Exploring the Interplay between Social Media Use, Identity Anxiety, Social Comparison, and Gender Attitudes in China: A Xiaohongshu-Based Perspective ()
1. Introduction
1.1. Theoretical Foundation
1.1.1. Social Comparison Theory
Social Comparison Theory, originally proposed by Festinger (1954), posits that individuals have an innate drive to evaluate their own abilities, achievements, and opinions by comparing themselves with others. Such comparisons may be upward (with individuals perceived as better off) or downward (with those perceived as worse off). Upward social comparisons often lead to feelings of inadequacy or envy, whereas downward comparisons can boost self-esteem. In the age of social networking sites (SNS), this theory is highly relevant: platforms like Xiaohongshu present curated images of others’ lives, making “everyone is better than me” illusions more likely.
1.1.2. Social Identity Theory
While social comparison focuses on individual self-evaluation, Social Identity Theory (Tajfel & Turner, 1979) emphasizes the role of group membership in one’s self-concept. According to Tajfel and Turner, people categorize themselves and others into social groups (e.g. by nationality, gender, or in-group vs out-group) as a way of defining their identity and gaining self-esteem. Individuals strive for a positive social identity by favorably comparing their in-group to relevant out-groups.
1.2. Xiaohongshu (RED): A Chinese Social Media Phenomenon
Xiaohongshu is a prominent Chinese social media platform that integrates content sharing, e-commerce, and community engagement. Launched in 2013, it has grown rapidly, boasting over 200 million users by 2023 (Zhang & Wang, 2023), with recent estimates indicating a monthly active user base exceeding 300 million (Chen et al., 2024). The platform’s demographic skews young and female―as of 2020, 70% of its users were born in the 1990s, and approximately 70% were female (Li & Liu, 2021). Notably, Xiaohongshu has also gained traction among international users, particularly “TikTok refugees”―foreigners who migrated to the platform after facing restrictions or cultural mismatches on Western social media (Wilson & Tan, 2022), sparking friendly cross-cultural interactions between American and Chinese users online (Park & Yang, 2023). This unusual influx highlights Xiaohongshu’s expanding global footprint, its growing influence, and its role as a unique space for cultural exchange and comparison (Guo, 2023). Given its algorithm-driven content management and strong community culture (Huang et al., 2022), studying Xiaohongshu’s user psychology offers valuable insights into digital identity construction (Goffman, 1959; Zhao et al., 2021a), parasocial relationships (Horton & Wohl, 1956; Xu & Li, 2023), and the impact of social media on mental health (Andreassen et al., 2017; Liu et al., 2024). Its popularity and unique user base make it an ideal environment for examining the relationship between social media use and psychological variables such as social comparison (Festinger, 1954; Feng & Kim, 2023) and identity anxiety (Erikson, 1968; Wang & Zhang, 2022), as well as social attitudes like views on gender equality (Inglehart & Norris, 2003; Li & Chen, 2024).
1.3. Active vs. Passive Social Networking Site Use
Social networking sites (SNS) play a central role in modern communication, offering both active and passive engagement. Xiaohongshu, one of China’s most widely used platforms, exemplifies this dual usage pattern. Active SNS use refers to direct interactions such as posting, commenting, or messaging, which can foster connection, self-expression, and social capital (Burke et al., 2010; Deters & Mehl, 2013; Ellison et al., 2007; Valkenburg et al., 2017a; Verduyn et al., 2015). On Xiaohongshu, such engagement may involve sharing lifestyle content or participating in discussions, reinforcing a sense of identity and community (Zhao et al., 2021b).
In contrast, passive SNS use involves consuming content without interaction, such as scrolling through feeds or viewing profiles (Krasnova et al., 2013; Liu et al., 2024). This mode of use has been linked to negative emotional outcomes through upward social comparison (Chou & Edge, 2012; Festinger, 1954; Fardouly et al., 2015; Tiggemann & Slater, 2017; Verduyn et al., 2020). On Xiaohongshu, browsing idealized posts may heighten feelings of inadequacy (Appel et al., 2016; Sagioglou & Greitemeyer, 2014; Wang et al., 2022). Research suggests that while active use can enhance social connectedness, passive use is associated with emotional declines via mechanisms like FoMO and vicarious emotional exhaustion (Liu et al., 2023; Meier & Reinecke, 2021; Przybylski et al., 2013; Verduyn et al., 2020). Understanding this distinction is critical for evaluating SNS-related psychological effects.
1.4. Social Comparison and Identity Anxiety
Social comparison and identity anxiety are two pivotal constructs influencing self-concept, emotional well-being, and behavior. Identity anxiety reflects distress caused by uncertainty or conflict in self-definition, especially under rapid societal change (Cheek et al., 1994; Erikson, 1968; Marcia, 1980). In China, youth from rural backgrounds often face such anxiety when adapting to urban life, navigating between rural and urban identities (Wang & Yeh, 2017; Zhang & Li, 2020). This uncertainty has been linked to reduced self-concept clarity and a drive for external validation (Baumeister, 1997), with rural students showing higher anxiety than urban peers (National Health Commission, 2022; Chen & Zhang, 2023; Wu & Leung, 2021). SNS platforms like Xiaohongshu further amplify this anxiety by offering spaces for identity performance and comparison (Goffman, 1959; Zhao et al., 2021a). Research has shown passive SNS use increases identity anxiety via upward comparisons, especially among rural students (Yang & Wang, 2023; Li et al., 2022).
Social comparison―evaluating oneself against others―is a core process in forming self-worth (Festinger, 1954; Gibbons & Buunk, 1999). While adaptive in moderation (Collins, 1996; Wood, 1996), frequent upward comparisons on SNS can harm self-esteem (Verduyn et al., 2015; Krasnova et al., 2013; Feinstein et al., 2013). On Xiaohongshu, aspirational content fosters such comparisons, distorting self-perception (Tesser, 1988; Feng & Kim, 2023; Chou & Edge, 2012). Studies show that social comparisons reduce self-esteem and increase body dissatisfaction and anxiety (Vogel et al., 2014; Fardouly et al., 2020; Xu & Zhang, 2022; Yang et al., 2022).
In China’s collectivist and competitive culture, social comparison and identity anxiety are deeply intertwined (Markus & Kitayama, 1991; Zhang et al., 2015). Societal benchmarks like academic and material success intensify this interplay, particularly amid globalization and rapid change (Wang et al., 2017; Higgins, 1987). These pressures are magnified for marginalized groups navigating conflicting values (Zhou, 2020). Social media further embeds traditional gender ideals, linking identity anxiety and comparison to shifting attitudes toward gender equality (Fardouly et al., 2020; Eagly & Wood, 2012). Understanding this interplay offers insight into how SNS engagement influences youth psychology and values (Twenge & Campbell, 2018; Inglehart & Norris, 2003).
1.5. Attitudes Toward Gender Equality in China
Attitudes toward gender equality encompass individuals’ beliefs, emotions, and behaviors concerning the roles and rights of men and women across domains like education, employment, and family life (Else-Quest & Hyde, 2016; Glick & Fiske, 2022; Inglehart & Norris, 2003; Ridgeway, 2011). These attitudes significantly influence life outcomes, including career choices, education, household labor division, and mental health (Bianchi et al., 2012; Charles & Bradley, 2009; Correll, 2004; Zhang, 2008). Traditional gender role attitudes reinforce stereotypes and limit opportunities for women, whereas egalitarian views promote flexibility and equity (Bolzendahl & Myers, 2004; Eagly & Wood, 2012). Gender attitudes also reflect broader cultural values and social progress (Glick & Fiske, 1996; Inglehart & Welzel, 2005).
Historically, many societies―including China―have favored sons and upheld rigid gender divisions (Das Gupta et al., 2003; Greenhalgh, 1985; Li & Jiang, 2021; Murphy et al., 2011). Although gender norms in urban China are shifting, traditional roles persist, especially in rural and working-class contexts (Ji, 2017; Zuo & Bian, 2021; Zhou, 2020). Implicit measures reveal lingering biases even among those expressing explicit support for gender equality (Charlesworth & Banaji, 2021; Greenwald et al., 2000; Nosek et al., 2007; Zhang, 2008).
Young Chinese generations show more egalitarian views, yet cultural contradictions endure―highly educated women face pressures to conform to traditional expectations around marriage and caregiving (China Youth Daily Social Survey Center, 2023; Hu & Scott, 2016; Ji, 2017; Zhou, 2020). While middle-class families increasingly embrace gender equality, many working-class men continue to justify unequal household contributions (Cooke, 2010; Sun & Chen, 2021; Zuo & Bian, 2001). China’s ranking of 106th in the Global Gender Gap Index underscores these persistent disparities (Attané, 2012; World Economic Forum, 2022). Thus, gender attitudes remain a critical variable for understanding social change and informing policy (Hannum et al., 2019).
1.6. Linking Social Media Use, Social Comparison, Identity Anxiety, and Gender Attitudes
A growing body of research has examined pairs of these variables, but a comprehensive picture integrating all four remains lacking. Prior studies have established several bilateral links. First, consistent with social comparison theory (Festinger, 1954), passive SNS use tends to foster upward social comparisons, which in turn can lead to negative psychological outcomes. For example, passive consumption of social media has been linked to envy and lower subjective well-being in both Western and Chinese samples (Verduyn et al., 2015; Zhao & Fu, 2022). In the Chinese context, upward comparisons on SNS have been specifically implicated in aggravating identity-related anxieties among youth (Liu et al., 2019). The aforementioned study of rural college students is illustrative: it found that upward social comparison mediated the relationship between passive social media use and identity anxiety (Wang & Zhang, 2021). In other words, those who scrolled Xiaohongshu or similar platforms without engaging were more likely to compare themselves to others’ seemingly superior lives, which then heightened their anxiety about their own social identity and status. This aligns with social identity theory (Tajfel & Turner, 1979)―when one’s in-group (e.g., rural students) is perceived as lower status compared to an out-group (urban peers on social media), identity threat and anxiety can result (Chen et al., 2020).
Despite limited direct evidence, we hypothesize that identity anxiety and social comparison may serve as pathways through which social media use influences gender-related attitudes. One possibility is that individuals experiencing identity insecurity (e.g., uncertainty in their social role) might cling to traditional norms, including gender roles, as a source of stability (Giddens, 1991). Conversely, those who are striving to establish a new identity (for instance, rural youth trying to assimilate into urban lifestyles) might adopt more progressive views, rejecting old norms as part of crafting a modern identity (Zhang & Sun, 2023). Past research in China hints at such dynamics: for example, education and exposure to urban values (often facilitated by media) correlate with more egalitarian gender attitudes (Li et al., 2022). Social networking sites like Xiaohongshu expose users to a plethora of content―from feminist discourse and career women role models to traditional family narratives―which could shape or reinforce their gender role attitudes (Wu & Leung, 2021). However, studies explicitly connecting SNS use or social comparison with gender ideology are limited (Anderson & Jiang, 2018). Similarly, while identity anxiety has been studied in relation to mental health (Stein & Kean, 2000), its relationship with social attitudes (like gender equality beliefs) remains under-examined.
In sum, prior research suggests that active/passive SNS use, social comparison, identity anxiety, and gender attitudes are interrelated in complex ways. Active engagement might buffer against some negative effects of social comparison (by providing social support or identity affirmation), whereas passive use can heighten comparison and anxiety (Valkenburg et al., 2017b). Social comparison appears to be a key mechanism linking social media use to psychological outcomes (Appel et al., 2016). And broader social identity factors likely play a role in shaping both identity anxiety and gender attitudes (Turner et al., 1987). What is missing is an integrative analysis that considers all four factors together, to see how they jointly influence one another in a specific cultural context.
1.7. Present Study: Research Gap and Objectives
Despite the rich insights from earlier studies, no prior work has, to our knowledge, examined these four variables in combination within the context of Xiaohongshu or any Chinese social media platform. This represents a clear research gap. Xiaohongshu provides a salient arena to explore these relationships, given its distinctive user base and content that could plausibly impact social comparison, identity formation, and gender norms. The present study aims to fill this gap by investigating the interplay among Active SNS Use, Passive SNS Use, Social Comparison, Identity Anxiety, and Attitudes Toward Gender Equality among young Chinese Xiaohongshu users. Grounded in Social Comparison Theory and Social Identity Theory, we propose a model in which different styles of SNS use (active vs. passive) may have differential effects on users’ tendency to engage in social comparison and on their feelings of identity anxiety, which in turn could influence their attitudes about gender equality. By examining all four variables together, this study moves beyond fragmented findings to provide a more holistic understanding of how digital behavior connects with psychological well-being and social values.
In doing so, our research offers several contributions. Theoretically, it integrates two prominent frameworks―Festinger’s and Tajfel & Turner’s―to explain social media effects in a non-Western context, enriching the cross-cultural applicability of these theories. Empirically, it extends previous Chinese studies on social media and mental health by incorporating the dimension of gender attitude, an outcome with societal significance. Practically, the findings can inform educators and policymakers about the potential impact of social media usage patterns: for instance, whether encouraging more active engagement (rather than passive scrolling) might mitigate identity-related anxieties or foster more egalitarian attitudes in the next generation. In summary, this study addresses an important gap at the intersection of online behavior, individual psychology, and social attitude formation. The following sections will detail our methodology and analytic approach to testing the proposed relationships, but first, this introduction has laid the conceptual groundwork and highlighted why examining these factors together is both novel and necessary.
2. Methods
2.1. Participants and Procedure
We employed a cross-sectional survey design to examine the relationships between Active and Passive Social Networking Site Use and Attitudes Toward Gender Equality. The independent variables consisted of Active Social Networking Site Use (e.g., posting content, commenting) and Passive Social Networking Site Use (e.g., browsing without interaction). The dependent variable was the participants’ Attitudes Toward Gender Equality. Additionally, two potential moderators were examined: Identity Anxiety and Social Comparison.
Data were collected in April 2025 through Wenjuanxing, a widely used Chinese online survey platform that enables the distribution and collection of electronic questionnaires. A total of 507 participants were recruited using snowball sampling techniques and completed the survey independently. The average completion time was approximately 15 minutes. Prior to participation, all respondents provided informed consent via a digital consent form. The form outlined the purpose and procedures of the study, potential risks and benefits, participants’ rights (including voluntary participation and the option to withdraw at any time), and the researchers’ contact information for further inquiries. To ensure data quality, several safeguards were implemented: 1) multiple submissions from the same IP address were blocked by the platform; 2) responses that showed evidence of inattentive or careless responding (e.g., straight-lining, abnormally fast completion times) were excluded; and 3) attention check items were embedded throughout the survey to assess participant attentiveness. All responses were collected anonymously.
2.2. Measures
2.2.1. Active and Passive Social Networking Site Use Scale
Based on the operational definitions and measurement strategies of Active and Passive Social Networking Site Use proposed by Verduyn et al. (2015), an 8-item self-developed questionnaire was constructed to assess participants’ social networking behavior. Given that Xiaohongshu (RED) is one of the most popular social media platforms in China―functionally similar to Instagram―the measure was tailored to reflect user behaviors specific to this platform. The Active Social Networking Site Use subscale consisted of four items, including: 1) “Posting content on Xiaohongshu,” 2) “Actively searching for content of interest on Xiaohongshu,” 3) “Liking, commenting on, or saving posts made by unfamiliar users,” and 4) “Liking, commenting on, or saving posts made by friends.” The Passive Social Networking Site Use subscale also included four items, such as: 1) “Viewing updates from followed users on Xiaohongshu without liking or commenting,” 2) “Viewing updates from friends on Xiaohongshu without liking or commenting,” 3) “Viewing content shared by friends without reposting, liking, or commenting,” and 4) “Browsing other users’ profiles on Xiaohongshu without interacting with them.” Participants responded on a 5-point Likert scale ranging from 1 (Never) to 5 (Always) to indicate the frequency of each behavior. Total scores for each subscale were calculated by summing all items, with higher scores representing greater frequencies of active or passive SNS use. In the present study, internal consistency (Cronbach’s α) was.77 for the Active subscale and.68 for the Passive subscale.
2.2.2. Social Comparison Scale
Social comparison was assessed using a revised version of the Social Comparison Scale developed by Gibbons and Buunk (1999), which was translated into Chinese by Bai et al. (2013) and adapted to the context of social networking sites (e.g., QQ, WeChat Moments). The revised version consisted of 12 items that reflected upward comparisons specifically in SNS contexts (e.g., “I often compare myself with people who are doing better than me on social networking sites”). Participants were instructed to rate each item based on how well it described their behavior, using a 5-point Likert scale ranging from 1 (Strongly disagree) to 5 (Strongly agree). Higher total scores indicated a greater tendency to engage in upward social comparisons on SNSs. The internal consistency of the scale in the present study was excellent (Cronbach’s α = 0.92).
2.2.3. Identity Anxiety Scale
Identity anxiety was measured using a 13-item scale adapted from the AIQ-IIIX (Cheek et al., 1994), and validated in the dissertation of Yu Dan (2024). The scale focused on individuals’ self-perceptions of identity and future prospects. Sample items include: “I often feel uncertain about my social class,” and “I often feel that studying and working have not changed my social status.” Several items were adapted to fit the Chinese sociocultural context of modernization and class mobility. For example, “I feel some pressure when I hear people say that college students should be the pillars and talents of the country,” and “I often envy the high social status of college students twenty years ago who were considered the ‘pride of the nation.’” Participants responded using a 5-point Likert scale ranging from 1 (Strongly disagree) to 5 (Strongly agree), with higher total scores indicating greater levels of identity anxiety. The scale demonstrated excellent internal consistency in the present study (Cronbach’s α = 0.93).
2.2.4. Attitudes toward Gender Equality
Attitudes toward gender equality were measured using the 14-item Gender Equality Attitude Test, translated by scholar Lin Yanyu and adapted from Wang Tao’s (2008) gender identity self-assessment scale. Participants responded to each item on a 5-point Likert scale ranging from 1 (Strongly disagree) to 5 (Strongly agree), reflecting their level of endorsement of traditional versus egalitarian gender roles. Higher total scores indicated more egalitarian attitudes toward gender equality, whereas greater differences between male and female respondents’ scores were interpreted as stronger gender-based identity preferences. Example items include: “Women should stay at home, and men should work outside,” and “Caring for children is the most important duty of women.” The scale demonstrated excellent internal consistency in the present study (Cronbach’s α = 0.93).
2.3. Data Analyses
Moderation analyses were conducted using Structural Equation Modeling (SEM) in R (Version 4.4.1; R Core Team, 2025) with the lavaan package (Rosseel, 2012). All continuous predictor variables―Active and Passive Social Networking Site Use, Identity Anxiety, and Social Comparison―were mean-centered to reduce multicollinearity and improve interpretability (Aiken & West, 1991). Interaction terms were created by multiplying the centered Active and Passive Social Networking Site Use variables with Identity Anxiety and Social Comparison, respectively. Separate SEM models were specified to test whether Identity Anxiety and Social Comparison moderated the relationships between Active or Passive Social Networking Site Use and Attitudes Toward Gender Equality. Maximum likelihood (ML) estimation was used, and model fit was assessed using standard SEM fit indices, including the Comparative Fit Index (CFI), Tucker-Lewis Index (TLI), Root Mean Square Error of Approximation (RMSEA), and Standardized Root Mean Square Residual (SRMR) (Hu & Bentler, 1999). Notably, as the models were fully saturated (df = 0), fit indices were perfect by definition and should not be interpreted as evidence of substantive model fit.
To probe significant interaction effects, simple slopes analyses were conducted at three levels of Identity Anxiety: low (−1 SD), mean (0), and high (+1 SD). Predicted values were plotted to visually illustrate how Identity Anxiety moderated the associations between social media usage and Attitudes Toward Gender Equality (Preacher, Curran, & Bauer, 2006).
3. Results
3.1. Participant Demographic Information
The final sample consisted of 507 participants, with a nearly equal distribution of gender (see Table 1): 50.3% identified as male (n = 255) and 49.7% as female (n = 252). The average age of participants was 34.41 years (SD = 6.61), with an average height of 169.02 cm (SD = 4.33) and an average weight of 63.42 kg (SD = 4.02).
A majority of participants were of Han ethnicity (60.4%, n = 306), while the remaining 39.6% (n = 201) identified as members of minority ethnic groups. Educational attainment was diverse: 2.6% had completed only primary school or below (n = 13), 7.5% junior middle school (n = 38), 23.3% vocational high school (n = 118), and 17.2% general high school (n = 87). A total of 24.5% had earned a junior college degree (n = 124), 20.5% held an undergraduate degree (n = 104), and a smaller proportion had obtained a master’s (0.8%, n = 4) or doctoral degree (3.7%, n = 19).
Regarding marital status (see Table 1), most participants were married (78.9%, n = 400), followed by unmarried individuals (13.8%, n = 70) and those who reported “other” marital situations (7.3%, n = 37). The majority resided in urban areas (62.1%,
![]()
Table 1. Demographic characteristics of the sample.
n = 315), while 37.9% (n = 192) lived in rural areas. Employment status was reported as employed (75.0%, n = 380), unemployed (14.8%, n = 75), or other (10.3%, n = 52).
More than half of the sample (60.9%, n = 309) reported being only children, and all participants (100.0%, n = 507) reported that their parents’ marriage was stable.
Participants represented a range of occupational types (see Table 1). Manual laborers, self-employed individuals, and skilled workers accounted for the largest group (34.7%, n = 176), followed by temporary workers, the unemployed, and low-skill agricultural laborers (33.5%, n = 170). Students comprised 12.2% (n = 62), general administrative and technical personnel 13.2% (n = 67), mid-level professionals and managers 3.9% (n = 20), and senior managers and professionals 2.4% (n = 12).
Most participants reported having children (66.9%, n = 339), while 33.1% (n = 168) did not. Monthly income levels varied: 25.2% (n = 128) earned ¥3000 or below, 25.0% (n = 127) earned ¥3000 - ¥6000, and 19.7% (n = 100) earned ¥6000 - ¥9000. A smaller portion reported incomes of ¥9000 - ¥12,000 (10.7%, n = 54), ¥12,000 - ¥15,000 (10.5%, n = 53), or above ¥15,000 (8.9%, n = 45).
3.2. Descriptive Statistics and Correlations
Descriptive statistics for the key variables are presented in Table 2. On average, participants reported moderate levels of Active Social Networking Site Use (M = 14.20, SD = 3.62) and Passive Social Networking Site Use (M = 14.50, SD = 3.15). The mean score for Identity Anxiety was 46.37 (SD = 11.10), and for Social Comparison, 42.37 (SD = 10.32). The mean score for Attitudes Toward Gender Equality was 48.75 (SD = 11.64), suggesting a generally moderate-to-high endorsement of gender egalitarianism.
![]()
Table 2. Pearson correlations among main study variables
aNote. N = 507. **p < 0.01 (2-tailed).
Pearson correlation coefficients were computed to examine the associations among the main study variables (see Table 2). Active and Passive Social Networking Site Use were strongly positively correlated (r = 0.864, p < 0.001), indicating that individuals who engaged more frequently in one type of SNS activity also tended to engage more in the other. No other significant correlations were observed among the core variables. Specifically, Attitudes Toward Gender Equality were not significantly correlated with Active Social Networking Site Use (r = 0.009, p = 0.846), Passive Social Networking Site Use (r = 0.061, p = 0.167), Identity Anxiety (r = −0.054, p = 0.228), or Social Comparison (r = −0.015, p = 0.729). These results suggest minimal bivariate associations between social networking site behaviors and Attitudes Toward Gender Equality, implying that interaction effects may play a more influential role in these relationships.
![]()
Table 3. Moderation analysis predicting gender equality from active social media usage, identity anxiety, and social comparison.
![]()
Table 4. Moderation analysis predicting gender equality from passive social media usage, identity anxiety, and social comparison.
3.3. Moderation Analysis
The first SEM moderation model demonstrated excellent fit for testing whether Identity Anxiety and Social Comparison moderated the relationship between Active Social Networking Site Use and Attitudes Toward Gender Equality (see Table 3). Neither Active Social Networking Site Use (β = 0.014, p = 0.745), Identity Anxiety (β = ?0.053, p = 0.228), nor Social Comparison (β = ?0.010, p = 0.814) individually predicted Attitudes Toward Gender Equality. However, the interaction between Active Social Networking Site Use and Identity Anxiety was statistically significant (β = ?0.114, z = ?2.585, p = 0.010), indicating that Identity Anxiety moderated this relationship. In contrast, the interaction involving Social Comparison was not significant (β = ?0.004, p = 0.935).
To further examine this significant interaction, simple slopes analyses were conducted at three levels of Identity Anxiety: low (?1 SD), mean (0), and high (+1 SD) (see Figure 1). Results revealed that at low levels of Identity Anxiety, Active Social Networking Site Use was positively associated with Attitudes Toward Gender Equality, suggesting that individuals with lower anxiety reported more egalitarian attitudes as their active SNS engagement increased. In contrast, among those with high Identity Anxiety, the relationship was negative, indicating that higher active use was linked with lower gender equality attitudes. No significant association was observed at the mean level of Identity Anxiety.
![]()
Figure 1. Interaction between active social media use and identity anxiety in predicting attitudes toward gender equality.
These findings indicate that the impact of Active Social Networking Site Use on Attitudes Toward Gender Equality is conditional on the level of Identity Anxiety. Specifically, active SNS engagement appeared beneficial for individuals with low identity-related distress, while it may have adverse effects among those with higher anxiety levels.
The second SEM moderation model, focusing on Passive Social Networking Site Use, also showed excellent model fit (see Table 4). Passive Social Networking Site Use alone was not a significant predictor of Attitudes Toward Gender Equality (β = 0.055, p = 0.210), nor were Identity Anxiety (β = ?0.048, p = 0.275) or Social Comparison (β = ?0.019, p = 0.675). However, the interaction between Passive Social Networking Site Use and Identity Anxiety was significant (β = ?0.101, z = ?2.281, p = 0.023), suggesting that Identity Anxiety also moderated this relationship. The interaction between Passive Social Networking Site Use and Social Comparison was not statistically significant (β = ?0.033, p = 0.455).
To further explore the significant interaction, simple slopes analyses were performed (see Figure 2). Among individuals with low Identity Anxiety (?1 SD), Passive Social Networking Site Use was positively associated with Attitudes Toward Gender Equality. Conversely, among individuals with high Identity Anxiety (+1 SD), Passive SNS Use was negatively associated with egalitarian attitudes. No significant association was found at the mean level of Identity Anxiety.
![]()
Figure 2. Interaction between passive social media use and identity anxiety in predicting attitudes toward gender equality.
Together, these results suggest that the association between Passive Social Networking Site Use and Attitudes Toward Gender Equality also depends on Identity Anxiety. While passive use may be beneficial for individuals with low anxiety, it may hinder egalitarian beliefs among those experiencing high identity anxiety.
4. Discussion
This study investigated how patterns of social networking site (SNS) use―namely, Active and Passive Social Networking Site Use―were associated with Attitudes Toward Gender Equality, and whether these relationships were moderated by Identity Anxiety and Social Comparison. Although neither active nor passive SNS use alone significantly predicted attitudes toward gender equality at the bivariate level, results from moderation analyses revealed that Identity Anxiety consistently moderated both relationships. Specifically, both active and passive social media use were positively associated with egalitarian gender attitudes among individuals with low identity anxiety, but this pattern reversed among those with high identity anxiety, such that increased social media use was associated with less egalitarian beliefs. These results suggest that the psychological context in which individuals engage with social media―particularly their level of identity-related anxiety―critically shapes how these platforms influence gender-related attitudes.
4.1. Interpretation in Light of Theories
Social Identity Theory (Tajfel & Turner, 1979) provides a valuable framework for understanding these results. The theory posits that individuals derive a sense of identity from their membership in social groups. In contexts where users perceive themselves as marginalized―such as rural-origin students navigating urban norms on Xiaohongshu―social media content may exacerbate concerns about status and belonging. This aligns with findings that individuals with high identity anxiety tend to interpret social content in ways that reinforce traditional or hierarchical attitudes (Cheek et al., 1994; Yu, 2024).
Social Comparison Theory (Festinger, 1954) suggests individuals evaluate themselves by comparing with others. SNSs like Xiaohongshu, with their curated, aspirational content, may intensify upward comparisons, which can lead to negative self-evaluations and reinforce conventional beliefs (Vogel et al., 2014; Verduyn et al., 2015). However, the current study did not find significant moderation effects for social comparison, suggesting that identity-related processes may be more salient in shaping gender attitudes than comparison orientation.
4.2. Role of Identity Anxiety
Identity Anxiety emerged as a key moderator in both models. This underscores its significance as a psychological vulnerability that amplifies the effects of social media exposure, a finding that aligns with recent work on digital stress (Wegmann et al., 2020). When individuals are confident in their social identity, they may interpret social media as empowering or informative, fostering egalitarian views (Duggan & Smith, 2013). In contrast, those with elevated identity anxiety may experience such content as threatening or exclusionary, leading to more conservative gender attitudes, consistent with research on defensive processing of identity-relevant information (Major et al., 2018). This pattern aligns with prior work showing identity threats can promote attitudinal rigidity (Cheek et al., 1994; Yu, 2024).
4.3. Passive vs. Active SNS Use
Although passive and active SNS use are often conceptualized as distinct―passive use linked to consumption and active use linked to creation―both interacted with identity anxiety similarly. This highlights that it is not merely the type of engagement but the psychological framing of that engagement that determines outcomes. Whether consuming or producing content, users high in identity anxiety appear to internalize social signals in ways that reinforce hierarchical views. (Toma & Hancock, 2013)
4.4. Limitations
Several limitations should be acknowledged. First, the cross-sectional design limits causal inference. Longitudinal or experimental studies are needed to establish directionality. Second, self-report measures may introduce social desirability or common method bias. Third, the sample was drawn from a single cultural context―contemporary China―and may not generalize to other populations or digital platforms. Finally, only linear moderation effects were tested; future work should examine more complex interaction patterns.
4.5. Practical Implications and Future Directions
Findings from this study have implications for education, digital literacy, and platform design. Initiatives aimed at reducing identity anxiety, such as community support programs, may help buffer users from the potentially polarizing effects of social media. Developers could consider content filters or personalized feedback tools that discourage excessive upward comparison. Future research should explore these interventions and also investigate gender identity development, intergroup dynamics, and cultural variation in shaping digital attitudes.
5. Conclusion
This study provides novel insights into how active and passive engagement with Xiaohongshu―one of China’s most popular social media platforms―relates to attitudes toward gender equality, contingent upon levels of identity anxiety. While no direct effects were found between SNS use and gender attitudes, identity anxiety consistently emerged as a significant moderator, reversing the direction of these associations depending on its level. These results align with Social Identity Theory and underscore the importance of psychological vulnerability―such as identity-related distress―in shaping social media’s impact.
Importantly, the findings extend prior literature by demonstrating that social media’s influence on social attitudes is not uniform but depends critically on the user’s internal context. While the curated content on Xiaohongshu can inspire or inform those with low identity anxiety, it may exacerbate insecurity or reinforce traditionalism among more anxious individuals. This highlights the dual role of SNS platforms as both empowering and alienating.
From a practical standpoint, these findings suggest that digital mental health interventions and more inclusive algorithmic designs may help reduce the negative consequences of identity anxiety. Future research should explore how these dynamics unfold across different life stages, subpopulations (e.g., migrants, youth), and platforms, ideally using longitudinal or experimental methods to establish causal directions.