Elsevier

Computers in Human Behavior

Volume 77, December 2017, Pages 132-139
Computers in Human Behavior

Full length article
Gendered discourse patterns on online social networks: A social network analysis perspective

https://doi.org/10.1016/j.chb.2017.08.041Get rights and content

Highlights

  • Social networks gendered discourse patterns are examined.

  • This study examined a social networking site with over 100,000 members.

  • Men write more posts in social networking sites, while women commented on other people's posts more often.

  • Women's posts received higher rankings than men's posts.

  • There is a link between activity network structure and content popularity.

Abstract

The study examines the gendered discourse patterns on a popular online social network, TheMarker Café, using social network analysis. Overall, the findings strengthen previous analyses that report evidence of men's assertive and dominant discourse style and social role versus women's more cooperative and supportive discourse style. Men wrote more posts, while women commented on other people's posts more often. Women's posts received higher rankings than men's posts, strengthening the notion that women receive more affirmations on online social networks. The study also examined the interplay between the structure of the TheMarker Café network and gendered discourse patterns. Our findings also confirmed a link between activity network structure and women content popularity.

Introduction

The role of gender in social conversations has attracted the interest of scholars for decades (Leaper and Ayres, 2007, Leaper, 2014, Maltz and Borker, 1982). A meta-analytic review of studies found strong evidence that men produce a more domineering and assertive discourse style in line with their perceived social status and competitive nature, while women's discourse style is more often characterized as affiliative and supportive (Leaper & Ayres, 2007). In light of the central role of online social networks in contemporary society, recent studies have moved to examine whether such traditional gendered discourse patterns persist in online social network environments or whether online social networks are a “game changer” (Brandtzaeg, 2015, Joiner et al., 2014, Joiner et al., 2016).

This study aims to contribute to studies of gendered discourse on online social networks via a social network analysis of gendered discourse using TheMarker,1 an online social networking site operated since 2007 by Haaretz, a prominent Israeli news outlet. While previous studies of online gendered discourse patterns mostly used surveys (e.g., Joiner et al., 2014) or experiments to examine how men and women respond to messages (Joiner et al., 2016), the social network analysis method used in this study allows a deeper examination of the connections between members in a real social network environment and on a much larger scale than most alternative methods (Hayat and Lyons, 2010, Lesser et al., 2016, Onnela et al., 2007, Szell and Thurner, 2013). Indeed, as noted below, the TheMarker Café community includes over 100,000 members, and over 1.7 million links, allowing a large scale and wide perspective on the issue.

The social network analysis method also makes it possible to identity the online network structure and thus illuminates not only the gendered discourse patterns but also the characteristic features of the network environment and their potential impact on community members’ gendered discourse patterns. In contrast to previous studies of gendered discourse that did not address network structure, our analysis maps the link between network structure and the gendered discourse patterns within the network.

Traditionally, it has been argued that certain network positions within offline social networks are associated with increased influence, which may lead to advantageous benefits in offline social settings (Burt, 1992, Erickson, 1996). Recent evidence suggests that these arguments are also true for online environments (Szabo & Huberman, 2010). Typically, online sites facilitate various means of interactions between individuals, such as following or responding. These social and communication ties form a network which provides an opportunity for Online Social Network (OSN) members to increase their visibility and exposure, interact with others, and gain real advantage from their network position (Ellison, Vitak, Gray, & Lampe, 2014). However, researchers have not yet systematically investigated whether, and to what extent, online network structures affect OSN members' abilities to increase the popularity of their content. This article attempts to fill this gap. Building on prior research, we propose that online content's popularity is highly correlated with social network structures, and even more so with the communication interactions network. Our study identifies preferred network positioning that might enhance popularity among OSN users.

The understanding of how certain network structures – which are often associated with social capital – can be mobilized to gain benefits in online environments may assist individuals and organizations seeking to promote their messages. Furthermore, acknowledging gender-specific differences in OSN behavior can provide gender-targeted insights for OSN administrators and members aspiring for influence and leadership.

In line with all the above, and in order to contribute to the question of the social networks gendered discourse patterns, the analysis was specifically designed to examine the following research questions:

  • RQ1 Which gender writes more posts?

  • RQ2 Which gender comments on posts more often?

  • RQ3 The posts of which gender are more popular?

  • RQ4 Which type of social ties (friendship ties or activity ties) are more strongly correlated with content popularity?

To address these research questions, in January 2012, we collected all the historic data – including all the posts and comments – from TheMarker Café since its inception in 2007. A detailed description of the friendship network can be found in Lesser, Tenenboim-Chekina, Rokach, and Elovici (2013), and this portion of the dataset is also publicly available.2 We collected the public portion of TheMarker dataset using a dedicated scraping tool, extracting information on two types of ties among its members — friendship and commenting — which were described by previous studies as the two main forms of communication between individuals on online social networking sites (Bohn et al., 2014, Ghosh and Lerman, 2010). From the over 100,000 members of the network, 42% identified themselves as males, and 36% as females, and the remainder chose not to indicate their gender. As we also did not analyze dormant members, our analysis includes a total of 21,413 TheMarker Café members (52% males and 48% females) who disclosed their gender, had at least one friendship tie, and posted at least one comment during the period analyzed in this research.

Importantly, the question of whether traditional gendered discourse patterns persist in online social network environments is central not only for gender studies but also for information studies, which widely examine whether the web and social networks serve as “social equalizers.” This question arose in light of the notion that online interactions allegedly blur some of the external social boundaries that are emphasized in real-life situations, consequently weakening or erasing traditional “social roles.” Amichai-Hamburger, McKenna, and Samuel-Azran (2008), who analyzed the web's effect on social connections and on social power plays, argue that the web empowers socially disadvantaged individuals and thus the socially poor get richer. Indeed, studies found that individuals who suffer from high levels of social anxiety benefit from interacting in small groups online because the factors that trigger their anxiety in face-to-face situations are absent in text-based Internet interactions (McKenna and Seidman, 2006, McKenna et al., 2002). Other studies found that the greater sense of control and security in online interactions benefits introverts, neurotics (Amichai-Hamburger, 2005, Maldonado et al., 2001), and people with physical disabilities (Raghavendra, Newman, Grace, & Wood, 2013).

In contrast to this optimistic notion, other studies show that the opposite effect occurs in many instances, as the socially rich accrue even greater social assets online. Thus, for example, studies show that attractiveness levels rise when Facebook members have good-looking friends (Walther, Van Der Heide, Kim, Westerman, & Tong, 2008), further compounding effects of the offline beauty bias (Rhode, 2010). Furthermore, in a gender studies context, a recent big data analysis (Brandtzaeg, 2015) that examined gender differences in civic engagement across countries found that traditional gendered discourse patterns persist in online social network environments, and concluded that Facebook is definitely not “the great equalizer.” Analyses by Joiner, Stewart, et al. (‘2014) and Joiner et al. (2016) also identified persistence of traditional women affiliative discourse versus males’ dominant discourse on social networks online.

To contribute to the debate over the web's role as the great equalizer and an e-empowerment tool, the current study contributes to the ongoing mapping of the issue not only by continuing examining the interplay between gender and social media discourse but also by examining the interplay between social network structure, gendered discourse, and content popularity, thus addressing a lacuna in contemporary studies. To place the study's hypotheses in context, analyses of online and offline gendered discourse patterns are reviewed below.

Section snippets

Social network analysis

In recent years, the rise of the Internet has facilitated the emergence of an enormous volume of traceable communication on frameworks such as OSN, providing an opportunity to study social networks on a larger scale (Onnela et al., 2007, Szell and Thurner, 2013). Often these studies have been based on the social network analysis (SNA) methodology (Wasserman & Faust, 1994). A key element of social networks is the ability for individuals to simultaneously interact in multiple social contexts by

Method

The current study uses the social network analysis method, grounded in the premise that social life is created primarily and most importantly by relations and the patterns formed by these relations. Social networks are formally defined as a set of nodes (representing network members) connected by one or several types of relations (Wasserman & Faust, 1994). Because network analysts view networks as the primary building blocks of the social world, they not only collect unique types of data, but

Results

Previous online social network studies have shown that information diffusion and the popularity of posts show heavy-tailed distributions (Cha, Mislove, & Gummadi, 2009). Thus, we used the Mann-Whitney U test to test whether significant gender differences exist for a given set of variables. Table 1 presents the average numbers of posts, comments, views, and stars for females and males respectively, as well as information on friendship degree (i.e., number of friendship ties an individual has –

Discussion and conclusions

The current study examines the gendered discourse patterns on TheMarker Café, a popular online social network, using the social network analysis method. Overall, the findings clearly strengthen former analyses that presented evidence of men's assertive and dominant discourse style and social role versus women's more cooperative and supportive role. In line with our first hypothesis, men wrote more posts, thus initiated and introduced more topics to TheMarker Café community, while women more

References (66)

  • P.B. Brandtzaeg

    Facebook is no “great equalizer”: A big data approach to gender differences in civic engagement across countries

    Social Science Computer Review

    (2015)
  • R. Burt

    Structural holes: The social structure of competition

    (1992)
  • B.M. Byrne

    Structural equation modeling with EQS: Basic concepts, applications, and programming

    (2013)
  • D. Cameron

    Gender

  • M. Castells

    Networks of outrage and hope: Social movements in the Internet age

    (2013)
  • M. Cha et al.

    A measurement-driven analysis of information propagation in the flickr social network

  • J. Coates

    Women, men and language: A sociolinguistic account of sex differences in language

    (2004)
  • J. Cohen

    Statistical power analysis for the behavioral sciences

    (1988)
  • J.S. Coleman

    Social capital in the creation of human-capital

    American Journal of Sociology

    (1988)
  • N.B. Ellison et al.

    Cultivating social resources on social network sites: Facebook relationship maintenance behaviors and their role in social capital processes

    Journal of Computer-mediated Communication

    (2014)
  • B.H. Erickson

    Culture, class, and connections

    American Journal of Sociology

    (1996)
  • F. Faul et al.

    Statistical power analyses using G Power 3.1: Tests for correlation and regression analyses

    Behavior Research Methods

    (2009)
  • R. Ghosh et al.

    Predicting influential users in online social networks

  • Z. Hayat et al.

    The evolution of the cascon community: A social network analysis

  • T. Hayat et al.

    Advice giving and receiving within a research network

    American Behavioral Scientist

    (2015)
  • T. Hayat et al.

    Al-Jazeera Sport's US twitter followers: Sport-politics nexus?

    Online Information Review

    (2016)
  • C.K. Jaccard et al.

    LISREL approaches to interaction effects in multiple regression

    (1996)
  • I. Kawachi et al.

    Social cohesion, social capital, and health

  • R. Lakoff

    Language and woman's place

    (1975)
  • C. Leaper

    Gender similarities and differences in language use

  • C. Leaper et al.

    A meta-analytic review of gender variations in adults' language use: Talkativeness, affiliative speech, and assertive speech

    Personality and Social Psychology Review

    (2007)
  • K. Lerman et al.

    Using a model of social dynamics to predict popularity of news

  • O. Lesser et al.

    The role of network setting and gender in online content popularity

    Information, Communication & Society

    (2016)
  • Cited by (0)

    View full text