12(23)#21 2023

Creative Commons licenca
This journal is open access and this work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.

DOI 10.46640/imr.12.23.7
UDK 316.647.8(=214.58)
Prethodno priopćenje
Preliminary communication
Primljeno: 12.1.2023.



Lidija Dujić i Dora Mesić

Department of Communication, Media and Journalism
University North, Koprivnica, Croatia
This email address is being protected from spambots. You need JavaScript enabled to view it.

Stereotypical depiction of Roma Population in the Media[173]

Puni tekst: pdf (403 KB), English, Str. 3873 - 3891




By analyzing the media content of the portal, the paper explores the extent to which the reporting on persons belonging to ethnic minorities is based on prejudice and stereotypes. Comparatively analyzed are 5 articles about the disappearance of the girl Barbara Vidović, published in August 2017, and 190 readers’ comments on those articles; and 3 articles about the disappearance of the girl Marija Marinela Grgić in the same month of the same year, and 141 readers’ comments on those articles. The analysis shows that two events of almost the same character are reported differently if the events pertains to members of a minority social group. With such approach to the topic they report on, the media become persecutors of minorities instead of the promoters of their rights and of the better coexistence of minorities with the majority population.


Key words: media coverage, ethnic minorities, prejudice, stereotypical representation of Roma,




In their reporting, the media adjust to the interests of the majority, which can be understood and justified from the market point of view, but in such a world, the voices of minority groups are muted. Content that would engage them or be subject of their interest is often pushed to the margins or absent altogether. In his book Manjine – između javnosti i stvarnosti [Minorities — between the public and reality], Igor Kanižaj analyzes the way ethnic minorities are represented in the Croatian media.[174] The research the book is based on was conducted in several stages and included a total of 600 articles and several print media (Večernji list and Jutarnji list, Slobodna Dalmacija, Novi list and Vjesnik). The results show that minority topics are present predominantly in political sections, while only traces are found in the cultural sections. This is an additional problem because the best results of integration of a minority into the general society are achieved in the cultural sphere. The dominant formats of reporting on minorities are the news and the feature reports, whose primary purpose is to inform. They do not provide deeper insights into the topics, nor offer value judgments, background to the problems, or possible solutions. Kanižaj (2006: 52-53) believes that the absence of investigative journalism as well as the journalists’ passivity may be the reasons. “In reporting on criminal activities, minorities are written about in a discriminatory way, through a sensationalist-discriminatory framework. (…) Journalists, however, testify that they are forced to provide information about members of ethnic minorities because of pressure from editors who believe that such information will increase the sales. Discrimination against minorities in the crime section often comes down to generalization. When a Roma commits a crime, the ethnicity is immediately emphasized, sometimes without disclosing the name of the person at all.” (Kanižaj 2006: 97)



The research part of the paper analyzes the way the Croatian online media report on the events in which the main figures are members of the Roma ethnic group and compares it with the way they report on the events in which the members of the Roma ethnic group do not participate. The subject of the study was the reporting on the disappearance of a 13-year-old girl, Barbara Vidović, in August 2017, involving Roma persons, and the reporting on the disappearance of a 13-year (or 14-year) old girl, Marija Marinela Grgić, in the same month of 2017, in which members of a minority were not involved in any way. Media content of the reporting on both cases on the web portal[175] was analyzed and compared. The aim of the research was to determine the quality of the journalistic coverage in the articles reporting on ethnic minorities on the basis of (non)respect for the standards of professional journalism. The readers’ comments on the published articles are also analyzed in order to reveal the commenters’ interest in the topics, the cultural level of communication, and the differences in the readers’ reactions to the articles in which members of ethnic minorities are mentioned and to those in which they are not. The aim is to determine the extent to which the stereotypes of ethnic minorities, in this particular case of the Roma, are present in the Croatian media, what enables their persistence, dissemination and creation of new ones. The basic subject of analysis was a journalist article on the case of Barbara Vidović, published on the portal, and the readers’ comments on that article (a total of 5 articles published between 08/16/2017 and 08/22/2017 and 190 readers’ comments), and a total of 3 articles published on August 17, 18 and 19, and 141 reader comments, on the case of Marija Marinela Grgić.



Entering keywords “Barbara Vidović” into the VL search facility gives 45 results, but only 5[176] refer to the disappearance of a 13-year-old girl who was in love with a 17-year-old member of the Roma ethnic minority and was suspected of being abducted. The articles are not sorted by the date of publication but by relevance, defined as number of individual ‘reads’ by the portal users. They were published between August 16 and 22. According to the Convention on the Rights of the Child[177], Croatia, as a signatory, is committed to giving priority to the welfare of the child in all situations. No right is above the rights of the child and they must always be protected, especially in media communication, because no public interest is above the rights of the child. For example, the Media Act guarantees the freedom of the media, but also prohibits “the publication of information revealing the identity of a child, if this threatens the welfare of the child”[178]. The Criminal Code protects the privacy of a child under the threat of imprisonment for one year to anyone “who brings or transfers into public view something from a child’s personal or family life, unlawfully posts a child’s photo or reveals the child’s identity, which causes anxiety in the child (…) or harms the welfare of the child in another way”[179]. For the same offense, the Code provides for a two-year prison sentence if the offense is committed “through print, radio, television, computer system or network”[180]. If the perpetrator is a media professional, the penalty is increased by another year. The Code of Honor of the Croatian Journalists’ Association (HND)[181] emphasizes that the freedom of thought and expression is one of the fundamental human rights, and so is the right to accurate and timely information, but not always and not under any circumstances. In Art. 14, the Code prescribes that journalists must protect a person’s intimacy from sensationalist and any other public disclosure, and respect everyone’s right to privacy. It is explained in more detail in the subsequent article, which clearly states the circumstances in which journalists may report and disclose the identities of persons who have, directly or indirectly, suffered misfortune or personal tragedy. This refers to cases where disclosure is in the public interest, but even then, the honor, reputation and dignity of the persons reported on should be first and foremost. The protection of identity and privacy of a child as the first priority, above all else, is mandated by the Code in Articles 19 and 20, which prohibit journalists from photographing or interviewing a child, even with parental consent, or consent of other responsible person, if it would endanger the child’s welfare. This is especially true when reporting on children involved in some form of violence or crime, when it is forbidden to identify a child in the media coverage. The Code is clear: “The welfare of the child is above the public interest”[182]. The legislative system and the professional organizations should be the first to set guidelines for the media professionals and define a framework within which they should operate. The media work is also overseen by some state institutions — in the context of this research, the Ombudsman for Children plays the most important role. The Office of the Ombudsman has the possibility to report cases in which the HND Code of Honor has been violated to the Journalists’ Honor Court. It can also warn the media of irregularities in their reporting, propose legal proceedings against a publisher or an editor-in-chief to the State Attorney’s Office. The Ombudsman is also authorized to issue recommendations. We wish to emphasize one of them — Preporuka o zaštiti privatnosti prilikom medijskog izvještavanja o nestaloj djeci [A Recommendation on the Protection of Privacy in Media Reporting on Missing Children] — in particular. It warns the media that they sometimes, instead of being “a helper in the search for a child” they become “abusers of the child’s rights”[183]. Namely, such stories are often of great public interest, and the media cover them above and beyond reasonable duty, publishing not only information that could help find a child, but also details of the child’s private life. The disclosure of the identity of missing children is justified only until the moment they are found. As the analysis of the reporting on the disappearances of Barbara Vidović and Marija Marinela Grgić shows, journalists continue to report on the case without real justification even after the child is found, revealing unnecessary details from the child’s private life and label him in the environment in which he lives. Only the first article reporting on the disappearance of a girl is justified; the subsequent ones represent a violation of children’s rights under the aforementioned regulations and recommendations. Not only should they never have been published, but according to the Ombudsman’s Recommendation for Children, even the first article should have later been deleted. The problem was not just about reporting on the 13-year-old Barbara Vidović, but also on the 17-year-old Robert Cindrić, a member of the Roma ethnic minority who was equally a statutory minor. The rights to privacy and the protection of dignity and reputation of other members of the boy’s and girl’s family featured in the reports were equally violated. This was not just a violation of regulations, but also a failure to respect the standards of professional journalism that every media professional should abide by. In his book, Osnove novinarstva [Fundamentals of Journalism], Stjepan Malović (2005: 18) defines it simply: “Modern journalism is based on the following professional standards: true, accurate, fair, impartial and balanced reporting.” Just one article studied[184] meets the requirement of truthfulness in terms of documenting the claims made therein (reliability of information sources). The others are based on assumptions made by non-objective sources of information (the girl’s father, whose statements are often qualified with terms “suspicion”, “assumption” and “probably”) or on statements that lack any attribution or stated basis. Two articles are based solely on the statements of the father, one has additional confirmation by the police or unnamed relatives, and only one article features three sources of information. We can say that the claims in that article are substantiated by evidence; in the rest they are based on mere speculation or suspicion. The deficient respect for truthfulness is also reflected in the fact that the first two published articles are not signed by a specific author but only by the portal, while in the last three the authors are stated. Journalist integrity, as a standard that requires all sides of the story to be considered, is not respected in any article studied. In the first two, based on the girl’s father’s suspicions and statements, the standard was not even applicable — at the time, it was not known who the girl had run away with. In the last three articles, the story is discussed unilaterally — either from the side of the girl and her father, or from the side of the boy with whom she ran away and his family. In the third article[185] the author demonstrates inappropriate bias, stating: “Luckily, Barbara’s family was persistent; they searched for the daughter themselves, and later, thanks to giving the story to the media, forced the police into action, too”, as well as “What is really scary about the whole story is the ease with which the kidnappers persuaded the girl to leave her parents”. Yet, there was no confirmation that the girl had indeed been abducted, or that anyone had actually persuaded her to leave. The information was not verified before publication. As for the balance of the articles in terms presenting positive and negative, the standard is not adhered to in three articles[186] — the author insists on the negative side of the story and much of the report is about blaming the two sides or pointing out police failures, implying a possible incestuous relationship, etc. Such is the for instance the statement of an unnamed cousin of a boy who is quoted in the last article: “We are not ashamed to be Roma. They are Roma, too. Barbara’s mother Branka is a native of the Gospić area. Her mother, Milka, is also from there. She was married to Stipe Cindrić, Robert’s uncle, for two years, since she turned 13. Also, Robert’s grandmother Mara is a sister to Branka’s grandfather Jurica. The children are blood relatives.” Would statements like this be published without checking if the subjects were not Roma? Would the journalist not, even if he had confirmation of the incestuous relations among minors, choose to conceal such information in order to protect their dignity, honor and reputation? We can assume that they would show greater sense of duty to protect the rights of the persons they report on had those persons not been members of a discriminated minority group. The stereotypes attached to this population include the widespread belief that incest is common among Roma; it is one of the reasons for the author including such ill-advised statements in the article. No article is of adequate length, if only for the reason that any article published after the girl was found can be deemed inappropriate. The first article, which reports on the girl’s disappearance, gives attention to her father’s suspicions without even asking for public support and concrete help in finding the child. The first and third articles have sensational headlines: “Missing Thirteen-Year-Old Girl, Father Suspects Abduction” and “Barbara Left with a Facebook Friend with Frightening Ease”. One sub-heading can also be considered sensationalist: “The father of the missing 13-year-old says stories are circulating that a minor has persuaded her to marry him”. Three articles are illustrated with photos of the 13-year-old, while the other two[187] feature links to the previously published articles that also contain a photo of Barbara Vidović. The second article already provides information that the young man who abducted the underage Barbara is a Roma. (The assumption of abduction later turned out to be inaccurate; the girl had accompanied him voluntarily.) This information is irrelevant to the report, even if the reporting does continue after all rationale for publication had run out. In the study mentioned above, Kanižaj (2006: 71) stated that “journalists are completely consistent and faithful to the strategy of emphasizing ethnicity and citizenship, even though it is in most cases irrelevant to the offense or the crime, especially while the investigation or court proceedings are still under way and no culpability has been proven yet”. The actual reporting dealt with suspected abduction. The writer quotes the father’s statement that the police did not respond to his reports of abduction of his daughter and that “the suspicious family was checked only perfunctorily; they were Roma in a Roma settlement, so they did not want to get too deeply involved”[188]. The journalist had to be aware of the stereotypes regarding the Roma population (criminals, people who often have problems with the law and who are generally not prosecuted). He thus had to know that highlighting a young man’s ethnicity, mentioning suspicions of incest or implying that a young man’s family was problematic, would increase the citizens’ interest in the topic, multiply the negative comments and thus increase the clickability of the article and affect its position on the portal as well as the presence on the internet. The author reports the girl’s father’s statement that the young man’s family was persuading her to marry him and come to live with them, that Barbara wanted to call father, but was not allowed to do so by the older members of the young man’s family. It is again a matter of disseminating and nurturing the stereotypical portrayal of Roma as a child-marrying population, living in almost tribal conditions with multiple generations of the same family under one roof, where the head of the family lords it over the other members. In the fourth article[189] all those claims are refuted by the protagonist of the story, Robert Cindrić, who points out that the girl brought her backpack with her, with plenty of clothing, proving that she was not abducted. The writer, who deliberately presents the ethnicity of the young man and his family, later builds further articles around the story of the Roma, citing details of the private life of the main actors as well as those who were just collateral victims of the lack of ethics in his reporting. He relies on the negative perception of the majority population towards the Roma. With such reporting, he encourages the creation and maintenance of stereotypes, nurtures prejudice and violates legal regulations and guidelines on the professional journalist work. The primary goal of such reporting is to get a large number of media users to write their opinions below the articles. The transfer of journalism to the internet has benefited everyone — both the readers and the media. It has allowed the readers to be active participants in content creation, not just its consumers. However, the greatest value of the internet for the user is the ability to express their opinion, attitude, disagreement, praise, etc. at any time and on any topic. In this way, the reader directs the media and gives feedback on the value of the content, while the media adapt to the customers’ desires, thus increasing the attractiveness of the content, which then has a positive effect on advertiser revenue. “The ability to comment increases the credibility of the journalists and the media because it enables feedback,” Brautović says in his book, Online novinarstvo [Online Journalism]. “Through comments, the audience shares its viewpoints on the story, their positions, opinions and disagreements. However, the content of the commentary can also be problematic in the sense of being offensive and fomenting hate on religious, ethnic, racial or other grounds. Because of that, news portals often require users to register, to avoid anonymous comments. However, this does not necessarily prevent the user from making offensive comments.” (Brautović 2011: 119-120). VL is linked with Twitter and Facebook at each article via the share button and the like button. The number of comments, the number of views, and the number of times the article was shared are all listed at the beginning of each article. This information makes it easy to deduce the article’s popularity. The portal allows users to comment on articles, but requires registration for that facility. In accordance with the Registration and Account Terms[190] published on the portal page, all information entered during registration must be true (the user is obliged to provide the personal ID number [OIB], first name, surname, address, contact number, e-mail address, etc.). The VL Terms state that the comments prohibit “swearing, insults, put-downs, attacking other users, hate speech, discrimination, defamation, expression of intolerance, all-caps, chatting, off-topic comments, spamming or advertising by putting links in comments or in any other way, copy/paste of full texts from other portals, blogs, etc., quoting offensive comments, or any other comments that are not allowed on the VL portal, using profane or offensive nicknames, or nicknames containing names of well known public persons, criticism of the administrator (…)”[191]. In particular, Article 16 states that “in communication, it is not allowed to use language that discriminates, offends others, especially on the basis of gender, religion, ethnicity, age, region, disability.”[192] At the same time, VL explicitly states that each user is personally responsible for the content they post because the portal cannot control all posts but reserves the right to delete a post it believes to be in violation of the VL Terms and the rules of commenting on the portal. Likewise, VL reserves the right to terminate any account that is contrary to the VL Terms. In this paper, readers’ comments are analyzed in several subcategories — whether they are positive, negative or neutral; to whom they are directed; do they have anything to do with the article itself; whether they offend and (if they do) on what basis; whether they incite others to violence or hatred; do they contain stereotypes. A total of 190 reader comments was found and analyzed.[193] The survey showed that more than half, specifically 108, readers’ comments were negative (56.84 percent) — mostly grossly insulting the persons the reports deal with, other commenters, institutions, or someone else. A total of 51 comments (26.84 percent) were rated neutral, with the commenters expressing opinions but without offensive elements. Only 31 comments (16.32 percent) were rated as positive, with the situation commented with apparent good intent. Most of the negative comments were offensive on the ethnic basis — 55 in total (28.95 percent). Most of those insulted the Roma people (e.g. claimed the Roma were marrying underage, giving birth while still immature, in childhood, demanded virginity as condition for marriage, treated women as property, habitual criminals, thieves, ignorant, backward…). In addition to ethnic insults, gender has also proven to be a fertile ground and inspiration for commenters that did not mince words. Of this kind, 26 (13.68 percent) were found. Gender stereotypes, mainly referring to women, are often mixed with ethnic ones and refer to Roma women, mentioning things like the high birth rate, a subordinate role in the household, lack of education, poor hygiene, etc. “A Roma woman can marry as early as thirteen. When she marries and gives birth, she can no longer find work because of duties to children and family, but also because of unfinished education. (…) Due to the traditionally patriarchal conception of family by fathers, husbands and brothers, Roma women are completely and legally unprotected.” (Rumbak 2003: 113-114) Readers’ comments vividly testify that such an understanding of the lives of individual Roma women is reflected in the overall picture of the Roma women, but women of other ethnic groups are not deprived of gender stereotypes either. A comment can also be found reflecting the stereotypical understanding that fathers are always weak towards their daughters and “blind” concerning their behavior. In addition to gender and ethnic stereotypes, there are also those referring to religion and which unfailingly — 100 percent — represent priests as pedophiles, homosexuals and perverted. The stereotypical portrayal of Roma as perpetrators of minor crimes, thefts, burglaries, etc. is a consequence of poverty, lack of education, poor living conditions and certain other factors. More significant, however, is the fact that such crimes are a priori connected to the Roma. When a felony is committed in the territory inhabited by the Roma, they are considered guilty until proven otherwise. In the eyes of commenters full of prejudice, a young man whose Roma affiliation immediately stands out has abducted or taken a girl with her consent with the intention of marrying her, and has also committed several offenses, such as car theft, but will not be held responsible for any of the above. An interesting thing is that a lot of comments, especially the negative ones, are addressed at other commenters — 42 in total (over 22 percent). The VL Terms mentioned above prohibit off-topic commenting, chat, and expressing abuse towards other users. Even if the comments were not insulting, as they often are here, or promulgating discrimination, the mere fact that the commenters comment on issues that have nothing to do with the article, talk to each other and respond to each other off topic, would be sufficient reason for VL to delete such comments or disable further commenting. However, VL does not take advantage of the remedies it has prescribed itself. In the comments, there is no lack of stereotypical portrayal of state institutions, either, most of all the police and courts as corrupt or idle. They are directly denounced in 18 comments (9.47 percent). The portal, its reporters, or journalist in general, are denounced in 9 comments, some of which are not related to the topic of the article at all, which is also prohibited by the VL Terms. The rest of the articles are mainly focused on the actors of the events themselves, either Barbara and her family, or Robert and his family. Most of those comments are negative and offensive on virtually all grounds, thus violating the Constitution of the Republic of Croatia and the applicable laws that proscribe discrimination, ensure equality and protect every person regardless of whether they belong to a particular minority group. A special question is whether such comments on portals, or communication on social networks in general, constitute hate speech. In the analysis of comments on VL articles conducted for the purposes of this paper, 8 comments (4.21 percent) were found to constitute a crime of public incitement to violence or hatred solely on the basis of ethnicity. It is unclear why the commenters are not sanctioned, if not legally, for which the grounds certainly exist, then at least by deleting the comments and the culpable users’ accounts from the portal. A total of 6 commenters, presented by first and last names[194], posted 10 comments, 4 (40 percent) of which were negative and 6 (60 percent) positive or neutral — they generally expressed the opinion that parents were responsible for children and that they should control them better. Compared to the total number of negative comments (108), the number of non-anonymous negative comments (4) is negligible (3.7 percent). On the other hand, the number of non-anonymous but positive or neutral comments compared to the total is twice as large (6 of 82, or 7.3 percent). The total number of anonymous comments is 180, of which 76 (42.22 percent) are positive or neutral and 104 (57.78 percent) negative. Given that non-anonymous and positive or neutral comments make up 60 percent compared to 42.22 percent of anonymous comments of the same kind, we can conclude that preventing the commenters from hiding their identity behind nicknames would reduce the number of negative comments, especially offensive ones. Anonymity on the portals reduces the self-reflection of commenters, prevents their identification and thus encourages them to express publicly totally unacceptable opinions. It is difficult to understand that such behavior in a public space, which is what VL’s internet space certainly is, is tolerated.



Entering keywords “Marija Marinela Grgić” into the VL search facility gives 16 results. Only two articles[195] are reports on the disappearance of the girl. One states she was 13, another that she was 14. A closer look at the content of the portal yields another article undoubtedly related to this case, despite the fact that it does not mention the girl’s name.[196] Marija Marinela Grgić was found four days after her disappearance was reported, in company of a 44-year-old man whom, according to the latest article, she had met via social networks. The disappearance was reported to the police on the same day. The first article was published on the VL portal three days later and included a photo of the girl. It provided information on her identity, family situation and earlier childhood, specifics of her looks in order to facilitate her recognition — and at the same time appealed to the citizens who had any information to call the phone numbers or contact the addresses listed in the article. Although this represented a violation of a child’s privacy — disclosure of her identity — it was not a violation of the child’s rights because the interest in finding the girl was greater than the interest in protecting her guaranteed rights. However, the fact that this article is still accessible on the portal represents a violation of the rights of the child in the light of the Recommendation of the Ombudsman for Children. Once a child has been found, all such content should be made inaccessible to the public. This also applies to another article, which informs the public that the girl has been found, and provides her photograph; the same one used the day before in an article pleading for help in finding her. Unlike the first one, this photo, as well as the article itself, represent violations of the rights of the child. This is especially true of a sentence in the article that says: “a variety of unverified information has been published on the case, but the true information and all the circumstances of the event will be known after the police investigation”. An article published on August 19, 2017, pertaining to this case without mentioning the name of the child, is a particular problem. Only a careful search of the articles published during and after the disappearance of Marija Marinela Grgić yields the article headlined “ 44-Year-Old, with Whom Missing Girl Was Found, Arrested”. The text does not mention the name of Marija Marinela Grgić, but it can be inferred from the information on the time of the girl’s disappearance, her age, the exact place of disappearance and the time of the finding, provided that news about the girl’s disappearance a day or two earlier was followed. The article would not be in violation of children’s rights had the journalists complied with the regulations protecting children’s rights, especially the part that suggests that all content published on a child’s disappearance be made unavailable when the child is found. Given that it has not been done, it is a gross violation of the rights of the child, further complicating the story of possible pedophilia and/or possible voluntary departure from home and love affair with an older man. We can agree that publication of the announcement that a 44-year-old was arrested on suspicion of criminal offense against a minor is of public interest, but linking it to a specific child is unacceptable. The article states that “police arrested a 44-year-old man and found with him a 14-year-old girl, who had been missing for four days (…) after the girl disappeared from the apartment in Tučepi”. The identification of the child is indisputable, though the article does not carry a picture of Marija Grgić but an illustration of a police car. Special mention should be made of the secondary and repeated violation of the rights of the aforementioned Barbara Vidović and all other actors in the events committed by a series of reports on the subject, especially on the young man she had escaped with, Robert Cindrić. All three articles related to the disappearance of Marija Grgić had links to articles about the disappearance of Barbara Vidović, well after Barbara had been found and the case clarified. The links also included Barbara’s photo. At the same time, in the first published article, on August 17, 2017, after asking for help in the search for the missing Marija Marinela Grgić, the author writes: “Before the disappearance of Marija Marinela Grgić, another 13-year-old, Barbara Vidović, had been missing, but was found in the meantime. Her father, Anton Vidović, announced yesterday that Barbara, who had disappeared in front of a house in Triblje near Crikvenica, was found five days after her disappearance. This was conformed by the police.” Those are certainly similar cases. The girls went missing, both 13-year-olds, in the same month and almost in the same way, leaving homes where they lived with their parents, and both had met the persons they eloped with through social networks. However, these cases are in no way related, and for all the reasons stated above, the links made by the media represent a violation of the rights of the child, the laws of the Republic of Croatia, and ethics in media reporting. Respect for truthfulness as the professional standard of journalism in the context of documenting the claims made, or the reliability of the sources of information, is applied selectively. Police reports, or news published by other portals, are cited, but so is unverified information, such as: “The man allegedly defends himself by claiming to have picked the girl up while driving on the Adriatic Highway, insisting he had never seen her or communicated with her before.” The articles are not signed by the author, but by the portal. We have already cited an example of the lack of respect for accuracy as a journalist standard — the failure to verify basic information, such as the age of the child at the time of disappearance. Honesty as a standard requires all sides of the story to be considered — the first article provides information about the girl’s disappearance and information that could help her find it, the second very briefly reports that the girl was found, and only the third brings a bit more information about the arrested 44-year-old man. The first two articles are justifiably one-sided, do not provide much information, nor should they, while the third article refers to a source, portal 24sata, which was the first to published the information on the arrest, but does not carry any statements the VL journalists have collected independently, nor can it be inferred from the article whether the portal has attempted to contact the competent institutions at all and obtain concrete information. Considering the nature of the articles, we can conclude that the requirement of balance, as a standard that requires an equal representation of the positive and the negative, was respected. Even the last article in line, which implies actions detrimental to the child, offers information (albeit alleged) that the man had never communicated with the girl through social networks before, leaving room for a different turn of events than the one most obvious at the first glance. Only the first article is of appropriate length. The second one cannot be because it should not have been published at all, and the same applies to the third one, which enables the identification of the child. The first two articles were furnished with informative headlines while the latter can be considered clickable, but not sensationalist. The same goes for the subheadings. In this paper, readers’ comments are analyzed in several subcategories — whether they are positive, negative or neutral; to whom they are directed; do they have anything to do with the article itself; whether they offend and (if they do) on what basis; whether they incite others to violence or hatred; do they contain stereotypes. A total of 141 reader comments were found.[197] The survey shows that 60 negative comments (42.55 percent) — mostly offend actors of the reported events, or other commenters, offer negative judgment about the work of state institutions and about education of the missing girl, use inappropriate vocabulary, contain swearing, or express unacceptable views.[198] A total of 56 comments (39.71 percent) were rated neutral — commenters expressed opinions or value judgments that were not insulting or disparaging. There were 25 positive comments (17.74 percent) — extremely well-intentioned, expressing the hope that the child would be found, giving advice or opinion that parents should have greater control over their children and the children’s use social networks, or, simply stating appropriate and civil comments on the situation. A more detailed analysis of the negative comments showed that they were rated as such for insults on grounds of ethnicity or gender and because of their negative focus on other commenters and state institutions. A total of 4 comments (2.84 percent) are offensive on the basis of ethnicity, mostly aimed at the Roma. Given that the disappearance of Barbara Vidović happened several days before the disappearance of Marija Marinela Grgić, and the portal reported on the disappearances in the space of two days, some commenters referred to the Vidović case in comments on the articles about Grgić. In one case, the same commenter offered almost identical comments on articles about the disappearances of both girls. Gender insults were featured in 10 comments (7.1 percent). They insulted and portrayed women stereotypically as using the body as the means to obtain economic security, and being themselves guilty of becoming victims, by leading immoral lives. There is also a mix of gender and ethnic stereotypes in the same comments. The state is censured in 3 comments (2.13 percent) that have nothing to do with the event itself. A high percentage of comments (27, or 19.15 percent) — in the majority of cases negative — are exclusively addressed at other commenters. Commenters insult each other and provoke a domino effect, generating one sinister comment after another. Some comments thus provoked over 15 replies, none of them having anything to do with the event described in the article. A total of 22 comments have nothing to do with the article, but do not address other commenters, either. VL conditions prohibit, among other things, chat and off-topic commenting. Portal administrators have the basis and the ability to delete such comments. They do not do it, just as they fail to act in the cases of insults, discrimination, foul language… While church and faith are an endless source of topics for bickering among commenters on Croatian portals, this topic has clearly not provided a trigger for discussion in that direction — just one comment was found referring to priests, in a negative context. Comments that contain elements of hate speech should be urgently removed, not only in accordance with the VL Terms, but also in accordance with a number of positive Croatian regulations. However, by analyzing the comments on articles about the disappearance/finding of Marija Marinela Grgić, 5 comments (3.55 percent) were found to constitute public incitement to violence or hatred. Interestingly, such comments on the first two articles are directed exclusively at the missing girl. Such comments on the last article, which reports that a middle-aged man who found a 14-year-old girl was arrested, are directed exclusively at that man. They call for disciplining a girl by the use of physical force or mutilating a man with whom the girl was found, who is labeled a pedophile in the comments. One can assume from the nicknames of two commenters that they are their real names, too, but we cannot be sure that those are real identities. The nicknames are jurajurić116 and ivica.grgic58. The latter commenter seems to be the father of the girl. He thanks everyone who contributed in any way to finding his daughter, while also expressing his dissatisfaction with the negative comments. It is the only comment published that can be rated as neutral. User jurajurić116 posted a total of 5 comments, only one of which was rated as negative. Of the total number of comments by non-anonymous commenters (6), only one (16.67 percent) were negative and 5 (83.33 percent) were rated positive or neutral. Compared to the total number of negative comments (60), the number of non-anonymous negative comments (1) is negligible (1.67 percent). On the other hand, the number of non-anonymous but positive or neutral comments (5) is several times higher than the total number of such comments (81) (6.17 percent). The total number of anonymous comments is 135, of which 76 (56.3 percent) are positive or neutral and 59 (43.7 percent) are negative. Given that 83.33 percent of the positive or neutral comments are non-anonymous, compared to 56.3 percent that are anonymous, we can conclude that disabling the hiding of commenters behind different nicknames would reduce the number of negative comments.



The research part of the paper analyzes VL’s reporting on two completely separate cases of missing girls, as well as comments on those reports. In both cases a 13-year-old girl went missing in August 2017, the disappearance was reported to police, and the girl was found unharmed after a few days and returned to her family. As it turned out later, both girls voluntarily eloped with the men they had contacted through social networks. In the Vidović case, the reports initially, completely without justification, promulgated the theory of abduction, which was refuted in the later reports. Barbara Vidović was found with 17-year-old Robert Cindrić, who claimed the two of them were in a love affair. Marija Marinela Grgić was found in the home of an unnamed 44-year-old man. In the latter case, the portal did not indulge in interpreting the nature of their relationship. Although very similar, the two cases were dealt with completely differently in the media, and saw a completely different interaction of the readership. The key difference was the fact that Robert Cindrić, with whom Barbara Vidović had eloped, was a member of the Roma ethnic minority. This is precisely what has determined the manner of reporting and numerous omissions in terms of compliance with the legal norms and the prescribed guidelines for professional journalism. Highlighting this information in the reports has resulted in numerous negative comments offensive on the ethnic basis. In both cases, the portal unjustifiably reported on the disappearance/return of the girls after they had been found, published their photos and details of their lives. More than a year and a half later those articles were still accessible to the public. As already mentioned, this constitutes a violation of children’s rights. However, continued reporting on the two cases is by no means similar. In the case of Grgić, the first article pleads for the help of citizens in finding the child. The second informs the public that the child was found, very sparely and appropriately. The third article cannot be viewed in the same way. It can be related to the specific event if the reporting of the girl’s disappearance was followed closely, or by a careful subsequent reading of all material published at that time. Reporting on possible offenses committed against a child may be justified as being in the public interest, but not in a way that identifies the specific child. Considering the previously published articles on the disappearance of Marija Marinela Grgić, which were not subsequently removed, and the details published in the third article, one can undoubtedly connect one and the other. Yet, the author of the third text did not reveal the identity either of the man or of the girl, did not publish the girl’s picture, nor did he continue to follow the events later on. In comparison with the Vidović case, we can be pleased with the reporting on this one. In the Vidović case, the first article informs about the girl’s disappearance, communicates unsubstantiated suspicions and assumptions of the girl’s father, and discloses irrelevant information. After she was found, 4 more articles were published, and the first of them highlighted the ethnicity of the young man with whom Barbara had eloped. The factually questionable statements of the girl’s father, statements of the relatives of the young man, the neighbors and the main actors of the story come out, refuting the previously stated claims. In addition to highlighting the Roma ethnicity, the writer also gives a stereotypical portrayal of members of the Roma community. Among other things, he relays a statement by an unnamed cousin that Barbara and Robert were in an incestuous relationship, being related by blood. Even if the events did not involve children, no one has the right to make such claims without verification, and even then, it should only be done if there is an overriding public interest in doing so. In this case, such interest did not exist at all. This is not the only stereotypical rendering of the Roma minority, either. The articles repeatedly portray them as criminals and people who get married early by forcing women into marriage. Women are presented as disenfranchised and living in patriarchy. Roma are said to violate laws with impunity, which is allegedly the reason why the police do not want to interfere in the cases in which the Roma are involved. The portal not only continues to cover the events surrounding the disappearance of Barbara Vidović even after any public interest has ceased, but articles about the later disappearance of Marija Marinela Grgić also feature links to the articles on the Vidović case. The writer clearly relies on the negative perception of the Roma held by the majority population and builds a story around the irrelevant ethnicity of the young man, knowing that an article intoned in such a way will attract greater public attention, increase the readership of the articles and boost the interaction of readers with the portal. However, as the analysis of the comments shows, what got increased was not so much the amount of interaction of the readers, but the tenor of their comments. It became unacceptable from the point of view of cultural and communication norms. As for the analysis of the comments in both cases, it shows that they do not differ much in quantity but do differ substantially in content. On the average, the Vidović case provoked 38 comments per article, and the Grgić case 47. The positive and neutral comments in the Vidović case took up 43.16 percent and in the Grgić case 57.45 percent of the total. Negative comments in the Vidović case represented 56.84 percent, and in the Grgić case 42.55 percent of the total. In both cases, the difference between the quantities of positive/neutral and negative comments was thus about 15 percent. Also, in the Grgić case, there was an obvious error in posting one negative comment 8 times in a row, which makes the actual number of negative comments even smaller. However, a more detailed analysis of the negative comments found significant differences in their content, as well as in the cause and motive for their posting. In the case of Vidović, the largest number of negative comments is one that offends on the ethnic basis (as much as 28.95 percent), while in the Grgić case, the quantity is just a tenth of that (only 2.84 percent). It is quite clear that this is a direct consequence of the manner of reporting — the highlighting of the Roma ethnicity of the actors of the reported events, and the stereotypical representation of the Roma population. In general, the cross section of comments on articles published on Croatian portals reflects the situation in the society, the general dissatisfaction and frustration, rather well. In both cases, in addition to comments based on ethnic stereotypes, some were based on gender stereotypes. There is a significant difference between the two cases analyzed in this respect, too. In the Vidović case, 13.68 percent of comments were of this kind, while only half as many (7.1 percent) were made in the Grgić case. This should be surprising because both cases involve women — missing girls — and any possible gender stereotypes should be due to the stereotypical understanding of the woman and her role in general. However, the emphasis on the Roma ethnicity in the Vidović case was sufficient reason for commenters to make a series of discriminatory and extremely offensive comments directed at women in general, and the Roma women in particular. A significant difference in percentage is also evident in the number of comments disparaging and censuring state institutions, especially the police, for their failure to act. In the Vidović case, such comments took up 9.47 percent, and in the Grgić case 2.13 percent of the total. There is generally no significant difference in the number of comments that give a stereotypical portrayal of the church, other commenters, journalists, and the media. Regarding the number of comments in the most malicious group — those which merit qualification as crimes of public incitement to violence or hatred — there is no major difference in the cases analyzed. In the Vidović case, 4.21 percent of such comments were found, in the Grgić case slightly fewer (3.55 percent), but again, they differed in content. All 8 comments representing public incitement to violence or hatred published on the articles about Barbara Vidović’s disappearance, without exception, do so in connection to the Roma, while two in the Grgić case relate to disciplining a girl and three to punishing the man she was found with, who is labeled a pedophile. By separating the comments posted by the commenters using first and last names, and those using nicknames, we have come to the conclusion that there is a significant difference in the categorization of the comments. The number of non-anonymous comments rated negative is a negligible part of the total number of negative comments. Depending on the case, it is 3.7 percent and 1.67 percent. If we consider only the non-anonymous comments, the ratio of positive/neutral versus negative ones is 60:40 percent and 83.33:16.67 percent, depending on the case. On the other hand, when we talk only about anonymous comments, the ratio of positive/neutral versus negative is 42.22:57.78 percent and 56.3:43.7 percent, respectively. The results of the survey indicate that the number of negative comments is lower in the case of commenters who are registered with their first and last names (or we can at least assume that this is the case). Anonymity makes it impossible to publicly identify commenters who are encouraged to voice their frustrations, offend participants in the reported events or other commenters without mincing words, often going as far as doing verbal discrimination, humiliation and even invoking violence. As the portal itself has mandated the VL Terms, which must be adhered to by all parties involved in communication under the threat of deleting comments and (or) accounts, it remains unclear why the portal does not use those powers. The great number of comments or the lack of portal administrators cannot be an excuse.



Discrimination, degradation and prejudice are part of the everyday life of members of minority social groups. The reasons for this are differences, ignorance and rejection by the majority population. As a front line of the presence of the society, the media play a key role in representing minority groups, their history, culture and current position — with a view to better understanding, faster integration, and prevention of conflicts and tensions. Journalism can be seen as the right of every person to be informed. However, instead of being a promoter of basic human rights, that same journalism often becomes their main nemesis. When reporting on events involving members of ethnic minorities, especially topics dealt with in the crime section, the media highlight the ethnicity of persons or a characteristic that separates them from the majority population and no longer views them as individuals, citizens, but as Serbs, Muslims, gays, etc., portraying them through established stereotypes, fostering prejudice and broadening divisions. The result is a distorted perception by the majority population, intolerance, discrimination and marginalization. The Roma population is in a particularly unenviable position. Its culture and way of life are the origin of stereotypes that have been disseminated and maintained for generations. The Guidelines for Reporting on Ethnic Minorities, which can be broadened to include reporting on all minority groups in the society, require that the reporting be affirmative, indicate the problems the minorities face, and make additional efforts to sensitize the majority population to their needs. They particularly prohibit the use of stereotypes, pejorative expressions, the spread of prejudice, and the promotion of discrimination. The conducted research, especially the analysis of readers’ comments on the published articles on the portal, in which there is a clear difference in the reports of events in which they were and in which Roma did not participate, proves that the Guidelines are not respected. This disregard produces a domino effect that removes all barriers to the expression of intolerance by the majority population. In addition to violating the standards of professional journalism to a notable extent, the authors also violate the legal regulations of the Republic of Croatia in relation to basic human rights, protection of children, protection of privacy and protection of the rights of ethnic minorities. The report in question relies entirely on stressing the Roma ethnicity of one of the main actors in the events. The continuation of the story looks as if it was consciously built upon the stereotypical portrayal of that minority population, going beyond the boundaries of what the author would probably have permitted himself were the articles not about the Roma. The best example is the citing of suspicion of incestuous relationship between the two Roma children, uttered by “an unnamed relative”. On the other hand, although the author continues to report on events even when the interest of the public ceases to be relevant, there are no other examples of gross violation of standards of professional journalism. He does not enter into an independent interpretation of events and does not base further reporting on prejudice or stereotypes. The different approach to writing about the topic also results in a different interaction of readers of the portal. The number of comments does not change as much as their nature. It is striking, though not surprising, that the percentage of negative comments that offend, denounce or disseminate ethnically based stereotypes is several times higher when the reporting is about the Roma. There percentage of gender stereotypes doubles in comparison with comments on the reports that do not include minority groups. The research has identified comments containing elements of the crime of public incitement to violence or hatred — in the case of reports dealing with the Roma, such comments refer to the Roma in total, 100-percent Given that the portal has the mechanisms and the legitimate basis for removal of such comments from the website, and even for deletion of the user profiles of the commenters who post such comments, one may wonder why it fails to implement them. It is illusory to expect it to be penalized for such offenses because it obviously does not want to discipline its own journalists. What portals are mostly interested in is numbers — readership, content sharing, and number of interactions. Given that the negative content boosts those parameters much more than the positive, and boosts the clickability of articles, we can reasonably assume that the portal, or its journalists, knowingly produce reports of inappropriate kind in order to pander to the supposed taste of the audience, appeal to the base instincts and the accepted prejudices. In any case, a stronger response from the institutions that can sanction “unruly” media and/or “unruly” media users is wholly absent. A more targeted education of media professionals and an emphasis on the importance of ethical conduct in the professional activities would greatly contribute to the general culture of media communication. So would a strategic turn of the behavior of all state institutions from the negative discrimination of minorities to positive. All of the above is just a reflection of the situation in the society in general — it is not the media that have invented stereotypes, nor have the stereotypes appeared as a result of the readers’ ability to comment on articles. A culture of tolerance of and respect for diversity, a culture of avoidance of generalization about members of any social group, can be a step towards a necessary change in the social climate.


[173]  This work has been fully supported by the University North.

[174]  Before presenting the results of his own research, he states that there is a large number of international studies on the same topic and all of them coincide in two dimensions. “First, the media generally ignore such topics, and second, when they pay attention, the reporting is marked with the following characteristics: 1. Unusual events, such as tragedies, crime, violence, extremism or threats to public order and peace are reported in a negative context. 2. Members of minority groups are presented as different, dangerous and irrational, with the author’s frequent use of prejudice, exaggerated adjectives, generalizations and stereotypes. 3. Members of a minority group are blamed for their own fate and deemed incapable of changing their own reality. 4. Minority-related social, economic and political developments are presented superficially, without serious analysis, or clarification, or the context in which the events take place.” (Kanižaj 2006: 31)

[175]  Hereinafter referred to as “VL”. According to the Gemius Media Tracking Page, is the second most visited portal in Croatia, just behind (

[176]  Two published on August 16, 2017 and one each published on August 18, 19 and 22, 2017

[177]  It is the most important international document regulating the rights of children, dated September 2, 1990. It specifies that a child is considered to be “a human being under 18 years of age, unless the law applicable to the child considers an earlier age to be the majority”. Službeni list SFRJ, Međunarodni ugovori, No. 15/1990, Art. 1.

[178]  Narodne novine, No. 59/04, 84/11, 81/13, Art. 16. p. 1.

[179]  Narodne novine, No. 125/2011, 144/2012, 56/15, 61/15, Art. 178. p. 1.

[180]  Narodne novine, No. 125/2011, 144/2012, 56/15, 61/15, Art. 178. p. 2.

[184]  Posted on 08/19/2017

[185]  Posted on 08/18/2017

[186]  Posted on August 16, 18 and 22, 2017

[187]  Posted on August 19 and 22, 2017

[188]  Posted on 08/16/2017

[189]  Posted on 08/19/2017

[190]  Hereinafter referred to as “VL Terms”.

[192]  ibid.

[193]  11 comments on the first article, 73 on the second, 40 on the third, 63 on the fourth, and 3 on the fifth.

[194]  Although we cannot be sure that they did not choose an invented name: ribaric.tomislav, biserkablagus4, zeljkovameli, jurajuric116, Nemanja Stoisavljenic and petar.visak.

[195]  Posted on August 17 and 18, 2017

[197]  25 comments on the first published article, 60 on the second, and 56 on the third.

[198]  It should be noted that the negative comments include the same comment 8 times in a row, so obviously this was an error in posting comments.



Brautović, M. 2011. Online novinarstvo. Zagreb: Školska knjiga. 

The Declaration on the Rights of Persons Belonging to Ethnic, National, Religious and Linguistic Minorities,, accessible on 04/04/2019, accessible on 04/11/2019, accessible on 04/10/2019, accessible on 04/11/2019, accessible on 04/15/2019

Kanižaj, I. 2006. Manjine – između javnosti i stvarnosti. Nacionalne manjine u dnevnim novinama 2001-2005. Zagreb: Sveučilišna knjižara. 

Kazneni zakon Republike Hrvatske, Narodne novine 125/11, 144/12, 56/15, 61/15, 101/17. Kodeks časti hrvatskih novinara,, accessible on 04/10/2019

Malović, S. 2005. Osnove novinarstva. Zagreb: Golden marketing-Tehnička knjiga. 

Narodne novine, Međunarodni ugovori, No. 12/1993.

Rumbak, I. 2003. Potrebe/problemi romske populacije u Republici Hrvatskoj. Integracija bez asimilacije. Zagreb: Landex. 

Službeni list SFRJ, Međunarodni ugovori, No. 15/1990.

Ustav Republike Hrvatske,, accessible on 04/04/2019.

Ustavni zakon o pravima nacionalnih manjina,, accessible on 04/04/2019

Uvjeti o registraciji i korisničkom računu VL,, accessible on 04/11/2019

Zakon o suzbijanju diskriminacije, Narodne novine 85/08, 112/12.

Zakon o medijima, Narodne novine 59/04, 84/11, 81/13.


Stereotipni prikaz romskog stanovništva u medijima




Analizirajući medijske sadržaje portala, rad istražuje u kojoj je mjeri izvještavanje o pripadnicima nacionalnih manjina utemeljeno na predrasudama i stereotipima. Usporedno je analizirano 5 članaka o nestanku djevojčice Barbare Vidović, objavljenih u kolovozu 2017., te 190 komentara čitatelja na te članke; i 3 članka o nestanku djevojčice Marije Marinele Grgić u istom mjesecu iste godine, te 141 komentar čitatelja na te članke. Analiza pokazuje da se o dva događaja gotovo istog karaktera različito izvještava ako se događaji odnose na pripadnike manjinske društvene skupine. Takvim pristupom temi o kojoj izvještavaju mediji postaju progonitelji manjina umjesto promicatelji njihovih prava i boljeg suživota manjina s većinskim stanovništvom.


Ključne riječi: medijska pokrivenost, nacionalne manjine, predrasude, stereotipno predstavljanje Roma,