“Can I write this is ableist AF in a peer review?”: A corpus-driven analysis of Twitter engagement strategies across disciplinary groups

Xiaoyu Xu, Jeroen Gevers, Luca RossiIbérica 2023 · DOI

TL;DR

While scientists are often told to use more "flair" on social media, this study reveals a surprising paradox: Social Scientists use the fewest images and emojis but enjoy the highest engagement rates. The secret sauce isn't just visual—it’s the use of explicit critique and connecting academic expertise to trending social and political issues.

What Problem Does This Paper Address?

Scholars are increasingly pressured to step out of the ivory tower and onto social media to disseminate knowledge. However, most "best practices" for science communication are one-size-fits-all. This paper investigates whether disciplinary background—the "academic tribes" we belong to—actually shapes how we tweet and how the public responds.

The authors explore "context collapse," where a single tweet must simultaneously satisfy fellow experts, journalists, and the general public. They specifically look for the rhetorical and multimodal strategies that drive engagement across four broad groups: Arts and Humanities (AH), Social Sciences (SS), Life Sciences (LS), and Physical Sciences (PS).

Methods

The researchers employed a "corpus-driven" approach, combining quantitative computational linguistics with qualitative close reading.

  • Analytical Framework: They adapted Tardy’s (2023) engagement model and Hyland’s (2005) interaction framework.
  • Metrics: They tracked "multimodal elements" (emojis, images, GIFs, URLs), "tweet types" (single tweets vs. threads), and "interaction markers" (hedges like perhaps, boosters like clearly, and reader pronouns like you).
  • Sentiment Analysis: They used VADER, a sentiment analysis tool, to determine the "polarity" (positivity vs. negativity) of attitude markers in the tweets.

Datasets

The study analyzed a newly curated corpus of 4,000 English-language tweets collected between August 2021 and April 2022.

  • The Sample: 40 scholars (10 per disciplinary group).
  • Selection Criteria: Scholars had to have between 2,000 and 40,000 followers, be research-active, and tweet at least once a week. The authors also balanced for gender and race to ensure a representative data slice.
  • Comparison: Unlike previous studies that might look at a single field (like epidemiology), this dataset allows for direct cross-disciplinary comparison using the Nesi et al. (2005) grouping system.

Key Insights and Ideas

  • The Social Science Engagement Paradox: SS scholars were the "least multimodal" (fewer images/emojis) and the "least interactive" (fewer hedges/boosters), yet they had the highest engagement. Why? Because they frequently linked to news articles and used "assertive, critical language" to address social issues.
  • Life Sciences as the "Visual Storytellers": LS scholars are the heavy hitters of multimodality. They use the most mentions, emojis, and GIFs. They also utilize long threads to "unfold" complex scientific narratives, effectively turning Twitter into a digital poster session.
  • Disciplinary Habits Die Hard: Even on a 280-character platform, old habits persist. AH scholars wrote the longest tweets (averaging 159.9 characters), reflecting the "soft" discipline tendency toward interpretation and wordiness. PS scholars, conversely, used the fewest words, relying on videos and images to demonstrate physical phenomena.
  • Positivity vs. Critique: LS tweets were overwhelmingly positive, focusing on community building and celebrating colleagues. SS tweets were significantly more negative/critical. Interestingly, the "snarky" or critical tone of SS scholars seemed more effective at provoking responses and retweets than the "collegial" tone of the life sciences.

Key Results

  • Engagement Rates: Social Sciences led the pack with an average engagement rate of 0.82%, nearly double that of Life Sciences (0.42%) and Physical Sciences (0.45%), and nearly triple Arts and Humanities (0.30%).
  • Multimodality: LS used 102.3 mentions per 100 tweets, compared to just 37.8 for AH. LS also used nearly 5x more emojis than SS.
  • Threads: LS used threads significantly more often (9% of posts) than SS (3.3%).
  • Sentiment: SS was the only group where "negative" attitude markers (e.g., ableist, terrifying, oppression) were a primary driver of engagement.

Limitations and Future Work

The authors acknowledge that a sample of 40 scholars is a small "early-stage" analysis. The data is also limited to English-speaking accounts, potentially missing different cultural norms in global science communication.

Future research could involve "behind the scenes" interviews with high-engagement scholars to see if their strategies are intentional. Additionally, as Twitter has transitioned to "X," the platform's changing algorithms and features (like long-form posts) may shift these disciplinary dynamics. The authors suggest that science communication training should move away from generic advice and instead help scholars leverage their specific disciplinary strengths—whether that’s the visual data of a biologist or the critical eye of a sociologist.

Summary updated

Abstract

At a time when scholars are increasingly expected to participate in public knowledge dissemination, social media platforms like Twitter hold great promise for engaging both experts and non-experts. However, it remains unclear in what ways academic tweets are shaped by disciplinary concerns and how this might, in turn, impact audience engagement. Our paper reports an early-stage corpus-driven analysis of 4,000 English tweets from 40 scholars’ Twitter accounts across four disciplinary groups: Arts and Humanities (AH), Social Sciences (SS), Life Sciences (LS), and Physical Sciences (PS). Engagement rates (Tardy, 2023), multimodal elements, tweet types, and interaction markers were quantitatively calculated using corpus and computational methods and qualitatively analysed through close reading. Our findings revealed some disciplinary variation in the corpus: specifically, LS used more multimodal elements than SS on Twitter; SS used fewer interactional markers than LS and PS on Twitter. We further found that LS also has the highest number of threads and the longest threads, often to unfold their multimodal information. Despite being the least multimodal and interactive disciplinary group, SS has the highest engagement rate. Our analysis suggests that explicit evaluation and critique plays an important role in eliciting responses on Twitter, particularly with regard to current social or political issues —a finding that resonates with previous research on science communication and popularization (Orpin, 2019). The findings can be applied in science communication training to raise disciplinary awareness in shaping one’s social media presence.
BibTeX
@inproceedings{xu2023can,
  author = {Xiaoyu Xu and Jeroen Gevers and Luca Rossi},
  title = {{“Can I write this is ableist AF in a peer review?”: A corpus-driven analysis of Twitter engagement strategies across disciplinary groups}},
  booktitle = {Ibérica},
  year = {2023},
  doi = {10.17398/2340-2784.46.207},
}
Figure