original scientific article UDC 316.775:179.8 received: 2014-06-01 READERS OF ONLINE NEWS COMMENTS: WHY DO THEY READ HATE SPEECH COMMENTS? Karmen ERJAVEC University of Ljubljana, Faculty of Social Sciences, Department of Communicology, Kardeljeva pl. 5, 1000 Ljubljana, Slovenia e-mail: karmen.erjavec@fdv.uni-lj.si ABSTRACT In this study, 378 readers of online hate speech comments were surveyed to examine the motivations of audiences for reading the hate speech comments that are published below online news texts. The research's results revealed that respondents were motivated to read them for guidance, entertainment or social utility, convenience, and to seek information. Seeking guidance was the primary motivation for reading online hate speech comments. The respondents' age, level of education, and gender were negatively correlated with all the motivations for reading them. Income was not correlated significantly to motivations. Those who conducted a greater amount of online activities read comments for entertainment. Hate speech online comments attracted greater numbers of politically affiliated people. Key words: Internet, audience, hate speech, reader online comments, motives, uses and gratification approach. LETTORI DI COMMENT! DELLE NOTIZIE ONLINE: PERCHE LEGGONO I COMMENTI INCITANTI ALL'ODIO? RIASSUNTO Nella ricerca sono stati intervistati 378 lettori di commenti incitanti all'odio, pubblicati come articoli giornalistici. Risultati dimostrano che gli intervistati leggono questi articoli per il loro orientamento, divertimento / per socializza-re, comoditä, cercando le informazioni. La ricerca delle informazioni/consiglio era riportata come motivo principale per la lettura di commenti. L'etä dell'intervistato, educazione e genere sono in correlazione negativa con sopra elencati motivi per la lettura di commenti incitanti all'odio. Il reddito non e statisticamente correlato con I motivi per la lettura di questo tipo di commenti. Cli intervistati che sono piu attivi e presenti sull'internet leggono anche piu commenti. Commenti incitanti all'odio attirano di piu I lettori con forte affiliazione al partito politico. Parole chiave: internet, pubblico, discorso incitante all'odio, lettori di commenti online, motivi, approccio AgendaSetting. INTRODUCTION Whereas traditional media readers were limited to writing letters to the editor, an online version of such public forum is available to online news media readers. They can immediately publish their opinions below online news texts. Scholars have labeled such practice in different manners. For example, some have labelled it as "daet-geul" (e.g., Cho, 2007; You et. al., 2011), "readers blogs" (e.g., Domingo, Heinonen, 2008; Hermida, Thurman, 2008), "audience participation opportunities" (Domingo et al., 2008), "(online) reader comments" (e.g., McCluskey, Hmielowsky, 2012; Nilsen, 2010; Santana, 2010; Wardle et. al., 2009; Weber, 2013). I used the term "online reader comments because its use is extremely commonly among scholars and the Slovenian public (Erjavec, 2012). Online reader comments have been presented as a new form of interactivity that could provide a larger public forum and a greater level of civic participation (e.g., Cho, 2007; Rosenberry, 2011) and that could "save" journalism (e.g., Bowman, Willis, 2003). However, the possibilities of interactivity have also increased the likelihood that hate speech, which is commonly defined as any form of expression that is directed at objects of prejudice that perpetrators use to wound and denigrate their recipient, might appear among news websites' comments (Erjavec, Poler Kovačič, 2012a). Therefore, numerous authors (e.g., Cammaerts, 2009; Domingo et al., 2008; Erjavec, Poler Kovačič, 2012a, 2012b; Santana, 2012) have highlighted that online hate speech comments have increased in number and have emphasized the need to analyze them. A study of hate speech victims demonstrated that the consequences of hate speech that they experience are similar to those that the recipients of other types of trauma experience (Leets, 2002). Studies on online hate speech comments have predominantly included analyses of regulations (e.g., Milosavljevic, 2012; Motl, 2010; Nemes, 2002; Santana, 2012), discourse (e.g., Erjavec, Poler Kovačič, 2012b), ethical dimensions (e.g., Vobič et. al., 2013) and the characteristics of writers of hate speech comments (e.g., Erjavec, Poler Kovačič, 2012a). However, the reason why people read hate speech online comments has not been the focus of any research. In this research, I contribute to uncovering what motivates readers to read online hate speech comments. In this study, I surveyed 378 readers of hate speech comments that were published below online news texts by their readers to examine their motivations for reading them. I employed a uses and gratification approach to analyze the readers' motivations because it is useful for studying people what motivates people to use the Internet (Kaye, Johnson, 2004). In the first chapter, I outline the theoretical background; in the two chapters that follow, I outline the methodology of this research, and present the results of the survey, respectively. In the concluding chapter, I summarize and discuss this research's results. THEORETICAL BACKGROUND Motivations for Reading Internet Components According to Sundar and Limperos (2013), the uses and gratification approach based on Katz and his colleagues (1974) regards (a) the social and psychological origins of (b) needs, which generate (c) expectations from (d) the mass media or other sources, which lead to (e) differential patterns of media exposure (or engagement in other activities), resulting in (f) need gratification and (g) other consequences, perhaps mostly unintended ones. The assumption inherent in this approach is that people have innate needs that can be satisfied by the media (Katz et al., 1974). Early Internet media consumption studies had already included an approach to the Internet as a medium that had a greater level of goal-orientation than the traditional mass media (Chan, 2011; Eighmey, 1997; Kaye, 1998; Kaye, Johnson, 2004; Lin, 2001, 2002). Scholars believe that online users are aware of the motives they are attempting to gratify a variety of motives (Chan, 2011; Eighmey, 1997). Studies that are based on the uses and gratification approach include the assumption that audience members actively search for media messages to satisfy specific needs (McLeod and Becker, 1981; Palmgreen, 1984; Kaye, Johnson, 2004). Scholars consider the uses and gratification approach to be optimally suited to study communication over the Internet. They have examined how people use the Internet in general (Chan, 2011; La-Rosa Eastin, 2004; Papacharissi, Rubin, 2000) and specific Internet components (Kaye, Johnson, 2004). According to Chan (2011, 69), four primary motivations exist for using the Internet: information seeking, socialization, entertainment, and pastime. When people are motivated to seek information, they use the Internet for self-education and information purposes. The socialization motivation comes into play the Internet becomes "a facilitator of interpersonal communication and activities" (Korgaonkar, Wolin, 1999, in Chan, 2011). People use the Internet for enjoyment and relaxation purposes when they are motivated to seek entertainment. The pastime motivation refers to instances where people might use the Internet to occupy their free time without seeking a productive interest. Although this set of motivations for using the Internet does not match exactly that relating to the use of traditional news media, the four primary dimensions hold promise for exploring the gratification-seeking motivations in the context of online newspapers because of their broader range and their commonalities with the motivations that were identified in traditional news media studies (Chan, 2011). Whereas most studies have included examinations of the Internet as a single entity, Kaye and Johnson (2004) specifically considered the Web, bulletin boards or electronic mailing lists, and chat rooms, and discovered that each Internet component satisfied slightly different needs. The 442 respondents of their study were motivated to use the Web to seek for political information concerning advice on how to vote, entertainment or social utility, and convenience, and to seek information. Entertainment or social utility, information seeking, convenience, and guidance were the four motivators for using message boards or electronic mailing lists. Respondents were drawn to chat rooms for guidance or because they were seeking information, entertainment or social utility, and convenience. Online Reader Comments Reader comments that are published below online news texts have specific characteristics: (a) they relate to news texts and can constitute an "intertextuality" between the news texts and comments that other users post below them; (b) they share the same space with the original messages, and are considered "parasitic text"; (c) they are "communicative text[s]," in that they are a response to the published news items; and (d) they are purposive and asynchronous (You et al., 2011, 5). Journalists are ambivalent about online reader comments, expressing concerns over the posts, but acknowledging that posts occasionally affect stories (McCluskey, Hmielowski, 2012, 307). According to Mc-Cluskey and Hmielowski (2012), although most journalists did not find the comments useful (Nielsen, 2010), the comments helped them to form story ideas (Robinson, 2010; Santana, 2010; Wardle et al., 2009). In addition, the journalists added story content (Robinson, 2010; Santana, 2010) and fixed inaccuracies (Robinson, 2010) in response to online readers' comments. In a recent study, Nilsen (2013) found that journalists largely ignored reader comments. They felt that anonymous reader comments were primarily a forum for readers to interact with other readers. The fact that journalists largely supported the idea of having online comments, but did not read them, suggests that journalists viewed comment spaces as a third place for readers. Using this conception, technology had satisfied the users' needs, making it a closed technology rather than a tool that allowed mutual shaping. Whereas users might have participated in ongoing versions of comments by conversing with other readers, journalists maintained the "we publish, you read" mentality (Nilsen, 2013). According to Torres da Silva (2013, 179), the principal problem that emerges from online reader comments is the anonymity of those responding online, which introduces concerns relating to verification, accountability, and accuracy. Some scholars argued that anonymity and the use of nicknames foster greater openness in debates because the participants felt freer to express their opinions online, could encourage the expression of parts of the self that were repressed in offline interactions, and could remove the fear of being personally banned. However, others referred to the opportunity for the conscious deception of identity as potentially leading to undermining the trust that exists within online groups and to misinformation and inaccuracy, as well as to a greater numbers of verbal attacks (leading to injury and humiliation), among other forms of incivility (Torres da Silva, 2013). Aside from anonymity, Torres da Silva also referred to other problems: posts and comments often included a focus on personal viewpoints and did not meet other arguments, nor were they written in response to other participants. The pressure to reply quickly could limit the likelihood of the participants carefully considering and re-developing their positions, and could lead particular individual persons or groups to monopolize the readers' attention (Torres da Silva, 2013). An analysis of people's reasons for reading online reader comments showed that the comments being read were strongly associated with the use of general and entertainment news (You et al., 2011). In particular, reading entertainment news was the most powerful indicator that readers read comments that are published below online news texts. This means that reading online comments could be a behavior that is strongly motivated by entertainment seeking. The probability that readers would read comments decreased as the age of the readers who engaged in online communities increased, and increased with the amount of online news that they read (Torres da Silva, 2013). Similarly, Tenenboim and Cohen (2013), in their study of the characteristics of heavily read versus highly commented-upon news items, found that 40%-59% of the heavily read items were different from the highly commented-upon items. Whereas sensational topics and curiosity-arousing elements were more widely read than the highly commented-upon items, political or social topics and controversial elements were among the more frequently commented-upon items. Readers were overwhelmingly engaged in a dialogue with each other in the comments section on political topics, such as racism (Loke, 2012). The quality of dialogue does not reflect the most thoughtful exchanges. Such reader comments do not need to foster an intellectual dimension; rather, they need to foster a civic dimension (Craig, 2010). Loke (2012, 249) claimed that racism is more difficult to combat now because anonymous reader comments' section hosts racist dialogues. Online Hate Speech Reader Comments The new public sphere that online news sites have created, which was meant to promote democracy, might have had the effect of excluding specific marginalized groups further and might have increased the possibility that news online sites' comments might contain hate speech. However, hate speech must be defined before its appearance in online readers' comments can be dis- cussed. Although, no definition of hate speech is universally preferred, certain common elements of hate speech have emerged. Hate speech refers to an expression that is abusive, insulting, intimidating, harassing, and/or incites to violence, hatred, or discrimination (Erjavec & Poler Kovačič, 2012a). It is directed against people based on characteristics such as race, ethnic origin, religion, gender, age, physical condition, disability, sexual orientation, political conviction (Erjavec, Poler Kovačič, 2012a). Social scientists agree that the regulation of hate speech should not be left solely to the law because laws and regulations are only possible within the legal system, which is consensually created from the external interests of political and civil actors (Dragoš, 2007). The reduction of complex issues to legal matters means transferring responsibilities to one sector and discharging the responsibilities of other sectors (Dragoš, 2007). The research on hate speech comments that are published below online news texts is limited but growing. Online reader comments contain greater numbers of hate speech messages than written letters to the editor because hate speech messages in letters can be eliminated from publication more easily. Editors deliberately permit the publication of online reader comments because they perceive them as a means of keeping old users of online news media and attracting new ones, and because reporters and editors do not know how to identify hate speech (Erjavec, Poler Kovačič, 2012a, 2012 b; Motl, 2010). Research of how reporters define and understand hate speech has demonstrated that over one third of interviewees from the mainstream Slovenian media cannot define hate speech, whereas other interviewees can differentiate between hate speech and offensive speech (Erjavec, 2012). An analysis of editors' understanding of hate speech produced similar results. Specific editors among the key Slovenian news media editors do not differentiate between hate-speech and other forms of banned speech and do not employ precise regulatory and self-regulatory systems and, thus, numerous decisions regarding hate-speech comments are arbitrary (Milosavljevic, 2012). Research on discourses in readers' hate speech comments that were published in mainstream Slovenian news web sites demonstrated that readers used different strategies, mostly rearticulating the meaning of news items (Erjavec, Poler Kovačič, 2012b). Hate speech referred to race (against the Roma), nationality (against Croats and other nations from the ex-Yugoslav republics), sexuality (against homosexuals), politics (against political opponents), and religion (against the Catholic Church, Muslims, and Jews) (Erjavec, Poler Kovačič, 2012b). An analysis of the personality characteristics of online producers of hate speech comments identified two groups of writers: the first group of writers consisted of "soldiers" who were organized, whereas the second group of non-organized producers includes "believers", "players," and "watchdogs" who acted on their own initiative (Erjavec, Poler Kovačič, 2012a). The principal motivation of the soldiers and believers was to complete their mission; they shared the characteristics of an authoritarian personality. Seeking thrills and fun motivated players, whereas social injustices drew the attention of watchdogs; these two subgroups had the characteristics of a libertarian personality. One of the most noted measures of the regulation of the hate speech in online reader comments is the removal of anonymity (Santana, 2012). Kling and colleagues (1999, 84) have claimed that people might express anonymously views that they would not express if they believed they could be identified and held responsible for them. However, Boyd (2011) indicated that the people who most heavily rely on pseudonyms in online spaces are those who are most marginalized by the systems of power. Thus, the removal of anonymity is not empowering; it is an authoritarian assertion of power over vulnerable people. Because studies on the motivations for reading online reader hate speech comments are absent in the existing literature, this research constitutes a partial attempt to fill this research gap and answer the following research questions: RQ1: What are the motivations for reading the hate speech online reader comments? RQ2: How strong is the correlation between the motivations for reading the online hate speech comments of readers and demographics, the strength of political party affiliation, and the use of the Internet? RQ3: Can demographics, political attitudes, the periods spent using the Internet, and the number of online activities predict what motivates readers to read the online hate speech comments of other readers? METHOD To answer the research questions, I posted an online survey on the World Wide Web in March and April 2014. Internet readers were asked to participate in the survey through announcements that I placed on online news sites, electronic mailing lists, and social networks sites, such as Facebook and Twitter. Analogously to the approach that Kaye and Johnson (2004) adopted in their study, in this study, I used a convenience sample of Internet users who were directed to the survey by the online announcements. In such survey, I could not calculate the response rate because knowing how many people might have seen the survey and might have refused to participate is impossible. A total of 378 readers of hate speech comments participated in this survey and replied to all questions in the questionnaire. Among the respondents, 83% (n = 314) were men and 17% (n = 44) were women; 49% (n = 185) were younger than 30 years of age; 28% (n=106) were between 30 and 35 years of age; 13% (n = 49) were between 46 and 60 years of age; and 10% (n = 38) were older than 61 years of age. To familiarize the respondents with the definition of hate speech, Iincluded the definition of hate speech into the introductory text: "hate speech comments include expression that are abusive, insulting, intimidating, harassing, and/or incite violence, hatred, or discrimination against people on the basis of their race, ethnic origin, religion, gender, age, physical condition, disability, sexual orientation, and political conviction." I developed a questionnaire that is based on Kaye and Johnson's (2004) study. Thus, I adopted their 22-statement scale, which they derived from past uses and gratification studies. The five factors were defined (Kaye, Johnson, 2004) as follows: (a) Guidance: People who look for advice and are interested in accessing information to guide their decisions; (b) Information seeking: Information seeking is an activity that is more purposeful than guidance and is defined as actively searching for specific information and monitoring the social landscape; (c) Entertainment: People seeking entertaining information for relaxation and amusement purposes; (d) Social utility: Using hate speech comments to reinforce decisions and arm people with information to use in discussions with others; and (e) Convenience: Seeking information from hate speech comments because it is more convenient than turning to traditional sources. Respondents indicated the extent of their agreement with reasons for reading the online hate speech comments by referring to a Likert scale ranging from 1 (strongly agree) to 5 (strongly disagree). The items were factored subsequently by conducting a principal components analysis that employed a varimax rotation. The summated indexes of each factor were created by summing the individual variables and conducting a reli- ability analysis. The factors were shown to be internally consistent (a = .901). The motivation factors were correlated with the strength of party affiliation. Respondents were asked to report whether they viewed themselves as a "strong party supporter," a "weak party supporter," or as "independent" (Kaye, Johnson, 2004). Based on Kaye and Johnson's study (2004), this study also accounted for Internet experience: the period spent using the Internet and the number of activities that respondents conducted while online. Respondents were asked to estimate how much time they spent on the Internet, they have been accessing the Internet, and to select from 22 online activities, such as sending e-mails, accessing news, surfing for interesting web sites, accessing bulletin lists, researching for school work, shopping, and downloading free software. In this study, I employed correlation and regression analyses to test the research questions. The motivation factors were correlated with the strength of party affiliation. A hierarchical regression was conducted to examine whether demographics (age, income, gender, and education), strength of party affiliation, and Internet experience (time spent using the Internet and number of online activities) could predict motivations for reading online hate speech comments. RESULTS Basic Information Of the 387 survey respondents, 16% (n = 62) were strong party supporters, 21% (n = 81) were weak party supporters, and 63% (n = 244) were independent. The amount of time that the respondents spent using the Internet on typical school, study or work days differed from their use of it on weekends or during holidays, in that it increased substantially (Table 1). Over a quarter of respondents used the Internet for 2-3 hr on Table 1: Time of Internet Use (n = 378): Tabela 1: Čas uporabe interneta (n = 378): How much time do you spend on the Internet? On a typical school/study/ work day On weekends or during holidays Approximately 10 min or less 4% 2% Approximately half an hr 8% 5% Approximately 1 hr 11% 8% 1-2 hr 15% 9% 2-3 hr 28% 21% 3-4 hr 18% 25% 4-5 hr 12% 22% 5 hr or more 4% 8% a typical weekday (28%, n = 106) and 3-4 hr (25%, n = 95) on weekends. According to the frequency is on the second place respondents use the Internet for 3-4 hr (18%, n = 68) on typical weekdays and 4-5 hr (22%, n = 83) on weekends. Fifteen percent (n = 57) of respondents use the Internet for 1-2 hr on typical school, study or work days, and 22% (n = 83) use it for 4-5 hr on weekends. The respondents using the Internet for the least period use it for approximately 10 min or less. The respondents engaged in an average of 10 of the 22 possible online activities, the most popular of which was posting e-mails (92.8%). This activity was closely followed by accessing news (92.3%), which included reading hate speech comments, and social network sites (91.7%) (Graph 1). In this study, 87.2% of respondents regularly surfed for interesting web sites. Respondents regularly engage in researching for school and work (68.7%), playing games (64.3%), banking or paying bills (64.1%), shopping (61.3%), searching services (60.8%), researching consumer products (55.2%), downloading or listening to music (49.3%), accessing bulletin boards or lists (45.4%), downloading free software (43.6%), and booking travel arrangements (41.6%). The second last group of online activities include chatting in forums (39.4%), instant messaging (37.2%), sending electronic postcards (32.5%), job searching (32.1%), checking stock or finance news (28.5%), and creating Web pages (22.8%). The last group of online activities includes respondents who are buying or selling stocks (10.3%) and auctioning (7.2%). Furthermore, slightly over eight out of 10 respondents (83.2%) had graduated with a college degree or had a higher university degree. Almost a half (48.7%) and slightly less than one quarter (24.6%) of the respondents reported a net monthly income between €1000 and €1500 and between €500 and €1000, respectively. The net monthly income of 11.3% and 8.3% of the respondents was, respectively, less than €500 and more than €2000. Finally, 7.1% of the respondents did not have an income. Motivations for Reading the Online Hate Speech Reader Comments Factor analysis revealed the following four motivations for reading online hate speech reader comments that were published below online news texts: guidance, entertainment and/or social utility, convenience, and information seeking (Table 2). Each factor had an eigenvalue of at least one. Guidance, entertainment and/ or social utility, convenience, and information seeking accounted for, respectively, 35.2%, 27.4%, 11.8%, and 10.6% of the variability, totaling 85% of the variance. Readers were drawn primarily to the online hate speech comments to seek guidance. Entertainment and/ or social utility was the second strongest reason for reading the hate speech comments that were published below online news texts. Convenience was the third Access news Surfing interesting sites Playing games 1 Shopping Research consumer products ■ Access bulletin boards/lists Book travel arrangements Instant messaging Job search Check web pages Auction 0,0% 10,0% 20,0%% 30,0%% 40,0%6 50,0%6 60,0%6 70,0%6 80,0%6 90,0%6 100,0%6 Percentage of users ( n=278) Figure 1: Online activities. Slika 1: Spletne dejavnosti. Table 2: Motivations for Reading Online Hate Speech Comments (n = 387): Tabela 2: Motivi za branje spletnih komentarjev s sovražnim govorom (n = 378): Factor matrix Guidance Entertainment and/or social utility Convenience Information seeking Hate comments use motivations ''I read hate speech comments Factor 1 Guidance To help me decide about important issues. .71 .02 .05 .15 To judge qualities of news texts. .71 .11 .21 .11 To see what other think about the issue. .69 .14 .08 .31 For unbiased viewpoints. .64 .05 .19 .11 To find out about issues affect people like me. .49 .15 .03 .49 Factor 2 Entertainment and/or social utility Because it is entertaining .02 .66 .03 .03 To enjoy the excitement of the reading different viewpoints .04 .77 .04 .07 To give me something to talk about with others .22 .58 .01 .09 Because it is exciting .19 .59 .39 .22 To use as ammunition in arguments with others .18 .55 .03 .29 Because it helps me relax .03 .58 .04 .22 To remind me of my protagonists strongest points .44 .55 .29 .08 Factor 3 Convenience To access information quickly .03 .02 .74 .29 Because information is easy to obtain .22 .01 .75 .17 To see how protagonists stand on issues .31 .12 .61 .01 To access information from home .28 .01 .59 .39 Factor 4 Information seeking To find specific information that I am looking for .06 .04 .21 .65 To keep up with main issues of the day .21 .19 .01 .71 To access information at any time. .21 .19 .39 .51 Eingevalue 7.3 4.2 3.9 2.7 Variance explained 35.2 27.4 11.8 10.6 Reliability .89 .77 .83 .69 strongest reason for reading hate speech comments. Lastly, locating specific information was the weakest motivator for reading hate speech comments (Table 2). Motivations for Reading Online Hate Speech Comments and Demographics The relationship between reasons for reading online hate speech comments and demographics, the strength party affiliation, and Internet's use is the focus of the second research question. Education was the strongest demographic correlation. It was significantly and negatively associated with guidance (r = -.28, p < .001), entertainment and/or social utility (r = -.23, p < .001), convenience (r = -0.26, p < .001) and information seeking (r = -.18, p < .01), suggesting that people with a lower level of education were more likely to read hate speech online comments. Age was significantly and negatively related to all motivations for reading online hate speech comments for guidance (r = -.16 p < .001), entertainment and/or social utility (r = -.30, p < .001), convenience (r = -.32, Table 3: Correlation between Motivations for Reading Online Hate Speech Comments, and Demographics, Strength of Party Affiliation, and Internet Use (n = 387): Tabela 3: Korelacija med Motivi za branje spletnih komentarjev s sovražnim govorom, Demografijo, Močjo strankarske pripadnosti in Uporabo interneta (n = 378): Reading hate speech motivations correlates (r) Guidance Entertainment and/or social utility Convenience Information seeking Gender -.21*** -.28*** -.25*** -.21*** Age -.16*** -.30*** -32*** - 13*** Income -.03 -.06 -.08 -.05 Education -.28*** -.23*** -.26*** -.18** Time spent using the Internet .05 .08 .03 .03 Number of online activities .27** .21* .02* .04* Strength of party affiliation .38*** .35*** .32*** .31*** *p < .05, **p < .01, ***p < .001 p < .001) and information seeking (r = -.13, p < .001). Younger respondents were more likely to read online hate speech comments especially for reasons of entertainment and/or social utility and convenience (Table 3). The results also showed that the respondent's gender was correlated negatively with all motivations of reading hate speech online comments. The males were more likely to read hate speech online comments because of guidance (r = -0.21, p < .001), entertainment and/or social utility (r = -.28, p < .001), convenience (r = - .25, p < .001) and information seeking (r = -21, p < .001) than females. Motivations for Reading Online Hate Speech Comments and Strength of Party Affiliation Strength of party affiliation was the variable with the strongest positive correlation to all motivations for reading online hate speech comments. If people were strongly motivated to read online hate speech comments when seeking information, it would be expected that their levels of strength of party affiliation would increase. Levels of strength of party affiliation were significantly and strongly correlated with all motivations of reading the hate speech online comments: guidance (r = 0.38, p < .001), entertainment/social utility (r = 0.35, p < .001), convenience (r = 0.32, p < .001), information seeking (r = 0.31, p < .001). Motivations for Reading the Online Hate Speech Comments and Internet Experience In this study, Internet experience is measured by accounting for both the Internet use periods and the number of online activities that respondents performed regularly. The respondents' periods of Internet use were positively but not significantly associated with all motivations (Table 3). The number of online activities were positively correlated with all motivations for reading online hate speech comments and were significantly correlated with reading the hate speech comments for entertainment and/or social utility (r = .21, p < .05) reasons and for guidance purposes (r = .27, p < .05). The number of online activities were weakly correlated with reading the hate speech comments for convenience (r = .02, p < .05) and information seeking (r = .04, p < .05) (Table 3). Predictors of Motivations for Using the Internet The third research question relates to whether demographics, strength of party affiliation, periods spent using the Internet, and number of online activities could predict the motivations for reading hate speech comments. The strength of party affiliation and Internet experience, but not demographics, were the strongest predictors of motivations for reading hate speech comments. The strength of party affiliation (ß = .31, p < .01) was positively correlated with guidance. The likelihood that readers would read online hate speech comments for guidance increased commensurately with the Internet's influence on the strength of party affiliation (Table 4). The strength of party affiliation and the number of online activities were the only two significant predictors of reading hate speech comments for entertainment and/ or social utility (ß = .21, p < .05; ß = .19, p < .01). The respondents who felt that their involvement in politics increased were more likely to seek hate speech reader comments for entertainment. Additionally, the likelihood of readers reading hate speech comments for entertainment and social reasons increased commensurately with the increased number of online activities that readers engaged in regularly (Table 4). The time spent on online activities was a significant predictor of and was positively correlated reading hate speech comments for reasons of convenience (ß = .13, Table 4: Hierarchical Regression Analysis of Predictors of Reading Online Hate Speech Comments (n = 387): Tabela 4: Hierarhična regresijska analiza prediktorjev branja spletnih komentarjev s sovražnim govorom (n = 378): Predictor variables Guidance Entertainment and/or social utility Convenience Information seeking Gender .01 -.09 .09 -.07 Age -.19 -.10 -.05 -.15 Income .05 .06 -.09 .07 Education .08 .07 -.01 .08 Time of Internet use .18 -.11 13*** .30*** Number of online activities .02 .19** .19 .09 Strength of party affiliation .31** .21* .18 .21* R2 .23 .24 .21 .22 Adjusted R .21 .20 .20 .21 Significance .000 .000 .000 .000 *p < .05, **p < .01, ***p < .001 p < .001). This indicates that readers who spent lengthier periods using the Internet were more likely to read hate speech comments for convenience (Table 4). The periods spent using the Internet and the strength of party affiliation were significant predictors of reading hate speech when readers were motivated to seek information. Periods spent using the Internet were positively correlated with information seeking (ß = .30, p < .001), indicating that readers who spent greater periods on the Internet, were more likely to read hate speech comments for information. Those readers who were strongly party affiliated read hate speech comments for information (ß = .21, p < .05) (Table 4). DISCUSSION AND CONCLUSION Past studies examined all components of hate speech comments that were published under online news, but not the reasons why audiences read them. This study constitutes an attempt to fill this research gap. Whereas hate speech messages are more numerous in online reader comments than in written letters to the editor (Erjavec, Poler Kovačič, 2012a, 2012b; Motl, 2010), analyzing the motivation for reading online hate speech comments is necessary. This study showed that 387 respondents were motivated to read hate speech comments for guidance, entertainment and/or social utility, convenience, and information seeking reasons. Guidance was the primary motivation for reading hate speech comments, suggesting that respondents trusted the producers of comments sufficiently to rely on their advice. Thus, respondents found hate speech comments reliable and trustworthy, even though their authors were anonymous, because readers trusted specific online news sites. Research on the characteristics of writers of hate speech comment showed that they wrote comments only on the news sites that they trusted (Erjavec, Poler Kovačič, 2012b). Convenience emerged as a week motivator for reading hate speech comments, which could indicate that readers might be seeking information on other media platforms. However, they remained loyal to specific online news sites. Those who conducted a greater number of online activities read hate speech comments for entertainment and/or social utility. Comments have benefit of real time and interpersonal exchange. In addition, You and colleagues (2011) found that reading the comments that are published below news texts could be substantially motivated by the need for entertainment. This study also includes an examination how the strength of party affiliation influences the motivations for reading online hate speech comments. Strong party affiliation was strongly correlated to all motivations for reading hate speech comments. Online hate speech comments attract greater numbers of politically affiliated people. Indeed, hate speech online comments might catalyze a greater involvement in politics. The results also revealed that the respondent's age, education, and gender were negatively correlated with all motivations for reading online hate speech comments. Males were the most likely to read hate speech comments than females were. Similarly, You and colleagues (2011) also found that males were more likely to write and read comments below online news texts then females were. Pedersen and MacAfee's (2007), whose findings could be used to explain this phenomenon, found that males concentrate on information and have a higher preference for anonymity, whereas females prioritize personal contact and focus on the social aspects. The younger respondents were more likely to read comments than the older respondents were for all mo- tivations. In addition, You and colleagues (2011) found that younger research participants were more likely to write and read comments than the older participants were. The reason for this could lie in the difference between young people who grew up surrounded by digital technologies and who are comfortable with their technological properties and those people who turned to Web 2.0 applications in later in life (Haferkamp, Krämer, 2008). Income was not significantly related to the motivations for reading hate speech comments. This could be explained by the relative economic egalitarianism, which, for decades, was described by the distribution and redistribution of income in Slovenia (Malnar, 2011). In the previous century and in recent years, income was distributed uniformly and, therefore, it did not represent a key element of differentiation. Education played a more substantial role. People with lower levels of education were more likely to read hate speech comments for guidance, entertainment and/ or social utility, convenience, and information seeking. This phenomenon could be explained by the finding that intellectual development is significantly related to levels of prejudice toward minor social groups (Chicker-ing, Reisser, 1993). The strength of party affiliation was a predictor of reading hate speech comments for all motivations, except for convenience. The periods spent using the Internet positively predicted whether the respondents read hate speech comments. Those respondents who spent greater periods using the Internet read hate speech for convenience and when seeking information. Those how conducted a greater amount of online activities read hate speech because it was convenient and entertaining. No demographic variables were predictors of reading hate speech comments. In addition, I should highlight the limitations of this study. Despite the fact that the questionnaire included a definition of hate speech, respondents might have used their own definition of hate speech or responded to motivations for reading all comments, and not just those including hate speech messages. Kaye and Johnson (2004) also emphasized that, even though researchers recognize online surveys as an effective method for collecting data, they still present a unique set of challenges and limitations that arise from the absence of a random selection. In situations where random probability sampling is not possible, such as with the Internet, a probability sampling is acceptable (Babbie, 1990) and commonly used when posting an online survey (Kaye, Johnson, 2004). Careful uses of this type of purposive sampling generates results that might be representative of a specific subset of Internet users, but might not of the larger population (Babbie, 1990). Although I cannot generalize this study's findings to the readers of online hate speech comments at large, this study does offer an insight into the online behaviors of readers of hate speech comments. In future studies, researchers could perhaps design a method where they could identify, randomly select, and solicit readers to complete their surveys through e-mail (Kaye, Johnson, 2004). In the future, motivations for reading hate speech comments could be derived from open-ended responses rather than from those found in other studies, and could include a question about the readers' definition of hate speech. BRALCI SPLETNIH NOVIČARSKIH KOMENTARJEV: ZAKAJ BEREJO KOMENTARJE S SOVRAŽNIM GOVOROM? Karmen ERJAVEC Univerza v Ljubljani, Fakulteta za družbene vede, Oddelek za komunikologijo, Kardeljeva pl. 5, 1000 Ljubljana, Slovenija e-mail: karmen.erjavec@fdv.uni-lj.si POVZETEK Ker so obstoječe študije proučevale vse prvine diskurza komentarjev s sovražnim govorom, objavljenih pod spletnimi novinarskimi prispevki, razen motivov za branje, skuša ta študija zapolniti raziskovalno vrzel. Anketiranih je bilo 378 bralcev komentarjev s sovražnim govorom, da bi ugotovili, zakaj bralci brejo komentarje s sovražnim govorom. Študija temelji na pristopu uporabe in zadovoljitve. Rezultati kažejo, da anketiranci berejo spletne komentarje s sovražnim govorom zaradi usmeritve, zabave / druženja, udobnosti in iskanja informacij. Iskanje nasveta je bil glavni motiv za branje spletnih komentarjev s sovražnim govorom. Rezultati so tudi pokazali, da so anketirančeva starost, izobrazba in spol negativno povezani z vsemi motivi branja komentarjev s sovražnim govorom. Bolj verjetno je, da moški berejo komentarje s sovražnim govorom kot ženske. Mlajši anketiranci bolj verjetno berejo komentarje s sovražnim govorom kot starejši zaradi vseh analiziranih razlogov. Nižje izobraženi posamezniki bolj verjetno berejo spletne komentarje s sovražnim govorom kot bolj izobraženi. Prihodek anketirancev ni statistično značilno povezan z motivi branja komentarjev s sovražnim govorom. Tisti, ki opravijo več spletnih dejavnosti, preberejo tudi več komentarjev s sovražnim govorom. Ta študija je tudi preučevala, kako je strankarska pripadnost povezana z motivi branja komentarjev s sovražnim govorom. Komentarji s sovražnim govorom bolj privlačijo bralce z močno strankarsko pripadnostjo. Ključne besede: internet, občinstvo, sovražni govor, bralci spletnih komentarjev, motivi, pristop prednostnega tematiziranja. REFERENCES Babbie, E. (1990): Survey Research Methods. Wadsworth, Belmont. Boyd, D. (2011): 'Real Names' Policies Are an Abuse of Power," Apophenia (blog), http://www.zephoria.org/ thoughts/archives/2011/08/04/real-names.html (4. 3. 2011). Bowman, S., & Willis, C. (2003): We media: How audiences are shaping the future of news and information. Reston, VA, Media Center, American Press Institute. Cammaerts, B. (2009): Radical pluralism and free speech in online public spaces. International Journal of Cultural Studies, 12, 4, 555-575. Chickering, A. W., Reisser. L. (1993): Education and Identity. San Francisco: Jossey-Bass. Chan, Y. Y. (2011): Modeling Audience Interactivity as the Gratification-Seeking Process in Online Newspapers. Communication Theory, 21, 1, 67-89. Cho, K. (2007): A study on textual typology of Internet "Daet-geul." Korean Journal of Text Linguistics, 3, 2, 203-230. Craig, D. A. (2010): Excellence in online journalism. Thousand Oaks, Sage Publications. Domingo, D., Heinonen, A. (2008): Weblogs and journalism. Nordicom Review, 29, 1, 3-15. Domingo D, Quandt T., Heinonen A., Paulussen S., Singer J.B., Vujnovic M. (2008): Participatory journalism practices in the media and beyond. Journalism Practice, 2, 3, 326-342. Dragoš, S. (2007): Sovražni govor. Socialno delo, 46, 3, 135-144. Eighmey, J. (1997): Profiling user responses to commercial Web sites. Journal of Advertising Research, 27 3, 59-69. Erjavec, K. (2012): Novinarsko opredeljevanje in razumevanje sovražnega govora. Teorija in praksa, 49, 1, 95-111. Erjavec, K., Poler Kovačič, M. (2012a): "You Don't Understand, This is a New War!" Analysis of Hate Speech in News Web Sites' Comments. Mass Communication and Society, 15, 6, 899-920. Erjavec, K., Poler Kovačič, M. (2012b): Novinarstvo in internet: značilnosti piscev sovražnega govora v komentarjih pod spletnimi novinarskimi prispevki. Teorija in praksa, 2012, 49, 1, 190-203, Haferkamp, N., Krämer, N. C. (2008): Entering the Blogosphere: Motives for Reading, Writing, and Commenting. Paper presented at the annual meeting of the International Communication Association, TBA, Montreal, Quebec, Canada: http://citation.allacademic.com/ meta/p232822_index.html (14. 12. 2013). Hermida A., Thurman N. (2008): A clash of cultures. Journalism Practice, 2, 3, 343-356. Kaye. B. K. (1998): Uses and gratifications of the World Wide Web: From coach potato to Web potato. The New Jersey Journal of Communication, 6, 1, 21-74. Kaye. B. K., Johnson, T. J. (2003): A Web for all reasons: uses and gratification of Internet components for political information. Telematics and Informatics, 21, 2, 197-223. Kling, R., Ya-ching, L., Teich Al., Frankel, M. (199). Assessing anonymous communication on the Internet: Policy deliberations. The Information Society 15, 1, 79-90. Katz, E., Blumler, J. G., Gurevitch, M. (1974): Utilization of mass communication by the individual. In: J. G. Blumler & E. Katz (Eds.): The uses of mass communications: Current perspectives on gratifications research. Beverly Hills, Sage, 19-32. LaRose, R., Eastin, M. S. (2004): A social cognitive theory of Internet uses and gratifications: Toward a new model of media attendance. Journal of Broadcasting & Electronic Media, 48, 3, 358-377. Leets, L. (2002): Experiencing Hate Speech: perception and Responses to Anti-Semitism and Antigay Speech. Journal of Social Issues, 58, 2, 341-361. Lin, C.A. (2001): Audience attributes, media supplementation, and likely online service adoption. Mass Communication and Society, 4, 1, 19-38. Lin, C.A. (2002): Perceived gratifications of online media services among potential users. Telematics and Informatics, 19, 1, 3-19. Loke, J. (2012): Public Expressions of Private Sentiments: Unveiling the Pulse of Racial Tolerance through Online News Readers' Comments. The Howard Journal of Communication, 23, 2, 235-252. McCluskey, M., Hmielowsky, J. (2012): Opinion expression during social conflict: Comparing online reader comments and letters to the editors. Journalism, 13, 3, 303-319. McLeod, J. M., Becker, L. B. (1974): Testing the validity of gratification measures through political effects analysis. In: Blumler, J.G., Katz, E. (Eds.), The Uses of Mass Communication: Current Perspectives on Gratifications Research. London, Sage, 137-162. Milosavljevic, M. (2012): Regulacija in percepci-ja sovražnega govora : analiza dokumentov in odnosa urednikov spletnih portalov. Teorija in praksa, 49, 1, 112-130. Motl, A. (2010): Sovražni govor v slovenskih medijih na spletu Medijska preža, 37, 1, 20-21. Nemes, I. (2002): Regulating hate speech in cyberspace. Information & Communication Technology Law, 11, 2, 193-220. Nielsen, C. (2010): Community conversation or 'the new bathroom wall'? Anonymous online comments and the journalist's role. Presented at the annual meeting of the Association for Education in Journalism and Mass Communication, Newspaper Division, Denver, CO, 4. 8. 2010. http://citation.allacademic.com/meta/p_mla_ apa_research_citation/4/3/4/7/7/p434772_index.htm. (1. 4. 2014). Nilsen, C. (2013): Coproduction or cohabitation: Are anonymous online comments on newspaper websites shaping news content? New Media & Society, http://nms.sage-pub.com/content/early/2013/06/05/1461444813487958. full.pdf (1. 4. 2014). Palmgreen, P. (1984): Uses and gratifications: A theoretical perspective. In: Bostrom, R. N. (Ed.): Communication Yearbook 8. London, Sage, 61-72. Papacharissi, Z., Rubin, A. M. (2000). Predictors of Internet use. Journal of Broadcasting & Electronic Media, 44, 2, 175-196. Pedersen, S., Macafee, C. (2007): Gender Differences in British Blogging. Journal of Computer-Mediated Communication, 12, 4, 1471-1492. Robinson, S. (2010): Traditionalists vs. convergers: Textual privilege, boundary work, and the journalist- audience relationship in the commenting policies of online news sites. Convergence: The International Journal of Research into New Media Technologies, 16, 1, 125-143. Santana A. (2010): Conversation or cacophony: Newspapers reporters' attitudes toward online reader comments. Presented at the annual meeting of the Association for Education in Journalism and Mass Communication, Newspaper Division, Denver, CO, 4. 8. 2010. http://citation.allacademic.com/meta/p_mla_apa_re-search_citation/4/3/4/5/6/p434567_index.html (1. 4. 2014). Santana, A. (2012): Civility, Anonymity and the Breakdown of a New Public Sphere. Unpublished doctoral dissertation: https://scholarsbank.uoregon. edu/xmlui/bitstream/handle/1794/12420/Santana_ore-gon_0171A_10411.pdf?sequence=1. Sundar, S. S., Limperos, A. M. (2013): Uses and Grats 2.0: new Gratifications for New Media. Journal of Broadcasting & Electronic Media, 57, 4, 504-525. Tanenboim, O., Cohen, A. A. (2013): What propts user to click and comment: A longitudinal study of online news. Journalism, http://j'ou.sagepub.com/content/ early/2013/12/27/1464884913513996.refs.html. Torres da Silva, M. (2013): Online forums, audience participation and modes of political discussion: readers' comments on the Brazilian presidential election as a case study. Communication & Society, 26, 4, 175-193. You, K. , Lee, M., Oh, S. (2011): Why Use Online Comments?: Examining the Relationship Among Online Comments, Civic Attitudes, and Participation Intention. Paper presented at the annual meeting of the International Communication Association, TBA, Boston: Http://citation.allacademic.com/meta/p489682_index. html (24. 10. 1. 2014). Vobič, I., Erjavec, K., Poler Kovačič, M. (2013): Journalists' strategies of news reporting on parliament members' hate speech. Journal of applied journalism & media studies, 2, 3, 505-519. Wardle, C., Williams A., Wahl-Jorgensen, K. (2009): ugc@thebbc: A production study examining the ways in which audience material is used at the BBC. Presented at the annual meeting of the International Communication Association, Chicago, IL, 20. 5. 2009. http://cita-tion.allacademic.com/meta/p_mla_apa_research_cita-tion/3/0/0/0/5/p300053_index.html (12. 12. 2013). Weber, P. (2013): Discussions in the comments section: Factors influencing participation and interactivity in online newspapers' readers comments. New Media & Society, http://nms.sagepub.com/content/ear-ly/2013/07/05/1461444813495165 (1. 4. 2014).