| Popis: |
Italian Abstract: L’affermazione sociale di piattaforme digitali come Facebook, Twitter o Youtube ha portato con se un mutamento anche nel modo di pensare (e praticare) il consumo di prodotti televisivi. Alle dinamiche fruitive di tipo mass-mediale o broadcast si e affiancata la costituzione di "audience partecipative" emergenti dall’integrazione tra reti di relazioni, comunicazioni ed interessi ego-centrati. A livello accademico, cio si e tradotto in un crescente interesse scientifico orientato allo studio sia delle pratiche di online engagement che, piu in dettaglio, di specifici prodotti comunicativi e forme espressive messi in campo durante la fruizione di programmi televisivi. Nonostante lo scenario ricco di casi empirici, sono praticamente assenti studi cross-genere che mettano a confronto format televisivi diversi e le corrispondenti pratiche di "social TV". In quest’ottica, il presente lavoro costituisce il primo tentativo italiano di confrontare le pratiche di fruizione partecipativa su Twitter attivate da un programma di intrattenimento (X Factor) ed uno di approfondimento politico (Servizio Pubblico). Le domande di ricerca cui si provera a dare risposta sono le seguenti: a) Quali momenti o aspetti dei due format innescano il coinvolgimento attivo del pubblico?; b) In che modo le forme espressive performate su Twitter dagli utenti-fruitori di talk show politici si differenziano da quelli di programmi d’intrattenimento? c) Quali sono (se vi sono) gli elementi di continuita/discontinuita tra questi "pubblici partecipativi" per quanto riguarda contenuti e stili comunicativi online?. La ricerca e stata condotta attorno a due dataset di tweet: il primo, relativo alla sesta stagione di X Factor, e formato da 772.018 tweet; il secondo riguarda l’edizione 2012/2013 di Servizio Pubblico per un totale di 611.393 tweet. Per ciascuno dei 37 episodi (9 di XF6 e 28 di Servizio Pubblico) abbiamo provveduto a individuare i momenti di maggiore coinvolgimento attivo del pubblico (picchi di produzione di tweet originali). Cio ha condotto all’individuazione di 39 picchi per Servizio Pubblico e 16 per X Factor 6. Ciascuno di questi momenti e stato classificato. Sulla base di questa classificazione, e stato costruito un sample di picchi su cui e stata condotta una hybrid content analysis di tipo comparativo.English Abstract: Trans-media usage practices, with a specific reference to the ones implying strong correlation between TV viewing and online content production, are quickly becoming a fascinating field of study for scholars and practitioners. According to Nielsen, during 2013, about 36 million people sent out 990 million of Tweets about TV. At the same time, growing number of studies focus on users’ communicative dynamics while watching specific TV programs or genres. From surveys about participatory viewing styles and repertoires to market-oriented researches, and more general attempts to go beyond Habermasian definition of public sphere. From media adoption perspective, Italy seems to be in line with general trend. Showing increasing shares both in TV viewing and Internet usage. In fact while, on the one side, average TV audience in a normal day raised of 1.8% from 2012 and "average time viewed" (the average daily time spent by every TV users) of 1.3%, on the other side, total amount of people using social networking sites like Facebook and Twitter, wikis or traditional websites passed from 53,8% to 58,7% ( 5%) during 2013. Despite this complex and interesting scenario, analyses and researches aimed at highlight some kind of correlations between different TV viewing styles and online contents actually represents an underestimated field of study. Opposite to this trend, this paper presents the first large-scale comparative study of Twitter contributors’ behavior in commenting two different types of TV format. Attempting to validate (or refute) the main assumption according to which, while TV programs based on pure entertainment are more likely to develop within participatory audience some kind of self-representative or conversational practice tout court, political TV show viewing should be mainly focused on developing a sort of conversation-based civic engagement in which information as well as communication production and sharing appear to be highly covered with political/civic sense. Therefore, we will especially attempt to answer the following research questions: RQ1. Are Twitter contributors taking part to conversations around political talk shows, different from contributors commenting around talent shows? Do they show different patterns of behavior concerning form, topic and the explicit Twitter recipient? RQ2. What are (if there is) the most significant elements of continuity/discontinuity between these contributors regarding contents as well as communicative styles? RQ3. To what extent socio-technical features of “Twitter as platform” influence the emergence of TV-based participatory practices? To answer the above research questions we collected, during TV season 2012/2103, two complete datasets of Tweets. The first consists of Tweets (n=474,167) belonging to one of the most engaging Italian political talk show named "Servizio Pubblico". During the 28 episodes aired during the season, we observed an average of 17,196 Tweets per episode. The second dataset contains Tweets (n=772,018 Tweets) published during the 9 live episodes aired during the season 2012/2013 of "X Factor" Italy. In both cases, we choose to collect only the Twitter messages marked by the official hashtags launched by TV program itself (#ServizioPubblico, #XF6). Instead of random sampling the data – a strategy not always effective with highly skewed distribution such the one we are dealing with – we decided to focus our analysis on peaks of Twitter activity. For both datasets we calculated a by minute time series of original Tweets (excluding both reply and RT). On these time series, we run an algorithm for peaks detection identifying 115 peaks in the #XF6 dataset and 127 peaks in the #ServizioPubblico one. A content analysis was carried on Tweets (n=3,000) created in a random sample of 10 peaks for each dataset. For these 10 peaks we also analyzed the corresponding TV scene. |
