Assessment of the Effectiveness of Identity-based Public Health Announcements in Increasing the Likelihood of Complying with COVID-19 Guidelines: An Online Experiment.

2020 
BACKGROUND Public health campaigns to curb the spread of COVID-19 are important in reducing its spread, but backlash to traditional information-based campaigns has been unexpectedly extreme. OBJECTIVE To understand if customizing public service announcements (PSAs) providing health guidelines to match individuals' identities increases compliance. METHODS We conducted a cross-sectional randomly controlled within- and between-subjects online experiment in July 2020. Participants viewed two PSA, one advocating wearing a mask in public settings and one staying at home. The control PSA provided information only, and the treatment PSA was designed to appeal to the identities held by individuals, either a Christian identity or an economically-motivated identity. Participants were asked about their identity and then received a control PSA and treatment PSA matching their identity in random order. The PSAs were about 100 words in length. RESULTS We recruited 300 social media users from Amazon Mechanical Turk following usual protocols to ensure data quality. Eight failed the data quality checks, leaving 292 for the analysis. The identity-based PSA changed the source of the PSA and inserted a phrase of about a dozen words relevant to the identity. A PSA tailored for Christians, when matched with a Christian identity, increased the likelihood of complying by 12 percentage points. A PSA that focused on economic values, when shown to individuals who identified as economically motivated, increased the likelihood of complying by 6 points. CONCLUSIONS Using social media to deliver COVID-19 public health announcements customized to individuals' identities is a promising way to increase compliance with public health guidelines. CLINICALTRIAL
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    48
    References
    0
    Citations
    NaN
    KQI
    []