Across the world, Civil Society Organizations (CSOs) and policymakers are often frustrated when their activities, including communication strategies, underperform. They then question why their strategies fail to translate into strong civic participation, despite offering benefits to the community. Evidence from Uganda reflects this reality. Afrobarometer (2022) reports that 22% of Ugandans attended community meetings and 17% collaborated to address community issues in the past year, with even fewer contacting leaders. These figures suggest that information alone is rarely enough to drive behavior. A citizen’s decision to engage is shaped not only by what is said, but also by who says it, who else is participating, and whether the message comes through trusted relationships within their street, village, or wider community. This points to the importance of social networks, the webs of trust, relationships, and influence that quietly determine whether communities participate in collective action or remain inactive. 

In behavioral science, this phenomenon is known as the “messenger effect”, where the person delivering a message is just as important, and sometimes more important, than the information itself. Our brains use a mental shortcut to decide if the information is trustworthy, relevant, and worth acting upon. In civic engagement, such trade-offs can involve deciding whether community elders are more effective at delivering certain messages than government officials or other civic actors. This is because elders are custodians of a community’s cultural norms and attitudes. In some instances, this is further complicated by the “halo effect”, a cognitive bias where our overall impression of a person influences how we feel and think about them and what they have to say. Essentially, if we like one thing about someone, we tend to assume everything else about them is great, too, and so on. If individuals look to others they trust when deciding whether to contribute to the common good, then social networks become the channels through which cooperation, volunteering, collective action, and civic norms spread.

Understanding prosocial behavior through networks

A social network perspective offers a different lens for viewing prosocial behavior, which rarely occurs in isolation. Instead, prosocial behaviors such as community support, civic participation, and collective action are shaped by the networks of family, friends, neighbors, and peers with whom a person interacts, and by the information shared through these networks. Theories such as Social Learning Theory (Bandura, 1977), Social Norms Theory (Bicchieri, 2016), and Social Capital Theory (Bourdieu, 1986) further emphasize the roles of observation, group expectations, trust, and reciprocity in influencing behavior. To put this into perspective, a person is more likely to trust information from friends than from strangers due to the messenger and halo effects we mentioned earlier. 

To explore this interplay, we conducted a social network study in Mbale District, Eastern Uganda, in October 2025, surveying 2,012 participants across rural (Busoba sub-county) and urban (Industrial Division sub-county) settings. We employed a cross-sectional quantitative design using a structured SurveyCTO survey with name-generating questions to map offline social ties related to advice-seeking and sharing within and beyond community boundaries. We started with randomly selected individuals from the village registers, then expanded outward through nominations. We applied social network analysis techniques to examine network structure and influential positions, including degree, betweenness, and closeness. In addition, behavioral attitudes towards engagement (e.g., self-efficacy, peer influence, community identity) and an engagement index based on participation in civic activities were analyzed to classify participants. Our study was designed to move beyond individual-level explanations of engagement and examine the social network structures that sustain or suppress prosocial behaviors. 

Rural networks are denser, more trusting, and more engaged than urban networks 

Think of a network like a web. A “dense” web has lots of threads connecting people. A “fragmented” web has gaps and broken threads.  The contrast between rural and urban networks revealed how network structure shapes behavior. In rural areas, people had about 4 connections each (family, friends, neighbors they trust). These connections overlapped heavily, your neighbor knew your elder, who happened to know your mother. This created dense clusters of mutual trust. 94% of rural participants said they trusted people within their community, compared to 82% in urban areas. In urban areas, people had similar numbers of connections (around 4), but the network was fragmented. People knew colleagues from work, friends from different areas, and neighbors they rarely saw. These ties didn’t reinforce each other. 

This difference showed up in their behavior: around 40% of rural participants were highly engaged or committed to community activities, compared to 29% in urban participants. Dense rural networks created a “culture of participation” through peer pressure, shared identity, and mutual accountability, encouraging people to show up and participate. Urban networks, broken apart by people moving for work and living among strangers, lacked repeated, trust-based interactions that make prosocial norms stick (Deshingkar & Grimm, 2005; Woolcock & Narayan, 2000). 

Influence requires more than just connections, it requires trust 

Highly engaged people didn’t just have more connections (7+ compared to 5 for the disengaged). They also had deeper trust within those connections as most relied on peers for decisions to participate (87%). However, being well-connected doesn’t automatically make you influential. Even well-positioned individuals in the network may struggle to mobilize others if trust is low or if shared norms do not support engagement (Panaite et al, 2024). This means campaigns that rely on people with “reach” but not “trust” will fail. Influence is most effective when it is embedded in trusted relationships and aligned with the community’s collective values, as seen in rural areas.

Women are socially connected but structurally sidelined 

An interesting finding was the gap between being connected and being influential. Women and men in our sample had nearly identical numbers of connections(average degree of 4). But men’s connections pointed toward other influential people, while women’s connections often pointed toward people on the social periphery, those with less power and reach. We have previously found that patriarchal social norms and cultural traditions in Uganda limit women’s civic and political participation by reinforcing male dominance in decision-making (Akinyi Kay, Liang, & Kombo, 2025). 

The latest findings tells us that women are not less social. They are structurally constrained by gendered norms about who can legitimately lead, speak publicly, or carry messages across the community limiting their structural reach (Ajrouch, Blandon & Antonucci, 2005). Two people with the same number of friends can have vastly different influence based on who those friends are and what power they hold.

People connect with those similar to them, which helps some and harms others 

Another pattern that emerged from the study was the role of similarity (homophily) in shaping connections. Individuals tended to connect with others who were similar to  them, in terms of location, gender, age, or education level. It’s natural. You bond easily with people who share your world. This reflects deep human roots, in which our socialization influences who we trust and how we see the world. Mental shortcuts also influence these, as we can easily connect with those who share our preferences or characteristics. 

Homophily has two sides. The good; it creates tight, trust-based clusters within groups. Rural communities benefit from this, shared identity and similar circumstances create natural coordination. The bad; information and ideas rarely cross group boundaries. Women stay connected to women, youth to youth, rural to rural.Therefore, important information and opportunities get trapped in echo chambers instead of reaching everyone who needs them.

Isolated individuals, mostly urban, female, and young, are left behind 

Finally, about 2% of our sample belonged to no network at all, they neither nominated anyone nor were nominated by others. They were from urban areas (97%), slightly more female (56%), younger (less than 35 years), and disengaged from community activities. This structural exclusion compounds over time. Isolated people miss information, opportunities, and the social reinforcement that motivates participation. Without deliberate intervention, they remain trapped outside the community web.

What this means for practitioners

Information alone does not change behavior. Civic engagement campaigns must be designed to leverage trusted messengers, community identity, social norms in networks and inclusion. 

The messenger strategy matters as much as message content. Different actors in the network serve different functions, and all community leaders should not be treated as interchangeable. Elders legitimize, middle-aged connectors, such as business owners, bridge different groups, youth spread messages quickly, and trusted influencers sustain adoption. Effective campaigns should align the right messenger with the right task.

Urban engagement requires a different theory of change. In fragmented urban contexts, CSOs cannot rely on existing community cohesion. They must actively build cohesion through micro-communities such as neighborhood groups, feedback loops that demonstrate visible outcomes, and hybrid digital-offline strategies that create the repeated interactions where trust grows.

Echo chambers are a risk. High gender and geographic homophily means information circulates within similar groups but rarely crosses social boundaries. There is a need for structured bridging activities, such as mixed-gender dialogues, inter-community and intergenerational exchanges, and youth-elder forums, to ensure information reaches all individuals.

Finally, reaching isolated individuals, i.e., those who are disconnected from the broader network, requires a deliberate trust-first approach before asking for action. Don’t parachute into someone’s life with a campaign. Build trust first through small-group, personal approaches. Only after a relationship exists should you ask for their participation. 

At its core, this research reinforces the idea that prosocial behavior is social. Prosocial decisions, whether to attend a meeting, contribute to a community effort, or speak up about an issue are shaped by the networks people are part of. When these networks are cohesive, trusted, and inclusive, they enable collective action. When they are fragmented or exclusionary, they constrain it. For practitioners and policymakers, this shifts the focus from individuals to relationships, from messages to messengers, and from awareness to social reinforcement. Understanding the invisible architecture of social connections, and who carries influence, who bridges divides, and who is left out, is essential to creating more sustainable and scalable impact.

References

1. Afrobarometer. (2022). Afrobarometer Round 9 survey, Uganda 2022. https://www.afrobarometer.org/wp-content/uploads/2023/08/UGA_R9_Summary-of-results_Afrobarometer-9aug23.pdf
2. Ajrouch K. J., Blandon A. Y., Antonucci T. C. (2005). Social networks among men and women: the effects of age and socioeconomic status. J. Gerontol. B Psychol. Sci. Soc. Sci. 60, S311–S317. doi: 10.1093/geronb/60.6.S311  
3. Akinyi Kay, M., Liang, X. T., & Kombo, S. (2025). Civic engagement in Uganda: Outlining our gender-specific findings (Busara Groundwork No. 20 [Lessons Learned]). Busara. https://doi.org/10.62372/KNRR1483 
4. Bandura, A. (1971). Social learning theory. New York. General Learning Press.
5. Bicchieri, C. (2016). Norms in the wild: How to diagnose, measure, and change social norms. Oxford University Press.
6. Bourdieu, P. (1986). The forms of capital. In J.G. Richardson (Ed.), Handbook of theory and research for the sociology of education. Greenwood Press.
7. Deshingkar, P., & Grimm, S. (2005). Internal migration and development: A global perspective. International Organization for Migration.
8. Panaite, A. C., Desroches, O. A., Warren, É., Rouly, G., Castonguay, G., & Boivin, A. (2024). Engaging with peers to integrate community care: Knowledge synthesis and conceptual map. Health expectations : An international journal of public participation in health care and health policy, 27(2), e14034. https://doi.org/10.1111/hex.14034 
9. Too, G. (2024). How can citizens’ voices enhance governance? Reflections from applied behavioral science research on what motivates citizen participation in East Africa. (Busara Groundwork No. 16) [Lessons learned]. Busara. https://www.doi.org/10.62372/NBDJ8943  
10. Woolcock, M., & Narayan, D. (2000). Social capital: Implications for development theory, research, and policy. World Bank Research Observer, 15(2), 225–249.