Improving responsible Media Reporting on Student Suicide to Reduce Negative Public Opinion: Text Mining Based on Sina Weibo in China

Abstract

The aims of this study were to explore whether student suicide reporting is consistent with media recommendations for suicide reporting; analyze public opinion and sentiments toward student suicide reports. A keyword search was performed on the WeiboReach platform. This study included 113 student suicide report posts and 176,262 readers’ comments on suicide news reports. Hierarchical generalized linear modeling was used to analyze the relationships between adherence to reporting recommendations and negative emotions in readers’ comments. None of the media reporting of student suicide was consistent with all of the media recommendations for suicide reporting. Netizens were less likely to post negative comments when the reports describe the suicide method used (OR 1.169, 95% CI 1.022∼1.337), and not specifying the cause of suicide was a protective factor for public negative emotion (OR 0.799, 95% CI 0.707, 0.905). The findings suggest improving responsible media reporting on student suicide to reduce negative public emotion.

Keywords

student suicide reporting media reporting recommendations public opinion text mining negative impact

Introduction

Suicide is an important global public health problem. In 2019, suicide was the fourth-leading cause of death among young people aged 15–29, accounting for 1.3% of all deaths worldwide (The World Health Organization, 2021). A total of 37 suicides were reported in 11 countries during the COVID-19 pandemic lockdowns from February to July 2020, and approximately one-third of the suicides were among students (Manzar et al., 2021). A survey of Chinese students aged 9–18 showed that approximately 32.09% of students had suicidal ideation (Tan et al., 2018). Among middle school students, 23.2% had nonsuicidal self-injury, and 3.2% attempted suicide (Liang et al., 2014). The incidence rates of suicidal ideation among college students are high (Mortier et al., 2018a, 2018b; Yao et al., 2014). Obviously, student suicides have become a public concern (Cheng et al., 2018).

The media can be a double-edged sword that can play a role in suicide prevention or increase the risk of suicide-related social media use behaviors (Choi & Noh, 2020; Liu et al., 2020; Unruh-Dawes et al., 2022). However, the majority of studies have concluded that there is a causal relationship between media coverage of suicides and actual suicide rates (Domaradzki, 2021). The WHO issued a recommendation list of “six don’ts and six dos” for media workers in suicide reporting in 2000 and revised it in 2017 (The World Health Organization, 2017). However, adherence to this recommendation has not been frequent (Arafat et al., 2021; Menon et al., 2020). Some scholars indicated that 51% of Chinese online media reports on suicides from 2015–2020 used sensational headlines, and 39% used photographs or videos of suicide sites/locations (Chu et al., 2021). Newspaper reporting of suicide in Kashmir during 2016–2020 barely adhered to the guidelines, with almost every report explicitly providing information about the identity of the deceased and almost none of the reports providing educational information related to suicide prevention (Shoib & Arafat, 2021). One study in Hong Kong found that inappropriate media coverage of student suicides, such as the release of photos and descriptions of suicide details that are not consistent with the WHO media guidelines, can increase suicidal ideation among adolescents, while student suicide rates can be reduced if the media reports more suicide prevention information (Cheng et al., 2018). Mainstream Chinese publishers most frequently reported suicide among youth (Lai et al., 2021). Some students with psychological disorders may be likely to generate suicidal ideation after harmful suicide exposure, which may lead to copycat behaviors. The adherence to media guidelines of student suicide reports in mainland China needs to be further discussed.

Audience Involvement and Media Reporting of Suicide

Audience involvement is an individual’s psychological relationship with a mediated persona—someone the individual does not know in any other way than through the media (Brown, 2015). Audience involvement with media persona involves four processes: transportation, parasocial interaction, identification, and worship (Brown, 2015). During the process of transportation, audiences become emotionally and psychologically involved in both the narrative and with the characters in the narrative. For example, a study found that celebrity suicide deaths can lead to significant, national increases in internet search volumes for suicide-related terms for celebrities of high prominence (Ortiz et al., 2019). In the process of parasocial interaction, netizens interact with others, including by viewing event-related comments, posting personal opinions, discussing events with other netizens and discussing related people. The identification process occurs when individuals adopt the attitudes, values, and beliefs or behaviors of other individuals or groups (Kelman, 1961). The process of worship involves a strong love for the persona; thus, worship is a much stronger psychological relationship than parasocial interaction. When posters published media reports of student suicide, netizens could become involved in those narratives through the four processes. Netizens generate emotional responses, express their opinions, and even spread their emotions. Social networking sites would be used to grieve and shape a wide range of emotional flow (Akhther & Tetteh, 2023; Moore et al., 2019). If the media reporting of student suicide is too detailed, including the location of the suicide, the method of suicide, and personal information of the people who died by suicide, netizens are more willing to become involved in suicide reporting. They are more likely to identify with suicidal ideation or post negative content than to post supportive content (Guidry et al., 2021).

Suicide Reporting and Public Opinion

As China’s Internet penetration rates and the number of Internet users have increased, online media have developed rapidly. The production mode of news has also changed in on media, and Weibo has the greatest influence on the news production mode. In the era of fragmented information, Weibo has become the center of public opinion due to its characteristics, including timeliness, fragmentation, instantaneousness, interactivity and civilianization. There were 224 million daily active users in September 2020, which makes it recognized as one of the most popular social media platforms in China (Feng & Johansson, 2019; Sina Weibo Data Center, 2020). Sina Weibo can increase news sources and communication effects for suicide reports. In addition, it could cause subtle psychological emotions and facilitate the generation and dissemination of public opinion (Han et al., 2020; Su, 2019; Yi et al., 2022).

To our knowledge, few studies have explored the public response and psychological effects of online media reporting of suicidal behaviors. A qualitative study of a case of suicide in Estonia showed that public comments on suicide reports in the five major online media outlets included value judgments, risk and protective factors, and criticism of the media. The most common emotional orientations of the comments were angry (34.5%), sarcastic (24.7%) and neutral (23.3%) (Sisask et al., 2012). The public reactions and psychological impacts of online suicide reports may vary with the social identities of the bereaved. The communication effects of media reporting on suicide related to celebrities and students were higher than those of reporting on the general population (Chu et al., 2021). Some scholars focused on public expression and emotions toward celebrities and found that people tended to glorify celebrity suicide (Qianchao et al., 2021). Little is known about readers’ reactions to media reports of student suicide. A content analysis of comments about student suicide reports using a manual judgment method found that readers were negative and irrational (Chiang et al., 2021). However, public opinion and the factors influencing the public’s emotional responses to online social media reports of student suicide remain unclear in China.

This study applied text mining methods to focus on suicide posts on Sina Weibo among full student groups, from elementary students to those with doctoral degrees. We classified the quality of student suicide reports using manual coding and explored public opinion and emotional orientation by cluster analysis and sentiment analysis. We aimed to explore the emotional effects that student suicide may have on the public and then find appropriate strategies to prevent public negative sentiment toward student suicide reports.

Method

Collection Process of Online-Reported Student Suicide-Related Events

Data were derived from the Harbin Institute of Technology joint laboratory for social networks and a data mining research platform named WeiboReach, which is one of the most advanced new media (including Weibo, WeChat, and other online media) news propagation analysis tools in China. The events in the platform inclusion criteria included high spreading speed, maintaining a large communication effect over a long period, and generating buzz in online social media.

To obtain online-reported student suicide-related events, this study first determined 19 keywords based on a literature review and an integrated national mortality surveillance system for death registration and mortality surveillance in China (Liu et al., 2016; Meng et al., 2020). These keywords included suicide, to die by suicide, exclusion of homicide, hanging, charcoal burning (a method of suicide involving burning of barbecue charcoal in a confined space, leading to carbon monoxide poisoning), jumping from a building, high fall, slitting the wrist, lightheartedness, taking poison, suicide note, exclusion of criminal case (removing scenarios involving deaths resulting from criminal offenses such as intentional injury and manslaughter), jumping from a cliff, falling from a cliff, and drowning. This study applied these keywords in Chinese to obtain events in the Weibo Reach database from 2015 to January 2022 by a keyword maximization search method. A total of 318 events were obtained, and 167 events remained after removing duplicate events and rumors. Ultimately, this study analyzed 31 student suicides after screening whether each person who died by suicide was a student in the event overviews. The flow diagram is shown in Figure 1. This study selected 13 online suicide reported events of students in which the standardized event impact index (EII) exceeded 65 (Table 1). The detailed information of the EII could be referred in Supplementary Materials.

Figure 1.

Flow chat of this study, EII, Event Influence Index.

Table 1.

The 13 Online-reported Suicide Events of Students (Sorted by the Event Impact Index).

Events	Event impact index
1. A student fell down from the building at Chengdu 49th middle school	77.1
2. A girl jumped from a building in Qingyang, Gansu Province	76.5
3. The wechat records of the girl at Peking University who committed suicide draw public attention	73.2
4. An elementary school student fell to death after the composition class	72.8
5. A junior two student fell to his death at Tai Fu middle school in Lu County, Sichuan Province	71.4
6. A sophomore student at North Central University jumped to his death after being caught cheating in a make-up exam	71.3
7. A former Peking University professor sexually assaulted a student	70.9
8. A doctor of Xi’an Jiaotong University drowned	70.3
9. Four left-behind children in Guizhou commit mass suicide by drinking pesticides	68.5
10. A shandong girl died in a nude fall	67.0
11. A graduate student of Nanjing University of posts and Telecommunications died by suicide	65.8
12. Jiangsu University reports that a student fell to his death	65.6
13. A female college student drowns after being trapped in pyramid schemes while working during summer vacation	65.3

Collection and Preprocessing of Corpus Data

Collection of Suicide Reports Corpus Data and Comments Corpus Data

This study used the above 13 online-reported suicide events of students as keywords to crawl all the search results by Python and GooSeeker, including the content of the posts, author of the post, number of likes, number of comments and number of forward. GooSeeker is a platform that offers web page crawling, information extraction, data extraction, data formatting, and data analysis capabilities (https://www.gooseeker.com/). The function of online data crawling has been well applied in previous studies (Bo & Di, 2020; Huang et al., 2021).

The posts needed to have some comments and spread effect; thus, this study set two standards: 1. number of comments ≥100; and 2. number of forwards ≥100. Posts meeting one or both criteria were included in this study. The posts were sorted according to the number of comments. According to the “top 10 effects” of public opinion in the cyberspace of mainland China, the opinions and comments of the top 10 netizens determine dozens or even hundreds of subsequent opinions and comments on certain news or social phenomena on the Internet, thus forming an online public opinion (Li & Yu, 2013). Thus, approximately 10 posts with the largest number of comments were selected for each event to form the student suicide report post corpus database. A total of 124 posts were obtained in this study.

All comments under each post in the suicide report post corpus database were crawled using GooSeeker to form the comment corpus database. Web crawler is an automated web browsing program capable of retrieving and collecting data from the Internet based on specific rules and algorithms. That is, this study collected all of the comments below the 124 posts of online-reported suicide events on Sina Weibo platform. Due to the long-time span, some bloggers had restricted the display of comments on Sina Weibo; thus, we were unable to obtain those Sina Weibo comments. Finally, a total of 176,262 comments from the 113 online suicide report posts were obtained.

Preprocessing of Student Suicide Posts and Comment Data

We used Python to conduct preprocessing, which included word separation and removing stop words. This study used the Jieba package to conduct word separation of the corpus. At present, the more commonly used Chinese stop words lists are the Chinese stop words list, Harbin University stop words list, and Sichuan University Machine Intelligence Lab stop words list. Then, we obtained the stop thesaurus by integrating and deduplicating these three stop word lists. This study removed the stop words after word segmentation.

Coding for Online Suicide Report Posts of Students

In 2017, the WHO updated its recommendations for the reporting of suicide in the media that included a list of six dos and six don’ts (Appendix table 1). We developed a coding scheme applicable to the online reporting of student suicide according to the list of six dos and six don’ts and the characteristics of media reporting on student suicide. There were 14 coding items and corresponding explanations in the coding scheme (Appendix Table 2). The correspondence between the 14 items and the six dos and don’ts is also presented in Appendix Table 2. Textual content analysis was conducted by three researchers, one of whom was an experienced coder. The three coders were trained at the same time, and the two junior coders encoded the 113 suicide reports of students separately. For posts with emotion feeling symbol, this study codes those posts according to the information expressed by the emotion feeling symbol. For example, a post includes negative emotional symbols such as sadness, fear, disgust, anger, we then code this post conform to item 1 of the coding scheme (Use language which sensationalizes suicide). For posts that were humorous or sarcastic, the coding results were decided by the research group after discussion. This study calculated the consistency between the two coders. Among the 14 coding items, the results of 9 coding items between the two researchers were completely consistent. We assessed intercoder reliability for the other 5 coding items. According to standards provided in the vast majority of scientific literature, coefficients >.80 are considered a high level of agreement, and ≥.70 is considered an acceptable level of agreement (Artstein & Poesio, 2008; Cohen, 1960; Hallgren, 2012). We calculated Cohen’s kappa coefficient for these 5 coding items. The researchers then discussed and revised the explanation of this coding item and then re-encoded it, if the Cohen’s kappa coefficient of a coding item was less than 0.8. And we calculated the Cohen’s kappa coefficient again. If Cohen’s kappa coefficient was 0.8 or more, the coding of the third experienced coder determined the final coding result. Finally, all of the Cohen’s kappa coefficients were more than 0.80, and the third researcher determined the coding results of those inconsistent posts.

Text Mining for the Comments on Online Student Suicide Posts

Text clustering was used to determine the topics of the comments of online students’ suicide posts. Sentiment analysis for the comments of online student suicide reports was conducted by the sentiment dictionary method. The detailed information, process, and effect evaluation of text clustering and sentiment analysis algorithm could be seen in Supplementary Materials.

Data Analysis

First, this study conducted content analysis to understand compliance with the guidelines for responsible reporting on suicide after the corpus data were collected and preprocessed. Second, this study used GooSeeker to conduct sentiment analysis for the 17,262 comments under 113 student suicide reports. We applied clustering analysis to mine public opinions of student suicide-related reports. Considering netizens expressed their comments under those posts of student suicide reporting, we used 2-level hierarchical generalized linear model (Bernoulli model) was used to analyze the relationships between the characteristics of student suicide posts and readers’ emotions (0 = positive sentiment, 1 = negative sentiment). In hierarchical generalized linear models, fixed effects are parameter estimates that do not vary across groups. Alternatively, random coefficients are parameter estimates that are allowed to vary across groups such as the level-1 regression coefficients. We first conducted random coefficient regression model to investigate whether there is significant variance in the intercepts across groups. According to the results, there was significant variance in the intercept term and the value of ICC1 was 0.219 (more than 0.12) (Chiang, 2023), which indicated that the effect of independent variables (characteristics of student suicide posts) on dependent variables (readers’ emotions) was different across post. We then conducted the intercepts-as-outcomes models with residuals open estimation to assess whether this variance is significantly related to the characteristics of student suicide posts. Seven characteristics (use language that normalizes suicide; use language that presents suicide as a constructive solution to problems; Titles are exaggerated or have a tendency to lead negatively; provide contact details of the bereaved, relatives or friends; report stories of how to get help; provide accurate information about where to seek help; and educate the public about the facts of suicide and suicide prevention) of student suicide posts were excluded from the regression analysis because their frequencies were less than 5%. About 7 item characteristics were included in the 2-level binary logistic regression model as 7 independent variables. We conducted three type of intercepts-as-outcomes models. First, we tested univariable associations between each characteristic of student suicide posts and netizens’ emotional orientation. Second, we adjusted for other characteristics in a model to test independent associations. Finally, we conducted the two-level logistic regression analysis after controlling the education level of student who have suicide behaviour, and the numbers of forwarding, likes, and comments.

Microsoft Excel (Microsoft Corporation) and SAS software 9.4 (SAS Institute, Inc) for statistical analyses, and hierarchical linear modeling (HLM) 7.0 (Copyright ©2011 By Scientific Software International, Inc All rights reserved.) for the statistical analyses.

Results

Adherence to WHO Media Guidelines on Suicide Reporting Among 113 Student Suicide Posts

The results regarding the WHO media guidelines for suicide reporting are shown in Table 2. Almost none of the reports were consistent with media recommendations for suicide reporting. Nearly 89.38% of the posts used suicide-related headlines, 79.65% described the method of suicide in detail, half of the posts (59.29%) exposed relevant photos and videos, and 56.64% provided details about the suicide site/location.

Table 2.

Adherence to WHO Recommended Guidelines Among Student Suicide Posts in Sina Weibo From 2015 to 2022 (n = 113).

Items	Frequency, n (%)
1. Use language which sensationalizes suicide	14 (12.39)
2. Use language which normalizes suicide	0
3. Use language which presents suicide as a constructive solution to problems	0
4. Describe the method used	90 (79.65)
5. Provide details about the site/location	64 (56.64)
6. Use suicide-related headlines	101 (89.38)
7. Titles are exaggerated or have a tendency to lead negatively (n = 101)	4 (3.96)
8. Use photographs or video footage of the bereaved people, related persons, or the site/location	67 (59.29)
9. Provide contact details of the bereaved, relatives or friends	0
10. Provide accurate information about where to seek help	0
11. Report stories of how to get help	5 (4.42)
12. Educate the public about the facts of suicide and suicide prevention	4 (3.54)
13. Do not specify the causes of suicide	52 (46.02)
14. Do not interview bereaved family or friends	84 (74.34)

Our findings also found that few online student suicide report posts provided useful information. Only five of the 113 posts (4.42%) provided stories about how to get help, but none provided accurate information about help, information that is critical for vulnerable people seeking it. Only four posts (3.54%) provided the public with knowledge of suicide prevention.

A minority of posts breached the media guidelines in relation to sensationalized language (12.39%) and used exaggerated or negatively guided titles (3.96%). Nearly 46.02% of posts did not specify an alleged cause of the suicide, and 74.34% did not interview bereaved family or friends. However, none of the online suicide reports about students used language that normalized suicide or provide contact details of the bereaved, relatives or friends. None of those reports presented suicide as a constructive solution to problems.

Text Clustering for the Comments of Online Student Suicide Posts

Cluster analysis was performed on the comments of all student suicides. This study used the elbow method to determine the number of clusters. The elbow method calculates the sum of squared distances from each point to its assigned center by the Hamming Distance. This study presents the sum of squared errors (SSE) obtained for different numbers of clusters (Appendix Figure 1). According to the “elbow principle”, we should select the number of clusters at the bend point as the optimal number of clusters. Therefore, this study chose 5 as the optimal cluster number. The silhouette coefficient was 0.63 in this study, which indicated that the clustering effect was relatively good in terms of cohesion and separation. Finally, the themes of each type of comment were summarized from the top 20 high-frequency words of each group (see Table 3). The five clusters of comments could be divided into two aspects, namely, the discussion of the event itself and the extended discussion.

Table 3.

Microblog Comment Clustering Themes and Representative Comments.

Aspects	Clusters	Representative comments
Event discussion	1. Truth	Hope to find out the truth
	2. Responsibility belonging	Don’t families have any responsibility?
		Schools and teachers must be to blame
		The psychological quality of the students is too poor
	3. Long-term impact	Who will ever send their children to school again?
	3. Long-term impact	Let’s think together about how to improve the current situation
Extended discussion	4. Relevant departments’ treatment measures	Hope the government strictly dispose…
		The attitude of the education department…
		The government do not response actively…
	5. The educational significance	Those who understand education will develop and guide their children…
		Education is going the wrong way
		There is something wrong with the teaching style

In terms of the discussion of the event itself, some netizens focused on the truth of the event. They analyzed, discussed and guessed the reasons related to the suicide reported in the media. There were also some comments about where the responsibility belonged. Most of the comments were emotional and irrational. Netizens blamed the bereaved, suicide-related schools, teachers, and parents. In addition, some comments were about the long-term impact of the media reporting on suicide. The comments included public worries about the future, such as being afraid to send children to school, how to protect themselves, and avoiding student suicidal behaviors. However, most of the comments were also negative and more about the negative impact and consequences.

There were two clusters in the extended discussions. One cluster was the evaluation of the strategies of the relevant organization or departments. Some of the comments were mild and rational, urging people to trust the government and wait for responses. However, some of the aggressive comments criticized the department’s treatment measures, arguing that it had failed in its duty. The other cluster was thinking about the educational significance; most of the comments indicated that the current education style did not prepare students to cope with life challenges. Some comments also discussed education styles.

The Evaluation of Sentiment Analysis Algorithm

The total accuracy of the sentiment analysis algorithm was 0.906. the precise rate of positive emotion was 0.727, and the recall rate was 0.955. The precise rate of negative emotion was 0.884, and the recall rate was 0.890, which was more consistent with the manual recognition.

Relationship Between Characteristics of Student Suicide Posts and Comment Emotions

This study used 2-level hierarchical generalized linear model (Bernoulli model) to explore the relationship between adherence to the WHO recommended guidelines and public negative or other sentiments (0 = positive and neutral sentiment, 1 = negative sentiment). The results are shown in Table 4. In univariable models, we found evidence of a positive association between the item of “use suicide-related headlines” and public emotion (Model 1). That is, netizens were more likely to post negative comments when posts used suicide-related headlines (OR = 1.305, 95% CI = [1.106.1.540], p = .002). And there was a negative between “Do not specify the causes of suicide” and public emotion. Not specifying the causes of suicide was a protective factor of public positing negative comments (OR = 0.811, 95% CI = [0.729.0.901], p < .001). Describing the suicide method and not specifying the causes of suicide had independent effect on public negative comments (Model 2). The posts were more likely to elicit readers’ negative emotions when they provided details of the suicide method (OR = 1.179, 95% CI = [1.052.1.321], p < .001). Associations between the items of “describing the suicide method” and “not specifying the causes of suicide” and public posting negative comments or not remained after adjustment for other characteristics of student suicide posts and all other confounders (Appendix Table 3).

Table 4.

Relationship Between Characteristics of Student Suicide Posts (n₁ = 113) and the Emotions Among Public Opinion (n₂ = 176,262).

Variables	Model 1			Model 2
Variables	β	p	OR (95%CI)	β	p	OR (95%CI)
- Use language which sensationalizes suicide	−0.099	0.249	0.906 (0.765, 1.073)	0.031	0.744	1.032 (0.854, 1.246)
- Describe the method used	0.084	0.137	1.088 (0.973, 1.215)	0.164	0.005	1.179 (1.052, 1.321)
- Provide details about the site/location	−0.096	0.069	0.908 (0.818, 1.008)	−0.038	0.477	0.963 (0.866, 1.070)
- Use suicide-related headlines	0.266	0.002	1.305 (1.106, 1.540)	0.155	0.108	1.168 (0.966, 1.411)
- Use photographs or video footage of the bereaved people, related persons, or the site/location	0.099	0.090	1.104 (0.985, 1.238)	0.006	0.917	1.006 (0.897, 1.129)
- Do not specify the causes of suicide	−0.210	<0.001	0.811 (0.729, 0.901)	−0.200	0.001	0.819 (0.725, 0.924)
- Do not interview bereaved family or friends	−0.079	0.192	0.924 (0.820, 1.041)	−0.035	0.541	0.965 (0.861, 1.082)

Note. CI: Confidence Interval. This study only included the items with a coverage rate of more than 5%. Model 1 represents the associations between each of those seven characteristics of student suicide posts and the emotions among public opinion. Model 2 represents the independent effect of each characteristic of student suicide posts on public emotions after adjusting for confounding by the other six characteristics.

Discussion

This study focused on the compliance of media reporting guidelines for suicide reporting updated by the WHO in 2017 and explored public opinion and appropriate strategies to reduce netizens’ negative sentiments from the perspective of enhancing responsible suicide reporting.

Student Suicide Reports are Not Consistent With all of the Media Reporting Guidelines

Previous studies found that there were common harmful reporting practices in reports containing suicide-related words in the headline and disclosing images of the bereaved, and there has been limited information about support programs in suicide reporting on social media (Chun et al., 2018; Lai et al., 2021; Pitman & Stevenson, 2015). In terms of student suicide reports, this study found that none of the posts under investigation in the current study were consistent with the WHO reporting guidelines. There is insufficient knowledge of suicide prevention provided in reporting, which is consistent with previous findings (Chu et al., 2018). Only 5 of the 113 media reports on student suicide provided information about suicide prevention and how to seek help. Our results were similar to previous findings about the quality of media reporting of suicide in some countries (Chu et al., 2018; Menon et al., 2020; Singh et al., 2022). In these reports, suicide prevention knowledge takes up very little space and largely fails to capture the reader’s attention. Information on sources of support is even omitted entirely in some media (Arafat et al., 2019; Pitman & Stevenson, 2015). High-intensity preventive reporting had a protective effect on reducing copycat suicide behaviors among students (Cheng et al., 2018). This study recommended that the media conduct more preventive reporting.

This study found that some online reports of student suicide provide detailed information, such as the detailed site/location, suicide method, and even identity of the student who died by suicide. The Internet and social media might have particularly important roles in spreading suicidal behaviour. Detailed descriptions of suicide can increase the risk of a vulnerable individual imitating the act (Knipe et al., 2022; Mars et al., 2015). The detailed site/location may make the people living in this area find something in common with the bereaved and increase their attention and exposure to media reporting student suicide events for longer periods. Some reports even have detailed photographs of the location where the bereaved committed suicide and bodies of suicide victims, which could have a substantial impact on readers. Media reporting about the detailed site or location could even have subsequent potential impact on individuals’ behaviors, such as suicidal ideation, self-harm, and copycat suicide behavior risk. Self-harm refers to any type of self-injurious behaviour, including suicide attempts and non-suicidal self-injury. One study found that the Internet could be directly associated with individuals’ self-harm thoughts and behaviours (Kirtley et al., 2021). The associations between social media use and self-injurious thoughts and behaviors have been verified in previous studies (Nesi et al., 2021; Tørmoen et al., 2023). In addition, those individuals with history of self-harm are more likely to search information related to suicide and to view media reports about self-harm and suicide (Mars et al., 2015). This vicious circle highlights the need for social media to be more cautious in reporting suicide events and avoid providing detailed descriptions of the site or location of suicides.

Including details on the suicide method may have a large impact on the number of suicides in the general population if they have high suicidal ideation or suicidal thoughts (Niederkrotenthaler et al., 2020). Some media reports of student suicides directly described the student’s identity and even the school’s name of the bereaved. A history of suicide among friends and acquaintances was significantly associated with an increased risk of suicidal behavior (Liu et al., 2019). Not adhering to the recommendations for suicide reports might affect readers’ suicide-related perceptions, information searches and psychology (Kim et al., 2015), and are likely to influence the population’s suicidal attempts and completed suicides (Chu et al., 2018). Responsible media reporting of suicide is important to prevent the Werther effect. This study recommended improving responsible reporting on student suicide, especially providing more prevention information, avoiding providing details about the site/location and methods, and avoiding using suicide-related photographs.

Guiding Public Opinion About Media Reporting of Student Suicidal Behaviors

Media portrayals of suicide are an important factor in shaping how the public understands the issue of suicide and reducing its stigma (O’Brien, 2021). Detailed and exaggerated media portrayals of suicide contribute to public discussion of the event. Some scholars found that comments about online media suicide reports were about family relations, blame for the people who died by suicide, suicide locations and methods, prevention, and media reporting styles. The comments on student suicide reports focused on the event itself and extended discussions about the event, such as the treatment measures taken by relevant departments and the educational significance. The public may discuss causes of suicide, blame people who are related to a suicide victim, and even be concerned about the further influence of these student suicide behaviors. Schools, teachers and parents are the focus of discussion in all student suicides. The public’s words of blame, abuse, and accountability may cause serious psychological harm to these groups. Insufficient public awareness and understanding of suicide issues, negative attitudes and prejudices, and labeling those people who died by suicide as selfish and weak have exacerbated the stigma and discrimination against suicides. Because of the stigma and taboo around mental illness and suicide, people with suicidal ideation are underrecognized and undertreated. Stigmatization also affects survivors of suicide (Witte et al., 2010). The grief of the survivors of suicide is not recognized by society, resulting in a lack of social support for the survivors of suicide, resulting in more guilt and self-blame, which can easily deteriorate into negative emotional reactions such as depression and suicidal ideation (Hanschmidt et al., 2016). A greater focus on prevention in media reporting could reduce public stigma and suicide copycat behavior (Sorensen et al., 2021). This study suggests that health educators disseminate correct suicide prevention information and carry out relevant life education and training to reduce stigma around suicide, thereby improving the effectiveness of suicide prevention efforts.

Government Responding to the Concerns of Netizens in a Timely Manner to Enhance Its Credibility

With respect to some media reporting on student suicide related to social issues, the objects of netizens’ discussions include the government, society, and even the country. If government information is not disclosed proactively and in a timely manner, it is easy for the public to cause misunderstandings or doubts. Evolving public opinion and risk have a negative impact on the government’s image and credibility. The Chinese government should be aware of the severity of negative public opinion on suicide reporting among students and take action to respond to the concerns of netizens.

Reducing the associations of student suicide reporting with potentially harmful characteristics on public negative emotions

Social media can be a platform on which netizens experience their emotions, feelings and thoughts. Celebrity suicide news reports can evoke the expression of negative emotions (Niederkrotenthaler et al., 2019; Rosen et al., 2020). This study found that most of the comments toward student suicide news were negative, and not adhering to media guidelines on student suicide reporting may have a negative effect on public sentiment. The psychological impact of suicide reports on readers may affect public mental health (Fahey et al., 2018). The suicide attempters may be more impulsive when they experience negative emotional states (You et al., 2020). There is an ongoing relationship in which suicide causes a negative mood, which in turn causes emotions of irrationality and agitation, leading to negative emotional contagion in the network. Thus, appropriate strategies need to be implemented to reduce the negative content in public opinion.

Language can play a large role in reflecting and shaping assumptions (Whorf, 2012), and it is a key tool in health care (Richards, 2019). Inappropriate language may strike a chord in netizens who are in situations similar to that reported. This study indicated that avoiding describing suicide method and using suicide-related headlines, and omitting specifying the causes of suicide are beneficial for reducing the likelihood of negative comments. More detailed information about student suicide reports and suicide-related headlines can attract the reader’s attention more by emphasizing and giving the essence of the story. This study highlights the importance of adhering to language recommendations for suicide reporting, and media professionals should be careful when writing headlines about suicides.

Limitations

First, this study analyzed the content of reports and comments on only the Sina Weibo platform, and only included 113 suicide reports in the content analysis. The Netizens also may discuss student suicide on other social media channels in China, such as WeChat, TikTok, BiliBili, and Little Red Book (Chinese equivalent of Instagram). As Weibo is one of China’s largest social media platforms, this study discussed Sina Weibo to make theoretical contributions to improving responsible student suicide reporting and reducing the negative impact of suicide reporting on public opinion. The 113 suicide reports include in the content analysis generated wide public attention and discussion. These posts may affect public sentiment infection or spread. This study analysed whether these posts are consistent with to the reporting guidelines, as well as analyzed the relationship between the adherence to the guidelines and commenter’s emotions. It would be of great significance in creating a preventive media environment for student suicide reporting. Certainly, future studies need to further expand the representativeness of samples. Adherence to student suicide reporting guidelines on other social media platforms and readers’ comments also need to be investigated. Second, this study analyzed public opinion toward student suicide reports based on cluster analysis and sentiment analysis. We did not investigate readers regarding the impact of student suicide media reports. Moreover, future studies should investigate the mechanism by which netizens’ opinions or emotions arise or spread to reduce the negative impact of student suicide media reports on public mental health. Finally, the sentiment analysis data of netizens’ comments in this study were mainly based on text content, and emoji- and image-based postings were excluded, which may have caused bias. Further analysis of video and picture content is needed in the future to obtain more convincing conclusions.

Theoretical Implications

Our findings add evidence that media reports of suicide impact public sentiment about student suicide events in China. This study has theoretical implications for leading netizens to be actively involved in the narrative of student suicide. When more details are provided in media covering suicide events (such as describing the method used; using suicide-related headlines; and reporting the causes of suicide), people have greater perceptions of the relevance of the suicide to themselves, thus stimulating negative public sentiment.

Practical Implications

This study also has practical implications for media professionals, social media platforms, society and netizens. We suggest that media workers be more rational and kinder when reporting student suicide behaviors. While maintaining the authenticity of their reports and adhering to the reporting guidelines, media professionals should pay more attention to the mental health of netizens. They should think about whether their reports will cause secondary harm to people who are related to the suicide victim and others and control the associations of the language used and manner of reporting on public mental health, including the stigma and prejudice around suicide.

Social media platforms should strengthen their control and management of the online environment by proposing and revising their suicide reporting guidelines. Given the distinctive characteristics of the student population, it is crucial for social media to approach reporting on student suicides with caution, considering the potential negative emotional impact on the public. Platforms should also exploit their strengths and take responsibility in promoting audience involvement in student suicide behavior, such as publishing suicide prevention public service announcements (Harris & Krishnan, 2023). They should monitor and identify public emotional orientations to suicide news early, especially for posts providing suicide details, using suicide-related headlines, or presenting the causes of suicide. Currently, China lacks national policies and guidelines for suicide intervention. We aspire for this study to contribute towards the formulation of suicide intervention laws and regulations, especially concerning media reporting guidelines.

Preventive efforts should target Weibo bloggers who irresponsibly report on student suicide and foster a good and positive cyber environment. The government should strengthen its public opinion response ability and awareness of public opinion risk prevention and control. We recommend that the government closely monitor and collect emerging public opinion and take the initiative to set agendas to beat the instant dissemination of irresponsible media reporting of student suicide on social media.

This study highlighted that netizens must remain rational and avoid being emotional when involved in narratives about student suicidal behavior. Authorities and social media platforms can adopt a combined approach of guidance and governance to get netizens to remain rational in such events. Firstly, social media platforms should publish responsible reports to prevent secondary emotional harm to the public. Secondly, after the student suicide events, positive guidance can be provided to encourage the public to enhance their self-judgment abilities and urge them not to post extremely emotional information, especially the influential users. Finally, social media platforms can also consider establishing monitoring and warning mechanisms. When there is extremely negative information, governance measures should be initiated, such as counseling for posters and commenters and post deletion. Moreover, it is critical to take actions to prevent the spread of this negative message. Those measures included the use of moderation strategies that remove inappropriate content (Boberg et al., 2018), requiring users to sign in to prevent anonymous commenting (Santana, 2019), and implementing interventions such as emotional tagging to promote implicit emotion regulation in online news comments, thereby encouraging people to be rational and reducing the further dissemination of negative information (Syrjämäki et al., 2022). The Internet provides a good platform for individual expression, but there are potential negative effects of free speech. Improving the ability to technically control the Internet can limit online communication to a certain extent, but it violates the right of citizens to freedom of speech. Therefore, solving the problem of online public expressions could focus on reacting and providing guidance. After the occurrence of suicide incidents, health education workers and social media managers should monitor public opinion in real time and guide citizens to publish suicide prevention information to reduce the associations between media reporting of suicide with potentially harmful characteristics on public emotion.

Conclusion

This study crawled social media information on online reporting of student suicide from 2015 to 2022 on Sina Weibo. We explored whether these reports reflected the WHO recommendations and discussed public opinion. Few online student suicide reports applying these WHO recommendations well. Moreover, student suicide news reports use emotional words. Qualified suicide reports can play a positive role in public mental health. This study found that using describing suicide methods was the independent risk factor associated with commenters’ negative emotions, and not specifying the causes of suicide was the independent protective factor after adjusting all of the confounders. These guidelines for responsible reporting should be more widely implemented and promoted, especially when reporting on the deaths of students by suicide. The government and social media should work together to improve media reporting of suicide and positively lead audience involvement to reduce the associations between the narrative of media reporting of suicide with potentially harmful characteristics on public negative emotion.

Supplemental Material

Supplemental Material - Improving responsible Media Reporting on Student Suicide to Reduce Negative Public Opinion: Text Mining Based on Sina Weibo in China

Supplemental Material for Improving responsible Media Reporting on Student Suicide to Reduce Negative Public Opinion: Text Mining Based on Sina Weibo in China by Meijie Chu, Chun-Yang Lee, Shiling Huang, Xiaoke Zhang, Lijie Wang, Tianmu Chen, and Yi-Chen Chiang in OMEGA - Journal of Death and Dying.

Footnotes

Authors’ Contributions

Conceptualization: Yi-Chen Chiang, Meijie Chu, and Chun-Yang Lee; Methodology: Xiaoke Zhang, Meijie Chu, and Shiling Huang; Formal analysis and investigation: Xiaoke Zhang, Meijie Chu, and Shiling Huang; Writing - original draft preparation: Meijie Chu, Shiling Huang, and Lijie Wang; Writing - review and editing: Meijie Chu, and Shiling Huang; Funding acquisition: Yi-Chen Chiang; Supervision: Yi-Chen Chiang, Chun-Yang Lee and Tianmu Chen.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by the Scientific Research Grant of Fujian Province of China (grant no. Z0230104), and JingZe Bio Cohort Study Center, School of Public Health, Xiamen University.

Ethical Statement

ORCID iD

Yi-Chen Chiang

Data Availability Statement

The data that support the findings of this study are available from the corresponding author, [Y-C C], upon reasonable request.*

Supplemental Material

Supplemental material for this article is available online.

Author Biographies

■■■

References

Akhther

Tetteh

D. A.

(2023). Global mediatized death and emotion: Parasocial grieving-mourning #stephenhawking on twitter. Omega, 87(1), 126–145. https://doi.org/10.1177/00302228211014775

Arafat

S. M. Y.

Mali

Akter

(2019). Quality of online news reporting of suicidal behavior in Bangladesh against World Health Organization guidelines. Asian Journal of Psychiatry, 40(1), 126–129. https://doi.org/10.1016/j.ajp.2018.10.010

Arafat

S. M. Y.

Menon

Bascarane

Kar

S. K.

Kabir

(2021). Variations in newspaper reporting of suicidal behavior in the WHO-South-East Asian region. Journal of Public Health, 43(4), E684–E688. https://doi.org/10.1093/pubmed/fdaa139

Artstein

Poesio

(2008). Inter-coder agreement for computational linguistics. Computational Linguistics, 34(4), 555–596. https://doi.org/10.1162/coli.07-034-R2

(2020). The visualization analysis of China civil public opinions in the early period of COVICD-19 epidemic: Based on Gooseeker data mining of Weibo. Journal of China Studies, 23(2), 71–84. https://doi.org/10.20288/jcs.2020.23.2.71

Boberg

Schatto-Eckrodt

Frischlich

Quandt

(2018). The moral gatekeeper? Media and Communication, 6(4), 58–69. https://doi.org/10.17645/mac.v6i4.1493

Brown

W. J.

(2015). Examining four processes of audience involvement with media personae: Transportation, parasocial interaction, identification, and worship. Communication Theory, 25(3), 259–283. https://doi.org/10.1111/comt.12053

Cheng

Chen

Lee

E. S. T.

Yip

P. S. F.

(2018). The role of media in preventing student suicides: A Hong Kong experience. Journal of Affective Disorders, 227(3), 643–648. https://doi.org/10.1016/j.jad.2017.11.007

Chiang

Y.-c.

(2023). Health services research methods. People’s Medical Publishing House.

10.

Chiang

Y.-C.

Chu

Lin

Cai

Chen

Wang

Rui

Zhang

Xie

Lee

C.-Y.

Chen

(2021). Capturing the trajectory of psychological status and analyzing online public reactions during the coronavirus disease 2019 pandemic through Weibo posts in China. Frontiers in Psychology, 12(eCollection 2021), 744691. https://doi.org/10.3389/fpsyg.2021.744691.

11.

Choi

D.-H.

Noh

G.-Y.

(2020). The influence of social media use on attitude toward suicide through psychological well-being, social isolation, and social support. Information, Communication and Society, 23(10), 1427–1443. https://doi.org/10.1080/1369118X.2019.1574860

12.

Chu

Lin

Cai

Chen

S. H.

Zhang

Man

Lee

C. Y.

Chiang

Y. C.

(2021). Appropriate strategies for reducing the negative impact of online reports of suicide and public opinion from social media in China. Frontiers in Public Health, 9(eCollection 2021), Article 756360. https://doi.org/10.3389/fpubh.2021.756360

13.

Chu

Zhang

X. Y.

Cheng

P. X.

Schwebel

D. C.

G. Q.

(2018). Assessing the use of media reporting recommendations by the World health organization in suicide news published in the most influential media sources in China, 2003-2015. International Journal of Environmental Research and Public Health, 15(3), Article 451. https://doi.org/10.3390/ijerph15030451

14.

Chun

Kim

Lee

(2018). Fidelity assessment of the suicide reporting guidelines in Korean newspapers. BMC Public Health, 18(1), Article 1115. https://doi.org/10.1186/s12889-018-6014-4

15.

Cohen

(1960). A coefficient of agreement for nominal scales. Educational and Psychological Measurement, 20(1), 37–46. https://doi.org/10.1177/001316446002000104

16.

Domaradzki

(2021). The werther effect, the papageno effect or No effect? A literature review. International Journal of Environmental Research and Public Health, 18(5), 2396. https://doi.org/10.3390/ijerph18052396

17.

Fahey

R. A.

Matsubayashi

Ueda

(2018). Tracking the Werther Effect on social media: Emotional responses to prominent suicide deaths on twitter and subsequent increases in suicide. Social Science & Medicine, 219(2018), 19–29. https://doi.org/10.1016/j.socscimed.2018.10.004

18.

Feng

Johansson

A. C.

(2019). Top executives on social media and information in the capital market: Evidence from China. Journal of Corporate Finance, 58(2019), 824–857. https://doi.org/10.1016/j.jcorpfin.2019.04.009

19.

Guidry

J. P. D.

O’Donnell

N. H.

Miller

C. A.

Perrin

P. B.

Carlyle

K. E.

(2021). Pinning despair and distress - suicide-related content on visual social media platform pinterest. Crisis, 42(4), 270–277. https://doi.org/10.1027/0227-5910/a000719

20.

Hallgren

K. A.

(2012). Computing inter-rater reliability for observational data: An overview and tutorial. Tutorials in quantitative methods for psychology, 8(1), 23–34. https://doi.org/10.20982/tqmp.08.1.p023

21.

Han

Wang

Zhang

Wang

(2020). Using social media to mine and analyze public opinion related to COVID-19 in China. International Journal of Environmental Research and Public Health, 17(8), 2788. https://doi.org/10.3390/ijerph17082788

22.

Hanschmidt

Lehnig

Riedel-Heller

S. G.

Kersting

(2016). The stigma of suicide survivorship and related consequences-A systematic review. PLoS One, 11(9), Article e0162688. https://doi.org/10.1371/journal.pone.0162688

23.

Harris

Krishnan

(2023). Exploring the association between suicide prevention public service announcements and user comments on YouTube: A computational text analysis approach. Journal of Health Communication, 28(5), 302–311. https://doi.org/10.1080/10810730.2023.2203077

24.

Huang

W. D.

Zhu

S. T.

Yao

X. K.

(2021). Destination image recognition and emotion analysis: Evidence from user-generated content of online travel communities. The Computer Journal, 64(3), 296–304. https://doi.org/10.1093/comjnl/bxaa064

25.

Kelman

H. C.

(1961). Processes of opinion change. Public Opinion Quarterly, 25(1), 57–78. https://doi.org/10.1086/266996

26.

Kim

E. Y.

Song

M. H.

Kim

Y. J.

(2015). The effect of newspaper suicide reporting on suicide-related perception: Focusing on the dynamic relationship between suicide contents and web-search activities. Korean Journal of Journalism & Communication Studies, 59(3), 94–124.

27.

Kirtley

O. J.

Hussey

Marzano

(2021). Exposure to and experience of self-harm and self-harm related content: An exploratory network analysis. Psychiatry Research, 295(2021), 113572. https://doi.org/10.1016/j.psychres.2020.113572

28.

Knipe

Padmanathan

Newton-Howes

Chan

L. F.

Kapur

(2022). Suicide and self-harm. Lancet, 399(10338), 1903–1916. https://doi.org/10.1016/s0140-6736(22)00173-8

29.

Lai

Peng

Zhao

(2021). Assessing suicide reporting in top newspaper social media accounts in China: Content analysis study. JMIR mental health, 8(5), Article e26654. https://doi.org/10.2196/26654

30.

(2013). The “Top 10 Effect” in internet public opinion an interpretation of the causes of internet public opinion. Shanghai Journalism Review, 2, 50–53. https://doi.org/10.16057/j.cnki.31-1171/g2.2013.02.014

31.

Liang

Yan

Zhang

Zhu

Situ

Huang

(2014). Differences between non-suicidal self injury and suicide attempt in Chinese adolescents. Asian journal of psychiatry, 8(1), 76–83. https://doi.org/10.1016/j.ajp.2013.11.015

32.

Liu

Lopez

A. D.

Wang

Cai

Page

Yin

Liu

You

Zhou

(2016). An integrated national mortality surveillance system for death registration and mortality surveillance, China. Bulletin of the World Health Organization, 94(1), 46–57. https://doi.org/10.2471/blt.15.153148

33.

Liu

X. C.

Chen

Liu

Z. Z.

Wang

J. Y.

Jia

C. X.

(2019). Prevalence of suicidal behaviour and associated factors in a large sample of Chinese adolescents. Epidemiology and Psychiatric Sciences, 28(3), 280–289. https://doi.org/10.1017/s2045796017000488

34.

Liu

X. Y.

Huang

J. S.

N. X.

Zhu

T. S.

(2020). Mediation effect of suicide-related social media use behaviors on the association between suicidal ideation and suicide attempt: Cross-sectional questionnaire study. Journal of Medical Internet Research, 22(4), Article e14940. https://doi.org/10.2196/14940

35.

Manzar

M. D.

Albougami

Usman

Mamun

M. A.

(2021). Suicide among adolescents and youths during the COVID-19 pandemic lockdowns: A press media reports-based exploratory study. Journal of Child and Adolescent Psychiatric Nursing: Official Publication of the Association of Child and Adolescent Psychiatric Nurses, Inc, 34(2), 139–146. https://doi.org/10.1111/jcap.12313

36.

Mars

Heron

Biddle

Donovan

J. L.

Holley

Piper

Potokar

Wyllie

Gunnell

(2015). Exposure to, and searching for, information about suicide and self-harm on the Internet: Prevalence and predictors in a population based cohort of young adults. Journal of Affective Disorders, 185(2015), 239–245. https://doi.org/10.1016/j.jad.2015.06.001

37.

Meng

Yin

Liu

Zhou

(2020). Suicide mortality - China, 2018. China CDC weekly, 2(36), 694–695. https://doi.org/10.46234/ccdcw2020.192

38.

Menon

Kaliamoorthy

Sridhar

V. K.

Varadharajan

Joseph

Kattimani

Kar

S. K.

Arafat

S. Y.

(2020). Do Tamil newspapers educate the public about suicide? Content analysis from a high suicide union territory in India. International Journal of Social Psychiatry, 66(8), 785–791. https://doi.org/10.1177/0020764020933296

39.

Moore

Magee

Gamreklidze

Kowalewski

(2019). Social media mourning using grounded theory to explore how people grieve on social networking sites. Omega, 79(3), 231–259. https://doi.org/10.1177/0030222817709691

40.

Mortier

Auerbach

R. P.

Alonso

Bantjes

Benjet

Cuijpers

Ebert

D. D.

Green

J. G.

Hasking

Nock

M. K.

O’Neill

Pinder-Amaker

Sampson

N. A.

Vilagut

Zaslavsky

A. M.

Bruffaerts

Kessler

R. C.

WHO WMH-ICS Collaborators (2018a). Suicidal thoughts and behaviors among first-year college students: Results from the WMH-ICS project. Journal of the American Academy of Child & Adolescent Psychiatry, 57(4), 263–273. https://doi.org/10.1016/j.jaac.2018.01.018

41.

Mortier

Cuijpers

Kiekens

Auerbach

R. P.

Demyttenaere

Green

J. G.

Kessler

R. C.

Nock

M. K.

Bruffaerts

(2018b). The prevalence of suicidal thoughts and behaviours among college students: A meta-analysis. Psychological Medicine, 48(4), 554–565. https://doi.org/10.1017/s0033291717002215

42.

Nesi

Burke

T. A.

Bettis

A. H.

Kudinova

A. Y.

Thompson

E. C.

MacPherson

H. A.

Fox

K. A.

Lawrence

H. R.

Thomas

S. A.

Wolff

J. C.

Altemus

M. K.

Soriano

Liu

R. T.

(2021). Social media use and self-injurious thoughts and behaviors: A systematic review and meta-analysis. Clinical Psychology Review, 87(2021), 102038. https://doi.org/10.1016/j.cpr.2021.102038

43.

Niederkrotenthaler

Till

Garcia

(2019). Celebrity suicide on Twitter: Activity, content and network analysis related to the death of Swedish DJ Tim Bergling alias Avicii. Journal of Affective Disorders, 245(2019), 848–855. https://doi.org/10.1016/j.jad.2018.11.071

44.

Niederkrotenthaler

Braun

Pirkis

Till

Stack

Sinyor

Tran

U. S.

Voracek

Cheng

Arendt

Scherr

Yip

P. S. F.

Spittal

M. J.

(2020). Association between suicide reporting in the media and suicide: Systematic review and meta-analysis. BMJ, 368(8238), m575. https://doi.org/10.1136/bmj.m575

45.

O’Brien

(2021). Reporting on mental health difficulties, mental illness and suicide: Journalists’ accounts of the challenges. Journalism, 22(12), 3031–3047. https://doi.org/10.1177/1464884920903855

46.

Ortiz

S. N.

Forrest

L. N.

Fisher

T. J.

Hughes

Smith

A. R.

(2019). Changes in internet suicide search volumes following celebrity suicides. Cyberpsychology, Behavior, and Social Networking, 22(6), 373–380. https://doi.org/10.1089/cyber.2018.0488

47.

Pitman

Stevenson

(2015). Suicide reporting within British newspapers’ arts coverage. Crisis, 36(1), 13–20. https://doi.org/10.1027/0227-5910/a000294

48.

Qianchao

F. U.

Lee

(2021). Social perception of suicide through comments posted on online suicide-related news: A comparison of celebrities’suicides and ordinary persons’suicides. Korean Journal of Broadcasting and Telecommunication Studies, 35(2), 35–67.

49.

Richards

(2019). The power of language: The importance of shaping language as a constructive tool in health care. Journal of Evaluation in Clinical Practice, 25(6), 1055–1056. https://doi.org/10.1111/jep.13172

50.

Rosen

Kreiner

Levi-Belz

(2020). Public response to suicide news reports as reflected in computerized text analysis of online reader comments. Archives of Suicide Research: Official Journal of the International Academy for Suicide Research, 24(sup1), 243–259. https://doi.org/10.1080/13811118.2018.1563578

51.

Santana

A. D.

(2019). Toward quality discourse: Measuring the effect of user identity in commenting forums. Newspaper Research Journal, 40(4), 467–486. https://doi.org/10.1177/0739532919873089

52.

Shoib

Arafat

S. M. Y.

(2021). Quality of newspaper reporting of suicide in Kashmir: Adherence to World health organization guidelines. Psychiatry, 84(3), 291–298. https://doi.org/10.1080/00332747.2021.1888605

53.

Sina Weibo Data Center . (2020). Weibo user development report (p. 2020). Sina Weibo Data Center. https://data.weibo.com/report/reportDetail?id=456

54.

Singh

Mahato

Khadka

Basnet

Bista

Karki

Arafat

S. M. Y.

(2022). Newspaper reporting of suicide in Nepal: Quality assessment against World Health Organization media guidelines. Health science reports, 5(2), Article e547. https://doi.org/10.1002/hsr2.547

55.

Sisask

Mark

Värnik

(2012). Internet comments elicited by media portrayal of a familicide-suicide case. Crisis, 33(4), 222–229. https://doi.org/10.1027/0227-5910/a000134

56.

Sorensen

J. B.

Pearson

Armstrong

Andersen

M. W.

Weerasinghe

Hawton

Konradsen

(2021). A qualitative analysis of self-harm and suicide in Sri Lankan printed newspapers. Crisis, 42(1), 56–63. https://doi.org/10.1027/0227-5910/a000687

57.

(2019). Exploring the effect of Weibo opinion leaders on the dynamics of public opinion in China: A revisit of the two-step flow of communication. Global Media and China, 4(4), 493–513. https://doi.org/10.1177/2059436419866012

58.

Syrjämäki

A. H.

Ilves

Kiskola

Rantasila

Isokoski

Olsson

Surakka

(2022). Facilitating implicit emotion regulation in online news commenting—an experimental vignette study. Interacting with Computers, 34(5), 129–136. https://doi.org/10.1093/iwc/iwad010

59.

Tan

Xia

Reece

(2018). Social and individual risk factors for suicide ideation among Chinese children and adolescents: A multilevel analysis. International Journal of Psychology: Journal International de Psychologie, 53(2), 117–125. https://doi.org/10.1002/ijop.12273

60.

The World Health Organization . (2017). Preventing suicide: A resource for media professionals, update 2017. World Health Organization. https://www.who.int/publications/i/item/WHO-MSD-MER-17.5

61.

The World Health Organization . (2021). Suicide. World Health Organization. https://www.who.int/news-room/fact-sheets/detail/suicide

62.

Tørmoen

A. J.

Myhre

M. Ø.

Kildahl

A. T.

Walby

F. A.

Rossow

(2023). A nationwide study on time spent on social media and self-harm among adolescents. Scientific Reports, 13(1), Article 19111. https://doi.org/10.1038/s41598-023-46370-y

63.

Unruh-Dawes

E. L.

Smith

L. M.

Krug Marks

C. P.

Wells

T. T.

(2022). Differing relationships between instagram and twitter on suicidal thinking: The importance of interpersonal factors. Social Media + Society, 8(1), Article 20563051221077027. https://doi.org/10.1177/20563051221077027

64.

Whorf

B. L.

(2012). In Carroll

J. B.

(Ed.), Language, thought, and reality. The MIT Press. https://www-jstor-org-443.web.bisu.edu.cn/stable/j.ctt5hhbx2

65.

Witte

T. K.

Smith

A. R.

Joiner

T. E.

Jr. (2010). Reason for cautious optimism? Two studies suggesting reduced stigma against suicide. Journal of Clinical Psychology, 66(6), 611–626. https://doi.org/10.1002/jclp.20691

66.

Yao

Y. S.

Chang

W. W.

Jin

Y. L.

Chen

L. P.

Zhang

(2014). Life satisfaction, coping, self-esteem and suicide ideation in Chinese adolescents: A school-based study. Child: Care, Health and Development, 40(5), 747–752. https://doi.org/10.1111/cch.12142

67.

Gina Qu

Zhang

W. J.

(2022). Depicting the emotion flow: Super-spreaders of emotional messages on Weibo during the COVID-19 pandemic. Social Media + Society, 8(1), Article 205630512210849. https://doi.org/10.1177/20563051221084950

68.

You

Lim

C. E.

Park

Ryu

Lee

H. J.

Choi

J. M.

Cho

Y. S.

(2020). Response inhibition in emotional contexts in suicide ideators and attempters: Evidence from an emotional stop-signal task and self-report measures. Psychology of Violence, 10(6), 594–603. https://doi.org/10.1037/vio0000351

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.36 MB