Analysis of Online Harassment Towards Black and Latinx Women PDF

Race and Social Problems (2022) 14:1–13 https://doi.org/10.1007/s12552-021-09330-7 What Did You Call Me? An Analysis of Online Harassment Towards Black and Latinx Women Sara C. Francisco1 · Diane H. Felmlee2 Accepted: 3 May 2021 / Published online: 14 May 2021 © Springer Science+Business Med...

Race and Social Problems https://doi.org/10.1007/s12552-021-09330-7 What Did You Call Me? An Analysis of Online Harassment Black and Latinx Women Sara C. Francisco1 · Diane H. Felmlee2 Accepted: 3 May 2021 / Published online: 14 May 2021 © Springer Science+Business Media, LLC, part of Springer Abstract Online harassment is a growing societal problem. Although attention, little research systematically explores on online harassment directed towards women of color. analysis) to messages posted on the social media platform messages directed towards two groups—Hispanic/Latinx that emerged within the communications. Messages towards In tweets that targeted Black women, one emergent that accused Black women of being overly sexual. In one recurring topic, with common terms related to Both groups of women also were subjected to feminine, munications are not idiosyncratic in nature, but instead As a result, these hostile messages contribute to Keywords Online harassment · Race · Gender · Social Introduction Communication on forms of social media is flourishing, ena- bling individuals to interact and distribute information easily and efficiently. On Twitter, for example, tweets are publicly available. They can be forwarded via retweeting or by shar- ing on Facebook or other media, which allows information to quickly spread across a larger audience. Unfortunately, the effects of network spread can be devastating when the con- tent of a message is negative, whether true or not. The focus of this research is on harmful, negative communication, or online harassment, that is directed at women of color. * Sara C. Francisco [email protected] media platform, which is the goal of this project. Studying the thematic content of online harassment is of particular data-cfemail="85e1ede3b4b7c5f5f6f0abe0e1f0">[email protected] interest, because knowledge of the content can aid in iden- tifying the social processes involved in aggressive, online communication. If consistent themes emerge in these harm- ful messages, it suggests that these tweets are not simply 13 Vol.:(0123456789) Race and Social Problems (2022) 14:1–13 racist and/or sexist stereotypes? Second, what is the pattern of connections among topics in these sets of negative tweets? Last, how does the content of aggressive instances differ between these two groups? facili- when to Background Social Processes in Online Harassment In this research, we examin online harassment, which refers to unwanted, problematic digital contact (Lenhart et al. 2016). Two closely related terms from the literature include cyber bullying (Xu et al., 2012; Bellmore et al., 2015), and cyber aggression (Felmlee et al., 2018a, 2018b; Grigg, 2010), both of which also involve intentional, harmful elec- tronic communication. Two social processes that contribute to the development of online harassment and aggression include the enforce- ment of social norms and the establishment of social hier- archies (Felmlee & Faris, 2016). For example, individuals who do not conform to the traditional, societal expectations of heterosexuality receive high levels of cyber aggression (e.g., Felmlee & Faris, 2016; Hinduja & Patchin, 2010). This behavior illustrates the endorsement of the social norms associated with “compulsory heterosexuality” (Rich, 1980). Furthermore, the motivation to increase one’s status among peers represents another fundamental mechanism involved in the development of online and offline aggression (e.g., Faris, 2012; Sijtsema et al., 2009). Much of school peer victimiza- tion occurs among relatively popular friends, for instance, many of whom are jockeying for similar positions, grades, and respect from peers (e.g., Faris et al., 2020). According to Ridgeway (2011), status hierarchies remain fundamental to inequality within societal gender systems. Online harassment aimed at women of color emanates from such a hierarchical social system within which women of color are placed on the lower strata. Demeaning messages on Twitter that target minority females, thus, likely repre- sent attempts to reinforce the subordinate position of these women in our society. Perpetrators also may view these actions as opportunities to advance their own status in the eyes of their peers and online followers. Furthermore, since women of color are marginalized on the basis of both their race/ethnicity and gender, it is likely they will be especially vulnerable to this kind of aggression. Pervasiveness of Online Harassment on Twitter in Some of the newest channels for interpersonal aggression occur on social media, and attacks within such domains are frequent. Starting in the 2000s, a rise in online harassment and cyberbullying occurred alongside the increased usage of 3 Others noted that minorities were overrepresented as targets of online harassment. For example, Wang et al. (2009) doc- umented that cyberbullying victimization was higher than whites among both Hispanic youth and among adolescents who defined themselves in terms of race as “other,” that is, a minority race. These types of variations in findings between studies regarding race and ethnic patterns of online aggression could be due to differences in the characteristics of schools and/or geographic locations under investigation. and Intersectionality Research Forms of oppression related to race, class, gender and sexu- ality are interconnected and affect one another. One way to consider the link between multiple identities is through the concept of intersectionality (e.g., Collins, 2000; Crenshaw, 1991; MacKinnon, 2013), which highlights the fact that one of these demographic identities alone is insufficient to gain an understanding of its connection to oppression. For instance, to better explain sexism completely, intersection- ality theory emphasizes that researchers need to acknowl- edge its relationship to racism and socioeconomic status, because the experiences of an individual on a daily basis vary based on distinct intersections of multiple identities (Hurtado, 1996). Studying intersectionality has its challenges methodologi- cally. While theoretically, intersectionality allows research- ers to explore the relationship between several identities, the perspective also may facilitate a misinterpretation of the explanatory theories underlying these differences (Collins, 2015; Davis, 2008). Methodologically, measuring multi- ple identities can be difficult without a resulting additive effect (Bowleg, 2008; West & Fenstermaker, 1995). Fur- thermore, when an additive effect is included, it failed to capture the complexities of inequality related to race, class, gender and sexuality (King, 1988). Thus, studying hostility toward women of color from an intersectional standpoint can be particularly challenging. Because minority women are often oppressed on more dimensions than their white counterparts (Hurtado, 1996), however, such an approach remains an essential endeavor. Hostility Towards Women of Color Hostility toward women is common on Twitter. One study identified 419,000 tweets per day that used one of four insulting terms aimed at women (e.g., “sl*t”) (Felmlee et al., 2019). In comparison to men, women also experience particular forms of online harassment online. For instance, aggressive tweets aimed toward minority women contained social norms that enforced traditional, negative race and gen- der stereotypes (Felmlee et al., 2018a, 2018b). In general, women face more harassment and bombardment than their 13 Race and Social Problems (2022) 14:1–13 thematic content of aggressive, sexist and racist, messages via the social media platform of Twitter. We examine the occurrence of aggressive and harmful Twitter messages directed towards two groups of women of color in the United States, that is, Hispanic/Latinx women and Black women. Next, we search for common themes that emerge within these types of communications, separately for each group. By applying an intersectional lens to understand online aggression towards women of color, we hope to better understand the influence of race and gender on the social media platform. For this research, we ask three questions. First, in what ways are women of color victims of online harassment? Sec- ond, what are common themes used to attack these groups of women? Last, are there distinctions between each group’s victimization and do these distinctions reflect stereotypes? For instance, are Black women likely to experience more victimization related to promiscuity and attractiveness? Additionally, are Latinx women apt to face victimization related to immigration and xenophobia? issue of Methods Search Procedure In collecting the data, we scraped Twitter’s public tweets and collected messages containing two types of search terms. The first terms consisted of aggressive and descrip- tive identifiers for minorities. This set of derogatory terms contained slurs for Blacks and Latinxs. The second type of terms included degrading words toward females. We utilized NodeXL (Smith et al., 2010) to acquire Twitter data directly from the Twitter API using keyword searches (see, for exam- ple, Sterner & Felmlee, 2017). the Data We collected a sample of data from Twitter over a period of two years from late 2015 to early 2017. We amassed over 25,000 tweets and analyzed a wide range of aggressive occurrences on Twitter using keywords searches. We down- loaded the data from the Twitter API, and the final dataset contained the tweet, and the actors who wrote, retweeted, and replied to a tweet, with information obtained from users who publish their tweets publicly. The dataset consisted of tweets that contained the key search terms (see Table 1). The first set of terms targeted minorities (specifically Latinxs and Blacks), some of which were neutral (Black, Latina, Mexican), and others that were offensive, racist terms. We used a common, insulting search term for Black individuals (i.e., “ni**er), since it is one of the most frequently used slurs to degrade individuals by 5 aggressive tweet Total tweets downloaded Time to find aggressive (percent total) tweet from downloaded Female term data (s) B*tch 16,585 (87.78%) 18 C*nt 2308 (12.21%) 16 Wh*re Sl*t 18,893 topics. In comparison to other models, the method can avoid overfitting data, and it utilizes a hierarchical Bayesian analy- sis to discover latent semantic groups in a collection of docu- ments (Blei, 2012). Specifically, LDA is a popular method for topic mod- elling with Twitter data (e.g., Negara et al., 2019; Becker et al., 2011), because it allows the algorithm to treat each document, or in this case each tweet, as a mixture of top- ics, and each topic is a distinct probability of distributions across all words. This approach has the advantage that it allows tweets to overlap each other in terms of content, and to include multiple topics within them, rather than having tweets separated into discrete groups. Various Twitter con- versations may use many of the same insulting sexist and racist terms and themes, for example, and yet address sepa- rate issues. Therefore, the ability of LDA topic modelling to allow overlap in topics was particularly useful for our study. For this analysis, we began by removing unnecessary words from the tweet as a pre-processing tool used to improve the quality of clustering results. The removed words included unimportant ones, such as prepositions or too fre- quently used terms (e.g. the, is, are, and, etc.). Then, we inferred the content of the topic based on the 20 words (or less) for which the probability was highest on the topic. The LDA model needs to be given the number of topics, so we determined the number heuristically. We note that in supple- mental analyses, in addition to frequently used English stop words (i.e., the, it, this, is, etc.), we also removed the most common slurs, and we varied the number of words in each topic. The main findings were unaltered in these additional analyses. Furthermore, we also manually coded tweets to ensure that tweets within each unsupervised, created, the- matic group matched the manually derived categories. and Semantic Networks We used semantic network analysis to provide further details regarding the main topics for both datasets and applied this technique to visually display interconnections between main words in tweets. Semantic network analy- sis represents a method to discover word and meaning 13 Race and Social Problems (2022) 14:1–13 Results First, Accessibility of Negative Tweets We examined the data to investigate the pervasiveness of hostility towards Latinx and Black women. We began by analyzing how quickly we could detect a tweet that was derogatory towards these two groups of women of color in a manual examination of tweet content. On average, it took approximately 18 seconds to detect an insulting, negative tweet directed to Black women, and it took approximately 16 seconds to locate such a message aimed at Latinx women (see Table 1). Overall, aggressive messages toward women of color were easily accessible and visible within the social media platform. Topic Analyses Latinx Women Next, we investigated the data systematically to uncover common words and themes in these Twitter messages. As can be seen in Table 2, three different themes emerged for Latinx women, including stereotyping/racism, media, and immigration. The term, “b**ner”, occured with the high- est word density in three of the clusters. Additionally, in this set of aggressive tweets, “Mexican” was the most men- tioned subgroup of Latinx groups and appears in five of the six clusters. Furthermore, the term with the second highest topic-word density was “Mexican”, suggesting that its use in aggressive messages was very common. Within the Latinx cluster analysis, two of the most frequent derogatory words toward women were “wh*re” and “sl*t”. The first cluster focused on stereotypes surrounding Latinx women (Table 2). The top four terms with highest word densities included “Mexican”, “race”, “wh*re”, and “dumb”. In Table 2, common words related to themes of racism/stereotypes for Latinx women were “wetback”, “n*gger”, “Mexican”, “cleaner”, “greasy”, “abuser”, and description/words used “sucks”, “ignorant”, “terrorist”, “racism”, “cabs”, “kill”, “black lives matter”, ‘slaves”, “job”, “muslim”, “ungrateful”, “americans” “gtf” [good to f!ck], “vanilla”, “wh!res”, “underwear”, “comment”, “beating” “poor”, “smelly”, “natural”, “fierce”, “ebony”, ‘acne” “n*gger”, “mexican”, “players”, “cleaner”, “greasy”, “abuser”, “sketchiest” “western”, “Mexican”, “built”, “moga”, “americans”, “Walmart”, “paisanos”, “guns” “n*ggas”, “evil”, “meek”, “pray”, “poc” 7 such as “looks” and “western”, and “spaghetti”. This clus- ter also highlighted stereotypes surrounding Latinx women, since it contained the term “cantina” [kitchen]. For instance, there was a message attacking a Latinx woman in the fol- lowing manner: “… She looks like a Mexican cantina wh*re extra in a 70 s spaghetti western.” This message targeted a woman using a derogatory, racist and sexist slur (“Mexican … wh*re”) and included references to a “Mexican cantina”, or kitchen. To amplify the attack, the user mentioned that the individual looks like she is in a “70 s spaghetti western,” suggesting that she was outdated in her appearance. Next, we used topic clustering to create a semantic net- work of connections among words in these tweets, based on the top 50 words in the sample of tweets. We visualized the resulting semantic network of words from the tweets. The semantic network contained three distinct subgroups that represented themes focusing on popular media, politics, and an aggressive message containing closely linked racist and sexist words. As shown in Fig. 1, the frequently used, insulting term “beaner” is located in a relatively central part of the seman- tic network (e.g., top right portion) for the Latinx tweets, Looking at some com- where centrality means that a word connects many of the other words in the network. The term “Mexican” also links multiple words in the semantic network, and it is situated in the middle, right section of the graph. These findings suggest that the terms “beaner” and “Mexican” are most often found within the same tweets and very close within the sentence structure of a tweet’s phrase. The derogatory terms toward women, “wh*re” and “sl*t”, also are located near the term “Mexican,” revealing that they, too, were often close together within tweets. In other words, the following 13 Race and Social Problems (2022) 14:1–13 In another subsample, we found terms such as “n*gger” and “muslim,” that were used to attack individuals (see Table 2). Furthermore, certain keywords reinforced negative, African American stereotypes such as “bad,” “fat,” and “ungrate- ful.” Focusing on gender, the most typical, negative insult- ing words for women were “b*tch” “c*nt,” and “sl*t”. In Table 2, frequent words related to themes of racism/stereo- types were “racism”, “stereotypes”, “ignorant”, “slaves”, and “kill.” For instance, one tweet stated: "you are a f*cking disgusting person and I hope you get cancer and die you black paki c*nt". The aggressor calls the woman “disgust- ing” and “c*nt”, in addition to pointing to race and ethnicity (Tables 3 and 4). The next topic theme insinuated promiscuity and hyper- sexuality, with many of the terms related to sexuality. For instance, words such as “horny”, “gtf’, and “vanilla” were used frequently. The terms “horny” and “gtf” also implied hyper sexualizing of Black women and endorsed Black women as promiscuous (see Table 2). The last topic con- tained insults related to attractiveness. Some typical terms included “ugly”, “poor”, “smelly”, and “ebony.” In gen- eral, aggressive tweets appeared to target Black women by employing stereotypes to attack their race, sexuality, and physical appearance (see Table 2). Next, we examined the semantic network results based on tweets collected with the search terms for Black women. We found that the network contained three distinct subgroups that centered on the themes of popular media, politics, and stereotypes (see Fig. 2). The keywords, “black” and “b*tch,” were the two most central terms in the Black semantic net- work, revealing that these words linked many of the other terms in the network. Importantly, the multiple edges between these two words revealed that they often occurred and “racist.” together within the same message, i.e., “black b*tch.” In sl*t" on a game, I don’t feel offended at all. I’m not mexican and I like the word "sl*t", i met at walmart and taco bell URL young girls have babies with differant scum mexican dudes like animals are you? Because there is no angry pussy flare in you. You are probably a loose mouth if every black, white, or Mexican Sl*t in America gets an abortion—I just don’t want to some greasy b**ner to cook my Paella while your at it Mr POTUS!! are you? Because there is no angry pu**y flare in you. You are probably a loose The f*ck do you mean. I’m a Mexican Muslim you c*nt URL 9 c*nt!!! Why do you have to be so racist c*nt?? Hope a black Muslim rapes you Shes a stupid black b*tch who wants people put in place because of skin color… wear a skirt and are inbred? Your most likely a black American inbred or a person and I hope you get cancer and die you black paki c*nt black LMAOOOO GTF URL MAD CAUSE MY MAIN BITCH VANILLA ???? f*cking all the birds black prick didn’t mean to get in ur way ugly c*nt ewww u brag about teams in 2017??UR represented some of the most typical sexist and racist slurs oriented toward women and/or Blacks and Latinx individu- als. Despite our relatively small collection of demeaning keywords, we quickly and easily found numerous aggres- sive messages, confirming that women of color face read- ily available, online hostility. Overall, aggressive messages contained themes that to note that in spite aimed to harm individuals, and these messages and themes often were spread and endorsed by the larger network of online followers. Examining the content of the negative tweets, we identified multiple hostile and humiliating themes that victimize minority women. Common themes relevant to women of Latinx backgrounds included tweets that focused on racist stereotypes, political issues such as “building the wall” between the U.S. and Mexico, and derogatory media references that insulted a woman’s appearance. Popular themes in tweets attacking Black women also involved racism/stereotypes and reflected top- ics of promiscuity and attractiveness slurs. 13 Race and Social Problems (2022) 14:1–13 more online attention than their positive counterparts (e.g., Tsugawa & Ohsaki, 2015), aggressive messages are apt to increase the status of the people who send these tweets in the eyes of their followers and the public, leading to more retweets, “likes,” and online followers. Racist and sexist tweets, therefore, contribute to social hierarchies on the basis of race and gender. These patterns of aggression, and the content used to attack individuals, reflect a similar discussion by Ridgeway (2011), who maintains that fundamental status processes contribute to the construction and maintenance of societal gender inequality. For instance, the messages telling people “to get the wall built,” and have a Latina cook, used negative stereotypes of Latinx immigrants that attempt to maintain status hierarchies and racial distinctions. Here we saw ways in which bullying in social media reinforced the low position of women of color in the dominant, societal hierarchy, and served to maintain inequality located at the intersection of both race/ethnicity and gender. This study was one of the first to examine online harass- ment topics on Twitter in a systematic way, and in doing so there are strengths and limitations to our research. First, we highlighted online harassment that attacked Black and Latinx women, but in doing so, we note that there are many other groups of people who are victimized online, but who were beyond the scope of the current project. Similarly, our focus on only monoracial categories also limited the reach of our study; women of multiracial backgrounds are apt to be included in our dataset. Some Latinx are also Black, for example, and these individuals can identify differently from non-Black Latinx and experience pressure to conform to a single, socially circumscribed race/ethnicity as Latinx/Mexi- can or Black (Jiménez, 2003; Romo, 2011). As dual-minor- ities, the context for their identity construction differs from that of White Latinx, who have a “white” identity option that few Black Latinx can exercise (Jiménez, 2003). Therefore, the ways in which Black Latinx and White Latinx women are harassed online likely differ, and in future studies the topic of multi-racialism remains an important issue to pursue. Likewise, we cannot claim that women of color are more likely to be targets of victimization in comparison to other groups of individuals. Our sample was not random, and it derived from a limited selection of tweets that Twit- ter releases for public downloading. The networks and the tweets we analyzed may be missing certain conversation threads, which could lead to confusion. Furthermore, the interpretation of message content and demographic charac- teristics on Twitter can be subjective and prone to error (due to researcher coding or intentional misinformation supplied by the user). For instance, men have been attacked online based on their sexuality (e.g., homophobic slurs) and by feminizing or emasculating them (Sterner & Felmlee, 2017). Although we failed to find clear evidence of such a trend 11 Another objective of this study is to encourage both indi- vidual internet users and social media platforms to explore additional ways of reducing harmful, online aggression. The findings and methods described herein could be used to aid in detecting certain cases of abuse on social media plat- forms, potentially leading to new algorithms in the removal of tweets based on patterns of word usage. Finally, knowl- edge of negative themes common to online harassment may encourage individuals to report these instances of aggression more often. problem Funding This research was supported by the National Science Foun- dation under IGERT Grant DGE-1144860, Big Data Social Science. Data Availability The datasets generated during and analyzed during the current study were collected via the Twitter Application Program- mer Interface and cannot be shared. They represent third-party data and are restricted by Twitter’s terms of service. However, we provided details of the search parameters used to construct the datasets in the Methods section. Declarations support for Conflict of interest The authors declare that they have no conflict of interest. - References to be aware Andreas, P., & Snyder, T. (Eds.). (2000). The wall around the West: State borders and immigration controls in North America and Europe. Rowman & Littlefield. Bailey, M. (2013). New Terms of Resistance: A Response to Zenzele Isoke. Souls, 15(4), 341–343 Bailey, M. (2016). Misogynoir in medical media: On Caster Semenya and R. Kelly. Catalyst: Feminism, Theory, Technoscience, 2(2), 1–31 Becker, H., Naaman, M., & Gravano, L. (2011). Beyond trending top- ics: Real-world event identification on Twitter. Icwsm, 11(2011), 438–441 Bellmore, A., Calvin, A. J., Xu, J. M., & Zhu, X. (2015). The five W’s of “bullying” on Twitter: Who, what, why, where, and when. Computers in Human Behavior, 44, 305–314 Blei, D. M. (2012). Probabilistic topic models. Communications of ACM, 55(4), 77–84 Bowleg, L. (2008). When Black+ lesbian+ woman≠ Black lesbian woman: The methodological challenges of qualitative and quan- titative intersectionality research. Sex Roles, 59(5–6), 312–325 Bossler, A. M., Holt, T. J., & May, D. C. (2012). Predicting online harassment victimization among a juvenile population. Youth & Society, 44(4), 500–523 Carley, K. M. (1997). Network text analysis: The network position of concepts. Text Analysis for the Social Sciences, 4, 79–100 Chavez, L. R. (2001). Covering immigration: Popular images and the politics of the nation. University of California Press. Chaudhry, I. (2015). #Hashtagging hate: Using Twitter to Track Rac- ism Online. First Monday. https://doi.org/10.5210/fm.v20i2.5450 Chatzakou, D., Kourtellis, N., Blackburn, J., De Cristofaro, E., Strin- ghini, G., & Vakali, A. (2017). Mean birds: Detecting aggression 13 Race and Social Problems (2022) 14:1–13 Grigg, D. W. (2010). Cyber-aggression: Definition and concept of cyberbullying. Journal of Psychologists and Counsellors in Schools, 20(2), 143–156 Harris-Perry, M. (2011). Sister citizen: Shame, stereotypes, and black women in America. (p. 9780300165548). Yale University Press. dilemmas. Annual Hinduja, S., & Patchin, J. (2010). Bullying, cyberbullying, and s. Archives of Suicide Research, 4(3), 206–221. https://doi.org/10. & Negro, P. (2014). 1090/13811118.2010.494133 PMID:20658375. Huang, Y. Y., & Chou, C. (2010). An analysis of multiple factors of cyberbullying among junior high school students in Taiwan. Com- puters in Human Behavior, 26(6), 1581–1590 Hurtado, A. (1996). The color of privilege: Three blasphemies on race and feminism. University of Michigan Press. Jiménez, T. R. (2003). Multiethnic Mexican Americans in demographic and ethnographic perspectives. In M. Coronado, R. P. Guevarra, J. Moniz, & L. F. Szanto (Eds.), Crossing lines: Race and mixed race across the geohistorical divide. (pp. 161–188). University of California, Santa Barbara, Multiethnic Student Outreach. King, D. (1988). Multiple jeopardy, multiple consciousness: The con- text of a Black feminist ideology. Signs, 14(1), 42–72 social structure from Kowalski, R. M., & Limber, S. P. (2013). Psychological, physical, and for network text academic correlates of cyberbullying and traditional bullying. Journal of Adolescent Health, 53(1), S13–S20 Kowalski, R. M., & Giumetti, G. (2014). Wall posts and tweets and blogs, oh my! A look at cyber bullying via social media. Social Networking as a Criminal Enterprise, 8, 91–110 Lenhart, A., Ybarra, M. L., Zickuhr, K., & Price-Feeney, M. (2016). Online harassment, digital abuse, and cyberstalking in America. 7/07/11/online-haras Data & Society Research Institute | Center for Innovative Pub- lic Research. https://www.datasociety.net/pubs/oh/Online_Haras sment_2016.pdf Li, Q. (2006). Cyberbullying in schools: A research of gender differ- ences. School Psychology International, 27(2), 157–170 to traditional MacKinnon, C. A. (2013). Intersectionality as method: A note. Signs: Journal of Women in Culture and Society, 38(4), 1019–1030 Mantilla, K. (2013). Gendertrolling: Misogyny adapts to new media. Feminist Studies, 39(2), 563–570 Miller, G. H., Marquez-Velarde, G., Williams, A. A., & Keith, V. M. (2020). Discrimination and black social media use: Sites of attainment in inter- oppression and expression. Sociology of Race and Ethnicity. https://doi.org/10.1093/oso/9780195160369.001.0001 of social combat: School Monge, P. R., Peter, R., Contractor, N. S., Contractor, P. S., & Noshir, S. (2003). Theories of communication networks. Oxford Univer- oi.o rg/1 0.1 177/ sity Press. National Hispanic Media Coalition. (2012). NHMC Poll Finds that With Friends like these: Negative Media Portrayals are Fueling Stereotypes about Immi- grants and Latinos. Retrieved from https://www.nhmc.org/nhmc- poll-fi nds-t hat-n egati ve-m edia-p ortra yals-a re-f uelin g-s tereo types- about-immigrants-and-latinos/ Negara, E. S., Triadi, D. & Andryani, R. (2019). Topic Modelling Twitter Data with Latent Dirichlet Allocation Method. 2019 Inter- national Conference on Electrical Engineering and Computer Sci- ence (ICECOS), Batam Island, Indonesia, pp. 386–390. https:// doi.org/10.1109/ICECOS47637.2019.8984523 rg/1 0.1 177/0 19027 25166 56585 Nevins, J. (2002). Operation gatekeeper: The rise of the" illegal alien" Sl*ts, Wh*res, and and the making of the US-Mexico boundary. Psychology Press. Pew Research Center. (2012). The 10 Largest Hispanic Origin Groups: Characteristics, Rankings, Top Counties. Retrieved from http:// www.pewinternet.org/ Pew Research Cente.r (2017). Online Harassment 2017. Retrieved from http://www.pewinternet.org/ Pew Research Center. (2021). The State of Online Harassment. Retrieved from https://www.pewresearch.org/inter net/2021/01/ size of minority 13/the-state-of-online-harassment/. Popping, R. (2003). Knowledge graphs and network text analysis. Social Science Information, 42(1), 91–106 13 Taylor, M. C. (1998). How white attitudes vary with the racial composi- tion of local populations: Numbers count. American Sociological Review, 63, 512–535 Tsugawa, S., & Ohsaki, H. (2015). Negative messages spread rapidly and widely on social media. In Proceedings of the 2015 ACM on conference on online social networks (pp. 151–160). U.S. Census Bureau. (2016). U.S. Census Bureau QuickFacts: United States. Retrieved from https://www.census.gov/quickfacts/fact/ table/US/PST045217 slur-mexican-americans- Van der Brug, W., Fennema, M., & Tillie, J. (2000). Anti-immigrant parties in Europe: Ideological or protest vote? European Journal of empathy train- of Political Research, 37(1), 77–102 Wang, J., Iannotti, R. J., & Nansel, T. R. (2009). School bullying among adolescents in the United States: Physical, verbal, rela- (1981). Attitude and atti- tional, and cyber. Journal of Adolescent Health, 45(4), 368–375 Wang, W., Chen, L., Thirunarayan, K., & Sheth, A. P. (2014). Cursing in English on Twitter. Proceedings of the 17th ACM Conference Salmivalli, C. (2009). on Computer Supported Cooperative Work & Social Computing direct goals, (pp. 415–424). Retrieved from http://corescholar.libraries.wright. edu/knoesis.590/ Werner, N. E., Bumpus, M. F., & Rock, D. (2010). Involvement in internet aggression during early adolescence. Journal of Youth Russell, S., & Tip- and Adolescence, 39(6), 607–619 West, C., & Fenstermaker, S. (1995). Doing difference. Gender & Society, 9(1), 8–37 Willard, N. E. (2007). Cyberbullying and cyberthreats: Responding to the challenge of online social aggression, threats, and distress. Research press. Xu, J. M., Jun, K. S., Zhu, X., & Bellmore, A. (2012, June). Learning from the Social Media from bullying traces in social media. In Proceedings of the 2012 conference of the North American chapter of the association for computational linguistics: Human language technologies (pp. Psychosocial risk factors 656–666). Association for Computational Linguistics. Ybarra, M. L., & Mitchell, K. J. (2004). Online aggressor/targets, aggressors, and targets: A comparison of associated youth char- chiatry.2010.79 acteristics. Journal of Child Psychology and Psychiatry, 45(7), networks of cyberbully- 1308–1316 (IJT), 8(2), Publisher’s Note Springer Nature remains neutral with regard to and cyberbullying: A jurisdictional claims in published maps and institutional affiliations. high school. Comput- org/10.1016/j. 13

Analysis of Online Harassment Towards Black and Latinx Women PDF

Document Details

Tags

Related

Summary

Full Transcript

Upgrade to continue