Disability, hate speech and social media: video replies to haters on TikTok

Discapacidad, discursos de odio y redes sociales: video-respuestas a los haters en TikTok

Victoria García-Prieto

University of Seville. Spain.

vgarcia8@us.es

Mónica Bonilla-del-Río

The European University of the Atlantic. Spain.

monica.bonilla@uneatlantico.es

Juan C. Figuereo-Benítez

University of Seville. Spain.

figuereo@us.es

The author Juan C. Figuereo-Benítez is the beneficiary of a PIF pre-doctoral contract funded by the VI PPIT-US (Own Research and Transfer Plan of the University of Seville), within the framework of the research group SEJ-675: Communication, power and critical thinking in the face of global change (Compoder, for its Spanish acronym), of the University of Seville.

How to cite this article:

García-Prieto, Victoria; Bonilla-del-Río, Mónica, & Figuereo-Benítez, Juan C. (2024). Disability, hate speech and social media: video replies to haters on TikTok [Discapacidad, discursos de odio y redes sociales: video-respuestas a los haters en TikTok]. Revista Latina de Comunicación Social, 82, 1-20. https://doi.org/10.4185/rlcs-2024-2258

ABSTRACT

Introduction: In recent years, the spread of hate speech through social media has grown to become an increasingly worrisome problem in our societies. This study focuses on hate speech against people with disabilities. The first objective is to analyze these people's video replies to haters on TikTok, considering aspects such as content, reach, interaction, engagement, tone and intentionality of the reply itself. The second objective is to delve into the perspective of the authors of the video replies regarding hate speech and their stance towards this phenomenon. Methodology: The study includes a mixed methodology, consisting of content analysis of 64 TikTok video replies posted by users with disabilities in response to hate messages received on this social network, and 14 structured interviews to users among the profiles that integrate the sample. Results: The results reflect that, in addition to negative comments, insults, belittling or mockery received by other groups, there is other content related to ableism or disability denial. Conclusions y discussion: The findings reveal that the expansion of hate speech is greater on TikTok and point to causes such as the age of the users of this social network, the algorithm operation or the anonymity allowed by the social media. Ways to reduce hate speech in social networks are explored, focusing on the legal framework, education and mental health promotion.

Keywords: disability; ableism; social media; TikTok; internet; hate speech; haters.

RESUMEN

Introducción: En los últimos años, la propagación de discursos de odio a través de redes sociales ha aumentado hasta convertirse en un problema cada día más preocupante en nuestras sociedades. Este estudio se enfoca en los discursos de odio hacia el colectivo de personas con discapacidad. El objetivo es analizar sus video-respuestas a los haters en TikTok, considerando aspectos como el contenido, el alcance, la interacción, el engagement, el tono y la intencionalidad de la propia respuesta. Así como profundizar en la perspectiva de los autores de las video-respuestas respecto al discurso de odio y su postura frente a este fenómeno. Metodología: La investigación incluye una metodología mixta, compuesta por análisis de contenido de 64 video-respuestas de TikTok publicadas por usuarios con discapacidad en respuesta a mensajes de odio recibidos en esta red social, y por 14 entrevistas estructuradas a usuarios entre los perfiles que integran la muestra. Resultados: Los resultados reflejan cómo, a los comentarios negativos, insultos, menosprecios o burlas que reciben otros colectivos, a este se suman otros contenidos como el capacitismo o la negación de la discapacidad. Conclusiones y discusión: Los hallazgos coinciden en que la expansión de discursos de odio es mayor en TikTok y apuntan a causas como la edad de los usuarios de esta red social, el funcionamiento del algoritmo o el anonimato que permiten las redes. Se exploran vías de reducción de los discursos de odio en redes sociales, enfocadas en el marco legal, la educación y la promoción de la salud mental.

Palabras clave: discapacidad; capacitismo; redes sociales; TikTok; internet; discurso de odio; haters.

1. Introduction

Hate speech is not a recent phenomenon (Losada-Díaz et al., 2021). The United Nations (UN) (2019, p. 3) understands hate speech as “any form of oral, written or behavioral communication that is an attack on a person or group on the basis of religion, ethnicity, nationality, race, color, descent, gender or other identity factor.” These speeches encompass both expressions that induce to commit discriminatory or violent acts for some xenophobic, racial, sexual orientation or disability reason (Cabo-Isasi and García-Juanatey, 2016) and manifestations that promote intolerance and prejudice towards certain social groups, as they create a hostile environment that can foster discrimination and violence (Gagliardone et al., 2015). This phenomenon is based on several polarizing components, such as cultural identity, gender, aporophobia, migration and democratic values (Fuentes-Lara and Arcila-Calderón, 2023).

Bustos-Martínez et al. (2019) consider that we are in the presence of hate speech if the following criteria are met: 1) it refers to a vulnerable or discriminated against group; 2) there are representative symbols that humiliate this social group; 3) there is malice and an invitation to denigrate these people; and 4) if there is an evident intention to discriminate against, humiliate and exclude them. In this framework, the group with disabilities, which has historically experienced rejection and exclusion (Santana-Valencia, 2022), is established as one of the current targets of hate speech (Muniesa-Tomás et al., 2022). Likewise, there is international evidence that children, teenagers and young people with disabilities are more exposed to violence or abuse, which violates their rights and has a negative impact on their development (UNICEF, 2022).

Hate speech can become a felony, establishing itself as “one of the main threats to peaceful coexistence in our societies” (Fuentes-Lara and Arcila-Calderón, 2023, p. 226). The Internet has facilitated its spread, especially through social networks, leading to a problem of global dimensions by generating conflicts and violence in the digital arena, but also in the offline context. Specifically, 1,869 criminal offenses and hate incidents were investigated in Spain in 2022, which is 3.7% more than the previous year (Muniesa-Tomás et al., 2022).

This study investigates hate speech against people with disabilities on TikTok by analyzing the video replies posted by the users themselves on the social network and structured interviews that give a first-person voice to these profiles. The relevance of the object of study lies in the topicality of the subject matter and the need to give visibility to this type of practices, in order to analyze and reflect on the social and exclusionary implications that they could have on the group.

1.1. Hate speech on social networks

In 2022, 4.76 billion people had a profile on social networks, representing 59.4% of the world's population (We Are Social, 2023). Among these platforms, according to the same report, TikTok stands out as the network where users spend the most time (more than 23 hours/month on average) and it is especially attractive to young people. In this context, and taking into account that the virtual world is an ideal ground for emotional content, the Internet and social networks have proven very efficient in the propagation of these violence-inducing messages, generating a worrying global problem (Cermi, 2021).

Digital platforms and social networks have transformed the engagement of citizens in the Internet era, posing new challenges to the role of the prosumer (producer and consumer of digital content). This has created new forms of representation and construction of digital identity, which significantly influences social interactions (Aguaded et al., 2022). However, there may also be certain risks associated with them despite the opportunities they offer, due to the permanence of content, the itinerant nature of platforms, transnationality and the publication of content from anonymous profiles or under pseudonyms (Gagliardone et al., 2015). Likewise, anonymity and emotional detachment, derived from the physical distance between participants, reduce empathy and create a sense of security to make comments without apparent consequences (Bustos-Martínez et al., 2019). In this sense, Pahor de Maiti et al. (2023) point to the emergence of haters, issuers of hate speech to certain individuals or social groups because of their ethnicity, religion, sexual orientation or gender, etc.

Undoubtedly, hate messages can cause significant emotional or psychological damage at the individual level, leading to depression or even suicide (Phanomtip et al., 2021). They can also cause indirect harm by damaging the reputation and dignity of members of the discriminated group (Cermi, 2021). Hate speech, generally composed of negative comments, insults, belittling and mockery (Martínez-Valerio, 2022), favors the perpetuation of stereotypes, stigmatization, marginalization and even dehumanization of certain groups (Cabo-Isasi and García-Juanatey, 2016; Raffone, 2022). Moreover, this occurs in a social environment with high rates of mental health problems that especially affect teenagers and young people (Westberg et al., 2022), increased in the wake of the COVID-19 pandemic (Cenat et al., 2022).

Although hate speech can constitute a crime, the paradox is that, although its dissemination through the media or social networks is an aggravating factor, the limits to freedom of speech are also very hard to determine, which is why it is difficult to convict people for these acts (Cermi, 2021; Ramírez-García et al., 2022). Thus, despite growing evidence in the rise of cyber hate, this type of speech goes largely unpunished, also due to the difficulties in monitoring online spaces (Burnap and Williams, 2016). A recent study by Weimann and Masri (2023) highlights the ineffectiveness of TikTok in monitoring and controlling hate speech related to far-right groups, highlighting the need to push for legislation that would force the platform to improve its monitoring practices and exercise more effective control. Henson (2023) highlights the existence of legal loopholes and advocates for national and international cooperation that actively involves social networks in the creation of a coherent and comprehensive legal framework, thus facilitating effective detection and adequate control of hate speech. In line with these findings, Chetty and Alathur (2018) conclude that a joint effort among governments, internet providers, and social networks is essential to implement effective policies to fight both hate speech and terrorism.

In this context, and in the face of the challenges posed by the regulation of these platforms, Sabater et al. (2021) stress the importance of education in the formation of a critical citizenship capable of unraveling the hate speech present in social networks and other online sites, as well as of reconstructing them as counter narratives. It is especially important in the case of teenagers and young people (Gámez-Guadix et al., 2020), a social group with high participation in platforms in general and in TikTok in particular.

Hate speech in social networks is the manifestation of a deeper social and structural problem. Cabo-Isasi and García-Juanatey (2016) warn of how all intolerant ideologies have found in these platforms a privileged space of expression, where they enjoy impunity for anger and hatred, to the point of constituting a problem in European and international organizations' agendas. Ramírez-García et al. (2022) refer to their normalization as a fertile ground for radicalization, social polarization or the proliferation of violent incidents, and defend the need to take comprehensive and preventive measures.

1.2. Disability and hate speech: from the offline world into TikTok

Disability has been constructed between the “medical” model, which conceives it as the result of the inability of the body's functions to work normally, and the “social” model, which considers it as a disadvantage created by society that can generate social exclusion (Raffone, 2022). Thus, the social model suggests that for the elimination of sociocultural barriers, we need to consider changes aimed primarily at society and not at the individual (Fisher and Purcal, 2016). In this line, and within these necessary changes, Figuereo-Benítez et al. (2023) point to the importance of increasing the visibility of the group's and promoting its representation as an independent citizenry with full rights. Likewise, the issue of self-representation or visibility of disability is one of the main values of social networks for vulnerable groups or those at risk of exclusion, as it enables them to tell their own story (Bonilla-del-Río et al., 2022).

However, these virtual environments are also being used to disseminate messages of hate and discrimination against different groups, including people with disabilities. These are cases of cyber hate, responsible for a negative and significant impact on their well-being and involving more people with visible disabilities (Alhaboby et al., 2021). Trolling and flaming are probably the most common forms of hate against these people in the digital space (Sherry et al., 2019). According to this study, these narratives seek to provoke or offend, as well as divert attention to superfluous and meaningless discussions, suffered especially by groups experiencing social inequality. Likewise, messages are often directed towards this group from ableism, understood as a form of discrimination that judges people based solely on their apparent physical, psychological or intellectual capacity. This misguided thinking induces to anticipate behaviors in an inflexible manner, without prior knowledge of the person, and can manifest itself in a hostile manner, with violence or rejection, either directly or indirectly, or adopting a benevolent form by offering help and advantages that, in fact, would be unnecessary (Sánchez-Rojo, 2023).

In the specific case of TikTok, Raffone (2022) exposes how intolerant attitudes towards disability are underpinned by deeply rooted mental models and beliefs. Based on the medical model, they deal with disability from a “normal- abnormal” perspective, which favors the perpetuation of stigmatization and the propagation of hate speech. According to this study, the result is the construction of disability, also in TikTok, through the process of “otherness”.

Research in this area is still incipient, due to the fact that, among the various forms of discrimination, hate speech against people with disabilities is an area that has received less interest from academia and is, therefore, underrepresented compared to other social groups (Raffone, 2022). Furthermore, there is the fact that research on TikTok is still in an embryonic state, as this platform arrived in Spain recently, in 2018, and was undervalued thinking that it would only be about people doing silly dance moves (Figuereo-Benítez et al., 2022).

However, some research points to the fact that propagating hate speech on TikTok might be easier than in other platforms, due to its particular and distinctive characteristics (Weimann and Masri, 2023). Whereas in other networks users see the content of those profiles they follow, on TikTok, content is displayed based on users' actions and interactions through an algorithmic system that generates personalized recommendations (Boeker and Urman, 2022), but also random ones. Likewise, the peculiarity of this platform lies in its “For you” feed, which shows popular content on TikTok, but also videos with low views, promoting greater equality with respect to other social networks (Jain and Arakkal, 2022) and allowing videos to reach a more varied audience and to cause greater reactions, both positive and negative. Its popularity among teenagers and young users (Muñoz-Rodríguez et al., 2023) and the lack of control over content (Weimann and Masri, 2023) have also been pointed to as causes that could favor the proliferation of hate content on this network.

Thus, despite the potential of the digital context for the visibility and self-representation of disability, discrimination against this group in general, and hate crimes in particular, persist still. According to the latest report of the Ministry of the Interior on hate crimes in Spain (Muniesa-Tomás et al., 2022), between 2020 and 2022, 95 hate crimes were committed against people with disabilities. Likewise, 590 hate crimes and incidents were registered on the Internet and social networks, 21 of them against people with disabilities. However, according to the estimate of the European Union Agency for Fundamental Rights (FRA) (2021), 9 out of 10 people do not report these crimes, whether they are violent experiences due to racism or intolerance, religion, sexual orientation or disability.

2. Methodology

The importance of social networks and their relationship with social movements poses the challenge of researching the quantitative and qualitative impact on certain groups and collectives (Arévalo-Salinas, 2018). This approach involves analyzing the speech of the speakers, the reply of the recipients and the strategies aimed at fostering a debate on current structural and social problems. This research follows the proposal of Arévalo-Salinas (2018), adapted to study TikTok. Thus, a mixed approach is employed through the application of two research techniques, content analysis and structured interviews, which complement each other and are set out in detail in the following subheadings. This methodological approach pursues the following objectives:

O1. To analyze the video replies of users with disabilities to hate messages received on TikTok, considering aspects such as content, reach, interaction, engagement, tone and intentionality of the video itself.

O2. To delve into the perspective of users with disabilities who post video-replies regarding the hate speech received and their stance towards this phenomenon.

2.1. Research techniques

First, in relation to the first objective, the authors perform a content analysis of 64 video replies to hate messages received by people with disabilities on TikTok, posted by 32 different users (23 women and 9 men)

To select the sample, the hashtag #discapacidad was used in conjunction with one of the following: #odio, #hate, #ciberodio, #bullying, #ciberbullying, #hater, #haters, selected through a previous process of digital ethnography and participant observation (Angrosino, 2012). To this end, an active and engaging scan of behavior and interactions related to hatred towards people with disabilities on TikTok was conducted in order to better understand the dynamics, trends and expressions related to this phenomenon. This approach allowed for a more informed selection of the hashtags used to collect video replies.

From the results obtained, only videos that met the following requirements were selected: 1) to be made by people with disabilities; 2) to be replies to specific hate comments or messages; 3) that the original hate message is explained or shown in the video. All the video replies and profiles selected for the study correspond to Spanish-speaking users from Spain and Latin American countries.

Nine different variables were analyzed on this sample, in addition to the classification of the profiles by sex. Five of these variables focused on measuring the interaction generated by the video reply and the rate of engagement achieved by the content, while the remaining four were used to analyze the content of the publications.

From the first group, four variables correspond to the interaction options provided by TikTok: 1) number of followers; 2) number of likes; 3) number of comments; 4) number of times shared. These data were obtained directly from the platform. In addition, the fifth variable corresponds to the engagement index on TikTok, calculated by the following formula: (likes + comments + shared) / (total followers) x 100 (Bressler and Zampella, 2020).

From the second group, corresponding to content-related variables, the authors measured: 1) the content of the hate message received; 2) the tone of the video reply; 3) the intentionality of the video reply; and 4) the presence of one's own speech. First, for the variable related to the content of the hate message received, the authors established the following categories, as described in Martínez-Valerio (2022), expanded and adapted to this study:

Criticism: Negative statements that aim to establish a value judgment of the recipient and sometimes inflict harm or offense.
Insults: Words that directly offend or humiliate a person, especially with hurtful messages, always with a negative connotation.
Belittlement: Negative attitude that considers someone below his or her real value and reflects contempt. In this study, contempt is related to ableism, considered as the specific undervaluation and discrimination against people with disabilities.
Threats: Comments that refer to danger or possible risk by means of a warning that arises from an event that has not yet happened.
Mockery: Actions or words that are intended to ridicule.
Disability denial: Comments that deny the existence of the subject's disability or minimize the effects exposed by users.
Other: Messages whose content does not correspond to any of the above categories.

The tone of the reply is understood as the speaker's attitude towards the message received (Phelan, 2014). The authors divided this variable into the following categories:

A. Humorous: Resource whose objective is to cause laughter. For this research, the authors rescue the concept of humor as an action whose social purpose is to laugh “by way of protest against those who try to break universal rights and values” (Cortés-Kandler and Arroyo-Carvajal, 2021, p. 80) and that functions as social criticism (Rodríguez-Montemayor, 2020).

B. Ironic: Resorting to its origin as antiphrasis, it refers to the use of words or intonation to give a different or even opposite meaning (Caro-Lopera et al., 2018). It can be employed with positive, negative or neutral charge. Currently, non-linguistic indicators of irony go beyond voice intonation and include others such as memes, emojis, laughter, emoticons or punctuation marks (Cortés-Kandler and Arroyo-Carvajal, 2021).

C. Aggressive: Belligerent style in which the person is explosive, hostile or angry. It is characterized by contempt and humiliation of the other (Briones and Ortiz, 2014).

D. Assertive: Effective transmission of positions, opinions, beliefs and feelings, expressing emotions towards another person in an appropriate manner (Briones and Ortiz, 2014).

E. Other: Intonations or emotional charge not corresponding to the previous categories.

The intentionality of the answer refers to the main objective of the user with his or her video reply. For this variable, the authors adapted the categories set forth by Bonilla-del-Río and Soares (2021), adding the last two after the first viewing of the sample:

A. Visibilization: The action of making people with disabilities visible.

B. Awareness: Action to make society aware of these people's reality and the barriers and speeches present in society.

C. Vindication: Action of claiming for the rights of people with disabilities, arguing the need to guarantee them effectively.

D. Denounce/Criticism: Action of publicly exposing the hate speech received.

E. Other: This category includes messages whose purpose does not correspond to the previous ones.

Finally, the variable of one’s own speech refers to the presence or absence of speech in oral or written expression by the user him/herself. It is related to the diversity of TikTok content, where one can find videos with original speeches, but also numerous dance videos (Suárez-Álvarez and García-Jiménez, 2021), challenges (Ahlse et al., 2020), scenes with audios that are trending at the time or acted lip syncing (Mackenzie and Nichols, 2020). In these cases, the authors considered that, although there was intentionality, the user was not communicating a speech of his or her own.

The coding of the sample involved a first complete viewing of all the units of analysis, guaranteeing a correct categorization of the variables. Subsequently, two coders performed the coding independently, and Krippendorff's Alpha coefficient was calculated on 100% of the total analyzed to assess reliability, obtaining a value of 0.92 (Krippendorff, 2004). The data were coded in Excel format and processed in SPSS v.29 software for analysis. Given that some variables could change over time, coding was performed on June 9th, 2023. To establish the existence of connection between variables, Pearson's correlation was used for nominal variables (Islam and Rizwan, 2022) and Spearman's correlation (Schober et al., 2018) for ordinal, scale or mixed variables.

To cover the second research objective, the authors applied a qualitative approach by conducting structured interviews. They used three axes as reference to elaborate the interview script, extracted from Alhaboby et al. (2017): 1) identification of the content of hate speech, by the people who receive it; 2) attitudes, emotions and opinions towards hate speech; and 3) effects of hate speech and intention with which it is engaged in counter-speeches or reports.

The authors proceeded to contact the 32 users in the sample analyzed (Spanish-speaking people with disabilities who issued the 64 video replies to hate messages on TikTok) and asked whether they were willing to participate in the structured interviews. Contact was established through their social networks, preferably TikTok or Instagram. In all cases, they were informed of the research objectives and, once they agreed to participate in the study, they received the complete interview script, along with the possibility of responding through the social network or by videoconference to facilitate their participation. Finally, 14 people (10 women and 4 men) with different types of disability agreed to the interview. Eleven interviews were conducted through social networks and three by videoconference. The interviews were conducted between June and July, 2023. The complete interview script and answers can be found at: https://bit.ly/48bwdLR

In conclusion, in line with Arévalo-Salinas (2018), this methodological approach studies the three areas involved: 1) speaker, by analyzing the profiles through TikTok's own data and structured interviews; 2) the message-speech, studied through content analysis; 3) reactions of the recipients, by measuring the interactions and engagement of the selected sample.

3. Results

3.1. Content analysis

The authors analyzed a sample of 64 TikTok video replies posted by 32 users with disabilities in response to hate messages received on this social network. None of these profiles corresponds, according to the classification proposed by Alassani and Göretz (2019), to the category of mega-influencers, since none reaches more than 1,000,000 followers. As for macro influencers, they constitute 46.87% (n=15) of the total number of profiles, encompassing those with an audience ranging from more than 100,000 to 999,999 followers. Micro influencers represent another 50% (n=16) of the set, being those with followers ranging from 1,000 to 99,999. Finally, Nano influencers, with a maximum number of 1,000 followers, make up 3.13% (n=1) of the total number of profiles, with a more limited reach.

Interaction is most often reflected in likes, followed by comments and shares. Specifically, 40.63% (n=26) of posts do not exceed 1,000 likes, 42.19% of video replies (n=27) reach between 1,000 and 10,000 likes and 17.18% (n=11) exceed 10,000 likes. Data regarding the profiles and interactions of the video replies are available at: https://bit.ly/3vhloJo

When examining the descriptive statistical data related to interaction and engagement, summarized in Chart 1, one can observe a notable difference in the scope of the analyzed content. The engagement rate shows a wide variation, as does the number of followers, indicating that hate messages are not limited only to profiles with a large influence, but also target users with a significantly smaller following.

Chart 1. Descriptive statistics on interaction achieved in the video replies.

	Range	Minimum	Maximum	Total	Average	Standard Deviation
Followers	650,401	299	650,700	5,312,103	166,003,22	180,501.4119
Likes	277,275	25	277,300	862,618	13,478.41	41.269.7051
Comments	11,200	0	11,200	32,446	506.969	1,614.1432
Times Shared	1,501	1	1,502	4,485	70.08	234.651
Engagement	120.473	.060	120.533	465.402	7.272	17.294

Source: Elaborated by the authors.

One can observe a moderate positive correlation between the number of followers and the extent of interactions in the form of likes and comments, while the correlation in relation to the number of times shared was weak. In addition, the authors found a strong positive correlation between the number of likes, comments and times shared, and between these forms of interaction and the engagement index. However, the authors identified a negative yet weak correlation between engagement and the number of followers. These data, shown in Chart 2, indicate that, although the number of likes, comments or times shared play a determining role in the engagement index, the number of followers does not have the same influence.

In relation to sex, differences can be seen both in the number of followers, where the average (x̄) is higher in men (x̄=252980) than in women (x̄=163378.87), and in the engagement index, which is higher in women (x̄=10.167) compared to men (x̄=2.324).

Chart 2. Spearman's correlation between interaction variables.

	Value	Sig. (bilateral)
Followers * Likes	.439**	<.001
Followers * Comments	.478**	<.001
Followers * Times Shared	.205	.104
Followers * Engagement	.241	.055
Likes * Comments	.765**	<.001
Likes * Times Shared	.765**	<,001
Likes * Engagement	.643**	<.001
Comments * Times Shared	.713**	<,001
Comments * Engagement	.512**	<.001
Times Shared * Engagement	.638	<.001
Sex * Followers	-.368**	.003
Sex * Engagement	.363**	003
*. Significant correlation at the 0.01 level. . Significant correlation at the 0.05 level.

Source: Elaborated by the authors.

The analysis focused on both the content of the hate speech received and the reply issued. The hate speech received consisted mainly of mockery in 34.4% of the cases (n=22). These messages included, for example: “watch my bike”, sent to a user with visual disability; “looks like a transformer”, to a user with motor disability; “you overindulged in cigarettes because of that” (sic), to a woman with tracheotomy; or “I remember when I played Mario Bros when I look at you” (sic), to a user with cerebral palsy. This type of content is followed by belittling in 21.9% (n=14), related in all cases to ableism. Comments such as “you are worthless”, “your only talent is to be immobile”, “but I can walk” or “the disability, as the name itself indicates, is that you are less capable than others, now that you strive to be more normal (sic) is up to you”.

In addition, the authors found criticism (20.3%; n=13) such as “you don't speak well”, “you have very thin skin” or “no one is going to feel like pushing your wheelchair up a half-kilometer hill”. The authors also found disability denial in 14.1% (n=9), in messages such as “you are making that up”, “you are full of it”, “I don't believe it at all”, “fake” or “don't pretend disability for visits”. In smaller proportions, the authors found insults as well (9.4%; n=6) such as “faggot and lame”, “ridiculous, you're half asleep, open your other eye” or “disabled, liar, you like to eat, you're just being defensive”.

In the analysis of the content of the video replies, the authors evaluated tone, intentionality and the presence of one’s own speech. First, regarding tone, they observed that 34.4% (n=22) of the videos used an assertive tone, followed by irony present in 26.6% (n=17). The authors detected an angry or aggressive tone in 17.2% (n=11) of the video replies, a humorous tone in 15.6% (n=10), and other tones in 6.3% (n=4) of the videos.

Second, in relation to the purpose, 46.9% (n=30) of the video replies aimed at publicly denouncing the hate message received. To a lesser extent, the authors found videos aimed at awareness-raising (21.9%; n=14), visibility (15.6%; n=10) and vindication (7.8%; n=5). Some 7.8% (n=5) of the videos did not correspond to the above options and were therefore coded in the “other” category.

Finally, the authors observed that 82.8% (n=53) of the video replies presented the issuers' own speech, while only 17.2% (n=11) lacked it. This indicates that, in the sample analyzed, hate messages are publicly responded to through this video format with the intention of conveying a message of their own, either in the form of text or orally. The videos that did not have their own speech were limited only to showing the hate message accompanied by a song, generally popular at the time, or used external audios to perform a lip sync. All video replies lacking original oral or textual speech were delivered by women.

The tone of the video reply is also related with the content of the hate message and the presence of one’s own speech (Figure 1). Humorous tone is used more frequently when being mocked. The assertive tone appears in all cases, but especially in the face of criticism. On the other hand, when dealing with a comment that includes contempt, an aggressive tone is used more frequently.

Figure 1. Tone of the reply according to the hate message received.

Source: Elaborated by the authors

Source: Elaborated by the authors.

Likewise, Figure 2 shows the correlation between the tone and the intentionality of the video reply. Assertive tones predominate when the objective is to raise awareness, but ironic and aggressive tones stand out when the purpose of the video reply is to denounce or criticize the message received.

Figure 2. Tone used according to the intentionality of the video reply.

Source: Elaborated by the authors.

The intentionality of the reply and the presence of one’s own speech are also correlated variables, as shown in Chart 3, because those videos in which the user does not create a new original speech focus exclusively on denouncing or criticizing the message received. For the rest of the objectives, users contribute new speech of their own, either orally or in writing.

Chart 3. Pearson's correlation between variables on content

	Value	Sig. (bilateral)
Sex * Hate Message Content	.112	.378
Sex * Intentionality	.160	.207
Sex * Own Discourse	.365**	.003
Hate message content * Own Discourse	.146	.249
Tone * Hate Message Content	-.244	.052
Tone * Intentionality	-.328**	.008
Intentionality * Hate message content	.409**	.001
Intentionality * One’s Own Speech	.260*	.038
*. Significant correlation at the 0.01 level. . Significant correlation at the 0.05 level.

Source: Elaborated by the authors.

Once the results related to the interaction and content variables were described separately, the authors proceeded to analyze the correlations between both variables (Chart 4). In these cases, the average (x̄) is the reference instead of the total, since there is a different number of video replies in every category. There is correlation between sex and the number of likes, where women (x̄=18315.92) receive more likes compared to men (x̄=5928.36). Likewise, sex and comments are also correlated, with more comments in females (x̄=515.64) compared to males (x̄=493.52).

In addition, there is a moderate correlation between the content of the hate message and the number of likes received. Disability denial messages have a higher average number of likes (x̄=47707.67) compared to criticism (x̄=7592.31), belittling (x̄=7541.13), mockery (x̄=9196.23) and insults (x̄=4427.33). The tone of the reply is also correlated to the number of comments, with videos with an aggressive (x̄=796.91) and assertive (x̄=785.91) tone being those with the most comments, with a clear difference with respect to humorous (x̄=292.1), ironic (x̄=142.59) and those using other tones (x̄=261.75). In contrast, the intentionality of the reply was not shown to have an influence on the number of interactions. Finally, the presence of one’s own speech shows a weak-to-moderate correlation with the number of likes, comments and times shared. The authors observe that video replies with original speech receive more comments (x̄=606.34 vs. x̄=28.36) and are shared more times (x̄=83.32 vs. x̄=6.27) compared to those lacking one’s own speech. However, video replies without one’s own speech receive more likes on average (x̄=17418.45 vs. x̄=12659) compared to those with one’s own speech.

Chart 4. Spearman's correlation between variables on interaction and content.

	Value	Sig. (bilateral)
Hate Message Content * Likes	.202	.100
Tone * Comments	.189	.134
One’s Own Speech * Likes	-.256**	.042
One’s Own Speech * Comments	-.448**	<.001
One’s Own Speech * Times Shared	-.381**	.002
*. Correlación significativa en el nivel 0,01. . Correlación significativa en el nivel 0,05.

Source: Elaborated by the authors.

3.2. Structured interviews

The authors conducted 14 structured interviews among the profiles included in the content analysis sample. First, users were asked whether they usually received hate messages on social networks, to which all profiles responded affirmatively, although with a very unequal perception of frequency or intensity, ranging from those who admit “that they were very few” (Interviewee 1; hereafter E1) to those who say they receive “more hate than you could imagine” (E3). According to the profiles interviewed, the more users are exposed, the more followers they have, the more controversial the content or the more repercussion a video has, the greater the chances of receiving hate messages. However, the study included profiles that had from 299 to more than 650,000 followers, so although hate increases when exposure is greater, the number of followers does not seem to be a crucial variable when it comes to receiving hate, as revealed by the content analysis results.

Regarding the content of the hate messages, the interviewees acknowledge receiving numerous comments related to their bodies. Ableist messages are frequent, as well as insults and mockery related to disability. Several participants relate ableism to lack of knowledge or stereotypical representation of people with disabilities. Another user (E2) accuses organizations of people with disabilities of presenting a biased and stereotypical view that exploits pity: “So when other people see individuals with disabilities living normal lives, working and participating in everyday activities, it challenges their misperceptions.”

Something similar happens in the profiles of people with autism, who unanimously reveal that they receive messages that deny their disability or accuse them of “romanticizing” the disability (E4). In addition to this common experience, there are also comments that point to the difficulty of finding a partner, having sexual relations or having children as a person with a disability. Likewise, there is consensus among people who also belong to the LGTBIQ+ group in the reception of more aggressive hate messages related to this last reason. One of the interviewed profiles (E3) acknowledges having even received death threats.

According to the interview, 12 out of the 14 participants acknowledge having received more hate messages through TikTok, although they also mention other networks such as Instagram or Facebook. In addition to the argument of the greater number of followers, there is the very way TikTok's algorithm works, which shows the videos to a wider audience, and not only to the followers or the community created by the user. Some of the interviewed profiles (E1, E2 and E7) agree in describing this network as “toxic”, although the higher level of hate speech on TikTok is also attributed to its greater virality. Several interviews talk about the presence of more young people with little knowledge about disability to justify the toxicity of the platform and hate speech. Only one person claims not to have received hate through this network, and only one through others such as Instagram. Regarding the origin of the messages, several profiles claim to receive mainly hate content in public from anonymous profiles, while others acknowledge receiving these messages mostly privately. Specifically, interviewee 2, a trans woman with visual disability, acknowledges receiving numerous private messages that show explicit “more verbal violence”, whose purpose is to strike “fear” in her.

When asked about how receiving hate speech affects them emotionally, many interviewees agree that time has made it affects them less, although there are clear differences. Some claim not to have noticed any negative consequences, although most of them acknowledge that it has at some point affected their self-esteem, caused depression, sent them to therapy, made them consider stopping creating content, or temporarily caused them to abandon social networks. One interviewee mentions feelings of helplessness, anger and rage. In addition, some participants acknowledge that the impact depends on their emotional state and the given date. Answers to this question underlines the relationship of hate speech to self-esteem and mental health in understanding the emotional impact on those on the receiving end.

At another point in the interview, and when asked about their opinion of haters, most reflect a negative perception. Several interviewees mention concepts such as “low self-esteem”, “envy” and describe haters as “very boring or sad people”, “with nothing to do” or “damaged”. They are characterized as “disengaged and insecure people who try to project their own insecurities onto other people” (E14). Two users relate the presence of haters on social networks to the idea that these platforms act as a channel for society to vent its frustration, and are a reflection of social problems resulting from ableist, fascist, homophobic, racist or misogynist thoughts and behaviors. These opinions reflect the importance of not limiting ourselves to addressing hate speech only through social network control policies or punitive measures, but rather looking for the roots of hatred.

Users recognize that they are selective when responding to hateful comments and, when they do so, their objectives range from educating, outreaching or teaching until they can finally curb hate. They also aim to correct ableist speech and false arguments, identify violence, demonstrate their ability to defend themselves, and even “make fools of haters” (E9). Interviewee 12 defines it as a “therapy” because it allows her to expose her feelings and receive support. The coincidence in the answers of users E2 and E5 is striking. Both profiles belong to people with disabilities and members of the LGTBIQ+ community, who claim to reply when they can “turn around” the message received. Only one user (E11) manifests a different strategy, as she reply to these hate messages in her own comments to increase the visibility of her content. Likewise, interviewee 7 describes her intention in replying to a hateful message as follows:

Let us not allow this message to have the power of symbolic violence. I am mature enough to see the root and the reason for that message, but TikTok is an application full of teenagers and there are many who do not know how to analyze it and I do not want a girl to read it and feel that what they are saying to me is valid. If we keep quiet in the face of bullies, what we do is to perpetuate the problem. There will always be bullies and haters, the real problem is that the rest of us are silent and therefore give power to that comment. So, by answering, what I want to make clear is that those comments have no power and make it clear that many of the arguments they use are false and worthless.

When asked whether they report accounts that send them hate messages, five people said they do not, while eight said they do, although with differences in their approaches. Among those profiles that choose not to report, two reject this idea outright, another acknowledges to have done it in the past but not anymore, and two others argue that it is pointless because TikTok does not consider such content inappropriate. On the other hand, those who do choose to report or block accounts show some similarity in their profiles, with three young women with autism explaining that they only do it when it is too offensive or it “crosses the line” (E6). Other interviewees do it when there are insults, community norms or human rights violations. In addition, one participant acknowledges reporting offensive comments directed at other people with disabilities, in contrast to another user who limits herself to doing so only in cases of bullying, harassment or sexual harassment that she receives through direct messages.

The interviews outline three main ways to fight against hate speech. The first one is to denounce, block and prevent disseminating hatred-promoting, whether towards people with disabilities, LGTBIQ+ or others. They also demand greater controls on algorithms and monitoring by TikTok and other networks, as well as controls for creating accounts on these platforms, since many messages that incite hatred come from anonymous accounts. Participant E13 notes that social networks should be for adults only, while profile E7 argues that effective legal avenues must be created: “It cannot be illegal to threaten someone on the street and legal to do it on the internet. We have already reached the point where the internet is part of our reality so it is time to legislate online as well.”

The second line of actions to combat hate speech focuses on education, both in the family and at school. The interviews highlight the importance of “educating in values of diversity and inclusion” (E5) and of exposing all citizens from an early age to the diversity of bodies, people, realities, families... in order to favor “social inclusion” and “empathy” (E4; E14). Respect and non-questioning of other people is also fundamental for user E2, who applies it to disability, but also to the LGTBIQ+ community. She considers it essential to raise awareness about disability with an adjusted and realistic image, which does not exclusively promote compassion or “pity”.

Finally, the interviewees point to a third line of actions to combat hate speech, which is framed in the promotion and care of mental health. The fact is that some profiles link the hate messages they receive with self “insecurities” (E11) and “emotional wounds” (E14) of those producing the messages. Participant E1 sums it up as follows: “If we had good mental health, people would not go to social networks to vent or to discriminate against others.”

4. Discussion and conclusions

The analysis of the video replies showed, first of all, that it is not necessary to be a user with a large number of followers to become the target of hate speech on TikTok, since even profiles with a few hundred followers receive this kind of messages. Interviews point to a greater amount of hate speech on this network compared to others, and even describe it as “toxic”. Other than exposure or number of followers, the interviews offer many reasons. For example, they highlight other causes exposed in previous scientific studies such as the functioning of the TikTok algorithm (Boeker and Urman, 2022), or the immaturity of the users of this social network, which is very popular among teenagers and young people (Muñoz-Rodríguez et al., 2023). The reasons are completed with the lack of control and the way users feel protected by freedom of speech in the platform. There is also the use of anonymous accounts that allow the dissemination of hate messages without the possibility of identifying their authors (Raffone, 2022), which hinders their prosecution and promotes their proliferation (Chetty and Alathur, 2018).

The hate speech received by the profiles addressed both in the content analysis and in the interviews includes not only negative comments, insults, belittling or mockery (Martínez-Valerio, 2022), but also ableism and/or disability denial. The latter accounts for a large part of the hate speech that directly affects people with disabilities, either individually or collectively, and justifies the need for research that focuses specifically on this social group. In addition, video replies to hate speech denying disability reached a higher average number of likes than others, although a more specific and detailed analysis is required to better understand this phenomenon.

The study highlights how profiles with disabilities, when replying to hate speech through TikTok, do so mainly to educate, raise awareness and outreach, or to correct ableist speech and false arguments. Along with this motive, comes the denounce and exposure of the hate message, identification of violence, and self-defense itself. This finding ratifies the potential that social networks offer for vulnerable groups to tell their own story (Bonilla-del-Río et al., 2022) It goes even further by enabling the creation of a counter-narrative.

Through structured interviews, the article addresses three key approaches to combat hate speech in social networks. The first one is precisely education, especially in the case of teenagers and young people, as proposed by Gámez-Guadix et al. (2020). This strategy pursues the formation of citizens with a critical thinking, capable of unraveling the hate speech in networks and other digital platforms, and reconstructing them as counter narratives (Sabater et al., 2021).

Likewise, participants highlight the importance of regulation as a mechanism to control hate content. Previous studies have demonstrated the existence of legal loopholes in this area (Henson, 2023), as well as the ineffectiveness of TikTok for a proper self-regulation in monitoring and reducing hate speech (Weimann and Masri, 2023). These works advocate national and international cooperation to push for standards that force networks to improve their monitoring techniques and exercise more effective control. Governments, internet service providers and social networks are established as necessary agents to implement truly effective policies in the face of hate speech (Chetty and Alathur, 2018). This study, through interviews with victims of hate speech, proposes specific measures such as the regulation of anonymous accounts, age restrictions and implementation of punitive measures.

Finally, a third strategy for dealing with hate speech focuses on the mental health of both the victim and the hater. Several interviewees acknowledge that these contents negatively affect their self-esteem, generate negative emotions and often force them to periodically disconnect from social networks. These consequences match with those presented in previous literature (Phanomtip et al., 2021; Raffone, 2022) in relation to victims of hate speech. However, participants also relate mental health problems to the haters themselves and to the high rates of mental health problems detected in Western societies, which have increased in the wake of the COVID-19 pandemic (Cenat et al., 2022) and especially affect teenagers and young adults (Westberg et al., 2022). Thus, they consider that promoting mental health care in society in general could be beneficial to counteract these contents.

In conclusion, this research contributes to the visibility of a problem around hate speech, discrimination and derogatory messages on social networks against users with disabilities. Analyzing their own video replies shared on their TikTok profile, as well as their perception and opinion about this phenomenon through interviews, allows the authors to delve deeper into how people with disabilities are affected by this type of practices and the strategies that could be adopted by platforms and governmental or educational institutions to achieve a more empathetic, fair and inclusive society. Therefore, understanding the reactions of users with disabilities is not only relevant at an academic level, but also from a social point of view.

Future lines of research include the possibility of extending the study to other social networks, in order to identify similarities and differences that could occur in the various platforms with respect to the type of replies generated by users with disabilities to negative comments or hate speech. It would also be possible to observe the evolution of the phenomenon and analyze whether society tends more or less to producing hate speech against the group. Likewise, it would be of great interest to delve into the regulations of the transnational platforms themselves on hate speech, in order to analyze the policies and conditions established and their compliance.

5. References

Aguaded, I., Civila, S., & Vizcaíno-Verdú, A. (2022). Paradigm changes and new challenges for media education: Review and science mapping (2000-2021). El Profesional de la información, 31(6), e310606. https://doi.org/10.3145/epi.2022.nov.06

Ahlse, J., Nilsson, F., & Sandström, N. (2020). It’s time to TikTok: Exploring Generation Z’s motivations to participate in #Challenges. [Doctoral dissertation, Jönköping University]. https://bit.ly/3PkoDpX

Alassani, R., & Göretz, J. (2019). Product placements by micro and macro influencers on Instagram. In G. Meiselwitz (Ed.). Social computing and social media (pp. 251-267). Springer. https://doi.org/10.1007/978-3-030-21905-5_20

Alhaboby, Z. A., Al-Khateeb, H. M., Barnes, J., Jahankhani, H., Pitchford, M., Conradie, L., & Short, E. (2021). Cyber-disability hate cases in the UK: the documentation by the police and potential barriers to reporting. In H, Jahankhani; A, Jamal & S. Lawson (Eds), Cybersecurity, Privacy and Freedom Protection in the Connected World: Proceedings of the 13th International Conference on Global Security, Safety and Sustainability (pp. 123-133). Springer. https://bit.ly/3RzJXJ4

Alhaboby, Z. A., Barnes, J., Evans, H., & Short, E. (2017). Challenges facing online research: Experiences from research concerning cyber-victimisation of people with disabilities. Cyberpsychology: Journal of Psychosocial Research on Cyberspace, 11(1). https://doi.org/10.5817/CP2017-1-8

Angrosino, M. (2012). Etnografía y observación participante en investigación cualitativa (Vol. 3). Ediciones Morata.

Arévalo-Salinas, Á.I. (2018). Propuesta metodológica para el análisis de YouTube y su relación con los movimientos sociales. En J. Candón-Mena (Coord.), Actas del II Congreso Internacional Move.net sobre Movimientos Sociales y TIC (pp. 25-34). Grupo Interdisciplinario de Estudios en Comunicación, Política y Cambio Social de la Universidad de Sevilla (COMPOLÍTICAS). https://bit.ly/3LxU3I7

Boeker, M., & Urman, A. (2022). An Empirical Investigation of Personalization Factors on TikTok. In F. Laforest.; R. Troncy.; E. Simperl.; D. Agarwal.; A. Glonis.; I. Herman, & L. Médini (Eds.), WWW '22: Proceedings of the ACM Web Conference (pp. 2298-2309). Association for Computing Machinery. https://doi.org/10.1145/3485447.3512102

Bonilla-del-Río, M., & Soares, I. (2021). Visibilidad y concienciación en el día mundial del síndrome de down: Movimientos en redes sociales hacia la inclusión. En M. C. Tornay-Márquez, Sánchez-López y D. Jaramillo-Dent (Eds.), Inclusión y activismo digital: participación ciudadana y empoderamiento desde la diversidad (pp. 29-53). Dykinson. https://bit.ly/3swIgn1

Bonilla-del-Río, M., Figuereo-Benítez, J. C., & García-Prieto, V. (2022). Influencers with physical disabilities on Instagram: Features, visibility and business collaboration. El Profesional de la Información, 31(6). https://doi.org/10.3145/epi.2022.nov.12

Bressler, M. Y., & Zampella, J. G. (2020). Response to: “Dermatology without dermatologists? Analyzing Instagram influencers with dermatology-related hashtags”. Journal of the American Academy of Dermatology, 83(6), 447-448. https://doi.org/10.1016/j.jaad.2020.07.098

Briones, H., & Ortiz, D. (2014). Ciberbullying y su influencia en la asertividad. Tendencia en psicología, 1(1), 34-39. https://bit.ly/48j485z

Burnap, P., & Williams, M. L. (2016). Us and them: Identifying cyber hate on Twitter across multiple protected characteristics. EPJ Data Sci., 5, 11. https://doi.org/10.1140/epjds/s13688-016-0072-6

Bustos-Martínez, L., De-Santiago-Ortega, P. P., Martínez-Miró, M. Á., & Rengifo-Hidalgo, M. S. (2019). Discursos de odio: Una epidemia que se propaga en la red. Estado de la cuestión sobre el racismo y la xenofobia en las redes sociales. Mediaciones Sociales, 18, 25-42. https://doi.org/10.5209/meso.64527

Cabo-Isasi, A. C., & García-Juanatey, A. (2016). El discurso del odio en las redes sociales: un estado de la cuestión. Ajuntament de Barcelona. https://bit.ly/450CLu8

Caro-Lopera, M. Á., Camargo Martínez, Z., & Uribe Álvarez, G. (2018). La ironía revisitada en sus dimensiones pragmáticas, comprensivas, discursivas y didácticas. Lenguaje, 46(1), 95-126. https://doi.org/10.25100/lenguaje.v46i1.6198

Cenat, J. M., Farahi, S. M. M. M., Dalexis, R. D., Darius, W. P., Bekarkhanechi, F. M., Poisson, H., Broussard, C., Ukwu, G., Auguste, E., Nguyen, D. D., Sehabi, G., Furyk, S. E., Gedeon, A. P., Onesi, O., El-Aouame, A. M., Khodabocus, S. N., Shah, M. S., & Labelle, P. R. (2022). The global evolution of mental health problems during the COVID-19 pandemic: A systematic review and meta-analysis of longitudinal studies. Journal of Affective Disorders, 315, 70-95. https://doi.org/10.1016/j.jad.2022.07.011

Cermi (2021). Los delitos de odio contra las personas con discapacidad. ¿Cómo identificarlos? ¿Cómo atajarlos? Guía informativa 2021. Comité Español de Representantes de Personas con Discapacidad. http://hdl.handle.net/1111/6500

Chetty, N., & Alathur, S. (2018). Hate speech review in the context of online social networks. Aggression and violent behavior, 40, 108-118. https://doi.org/10.1016/j.avb.2018.05.003

Cortés-Kandler, M., & Arroyo-Carvajal, Y. (2021). Análisis pragmático de los actos de habla irónicos y humorísticos en comentarios publicados en las páginas de Facebook de cuatro noticieros costarricenses. Textos en Proceso, 7(1), 76-99. https://doi.org/10.17710/tep.2021.7.1.5cortes_arroyo

European Union Agency for Fundamental Rights [FRA] (2021). Hate crime reporting infographic. FRA. https://bit.ly/44TaVjn

Figuereo-Benítez, J. C., García-Prieto, V., & Bonilla-del-Río, M. (2023). Percepción de las organizaciones de personas con discapacidad sobre el tratamiento mediático del colectivo en espacios informativos. Revista ICONO 14, 21(2). https://doi.org/10.7195/ri14.v21i2.2018

Figuereo-Benítez, J. C., Oliveira, J. S., & Mancinas-Chávez, R. (2022). TikTok como herramienta de comunicación política de los presidentes iberoamericanos. En I. Aguaded; A. Vizcaíno-Verdú; Á. Hernando-Gómez, & M. Bonilla-del-Río (Eds.), Redes sociales y ciudadanía: Ciberculturas para el aprendizaje (pp. 103-112). Comunicar Ediciones. https://doi.org/10.3916/Alfamed2022

Fisher, K. R., & Purcal, C. (2016). Policies to change attitudes to people with disabilities. Scandinavian Journal of Disability Research, 19(2), 161-174. http://dx.doi.org/10.1080/15017419.2016.1222303

Fuentes-Lara, C., & Arcila-Calderón, C. (2023). El discurso de odio islamófobo en las redes sociales. Un análisis de las actitudes ante la islamofobia en Twitter. Revista Mediterránea de Comunicación, 14(1), 225-240. https://doi.org/10.14198/MEDCOM.23044

Gagliardone, I., Gal, D., Alves, T., & Martínez, G. (2015). Countering online hate speech. UNESCO. https://bit.ly/3ZlsLuo

Gámez-Guadix, M., Wachs, S., & Wright, M. (2020). “Haters back off!” Psychometric properties of the coping with cyberhate questionnaire and relationship with well-being in Spanish adolescents. Psicothema, 32(4), 567-574. https://doi.org/10.7334/psicothema2020.219

Henson, A. (2023). Virtual Whac-A-Mole: Addressing the Patchwork Regulation of Online Hate Speech. Michigan State International Law Review, 31(1), 115-156. https://doi.org/10.17613/aky5-jz66

Islam, T. U., & Rizwan, M. (2022) Comparison of correlation measures for nominal data. Communications in Statistics - Simulation and Computation, 51(3), 698-714. https://doi.org/10.1080/03610918.2020.1869984

Jain, P., & Arakkal, J. (2022). Perceived Popularity of TikTok among Gen-Z. CARDIOMETRY, 24, 896-902. https://doi.org/10.18137/cardiometry.2022.24.896902

Krippendorff, K. (2004). Content analysis: An introduction to its methodology. Sage.

Losada-Díaz, J. C., Zamora-Medina, R., & Martínez-Martínez, H. (2021). El discurso del odio en Instagram durante las Elecciones Generales 2019 en España. Revista Mediterránea de Comunicación/Mediterranean Journal of Communication, 12(2), 195-208. https://www.doi.org/10.14198/MEDCOM.19142

Mackenzie, S. A., & Nichols, D. (2020). Finding ‘Places to Be Bad’ in social media: The case of TikTok. In D. Nichols y S. Perillo (Eds.), Urban Australia and post-punk (pp. 285-298). Palgrave Macmillan. https://doi.org/10.1007/978-981-32-9702-9_22

Martínez-Valerio, L. (2022). Mensajes de odio hacia la comunidad LGTBIQ+: análisis de los perfiles de Instagram de la prensa española durante la “Semana del Orgullo”. Revista Latina de Comunicación Social, 80, 363-388. https://www.doi.org/10.4185/RLCS-2022-1749

Muniesa-Tomás, M. P., Fernández-Villazala, T., Máñez-Cortinas, C. J., Herrera-Sánchez, D., Martínez-Moreno, F., San-Abelardo-Anta, M. Y., Rubio-García, M., Gil-Pérez, V., Santiago-Orozco, A. M., Gómez-Martín, M. Á., Méndez-Matos, G., Gómez-Esteban, J., Amado-Hernández, M. P., González-González, M., & Matilla-Molina, A. (2022). Informe sobre la evolución de los delitos de odio en España. Ministerio del Interior. https://bit.ly/48ezpXr

Muñoz-Rodríguez, J. M., Patino-Alonso, C., Pessoa, T., & Martín-Lucas, J. (2023). Identity profile of young people experiencing a sense of risk on the internet: A data mining application of decision tree with CHAID algorithm. Computers & Education, 197(104743). https://doi.org/10.1016/j.compedu.2023.104743

Organización de las Naciones Unidas [ONU] (2019). La estrategia y plan de acción de las Naciones Unidas para la lucha contra el discurso del odio. Naciones Unidas. https://bit.ly/3EKU9sc

Pahor de Maiti, K., Franza, J., & Fišer, D. (2023). Haters in the spotlight: gender and socially unacceptable Facebook comments. Internet Pragmatics. https://doi.org/10.1075/ip.00093.pah

Phanomtip, A., Sueb-in T., & Vittayakorn, S. (2021). Cyberbullying detection on Tweets. 18th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology (ECTI-CON), 295-298. https://doi.org/10.1109/ECTICON51831.2021.9454848

Phelan, J. (2014). Voice, tone, and the rhetoric of narrative communication. Language and literature, 23(1), 49-60. https://doi.org/10.1177/0963947013511723

Raffone, A. (2022). “Her leg didn’t fully load in”: A digitally-mediated social-semiotic critical discourse analysis of disability hate speech on TikTok. International Journal of Language Studies, 16(4), 17-42. https://bit.ly/44Ua3Lm

Ramírez-García, A., González-Molina, A., & Moyano-Pacheco, M. (2022). Interdisciplinariedad de la producción científica sobre el discurso del odio y las redes sociales: Un análisis bibliométrico. Comunicar, 30(72), 129-140. https://doi.org/10.3916/C72-2022-10

Rodríguez-Montemayor, A. (2020). El humor en la crítica social. 11º Coloquio inteinstitucional de docentes. https://bit.ly/3ZtXcyI

Sabater, M. M., García-Ruiz, C. R., & González-Monfort, N. (2021). Contrariar el odio. Bellaterra Journal of Teaching & Learning Language & Literature, 14(2), e909-e909. https://bit.ly/3t7ceOM

Sánchez-Rojo, A. (2023). Por una educación inclusiva libre de capacitismo: La importancia de la alteridad como fundamento. Estudios Pedagógicos, 49(2), 341-356. https://doi.org/10.4067/S0718-07052023000200341

Santana-Valencia, E.V. (2022). El derecho a la educación de las personas con discapacidad. Perfiles educativos, 44(178), 202-209. https://doi.org/10.22201/iisue.24486167e.2022.178.61096

Schober, P., Boer, C., & Schwarte, L. A. (2018). Correlation coefficients: appropriate use and interpretation. Anesthesia & analgesia, 126(5), 1763-1768. https://doi.org/10.1213/ANE.0000000000002864

Sherry, M., Olsen, T., Solstad Vedeler, J., & Eriksen, J. (2019). Disability hate speech: Social, cultural and political contexts. Routledge.

Suárez-Álvarez, R., & García-Jiménez, A. (2021). Centennials en TikTok: tipología de vídeos. Análisis y comparativa España-Gran Bretaña por género, edad y nacionalidad. Revista Latina de Comunicación Social, 79,1-22. https://www.doi.org/10.4185/RLCS-2021-1503

UNICEF (2022). Voces de adolescentes con discapacidad sobre la violencia y discriminación en la familia, escuela y comunidad. Una consulta rápida en 11 países de América Latina. Fondo de las Naciones Unidas para la Infancia (UNICEF). http://bit.ly/462uBmo

We Are Social (2023). Special Report. Digital 2022. La guía definitiva para un mundo digital cambiante. We Are Social / Hootsuite. https://bit.ly/3PCXQ9I

Weimann, G., & Masri, N. (2023). Research note: Spreading hate on TikTok. Studies in conflict & terrorism, 46(5), 752-765. https://doi.org/10.1080/1057610X.2020.1780027

Westberg, K. H., Nyholm, M., Nygren, J. M., & Svedberg, P. (2022). Mental health problems among young people-a scoping review of help-seeking. International journal of environmental research and public health, 19(3), 1430. https://doi.org/10.3390/ijerph19031430

AUTHORS’ CONTRIBUTIONS, FUNDING AND ACKNOWLEDGEMENTS

Authors' contributions:

Conceptualization: García-Prieto, V.; Bonilla-del-Río, M.; Figuereo-Benítez, J. C. Software: García-Prieto, V. Validation: García-Prieto, V. Figuereo-Benítez, J. C. Formal analysis: García-Prieto, V. Data curation: García-Prieto, V.; Bonilla-del-Río, M.; Figuereo-Benítez, J. C. Writing and Preparation of the original draft: Bonilla-del-Río, M.; García-Prieto, V. Drafting, Revision and Editing: Bonilla-del-Río, M.; García-Prieto, V.; Figuereo-Benítez, J. C. Visualization: García-Prieto, V. Supervision: Bonilla-del-Río, M.; García-Prieto, V.; Figuereo-Benítez, J. C. Figuereo-Benítez, J. C. All authors have read and accepted the published version of the manuscript: García-Prieto, V.; Bonilla-del-Río, M.; Figuereo-Benítez, J. C.

Funding: This research did not receive external funding.

AUTHORS:

Victoria García-Prieto

University of Seville.

PhD in Communication from the University of Seville, graduated in Journalism and M.A. in Communication and Culture. Interim professor at the University of Seville and lecturer at the EUSA University Center, she teaches Journalism, Audiovisual Communication and Advertising, and Public Relations. Member of Media, Communication Policies and Democracy research groups in the European Union (DemocMedia), the Comunicar Group and the Ibero-American network of researchers Alfamed. She has been a visiting researcher at the universities of Cambridge and Westminster (UK), and Nova de Lisboa (Portugal). Her main lines of research include audiovisual accessibility, cultural pluralism and media literacy for people with disabilities.

vgarcia8@us.es

Índice H: 6

Orcid ID: https://orcid.org/0000-0003-4973-7583

Scopus ID: https://www.scopus.com/authid/detail.uri?authorId=57193826726

Google Scholar: https://scholar.google.com/citations?user=ZqN1YpwAAAAJ&hl=es&oi=ao

ResearchGate: https://www.researchgate.net/profile/Victoria-Garcia-Prieto

Academia.edu: https://independent.academia.edu/VictoriaGarc%C3%ADaPrieto

Dialnet: https://dialnet.unirioja.es/servlet/autor?codigo=3085709

idUS: https://idus.us.es/author-profiles/author?id=4655

Mónica Bonilla-del-Río

European University of the Atlantic.

PhD from the Interuniversity Communication Program in the line of Communication Education and Media Literacy (UHU). Professor of Journalism, Audiovisual Communication, Advertising and Public Relations at the European University of the Atlantic (Santander). Member of the Agora Group (Andalusian Research Plan: HUM-648), member of Comunicar Group, a veteran Andalusian group of professionals in communication education and of Alfamed Joven (https://www.redalfamed.org/alfamed-joven), belonging to the Euroamerican Interuniversity Network Alfamed (https://www.redalfamed.org), dedicated to the study of media competence for citizenship. Interuniversity M.A. in Communication and Audiovisual Education by the International University of Andalusia and the University of Huelva. M.A. in Emotional, Social and Creativity Education by the University of Cantabria. Graduated in Early Childhood Education by the University of Cantabria.

monica.bonilla@uneatlantico.es

Índice H: 12

Orcid ID: https://orcid.org/0000-0003-2476-8922

Scopus ID: https://bit.ly/3mEYuYf

Google Scholar: https://scholar.google.es/citations?user=Tel_fQwAAAAJ&hl=es

ResearchGate: https://www.researchgate.net/profile/Monica-Bonilla-Del-Rio

Academia.edu: https://uhu.academia.edu/M%C3%B3nicaBonilladelR%C3%ADo

Dialnet: https://dialnet.unirioja.es/servlet/autor?codigo=4472575

Juan C. Figuereo-Benítez

University of Seville.

Pre-doctoral Researcher Professor in the Department of Journalism II at the University of Seville. He has a Bachelor's Degree in Journalism and a Master's Degree in Political and Institutional Communication from the University of Seville. He currently studies the Interuniversity PhD in Communication at the Universities of Cadiz, Huelva, Malaga and Seville. Member of the Research Group Communication, Power and Critical Thinking in the Face of Global Change (Compoder), with official code SEJ-675. He has been a visiting researcher at the Universities of Havana (Cuba), El Salvador and Francisco Gavidia (El Salvador), Autonomous University of Baja California and UNAM (Mexico), Cartagena (Colombia), National University of San Agustin de Arequipa (Peru) and UFSC (Brazil). His research interests are political and institutional communication, electoral campaigns, social networks and accessibility.

figuereo@us.es

Índice H: 4

Orcid ID: https://orcid.org/0000-0002-9061-8482

Scopus ID: https://www.scopus.com/authid/detail.uri?authorId=57744030100

Google Scholar: https://scholar.google.es/citations?user=ZFfR2YsAAAAJ

WoS: https://www.webofscience.com/wos/author/record/HDM-9072-2022

Dialnet: https://dialnet.unirioja.es/servlet/autor?codigo=4900967

IdUS: https://idus.us.es/author-profiles/author?id=7741