Write like i write: Herding in the language of online reviews

“Marwick and boyd (2010) conducted a qualitative study of Twitter users, demonstrating some peculiar properties of social media with respect to audience: social media audiences are largely imagined by participants, and “context collapse” — where participants find a diverse set of people, from close…”

“Bell’s theory of audience design (Bell 1984) provides an explanation for style shifting in the absence of direct interaction.”

“Linguistic style varies depending on a range of social characteristics, such as socio-economic status or gender (Labov 1990), and authorship attribution techniques rely on this social (i.e., interspeaker) variance in order to predict the likely author of a given text (Argamon et al. 2006).”

“Many argue that stylistic change takes place subconsciously, and that its underlying driver is social acceptance (Giles, Coupland, and Coupland 1991).”

“…some researchers argue that it is “nearly impossible” to distinguish genuine herding behaviors from uninfluenced behaviors using observational data (Muchnik, Aral, and Taylor 2013), others have demonstrated that under certain conditions, it might be possible to infer causal relationships in such…”

“Previous work showed that users presented with such a ranked list of items are unlikely to look past the first page of results; additionally, the time spent engaging each item decreases as one moves down the list (Pan et al. 2007).”

“This is true of both dyadic (Scissors, Gill, and Gergle 2008; Danescu-Niculescu-Mizil, Gamon, and Dumais 2011) and group communication (Postmes, Spears, and Lea 2000; Cassell and Tversky 2005).”

“The trend is clear even for the remaining features where this complete compliance is not observed, and we expect that given more data and the ability to examine sufficiently larger contexts (i.e., values of L) we would still get a complete compliance.”

“Raafat and Chater (2009) advocate modeling collective behavior at two levels: pattern (i.e., the structure of connections between individuals) and transmission (i.e., the mechanisms that govern the transfer of information between individuals).”

“Of course, an individual is more likely to be influenced by others in her immediate environment, and is more likely to change her behavior the more sources of such influence there are (Latane 1996).”

“, a newcomer in the process of acquiring a group’s vocabulary versus an older participant who has stopped adjusting to group norms and may soon cease participating (Danescu-Niculescu-Mizil et al. 2013).”

“It has been argued that style also changes with life experience (Nguyen et al. 2013).”

“The language of communicators of a lower social status is expected to converge to the patterns of more powerful or prestigious interlocutors (Herring 1996; Wang, Fussell, and Setlock 2009; Danescu-Niculescu-Mizil et al. 2012).”

“Emoticons on Twitter: Is the use of emoticons a social norm, transmitted from user to user? (Park et al. 2013) A user’s network of contacts with whom she has exchanged replies at least twice.”

“The language of communicators of a lower social status is expected to converge to the patterns of more powerful or prestigious interlocutors (Herring 1996; Wang, Fussell, and Setlock 2009; Danescu-Niculescu-Mizil et al. 2012).”

“The language of communicators of a lower social status is expected to converge to the patterns of more powerful or prestigious interlocutors (Herring 1996; Wang, Fussell, and Setlock 2009; Danescu-Niculescu-Mizil et al. 2012).”

“This is true of both dyadic (Scissors, Gill, and Gergle 2008; Danescu-Niculescu-Mizil, Gamon, and Dumais 2011) and group communication (Postmes, Spears, and Lea 2000; Cassell and Tversky 2005).”

“This is true of both dyadic (Scissors, Gill, and Gergle 2008; Danescu-Niculescu-Mizil, Gamon, and Dumais 2011) and group communication (Postmes, Spears, and Lea 2000; Cassell and Tversky 2005).”

“For instance, second person pronouns are characteristic of spoken conversation (Yates 1996) and therefore are likely to be less common in more formal media (Hu, Talamadupula, and Kambhampati 2013).”

“Similarly, he cites his own study of the styles of newscasters across two radio stations (Bell 1982).”

“For instance, language behaviors in Twitter have been used to infer user characteristics including gender (Fink, Kopecky, and Morawski 2012), age (Nguyen et al. 2013) and even ethnicity (Pennacchiotti and Popescu 2011).”

“In this section, we establish the underpinnings of our model of herding in order to study the dynamics of linguistic style.”

“(Barton 2009) Network of contacts established online (email senders and receivers) and offline (e.g., colleagues).”

“For instance, prior ratings and reviews introduce significant bias in individual rating behavior and writing style, creating a herding effect (Muchnik et al., 2013; Michael and Otterbacher, 2014).”

“Review sites, such as TripAdvisor (Michael and Otterbacher 2014), IMDb (Hemphill and Otterbacher 2012) and beer review communities (Danescu-Niculescu-Mizil et al. 2013b), have also been used in studies on language in online communities.”

“[Michael and Otterbacher, 2014] found that stylistic”

“Humans have been shown to entrain to their interlocutor's language at the lexical [10] or syntactic [11] [12] level, and on linguistic style [13] [14] [15].”

“, 2010), and on what (Niederhoffer and Pennebaker, 2002) called linguistic style, which includes, among other features, the use of pronouns and verb tenses (Niederhoffer and Pennebaker, 2002; DanescuNiculescu-Mizil et al., 2011; Michael and Otterbacher, 2014).”

“Other authors uncovered the existence of herding effects in the language [20, 51] and rating [35, 36] of online reviews.”

“…shown that speakers and writers alike adapt their speaking and writing styles to their interlocutors in a variety of different on-and offline contexts (see, e.g., Bunz and Campbell, 2004; Gnisci, 2005; Michael and Otterbacher, 2014; Muir et al., 2016, 2017; Shin and Doyle, 2018; Hilte et al., 2020).”

“…style of particular media, in order to better understand how users exploit and expand their communication affordances (e.g., (Hu, Talamadupula, and Kambhampati 2013)) or when and why people might change their linguistic style (Danescu-Niculescu-Mizil et al. 2013; Michael and Otterbacher 2014).”

“While the resulting public recognition may be motivating to some, a wide range of studies has shown that negative feedback in such public settings can be demotivating [7, 10, 15, 23, 29, 30, 32, 35, 36].”

“Individuals are known to adopt the norms and behaviors of communities upon joining [47] and this tendency has been seen in topical online communities with changes to linguistic behaviors [76, 24, 59, 55, 14].”

“For example, consumers are influenced by previous reviews on platforms like TripAdvisor (Michael and Otterbacher, 2014), and the tone of individuals’ comments is often shaped by the remarks of others in online discussions (Matook et al., 2022).”

“Moreover, the paper augments the scholarly understanding of how rating, readability and usefulness of reviews change as a function of the stages in the reviewer lifecycle as well as the rate of reviewers’ maturity [6,9,13,16].”

“In many settings, mean-field approximations are reasonable for binary choice decisions: many decisions one makes on a day-to-day basis are influenced by many peers acting around an individual, for example, deciding whether to invest in cryptocurrency [22], or the language one uses in online reviews [23].”