Purpose The purpose of this paper is to recommend behavioral targets for future interventions to reduce greenhouse gas emissions at college campuses and to advise interventionists on how to choose between many potential behavioral targets. Design/methodology/approach The authors used the community-based social marketing (CBSM) methodology over two studies. In Study 1, the authors assessed adoption rates (i.e. penetration) and likelihood of adoption (i.e. probability) for 16 potential behavioral targets. In Study 2, the authors used quantitative and qualitative methods to assess the barriers and benefits of engagement in five of the top-performing behaviors from Study 1. Findings The findings suggest that an intervention to promote purchasing green energy credits (GECs) has a high potential to reduce emissions. Purchasing GECs has a small penetration (<7%) and a large impact (1,405 kgCO 2 e/person/year). Compared to the other four behaviors the authors examined in Study 2, purchasing GECs is also more convenient and requires very little time. Thus, the behavior should be appealing to many individuals interested in reducing emissions or protecting the environment. Originality/value The authors performed a holistic evaluation of potential behavioral targets that included a barrier and benefit analysis, in addition to the traditional CBSM method of combining impact, probability and penetration.

School closures during the COVID-19 pandemic have highlighted the importance of in-person learning on child health and wellness. Improving indoor air quality (IAQ) is a critical undertaking to keep children in school during outbreak events. Our primary objective was to evaluate the impact of Enhanced IAQ credit achievement among LEED-certified schools on the ability to remain open during the pandemic. In this analysis, schools that achieved LEED Enhanced IAQ credits for increased ventilation or outdoor air delivery monitoring were assigned to the treatment group; LEED schools without these credits were in the control group. We used LASSO regression to select potential confounders for treatment models. Inverse treatment probability weights were used to control for confounding in mixed effects Poisson models in which the response variable was the number of COVID-related school closure days during the 2021-22 academic year. Although effect estimates for the treatment were not significant, they were consistently in the inverse direction (incidence rate ratio (IRR) [95% CI]: 0.90 [0.80, 1.01]). Models may have lacked power to detect significance due to many schools having zero COVID-related closures during the 2021-22 academic year. An important secondary finding was a 53% decrease in COVID-related closure days (95% CI: 7%–77%) associated with a 10% increase in county residents who reported 'Always' using a mask in public. This study contributes to the current understanding of the indoor environment and airborne disease transmission. Efforts to monitor and improve IAQ in schools are important avenues to support community, student, and staff health.