Giglietto, F., Artieri, G. B., Gemini, L., & Orefice, M. (2015). Understanding Engagement and Willingness to Speak Up in Social-Television: A Full-Season, Cross-Genre Analysis of TV Audience Participation on Twitter. Elsevier BV. https://doi.org/10.2139/ssrn.2611083
Summary
This paper examines Twitter audience participation around two Italian television programmes from the 2012–2013 season — the talent show The X Factor and the political talk show Servizio Pubblico — using a full-season, cross-genre design. Analysing nearly 1.4 million tweets, the authors move beyond the single-episode or single-event focus that characterised much early social-TV research, arguing that this broader lens surfaces both genre-dependent and genre-independent audience behaviours. Their central claim is that television exerts a strong framing effect on the structure, volume, and topics of online conversation, and that the narrative selectors of suspense and surprise catalyse engagement differently across entertainment and political genres. A further finding concerns “willingness to speak up”: viewers express political opinions more cautiously — via jokes or information-bolstered claims — than they do about entertainment.
Key Contributions
- A methodological strategy pairing peak-detection algorithms with hybrid scene/tweet content analysis, enabling study of large, skewed social-TV datasets without conventional sampling.
- A rare full-season, cross-genre comparison that reveals a strong TV framing effect, nuancing earlier single-event studies that emphasised user agency and divergence.
- Introduces Luhmann’s surprise/suspense media selectors as analytical tools for understanding what triggers online audience engagement.
- Advances the notion of “willingness to speak up” in social TV, linking shifts in opinion-expression strategy (jokes, retweets, information-bolstering) to topic sensitivity and to spiral-of-silence / couch-participation debates.
Methods
The authors collected 1,383,414 tweets via the Twitter Firehose (DiscoverText/GNIP Powertrack) from August 2012 to June 2013, using the official hashtags #xf6 (772,018 tweets, 9 episodes) and #serviziopubblico (611,396 tweets, 28 episodes). Data were segmented by air time into per-minute time series (tweets, replies, retweets, originals), and Marcus et al.’s (2011) peak-detection algorithm was applied to the stream of original tweets, yielding 16 peaks for The X Factor and 39 for Servizio Pubblico. All 55 peaks underwent scene analysis coded with a Luhmann-informed codebook (surprise, suspense; α = 0.68). The two highest- and lowest-original-share peaks per show (8 peaks, >12,000 tweets) were then content-analysed with an adapted AEIO-style codebook capturing information, opinion, opinion-as-joke, emotion, attention-seeking, and interaction (α = 0.71), with chi-square significance testing.
Findings
- The X Factor generated far more activity per minute (337.78 vs. 99.61 tweets/min), more original tweets (62% vs. 55% seasonally; 70% vs. 63% in peaks), and fewer retweets within peaks (25% vs. 33%).
- The X Factor drew on both surprise (50% of peaks) and suspense (56.2%); Servizio Pubblico relied mainly on surprise (48.7%) with almost no suspense (5.1%).
- Original tweets were relatively more frequent during scenes featuring a peer-like figure (an audience member speaking or a contestant performing).
- Opinion was the most common tweet category for both shows (64% political, 58% entertainment), but emotional tweets were far more common in entertainment (33% vs. 3%).
- Political tweets were more often framed as jokes (25% vs. 11%) and more often combined opinion with information (10% vs. 4%).
- The opinion-plus-information combination for politics was significantly more frequent in low-original-share peaks (13% vs. 7%), suggesting credibility-bolstering or avoidance of risky direct opinion; no equivalent effect for entertainment.
- Explicit requests for change (interaction) were only ~10–11% of coded tweets and not higher for politics.
Connections
This is an early study within the author’s broader programme on Italian digital political communication and audience participation; it connects most directly to other work by the same group on Italian hybrid media and public opinion, such as Iannelli2015-e0818c3e and Giglietto2012-32b6b551. Its concern with cautious, indirect political opinion expression resonates conceptually with spiral-of-silence and partisan-engagement dynamics explored in the broader topic set, though most of those papers address platforms and periods well beyond this 2012–2013 social-TV context.