operate across the country, hopefully, they will consider the impact their outreach could have on neighboring communities. With enough conversations in the sector, nonprofits could create the exposure needed to kick-start collaborations in those communities that need it most.

¹ Wood, D. J., & Gray, B. (1991). Toward a comprehensive theory of collaboration. The Journal of Applied Behavioral Science, 27(2), 139–162.

Original article:

Boyer, A.-M., Cooper, K. R., Dougherty, S. M., Wang, R., & Shumate, M. (2020). Predicting Community Adoption of Collective Impact in the United States: A National Scan: Nonprofit and Voluntary Sector Quarterly.

Learn more about collective impact networks in: 

Networks for Social Impact by Michelle Shumate and Katherine Cooper – expected publications July 2021 by Oxford University Press.