Social norms and cooperation: a meta-analysis
Social norms prescribe how people should behave in social situations and are argued to play a crucial role in sustaining cooperative relationships and coordinating collective action. Despite their theoretical and practical importance, there is no systematic review on the impact of norms on cooperation decisions. In fact, most social norm research takes the effect of social norms for granted and uses norms as a post-hoc interpretation of observed behavior.
To fill this gap, we first conducted a systematic search for all studies that elicit norms in standard games measuring cooperation and prosociality. We keep those studies that use established incentivized approaches to social norm elicitation, either Bicchieri (2006) or Krupka and Weber (2013), run in the lab or online (not in the field) and use monetary incentives for choice and norms elicitation. Until now we are able to identify 61 papers and 227 treatments.
We will then conduct a meta-analysis to assess the effect of social norms on cooperative behaviour. Specifically, we test if there exists a correlation between social norms and cooperation, if compliance with social norms is related to the norm strength, and if punishment affects norms and actions. To do so, we develop a specific a measure of consensus around a norm, called consistency.
The collected data will be used to build the SocialNorm Databank: an international, multidisciplinary open-access database for cataloging the history of empirical research on social norms in cooperation settings. It will enable social norms researchers to query past experimental data on social norms, integrate and systematize knowledge from multiple disciplines, upload new data from experiments and surveys, run several types of analyses (such as meta-analyses, meta-regressions, estimate publication bias and power analyses), and foster open-access and reproducible science. It will contain experimental data at the study level (e.g., aggregated cooperation levels and social expectations, effect sizes, features of the experimental game and design, social norm measurement methods) and the individual level (e.g., individual cooperation decisions, normative and empirical expectations, demographics, preferences and norm sensitivity). The SocialNorm Databank is complementary to existing resources, such as the Cooperation Databank (Spadaro et al. in press), which contains accessible data of empirical studies on human cooperation in social dilemmas.
To fill this gap, we first conducted a systematic search for all studies that elicit norms in standard games measuring cooperation and prosociality. We keep those studies that use established incentivized approaches to social norm elicitation, either Bicchieri (2006) or Krupka and Weber (2013), run in the lab or online (not in the field) and use monetary incentives for choice and norms elicitation. Until now we are able to identify 61 papers and 227 treatments.
We will then conduct a meta-analysis to assess the effect of social norms on cooperative behaviour. Specifically, we test if there exists a correlation between social norms and cooperation, if compliance with social norms is related to the norm strength, and if punishment affects norms and actions. To do so, we develop a specific a measure of consensus around a norm, called consistency.
The collected data will be used to build the SocialNorm Databank: an international, multidisciplinary open-access database for cataloging the history of empirical research on social norms in cooperation settings. It will enable social norms researchers to query past experimental data on social norms, integrate and systematize knowledge from multiple disciplines, upload new data from experiments and surveys, run several types of analyses (such as meta-analyses, meta-regressions, estimate publication bias and power analyses), and foster open-access and reproducible science. It will contain experimental data at the study level (e.g., aggregated cooperation levels and social expectations, effect sizes, features of the experimental game and design, social norm measurement methods) and the individual level (e.g., individual cooperation decisions, normative and empirical expectations, demographics, preferences and norm sensitivity). The SocialNorm Databank is complementary to existing resources, such as the Cooperation Databank (Spadaro et al. in press), which contains accessible data of empirical studies on human cooperation in social dilemmas.