Improving nonprofit engagement on social media: Using big data, machine-learning, and sentiment analysis to evaluate leading nonprofits’ message strategies on Twitter/X

Current research on nonprofit organizations (NPOs) indicates that interactive and engaging communication on social media platforms is preferrable to unidirectional communication. To gauge adherence to this notion among leading NPO communications, the current study incorporated big data, message features, sentiment analysis (SA), and social-media diffusion measures to examine the Twitter/X posts (N = 487,547) of the Forbes’ America’s Top 100 Charities in the US over a period spanning eight years. Results indicated that unidirectional strategies were much more common than engaging or dialogic strategies, although the latter two types of messages led to significantly higher positive sentiment and significantly lower negative sentiment within the message. Interactive features (media and hashtags) were included in most Tweets, and these features also led to greater positive sentiment. Interestingly, negative sentiment also was slightly but significantly correlated with greater diffusion on social media.