A
Brand Communication of Chinese Universities on Social Media: Evidence from JiangsuAbstract
Overseas social media platforms have become crucial channels for universities to communicate their international brand and engage global audiences. Situated within the broader context of international communication, this study examines, this study examines the official accounts of seven highly internationalized universities in Jiangsu Province on X (formerly Twitter), Facebook, and Instagram. Posts were coded and analyzed across four dimensions—frequency, presence, topic and platform choice—and combined with user engagement metrics (likes, comments, shares) and sentiment analysis on the comments (positive, neutral and negative). Findings indicate that posting frequency is generally low, with topics concentrated on teaching and contextual life, while research- and organizational topics are underrepresented. Most posts are proprietary, and Instagram emerges as the preferred platform. User engagement is largely limited to likes, with comments and shares comparatively rare, and sentiment responses is predominantly neutral or positive, with contextual posts eliciting the most positive responses. The results suggest that Chinese universities could enhance international visibility and engagement by increasing posting frequency, diversifying content themes, and fostering more interactive communication strategies. This study provides empirical evidence and theoretical insights into the practices and effectiveness of Chinese universities’ brand communication in the global digital environment.
Keywords:
university brand building
international communication
social media
user engagement
sentiment analysis
A
A
I Introduction
In the context of intensifying global competition in higher education, international branding has emerged as a strategic imperative for universities. Higher education institutions (HEIs) are increasingly recognized not only as producers of knowledge but also as global actors competing for students, faculty, funding, and strategic partnerships. A strong institutional brand functions as a critical intangible asset, shaping academic reputation, ensuring financial sustainability, and extending global influence (Oplatka & Hemsley-Brown, 2021). In a landscape where rankings and reputation metrics carry considerable weight, branding offers universities a mechanism for differentiation and strategic positioning within the international education sector.
Social media has become central to this process, offering immediacy, interactivity, and global reach. Unlike traditional communication channels, social media enables two-way dialogue and the co-creation of brand meaning with stakeholders (Peruta & Shields, 2017; Pera et al., 2022). Platforms such as X (formerly Twitter), Facebook, and Instagram allow universities to amplify visibility, convey institutional values, and foster trust among diverse audiences. From a branding perspective, social media is not merely a promotional channel but a dynamic space where identity, reputation, and relationships are continuously constructed and contested. Nevertheless, research suggests that universities often rely on one-way promotional messaging, missing opportunities for meaningful engagement and the cultivation of long-term loyalty (Sobaih et al., 2022). These observations underscore the need to move beyond descriptive accounts of online activity toward a more nuanced view of how content, interaction, and affective responses collectively shape institutional brand identities.
For Chinese universities, leveraging international social media platforms is particularly critical. National initiatives such as the “Double First-Class” plan, the Belt and Road Initiative, and the broader “Going Global” strategy have intensified pressure on institutions to build recognizable brands and contribute to China’s soft power. Yet structural constraints—including language barriers, platform restrictions, cultural differences, and limited expertise in localized communication—often limit the effectiveness of global branding efforts. While many institutions maintain a robust domestic presence on platforms such as WeChat and Weibo, their international visibility remains uneven in both scope and quality. Consequently, despite their substantial contributions to research, innovation, and education, Chinese universities face the risk of underrepresentation in the global higher education brand landscape.
A
Existing scholarship on higher education branding has provided instructive information about strategies, stakeholder perceptions, and institutional outcomes. However, three key gaps persist. First, most studies focus on Western contexts, where branding practices are strongly influenced by market logics and ranking-driven imperatives (Capriotti & Zeler, 2023; Jiang et al., 2025), leaving non-Western experiences—particularly those of Chinese universities—underexamined. Second, prior research often adopts institutional or student-choice perspectives, paying insufficient attention to social media as both a communicative arena and an empirical lens for evaluating brand performance (Iram, 2019; Ladogina et al., 2020). Third, studies examining online branding frequently treat content strategies or engagement behaviors in isolation, with limited consideration of how content, interaction, and emotional responses dynamically interact to shape brand resonance (Eger & Gangur, 2024; Kim & Yang, 2017).Against this backdrop, the present study investigates how Chinese universities utilize international social media platforms to advance brand-building efforts. Focusing on seven highly internationalized universities in Jiangsu Province, it examines content strategies, user engagement, and emotional responses across X, Facebook, and Instagram. By integrating content analysis, engagement metrics, and sentiment analysis, the study seeks to illuminate the mechanisms through which institutional brands are constructed, negotiated, and evaluated in global digital spaces. In doing so, it addresses critical gaps in the literature while offering actionable insights for universities aiming to enhance their visibility, credibility, and resonance in the international higher education arena.
II. Literature Review
1. Posting on Social Media
International communication and branding in higher education increasingly depend on the strategic deployment of social media. Within the broader framework of higher education marketing, the design and management of communication content extend beyond operational concerns to constitute critical mechanisms for brand differentiation, stakeholder engagement, and the formation of institutional identity, as articulated through brand equity and reputation management frameworks. Accordingly, social media platforms have become essential channels for universities to disseminate information, convey institutional values, and engage with diverse international users.
Empirical evidence indicates that universities’ content strategies—including posting frequency, thematic focus, and presence—play a decisive role in shaping both user engagement and brand perception. Prior studies have consistently categorized university content into several functional areas, including teaching, research, social commitment, and organizational and contextual topics (Capriotti et al., 2023; Zeler et al., 2023). These categories reflect the multifaceted roles of higher education institutions, encompassing academic instruction, research production, societal engagement, governance transparency, and responsiveness to broader social, economic, and cultural environments. The relative emphasis and balance among these content types signal institutional priorities and shape the communicative positioning of each university, thereby influencing how stakeholders perceive and interact with the brand.
Despite these insights, research reveals notable disparities in practice. Content frequently prioritizes certain domains—such as teaching or contextual information—while underrepresenting others, particularly research outputs, resulting in a “functional misalignment” between social media outputs and institutional missions (Chapleo et al., 2011; Capriotti et al., 2023). This points out the need for systematic analysis of content strategies as mechanisms for enhancing brand visibility, stakeholder trust, and international engagement, especially in non-Western contexts where empirical evidence remains limited.
Platform-specific dynamics further shape content strategy. For example, Instagram, widely used by college students, is primarily employed for documentation, creative expression, social interaction, and staying informed about others, reflecting a range of motives that extend beyond simple communication (Sheldon & Bryant, 2016). Engagement patterns vary across platforms: in this study, Facebook generated significantly higher engagement in likes, comments, and shares compared to Instagram and LinkedIn, highlighting measurable differences in user interaction across social media channels (Mishnick & Wise, 2024). In terms of types of presence, most institutions rely heavily on original posts, reflecting a “centripetal” strategy that concentrates audience attention on the university’s own discourse system and reinforces user retention (Capriotti & Zeler, 2023), whereas sharing and reposting represent a “centrifugal” approach, situating the university within broader informational networks (Scolari, 2009).
These findings raise key questions: How do different content types—teaching, research, social commitment, organizational, and contextual—affect international user engagement and sentiment? To what extent do posting frequency and sourcing strategies influence stakeholder trust, visibility, and long-term brand equity? How do these practices differ between Western and non-Western contexts? Addressing these questions is critical for advancing theoretical understanding and informing strategic practice. In response, the present study systematically examines the content strategies, engagement patterns, and emotional responses of Chinese universities on Facebook, Instagram, and X, offering an integrated perspective on how higher education institutions leverage digital communication to enhance their global brand presence.
2. User Engagement
A
User engagement on social media—typically operationalized through metrics such as likes, comments, and shares—has been widely recognized as a critical indicator of user interaction and communication effectiveness. In the higher education sector, however, engagement extends beyond these quantitative measures, reflecting stakeholders’ emotional connection, cognitive evaluation, and willingness to identify with and advocate for a university’s brand. In this sense, user engagement functions not only as an outcome of communication but also as a key component of brand equity and reputation-building processes within universities.Building on this perspective, prior scholarship has sought to conceptualize the dimensions of online engagement. Lovejoy et al. (2012) classify engagement behaviors into three types: information-responsive (likes), relational (comments), and dissemination-oriented (shares). Liking constitutes a basic form of affective feedback, commenting indicates cognitive evaluation and attitudinal expression, and sharing reflects a higher level of identification and the willingness to actively disseminate content. Extending this, Kim and Yang (2017) propose a three-dimensional framework encompassing sensory, cognitive, and dissemination-related dimensions, highlighting how interactive behaviors are driven by the interplay between emotional connection and information processing. Collectively, these models underscore that engagement is not merely a behavioral metric but a mechanism linking communication practices to long-term brand relationships.
Empirical evidence further demonstrates that content type significantly shapes engagement outcomes. Saxton and Waters (2014) report that posts related to teaching and campus life are more likely to elicit likes, whereas socially responsible content tends to generate higher levels of comments and shares. Similarly, Capriotti et al. (2023) show that different content functions produce distinct engagement patterns, with teaching and contextual posts stimulating greater interaction than research- or administration-focused content. Such findings are particularly relevant for higher education institutions, where strategic content design is increasingly recognized as integral to shaping institutional identity, visibility, and stakeholder trust.
In the Chinese context, research on engagement has predominantly focused on domestic platforms such as Weibo and WeChat. While these studies confirm the professionalization and scale of local communication practices, little is known about how international audiences interact with Chinese universities on global platforms, including Facebook, X, and Instagram. Compared to Western institutions, many of which have developed systematic strategies linking engagement metrics to branding objectives, Chinese universities’ international presence remains limited in scope, frequency, and content diversity. This constrained interactivity reduces their capacity to strengthen brand resonance, enhance global reputation, and cultivate stakeholder loyalty in international higher education markets.
To address these challenges, scholars have recommended expanding international channels, establishing specialized communication teams, and adopting content strategies tailored to overseas audiences. Nonetheless, there remains a paucity of integrated research examining how content, engagement, and sentiment interact to shape institutional brand equity in the global higher education arena. This theoretical and empirical gap demonstrates the value of studies that move beyond descriptive accounts of online activity to explicate the communicative mechanisms through which user engagement contributes to brand trust and international reputation for universities.
3. Sentiment analysis
Sentiment analysis, initially developed in commercial brand reputation research, utilizes sentiment lexicons, machine learning, or deep learning models to automatically classify the emotional orientation of user-generated text. Advances in natural language processing (NLP) and affective computing have enhanced its utility, rendering it an increasingly valuable tool for examining users’ attitudes, evaluative feedback, and affective responses on social media platforms. In the context of higher education, social media communication extends beyond the dissemination of information and basic interaction metrics, encompassing the mechanisms through which users’ emotional reactions, attitudinal preferences, and perceptions of institutional brand identity are shaped. Studies exploring universities' social media practices have increasingly integrated sentiment analysis.
Several studies have combined content classification with emotion recognition to investigate how different topics of university posts elicit affective responses. For instance, Iram (2019) found that posts related to education and campus culture tend to generate more positive emotions, whereas topics concerning organizational management or official announcements are often perceived as neutral. These findings suggest that sentiment analysis provides an effective means of capturing international users’ emotional reactions and assessing their implications for institutional brand perception. Nevertheless, much of the existing literature remains largely descriptive, offering limited empirical exploration of the dynamic relationships among emotional feedback, user engagement, and the development of institution–user relationships.
Despite growing attention to social media content, engagement behaviors, and affective responses, several critical gaps persist. First, research remains heavily concentrated on English-speaking contexts (Capriotti et al., 2023; Eger & Gangur, 2024; Jiang et al., 2025), leaving systematic studies of Chinese universities’ communication practices on international platforms scarce. Second, prior studies predominantly rely on qualitative content analysis or basic descriptive statistics, with computational social science approaches, including sentiment analysis, remaining underutilized. Third, although previous research has examined relationships between post content, engagement behaviors (e.g., likes, comments, shares), and emotional responses individually, the dynamic interplay among content type, user participation, and sentiment remains insufficiently investigated. From the perspective of international communication and higher education branding, empirical evidence is particularly limited regarding how content strategies and multi-platform operations contribute to global brand recognition.
In response to these gaps, the present study focuses on seven highly internationalized universities in Jiangsu Province and their communication practices on major overseas social media platforms (X, Instagram, and Facebook). Specifically, the study aims to systematically document social media content, including posting frequency, types of presence, topics, and platform distribution; analyze international users’ engagement behaviors across topics, examining interaction patterns, user preferences, and the structural relationships underlying participation; and apply sentiment analysis to assess users’ emotional responses, capturing the distribution and distinctive characteristics of positive, neutral, and negative sentiments.
Drawing on the multi-dimensional concept of consumer engagement (Dessart et al., 2015), the study conceptualizes international users’ interactions with university social media content as a form of engagement encompassing cognitive, emotional, and behavioral components. Prior research on branded YouTube channels suggests that engagement comprises attention, interactivity, cognition, and emotion, collectively shaping user responses and influencing brand perception (Brodie et al., 2013; Chan et al., 2014). Applied to higher education, universities’ content strategies similarly shape how international audiences perceive institutional identity, respond emotionally, and participate through observable behaviors, including likes, shares, and comments.
Building on this perspective, the study adopts a content–behavior–emotion framework to systematically examine how social media content types, posting frequency, and sourcing practices influence both engagement outcomes and sentiment responses. The cognitive dimension captures users’ attention and comprehension of content; the behavioral dimension measures observable interactions, including likes, shares, and comments; and the emotional dimension assesses sentiment responses, encompassing positive, neutral, and negative affective reactions. Integrating these three dimensions provides a comprehensive understanding of how content strategies shape institutional brand identity, foster stakeholder trust, and enhance international visibility. In doing so, the study contributes to theoretical advancements in higher education brand management within non-Western contexts and offers practical guidance for universities seeking to optimize social media strategies and strengthen their global presence.
Accordingly, the study addresses the following research questions:
Q1: What is the current state of Chinese universities’ communication on international social media platforms in terms of posting frequency, type of presence, posting content, and platform preferences?
Q2: How do international users engage with topics posted by Chinese universities, and how does user engagement vary across topics?
Q3: What are the overall sentiment tendencies of international users’ comments on posts by Chinese universities, and how are positive, neutral, and negative sentiments distributed?
III Methodology
(1) Participants
The study focused on seven higher education institutions in Jiangsu Province that demonstrate comparatively high levels of internationalization: Nanjing University (NJU), Southeast University (SEU), Nanjing University of Aeronautics and Astronautics (NUAA), Soochow University (SUDA), Nanjing University of Technology (NJTECH), Jiangnan University (JNU), and China University of Mining and Technology (CUMT). These institutions were strategically selected as representative cases of international communication practices within a province distinguished by a well-developed higher education system. Their inclusion provides a heterogeneous yet comparable set of contexts, facilitating both cross-institutional comparisons and in-depth analyses of platform-specific strategies.
(2) Social Media
Data were collected from three major international social media platforms—X, Facebook, and Instagram—selected for their global reach, extensive user bases, and distinctive functionalities that facilitate university brand communication. X, with an estimated 590 million users, supports real-time information dissemination and interactive engagement (Kemp, 2024). Facebook, which reaches approximately 3.07 billion users, offers multimedia capabilities that enable diverse content expression and foster community interaction between universities and external audiences (Kemp, 2024). Instagram, a visually-oriented platform with around 2 billion users, is particularly well-suited for showcasing campus culture, students’ life, academic achievements, and alumni activities (Kemp, 2024). Together, these platforms constitute critical channels for international academic communication, cultural promotion, and brand building, forming the empirical basis for the present study’s examination of Chinese universities’ global social media practices.
(3) Data collection and analysis
Data collection for this study spanned a 12-month period, from 1 March 2024 to 28 February 2025. During this interval, the research team systematically retrieved all content posted by the seven selected universities on their official accounts across X, Facebook, and Instagram. Each social media homepage was accessed regularly, and every post was examined individually to record key information, including post type (proprietary, shared, or hybrid), posting date and time, and interaction metrics (likes, comments, and shares). All data were compiled into a structured Excel database and independently cross-verified by a second researcher to ensure accuracy and completeness. Posting frequency and consistency varied across institutions and platforms; to address potential bias, platform-specific analyses were applied where appropriate, and data completeness was confirmed for each account. This rigorous and systematic approach produced a robust dataset, providing a solid foundation for subsequent analyses of content categories, user engagement, and sentiment responses.
3.1.2 Content Categories
Following data collection, all posts were categorized into five thematic areas reflecting the functional roles of universities: teaching, research, social commitment, organizational, and contextual content, following the framework proposed by Capriotti et al. (2023). Building on this framework, the research team developed operational criteria for topic identification and classification (see Table 1), specifically adapted to the communication practices of Chinese universities on international platforms, thereby enhancing both the validity and precision of the classification system. Once these criteria were established, Python was employed to perform preliminary automated classification of the dataset. Subsequent statistical analyses examined the distribution of posts across universities, platforms, and content categories, generating quantitative structural distribution maps. To ensure methodological rigor and the reliability of results, all automated classifications were independently reviewed and, where necessary, corrected by two researchers, thereby guaranteeing consistency and accuracy in data annotation.
Topic | Criteria | Examples |
|---|---|---|
Teaching | information relating to training activity and the teaching-learning process | Teaching activities, evaluations of faculty, awards, teaching publications |
Research | information related to the research activity of the university | research paper, doctorates, scientific publications, research conference |
Social commitment | information related to the university’s sustainable action, as well as its social activity and its link with the community. | social activity, Raising the spirit of Lei Feng |
Organizational | information on the general running and governance of the university | New Profile Picture |
Contextual | information on general issues (social, economic, cultural, etc.) of the environment | Campus, spring comes |
3.1.3 Types of presence
Posts were further classified according to the degree of university involvement in content production into three types—proprietary, shared, and hybrid posts—following the framework proposed by Capriotti et al. (2023). Operational definitions were refined in light of the observed characteristics of the posts to enhance coding consistency and ensure comparability across institutions (see Table 2).
Type | Criteria | Examples |
|---|---|---|
proprietary posts | Content created and disseminated by the university on its profiles | What is it like to spend a day at Nanjing University? Robert, one of our international students, obviously has a lot to share. |
shared posts | Content from other users shared without additional information | Let’s take a look! https://nju.edu.cn/en/info/1026/15231.htm… |
hybrid posts | Content from other users shared with additional customized information | #Beyond the Classroom NJU students in winter holiday. |
3.1.4 Engagement of International users
User engagement was assessed across three dimensions—likes, shares, and comments—using a manual coding approach to systematically quantify interaction behaviors. To ensure measurement consistency, the distinctive interactive features of each platform were integrated with observed user behavior to develop clear operational definitions for each engagement dimension (see Table 3).
behavior | criteria | examples | |
|---|---|---|---|
like | Appears on the post without requiring verbal expression | 
| |
comment | User enters text or emojis under a postt | ||
share | Post appears both on the poster’s page and the user’s profile |  |
3.1.5 Sentiment Analysis of posting
In order to examine international users’ emotional responses, we conducted a systematic sentiment analysis of user comments following the methodology proposed by Wajdi Aljedaani et al. (2022).This approach facilitated the identification of overall attitudinal trends, categorizing feedback as positive, negative, or neutral. The Python library TextBlob was employed to classify comments into these sentiment categories, providing a structured and replicable framework for analyzing user affective responses. While TextBlob offers a lightweight and reproducible tool suitable for large-scale datasets, its ability to capture nuanced expressions, such as sarcasm or informal language, remains limited. Future research may consider leveraging more advanced models, including VADER or BERT-based classifiers, to enhance classification accuracy.
sentiment | text |
|---|---|
positive | Wow, very useful! |
netural | Hello, is there anyone from Nanjing tech university, need help |
negative | Are you in contact with anyone from that university? Maybe let them know you're getting scrutiny meant for them. There is no place in this world for animal abuse, please stand against such atrocious behavior. |
All data analyzed were publicly available from official university accounts, and no personal or private data were collected. Ethical considerations related to data privacy and consent were therefore deemed minimal.
IV Results and Discussion
4.1 Posting content on social media
4.1.1 Posting performance on social media by types of presence
During the one-year study period, the seven selected Chinese universities collectively published 295 posts across three major overseas social media platforms—X, Facebook, and Instagram—with substantial variation in institutional activity.
University | X | Facebook | Instagram | Total | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
proprietary posts | shared posts | Hybrid posts | proprietary posts | shared posts | Hybrid posts | proprietary posts | shared posts | Hybrid posts | ||
NJU | 30 | 1 | 15 | 0 | 0 | 0 | 0 | 0 | 0 | 46 |
SEU | 0 | 0 | 0 | 0 | 0 | 0 | 10 | 0 | 0 | 10 |
JNU | 4 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 4 |
NUAA | 0 | 0 | 0 | 43 | 0 | 0 | 55 | 0 | 0 | 98 |
CUMT | 0 | 0 | 0 | 0 | 0 | 0 | 91 | 0 | 0 | 91 |
SUDA | 0 | 0 | 0 | 18 | 0 | 0 | 18 | 0 | 0 | 36 |
NJTECH | 0 | 0 | 0 | 6 | 0 | 0 | 5 | 0 | 0 | 10 |
Total | 34 | 1 | 15 | 66 | 0 | 0 | 179 | 0 | 0 | 295 |
As shown in Table 5, NUAA and CUMT were the most active, contributing nearly 90 posts each, whereas NJU and SUDA averaged around 40 posts, and JNU, SEU, and NJTECH each produced fewer than 10 posts annually. Across all institutions, this figure corresponds to an average of approximately 42 posts per university per year, or roughly 3.5 posts per month, highlighting a relatively modest posting frequency and considerable heterogeneity in engagement levels. Regarding posting frequency, the seven Chinese universities in this study occupy a lower-middle position compared with internationally active institutions such as Yale, the University of Pennsylvania, and Columbia University, which post an average of 24–27 times per week (Fähnrich, 2020). This relatively low frequency may constrain continuous visibility and limit opportunities for sustained engagement. This aspect is primarily because international communication efforts are hindered by limited systematic planning, fragmented teams, and weak coordination mechanisms. Moreover, universities often adopt cautious strategies to mitigate the risks of cultural misinterpretation or politically sensitive content (Rauschnabel et al., 2016). As a result, institutional accounts frequently remain “silent,” which further weakens international presence and reduces brand exposure, thereby limiting the potential for cross-cultural engagement and influence (Hofstede, 2001).
Table 5 further disaggregates posting activity by type of presence—proprietary, shared, and hybrid—across the three platforms. Proprietary posts, representing original content created by the universities, constituted the majority of activity, particularly on Instagram and Facebook. For example, NUAA and CUMT relied heavily on proprietary posts (55 and 91 posts, respectively, on Instagram and Facebook), reflecting an emphasis on institutional control over messaging. In contrast, hybrid posts were more prominent on X for NJU (15 posts), indicating occasional integration of original and shared content. Shared posts were relatively rare across all platforms, suggesting limited engagement in content co-creation or dialogic interaction. Overall, this distribution indicates that while the universities maintain a visible social media presence, their activity is largely unidirectional and centered on controlled content dissemination rather than participatory engagement (Lovejoy & Saxton, 2012).
Several factors may account for this distribution. The predominance of proprietary posts reflects strategic efforts to manage institutional image and ensure messaging consistency, whereas the limited use of shared or hybrid posts may stem from fragmented communication structures, insufficient coordination mechanisms, and risk-averse approaches to culturally or politically sensitive content (Rauschnabel et al., 2016). Together, these structural and strategic considerations help explain why universities’ social media activity emphasizes controlled content dissemination over dialogic engagement.
4.1.2 Posting performance on social media by topics
During the one-year study period, the seven Chinese universities collectively published 295 posts on overseas social media platforms, exhibiting notable variation in content distribution across institutions. Table 6 presents the distribution of posts by topic across X, Facebook, and Instagram. Teaching-related content was predominant, accounting for 129 posts (43.73%), followed by contextual content with 94 posts (31.86%). Social commitment content represented 30 posts (10.17%), organizational content 23 posts (7.80%), and research-related content was the least frequent, with only 20 posts (6.78%).
Teaching | Research | Social commitment | Organizational | Contextual | Total | ||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
N | % | N | % | N | % | N | % | N | % | ||
X | 8 | 16.00% | 11 | 22.00% | 7 | 14.00% | 5 | 10.00% | 19 | 38.00% | 50 |
Instagram | 77 | 46.39% | 7 | 4.22% | 14 | 8.43% | 10 | 6.02% | 58 | 34.94% | 166 |
Facebook | 44 | 55.70% | 2 | 2.53% | 9 | 11.39% | 8 | 10.13% | 17 | 21.52% | 79 |
Total | 129 | 43.73% | 20 | 6.78% | 30 | 10.17% | 23 | 7.80% | 94 | 31.86% | 295 |
The predominance of teaching-related content reflects a strategic emphasis on highlighting educational achievements and projecting an internationalized teaching image. Drawing on Capriotti’s (2011) framework of organizational communication functions, such posts constitute “core functional communication,” serving to convey institutional strengths in curriculum design, faculty quality, and student mobility. Regular dissemination of teaching-related content not only reinforces brand recognition but also consolidates institutional credibility, positioning these universities as high-quality, reliable, and internationally oriented actors within the global higher education landscape.
Contextual content, accounting for nearly one-third of all posts, demonstrates an effort to situate institutional messaging within broader social, cultural, or political frameworks, including international festivals, cross-cultural greetings, and global events. Such contextualization enhances the relevance and resonance of communication, signaling both social responsibility and global engagement (Fähnrich et al., 2020).
By contrast, topic pertaining to social commitment and organizational management is markedly underrepresented, reflecting a relative deprioritization of the universities’ “third mission,” encompassing community engagement, sustainability initiatives, and institutional transparency. This imbalance suggests that current social media strategies favor controlled dissemination of institutional information and brand projection over diversified, interactive communication approaches (Kent & Taylor, 2002). Similarly, research-oriented content remains limited, likely due to its technical complexity, reliance on conventional dissemination channels, and insufficient adaptation into accessible, social media–friendly formats. Low levels of researcher participation and the absence of institutional guidelines further constrain the translation of academic output into digital communication practices.
Collectively, these patterns indicate that while the universities achieve visibility through teaching and contextual posts, the limited coverage of social commitment, organizational, and research-related content constrains thematic breadth and functional diversity. Expanding content diversity, adapting research outputs for social media, and integrating strategies that balance brand projection with broader institutional representation may enhance international engagement and strengthen global communicative impact.
4.2 User Engagement on social media
During the study period, international user engagement with the seven Chinese universities’ social media accounts totaled 16,973 likes, 1,007 shares, and 841 comments across 295 posts (Table 7). Subscriber numbers varied markedly, ranging from 68 for CUMT to over 201,000 for NUAA, reflecting substantial disparities in audience reach.
university | subscribers | post | share | comment | like |
|---|---|---|---|---|---|
NJU | 2938 | 46 | 40 | 187 | 34 |
SEU | 4290 | 10 | 0 | 7 | 513 |
JNU | 7029 | 91 | 532 | 123 | 10658 |
NUAA | 201530 | 98 | 310 | 402 | 4523 |
CUMT | 68 | 4 | 0 | 5 | 0 |
SUDA | 228 | 36 | 77 | 18 | 517 |
NJTECH | 7112 | 10 | 48 | 99 | 728 |
Total | 223195 | 295 | 1007 | 841 | 16973 |
Across all universities, “likes” overwhelmingly dominated user interactions, constituting the most frequent form of engagement, while shares and comments occurred far less frequently, often in single- or double-digit counts. This pattern indicates that international users’ engagement is largely affective rather than substantive. The predominance of likes—low-cognitive, high-frequency signals of emotional resonance—is consistent with engagement patterns observed on visually oriented platforms (Lovejoy et al., 2012). In contrast, shares, which require users to evaluate content as valuable and personally relevant before disseminating it (Kaplan & Haenlein, 2010), were limited, suggesting that content has not fully penetrated users’ broader social networks. Similarly, the consistently low frequency of comments indicates minimal cognitively demanding, dialogic engagement, highlighting an underutilization of social media’s interactive affordances (Kent & Taylor, 2002).
From a strategic perspective, these disparities reveal both challenges and opportunities. Large-scale accounts such as NUAA benefit from extensive reach but must optimize content to encourage higher-effort engagement, whereas smaller accounts could prioritize highly salient, visually engaging posts to maximize impact with limited resources. Overall, the interaction structure across universities reflects a predominance of low-effort, affective participation, underscoring the need for strategies that promote deeper cognitive engagement and relational development.
An analysis of user engagement by content type (Fig. 1) shows a clear hierarchical distribution. Teaching-related content generated 6,542 likes, 518 shares and 153 comments, ranking among the most engaged categories. Contextual content demonstrated the highest overall appeal, with 7,784 likes, 923 shares and 259 comments. In comparison, research-focused posts attracted only 179 likes, 50 shares and 16 comments, while organisational announcements received 1,037 likes, 33 shares and 61 comments. Social commitment content occupied an intermediate position, generating 1,295 likes, 24 shares and 38 comments. Overall, teaching and contextual posts consistently scored highest across likes, shares and comments—key indicators of affective engagement—whereas research and organisational content recorded the lowest interaction levels, with social commitment content falling between these extremes. These findings indicate that content directly associated with teaching, campus life and cultural context elicited stronger emotional responses and higher levels of sharing.
These patterns can be interpreted through the lens of content salience and cognitive accessibility. Teaching and contextual posts typically incorporate visually rich, emotionally resonant elements, aligning with the concept of “instant-likeable content” that promotes rapid, affect-driven responses (Sheldon & Bryant, 2016). Contextually embedded content also leverages culturally meaningful cues, enhancing cross-cultural resonance and the likelihood of sharing (Kaplan & Haenlein, 2010). In contrast, research and organizational content often involves technical language, specialized knowledge, and reduced visual appeal, limiting accessibility and constraining both affective and cognitive engagement (Fähnrich et al., 2020).
From a strategic standpoint, the observed engagement hierarchy highlights a tension between immediate visibility and long-term institutional impact. While teaching and contextual posts generate rapid user response and baseline visibility, research-oriented content is critical for reinforcing scholarly reputation and positioning universities within global knowledge networks. To address this gap, universities could translate research findings into visually engaging, interactive formats, employ storytelling techniques, or implement participatory initiatives that invite user co-creation. Such strategies can foster deeper cognitive engagement, broaden the reach of complex content, and balance affective participation with knowledge-based institutional communication.
Figure 1 User Engagement statistics by topics
4.3 International user sentiment on comments posted by users
Table 8 reports the distribution of sentiment across posts from seven universities on X, Instagram and Facebook. A total of 365 comments were analysed. Activity was overwhelmingly concentrated on Instagram, which accounted for 315 comments (229 neutral, 86 positive and 6 negative), representing approximately 86 per cent of the dataset. Facebook generated 44 comments (27 neutral and 17 positive), while no comments were recorded on X. On Instagram, neutral sentiment was most prevalent (71.34%), followed by positive sentiment (26.79%) and a small proportion of negative comments (1.87%). Facebook comments similarly reflected only neutral and positive sentiment, with no negative comments observed. At the institutional level, NUAA produced the largest volume of comments (149), followed by CUMT (119) and NJTECH (74), whereas SEU (7), SUDA (16) and especially NJU and JNU (no recorded comments) exhibited minimal engagement across all platforms. These results indicate a pronounced platform bias towards Instagram and a predominance of neutral and positive sentiment across the sampled posts.
university | X | Instagram | Facebook | Total | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
negative | neutral | positive | negative | netutral | positive | nagative | neutral | positive | ||
NJU | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
SEU | 0 | 0 | 0 | 0 | 7 | 0 | 0 | 0 | 0 | 7 |
JNU | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
NUAA | 0 | 0 | 0 | 1 | 82 | 35 | 0 | 16 | 15 | 149 |
CUMT | 0 | 0 | 0 | 0 | 79 | 40 | 0 | 0 | 0 | 119 |
SUDA | 0 | 0 | 0 | 0 | 12 | 4 | 0 | 0 | 0 | 16 |
NJTECH | 0 | 0 | 0 | 5 | 49 | 7 | 0 | 11 | 2 | 74 |
Total | 0 | 0 | 0 | 6 | 229 | 86 | 0 | 27 | 17 | 365 |
Several factors may account for this pattern. First, the dominance of Instagram suggests that universities prioritize visually oriented content and interactive engagement with international audiences, consistent with prior research highlighting platform-specific strategies in higher education communication (Eger & Gangur, 2024). Second, the predominance of neutral and positive sentiments likely reflects the universities’ intention to maintain a professional and favorable international image, avoiding controversial or potentially negative content. Third, institutional capacity and resource allocation may influence posting frequency and sentiment management, as observed in the marked differences between larger institutions (e.g., NUAA, CUMT) and smaller or less active ones (e.g., NJU, JNU).
In sum, the sentiment distribution underscores a strategic emphasis on neutral-to-positive messaging, concentrated on Instagram, which aligns with broader trends in university international communication practices.
Fig. 2
Sentiment on comments posted by users
Figure 2 depicts international users’ emotional reactions to Chinese universities’ social media posts, categorized by topic. Neutral responses predominated across teaching, research, social commitment, and organizational posts, whereas contextual content elicited the most positive engagement. Specifically, teaching posts received 95 neutral and 49 positive reactions, reflecting their informational emphasis, such as course introductions, awards, or teaching achievements. Research-related posts were minimally engaging, with only 1 neutral and 1 positive response. Social commitment content generated 13 neutral and 5 positive responses, and organizational posts accounted for 21 neutral and 9 positive reactions. Contextual posts—featuring cultural festivals, campus landscapes, and student life—received 124 neutral and 39 positive responses, indicating comparatively stronger affective resonance. Across all topics, negative reactions were negligible.
This pattern can be attributed to several factors. Posts focusing on teaching, research, social commitment, and organizational management are largely informational, eliciting low-arousal, high-information responses consistent with prior findings (Sheldon & Bryant, 2016; Lovejoy et al., 2012). In contrast, contextual content combines visual appeal, culturally salient cues, and relatable narratives, aligning with emotional contagion theory (Kaplan & Haenlein, 2010) and facilitating positive affective engagement. Moreover, the non-commercial and public-interest positioning of universities likely constrains strong emotional reactions for institutional posts, whereas posts depicting student life or cultural events provide accessible stimuli for international users’ engagement.
Overall, these results highlight a clear differentiation among content types: informational posts maintain a neutral tone, supporting professional credibility, whereas culturally and visually enriched contextual content demonstrates greater potential to elicit positive affect. Universities may therefore strategically leverage narrative-driven, visually compelling posts within culturally resonant frameworks to enhance international engagement and audience affinity.
5. Conclusion and implication
This study investigated the international communication practices of seven highly internationalized universities in Jiangsu Province across X, Facebook, and Instagram. The findings reveal a paradoxical landscape: although these institutions maintain a visible international presence, their overall activity levels remain modest. Posting frequency is relatively low, with content disproportionately focused on teaching and campus life, whereas research and governance—key pillars of institutional reputation—receive minimal attention. Instagram emerges as the dominant platform, underscoring the value of visual storytelling; however, user engagement remains largely superficial, manifesting primarily through likes rather than dialogic interactions. Sentiment feedback is generally neutral to positive, yet the paucity of substantive exchanges suggests that universities’ brand narratives are not fully resonating with international audiences.
These patterns carry important implications for higher education marketing and international branding. Universities should recalibrate their content strategies, moving beyond routine campus updates to translate research achievements and governance initiatives into accessible, story-driven formats. Engagement practices must also evolve from one-way dissemination toward dialogic interaction, leveraging platform-specific features such as polls, Q&A sessions, and user-generated content to foster deeper participation. Moreover, platform differentiation warrants strategic consideration: Instagram may serve as a channel for visual appeal, X for real-time academic dialogue, and Facebook for community-building. Coordinating these distinct affordances within an integrated, cross-platform approach could enhance institutional visibility, authenticity, and trust in the global digital environment.
This study is subject to certain limitations. Its focus on a single province constrains the generalizability of the findings, and reliance on engagement metrics and automated sentiment analysis captures only surface-level responses, potentially overlooking deeper cultural or contextual meanings. Additionally, the analysis does not account for factors such as language choice, time-zone alignment, or audience segmentation, all of which may significantly influence international reach and resonance.
Future research could build on this work through comparative, multi-level designs incorporating universities from diverse regions and tiers, thereby illuminating variations in global branding practices. Complementary qualitative approaches—such as interviews with communication managers and focus groups with international students—could provide richer insights into strategic intent and audience interpretation. Beyond empirical extensions, future studies should engage more directly with theoretical debates in international communication and higher education marketing, particularly concerning how digital media mediate the interplay between local cultural narratives and global market pressures. Advancing this agenda would not only refine understanding of universities’ digital branding practices but also contribute to broader discussions on how higher education institutions negotiate identity, legitimacy, and competitiveness within the mediatized global landscape.
A
Author Contribution
A
Yawen Han and Yanzhen Tan wrote the main manuscript text and contributed equally to this work, sharing first authorship. Yawen Han also conceptualized the study and secured funding. Kaixuan Li prepared Tables 1–9 and Figures 1–2. Kaixuan Li collected the data together with Yanzhen Tan, and both Yanzhen Tan and Kaixuan Li performed the data analysis. Dandan Liu (corresponding author) conceived and supervised the overall study, coordinated contributions from all co-authors, approved the final version of the manuscript, and handled all correspondence with the journal. All authors reviewed and approved the final manuscript.References
Aljedaani W, Saad E, Rustam F, de la Torre Díez I, Ashraf I (2022) Role of artificial intelligence for analysis of COVID-19 vaccination related tweets: Opportunities, challenges, and future trends. Mathematics 10(17):3199. https://doi.org/10.3390/math10173199
Brodie RJ, Hollebeek LD, Jurić B, Ilić A (2013) Consumer engagement in a virtual brand community: An exploratory analysis. J Bus Res 66(1):105–114. https://doi.org/10.1016/j.jbusres.2011.03.002
Capriotti P (2011) Communicating corporate social responsibility through the Internet and social media. Revista Latina de Comunicación Social 66:539–567. https://doi.org/10.4185/RLCS-66-2011-925-539-567
A
Capriotti P, Zeler I (2020) Information and dialogue on social media: The use of corporate Facebook pages for CSR communication. Corp Soc Responsib Environ Manag 27(1):29–36. https://doi.org/10.1002/csr.1714Capriotti P, Oliveira A, Carretón C (2023) A model for assessing the active presence of institutions on social media: Application to universities worldwide. J Mark High Educ 00:1–21
Chan TSK, Yim CK, Lam SSK (2014) Antecedents and consequences of customer engagement in online brand communities. J Mark Analytics 2(2):73–85. https://doi.org/10.1057/jma.2014.2
A
Chandrasekaran S, De SK (2021) Decoding user interaction dynamics on Facebook fan page: A sentiment mining approach. Global Bus Rev. https://doi.org/10.1177/0972150918825078Dessart L, Veloutsou C, Morgan-Thomas A (2015) Consumer engagement in online brand communities: A social media perspective. J Mark Manage 31(13–14):1503–1527. https://doi.org/10.1080/0267257X.2015.1130738
Eger L, Gangur M (2024) How universities communicate with the public via social media: A content analysis. Masaryk University
Fähnrich B, Vogelgesang J, Scharkow M (2020) Evaluating universities’ strategic online communication: How do Shanghai Ranking’s top 50 universities grow stakeholder engagement with Facebook posts? J Communication Manage 24(3):265–283. https://doi.org/10.1108/JCOM-06-2019-0090
Hofstede G (2001) Culture's consequences: Comparing values, behaviors, institutions, and organizations across nations, 2nd edn. Sage
Iram A (2019) Sentiment analysis of students’ Facebook posts. In I. S. Bajwa, F. Kamareddine, & A. Costa (Eds.), Intelligent Technologies and Applications: Proceedings of INTAP 2018 (Communications in Computer and Information Science, Vol. 932, pp. 86–97). Springer. https://doi.org/10.1007/978-981-13-6052-7_8
Jiang S, Tahat K, El Gody A, Al Mansoori A (2025) Employing social media in building academic brands of the UAE’s universities: A comparative analysis. Online J Communication Media Technol 15(1) Article e202505. https://doi.org/10.30935/ojcmt/15872
Kaplan AM, Haenlein M (2010) Users of the world, unite! The challenges and opportunities of social media. Bus Horiz 53(1):59–68. https://doi.org/10.1016/j.bushor.2009.09.003
Kent ML, Taylor M (2002) Toward a dialogic theory of public relations. Public Relations Rev 28(1):21–37. https://doi.org/10.1016/S0363-8111(02)00108-X
A
Kietzmann JH, Hermkens K, McCarthy IP, Silvestre BS (2011) Social media? Get serious! Understanding the functional building blocks of social media. Bus Horiz 54(3):241–251. https://doi.org/10.1016/j.bushor.2011.01.005Kim S, Yang H (2017) Like, comment, and share on Facebook: How each behavior differs in terms of motivation and psychological outcomes. Comput Hum Behav 66:276–285. https://doi.org/10.1016/j.chb.2016.11.004
Ladogina A, Samoylenko I, Golovina V, Razina N, Petushkova E (2020) Communication effectiveness in social networks of leading universities. IC Revista Científica de Información y Comunicación 17:319–344. https://icjournal-ojs.org/index.php/IC-Journal/article/view/532
A
Liu B (2012) Sentiment analysis and opinion mining (Synthesis Lectures on Human Language Technologies, Vol. 5(1), pp. 1–167). Morgan & Claypool. https://doi.org/10.2200/S00416ED1V01Y201204HLT016Lovejoy K, Waters RD, Saxton GD (2012) Engaging stakeholders through Twitter: How nonprofit organizations are getting more out of 140 characters or less. Public Relations Rev 38(2):313–318. https://doi.org/10.1016/j.pubrev.2012.01.005
A
Melewar TC, Skinner H (2020) Territorial brand management: Beer, authenticity, and sense of place. J Bus Res 116:680–689. https://doi.org/10.1016/j.jbusres.2018.03.038A
McAllister-Spooner SM, Kent ML (2009) Dialogic public relations and resource dependency: New Jersey community colleges as models for website effectiveness. Atl J Communication 17(4):220–239. https://doi.org/10.1080/15456870903210113Mishnick N, Wise D (2024) Social media engagement: An analysis of the impact of social media campaigns on Facebook, Instagram, and LinkedIn. Int J Technol Educ 7(3):535–549. https://doi.org/10.46328/ijte.699
Oplatka I, Hemsley-Brown J (2021) A systematic and updated review of the literature on higher education marketing 2005–2019. In: Sinuany-Stern Z (ed) Handbook of operations research and management science in higher education. Springer, pp 35–80
A
Pang B, Lee L (2008) Opinion mining and sentiment analysis. Found Trends® Inform Retr 2(1–2):1–135. https://doi.org/10.1561/1500000011Pera R, Viglia G, Furlan R (2022) Who am I? How compelling self-storytelling builds digital personal reputation. J Interact Mark 35:44–55. https://doi.org/10.1016/j.intmar.2015.11.002
Peruta A, Shields AB (2017) Social media in higher education: Understanding how colleges and universities use Facebook. J Mark High Educ 27(1):131–143. https://doi.org/10.1080/08841241.2016.1212451
Rauschnabel PA, Kammerlander N, Ivens BS (2016) Collaborative brand attacks in social media: Exploring the antecedents, characteristics, and consequences of a new form of brand crises. J Mark Theory Pract 24(4):381–410. https://doi.org/10.1080/10696679.2016.1205452
Saxton GD, Waters RD (2014) What do stakeholders like on Facebook? Examining public reactions to nonprofit organizations’ informational, promotional, and community-building messages. J Public Relations Res 26(3):280–299. https://doi.org/10.1080/1062726X.2014.908721
Sheldon P, Bryant K (2016) Instagram: Motives for its use and relationship to narcissism and contextual age. Comput Hum Behav 58:89–97. https://doi.org/10.1016/j.chb.2015.12.059
Scolari CA (2009) The sense of the interface: Applying semiotics to HCI research. Semiotica 177:1–27
Sobaih AEE, Moustafa MA, Ghandforoush P, Khan M (2022) Higher education in and after COVID-19: The impact of using social network applications for e-learning on students’ academic performance. Sustainability 14(9):5195. https://doi.org/10.3390/su14095195
A
Data Availability
The social media data analysed in this study were obtained from publicly available platforms. In accordance with the platforms’ terms of service, the raw data cannot be shared, but processed datasets are available from the corresponding author on reasonable request.
Ethics Statement
This study was reviewed and approved by the Ethics Committee of Southeast University. All data were collected from publicly available social media platforms. No direct human participation was involved.