Predicting gender from electronic discourse

“For example, females used more emotion words [86,87] (e.”

“More recently, in the context of e-mail authorship characterisation, Thomson et al [34] have investigated the existence of gender-preferential language styles in e-mail communication.”

“For example, evidence indicates that gender is recognizable in text-based language use (Savicki, Kelley, & Oesterreich, 1999; Thomson & Murachver, 2001 ).”

“Much of the literature on cell phone use has focused on university-aged individuals (Harley, Winn, Pemberton, & Wilcox, 2007; Lee & Robbins, 2000; Thomson & Murachver, 2001; Walsh & White, 2007).”

“Taking the case of gender, this can occur through gendered names and message headers but may also be detectable well above chance by gendered language style (Thomson & Murachver, 2001).”

“For example, in one study of samesex dyadic e-mails with participant-determined topics, women referenced emotion more than men, but similar levels of questions and oppositions emerged (Thomson & Murachver, 2001).”

“Thomson and Murachver (2001) examined emails sent to assigned netpals of the same sex over a short period and found that not only was it possible to classify the gender of the sender with a very high level of accuracy on the basis of the frequency of linguistic and content features but that readers…”

“Furthermore, males are more likely to tease and insult (Lampert & Ervin-Tripp, 2006; Thomson & Murachver, 2001), whereas women are more likely to avoid such face-threatening situations.”

“Cues to gender were found to exist in the language used by CMC participants and as the work by Thomson and Murachver (2001) suggests, CMC users may be able to identify gender based on these”

“This finding is consistent with the findings of previous studies investigating communication on Facebook specifically [21], and social media more broadly [27].”

“Recent studies have continued topic-specific research into this direction, such as a study on gender in electronic discourse ( Thomson & Murachver, 2001 ), one on the influence of gender on the development of personal relationships via email (Boneva, Kraut, & Frohlich, 2001), and a study on the impacts of emoticons on…”

“Regarding emotional research, most studies report that women refer to emotions more often than men (Thomson and Murachver, 2001).”

“Earlier research found that women and men react and accommodate to gender-preferential language in electronic communication (Thomson & Murachver, 2001; Thomson, Murachver, & Green, 2001).”

“Conversely, males were found to give more opinions and use more insults (Thomson and Murachver, 2001).”

“Thomson and Murachver (2001) assigned netpals of the same sex to participants and examined the e-mails sent to them over a short period.”

“While some authors find that women express more negative emotions than men [46,47], others (e.”

“The operationalizations of the three language variables were modeled after past language and gender research (Palomares, 2008; S. A. Reid et al., 2003; Thomson & Murachver, 2001).”

“In addition, while some sex differences in CMC language are known (e.g. Herring, 1995), other research indicates that CMC users adapt to dyadic partners’ gender-linked language styles (Thompson & Murachver, 2001), rendering predictions about sex differences in online responses also tenuous.”

“Work in computer-mediated communication ( Colley & Todd, 2002 ; Hancock, 2004 ; Hussey & Katz, 2006a , 2006b; Thomson & Murachver, 2001 ) has successfully demonstrated that online methodologies provide considerable control over manipulated variables when participants engage in interactive language production in a conv…”

“Thomson and Muracvech [82] have shown that there is a gender-preferential language in informal electronic discourse and textual content could be used to identify the author’s gender.”

“Thus, guessing accuracy was significantly higher than chance (p Thomson & Murachver, 2001 ).”

“It has also been observed that genderspecific language is more prevalent in conversations consisting of only one gender when compared to pairs or groups of both genders [12].”

“Moreover, it appears that readers of CMC text are able to predict author gender with a fairly high accuracy (Savicki, Kelley, & Oesterreich, 1999; Thomson & Murachver , 2001). a wide range of language styles depending on whom they theory (CAT; see Coupland, Coupland, Giles, & Henwood, 1988;…”

“Differences found in language style mirrored findings in other forms of communication (e.g., Mulac et al., 1990; Thomson & Murachver, 2001).”

“Or , because of depersonalization in computer-mediated environments, does online communication foreground group identities and accentuate group discrepancies (Carr, 2018; Postmes et al., 2002; Thomson & Murachver, 2001)?”

“…communications has found that male and female users have identifiable differences in their language use (Savicki, Kelley, & Oesterreich, 1999; Thomson & Murachver, 2001, nicknames (Cornetto & Nowak, 2006; Herring & Martinson, 2004), choice of emoticons (Wolf, 2000), and methods of requesting…”

“These gender skews would be of less concern if men and women said the same things; however, this is not, of course, the case, as numerous previous authors have found [58, 24, 25, 64, 31].”

“In contrast, men tend to give more opinions (Aries, 1982, Tannen, 1990; Thomson & Murachver, 2001); dominate conversations by taking longer turns (Herring, Johnson, & DiBenedetto, 1995; Swacker, 1975, cited in Goddard & Patterson, 2000); refer more to facts, quantities, and geographical locations…”

“…that we are fairly good at identifying the sex of authors of e-mail messages, possibly because we attend to gender-typed patterns of language use (Thomson & Murachver, 2001), and e-mails that contain strong emotional words such as excited or wonderful elicit more positive impressions in online…”

“Even those who claim gender difference (e.g., Fitzpatrick, Mulac, & Dindia, 1995; Thomson & Murachver, 2001) acknowledge that language features are gender preferential rather than gender exclusive, and they vary with context (Dovidio, Brown, Heltman, Ellyson, & Keating, 1988; Murachver & Janssen,…”

“Furthermore, males are more likely to tease and insult [108,110], whereas women are more likely to avoid such face-threatening situations [111].”

“Thomson and Murachver (2001) have recently supported this finding.”

“In fact, using several sex-linked linguistic characteristics such as references to emotion, personal information, insults, self-derogatory comments, compliments to interactants, apolo-gies, and hedges, it was possible to identify the author’s sex with 91.4% accuracy (Thomson & Murachver, 2001).”

“Herring's findings correspond to the results of other research in the field of gender and CMC (Savicki, 1996; Thomson and Murachver, 2001).”

“The context of communication influences language style (Thomson and Murachver, 2001; Coupland, et al., 1988).”

“Further, the work of (Thomson and Murachver, 2001) discovered that gender-specific language is more prominent across conversations comprising people of solely one gender than across email exchanges encompassing participants of both genders.”

“…markers, such as references to emotion, insults, and compliments to interactants, it was possible to identify the author’s gender with 91.4% accuracy, which indicates that the communicator’s gender is betrayed through syntactic and pragmatic cues even in text-only CMC (Thomson & Murachver, 2001).”

“Author gender classification task has also been explored for emails ([51], [10]) with up to 91% accuracy in gender prediction; and university students’ essays [34] with discriminant analysis prediction of 72.”

“Journal compilation & 2006 Blackwell Publishing Ltd Journal of Computer Assisted Learning 22, pp360–367 CMC vary in frequency according to the sex of the author of the communication (Herring 1994; Thomson & Murachver 2001).”

“Current research [25-27], particularly that with e-mail, suggests that women (mainly student) writers are more likely than men (mainly student) writers to use higher frequencies of references to emotions, personal information, questions, compliments, self-derogatory statements, and apologies.”

“However, an increasing number of studies have demonstrated that the deindividualization effect (SIDE) of computer-mediated communication foregrounds group identities and increases group discrepancies (Carr, 2018; Herring, 2000; Postmes, Spears, & Lea, 2002; Thomson & Murachver, 2001).”

“Experimental work by Thomson and Murachver [12] found evidence for gender-preferential language in informal CMC and showed that readers of email messages used gender-linked language differences within messages to correctly identify the author’s gender.”

“Thomson and Murachver [51] have provided an examination of gender-preferential language style in electronic discourse with some interesting case study results.”

“There is much in the literature to support Ajana, Boler, and the others’ scepticism over cyber-disembodiment (see Lea & Spears, 1995; Scharlott & Christ, 1995; Thomson & Murachver, 2001; Whitty, 2003; Witmer & Katzman, 1997).”

“Thomson and Murachver (2001) examined the role of CAT in men’s and women’s email communication styles.”

“Thomson and murachver (2001) found that women are more likely to make references to emotional and personal information online. moreover, these researchers found that participants in their experimental study were able to identify the gender of the person they were communicating with online.”

“Men communicate in an adversarial way, by expressing more critical and opposed views [13, 49, 51], disagreeing with others [39, 103], using aggressive [65] or hostile [89] tones, and using swear words [36, 53, 100], insults [114, 115], or threats [51].”

“Several gender studies with adults in face-to-face settings report a similar alignment for language use (e.g., Hanna and Murachver 1999; Thomson et al. 2001).”

“Two studies (Savicki, Kelley, & Oesterreich, 1999 ; Thomson & Murachver, 2001 ) demonstrated that the gender of e-mail authors could be identified at rates above chance.”

“…focus to their communication (Holmes, 1995; Lang-Takac & Osterweil, 1992; Maltz & Borker, 1982) whereas men are more likely to focus on clarity, facts, and conveying their opinions (Mulac, Bradac, & Gibbons, 2001; Mulac, Studley, & Blau, 1990; Tannen, 1990; Thomson & Murachver, 2001).”

“Women and men use different vocabulary in self-presentation 20,21 , and women tend more to limit the visibility of their profiles 22 .”