Digital bridges or digital divides? Social media use and crosscultural interaction among students at a Chinese university

1 Introduction

Social media has expanded rapidly and is changing how young people engage in cross-cultural interaction. Zheng et al. (2024) demonstrate that online communication environments enhance interaction willingness and facilitate the expansion of cross-cultural social networks. Its role in cultural maintenance is particularly evident among Chinese students abroad, who often use WeChat to participate in discussions of domestic cultural topics (Kuang and Wu, 2019). Geographical boundaries are no longer as restrictive as they once were for communication across cultures. Previous research suggests that information cocoons may emerge through the interaction of algorithmic filtering and users’ selective exposure or cognitive inertia (Yu and Qu, 2020). Structural barriers to accessing particular platforms or content may further restrict exposure to culturally diverse information. In the present study, the information-cocoon item measured only respondents’ perceived limitation in exposure to cultural information and did not distinguish among these mechanisms.

Against this background, the present study examines social media use and cross-cultural interaction among students at Shandong University of Technology. The questionnaire focuses on social media use patterns, cross-cultural interaction behaviors and perceptions, and the difficulties encountered in online cross-cultural contexts. The remainder of the paper reviews the relevant literature, describes the research design, reports the quantitative findings, and discusses their implications and limitations.

2 Literature review

Research on social media and cross-cultural interaction can be broadly divided into three strands. These strands include theoretical explanations, empirical descriptions of social media use and cross-cultural interaction behaviors, and critical analyses of associated risks.

Acculturation theory and uses and gratifications theory are two major analytical frameworks for examining social media and cross-cultural interaction. Berry (1997, 2005) developed acculturation theory to explain psychological and behavioral changes following sustained cultural contact. Identity construction and conflict resolution in cross-cultural interactions were explained in the theory, which included the psychological adaptation process, behavioral strategy choices, and influencing factors after cultural contact. Katz et al. (1973) complemented acculturation theory by proposing the Uses and Gratifications (U&G) framework. Uses and gratifications theory analyzes media use from the perspective of audiences’ active needs. The argument that individuals actively use media to satisfy specific psychological and social needs, rather than passively receiving content, is widely acknowledged. In contemporary social media environments, young people are both recipients of information and active participants in content production and dissemination. Behaviors such as liking, commenting, and sharing are the most prevalent cross-cultural interaction behaviors, which can be regarded as important ways to meet cognitive, social, and self-identity needs.

Existing empirical studies have outlined the basic relationship between social media and cross-cultural interactions across different sample groups, including international students, urban youth, and users of transnational communities. These studies generally suggest that more intensive social media use is associated with more frequent cross-cultural contact, greater cultural openness, and higher cross-cultural awareness. Research on teenage YouTube uploaders showed that users expected audiences who were geographically and sociodemographically close and placed considerable value on online feedback, indicating that participatory media do not automatically produce broad cross-cultural exposure (Courtois et al., 2011). In China, Fan and Wu (2025) analyzed cross-cultural information-sharing videos posted by multilingual users on Bilibili and identified the roles of content characteristics, platform support, and interaction patterns in expanding information reach. Social media has become an important channel for everyday cross-cultural information exposure.

Although social media has significant advantages in promoting cross-cultural interactions, studies have also pointed out that the technical logic and cultural power structure of social media platforms may have negative impacts. A cross-national study of adolescents in 42 countries found that more frequent social media use was associated with greater involvement in cyberbullying (Craig et al., 2020). Shen (2024) discussed how social-media mechanisms can contribute to youth opinion polarization, suggesting that platform dynamics may intensify conflict and extreme expression. A content analysis of Chinese cultural content on ReelShort found that algorithmic recommendation may create a communication pattern characterized by weak geographical boundaries and strong interest-based clustering (Shi and Liang, 2024). Li (2020) found that Facebook and WeChat use were positively associated with bridging and bonding social capital among university students in Macao, while Kuang and Wu (2019) highlighted the role of WeChat in the cross-cultural adaptation of international students in China.

Taken together, previous studies have outlined the general relationship between social media use and young people's cross-cultural interaction. Building on this foundation, the present study focuses on university students and examines the relationships among social media use duration, platform types, cross-cultural interaction frequency, and perceived importance of cross-cultural interaction for personal development.

Accordingly, the following hypotheses were proposed: The study aims to provide context-specific quantitative evidence on the cross-cultural interaction behaviors and perceptions of students at one Chinese university, and to provide empirical references for cross-cultural education practices and social media platform optimization.

• H1a: Daily social media use duration is associated with students’ cross-cultural interaction frequency.
• H1b: The use of at least one overseas social media platform is associated with students’ cross-cultural interaction frequency.
• H2: Students’ cross-cultural interaction frequency is positively associated with their perceived importance of cross-cultural interaction for personal development.

3 Materials and methods

3.1 Participants

To efficiently obtain large-scale data, questionnaires were distributed online to college students at Shandong University of Technology via Tencent QQ from March to June 2025. The selection of Shandong University of Technology students as research participants was based on two considerations: (1) the university has actively promoted internationalization in recent years, with a growing number of exchange programs and international students in its International Education College, creating a multicultural campus environment that provides rich resources for cross-cultural interaction research; (2) as a comprehensive teaching-and-research university spanning multiple disciplines, it provides an appropriate institutional context for examining students’ social media use and cross-cultural interaction. However, the sample was drawn from one university and was not intended to represent all Chinese university students.

Participants completed only an anonymous online survey. No personally identifiable information, including names, student identification numbers, telephone numbers, or residential addresses, was collected. Only non-sensitive self-reported behavioral and attitudinal data of the participants were collected. All research procedures involving human participants complied with the ethical principles of the Declaration of Helsinki and its subsequent revisions. Given the anonymous and minimal-risk nature of the study, this questionnaire-based cross-sectional study received a formal Ethics Review Waiver from the Medical Ethics Committee of Shandong University of Technology before data collection in February 2025 (Waiver Reference Number: SDUT-ETH-WAIVER-2025-028).

On the opening page of the online survey platform, a standardized written informed notice was displayed. All respondents were required to read the notice before accessing the questionnaire. All respondents were informed about the research objectives, data-use procedures, guarantee of anonymity, voluntary nature of participation, and unconditional right to discontinue the survey at any time without negative consequences. The notice informed participants that voluntarily completing and submitting the questionnaire indicated their electronic informed consent. No participants under the age of 18 were recruited. To protect participant privacy, the data cleaning process de-identified all raw survey responses and only aggregated statistical results are presented in this manuscript.

3.2 Questionnaire and data collection

The questionnaires were distributed online through established student communication channels. Students accessed the online questionnaire by scanning a QR code. A total of 640 complete responses were retained for analysis. To minimize disruption to students’ regular schedules, the questionnaire was distributed during regular teaching weeks and avoided examination periods and major holidays.

3.2.1 Questionnaire development and pilot testing

Before the formal data collection, an initial version of the questionnaire was pilot-tested to evaluate item clarity, response-option design, and skip logic. Of the 104 initial responses, 48 were complete. The incomplete responses were primarily associated with problems in the questionnaire's skip logic. The questionnaire was subsequently revised, and a second pilot test produced 34 complete responses. Further adjustments were made to improve the clarity of the questions and the consistency of the response options before the final questionnaire was administered.

Because the questionnaire mainly consisted of demographic questions, single-item ordinal measures, and multiple-response items that assessed different aspects of social media use and cross-cultural interaction, the questionnaire was not treated as a single unidimensional psychometric scale. Therefore, an overall Cronbach's alpha and an overall factor analysis were not considered appropriate. Internal consistency was assessed only for the three conceptually related interaction-behavior items in Q8: liking, commenting, and sharing.

3.2.2 Questionnaire

The questionnaire consisted of three parts. The first part collected demographic information, including gender, grade, discipline, and hometown. Q4 collected respondents’ hometown information through an open-ended province/city response. During data preparation, the reported hometowns were recoded into three descriptive categories: first-tier cities, second-tier cities, and third-tier cities and below. This recoded variable was used to describe the regional composition of the sample and was not included in the hypothesis testing. The second and third parts examined social media use habits and cross-cultural interaction behaviors and perceptions, respectively. The specific questionnaire items are presented in Tables 1–3. Regional background was retained as a descriptive characteristic because previous research has reported regional variation in social media use among Chinese university students (Wang et al., 2025).

Table 1. Demographic information items.

Table 2. Social media usage habits items.

Table 3. Cross-cultural interaction behaviors and perceptions.

The section on social media usage habits included questions such as commonly used platforms, duration of usage, and purposes and frequency of interactive behaviors. The questionnaire used a five-point Likert scale dividing the frequency of interactive behaviors such as liking and commenting into five levels, with 1 indicating ‘never participate’ and 5 indicating ‘participate daily.’ The specific questions are shown in Table 2.

In the section on cross-cultural interaction, cross-cultural interaction was operationalized as respondents’ self-reported contact or communication on social media with people whom they regarded as having a different cultural background during the previous six months. Q11 measured the frequency of such interaction on a five-point ordinal scale ranging from 1 (never) to 5 (daily) and served as the primary cross-cultural interaction-frequency variable. Q12–Q14 separately recorded the reported geographic origins of interaction partners, interaction content, and interaction channels. These items were analyzed descriptively rather than combined into a composite score because their multiple-response format did not measure the frequency, duration, or depth of each form of engagement and provided no validated basis for differential weighting. Q8 measured general liking, commenting, and sharing behaviors on social media and was not interpreted as a measure of cross-cultural interaction. Geographic origin in Q12 was treated only as a descriptive characteristic and not as a proxy for cultural similarity or cultural distance.

In Q15, “information cocoon” was presented as limited exposure to cultural information and was treated as a self-reported difficulty rather than as a direct measure of algorithmic filtering. The item did not distinguish whether restricted exposure resulted from platform algorithms, users’ selective choices, or structural barriers to accessing particular platforms or content.

3.2.3 Data analysis tools

Statistical Package for the Social Sciences (SPSS), version 24.0, and Microsoft Excel were used for data analysis. Descriptive statistics were calculated to summarize demographic characteristics, social media use patterns, and cross-cultural interaction responses. Cronbach's alpha was calculated for the three interaction-behavior items in Q8. A one-way analysis of variance was used to compare cross-cultural interaction frequency across the four daily social media use duration groups. Because the Q9 and Q11 scores were measured using five-point ordinal scales, a Kruskal–Wallis test was used to compare Q9 scores across the five interaction-frequency groups. When the omnibus Kruskal–Wallis test was statistically significant, Dunn's pairwise comparisons with Holm adjustment were conducted. A chi-square test was used to examine the association between overseas-platform use and cross-cultural interaction frequency. Spearman's rank correlation was used to examine associations among the ordinal variables. All statistical tests were two-tailed, with the significance level set at p < 0.05. Excel was used for data organization and visualization.

4 Results

Among the 640 valid respondents, 346 (54.1%) were female and 294 (45.9%) were male. The sample was predominantly composed of freshmen (n = 527, 82.3%), while students majoring in engineering constituted the largest disciplinary group (n = 288, 45.0%). Based on the recoded hometown information, 7 respondents (1.1%) were from first-tier cities, 107 (16.7%) were from second-tier cities, and 526 (82.2%) were from third-tier cities and below. The complete demographic characteristics are presented in Table 4.

Table 4. Respondents’ descriptive statistics.

4.1 Internal consistency of the interaction-behavior items

Internal consistency was evaluated for the three interaction-behavior items in Q8: liking, commenting, and sharing. The three items demonstrated satisfactory internal consistency, with a Cronbach's alpha of 0.797. An overall reliability coefficient and factor analysis were not calculated for the complete questionnaire because its demographic, single-item ordinal, and multiple-response questions assessed distinct constructs and were not intended to form a single psychometric scale.

4.2 Difference analysis

A one-way analysis of variance was conducted to compare cross-cultural interaction frequency across the four daily social media use duration groups (Table 5). The omnibus test indicated a statistically significant but small overall group difference, F(3, 636) = 3.395, p = 0.018, η² = 0.016. However, the group means did not show a consistently increasing pattern with longer daily use duration.

Table 5. Cross-cultural interaction frequency by daily social media use duration.

M, mean; SD, standard deviation; η², eta squared.

A chi-square test was conducted to examine the association between overseas-platform use and cross-cultural interaction frequency (Table 6). The distribution of cross-cultural interaction frequency differed significantly between respondents who used at least one overseas social media platform and those who did not, χ²(4) = 11.358, p = 0.023, Cramér's V = 0.133. Although the association was statistically significant, its magnitude was weak.

Table 6. Overseas-platform use and cross-cultural interaction frequency.

Pearson's χ²(4) = 11.358, p = 0.023, Cramér's V = 0.133. Each respondent was classified once according to whether they used at least one overseas social media platform.

Because cross-cultural interaction frequency and perceived importance were measured using five-point ordinal scales, a Kruskal–Wallis test was conducted to compare perceived importance across the five interaction-frequency groups (Table 7). The results showed a statistically significant overall group difference, H(4) = 96.220, p < 0.001. Dunn's pairwise comparisons with Holm adjustment showed that respondents who reported daily cross-cultural interaction had significantly higher perceived-importance scores than respondents in each of the other four interaction-frequency groups (all adjusted p < 0.001). No other pairwise comparison remained statistically significant after Holm adjustment (all adjusted p ≥ 0.075). These findings indicate an association between interaction frequency and perceived importance but do not establish a causal relationship.

Table 7. Q9 scores by cross-cultural interaction frequency.

M, mean; SD, standard deviation; IQR, interquartile range; H, Kruskal–Wallis test statistic. Post hoc comparisons were conducted using Dunn's test with Holm adjustment.

4.3 Correlation analysis

Spearman's rank correlation was used because daily social media use duration, cross-cultural interaction frequency, and the Q9 score were measured as ordinal variables. Daily social media use duration was not significantly correlated with cross-cultural interaction frequency (r_s = 0.054, p = 0.170). In contrast, cross-cultural interaction frequency was significantly and positively correlated with the Q9 score (r_s = 0.285, p < 0.001). The complete correlation results are presented in Table 8.

Table 8. Spearman correlations among the main ordinal variables.

All correlations are two-tailed. r_s, Spearman's rank correlation coefficient.

4.4 Hypothesis testing

The results provided different levels of support for the three hypotheses. Daily social media use duration was not significantly correlated with cross-cultural interaction frequency (r_s = 0.054, p = 0.170); therefore, H1a was not supported. The distribution of cross-cultural interaction frequency differed significantly between students who used at least one overseas social media platform and those who did not, although the association was weak [χ²(4) = 11.358, p = 0.023, Cramér's V = 0.133]; therefore, H1b was supported. Cross-cultural interaction frequency was significantly and positively correlated with the perceived importance of cross-cultural interaction for personal development (r_s = 0.285, p < 0.001); therefore, H2 was supported. These findings indicate statistical associations rather than causal effects.

5 Discussion

The findings reveal a contrast between respondents’ generally positive perceptions of cross-cultural interaction and their low reported use of overseas social media platforms. Respondents also reported language barriers, limited exposure to diverse cultural information, cultural misunderstandings, and online mistreatment attributed to cultural differences. Because the data are cross-sectional and self-reported, these findings indicate co-occurring patterns rather than causal relationships among platform access, interaction difficulties, and online mistreatment.

5.1 Main findings

The usage rate of domestic platforms is significantly higher than that of overseas platforms. WeChat was the most frequently reported platform in the sample, with a usage rate of 97.50%. Some students also used niche platforms such as Xiao Hei He, Zui You, and LOFTER.

Regarding average daily usage time, students showed high-intensity use of social media. As shown in Figure 1, nearly half of the students used social media for more than four hours a day, and only 4.38% used it for less than one hour a day.

Figure 2 shows that social chatting was the most frequently selected purpose of social media use (93.59%). Other commonly selected purposes included obtaining information, acquiring knowledge, and sharing daily life.

Figure 3 shows the participation frequency in three interactive behaviors: liking, commenting, and sharing. Overall, liking and sharing were reported more frequently than commenting, and “rarely” was the most commonly selected category for commenting.

Figure 4 shows the perceived importance of cross-cultural interaction for personal development. Over 70% of students recognized its value; only 3.28% considered it ‘highly unimportant’ and 0.63% ‘unimportant,’ while 17.03% rated its importance as ‘moderate.’

As shown in Figure 5, among the factors increasing the risk of cyberbullying, ‘lack of understanding and respect for different cultures’ had the highest proportion (87.03%), followed by ‘inadequate management and supervision of platforms’ (74.69%), ‘anonymity lowering the threshold for inappropriate remarks’ (73.91%), ‘insufficient cross-cultural communication skills’ (62.66%), and ‘insufficient social tolerance for multiculturalism’ (58.59%).

Figures 6, 7 present students’ cross-cultural interaction frequency during the previous six months and the reported geographic origins of their interaction partners, respectively. As shown in Figure 6, 57.34% of respondents reported at least occasional cross-cultural interaction (occasionally: 32.19%; frequently: 12.81%; daily: 12.34%), whereas 42.66% reported rarely or never interacting across cultures. In Q12, 37.34% of respondents selected no cross-cultural social interaction. Among the full sample, the reported geographic origins of interaction partners included Asia excluding China (53.28%), Europe (24.69%), the Americas (20.31%), Oceania (10.00%), and other regions (4.84%). Because Q12 permitted multiple selections, these percentages do not sum to 100%. These findings describe the geographic distribution of reported interaction partners only and should not be interpreted as evidence of cultural proximity or cultural similarity.

Figures 8, 9 provide descriptive information on the content and channels of cross-cultural interaction. Entertainment topics, cultural customs, and daily-life sharing were among the most frequently selected content categories, whereas academic discussion was selected less frequently. Comment sections and chat windows were the most frequently selected interaction channels. However, these categories indicate only what respondents discussed and which platform functions they used; they do not measure the frequency, duration, or depth of engagement within each channel. Therefore, the results cannot distinguish brief or low-intensity engagement from sustained interpersonal dialogue.

Figures 10, 11 present the difficulties encountered and the frequency of cyberbullying due to cultural differences. The most frequently reported difficulties were language barriers (46.09%), perceived limited exposure to cultural information described in Q15 as an information cocoon (45.63%), and misunderstandings arising from cultural differences (45.00%). Regarding perceived online mistreatment, 39.84% of respondents selected “never,” and 32.66% selected “rarely,” while the combined proportion selecting “occasionally,” “often,” or “frequently” was 27.50%.

These descriptive results do not establish that the reported interaction difficulties caused online mistreatment or reduced respondents’ willingness to participate in cross-cultural interaction.

5.2 Dialogue with the literature

The findings align with relevant domestic and international research. Regarding social-media participation, Courtois et al. (2011) showed that teenage YouTube uploaders’ audience expectations and feedback preferences shape their platform behavior, which is consistent with the importance of liking, commenting, and sharing observed in the present study. Shi and Liang (2024) emphasized cultural adaptation and platform strategies in transnational dissemination, while Li (2020) documented positive associations between Facebook and WeChat use and university students’ social capital. Regarding problems and challenges, Shen (2024) discussed the role of social-media mechanisms in youth opinion polarization, which is relevant to the information-cocoon difficulty reported by participants. Craig et al. (2020) documented a cross-national association between frequent social media use and cyberbullying involvement among adolescents; the present findings extend this concern to perceived online mistreatment attributed to cultural differences.

The information-cocoon result should be interpreted as respondents’ perceived limitation in exposure to culturally diverse information rather than as direct evidence of algorithmic filtering. Such limited exposure may arise from the interaction of three mechanisms: platform algorithms may prioritize content similar to users’ previous behavior, users may selectively engage with familiar topics or viewpoints, and structural access restrictions may reduce the range of available platforms and content. The present questionnaire did not measure the relative contribution of these mechanisms.

6 Conclusions

This study examined the relationships between social media use and cross-cultural interaction among 640 students at Shandong University of Technology. The results showed that daily social media use duration was not significantly correlated with cross-cultural interaction frequency. However, the distribution of cross-cultural interaction frequency differed significantly between students who used at least one overseas social media platform and those who did not, although the association was weak. Cross-cultural interaction frequency was also positively correlated with students’ perceived importance of cross-cultural interaction for personal development. Given the cross-sectional design of the study, these findings indicate statistical associations rather than causal effects.

6.1 Practical implications

The findings suggest several targeted practical implications. First, interaction partners from Asia excluding China were reported most frequently, accounting for 53.28% of the full sample. Universities could broaden the geographic range of students’ cross-cultural engagement by establishing structured online exchange programs with partner institutions in Europe, the Americas, Oceania, and other regions. These programs could include small-group cultural discussions, collaborative assignments, and short virtual exchange projects. The geographic categories should be used to broaden opportunities for contact rather than as indicators of cultural similarity or distance.

Second, although 79.06% of respondents perceived cross-cultural interaction as important or highly important for personal development, 42.66% reported rarely or never engaging in such interaction. To address this perception–behavior gap, universities could integrate regular and low-threshold cross-cultural activities into language, communication, and general education courses. Examples include international peer matching, moderated online discussion groups, collaborative cultural-comparison tasks, and participation credits for sustained engagement. Because overseas-platform use showed only a weak association with interaction frequency and access to some overseas platforms is restricted, these activities could be organized through accessible domestic platforms in cooperation with overseas partner institutions.

Third, language barriers were the most frequently reported difficulty, selected by 46.09% of respondents. Universities could provide targeted language support before and during online exchanges, including topic-specific vocabulary, conversation prompts, bilingual peer facilitation, and guidance on clarifying culturally ambiguous expressions. Social media platforms could support these efforts by improving contextual translation, offering prompts for respectful cross-cultural communication, and strengthening moderation of culturally prejudicial or hostile content. Because the present study was cross-sectional, these recommendations should be piloted and evaluated rather than interpreted as interventions whose effectiveness has already been established.

6.2 Limitations and future research

This study has several limitations. First, this study used a convenience sample from a single university, and freshmen accounted for 82.3% of the respondents. The findings therefore cannot be generalized to all Chinese university students or to students with a balanced grade distribution. Future studies should employ stratified multi-institutional sampling to obtain more balanced representation across universities and grade levels. Second, respondents’ platform choices are constrained because access to overseas platforms is severely restricted within China's domestic network environment. This structural factor should be considered when interpreting and generalizing the findings. Thirdly, the study was confined to behavioral indicators. The underlying motivations and psychological processes driving cross-cultural interaction remain underexplored. Future research would benefit from incorporating qualitative or mixed-method approaches. Fourth, the measurement of cross-cultural interaction relied on respondents’ own interpretation of “different cultural backgrounds.” The questionnaire did not provide a standardized definition based on nationality, language, cultural identity, or perceived cultural distance. It also did not distinguish brief or low-intensity engagement from sustained interpersonal dialogue in terms of frequency, duration, and depth. Moreover, Q12 recorded only the geographic origins of interaction partners, and geographic proximity cannot be assumed to represent cultural similarity. Future research should employ validated multi-item measures that separately assess cross-cultural exposure, active participation, direct dialogue, interaction duration, and perceived cultural distance before constructing a composite engagement score.

Fifth, Q15 measured perceived limited exposure to cultural information through a single multiple-response option and did not distinguish among algorithmic filtering, user-initiated selective exposure, and structural access barriers. Future studies should measure these mechanisms separately using validated multi-item scales and, where possible, behavioral data on content exposure and platform use.

Finally, the cross-sectional design does not permit conclusions about temporal order or causality. The reported relationships should therefore be interpreted as associations rather than causal effects. Future longitudinal or experimental studies are needed to examine the direction of these relationships.

Despite these limitations, the study provides useful quantitative evidence for universities designing cross-cultural education programs and for social media platforms seeking to improve their intercultural features.

Generative AI statement

The author(s) declared that generative AI was used in the creation of this manuscript. During the preparation of this manuscript, the authors used AI tools for the purposes of language editing and translation assistance. The authors have reviewed and edited all AI-assisted content and take full responsibility for the content of this publication.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher's note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fcomm.2026.1927459/full#supplementary-material

Demographic information (Q1–Q4), social media usage habits (Q5–Q8), and cross-cultural interaction behaviors and perceptions (Q9–Q16) of the survey questionnaire. Supplementary material at the source

Footnotes

Edited by: Ke Zhang, Shandong University, China

Reviewed by: Ozge Kirezli, Yeditepe University, Türkiye

Chao Lu, University of Science and Technology Beijing, China