Can High Exposure Help to Close Gaps? The Influence of COVID-19 Preventive Messages on Behavioral Intentions by Educational Attainment in Chile

ABSTRACT

The literature in the field of health communication has shown the existence of gaps between groups of different socioeconomic levels both in exposure to health campaign messages and in the influence these messages exert on individuals’ health decisions. In this article, we examine the association between educational attainment and exposure to messages promoting COVID-19 preventive behaviors, namely, physical distancing, handwashing, and masks wearing in Chile. In addition, we model the association between exposure to these messages and the intentions to carry out the behaviors, as well as the differences attributable to educational attainment in the influence of exposure. A cross-sectional population survey combining online and telephone techniques was conducted among 3,592 adults in Chile. Regression analysis revealed that lower educational level and frequency of television use were positively associated with greater exposure to preventive messages. Exposure to preventive messages, in turn, was positively associated with intentions to maintain the three behaviors examined. Individuals in the lower educational attainment group had lower intentions to engage in two of the three behaviors, but those in these segments who were more exposed to preventive messages were as likely as their more educated counterparts to intend wearing masks and adhering to physical distance. The findings of this study underscore the importance of reaching the least educated segments with campaign messages in the context of public health crisis, since these messages can close gaps between the more and less educated group in their intentions to engage in preventive behaviors.

Acknowledgement

We thank Sebastián Rivera at University of California Irvine for his help with data analysis.

Disclosure statement

No potential conflict of interest was reported by the authors.

Supplemental data

Supplemental data for this article can be accessed on the publisher’s website.

Notes

1. Dataset available at http://doi.org/10.6084/m9.figshare.14885154

2. We examined whether this study’s findings held when other social position indicator was considered in the model, specifically ESOMAR’s (ABC1, C2, C3, D, and E) socioeconomic status (SES) classification. SES is a composite measure that includes educational attainment as well as occupation and household income. We found our main effects results to be consistent for the most part, as the higher SES segments had lower levels of preventive message exposure and reported greater likelihood of engaging in preventive behaviors compared to the lowest segments. Unfortunately, due to the high number of SES categories, the interaction between SES and message exposure was not tested because of the existence of cells with zero cases.

3. We found television to be the only media consumption variable that was consistently associated with exposure to preventive messages, while Facebook use was the only social media use significantly associated with exposure to social distancing messages. Additionally, except for Twitter use, we found that neither mass media nor social media use was associated with intentions to engage in the preventive behaviors. In order to rule out that measurement error was preventing us to find the contribution of other media and social media use for explaining exposure and intentions, we created composite measures of media use including television, radio, and online newspaper, and a composite measure of social media use, including Facebook, Twitter, and Instagram. We found that media use was associated with exposure to distancing and handwashing, whereas no association was found for any of the behavioral intentions analyzed. In turn, social media use was not associated with exposure nor with intentions to engage in preventive behaviors. We believe that from a campaign design and evaluation standpoint it is valuable to know which media and social media use are associated with exposure and with the target behavioral intentions and that is why we chose to include each media and social media use as individual predictors in the models.

4. In order to rule out the ceiling effect and to probe the interaction in a most robust manner, we divided the data into subsamples of educational attainment and estimated the model separately in each sub-sample. In the case of mask wearing, the coefficient of preventive message exposure falls across the subsamples as educational attainment increase. In a similar vein, we estimated models with intentions to wear a mask as dependent variable and educational attainment as the independent variable at different levels of the mask wearing message exposure and found that when exposure was low, the coefficient for educational attainment was the highest and when exposure was highest the coefficient for educational attainment was the lowest. The same trends were found in the case of distancing intentions.

Additional information

Funding

This work was supported by Agencia Nacional de Investigación y Desarrollo (ANID), grant # [ANID-COVID0508].