The project investigates the contributing factors associated with the choice to wear a mask consistently or not during the COVID-19 pandemic. Most attempts to increase mask-wearing during the pandemic have targeted correlates of mask compliance. The main purpose of our study was to learn about correlates of mask non-compliance and how to increase mask-wearing in non-compliant mask wearers.

The first study investigated the mask-wearing and risk-mitigating habits of undergraduate students at the University of Pennsylvania. Participants completed surveys that identified various factors, such as moral values, political affiliation, health anxiety, health locus of control, self-efficacy, and fear of COVID-19, that might relate to their mask-wearing habits. In Study 1, we found that political orientation and moral values could predict mask compliance. Students who were less compliant with mask-wearing tended to be more conservative and placed more value on respect for authority and loyalty over harm reduction and fairness. Furthermore, men were less likely to comply with mask-wearing than women.

The second study was a follow-up intervention study that attempted to increase mask-wearing rates while in public, particularly of individuals less likely to wear a mask. In this study, we put together two different PSA videos that would appeal to these less compliant participants by featuring authority figures from within and outside the Penn community. One focused on health care authorities, while the other focused on non-healthcare leaders in the Penn community.  We randomly assigned 102 Penn undergraduates to watch one of the two PSAs. After taking the preliminary survey and watching their assigned PSA, participants completed longitudinal daily surveys for each of the next seven days where they reported their mask-wearing habits for each day.

When asked to assess how authoritative each PSA video was, conservative students rated the two videos to be equally authoritative, while liberal students rated the PSA with health professionals to be significantly more authoritative than the PSA with Penn leaders. In the week after watching either PSAs, conservative participants significantly increased their mask-wearing rates compared to baseline, while liberal participants did not. For conservatives, both PSAs were equally effective.  

Because liberal students are already near the “ceiling” in terms of mask-wearing, it is unlikely that a significant increase in mask-wearing among liberals will occur following an intervention such as ours. However, among conservative students, the baseline rate of mask-wearing was significantly lower than that of liberals, meaning that there is more room to move the needle on their mask-wearing behavior with PSAs like the ones we created. The two PSAs, which targeted conservative students by featuring several authority figures, had the same effect of increasing mask-wearing among conservatives. Interventions that encourage mask-wearing can target the values and beliefs associated with non-compliance in order to increase mask-wearing rates in previously non-compliant mask wearers. Our findings further explain why people decide to either wear or not wear masks, along with being used to promote health and well-being on campus.