Abstract
Although various cases of inappropriate use of anonymous news sources have shaken journalistic practice in the past and encouraged media organizations to act and tighten editorial policies, they remain at the center of journalistic practice. In times of rising demands from media consumers for increased transparency in media content and journalism grounded on ethical principles, the issue of anonymous sources presents an important stage. In our study, we conducted content analysis on four Slovenian daily newspapers which held the largest readership share, during the period from October 1 to December 31, 2021, to explore the extent of the use of anonymous sources of information and how they are employed. In addition, we analyzed the relationship between the use of anonymous sources and various topics of the news stories, the type and sentiment of information. It was found that information from anonymous sources is often of a primary type and neutral in sentiment. According to our research, the sentiment of anonymous information is not related to the content. The connection between the topic of the news story and the type of anonymous information was recognized in the case of reporting on the public sector. Furthermore, articles contained minimal specific descriptions about the sources, providing very limited information about their use. Our study revealed an exceptionally low share of provided explanation on the approval of anonymity for media consumers with very general and generic reasons.
Keywords: Anonymous sources, Anonymity, Unnamed sources, News sources, Media sources
1. Introduction
Various high-profile examples of inappropriate use of anonymous journalistic sources, such as the journalistic “slips” of Janet Cooke [1], the case of Jayson Blair [2], Jack Kelley [3] and others, topped with criticism of their (excessive) use, prompted media organizations to take action and tighten their editorial policies. However, anonymous sources still play a significant role in journalistic practice [[4], [5], [6], [7], [8], [9], [10], [11], [12]].
In order to find how extensive the use of anonymous sources of information is and how they are used, we analyzed the reporting of four Slovenian daily newspapers with the largest share of readership. In our study we used a quantitative research method to focus on determining the frequency of occurrence of anonymous information in the daily press, and the characteristics of published information of anonymous sources, in order to determine how anonymous information is interrelated with the different topics, type and sentiment of information. In this regard, we were particularly interested in what information about included anonymous sources and reasons for granting their anonymity were available to media consumers, so that they could form their own opinions about the justification of their use and decide whether to trust published information.
Our research was prompted by the lack of recent studies on the subject and comprehensive analyses conducted regarding the use of anonymous sources. In addition, we are undertaking research at a time when there are growing demands for greater transparency in media content, and increasingly strong calls for the establishment of new ethical criteria to assist media consumers in making informed judgments about the information presented in the media [[13], [14], [15]]. Simultaneously, the profession was also grappling with a decline in public trust [16].
To maintain their credibility, journalists must fulfill their duty in reporting and using all available journalistic methods fairly and transparently [7]. The use of anonymous sources presents a conflict between two principles. On the one hand, it is the fulfillment of the media's duty to inform the public, and, on the other, the journalistic commitment to do so in a transparent and trustworthy manner [7]. Furthermore, in their daily reporting and decision-making, journalists must balance competing interests on the one hand, and ethical commitments on the other, the authors added. Using anonymous sources in journalism is in conflict with the journalistic standard of disclosing sources, which serves as a tool for verifyng the presented story [17].
Critical views still surround their use, but involving anonymous sources may be the only way for specific information to become public [[18], [19], [20], [21]]. Using information from anonymous sources is especially justified when it comes to information that is in the public interest [22,23]. Journalists often rely on confidential sources, frequently presented as anonymous sources in the media. They often depend upon confidential sources, especially when the story has an “investigative” nature [11]. They are allowing journalists to expose unethical or illegal behavior [24]. Concealment of identity can help protect a source's life, job, defend them from prosecution if the source is involved in criminal activities, protect them from embarrassment, and lead to a more open dialog with the source [21,22]. Osifelo [24] however, argues that the inclusion of anonymous sources is affected by the time pressure under which journalists work. Buttry [25] adds, they are included to beat the competition and get ahead of them by publishing such a story. Carlson [8] argued that they may also be using them to make their story seem more important.
Journalists rely on anonymous sources of information mainly when reporting about government, criminal acts, social problems, financial or economic issues [4], as well as in cases of “soft news” [26], Kratzer and Thorson [5] added, the largest share of such information includes topics about government, foreign relations and internal affairs. Purvis [9] made a similar observation in cases of media reporting on foreign relations, military, foreign affairs and national security. Shapiro [27] also noted that they were common in the reporting about criminal matters. Osifelo [24] noticed a higher share of anonymous sources in controversial or sensitive stories, especially about politics or the economy.
1.1. Journalists and their anonymous sources
Journalists must fulfill their duty and report using all available journalistic methods [7]. Citing sources is an an important journalistic duty that should only be violated in exceptional situations [[7], [28]]. Anonymous sources, however, should only be used as a last resort [11].
Knowing the source of information is essential in understanding the meaning of the message and its relevance [17]. However, media consumers demand complete identification of sources [18,29]. Journalists must provide enough information to the public and enable them with adequate information to judge the information of the source and decide whether to trust the source [30]. This allows them to verify the information presented independently [5,17] and to assess the credibility of the source [31].
An anonymous source may have valid reasons for remaining anonymous, and may provide accurate information. However, relying on anonymous sources can be risky. Journalists need to be aware of the motives of the sources that are sometimes not clear, or sources are hiding bad intentions [18]. Sources may provide certain information that supports their own interests, and may be driven by a desire to get their ideas out to the public, or just simply to get media attention [32]. Sheehy [33] argues, in the delicate relationship between journalists and sources, the latter try to manipulate the media. Some sources are even motivated by hostility or revenge [18]. Fernandez [11] argued that sources could lie intentionally to mislead or manipulate journalists. Journalists may, therefore, be more open to misinformation by including it. When anonymous information is published, it is the journalist who assumes the additional burden of responsibility [34]. Namely, anonymity allows sources to criticize without being held accountable for their words. Approving the confidentiality of sources, including their incomplete and incorrect information, can damage the reputation of journalists and employers [29]. When it turns out that the information was inaccurate, or even false, the credibility of the media is especially at risk [7]. Media reporting assumes that the sources are telling the truth, as this is also demonstrated by their actions, based on of which they have built reputation and trust [7]. Due to the promise of anonymity and the price of credibility, it may happen that journalists might have to support an anonymous source, even when it becomes questionable [8], therefore, they should be used sparingly and carefully [21].
Trust in the presented information of an anonymous source can be supported by the process of verification of information. Research on the Slovenian daily press by Košir and Poler [35] revealed that this duty is often violated. Journalists must be careful when handling unreliable sources to ensure that the reader is not misled [30]. Buttry [25] argued that journalists should avoid using unverified sources. Vobič and Poler Kovačič [23] noted that journalists are aware of the importance and necessity of their verification, while Duffy [36] pointed to the decreasing emphasis on the importance of independent verification of information in literature and ethical standards. In Duffy's study [36] most of the news stories included independently verified information, but in Lewis et all. [37] only half did. Franklin [38] added that, under time pressure and lack of resources, journalists are increasingly abandoning the routine protocols necessary for producing original, verified and accurate news []. Economic, institutional and organizational restrictions hinder their independent functioning [37]. Machill and Beiler [39] found that journalists spend almost half of their time researching, but only a quarter in this process verifying the probability of the correctness of information.
In addition, to achieve better transparency, when it comes to using anonymous sources, media should include explanations of the reasons why sources were not identified [20]. Carlson noted that contributions with anonymous sources usually provide only a loose explanation of why anonymity was granted [40]. The public is, thus, only given a fleeting insight into the relationship between the source and the journalist, and media consumers have no other option but to believe the information [40]. Sternadori and Thorson [6] believed that the absence of such explanations was a sign of journalistic incompetence. The share of anonymous statements with an explanation of the reasons for granting anonymity is generally relatively low. Duffy's study [36] showed that the share rose from 10 % (before 2008) to “only” 25 % (after 2008).
Carlson [40] also pointed out the importance of the descriptor of anonymous sources, which, in practice, are often only marked with vague identifiers. Anonymous sources are often described with general labels (e.g. “officials”, “people close to officials”. In the case of Wulfemeyer's [21] study, the most frequently used descriptors to describe anonymous sources were “officials”, “sources”, “experts” and “assistants”. Such generic labels usually do not include specific information about the sources [9], but often indicate the source's occupation or other connections.
2. Materials and methods
In order to find out how extensive the use of anonymous sources of information is and how they are used in the daily press of Slovenian daily newspapers, we analyzed the reporting of four daily newspapers with the largest share of readership, i.e. Večer, Delo, Slovenske novice and Dnevnik. An analysis of the readership of print media [41] showed that Delo (9.4 %), Slovenske novice (9.3 %), Večer (6.3 %) and Dnevnik (6.1 %) are the most read newspapers that are published on weekdays. In accordance with the size of the country, the Slovenian media market is small and characterized by a high concentration of media ownership [42]. The analyzed newspapers are typically classified as quality press, with the newspaper Slovenske novice often described as having a tabloid approach. Their content differs in language, selection of events, content, used visual material, journalistic comments, etc [43].
In the study, we used a quantitative research method, where we focused on determining the frequency of occurrence of anonymous information in the daily press and the characteristics of anonymous information, in order to determine how anonymous information is interrelated with different topics, type and sentiment of information.
For this purpose, we defined research questions.
RQ1
What is the prevalence of specific types of information within statements of anonymous media sources?
Information of anonymous sources were categorized by their type, where we modified the type of knowledge in the news, according to Ericson [44]: primary (factual; asking What happened?), secondary (explanatory; asking Why did it happen?), tertiary (descriptive; asking What was it like to be involved in what happened?), evaluative (moral; asking Was what happened good or bad?), and recommendation (asking What should be done about what happened?).
RQ2
Which sentiment dominates in daily reporting within statements of anonymous media sources?
The presentation of information in the media can influence how media consumers perceive results significantly, while their understanding often depends on the presented content [44]. Consequently, the way news media convey information, can shape public perception. In this context, the public interpretation of news medias’ portrayal of issues can also be influenced by the tone of the news 45. The tone, also known as sentiment or favorability (Macnamara 2018), employed by the media, is an additional element of media framing that has the power to influence media consumers and directs their thoughts in a certain way regarding a given issue. Moreover, it attaches emotional attributes [46]. According to Saxer and [47] emotional communication stands as one of the primary strategies employed by journalists.
Therefore, we extended the existing findings in this area by analyzing the interconnectedness of concepts, topics, type and sentiment of information. In the following we wanted to verify the interrelation between sentiment, the type of information and the topic of the news story.
RQ3
How does the sentiment used in statements of anonymous sources’ information relate to the type of anonymous information?
RQ4
How is the sentiment used in statements of anonymous sources related to the topic of the news item?
RQ5
Does the topic addressed in the analyzed article, which includes information of anonymous sources, relate to the type of anonymous information?
Journalists are obliged to provide media consumers with enough information to enable them to assess the sources’ credibility [30] and verify the presented information independently [5,17]. Therefore, the analysis included a review of included information on the verification of anonymous statements in the articles. Next, we were interested in stated explanations available in news stories to readers about the reasons for granting anonymity to their sources. Both pieces of information are essential for media consumers when they are deciding whether to believe and trust the presented information.
RQ6
What explanations about the reasons for granted anonymity do the journalists provide through news stories to media consumers?
Within the descriptive quantitave study and using a content analysis, a total of 260 printed copies of newspapers were examined, published during the weeks between October 1 and December 31, 2021. We analyzed 65 copies of each newspaper (Delo, Dnevnik, Večer, Slovenske novice). We chose exlusively editions issued from Monday to Friday for our selection, because there is a substantial divergence in content between weekday and weekend editions, making it unfeasible to compare the sections effectively. Only news stories under the central, main, leading section of the newspaper with current news from different themes, were selected, due to the comparability of the content and categories.
The basic unit of analysis was a news story, a subject of analysis to identify anonymous source information. Comments and opinions, agency announcements, editorials, promotional content, content such as weather forecasts, and sports news were not included in the analysis. To code news stories, two coders were selected as coders. To ensure inter-coder reliability, both were given a training in the rules of coding. Inter-coder reliability test was performed by selecting and double coding 14 % of the sampled news stories. Inter-coder reliability was examined using Krippendorff's Alpha, where the results showed very good agreement (from α = 0,872 for tone, α = 0,919 for type of information and α = 0,901 for information verification).
First, we analyzed the stories in each issue of the newspaper that contained anonymous sources manually. In 260 copies, we analyzed 3523 news stories in all four newspapers (see Table 1). 694 were Agency news, 333 were all citations from anonymous sources in 156 statements, which is 6.8 % of all (see Table 4).
Table 1.
Collected newspaper issues, analyzed articles and articles with anonymous information.
| Newspaper | Analyzed newspaper issues | Analyzed articles | Analyzed articles with anonymous information | Share of articles with anonymous information |
|---|---|---|---|---|
| Delo | 64 | 961 | 51 | 5,3 % |
| Dnevnik | 64 | 323 | 33 | 10,2 % |
| Slovenske novice | 64 | 585 | 42 | 7,2 % |
| Večer | 64 | 437 | 30 | 6,9 % |
| Skupaj | 256 | 2.306 | 156 | 6,8 % |
Table 4.
The number of all posts with an anonymous source of information and the number of posts with verified sources of anonymous information.
| Number of all news stories with an anonymous source of information |
News stories with verified sources of anonymous information |
||
|---|---|---|---|
| F | frequency | % | |
| Delo | 51 | 9 | 17,6 % |
| Dnevnik | 33 | 6 | 18,2 % |
| Slovenske novice | 42 | 11 | 26,2 % |
| Večer | 30 | 6 | 20,0 % |
| Total | 156 | 32 | 20,5 % |
An extraction form was prepared to facilitate the data. We defined categories: newspaper title, frequency, depth of coverage, and the issue date. We conducted a thorough manual analysis of the news stories, to identify and extract statements containing anonymous information. Only references within news stories that indicated the use of anonymous information clearly within the use of specific words and phrases were considered. When in doubt, the coders needed to reach an agreement whether to include the quotes. Next, each news story with included anonymous information underwent additional coding, where elements such as title, author, date of publication, dominant theme of the article, regardless of the actual section in the newspaper were determined (Politics, Police, Health Economy, Judiciary, Public Sector). We counted the number of different anonymous statements, stated the exact use of words, anonymous source labels, information regarding the stated reasons for granting anonymity, and verified information of the anonymous source. We counted the number of sources that confirmed the information of anonymous sources and determined the tone and type of information.
To assess the tone, we employed the categories positive, neutral, and negative. Each phrase was assessed individually to determine the impression. In cases of balanced presentation of a favorable and unfavorable evaluation of lack of clear direction, the phrase was categorized as neutral.
We conducted a data analysis using quantitative statistical methods, used to identify the existence of specific words, themes, and concepts. According to Macnamara [48] the method relies primarily on human coding of text into categories such as issue, source or message, and involves making a series of judgments regarding text characterics based on predetermined parameters. Manual coding requires careful analysis, because heavy reliance on human interpretation is prone to increased error, particularly when employing relational analysis to achieve a more advanced level of interpretation, as was the case in our study.
We calculated both actual and relative frequencies, to illustrate the occurrence of each studied element. To calculate statistically significant differences in dependent variables, we employed the likelihood ratio test. The null hypothesis was tested at the level of p = 0.05. Differences where the risk of rejecting the null hypothesis was less than 0.05 were considered statistically significant. Data analysis was carried out using the IBM SPSS 25.0.
3. Results
Out of a total of 2306 news stories, 156 (6.8 %) contained an anonymous source. The share varied slightly from one newspaper issue to another – the largest share of them containing anonymous sources was found in Dnevnik (10,2 %), while the lowest number of stories with anonymous sources was found in Delo (5,3 %).
During the observed period, 228 statements with anonymous pieces of information appeared in a total of 156 news stories in the analyzed newspaper issues, or an average of 1.46 pieces of anonymous information per one issue (SD = 0.95).
Almost three quarters of the analyzed articles that contained anonymous information included one anonymous piece of information each. Just under a fifth of them contained two pieces of anonymous information each.
The labels of anonymous sources were mainly generic, for an example » unofficial«, »unofficial data«, »our data«, »out data«, »unofficial data«, »townsmen, villagers, citizens«, »neighbors«, »our source«, »interlocutor«.
Depending on the individual topics, anonymous sources were most often present in news stories dealing with political topics – as much as half of all containing an anonymous source addressed this topic. Approximately one in seven news stories with anonymous sources dealt with the field of Healthcare, and one in eight with the economy (see Table 2). The rest of the topics were represented in news stories with an anonymous source in less than 10 %. A high degree of co-occurrence was observed in the topics of economy and politics, as shown in Table 3, where just under two-thirds of articles on the topic of the economy also discussed the topic of politics (but only just under 17 % on the topic of politics discussed the topic of the economy). News stories dealing with the topic of the police most often (one out of five) also dealt with the topic of politics. One out of four news stories from the field of Healthcare also dealt with the topic of politics or the police.
Table 2.
Number of news stories containing an anonymous source, by topic (above 5 %).
| frequency | % | |
|---|---|---|
| politics | 77 | 49,4 % |
| police | 58 | 37,2 % |
| health | 23 | 14,7 % |
| economy | 20 | 12,8 % |
| judiciary | 15 | 9,6 % |
| public sector | 14 | 9,0 % |
Table 3.
Share of news stories, dealing with a single topic, by topic.
| n | politics | police | health | economy | judiciary | public sector | |
|---|---|---|---|---|---|---|---|
| politics | 77 | 14,3 % | 7,8 % | 16,9 % | 5,2 % | 6,5 % | |
| police | 58 | 19,0 % | 10,3 % | 0,0 % | 6,9 % | 0,0 % | |
| health | 23 | 26,1 % | 26,1 % | 4,3 % | 4,3 % | 13,0 % | |
| economy | 20 | 65,0 % | 0,0 % | 5,0 % | 0,0 % | 0,0 % | |
| judiciary | 15 | 26,7 % | 26,7 % | 6,7 % | 0,0 % | 0,0 % | |
| public sector | 14 | 35,7 % | 0,0 % | 21,4 % | 0,0 % | 0,0 % |
Note: reading of the Table is linear, for example: 14.3 % of the news items that dealt with the topic of politics also dealt with the topic of policing.
The share of statements with an explanation on the approval of anonymity was extremely low (n = 2), including mainly general reasons (»they don't want to be exposed«, »they don't want to be named«).
Half of the anonymous information that appeared in the analyzed news items included primary type (example: “According to our information, the accused has already been sentenced to probation”), a quarter of secondary (“unofficially, the suspect is believed to have mental health issues”), and a fifth of tertiary (“according to unofficial information, police investigators even found a sheet of instructions at his home on how to act during a house search”) type of knowledge. To a small extent, evaluation (“Sad, as one of the neighbors described the tragedy”) and recommendation of anonymous information (“according to our information, subsequent changes to the received decision will strain relations further”) can be found in the news stories.
In the following, we assigned a sentiment to the anonymous information, which was marked as either positive, negative or neutral. As shown in Table 5, most, more than half of the anonymous information was of neutral sentiment (for example “according to our information, they talked about cooperation”), and two-fifths of it was negative (“according to unofficial information the limitation will harm companies”). An extremely low proportion of anonymous information was positive (for example “according to some information, representatives promised they would receive the documents shortly”).
Table 5.
Numbers of anonymous information in stories from the analyzed section at the overall level, by type of anonymous information and sentiment of the content of anonymous information.
| frequency | % | ||
|---|---|---|---|
| No. of anonymous information in the news story (in the analyzed section) | |||
| total | 228 | 100,0 % | |
| By type of anonymous information | |||
| primary | 116 | 50,9 % | |
| secondary | 55 | 24,1 % | |
| tertiary | 46 | 20,2 % | |
| evaluation | 8 | 3,5 % | |
| recommendation | 3 | 1,3 % | |
| By the sentiment of anonymous information | |||
| positive | 10 | 4,4 % | |
| neutral | 128 | 56,1 % | |
| negative | 90 | 39,5 % | |
In the following we verified whether the topics were related to the type of anonymous information. The connection between the topics and the type of anonymous information appearing in the news stories was assessed using the likelihood ratio test. We observed that, only in the case of stories dealing with the public sector, were there statistically significant differences (at the p < 0.05 level) in what types of anonymous information they contained compared to stories that do not deal with the topic public sector. In news stories dealing with the public sector, we found significantly more primary anonymous information than in the news that did not deal with this topic, while the representation of the remaining types of anonymous information in the news stories on the topic of the public sector was smaller.
The topic of the police was also on the borderline of statistical characteristics (p = 0.085). News stories dealing with the topic of policing and/or crime contained less primary and more tertiary anonymous information compared to news not dealing with the topic of policing.
In order to determine whether the sentiment was related to the type of anonymous information (see Table 8), we performed a likelihood ratio test to test the association between the sentiment of the content of the anonymous information and the type of anonymous information. We did not find the existence of statistically significant differences in the type of anonymous information in relation to the sentiment of the content of the anonymous information (LR = 14,006; df = 8; α = 0,082), but, nevertheless, a trend towards a greater presence of secondary anonymous information was indicated when the sentiment of the content of the anonymous information was positive, while news stories containing anonymous information with neutral sentiment were dominated by primary anonymous information (see Table 6).
Table 8.
A test of differences in anonymous information sentiment by topic.
| THEME | n | SENTIMENT |
Relative belief ratios (df = 2) |
|||
|---|---|---|---|---|---|---|
| positive | neutral | negative | LR | Α | ||
| Politics | 98 | 2,0 % | 61,2 % | 36,7 % | 3474 | 0,176 |
| Police | 76 | 3,9 % | 47,4 % | 48,7 % | 4029 | 0,133 |
| Health | 31 | 0,0 % | 61,3 % | 38,7 % | 3093 | 0,213 |
| Economy | 28 | 7,1 % | 50,0 % | 42,9 % | 0,789 | 0674 |
| Judiciary | 19 | 0,0 % | 57,9 % | 42,1 % | 1787 | 0,409 |
| Public sector | 16 | 0,0 % | 56,3 % | 43,8 % | 1532 | 0,465 |
Table 6.
Numbers of anonymous information from the analyzed section at the overall level, by type of anonymous information and sentiment of the content of anonymous information.
| frequency | % | |
|---|---|---|
| Nos. of anonymous information in the news stories (in the analyzed section) | ||
| total | 228 | 100,0 % |
| By type of anonymous information | ||
| primary | 116 | 50,9 % |
| secondary | 55 | 24,1 % |
| tertiary | 46 | 20,2 % |
| evaluation | 8 | 3,5 % |
| recommendation | 3 | 1,3 % |
| By the sentiment of anonymous information | ||
| positive | 10 | 4,4 % |
| neutral | 128 | 56,1 % |
| negative | 90 | 39,5 % |
The connection between the topics and the sentiment of the content of the anonymous information appearing in the articles was verified using a likelihood ratio test. We found that the sentiment of the anonymous information was similar in the news, regardless of the topic they discussed. We did not find any statistically significant differences in the occurrence of individual sentiments of anonymous information according to the topics of the news (Table 7). We can conclude that the sentiment of anonymous information does not appear to be related to the content of the news stories.
Table 7.
A test of differences in the type of anonymous information according to the sentiment of the content of the anonymous information.
| SENTIMENT | n | TYPE OF ANONYMOUS INFORMATION |
||||
|---|---|---|---|---|---|---|
| primary | secondary | tertiary | recommendation | evaluation | ||
| positive | 10 | 30,0 % | 50,0 % | 20,0 % | 0,0 % | 0,0 % |
| neutral | 128 | 56,3 % | 20,3 % | 21,1 % | 1,6 % | 0,8 % |
| negative | 90 | 45,6 % | 26,7 % | 18,9 % | 1,1 % | 7,8 % |
4. Discussion
One of the central contemporary journalism challenges is the question of what ethics should guide modern news production [14]. At the forefront of discussions are also transparency and anonymity [40], with the highlighted desire of media consumers and their need for transparent reporting and journalism based on ethical decision-making [13,15]). The question of transparent media reporting based on ethical foundations also applies to the use of anonymous journalistic sources. The share of sources that require their identities hidden in the media has decreased significantlyin recent decades, although they remain at the core of journalism practice [[4], [5], [6], [7],10,40]. Relying on such sources carries risks. Furthermore, with approved anonymity, the actual power of actors in society, their expertise, ideological positions, goals, and their motives, are hidden to the public [49], leading to compromised credibility of the media [7].
In the Slovenian print media landscape journalists do not rely on anonymous sources frequently. The analyzed percentage of anonymous sources used was relatively low, and consistent with the findings of recent studies in this area (see Osifelo [24]). Not only the inclusion of anonymous sources, but also the manner in which they are included, is crucial. Fernandez 11 argued that readers should be provided with as much information as possible about an identified source. However, the journalists in our study did not provide media consumers with a substantial amount of information which would facilitate the public in their assessments. The key information that assists them in this regard – the used anonymous source descriptors were very general. Phrases provided only minimal information about resources, their background, function, work, and other details. By using descriptive adjectives, journalists aimed to emphasize that these sources were either firsthand, or derived from individuals close to the organizations they were reporting on. The problematic aspect is the use of verbs and other phrases that indicated the utilization of anonymous information, but do not express it clearly (for example: “we found out,” “according to our information”), and readers could potentially interpret these phrases wrongly. A possible reason for this might be the limited and confined Slovenian media space, where additional source labels could pose a risk of revealing the source's identity. In this regard we agree with Fernandez [,11, who argued that such “passive” approach in identifying sources diverts attention from the identity of the source. In general, news stories may include a specific reference to the source's anonymity, often indicating why the source has been granted anonymity, such as “the sensitivity of the matter”, “the fact that the source is not authorized to provide information”, etc. [9,21]. In our study, texts very rarely included the provided explanations for granting anonymity, which gives the impression that journalists do not particularly want to highlight the use of anonymous sources. This finding deviates negatively from the findings of similar studies in this area, where such a percentage was indeed much higher, but, in most cases, still inconclusive (see Purvis [9]).
The analyzed anonymous sources stood out according to the dominant topic, with politics leading the way. A good third of the analyzed statements which contained an anonymous source dealt with the police, or they were connected with crime as the main topic. As Cheung 50 wrote, police sources are among the most influential official sources. At the same time, anonymous sources are used more often when the source of information is limited [5], and this principle applies to reporting on criminal matters. The fact that Healthcare followed in third place in terms of frequency was most probably also contributed to by the pandemic period and increased attention to news about Covid-19. The fact that the media reporting analysis was conducted during the pandemic, is also one of the significant limitations of this study, as media coverage at that time was focused heavily on Healthcare.
The connection between the topic and the type of anonymous information was recognized in the case of reporting on the public sector. In the news dealing with the topic of the public sector, we found significantly more primary anonymous information in the news that did not deal with this topic, while the representation of the remaining types of anonymous information in the news on the topic of the public sector was smaller. Given that the articles in this regard related primarily to the decision-making processes of employees in public sector and also to personnel employment in the public sector, this can suggest journalists have a harder time obtaining official explanations about fundamental information concerning the broader public sector, or that anonymous sources willing to provide information about the public sector are willing to comment only on primary and factual information. For a more detailed analysis of the use of types of information in relation to anonymous journalistic sources, it would be beneficial to conduct a longitudinal study of media content, which would demonstrate more clearly how specific topics are related to the particular types of information. For instance, articles in the field of Politics often pertain to political decision and searching information on background of these decisions, while, in the realm of crime reporting, journalists are frequently seeking witnesses to criminal acts and sources who could reveal the circumstances surrounding these events. In the case of reporting on criminal matters, in that sense it is no surprise that the topic of the police was also on the borderline of statistical characteristics. News stories dealing with the topic of policing and/or crime contained ess primary and more tertiary anonymous information compared to news stories not dealing with the topic of policing. As reporting of police work is usually about criminal acts, this is to be expected, as tertiary information provides details about events, often provided by victims, witnesses, neighbors.
Our study revealed a shift towards neutrality in tone when anonymous sources were used, with a close presence of a negative tone. Tone is an important element of media framing that holds the power to influence media consumers, and directs their thoughts in a certain way regarding a given issue. Our data suggested a shift towards neutrality when utilizing anonymous sources. Although the existence of statistically significant differences in the type of anonymous information according to the sentiment of the content was not established, a trend of higher appearance of secondary anonymous information when the sentiment of the content of the anonymous information was positive, was detected, while the texts containing anonymous information with a neutral sentiment had predominantly primary anonymous information. This could be explained by the fact that the secondary type of information allows the sources more subjectivity and details, that bring a positive sentiment.
5. Conclusion
Studying the structure of journalistic sources, their relations and activities is crucial for comprehending media content. Furthermore, the variety of sources and their characteristics serve as an indicator of journalistic quality [51]. The manner in which a journalist has obtained information and the types of sources they have chosen plays a pivotal role in shaping the information that reaches the public and the portrayal of society [52].
By analyzing the reporting in four Slovenian daily newspapers, we examined the prevalence of the use of anonymous sources and their utilization in the daily reporting. To deepen the understanding of their utilization, we were interested in whether their use was associated with specific concepts, such as sentiment, types of information and themes. Although the share of used anonymous sources is relatively low, journalists could ensure the transparency of their usage better. Media consumers could benefit from additional information about their identity and explanations for granting anonymity. It was found that information from anonymous sources was often of a primary type and neutral in sentiment, leading to the conclusion journalists are using them to obtain critical information while striving for balanced reporting. According to our research, the sentiment of anonymous information is not related to the content. The connection between the topic of the news story and the type of anonymous information was recognized in the case of reporting on the public sector.
Data availability statement
Data associated with this study has not been deposited into a publicly available repository. Data will be made available on request.
CRediT authorship contribution statement
Anita Kovačič Čelofiga: Visualization, Validation, Supervision, Resources, Project administration, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Tina Tomažič: Visualization, Validation, Supervision, Resources, Methodology, Investigation, Formal analysis, Data curation, Conceptualization.
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper
References
- 1.Kurtz H. Washington Post; 1996. Janet Cook's Untold Story.https://www.washingtonpost.com/archive/lifestyle/1996/05/09/janet-cookes-untold-story/23151d68-3abd-449a-a053-d72793939d85/ [Google Scholar]
- 2.Barry D., Barstow D., Glater D.J., Liptak A., Steinberg J. New York Times; 2003. Correcting the Record; Times Reporter Who Resigned Leaves Long Trail of Deception.https://www.nytimes.com/2003/05/11/us/correcting-the-record-times-reporter-who-resigned-leaves-long-trail-of-deception.html [Google Scholar]
- 3.Morrison B. USA Today; 2004. Ex-USA TODAY Reporter Faked Major Stories.https://usatoday30.usatoday.com/news/2004-03-18-2004-03-18_kelleymain_x.htm [Google Scholar]
- 4.Weisenhaus D. Newsgathering practices: Hong Kong journalists' views and use of controversial techniques. Global Media J. 2005;4 [Google Scholar]
- 5.Kratzer R.M., Thorson E. Use of anonymous sources declines in U.S. Newspapers. Newspaper Res. J. 2007;28:56–70. [Google Scholar]
- 6.Sternadori M.M., Thorson E. E., anonymous sources harm credibility of all stories. Newspaper Res. J. 2009;30 [Google Scholar]
- 7.Duffy J.M., Freeman C.P. Anonymous sources: a utilitarian exploration of their justification and guidelines for limited use. J. Mass Media Ethics. 2011;26:297–315. [Google Scholar]
- 8.Carlson M. University of Illinois Press; Oxfordshire: 2012. On the Condition of Anonymity. [Google Scholar]
- 9.Purvis H. Anonymous sources: more or less and why and where? Southwestern Mass Communication Journal. 2015;30 [Google Scholar]
- 10.Gottfried J., Walker M. Pew Research Center; 2020. Most Americans See a Place for Anonymous Sources in News Stories but Not All the Time.https://www.pewresearch.org/fact-tank/2020/10/09/most-americans-see-a-place-for-anonymous-sources-in-news-stories-but-not-all-the-time [Google Scholar]
- 11.Fernandez J.M. Routledge; 2021. Journalists and Confidential Sources. [Google Scholar]
- 12.Gottfried J., Walker M. Pew Research Center; 2020. Most Americans See a Place for Anonymous Sources in News Stories but Not All the Time.https://www.pewresearch.org/fact-tank/2020/10/09/most-americans-see-a-place-for-anonymous-sources-in-news-stories-but-not-all-the-time Retrieved August 12 2020: [Google Scholar]
- 13.Auman A., Stos S., Susan, Burch E. Ethics without borders in a digital age. Journal. Mass Commun. Educat. 2020;27 [Google Scholar]
- 14.McBride K., Rosenstiel T.B. Poynter Institute; 2014. The New Ethics of Journalism: Principles for the 21st Century. [Google Scholar]
- 15.Sullivan M. NY Times; 2013. The Disconnect on Anonymous Sources.https://www.nytimes.com/2013/10/13/opinion/sunday/the-public-editor-the-disconnect-on-anonymous-sources.html [Google Scholar]
- 16.Newman N., Fletcher R., Shulz A., Nielsen R.K. Reuters Institute; 2020. Digital News Report 2020.https://reutersinstitute.politics.ox.ac.uk/sites/default/files/2020-06/DNR_2020_FINAL.pdf [Google Scholar]
- 17.Boeyink E.D. Anonymous sources in news stories: justifying exceptions and limiting abuses. J. Mass Media Ethics. 1990;5:233–246. [Google Scholar]
- 18.Poler M.K. Magnolija; Ljubljana: 1997. Novinarska Etika. [Google Scholar]
- 19.Franklin B., Hogan M., Quentin L., Mosdell N., Pill P. Sage Publications; 2009. P. Key Concepts in Public Relations. [Google Scholar]
- 20.Farhi P. 2013. Anonymous Sources Are Increasing in News Stories, along with rather Curious Explanations.https://www.washingtonpost.com/lifestyle/style/anonymous-sources-are-increasing-in-news-stories-along-with-rather-curious-explanations/2013/12/15/5049a11e-61ec-11e3-94ad-004fefa61ee6_story.html [Google Scholar]
- 21.Wulfemeyer K.T. L.M, anonymous attribution in net- work news. Journal. Q. 1986;63:468–473. [Google Scholar]
- 22.Harcup T. Sage Publications; 2007. The Ethical Journalist. [Google Scholar]
- 23.Vobič I., Poler Kovačič P. Watchdog journalism and confidential sources: a study of journalists' negotiation of confidentiality with their sources. Teorija Praksa. 2015:52. [Google Scholar]
- 24.Osifelo E.T. The challenges of anonymous source stories. A case study of Solomon Islands daily newspapers. Pac. Journal. Rev. 2017;23 [Google Scholar]
- 25.Buttry S. 2005. Handout on Confidential Sources.https://stevebuttry.wordpress.com/2013/12/19/a-2005-handout-on-confidential-sources-you-can-quote-me-on-that/ [Google Scholar]
- 26.Malling M. In: Close and Distant: Political Executive–Media Relations in Four Countries. Johansson J., Nygren G., editors. Nordicom; 2019. Power and exchange in formal andinformal interaction between journalists and their sources. [Google Scholar]
- 27.Shapiro I., Brin C., Bedard-Brule I., Mychajlowcysz K. Verification as a strategic ritual. Journal. Pract. 2013;7:657–673. [Google Scholar]
- 28.Laban V., Poler Kovačič M. Družboslovne razprave; 2007. Poročevalska Sklicevalnost V Agencijskih Besedilih; p. 23. [Google Scholar]
- 29.Ahadzadeh A.S., Fong L.K. Anonymous source undermines credibility: an elaboration likelihood model. Research Process. 2017;5(2017):85–96. [Google Scholar]
- 30.Frost C. Pearson Education Limited; 2011. Journalism. Ethics and Regulation. [Google Scholar]
- 31.Hoyt C. 2008. Culling the Anonymous Sources.https://www.nytimes.com/2008/06/08/opinion/08pubed.html [Google Scholar]
- 32.Gans J.H. Northwestern University Press; 1979. Deciding What's News: A Study of CBS Evening News, NBC Nightly News, Newsweek, and Time. [Google Scholar]
- 33.Sheehy M., M (2008). Foreign news stories more likely to include unnamed sources. Newspaper Res. J. 2008;29:24–37. [Google Scholar]
- 34.Sigal V.L. D C Heath & Co; 1973. Reporters and Officials: the Organization and Politics of Newsmaking. [Google Scholar]
- 35.Košir M., Poler Kovačič M. Magnolija; Ljubljana: 1997. Novinarska Etika. [Google Scholar]
- 36.Duffy J.M. Georgia State University; 2010. Unnamed Sources: A Longitudinal Review of the Practice and its Merits. [Google Scholar]
- 37.Lewis J., Williams A., Franklin B. A Compromised fourth estate? Journal. Stud. 2008;91:1–20. [Google Scholar]
- 38.Franklin B. In: Journalists, Sources, and Credibility. Franklin B., Carlson M., editors. Routledge; New York: 2011. Sources, credibility and the continuing crisis of UK journalism. [Google Scholar]
- 39.Machill M., Beiler M. The importance of the internet for Journalistic Research. Journal. Stud. 2009;10 [Google Scholar]
- 40.Carlson M. In: Journalists, Sources, and Credibility: New Perspectives. Franklin B., Carlson M., editors. Taylor and Francis; New York: 2011. Whither anonymity? Journalism and unnamed sources in a changing media environment; pp. 37–48. [Google Scholar]
- 41.Mediana . 2022. Aktualni Podatki O Branosti Tiskanih Medijev Mediana TGI 2021.https://www.mediana.si/aktualni-podatki-o-branosti-tiskanih-medijev-mediana-tgi-2021/ [Google Scholar]
- 42.Milosavljević M., Biljak Gerjevic R. Centre for Media Pluralism and Media Freedom; 2021. Spremljanje Medijskega Pluralizma V Digitalni Dobi.https://cadmus.eui.eu/bitstream/handle/1814/71962/slovenija_results_mpm_2021_cmpf.pdf?sequence=3&isAllowed=y [Google Scholar]
- 43.Bučar Ručman A., Poročanje o, v Sloveniji kriminaliteti. Revija kriminalistiko kriminologijo. 2011;62:23–38. [Google Scholar]
- 44.Ericson R.V., Baranek P.M., Baranek P.M., Chan J.B.L. Open Universirty Press; 1991. Representing Order: Crime, Law, and Justice in the News Media. [Google Scholar]; (a) Vreese C.H., Boomgaarden H.G., Semetko H.A. Direct framing Effects: the Effects of news media framing on public support for Turkish Membership in the European union. Commun. Res. 2010;38 [Google Scholar]
- 45.Cox S., Ficher R.J., Campbell T. vol. 5. Law Enforcement Executive Forum; 2005. https://iletsbeiforumjournal.com/images/Issues/FreeIssues/ILEEF%202005-5.5.pdf#page=22 (Police Officers and Reporters: Understanding the Relationship). [Google Scholar]
- 46.Sheafer T. How to evaluate it: the role of story-evaluative tone in agenda setting and priming. J. Commun. 2007;57:21–39. [Google Scholar]
- 47.Saxer U., Mediengefühlskultur M. Märki-Koepp“. Ölschläger; Munich, Germany: 1992. Zielgruppenspezifische Gefühlsdramaturgie Als Journalistische Produktionsroutine. [Google Scholar]
- 48.Macnamara J. Routledge; Oxon: 2018. Evaluating Public Communication. [Google Scholar]
- 49.Jelen A. FDV; Ljubljana: 2012. Ustvarjanje Medijske Podobe Predsednika Države: Vloga Uradnih Virov V Procesu Ouokvirjanja Televizijskih Novic. Doktorska Disertacija. [Google Scholar]
- 50.Cheung F.M., Wong C.T. Censorship and changing Gatekeeping roles: Non- routine news coverage and news routines in the context of police Digital communications in Hong Kong. Journal. Mass Commun. Q. 2016;93:1091–1114. [Google Scholar]
- 51.Tiffen R., Jones P.K., Rowe D., Aalberg T., Coen S., Curran J., Hayashi K., Iyengar S., Mazzoleni G., Papathanassopoulos S., Rojas H., Soroka S. Sources in the news. Journal. Stud. 2014;15:374–391. [Google Scholar]
- 52.Soloski J. Sources and channels of local news. Journal. Q. 1989;66:864–870. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
Data associated with this study has not been deposited into a publicly available repository. Data will be made available on request.