During a public health crisis, officials need to communicate with the public quickly and accurately. Social networking sites (SNS) have been identified as an appropriate channel for this type of communication; however, few studies have examined what makes SNS messages effective. Further limiting research in this area is a lack of attention to theoretical constructs that may explain message effectiveness in SNS.

In this dissertation, I propose that diffusion of innovations (DOI) be used to understand SNS and message success on SNS. In doing so, I compare traditional message success (persuasion) to message success on SNS platforms (amplification) and provide a brief overview of relevant message design constructs. I then conduct a study to analyze Twitter messages from state and local health departments and federal government agencies charged with communicating to the public during a public health crisis to test these theoretical claims and identify message elements that increase SNS message amplification. The context of the study is the fall 2014 Ebola crisis in the United States. The messages are first classified using content analysis methods to identify message design elements related to content, structure, and style. The success of those elements, in terms of the influence they have on messages amplification, is then evaluated using negative binomial regression.

The results suggest that specific content (hazard information, response instruction, and official action), effective structure (word and image graphics), and key style choices (using figurative language, one hashtag or a keyword hashtag, and the first person) improve the amplification value of a message. Other choices, like mentioning another user, reduce the amplification value. These findings add to the evidence that suggests that DOI enhances scholars’ understanding of communication on SNS. In addition, the results demonstrate that messages can be conceptualized as innovations, and, as such, their characteristics influence the likelihood that they will be diffused through SNS platforms. The results suggest that those charged with communicating during a public health crisis use specific message strategies for SNS messages. These strategies include recommendations related to message content, message structure, and message style. Finally, the results suggest that scholars should continue research to understand the relationship between message design and message amplification in order to improve our knowledge of communication on SNS and help practitioners identify effective communication practices on this new and important channel. Research should also examine the relationship between persuasion and amplification in order to understand how amplification influences attitudes, behavioral intentions, and behavior in both those who amplify the message and in those who receive the message as a result of that amplification.