#MeToo as a Connective Movement: Examining the Frames Adopted in the Anti-Sexual Harassment Movement in China

Abstract

This study was aimed to contribute to understanding how networked yet fragmented online actors create meaning in digital media–enabled movements like #MeToo. By drawing upon a multidimensional framing analysis, this study investigated how personal action frames, collective action frames, and issue-specific frames were adopted in #MeToo movement in China, and it also shed light on how different groups of social actors respond to sexual harassment issues on Sina Weibo, a Chinese social media platform. This study employed computational content analysis to extract frames from a huge amount of traceable data (i.e., 16,187 Weibo posts) and uncovered seven specific types of frames categorized as personal experiences and emotional commentary (as personal action frames), injustice and opposition (as collective action frames), and problem definition, treatment recommendation, and related news (as issue-specific frames). The results revealed that personal action frames and collective action frames were widely adopted by females and ordinary users, whereas issue-specific frames were more commonly applied by males and organizational users. These empirical findings enhance our understanding of meaning construction with regard to digital media–enabled movements.

Keywords

online movement connective action framing sexual violence topic modeling digital media

Digital technologies have contributed considerable benefits to individuals’ engagement in social changes, which include lowering the threshold of participation, facilitating participation with easy-to-use technological affordances, and expanding movement action repertoires via sufficient technological features (Vaast et al., 2017; Van Laer & Van Aelst, 2010). With the aid of the digital media—especially social media platforms—segmented individuals can now work together and publicly react to social issues through joint actions (Castells, 2015; Van Laer & Van Aelst, 2010). Due to social media’s development, various social movements that occurred in recent years have gained momentum, such as the Arab Spring, the Umbrella Movement in Hong Kong, the Occupy movement, and the #MeToo movement.

It is worth noting that, in the aforementioned movements, collective actors or conventional movement organizers played a noncentral or peripheral role, whereas segmented individuals are becoming increasingly important in leveraging actions (Anduiza et al., 2014; Lee & Chan, 2016; Leong et al., 2018). These atomized individuals, who were originally regarded as peripheral to social movements, are now able to play a central role in organizing and expanding local movements into large-scale events via the facilitation of digital media. Recent research agendas posit that digital media shift social movements’ organizational patterns from the logic of collective action (i.e., enabled by conventional organizations) to the logic of connective action (i.e., enabled by connected or networked individuals; Bennett & Segerberg, 2013). A growing body of research has investigated various aspects of connective movements, including their organizational structures and processes (Agarwal et al., 2014), the characteristics of mobilization channels, participants and involved organizations (Anduiza et al., 2014), the modes of participation (Lee & Chan, 2016), and the technological affordances of social media adopted into such movements (Vaast et al., 2017).

When engaging in social movements, actors construct meanings for the focused issue to mobilize resources. Different from conventional movements, the symbolic packages or frames are constructed by segmented individuals in a personalized manner rather than by central organizations via collective action frames (Bennett & Segerberg, 2013). Although collective action frames have been extensively studied (for a review, see Benford & Snow, 2000), to date, scarce research has empirically examined the development of personalized action frames in connective movements. Moreover, connective movements normally take place in online environments, where a variety of social actors (e.g., movement actors, organizational agencies) constantly construct meanings of the same issue according to diverging orientations (Wright, 2015). The complexity of participants’ composition makes it difficult to illustrate how a connective movement is organized and expanded. This study’s aim is thus to attain an empirical understanding of how distinct groups of social actors construct and utilize various frames (including personalized action frames) to mobilize resources in a connective movement.

To achieve this goal, this study sheds light on the #MeToo movement in China by examining frames that have emerged from social media posts related to this movement. Collective movements have rarely occurred in China due to the ubiquitous government censorship and the limited role of movement organizations or activists. As such, the proliferation of #MeToo movement in China’s unique sociopolitical context provides an opportunity to study how a social movement could be mobilized by atomized social actors with the aid of social media platforms. In addition, by uncovering frames embedded in the discourse of social movements, researchers can capture movement actors’ perceptions regarding the concerned issues as well as their participation modes (Hon, 2016; Snow & Benford, 1988). Hence, this study employs framing analysis as the analytical tool to illustrate meaning construction in the public discourse on China’s #MeToo movement. Considering the difficulty of manually extracting meaningful themes from the considerable number of unstructured social media posts, this study relies on the computational content analysis method (i.e., coding the content of a set of texts through computational techniques), which can exploratorily, automatically, and objectively detect meanings from enormous yet unstructured documents such as social media posts (Nelson, 2017).

Literature Review

#MeToo in China as a Connective Movement

“MeToo” has been a globally prominent topic on social media since October 2017, during which time the public used #MeToo as a hashtag to reveal and call attention to the magnitude of and the frequency of sexual violence. By encouraging victims to expose the sexual violence they have suffered, #MeToo movement aimed to raise public awareness about sexual violence and provoke responses from society (Xiong et al., 2018). Within a short amount of time, the movement spread widely around the world including China. This movement began in late December 2017 in China, when a female graduate from a university located in Eastern China published a blog article on Weibo (a microblogging website akin to Twitter) revealing that she was sexually assaulted by a professor for 7 months during her course of study. Shortly after, on the first day of 2018, a former doctoral student made an allegation on WeChat (a social media application akin to Facebook) against her doctoral supervisor and claimed that she was sexually harassed and assaulted several times by her supervisor 12 years before. Inspired by the global #MeToo campaign, the two victims used the hashtag #MeToo to call for public attention. These two allegations spurred heated public discussion about sexual misconduct in colleges and kicked off #MeToo movement in China. Since then, a number of professors, prominent intellectuals, and civil rights activists were accused of sexually harassing or assaulting females. These exposed cases made Chinese netizens begin continuously responding to the issue of sexual harassment on social media.

Compared to its counterpart in the United States or other countries, #MeToo movement has home-grown characteristics rooted in the Chinese context. On the one hand, sexual assault in colleges is central to China’s #MeToo, as this movement was initiated by influential cases that occurred in colleges, and the majority of the overall allegations (31 of 56 reported sexual violence cases; Mat-X, 2018) were against university professors and made by current or former students. This distinctive staging ground for China’s #MeToo is related to (a) the staggeringly unequal power structure between teachers and students in educational sectors, especially in colleges and (b) the rising awareness of gender and sexual issues among college students (Z. Chen et al., 2019; Lin & Yang, 2019). On the other hand, because the Chinese government is very vigilant about and intolerant of any form of collective action, the development of this movement constantly encountered online censorship from the authorities (King et al., 2013). For instance, a series of hashtags (including #MeToo) used for indicating #MeToo movement were banned, and the social media accounts of many feminist activists and movement organizations were rigorously monitored and even suspended (Zeng, 2019). To bypass the censorship, sophisticated Chinese netizens employed various strategies (e.g., restoring and hiding previously deleted content; twisting and camouflaging texts and images) to show their sympathies for victims and express personal concerns and opinions (Zeng, 2020). During such public discussions, Chinese netizens encouraged the public to break the silence regarding the issue of sexual harassment in daily life, called for actions to address this social issue, and thereby mobilized the development of the #MeToo movement.

Similar to other countries, this anti-sexual harassment movement was initiated and popularized by the virtue of social media. More specifically, it was largely enabled and mobilized by networked yet distributed participants on social media (Lin & Yang, 2019), which is akin to other digital media–enabled movements (e.g., the Arab Spring, the Occupy movement). Bennett and Segerberg (2013) distinguished between the logic of collective action and the logic of connective action to illustrate the differences in organizational patterns between conventional movements and digital media–enabled movements. Following the logic of collective action, conventional social movements emphasize resource mobilization by core activists or central social organizations, formal collective identity, and actions coordinated with explicit goals aimed toward social changes (Leong et al., 2018; Obregón & Tufte, 2017).

Different from collective action, the logic of connective action demonstrates a new form of social movement in which separate individuals with no identified collective identity spontaneously and informally network together to respond to social issues by expressing their personal thoughts and becoming involved in actions via information technologies (Bennett & Segerberg, 2013). Although there are still core organizers involved in connective movements, the participants are less likely to be coordinated by any organizer. Rather, they mainly act in their individualized manners, including the use of memes and hashtags as well as the personalization of their social media profiles (Anduiza et al., 2014; Bennett & Segerberg, 2013; Lee & Chan, 2016). The logic of connective action is saliently reflected in China’s #MeToo movement (Zeng, 2020). As mentioned previously, due to the authorities’ online surveillance, an identifiable organizational resource is lacking, and the collective identity of the involved public remains implicit in this movement. The movement has retained its vitality from the involved crowd’s constant engagement. Due to a series of sexual assault cases being exposed by victims, the crowd has continuously focused on and responded to such cases (Zeng, 2020). These individuals expected the sexual harassment issue to be fixed in the society; thus, they were self-motivated (due to personal reasons) to act and call for action via social media platforms.

It is clear that China’s #MeToo movement follows the logic of connective action, but it is unclear how the movement actors construct meaning from sexual harassment to act and mobilize others’ actions. Meaning construction is considerably important for social movements, whereby movement actors communicate with external audiences and expand the influence. In collective movements, the meaning of the focused issue is framed by central organizations, whereas the distributed self-organizing actors accomplish this task in a connective movement such as #MeToo. To understand how these networked yet distributed participants create the meaning of #MeToo, we now turn to framing scholarship.

Framing Analysis in Social Movements

A frame refers to the schemata embedded in a message to organize various aspects of perceived reality (Entman, 1993; Scheufele, 1999). Framing hence reflects the practice of constructing the meaning of social reality and is influenced by various factors such as individuals’ beliefs and values, the impact of organizational ownership, and a social system’s ideological impact (Brüggemann, 2014; Shoemaker & Reese, 2013). In short, framing is subject to distinct groups of social actors who possess distinct purposes.

Previous literature has identified several categories of frames that reflect how different social actors construct meaning from social problems. Initially, Entman (1993) specified that mass media employ “issue-specific frames” in news stories and media discourses to influence audiences. Issue-specific frames describe four aspects of reasoning with a specific social problem, namely problem definition (i.e., providing a definition of the issue), causal interpretation (i.e., offering a causal interpretation), moral evaluation (i.e., evaluating moral effects), and treatment recommendation (i.e., recommending possible treatments). From the social movement perspective, Benford and Snow (2000) proposed that central organizations in social movements develop “collective action frames,” which entails “the production of mobilizing and countermobilizing ideas and meanings” (p. 613) to build collective identity and coordinate actions. In digital media–enabled movements, however, individual actors mostly express their personal appeals through “personal action frames,” referring to individual actors’ personalized expressions of their hopes, lifestyles, and grievances with the aid of technological affordances (Bennett & Segerberg, 2013).

The aforementioned three categories of frames would eventually manifest in China’s #MeToo movement. According to Bennett and Segerberg (2013), personal action frames are predominantly used by self-organized actors in connective movements. On the one hand, the focused issue in a connective movement (e.g., sociopolitical inequality in the Occupy movement) provides a common ground (e.g., using the slogan “We Are the 99%”) that allows people to frame their personal appeals (Bennett et al., 2014). On the other hand, by sharing the “small” problematization or personal calls on social media, individual actors help the movement attain global relevance and significance and thereby mobilize external audiences (Bang & Halupka, 2019). As such, personal action frames function as organizational hubs in connective movements. By examining personal action frames adopted in China’s #MeToo, we can uncover which aspects the participants are most concerned about regarding sexual harassment issues in their daily lives and understand how these concerns help mobilize coordination.

Collective action frames can also be useful in #MeToo movement. Although conventional organizations play a less central role in connective movements, they are nevertheless involved in these movements via collective action frames to cocreate shared meanings with individual actors. A recent study on #MeToo movement identified frames that movement organizations pull from Twitter hashtags, including movement-related, action-oriented, and event-specific frames (Xiong et al., 2018). Previous literature has revealed that central organizations predominantly adopt the injustice frame (centered on victims of perceived injustice caused by identifiable parties) and the oppositional frame (centered on activating oppositional consciousness to motivate participation) in sexual violence–related or female-focused movements (Marshall, 2003; Trumpy, 2016). Given that #MeToo movement relates to sexual harassment and gender inequality, this study attempts to examine whether and how such collective action frames are adopted in China’s #MeToo movement.

As contended previously, apart from movement actors, other stakeholders (e.g., media or institutional agencies) are involved in #MeToo movement. A prior study (Lim, 2013) noted that media agencies framed news to influence public perceptions, which further influenced the orientation of digital media–enabled movements (e.g., the Arab Spring in Tunisia). Hence, this study incorporates into the analysis issue-specific frames that are normally adopted to shape media discourses in order to investigate how #MeToo is framed to influence the Chinese public’s perceptions regarding sexual harassment.

Research Questions

Framing analysis—that is, the identification of thematic units embedded in messages—has been widely applied to uncover the meanings of social issues formed by various social actors such as mass media, social movement actors, or even individual actors. This study employs a multidimensional approach to extract frames manifested in social media posts regarding China’s #MeToo movement by taking into account the above-described three genres of frames. Because connective movements primarily occur on social media platforms (e.g., Weibo), relying on a single type of frame would fail to capture the complexity and diversity of meaning constructions in China’s #MeToo movement. Prior studies have applied the multidimensional approach to analyze social media posts (e.g., Nip & Fu, 2016), as social media afford space for different voices from fragmented actors (e.g., media agencies, organizations, elites, individual users) who create a diversity of frames (DeLuca et al., 2012). Likewise, the decentralized nature of connective actions renders the necessity of adopting a multidimensional approach in this study. Therefore, this study incorporates a diverse group of frame genres (i.e., personal action, collective action, and issue-specific frames) and attempts to empirically examine how these frames were adopted in China’s #MeToo movement. We thus propose the following question:

Research Question 1: Which prominent frames among the personal action frames, collective action frames, and issue-specific frames emerged during China’s #MeToo movement?

In connective movements, frames adopted by segmented individuals may not work concordantly in influencing how the movement develops. Instead, these frames can have different and even competing orientations to steer the movement in distinct directions because segmented individuals attach different meanings to the focused social issue via personalized and diverse expressions. Indeed, as described above, framing is inherently subject to the individual’s identity; thus, how social actors frame a concerned issue can be impacted by their identities (W. Y. Wang, 2013). Hence, in this study, we mainly focus on the influence of two of the actor’s attributes: gender and account type.

Gender’s impact on framing #MeToo should be examined, as sexual violence is predominantly committed by men to prey upon women (Marshall, 2003). The current patriarchal structures normalize “power over” females in the dominative sense (Uggen & Blackstone, 2004), and the imbalance of power relations between genders results in a gender gap in relevant movements (Swank & Fahs, 2014). Specifically, females are inclined to participate in sexual violence–related movements on their own behalf to seek social justice frames (Fileborn, 2016; Skalli, 2014), and they tend to apply injustice frame when identifying predators’ responsibility for the offense (Marshall, 2003). In contrast, males are reluctant to engage in gender-related issues that are perceived to challenge male privileges (Liss et al., 2004; Swank & Fahs, 2014). Consequently, gender may impact the frames social actors adopt regarding the #MeToo movement, and we propose the following question to address this possibility:

Research Question 2: How does gender impact the adoption of frames that emerged during China’s #MeToo movement?

In this study, we are also motivated to understand the influences of different social media account types, as each reflects distinct groups of social actors on social media. Previous research has proposed that Weibo users with distinct account types set the agenda differently through framing to affect public opinion (e.g., Z. Chen et al., 2019; H. Wang & Shi, 2018). Therefore, the individual’s account type, representing their online identity, influences the frames they use in #MeToo movement. Weibo accounts are categorized into three types based on the platform’s authentication system—namely, organizational accounts (representing corporations, institutions, government affiliations, and other social groups), individual influencers’ accounts (mostly celebrities and social influencers who are authenticated and possess relatively large follower bases), and ordinary individual accounts (which are typically unauthenticated; Z. Chen et al., 2019; H. Wang & Shi, 2018). Because they are subject to organizational ownership and current political ideologies, organizations usually frame everyday contests within mainstream politics (Shoemaker & Reese, 2013), and their social media accounts primarily function as news media broadcasting information to maintain relationships with their followers (Lovejoy & Saxton, 2012). Individual influencers are normally considered opinion leaders on Weibo, in that they are capable of influencing numerous followers by drawing on shared values (e.g., equality, justice) to frame social problems (Veenstra et al., 2008). Ordinary users, however, mainly correspond with personal concerns in their online expressions (Bennett & Segerberg, 2013). Although ordinary users are solely effective via a relatively narrow scope, their expressions are crucial for sustaining and expanding a movement when many of them act together (Bennett & Segerberg, 2013). Apparently, distinct account holders may implement varying frames on social media when responding to the issue of sexual harassment, and we propose the following question to make this determination:

Research Question 3: How do the frames that emerged during China’s #MeToo movement differentiate across Weibo users holding distinct account types?

Method

Data Collection

To answer the research questions, we collected data on Sina Weibo (https://weibo.com/) in October 2018. As one of the most popular social media platforms in China, Weibo is akin to Twitter, where users follow and are followed by others, acquire information, and publicly express their opinions. Despite censorship, Weibo provides a public sphere for Chinese netizens’ civic participation in various social issues including this anti-sexual harassment movement (Rauchfleisch & Schäfer, 2015; H. Wang & Shi, 2018). We collected Weibo posts related to #MeToo, the account type, and gender of each poster (both indicated in their profile) using Weibo’s application programming interface (API).

When adopting computational techniques to collect and analyze data, a specific and clear boundary for accurate data scrapping is needed in order to maximize the large-scale capacity of big data on the one hand and to avoid a huge amount of noise (i.e., irrelevant posts) on the other (Lewis et al., 2013). As such, considering the key characteristic of #MeToo movement (i.e., advocating voluntary disclosure of sexual harassment and sexual assault) and the unique characteristics of this movement in China (i.e., the prevalence and severity of sexual assault in colleges; H. Chen et al., 2020), in this study, the collected posts need to satisfy two criteria simultaneously: (a) The posts must have discussed sexual harassment or sexual assault cases voluntarily exposed by the victims themselves or their close friends and (b) the posts must have been about sexual harassment or sexual assault cases that occurred in colleges.

In particular, the data were collected in two steps. In the first step, Weibo posts were collected via a keyword-based searching method. Two Chinese keywords that are centrally related to the #MeToo movement, “Xing Sao Rao” (sexual harassment) and “Xing Qin” (sexual assault), were searched for to identify posts published from December 1, 2017, to September 1, 2018. We set September 2018 as the end point of data collection because few allegations about sexual harassment or sexual assault in colleges have been exposed since then (Zeng, 2020), and the public discussion about #MeToo has also declined and remained at a relatively low level (see Figure 1 for details). After the first step, we obtained a huge data set containing Weibo posts that broadly discussed sexual harassment and sexual assault, although a significant amount of it was ultimately irrelevant to our research focus.

In the second step, therefore, we filtered the posts that satisfied the aforementioned selection criteria by using a list of keywords. As previously mentioned, 36 (of 51) voluntarily exposed allegations regarding sexual abuse in colleges were identified (Mat-X, 2018). Because the development of #MeToo movement in China predominantly gained traction from the public discussion around these cases (Zeng, 2020), we created a list of 122 keywords through close observation that were frequently used while discussing these 36 allegations to select appropriate posts in the final dataset. These keywords included three dimensions: (a) the names of involved stakeholders (including 62 keywords indicating the specific predators, victims, and institutes); (b) frequently used Chinese words that address “teachers” and “students” in colleges (including 33 keywords related to “professors,” “supervisors,” etc.); and (c) frequently used Chinese words that describe the incidents’ settings (including 27 keywords related to “colleges,” “universities,” “academia,” etc.). To ensure the accuracy of data collection, we randomly selected 10% of the collected posts, and two of the authors thoroughly read these sample posts and found all of them reliable.

After carrying out these two steps, 16,187 posts were entered into the final data set. The time series of these Weibo posts are shown in Figure 1. In terms of account type, we included 1,502 (9.3%) organizational users, 2,235 (13.8%) influencers, and 12,450 (76.9%) ordinary users. Regarding gender distribution, among the individual users (influencers and ordinary users), 7,630 (52.0%) females and 7,054 (48.0%) males were included.

Figure 1.

Time series of Weibo posts related to #MeToo movement.

Computational Content Analysis: Topic Modeling

This study employed an unsupervised, computational content analysis method, particularly topic modeling, to effectively extract patterns of meaning from the large number of unstructured Weibo posts (to address Research Question 1). Topic modeling is a form of semantic analysis that applies statistical algorithms to detect the abstract or hidden “topics” that emerge from a large collection of documents. In this study, a specific algorithm of topic modeling, latent Dirichlet allocation (LDA), was used. The LDA algorithm clusters bags of words in individual documents into various topics, each of which is reflected by an extracted cluster of keywords based on the computed probability of the coappearance of keywords (Blei, 2012). To perform topic modeling, we first removed the common Chinese stop words (e.g., “the,” “a,” and “an”) by applying a manually curated stop word dictionary because the inclusion of such words produces noise and prevents the algorithm from recognizing patterns in content-bearing words (Schofield et al., 2017). Second, unlike words in the English language, the basic structural units of the Chinese language are Chinese characters written without spaces between them. Therefore, we used the open-source corpus “Jieba” (https://github.com/fxsjy/jieba) to segment the original Weibo posts into a set of meaningful texts (i.e., bags of words) to make the analysis feasible. Third, LDA was implemented to analyze the segmented texts. By assessing the exclusiveness between topics and the clarity of an individual topic for each model, after several attempts, we identified the model with 17 topics being optimal for interpretation. As shown in Table 1, for each topic the algorithm generated is represented by a set of similar keywords.

Finally, we manually recoded the 17 extracted topics into seven frames by reading the 10 most representative posts and the top-weighted keywords under each topic via textual analysis. This step is in line with the proposition that the topics generated from LDA can be recoded or regrouped into meaningful, condensed frames based on conceptual similarities (Nelson, 2017), which has been widely adopted in the existing literature (e.g., Choi, 2018; Maier et al., 2018). As demonstrated in the Literature Review section, previous studies have identified several key frames (which can be divided into different aspects of personal action, collective action, and issue-specific frames) that could be adopted in digital media–enabled movements. Using these frames as guidelines, we sought conceptual similarities among the 17 extracted topics and combined similar topics for the sake of conceptual clarity, clear distinction between topics, and parsimony (Nelson, 2017).

The 17 topics were eventually coded into seven frames (shown in Table 1): personal experiences, emotional commentary, injustice, opposition, problem definition, treatment recommendation, and related news. In particular, personal experiences and emotional commentary were the two subcategories of personal action frames (Bennett & Segerberg, 2013). Injustice and opposition reflected collective action frames because these two topics are conventionally considered action frames that leverage collective participation (Marshall, 2003; Trumpy, 2016). Problem definition, treatment recommendation, and related news indicated different aspects of the issue-specific frames (Entman, 1993).

As shown in Figure 2, we also performed a semantic network analysis based on the results of LDA to visualize our findings (Choi, 2018), and the semantic network analysis confirmed the results of coding the 17 topics into seven frames. Figure 2 also shows the graphical representation of the seven frames with the keywords of each frame and their semantic relationships. To examine the impact of gender and account type on frame adoption (i.e., Research Questions 2 and 3), we respectively performed a χ² test and a post hoc cellwise test on the contingency tables to locate discrepancies (Beasley & Schumacker, 1995).

Figure 2.

Semantic networks of the extracted frames’ distribution. Note. The irrelevant keywords are indicated by gray nodes; all keywords were translated from Chinese to English.

Results.

Through collecting and analyzing 16,187 Weibo posts, this study uncovered a variety of frames social actors adopted when participating in #MeToo movement in China. As noted above, seven frames were extracted from the unstructured Weibo posts (shown in Table 1), namely personal experiences, emotional commentary, injustice, opposition, problem definition, treatment recommendation, and related news. In response to Research Question 1, these seven frames reflect the Chinese public’s framed #MeToo via three dimensions—personal action frames (e.g., personal experiences and emotional commentary), collective action frames (e.g., injustice and opposition), and issue-specific frames (e.g., problem definition, treatment recommendation, and related news). The sample posts of each frame are presented in Table 2.

Table 1.

List of Extracted Topics and Keywords.

Frame	Topic	n (%)	Keywords
Personal experience	0	1,285 (7.9)	girls; recall; midnights; men; scared; horrible; harms; life
Personal experience	1	1,923 (11.9)	female; experience; friend; men; harassment; girls; unwilling; tell; disgusting; hurt
Emotional commentary	2	1,445 (8.9)	[disgusting]; [smile]; [Facepalm]; female; workplace; company; college; men
Injustice	3	1,896 (11.7)	victim; power; gender; hope; justice; rights; relations; change
Opposition	4	598 (3.7)	female; exposed; dressing; claim; combat; every walk of life; Korea drama; daily life
Opposition	5	636 (3.9)	encounter; silence; bravely; female; victim; workplace; bus; embarrassment
Problem definition	6	902 (5.6)	report; autonym; cancel; teacher qualification; female PhD student; investigate; repeal; announce
	7	838 (5.2)	tolerate; first; suffer sex assault; college teachers; domestic; first case; obscenity; research
	8	659 (4.1)	report; autonym; University; professor; obscenity; slap in the face; college students; expose
	9	921 (5.7)	suicide; investigate; report; faculty of arts; police; suspected; punishment; autonym
Treatment recommend	10	972 (6.0)	law; hope; investigate; protect; crimes; harassment; obscenity; college student
	11	254 (1.6)	mechanism; protect; establish; national; system; juvenile; Education Bureau; discipline
	12	1,234 (7.6)	educate; children; hope; life; father and mother; prevent; truth; world
	13	875 (5.4)	protect; children; educate; parents; anti-sexual abuse; knowledge; juveniles; prevent
Related news	14	829 (5.1)	US; accuse; sexual abuse; victims; Hollywood; again and again; suffer; public opinion
	15	573 (3.5)	kindergarten; sexual assault case; college students; girl; security guard; obscenity; call police
	16	347 (2.1)	US; children; school doctor; overseas student; teaching; sexual assault; education; California

Note. All keywords were translated from Chinese to English.

Table 2.

Frames and Sample Posts.

Meta Frame	Frame	n (%)	Sample Posts
Personal action	Personal experience	3,208 (19.8)	#metoo# I remember when I was in middle school, every day I walked to my home through a shabby neighbourhood. One night, a man suddenly came out and grabbed my breast. I didn’t realize I was sexually harassed until he ran away. I was too ashamed to tell my parents at that time.
Personal action	Emotional commentary	1,445 (8.9)	Sexual harassment is not rare in workplaces especially when your boss is male. The VP in my last company is a person like this. One day, a female colleague told him “you look sharp in this new outfit.” His response was, “you too, and you would be even prettier if having no clothes on.” [disgusting]
Collective action	Injustice	1,896 (11.7)	The recent sexual violence cases remind me that each relationship, like teacher–student relation, love relation and family relation, is intertwined with power relation. The “power principle” can even explain the sex-pests’ behaviors in buses, they usually target at the girls who look reticent.
	Opposition	1,234 (7.6)	1. Regarding Metoo Movement and female self-protection, don’t be silent, don’t be afraid, just speak up! Repost for our own rights and say “No” to sexual harassment.
	Opposition	1,234 (7.6)	2. Females being sexually assaulted is because they are scantily-clad? Let’s see how the lady in Korean TV series “Ms. Hamraby” beats back such absurd comments wisely!
Issue-specific	Problem definition	3,320 (20.5)	All three universities talk about sexual assault, but how is it determined? Is there any formal decision that draws such conclusion? Is there any fact that supports the conclusion? Just determined by public opinion?
	Treatment recommendation	3,335 (20.6)	1. The result is fine, but anti-sexual harassment mechanism still needs to be built. After all, protecting safety and legal interests of teachers and students is the top priority.
	Treatment recommendation	3,335 (20.6)	2. Everyone should know the sexual abusers may be the person you know or even your teacher. Parents should pay attention and give guidance to their children in this aspect. When tragedy happens, parents should pay more attention to children instead of blaming them.
	Related issues	1,749 (10.8)	The sexual harassment cases are continuously exposed in Hollywood. After Harvey Weinstein, three women have now accused Ed Westwick, who played Chuck Bass on The CW’s “Gossip Girl,” of sexual assault….

Theoretically, both personal action frames and collective action frames indicate movement actors’ mobilized actions in #MeToo movement. Therefore, for the sake of discussion, this study groups these two dimensions as movement-oriented frames. The results of the semantic network analysis also supported such grouping (shown in Figure 2), as the cluster of personal action frames (especially personal experiences) and of collective action frames (e.g., injustice and opposition) is relatively close to one another.

Movement-Oriented Frames

Regarding personal action frames, personal experiences (n = 3,208, 19.8%; Topic 0, 1), representing the most frequently adopted movement-oriented frames, revealed that users shared and ruminated their negative experiences with sexual harassment from which they or their friends suffered in daily life. Emotional commentary (n = 1,445, 8.9%; Topic 2) revealed that users expressed their personal feelings, which were primarily negative emotions, when commenting on the prevalence of sexual violence cases. Notably, emotional commentary was mostly conveyed by the emoji devices provided by the Weibo platform (e.g., [disgust], [facepalm] present in Topic 2, Table 1).¹

For collective action frames, the two commonly used frames in gender-related movements—that is, injustice frame and oppositional frame—also appeared during China’s #MeToo movement. Injustice (n = 1,896, 11.7%; Topic 3) amplified the scope of victimization from academia to other fields and attributed sexual harassment to the power relations of male dominance over females, thereby indicating Weibo users’ awareness of the origin of sexual violence problems (Marshall, 2003). Oppositional frame (n = 1,234, 7.6%; Topic 4, 5), which underscored social actors’ agency to combat sexual violence, was adopted during China’s #MeToo movement to evoke oppositional consciousness against sexual harassment in everyday situations, such as “buses” or “workplaces” (Sandoval, 1991). One aspect of oppositional frame was anti-victim blaming; as discussed in Topic 4’s sample post and keywords, users refuted the victim-blaming claim by sharing a video clip from a Korean drama in which the female character harshly criticizes the prevalent claim that attributes sexual violence problems to women dressing scantily. This video clip was widely shared to call for the public’s consciousness of opposing to victim-blaming claim.

It should be noted that personal action frames were relatively more prevalent than collective action frames (28.7% vs. 19.3%), suggesting that individual actors mobilized the movement primarily through expressing their personal appeals or opinions. Indeed, through collective action frames, people could share common concerns regarding sexual violence (i.e., attributing the cause of which to injustice against women and evoking oppositional consciousness) to scale up the development of #MeToo.

Issue-Specific Frames

The other half of the Weibo posts (n = 8,404, 51.9%) was constructed through issue-specific frames focusing on three aspects: problem definition, treatment recommendation, and related news. Posts that adopted such frames mainly concentrated on specific sexual violence cases at colleges apart from the prevalence of sexual violence present in the society.

Problem definition (n = 3,320, 20.5%; Topic 6, 7, 8, 9) was used to portray the facts of exposed cases from different angles, such as the definition of sexual harassment/assault, the severity and salience of the focal cases, and the responsibilities of the involved stakeholders (e.g., predators, universities, and policy makers). Treatment recommendation (n = 3,335, 20.6%; Topic 10, 11, 12, 13) was applied to provide suggestions for preventing sexual violence inflicted upon students and children. The suggestions ranged from establishing laws or systematic mechanisms on the macro-level to educating children by their parents on the microlevel. In other words, treatment recommendation showed that people lay hopes on government or relevant departments or educating children by their parents (other than calling the public for active participations to change the situation). Related news (n = 1,749, 10.8%; Topic 14, 15, 16) was manifested in posts that simply reported or described sexual violence cases in other countries, such as “the US” or, more specifically, “Hollywood.” The appearance of this frame suggests that with the aid of information technologies, people connected local issues with global events or other relevant local issues to make a discussion.

The Influence of Online Identity

To evaluate gender’s influence on framing (Research Question 2), we firstly filtered out organizational users (as gender is not applicable to them). As a result, 14,684 Weibo posts were included to compare gender differences. The χ² test results revealed that males and females adopted different frames when they became involved in the movement (χ² = 791.90, p < .001). As reflected by Figure 3, females were more likely to use movement-oriented frames. In detail, the post hoc tests reported that females adopted certain frames more so than males, such as personal experiences (28.0% vs. 14.3%; χ² = 412.09, p < .001), emotional commentary (11.3% vs. 7.6%; χ² = 57.76, p < .001), and injustice (13.5% vs. 11.3%; χ² = 15.21, p < .001). Meanwhile, some issue-specific frames were more prevalent among males than females, including problem definition (27.0% vs.13.1%; χ² = 449.44, p < .001) and related news (12.0% vs. 9.8; χ² = 38.44, p < .001). No significant difference was identified between females and males in their use of oppositional frame (7.1% vs. 8.0%; χ² = 4.41, p > .05) or treatment recommendation (18.3% vs. 19.8%; χ² = 5.76, p > .05).

Figure 3.

Relative prevalence of frames by genders.

In terms of Research Question 3, the results unveiled that significant differences occurred across the frames adopted by different Weibo account holders (χ² = 863.03, p < .001). As presented in Figure 4, the post hoc results revealed a descending trend across ordinary users, influencers, and organizational users in their adoption of movement-oriented frames, including personal experiences (22.9% vs. 13.1% vs. 4.4%, χ² = 320.41, p < .001), emotional commentary (9.9% vs. 7.2 vs. 3.3, χ² = 67.42, p < .001), and injustice (12.8% vs. 10.3% vs. 4.7%, χ² = 62.41, p < .001). Regarding issue-specific frames, ordinary users were far less likely to adopt problem definition than influencers and organizational users, and no significant difference was identified between the other two groups (17.9% vs. 30.3% vs. 27.5%, χ² = 225.00, p < .001). Indeed, organizational users registered a higher ratio of treatment recommendation uses than ordinary users and influencers, and no significant difference was determined between the other two groups (ordinary users and influencers; 36.4% vs. 19.0% vs. 18.7%; χ² = 81.00, p < .001). No significant differences were found in the adoption of oppositional frame (7.4% vs. 8.1% vs. 8.9%; χ² = 4.41, p >.05) or related news (10.0% vs. 12.3% vs. 14.9%; χ² = 6.25, p >.05) among ordinary users, influencers, and organizational users.

Figure 4.

Relative prevalence of frames by account types.

Discussion

By examining the #MeToo movement in China, this study represents one of the first attempts to empirically investigate meaning construction in connective movements. This study has uncovered seven specific types of frames that emerged from China’s #MeToo movement—that is, personal experiences and emotional commentary (as personal action frames), injustice and opposition (as collective action frames), and problem definition, treatment recommendation, and related news (as issue-specific frames). Moreover, it is herein illustrated that social media users’ adoption of frames was subject to their identities (i.e., gender and account types). These findings greatly enhance our understanding of the public’s engagement in digital media–enabled movements.

Our results reveal that personal action and collective action frames were found to be of use in China’s #MeToo movement in orienting the public discourse toward the direction of a social movement. The emergence of two movement-oriented frame genres confirms the proposition that the organizing mechanisms in connective action movements are accomplished via mixed (collective and personal) framings (Lim, 2013; Wright, 2015). Within movement-oriented frames, personal action frames were adopted more frequently, thereby indicating that connective movements are primarily mobilized by personalized actions (Bennett & Segerberg, 2013). These results further render empirical support for the importance of personal action frames during the meaning construction of connective movements. It is herein demonstrated that segmented individuals actively shared their personal experiences and expressed emotional commentary to participate in #China’s #MeToo. By sharing their personal experiences, segmented individuals—playing the role of movement actors—echoed the movement’s initial advocacy and pushed forward its development. Through emotional commentary, movement actors explicitly expressed their personal opinions regarding sexual harassment and facilitated the participation of their connected friends. In particular, the emojis used in emotional commentary can elicit empathy and sympathy and further generate feelings of solidarity in online movements (Santhanam et al., 2018). This finding confirmed that technological affordances largely promote ordinary people’s engagements in public issues. These results, taken together, imply that movement actors expressed distinct aspects of their personal lives or thoughts in response to movement themes and that they, through self-organizing, became involved in the movement with no facilitation from central organizations.

As discussed, this study also found that collective action frames were employed to organize coordination in a connective movement such as #MeToo. The injustice frame conveyed a common concern (i.e., most people may be considered victims in unequal power relations), whereas the oppositional frame was adopted to advocate action toward the common good (i.e., combatting the victim-related stigma is regarded as beneficial to the whole society). This means that, despite the fact that the Weibo posts were separately produced by segmented individuals, these separate posts similarly reflected common concerns or the common good and thereby led to a loose manner of collective mobilization. These findings suggest that although no centralized organization is involved in mobilizing the movement, individual actors can actually form a common ground from the perspective that the whole society will benefit (Bennett & Segerberg, 2013). Hence, in connective movements, people are driven to participate not only for personal reasons but also because of common concerns or the common good.

These specific movement-oriented frames indicate that even in an authoritarian sociopolitical context such as China, segmented social actors can still promote the development of social movements by utilizing various technological affordances of social media and accomplish online mobilization (Lin & Yang, 2019; Zeng, 2020). However, although #MeToo movement gained great influence in China, the findings also reveal the limitations of developing connective movements by relying solely on fragmented individuals who are armed with digital technologies but do not receive clear guidance and potent support from movement organizations or activists (partially due to the censorship from the authorities). As reflected by these movement-oriented frames, social media users contributed to the movement mostly in the form of communicative actions, such as sharing personal experiences or expressing emotional commentaries, which could not lead to social changes immediately and directly (Wang & Shi, 2018; Zeng, 2020).

Apart from movement-oriented frames, issue-specific frames were additionally manifested in China’s #MeToo movement. Different from movement-oriented frames, which constructed the exposed sexual harassment cases as a pervasive societal problem, issue-specific frames served to set the boundary (within the scope of college campuses) for public debate concerning such cases. Such a finding reflects the attempt of the fragmented public engaged in connective movements to frame the focused issue in different or even competing directions and influence public opinion.

Particularly, the study reveals a gender gap in the framing of the sexual harassment issue, as the results imply that movement-oriented frames were more prevalent among female users and issue-specific frames were more prevalent among male users (see Figure 3). This finding is reasonable because females are more vulnerable to sexual violence in the current social structure. #MeToo offered Chinese females the opportunity to voice themselves on social media by actively sharing their own stories (personal experiences), emotionally expressing their personal opinions (emotional commentary), and condemning unjust power relations (injustice). Inversely, under the “male-dominant” power structure (Uggen & Blackstone, 2004), males are conditioned to be conservative toward gender-related issues (Armstrong & Mahone, 2017). Therefore, when participating in #MeToo movement, males were less ardent and less action-oriented by framing sexual harassment as a phenomenon limited to certain fields (i.e., colleges) or by focusing more on cases in societies other than those in which they lived.

The study found that Weibo users framed sexual harassment differently according to their account types. This result indicates that social media empower ordinary people (Leong et al., 2018), as ordinary Weibo users served as movement actors in China’s #MeToo movement by emphasizing the issue’s salience and mobilizing users’ participation (i.e., via movement-oriented frames). Meanwhile, given the impact of China’s political ideology and censorship, organizational users strategically framed the exposed sexual harassment cases as a particular problem in educational sectors to guide public conversations; in a similar vein, ordinary and organizational users framed the same issue to compete for prominence and to increase their influence, thereby suggesting that framing is a strategic action (Scheufele, 1999). Notably, Weibo influencers were less active in employing movement-oriented frames than ordinary users as well as less active in proposing recommendations than organization users. This finding indicates that, due to the Chinese government’s censorship, influencers tend to “play it safe” when discussing sensitive social problems by, for instance, primarily relying on the problem definition frame to engage in #MeToo movement (King et al., 2013).

Conclusion

A few limitations of this study should be noted. For example, we primarily concentrated on communicative actions in the online context and thus concluded that social media facilitate the #MeToo movement in China. However, social media’s effectiveness in facilitating concretely off-line civic participation remains unclear and is thus worth studying in the future. This aspect is especially important in Chinese society, where collective civic engagement in offline settings is traditionally unappreciated by the government. In addition, to select Weibo posts related to sexual harassment and sexual assault at colleges, we employed a key word-based searching method. Although we adopted a comprehensive key word list to collect accurate data to the greatest extent possible, we could not avoid the common drawback of scrapping a large amount of online data—some irrelevant data would be included in the data set (boyd & Crawford, 2012). For instance, some posts that mentioned the listed keywords by coincidence but were essentially unrelated to #MeToo movement could be selected in our final data set and thereby bias our interpretation to a certain extent. Future research may thus utilize a more delicately designed data collection method to attain more comprehensive insights into this issue.

In conclusion, by taking into account personal action frames, collective action frames, and issue-specific frames, this study provides a useful analytic tool—that is, the multidimensional framing analysis—for examining public discourse on connective movements. By conducting a computational content analysis of Weibo posts, this study examined frames that emerged during China’s #MeToo movement, and the results offer empirical understandings regarding the scholarship of connective movements. The specific types of personal action frames identified in this study (i.e., personal experiences and emotional commentary) explicate in which aspects segmented individuals construct meanings to contribute to connective movements, whereas the different framings that emerged during China’s #MeToo movement reveal how the mixed organizing mechanisms mobilize resources in connective movements. Furthermore, the findings present a valuable snapshot for learning how the Chinese public perceives and responds to sexual harassment and related issues.

Footnotes

Data Availability

For the replicability of our study, we will provide the final data set of Weibo posts by responding to email requests (leepengxiang@hotmail.com or caf16@ustc.edu.cn).

Software Information

Topic modeling was conducted through Python Environment Version 3.6.5 (http://https://www.python.org/), and the major Python library used for the analysis included: LDA, Pandas, NumPy, and Jieba. The major code for Python is available online (https://github.com/xiaoanzi123/Topic-Modeling_LDA). Figure 1 was visualized through Gephi Version 0.9.2 ().

Declaration of Conflicting Interests

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

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Note

References

Agarwal

S. D.

Bennett

W. L.

Johnson

C. N.

Walker

(2014). A model of crowd-enabled organization: Theory and methods for understanding the role of Twitter in the occupy protests. International Journal of Communication, 8, 646–672.

Anduiza

Cristancho

Sabucedo

J. M.

(2014). Mobilization through online social networks: The political protest of the Indignados in Spain. Information, Communication & Society, 17(6), 750–764. https://doi.org/10.1080/1369118x.2013.808360

Armstrong

C. L.

Mahone

(2017). “It’s on us.” The role of social media and rape culture in individual willingness to mobilize against sexual assault. Mass Communication and Society, 20(1), 92–115. https://doi.org/10.1080/15205436.2016.1185127

Bang

Halupka

(2019). Contentious connective action: A new kind of life-political association for problematizing how expert systems operate. Information, Communication & Society, 22(1), 89–104. https://doi.org/10.1080/1369118x.2017.1355402

Beasley

T. M.

Schumacker

R. E.

(1995). Multiple regression approach to analyzing contingency tables: Post hoc and planned comparison procedures. The Journal of Experimental Education, 64(1), 79–93. https://doi.org/10.1080/00220973.1995.9943797

Benford

R. D.

Snow

D. A.

(2000). Framing processes and social movements: An overview and assessment. Annual Review of Sociology, 26(1), 611–639. https://doi.org/10.1146/annurev.soc.26.1.611

Bennett

W. L.

Segerberg

(2013). The logic of connective action: Digital media and the personalization of contentious politics. Cambridge University Press.

Bennett

W. L.

Segerberg

Walker

(2014). Organization in the crowd: Peer production in large-scale networked protests. Information, Communication & Society, 17(2), 232–260. https://doi.org/10.1080/1369118x.2013.870379

Blei

D. M.

(2012). Probabilistic topic models. Communications of the ACM, 55(4), 77–84. https://doi.org/10.1145/2133806.2133826

10.

boyd

Crawford

(2012). Critical questions for big data: Provocations for a cultural, technological, and scholarly phenomenon. Information, Communication & Society, 15(5), 662–679. https://doi.org/10.1080/1369118X.2012.678878

11.

Brüggemann

(2014). Between frame setting and frame sending: How journalists contribute to news frames. Communication Theory, 24(1), 61–82. https://doi.org/10.1111/comt.12027

12.

Castells

(2015). Networks of outrage and hope: Social movements in the internet age (2nd ed.). Polity Press.

13.

Chen

Huang

Jiang

(2020). Empowering Chinese college students to prevent sexual assault in post-MeToo Era: An empirical study of the bystander intervention approach. Journal of Interpersonal Violence. Advance online publication. https://doi.org/10.1177/0886260520917515

14.

Chen

C. C.

Chen

(2019). Top-down or bottom-up? A network agenda-setting study of Chinese nationalism on social media. Journal of Broadcasting & Electronic Media, 63(3), 512–533. https://doi.org/10.1080/08838151.2019.1653104

15.

Choi

(2018). When digital trace data meet traditional communication theory: Theoretical/methodological directions. Social Science Computer Review, 38(1), 91–107.

16.

DeLuca

K. M.

Lawson

Sun

(2012). Occupy wall street on the public screens of social media: The many framings of the birth of a protest movement. Communication, Culture & Critique, 5(4), 483–509. https://doi.org/10.1111/j.1753-9137.2012.01141.x

17.

Entman

R. M.

(1993). Framing: Toward clarification of a fractured paradigm. Journal of Communication, 43(4), 51–58. https://doi.org/10.1111/j.1460-2466.1993.tb01304.x

18.

Fileborn

(2016). Justice 2.0: Street harassment victims’ use of social media and online activism as sites of informal justice. British Journal of Criminology, 57(6), 1482–1501. https://doi.org/10.1093/bjc/azw093

19.

Hon

(2016). Social media framing within the million hoodies movement for justice. Public Relations Review, 42(1), 9–19. https://doi.org/10.1016/j.pubrev.2015.11.013

20.

King

Pan

Roberts

M. E.

(2013). How censorship in China allows government criticism but silences collective expression. American Political Science Review, 107(2), 326–343. https://doi.org/10.1017/s0003055413000014

21.

Lee

F. L. F.

Chan

J. M.

(2016). Digital media activities and mode of participation in a protest campaign: A study of the Umbrella Movement. Information, Communication & Society, 19(1), 4–22. https://doi.org/10.1080/1369118X.2015.1093530

22.

Leong

Pan

S. L.

Bahri

Fauzi

(2018). Social media empowerment in social movements: Power activation and power accrual in digital activism. European Journal of Information Systems, 28(2), 173–204. https://doi.org/10.1080/0960085x.2018.1512944

23.

Lewis

S. C.

Zamith

Hermida

(2013). Content analysis in an era of big data: A hybrid approach to computational and manual methods. Journal of Broadcasting & Electronic Media, 57(1), 34–52. https://doi.org/10.1080/08838151.2012.761702

24.

Lim

(2013). Framing Bouazizi: “White lies,” hybrid network, and collective/connective action in the 2010–11 Tunisian uprising. Journalism, 14(7), 921–941. https://doi.org/10.1177/1464884913478359

25.

Lin

Yang

(2019). Individual and collective empowerment: Women’s voices in the #MeToo movement in China. Asian Journal of Women’s Studies, 25(1), 117–131. https://doi.org/10.1080/12259276.2019.1573002

26.

Liss

Crawford

Popp

(2004). Predictors and correlates of collective action. Sex Roles, 50(11), 771–779. https://doi.org/10.1023/B:SERS.0000029096.90835.3f

27.

Lovejoy

Saxton

G. D.

(2012). Information, community, and action: How nonprofit organizations use social media. Journal of Computer-Mediated Communication, 17(3), 337–353. https://doi.org/10.1111/j.1083-6101.2012.01576.x

28.

Maier

Waldherr

Miltner

Jähnichen

Pfetsch

(2018). Exploring issues in a networked public sphere: Combining hyperlink network analysis and topic modeling. Social Science Computer Review, 36(1), 3–20.

29.

Marshall

A. M.

(2003). Injustice frames, legality, and the everyday construction of sexual harassment. Law & Social Inquiry, 28(3), 659–689. https://doi.org/10.1111/j.1747-4469.2003.tb00211.x

30.

Mat-

(2018). The collection of cases about MeToo movement in China. https://matters.news/@jidongjie

31.

Nelson

L. K.

(2017). Computational grounded theory: A methodological framework. Sociological Methods & Research. https://doi.org/10.1177/0049124117729703

32.

Nip

J. Y.

K.-w.

(2016). Networked framing between source posts and their reposts: An analysis of public opinion on China’s microblogs. Information, Communication & Society, 19(8), 1127–1149. https://doi.org/10.1080/1369118x.2015.1104372

33.

Obregón

Tufte

(2017). Communication, social movements, and collective action: Toward a new research agenda in communication for development and social change. Journal of Communication, 67(5), 635–645. https://doi.org/10.1111/jcom.12332

34.

Rauchfleisch

Schäfer

M. S.

(2015). Multiple public spheres of Weibo: A typology of forms and potentials of online public spheres in China. Information, Communication & Society, 18(2), 139–155.

35.

Sandoval

(1991). U.S. third world feminism: The theory and method of oppositional consciousness in the postmodern world. Genders, (10), 1–24. https://doi.org/10.5555/gen.1991.10.1

36.

Santhanam

Srinivasan

Glass

Shaikh

(2018). I stand with you: Using Emojis to study solidarity in crisis events [Paper presentation]. Proceedings of the First International Workshop on Emoji Understanding and Applications in Social Media (Emoji2018), Stanford, CA, United States.

37.

Scheufele

D. A.

(1999). Framing as a theory of media effects. Journal of Communication, 49(1), 103–122. https://doi.org/10.1111/j.1460-2466.1999.tb02784.x

38.

Schofield

Magnusson

Mimno

(2017, April). Pulling out the stops: Rethinking stopword removal for topic models. Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics. https://www.aclweb.org/anthology/E17-2069/

39.

Shoemaker

P. J.

Reese

S. D.

(2013). Mediating the message in the 21st century: A media sociology perspective. Routledge.

40.

Skalli

L. H.

(2014). Young women and social media against sexual harassment in North Africa. The Journal of North African Studies, 19(2), 244–258. https://doi.org/10.1080/13629387.2013.858034

41.

Snow

Benford

(1988). Ideology, frame resonance, and participant mobilization. In Klandermans

Kriesi

Tarrow

S. G.

(Eds.), From structure to action: Comparing social movement research across cultures ( International Social Movement Research , pp. 197–217). JAI Press.

42.

Swank

Fahs

(2014). Predictors of feminist activism among social work students in the United States. Social Work Education, 33(4), 519–532. https://doi.org/10.1080/02615479.2013.860096

43.

Trumpy

A. J.

(2016). “I Hate It, But it still sounds good”: Collective identity and the evaluation of oppositional frame resonance. Social Movement Studies, 15(2), 164–179. https://doi.org/10.1080/14742837.2015.1060157

44.

Uggen

Blackstone

(2004). Sexual harassment as a gendered expression of power. American Sociological Review, 69(1), 64–92. https://doi.org/10.1177/000312240406900105

45.

Vaast

Safadi

Lapointe

Negoita

(2017). Social media affordances or connective action: An examination of microblogging use during the Gulf of Mexico oil spill. MIS Quarterly, 41(4), 1179–1205. https://doi.org/10.25300/MISQ/2017/41.4.08

46.

Van Laer

Van Aelst

(2010). Internet and social movement action repertoires: Opportunities and limitations. Information, Communication & Society, 13(8), 1146–1171. https://doi.org/10.1080/13691181003628307

47.

Veenstra

A. S.

Sayre

Shah

D. V.

Mcleod

D. M.

(2008). Frames and knowledge in mixed media: How activation changes information intake. Cyber Psychology & Behavior, 11(4), 443–450. https://doi.org/10.1089/cpb.2007.0033

48.

Wang

Shi

(2018). Weibo use and political participation: The mechanism explaining the positive effect of Weibo use on online political participation among college students in contemporary China. Information, Communication & Society, 21(4), 516–530. https://doi.org/10.1080/1369118X.2017.1289234

49.

Wang

W. Y.

(2013). Weibo, framing, and media practices in China. Journal of Chinese Political Science, 18(4), 375–388. https://doi.org/10.1007/s11366-013-9261-3

50.

Wright

(2015). Populism and downing street e-petitions: Connective action, hybridity, and the changing nature of organizing. Political Communication, 32(3), 414–433. https://doi.org/10.1080/10584609.2014.958256

51.

Xiong

Cho

Boatwright

(2018). Hashtag activism and message frames among social movement organizations: Semantic network analysis and thematic analysis of Twitter during the #MeToo movement. Public Relations Review. https://doi.zorg/10.1016/j.pubrev.2018.10.014

52.

Zeng

(2019). You say #MeToo, I Say #MiTu: China’s online campaigns against sexual abuse. In Fileborn

Loney-Howes

(Eds.), #MeToo and the politics of social change (pp. 71–83). Palgrave Macmillan. https://doi.org/10.1007/978-3-030-15213-0_5

53.

Zeng

(2020). #MeToo as connective action: A study of the anti-sexual violence and anti-sexual harassment campaign on Chinese social media in 2018. Journalism Practice, 14(2), 1–20. https://doi.org/10.1080/17512786.2019.1706622