Research objective⭕️: How can we better understand how people respond to and gather information about a pandemic on Facebook
Project brief: The research team wants to better understand how people respond to a pandemic. We expect to deliver insights about how people respond to and gather information about a pandemic on Facebook in order to help them to formulate a relevant solution in future communication.
Method:
In order to focus on understanding the characterization of CDC communications via Facebook regarding the COVID-19 pandemic and its effects on public engagement and response, we think it is important to gather behavioral data from people(how people engaged to a specific post) and behavioral data from CDC ( communication characteristics). Therefore, we used quantitative content analysis to investigate how people respond to CDC's message on Facebook.
Coding Process
19 posts (4% of the data) scattered throughout the time period were used to create the codebook for CDC communications. A training process was used to train the coders. 25 new posts (5.3% of the data) were used for the coders to practice coding and calculation of intercoder reliability. Cronbach’s or Krippendorff’s alpha was calculated for this training data, reaching a threshold of 0.7 or above for the alpha of each category.
Statistical Analysis
Linear regression analysis was conducted to investigate the relationship between the dependent variable of response (as indicated by the number of negative/neutral/positive comments) and the independent variables (topic, medium, purpose, advice, tone, and evidence). Additionally, the linear regression model was also used to test the relationships between the dependent variables of engagement (number of reactions/comments/shares) of the posts and the independent variables (topic, medium, purpose, advice, tone, and evidence). In order to further understand if the CDC tended to use certain versus uncertain language for particular topics, a Chi-square analysis was run between topics and tone certainty to check for the association.
Implications
This work provides insight into how scientists and public health officials can interact with the public.
1.👍 Videos seem to be a good tool to present information that needs to have wide dissemination and engagement. Interestingly, the CDC only used video 16.4% of the time; adding more videos may be a simple way to increase engagement.
2. 👎Advice on personal behaviors led to an increased negative response. People tend to engage with posts that provide information and no advice.
Future Studies
In the future, we will use survey experiments to further investigate public responses to the communication of emerging scientific issues. We can modify the language used to describe the issue, including the use of more definite uncertain language, in order to gauge how this affects understanding, trust, and behavioral response to scientific communications. This will help us provide more input for scientists and public health officials on how to tailor their message to achieve the desired response from diverse audiences.
Challenges
There are hundreds to thousands of comments in response to each Facebook post. The Facebook platform automatically puts some comments near the top as “most relevant” according to their algorithm; these often have the most likes and replies. We need to make sure the comments we selected are representative.
Since this is a collaboration with the Department of Microbiology and the Department of Biomedical Engineering, we started this project by aligning our research goals. We ask how each other perceive the actual scope of inquiry. Questions like: What areas do we want to investigate? What are the key information and methods gaps that we have? Out of all the research questions, what is it that we need immediately?