The role of the environment: how mask wearing varies across different activities

Abstract

Background: People's decisions to engage in protective health behaviors, such as mask wearing during the COVID-19 pandemic, are influenced by environmental and social contexts. Previous research on mask wearing used a single question about general mask usage in public, which may not reflect actual behavior in every setting. The likelihood of wearing a mask during one activity is also related to the likelihood of wearing a mask in another or avoiding an activity entirely. This analysis compared responses between a general question and activity-specific questions and identified patterns of mask-wearing behavior across activities.

Methods: Online, opt-in, cross-sectional surveys were conducted every 2 months from November 2020 to May 2021 (n = 2508), with quota sampling and weighting to achieve a representative sample of the U.S.

Population: Respondents were asked how frequently they wore a mask in public and during 12 specific activities including: on public transportation, while shopping, and attending social gatherings indoors and outdoors. Spearman's rank order correlation was used to compare the frequency of mask wearing reported using a general question versus an activity specific question. Additionally, a latent class analysis was conducted to identify patterns of mask wearing behavior across activities.

Results: There was little to no correlation (r = .16-0.33) between respondents' general attitudes towards mask wearing and their reported frequency of mask wearing in different activities. Latent class analysis identified six distinct groups based on their mask wearing behaviors and avoidance of certain activities. The largest group (29%) avoided ten of the twelve activities and always wore a mask during activities that could not be avoided. Additional groups included those who avoided most activities but made exceptions when around friends or family (20%), part time mask wearers (18%), and never mask wearers (6%).

Conclusions: The findings suggest that activity-specific questions provide more accurate and useful information than a single general question. Specific, context based, questions allow for analyses that consider the nuances of people's decision-making regarding engaging in protective health behaviors, such as mask wearing, thus enabling public health professionals to create targeted guidelines and messages.

Keywords: COVID-19; Face mask wearing; Latent class analysis.

Fig. 1

Predicted probabilities of either activity avoidance or mask wearing across activities for the 6-group model