Do communication content functions drive engagement among interest group audiences? An analysis of organizational communication on Twitter

Interest group organizations (IGOs) utilize social media, especially Twitter, to engage audiences and advance their causes. User engagement, reflecting perceived value and interactivity, is a crucial metric for evaluating the effectiveness of these communication efforts. While many studies analyze engagement metrics like likes, shares, and comments, few examine the underlying drivers, particularly within the context of IGO communication content. This study addresses this gap by investigating the relationship between the functional use of communication content by IGOs on Twitter and user engagement, examining whether multimedia inclusion modifies this relationship. The increasing reliance on social media for communication necessitates a deeper understanding of effective engagement strategies. Successful campaigns, such as #bringbackourgirls or #EndSARS, highlight the power of social media to mobilize collective action, while others illustrate the challenges of achieving widespread engagement. Existing research examines various factors influencing engagement, including user account characteristics, post context, and content features, but the role of communication content functions remains relatively unexplored. IGOs face unique challenges in mobilizing collective action, navigating diverse audiences and causes. This study seeks to understand how communication content functions influence engagement actions on Twitter, focusing on three key research questions: (a) What are the communication content functions employed by IGOs on Twitter? (b) How do these communication content functions affect user engagement? (c) What is the effect of multimedia inclusion on the relationship between communication content functions and user engagement?

Existing literature offers diverse perspectives on IGOs' use of social media. Some view social media as a resource-leveling tool for resource-poor organizations, while others see it as an advantage for already established entities. However, the key determining factor isn't resource capacity, but rather how organizations conceptualize social media's role. Studies on non-profit organizations' social media use highlight motivations such as shaping news and policy debates, and managing organizational image. Social media's potential for agenda-setting and image-building is well-established, but the effectiveness of these uses remains an area of ongoing investigation. The challenges faced by IGOs in capturing user attention underscore the need to understand the factors driving user engagement, which is vital for building positive reputations and fostering strong audience relationships. Research on online engagement emphasizes the value of user-initiated activities in co-creation. Engagement manifests in behavioral, cognitive, and affective forms. Several factors influence user engagement on social media, including user account characteristics, post context, and content-specific features. While studies have focused on linguistic features and other content aspects, research on the influence of communication content functions remains limited. The importance of multimedia inclusion in enhancing user experience is also acknowledged, though existing findings offer mixed results depending on the platform and content type.

This study employs a mixed-methods approach to investigate the impact of communication content functions on user engagement on Twitter. First, a quantitative approach utilizing text mining and topic modeling techniques is used to analyze the content of tweets from IGOs. The bi-term topic model, chosen for its suitability for analyzing short text, was employed to identify latent topics and communication functions within the tweets. Data was collected from 121 official Twitter accounts of IGOs, with a total of 12,766 tweets collected from September 2017 to August 2019. A sample of 1600 tweets was coded to establish inter-coder reliability using Cohen's Kappa. The study used an inductive coding scheme, developing nine communication content functions: Event (activities, updates), Report (research, group updates), Trade (business, news), Period (date, time), Cancer (health information), People (appreciation, recognition), Unite (call to action), Sign (online activism), and Glean (call for support). The frequencies of each function were recorded for each post. The outcome variable, user engagement, was measured using retweets. A multilevel generalized linear regression model was implemented to analyze the relationship between communication functions (independent variables), multimedia inclusion (dichotomous variable), and the number of retweets (dependent variable), controlling for the number of followers. The model allowed for the examination of effects both with and without multimedia inclusion.

The analysis of 12,766 tweets from 121 IGOs revealed that the most prevalent communication functions were Event (48.93%), People (14.40%), and Period (10.22%). The bi-term topic modeling analysis successfully identified nine distinct communication functions. The top 25 most frequent words, comprising 15% of the total word tokens, included terms related to events, activities, reports, and time specifications, highlighting the IGOs' focus on sharing timely information and updates on their activities. The multilevel generalized linear regression analysis showed significant relationships between communication functions and user engagement, with effects varying depending on multimedia inclusion. Without multimedia, Report, Event, and Period functions had negative effects on retweets, while People, Unite, Sign, and Glean functions had positive effects. However, with multimedia inclusion, Report and Event functions showed positive associations, whereas Unite, Sign, and Glean functions showed negative associations. This suggests that the effectiveness of different communication functions depends on the presence or absence of multimedia elements. The model fit statistics indicated good model fit, confirming the validity of the analysis.

The findings demonstrate that IGOs primarily use Twitter to share event and activity information and reports, corroborating previous research. The high frequency of time-related communications emphasizes the importance of timely updates. The differing effects of communication functions on user engagement highlight the nuanced roles of various content types. Specifically, the positive association of People-related communications without multimedia is consistent with prior findings that personally relevant messages are frequently shared. The contrasting effects of certain functions with and without multimedia illustrate the complex interaction between content and presentation format. The results challenge some earlier assumptions about the limitations of social media for in-depth communication. The finding that reports and policy updates can generate engagement on Twitter suggests that social media can effectively serve as an inside strategy tool for IGOs, enabling them to shape discussions and influence policy indirectly by sharing information and updates. The contrasting effects of some functions when combined with multimedia highlight the crucial need for understanding the compatibility of function and form. The study offers practical recommendations for IGOs: using multimedia to enhance event and activity communications while keeping calls to action primarily text-based.

This study contributes significantly to the understanding of organizational communication on social media, particularly for IGOs. It reveals the importance of tailoring communication content functions to the platform and audience, highlighting the nuanced interplay between content type, multimedia inclusion, and user engagement. The findings suggest that IGOs can leverage Twitter for both disseminating information and directly influencing policy, using a blend of strategies to maximize audience engagement. Future research could explore the effects of other engagement metrics, such as likes and comments, across multiple social media platforms and investigate the influence of contextual factors on the observed relationships.

This study is limited by its focus on Twitter data, potentially limiting generalizability to other platforms. The use of retweets as the sole engagement metric may not fully capture the complexity of user interactions. Future studies could explore additional engagement measures and investigate platform-specific differences in the relationships between communication functions and user engagement. The cross-sectional nature of the study also limits inferences regarding causality.