We analyzed Twitter accounts across ten regional communities to examine social media communication patterns and found certain words that led to increased engagement.
One common question asked by many nonprofit communication professionals is how to maximize their reach and engagement on social media, especially with a limited budget. As an online network communication tool, social media provides the opportunity to reach new and wider audiences, without the price of traditional advertising. However, the mechanisms by which to reach that audience can be harder than expected, especially with changing metrics and algorithms on each of the platforms.
During our analysis of the Communities RISE Together (RISE) grant, funded by the Centers for Disease Control and Prevention, we focused on demonstrating the reach and effectiveness of messages to address vaccine hesitancy to further health equity. One way we looked to evaluate effectiveness is through the messages posted on social media accounts of the ten regional communities we were working with. While many of these organizations had accounts with several social media platforms (Facebook, Instagram, Twitter, YouTube), we chose to focus on Twitter. This was due in part to ease of access as a social media platform, and that it also provided the easiest application programming interface (API) to create a dataset.
We began by identifying all Twitter accounts affiliated with one of the ten community coalitions within the RISE collaboration, resulting in 48 accounts, about 75% of the total number of organizations (n=64). Then, using the Twitter application programming interface (API) and the R package rtweet, we pulled data containing the Twitter timelines for each organization. Our final dataset contained over 83,000 unique tweets across 48 different organizations, with an average of 8,236 (sd = 5,550) tweets per coalition and 1,915 (sd=1,209) tweets per organization.
In order to gauge effectiveness of tweets within the dataset, we focused on Twitter post likes and retweets. What we found displayed a pattern of increased engagement based on specific words used within the tweet’s content. Our findings identified certain words such as food, older adults, equity, and COVID that were most associated with increased likes and retweets on the platform. We realized that, through examining social media communication patterns, we could help guide organizations on how to increase messaging engagement. Additionally, while our focus was on vaccine messaging, we chose to look at how RISE partners were using Twitter historically to understand both how they could improve vaccine messaging specifically and their messaging and engagement with tweets overall.
It is important to note that the strongest predictor of receiving likes and retweets is the number of followers, as the more followers an account has, the more potential likes and retweets they can get. Additionally, sentiment was a significant, but not meaningful, predictor of tweet engagement, so making a tweet more or less negative will not necessarily predict whether the tweet will have more engagement.
Read our article in the Journal of Public Health Management and Practice:
Kalie M. Mayberry is a social impact researcher and educator, exploring activism practices and community governance models at Berkman Klein Center for Internet and Society at Harvard University. She has previously conducted research at Columbia Business School and Wharton School of Business, and holds an MPA from University of Pennsylvania.
Jonathan P. Scaccia, PhD, is the Principal of the Dawn Chorus Group. He is a community psychologist and evaluator with 20+ years of experience working in community-based settings. He has extensive experience helping organizations select, adapt, implement, and evaluate community-based improvement interventions. Dr. Scaccia was one of the initial developers of the R=MC2 readiness model that helps organizations understand and build their capacity and motivation for change. His current research focuses on developing comprehensive methods to better integrate data-informed decision making in community-based settings, especially those supported by artificial intelligence paradigms. Dr. Scaccia also founded the research synthesis website PubTrawlr. He received his Ph.D. in clinical-community psychology from the University of South Carolina and completed a research fellowship in the US. Dept of Health and Human Services, Office of the Assistant Secretary for Health’s Public Health Systems, Finance, and Quality Program.
Dr. Mary Mitsdarffer is an assistant professor in the Biden School with primary research responsibilities in the Center for Community Research and Service. As an interdisciplinary scholar, she uses a mix of applied quantitative, qualitative and archival methods. Her research focuses on the intersection of policy, ethno-racial health disparities, and life course development.