Replication Data for: Channels for Influence or Maps of Behavior? A Field Experiment on Social Networks and Cooperation

Communities in developing countries often must cooperate to self-provide or co-produce local public goods. Many expect that community social networks facilitate this cooperation, but few studies directly observe real-life networks in these settings. We collect detailed social network data in rural Northern Ghana to explore how social positions and proximity to community leaders predict donations to a local public good. We then implement a field experiment manipulating participants’ opportunity to communicate and apply social pressure before donating. We find clear evidence that locations in community social networks predict cooperative behavior, but no evidence that communication improves coordination or cooperation, in contrast to common theoretical expectations and laboratory findings. Our results show that evolved, real-life social networks serve as a mapping of community members’ already-engrained behaviors, not only as an active technology through which social influence propagates to solve collective action problems.