By Lisa Lundgren
As members of the FOSSIL project team, Dr. Kent Crippen and I have been exploring ways to effectively engage the public with science using social media. In one recent project, we examined the impact of three elements of social media posts–mentions, hashtags, and URLs–on engagement. “Mentions” occur when a post includes someone’s username, such as @FloridaMuseum or @AndyFarke. Hashtags involve adding a # to a term or phrase (e.g., #paleontology) used in the post. As a result, that post is pooled with all other public posts on that social media platform that include #paleontology which makes it easier for users interested in a specific topic to find relevant posts. Lastly, users can add URLs–links to websites with additional content–to their messages.
To explore the FOSSIL Project’s use of these elements, we analyzed messages posted on the FOSSIL Project’s Facebook and Twitter pages from May 2014 through December 2016. In total, we analyzed 1450 messages, with approximately half the messages coming from Facebook and half from Twitter. We found that there were seven combinations of messaging elements that the FOSSIL Project used in their posts as well as posts that did not include any elements. Some posts included only mentions, while others included hashtags, mentions, and URLs, while others included mentions and URLs. But which elements made posts engaging?
To help us determine which message elements, if any, are most likely to interest viewers, we used data analytics provided by Facebook and Twitter. Specifically, Facebook and Twitter provide what is called engagement rate. Engagement rates are calculated by determining the number of people who interacted with a message by either “liking” it, sharing it with others, or commenting on it and dividing that total by the number of people who saw a message (interactions/number of people who saw a message). The engagement rate is usually displayed as a percent, such as 3.2 percent.
To our surprise, we found that messaging elements actually tended to lower engagement rates! The posts with the highest engagement rates were those that had no messaging element, averaging an engagement rate of 4.7 percent. Posts that only included a hashtag dropped the engagement rate only slightly, to 4.3 percent. However, we found statistically significant differences in engagement rates when the other two elements, mentions and URLs, were included in the post. In posts that included only a URL, the engagement rate was 3.3 percent. Including only the mention element significantly lowered engagement rates to a rate of 2.3 percent. Interestingly, including mentions and hashtags in a post lowered the rate to 1.9 percent!
What we took away from these findings is the ways that message elements structure a social media message and therefore affect engagement. Mentions are a way to call out a specific person or organization, and therefore create barriers to community conversations; however, mentions are helpful and important to communicating with specific people. If you are interested in creating social media messaging to include communities, social media messages should include elements such as hashtags, as these do not significantly lower your engagement rates and are designed to include a community through aggregating information. We suggest that scientists, museums, and clubs can get their messages out more effectively by following these strategies.
We recently presented these findings at the largest international science education research conference, NARST. To read the full paper and view the research poster with associated figures, please visit http://bit.ly/fossilnarst2018.
Lundgren, L. & Crippen, K. J. (2018, March). Educative social media for informal science learning: Effective message design across two digital niches. Annual International Conference for the National Association for Research in Science Teaching (NARST), Atlanta, GA