Speaking the Right Digital Language: How Post Format and Communication Impact University Facebook Engagement

Abstract

Social media has become a vital communication tool for higher education institutions (HEIs) to reach larger targets, attract followers, and engage with diverse audiences. This study conducted a quantitative and qualitative analysis of 4148 Facebook posts from 16 public and private HEIs in the United Arab Emirates (UAE). The aim of the study is to evaluate users’ engagement through their reactions to various post characteristics, including format, language, and content type. The posts generated 177,022 emotes, 17,269 shares, and 8374 comments. The results showed that images are an efficient format for boosting interaction, whereas plain text posts did not generate high engagement. The English language was more conducive for generating shares, while Arabic-language posts generated more emotes and likes. The comparative analysis results showed that private HEIs are more active on their Facebook pages than public HEIs. The content analysis suggested that student-related posts generate the highest level of engagement, while announcements and faculty- and research-related posts drive the lowest levels of engagement. These results offer valuable insights into how HEIs can optimize their social media strategies to enhance user engagement.

1. Introduction

Social media has become part of daily life, and companies realize that social media platforms are crucial to attracting and retaining customers and understanding their behavior (Hanandeh et al., 2024). Today, more students are using new technologies and frequently employ digital resources to collect information (Peruta & Shields, 2017). HEIs especially expanded their digital strategies with the COVID-19 pandemic and the shift to the online mode (Graham et al., 2023; Kummitha et al., 2021; Papademetriou et al., 2022). In addition, students’ Facebook use, the time spent on this platform, and interactions have significantly increased (Raza et al., 2020). In the UAE in particular, the usage of social media is very high, with 79% of the population using Facebook (GMI, 2023). This widespread adoption makes Facebook a vital channel for UAE universities to connect with their communities and strengthen their institutional visibility. In fact, Facebook provides an interactive system for discussions and connections where users can comment, post, and share their activities (Ellison et al., 2013). Nowadays social media has become a crucial tool for HEIs to build strong images and communicate their brand information (Peruta & Shields, 2017). Based on statistics, young adults whose ages are between 18 and 29 represent more than 90% of users, which highlights the high level of reach among HEI-age audiences. (Maresova et al., 2020). Facebook represents one of the major social media platforms, with 2.9 billion active users. This platform enables HEIs to promote their culture and news and engage with their students, in addition to their alumni and other stakeholders. (Capriotti et al., 2023). During the COVID-19 pandemic, social media proved to be crucial for HEIs to maintain connections and inform students remotely, underscoring the value of social media as an official communication channel in academia. Despite the global expansion of social media use, few studies have analyzed social media platforms’ impact on user engagement in higher education, especially regarding interactions with Facebook posts (Fähnrich et al., 2020; Gharbi et al., 2023; Sörensen et al., 2023).

Although the amount of research related to social media use by universities has increased, several key gaps persist in our understanding of Facebook engagement in the context of higher education. A focused review of existing studies reveals three specific areas that need deeper investigation. A few studies focused on public universities, they but did not differentiate between the social media performance of public and private HEIs (Almeida & Morais, 2020; Halaweh et al., 2020). Prior research examined all universities together or studied the top-ranked institutions, without considering the type of university and how it might influence Facebook posts’ engagement (Capriotti et al., 2023). Public and private universities often have distinct missions, funding models, and audience demographics, which could lead to differences in how they use Facebook and how stakeholders respond. However, we lack detailed comparative analysis of users’ engagement and reaction (such as comments, emotes, and shares) between public and private UAE HEIs. This gap leaves open questions about whether public universities with broader public mandates engage their audience differently (or more effectively) than private institutions, which may have smaller communities but perhaps more marketing-driven social media strategies. While it is recognized that content format (text vs. media) can impact user interaction, existing research has not thoroughly examined the effectiveness of different Facebook post formats for universities in this region. Past studies, largely in Western contexts, suggest that visual posts outperform text-only posts in engagement (Sabate et al., 2014) and some have cataloged the types of content universities tend to post (Bularca et al., 2022). Yet, there is a paucity of research isolating how each format—status updates, image-based posts, video posts, or link shares—specifically drives engagement within Middle Eastern HEIs. It remains unclear whether the general findings about multimedia content hold true for universities in the UAE, where audience preferences and platform algorithms may exhibit unique patterns. By examining the engagement with different post formats, the present study addresses a gap in determining which forms of content delivery are most effective for university pages in this context. An additional gap concerns the role of language in social media engagement, given that universities in the UAE operate in a culturally diverse setting and normally communicate in both English and Arabic to reach both international and local audiences. Global studies on university Facebook pages typically assume a single language of communication (English is common, especially for internationally ranked universities (Maresova et al., 2020)). This leaves unanswered questions about engagement dynamics—for example, do posts in Arabic elicit a different volume or type of response compared to posts in English? Does using the audience’s native language (Arabic) foster greater sense of community and thus more reactions, or does English (as the lingua franca of many UAE universities) generate broader engagement due to its wider reach? The influence of language choice on metrics like comments, shares, and “love” or “angry” reactions has not been empirically studied in the UAE’s higher education social media, representing a clear research gap that this study will explore. By addressing these gaps—institutional differences, post format efficacy, and language effects—the study seeks to extend the existing literature and provide a more nuanced understanding of Facebook engagement in UAE higher education.

Research Objectives

In light of the above gaps, this study adopts a quantitative and qualitative content analysis to comprehensively examine Facebook engagement by UAE HEIs. The main objectives of the research are as follows:

• To compare engagement levels on Facebook between public and private HEIs in the UAE. This includes analyzing and contrasting key engagement metrics for public vs. private universities in order to detect the differences in audience interaction.
• To evaluate the effectiveness of different Facebook post formats, such as posts with text only, posts with images, video posts, and link-sharing posts, in terms of boosting engagement through emotes, comments, and shares.
• To examine the influence of language on user interaction and reactions. Specifically, the study will assess whether posts published in English versus those in Arabic result in different engagement outcomes.
• To qualitatively analyze the themes and messaging content of universities’ posts. Through content analysis of the posts’ textual content, the study will identify common themes and assess how these thematic elements correlate with engagement.

This paper provides theoretical and practical contributions, with implications for higher education institutions, social media practitioners, and higher education policymakers. Each of these contributions underscores the significance of the study in bridging academic knowledge and practical application. By systematically analyzing Facebook engagement in UAE HEIs through a mixed-methods lens, the research not only advances scholarly understandings of social media in the context of higher education but also provides actionable guidance to those responsible for managing and governing these platforms.

2. Literature Review

2.1. Theoretical Framework

This study is grounded in two complementary theoretical frameworks: Engagement Theory and Media Richness Theory. Engagement Theory posits that learners and users are more likely to participate actively in an online environment when the experience is meaningful, interactive, and socially engaging (Kearsley & Shneiderman, 1998). In the context of higher education institutions (HEIs) and social media, this theory suggests that posts fostering community involvement such as student achievements or events are more likely to generate user reactions, including likes, comments, and shares. Meanwhile, Media Richness Theory proposes that different communication media vary in their capacity to convey information effectively. Richer media such as videos and images are more effective at reducing ambiguity and enhancing understanding due to their ability to transmit visual and emotional cues (Daft & Lengel, 1983). Applying this theory, we hypothesize that Facebook posts using rich media formats (e.g., images and videos) will yield higher engagement levels compared to plain text or link-based posts. Together, these frameworks offer a robust lens for analyzing how post characteristics influence user interaction in a bilingual, multicultural context such as the UAE.

2.2. Facebook Content for Higher Education

Content analysis has become an essential method for understanding how higher education institutions engage with audiences on social media platforms. The rapid growth of social networks has changed how companies share and transmit content online (Gunn, 2018), where they use social media to increase their interaction with users by publishing different post formats and working to increase users’ engagement by combining text, photos, and videos (De Vries et al., 2012). HEIs must be proactive and strategic on social media to develop visual Facebook content and tell effective stories (Taecharungroj, 2017). Hence, HEIs use social media to generate content, share information, interact, and improve communication (Khobzi & Teimourpour, 2015). Taecharungroj (2017), for example, examined Facebook posts from universities in the United States and Thailand and categorized them into 12 types: curriculum, students, research, faculty, campus, alumni, events, products, announcements, industry, image and reputation, and others. The author built on (Chapleo et al., 2011) research which categorized HEIs based on their website communication and highlighted several key variables to feature on their websites: excellence in teaching, campus environment, research and management, innovation, corporate social responsibility, and international projection. Further studies support and expand these content typologies. An exploratory study conducted in the context of Poland, the Czech Republic, and Slovakia on top public universities categorized posts into 12 codes: Persons, Campus, Study, Update photo, Funny, Promotion, Students event, Ecology, Management, Offer service, Sports, Award, and Others. Its results show that posts about people’s stories who are connected to the university, entertainment, and jokes are important in communication via social media (Eger et al., 2021). Another piece of research in the context of 19 Indonesian universities that used Instagram accounts found that nonverbal information such as video posts had no effect on likes and comments. As per the post content analysis, the study identified 10 topics (information, product, achievement, coverage, holiday, entertainment, cooperation, charity, competition, and promotion). Only information and achievement had a positive effect on social media engagement (Wahid & Gunarto, 2022). Peruta and Shields (2018) analyzed 5932 Facebook posts from top US universities, and during their study period they identified 17 content categories. They found that some content categories are positively related to engagement, such as news-related, school spirit, athletics, admissions, and promotion posts, while other content, such as campus events, overall informative, research/scholarly/creative, student achievements, and academic events, can decrease engagement. All these studies highlight the importance of content strategy in higher education social media and the need for institutions to tailor their digital content to increase engagement.

2.3. User Reactions to Facebook Posts

Understanding how users engage with different Facebook posts is important for HEIs to analyze their social media strategies. In February 2016, Facebook introduced five new reactions (love, haha, wow, sad, and angry) to the original thumbs-up icon (Spottswood & Wohn, 2019), while the “cares” reaction was added in April 2020 (CNET, 2022). These “emote” reactions allow users to express their emotions without text through different icons and facial expressions. Reacting with icons involves less effort and emotional expressivity than commenting, which requires verbal expression (Burke & Kraut, 2016). Few studies have explored some of these reactions in HEI’s Facebook accounts, examined their impact on user engagement (Bélanger et al., 2014), or measured users’ emote responses (Sörensen et al., 2023; Spottswood & Wohn, 2019). Using likes suggests that people are reacting to posts in a minimal manner, while shares are considered as a more voluntary behavior to increase post visibility Comments, on the other hand, are considered as a two-way communication tool (Lappas et al., 2022).

Facebook content is highly flexible, allowing users to share links, photos, videos, and text (Ellison et al., 2015), which can enhance interaction among users (De Vries et al., 2012). Numerous studies confirm that posts containing visual content and featuring photos and videos generate more engagement than those with plain text only (Gharbi et al., 2023; Pletikosa Cvijikj & Michahelles, 2013; Sabate et al., 2014). In addition, visual content more effectively triggers younger student engagement than older population engagement (Delello & McWhorter, 2017). Valerio et al. (2015) analyzed Facebook posts from Mexican universities and found that the most frequently used formats were links, and the least-used formats were images. Social media images are a promising source of information (Landwehr & Carley, 2014), and are often used to attract attention and illustrate important ideas (Brindley et al., 2019). Images are also a significant indicator of relevant content and can be used to enhance awareness (Peters & De Albuquerque, 2015). However, some informational videos cause distress, which might discourage users from reacting with a like, share, or comment (Rus & Cameron, 2016). Other studies analyzed the effect of the type of post (link, photo, status, video), frequency of posts, and timing on engagement with private universities. Posts with photos generate the maximum amount of engagement (52%), followed by videos (30%), while posts containing links represent only 18%, and statuses less than 0% (Mehmood et al., 2022). In addition to format, the language used in posts has also been identified as a variable influencing online engagement (Cheung et al., 2020; Gharbi et al., 2023).

Although a variety of studies have assessed university social media strategies, most have focused on Western contexts and highly ranked institutions, often overlooking regional and institutional diversity. For example, prior research confirms that visual content generally garners more engagement than plain text (Pletikosa Cvijikj & Michahelles, 2013; Sabate et al., 2014). Yet few studies have explored whether these findings extend to the Middle East or account for differences between public and private HEIs. Furthermore, while global studies typically assume a single language of communication, usually English, this does not reflect the linguistic realities of countries like the UAE, where both Arabic and English are widely used in educational communication. Therefore, this study addresses key gaps by comparatively analyzing post format, content themes, and language use across public and private HEIs in Abu Dhabi, with a focus on their respective impacts on user engagement.

3. Methods

3.1. Context and Procedure

Using data from the Ministry of Higher Education’s official website for the period between 1 September 2019, and 31 August 2020, we identified 18 accredited and active higher education institutions (HEIs) in Abu Dhabi, categorized as either public or private. We then verified whether each institution maintained an active official Facebook account. Two institutions did not have active accounts and were therefore excluded, resulting in a final sample of 16 HEIs (5 public and 11 private). The official Facebook pages’ posts and their reactions were then analyzed. The measurement of Facebook post engagement was conducted using methodologies outlined in the studies of Valerio et al. (2015) and Alsufyan and Aloud (2017). Facebook post engagement was measured in terms of likes, comments, and shares. This study also measured “emote” reactions (as the sum of like, care, wow, haha, love, sad, and angry reactions) in addition to comments and shares. From this point forward, “engagement” and “reaction” are used interchangeably.

3.2. Sample and Data Collection and Data Analysis

To construct a comprehensive and context-specific sample, we identified 18 accredited higher education institutions (HEIs) in Abu Dhabi from the Ministry of Higher Education’s official records for the 2019–2020 academic year. Out of the 18, only 16 maintained active and regularly updated official Facebook pages during the study period, resulting in our final sample. The chosen timeframe, from 1 September 2019, to 31 August 2020, spans the pre-pandemic and early COVID-19 phases. This period was selected intentionally to observe how HEIs adapted their digital engagement strategies amid unprecedented shifts to remote learning and online communication. Research has shown that this era marked a substantial increase in institutional reliance on digital platforms, thus providing a rich context for evaluating social media engagement.

A Python software (Python 3.8.10) data-scraping method was used to retrieve all published Facebook posts for the 16 HEIs and their primary characteristics, since Python offers the most appropriate language for building scrapers that can hop from one domain to another (Thomas & Mathur, 2019). To retrieve Facebook page URLs, a customized Python code was used due to its suitability for building scrapers. This code inspected the URLs to collect HTML data, which was then parsed and stored in a relational MySQL database. The collected data included post ID, date and time of creation, links, and content, in addition to post format (plain text, image, video, or link), HEI type (public or private), language (English, Arabic, or English and Arabic), and user reactions (comments, shares, and emotes [the sum of like, care, wow, haha, love, sad, and angry reactions]). Finally, the data was exported to Microsoft Excel and subsequently imported into SPSS 29 for analysis. The scraping process started on 1st of February 2021 and lasted 10 days. The posts that were extracted were published by the 16 HEIs during the period from 1 September 2019, to 31 August 2020, which represented the most recent academic year. A total of 4148 Arabic- and/or English-language posts were published during the data collection period. This study was part of a project that covers other aspects and other social media platforms, for which the HEIs’ Facebook post content was analyzed with a combination of qualitative and quantitative methods. All posts that were created by the page administration during the timeframe were included. This contains text posts, posts with images and videos, and links using the Arabic and English languages. The main objective of this study is to find the main themes, along with the post format and the language, that generate the highest reaction regardless of the audience of the Facebook post, as follows:

Quantitative methods were adopted and descriptive and inferential approaches in data analysis were used to provide a comprehensive understanding. This allowed the study to gain insights into the underlying patterns and relationships within the data and have comparative views. The descriptive analysis summarizes the posts’ characteristics in terms of reactions, language, and format. The inferential analysis used an independent t-test to compare the reactions between private and public HEIs. Furthermore, the chi-squared test was used to determine the significant associations between post reactions (emotes, comments, and shares) and post format (text, images, videos, or links). Users’ reactions to HEIs’ Facebook posts (emotes, comments, and shares) were analyzed. In addition, a comparative approach was employed to assess how the language used in posts (Arabic, English, or both Arabic and English) affected users’ reactions (emotes, comments, and shares).

Qualitative methods have also been applied; all collected posts were compiled into a spreadsheet and categorized by university. This indexing made it easier to classify and trace the data during the analysis. The prepared dataset served as a database for coding. By having all posts stored by university name, this allowed us to work through content more easily and to look at posts from different angles, including university name, post theme, time, and other patterns. The core of the qualitative content analysis was to transform the post content into meaningful categories. The researchers carefully read each post and assigned it a code/theme. Whenever the post did not fit an existing code, a new subtheme was created; all emerging codes were added to the codebook.

As the study of social media content analysis in the UAE is a new topic, the researchers relied initially on the study of Taecharungroj (2017), who identified 12 types of content. As per the present study, some categories were retained, others adapted to the sample, and new themes were added. The researchers reviewed the codes to merge the duplicates; any two codes having a similar meaning or overlapping were consolidated into one broader code, and definitions were developed to reduce the ambiguity and to maintain the consistency of all posts. To strengthen the reliability of manual coding, the four researchers performed self-checks by coding a random sample of 200 posts to ensure that the same codes were applied. Researchers reviewed the coded posts and discussed any discrepancies. This discussion between researchers helped ensure that the coding scheme was logical and consistently applied. Any disagreements or confusions were solved by referring to the code definition. Once coding was completed, a list was made for all codes/themes and subthemes. The result of this coding process was the creation of a set of 22 subthemes covering the range of topics found in the Facebook posts for the 16 universities. Later, subthemes that covered similar subjects or reflected different components of a larger concept were clustered together. For example, separate codes like “promoting a new program/course” and “highlighting online learning offerings” were both related to educational content; these were grouped under a larger education theme. Each cluster of subthemes was given a key theme name that captured the common link among those subthemes. At this stage, six main themes emerged. The six themes represented broad content categories such as announcements, educational content, events, research/faculty focus, reputation building, and student-centric posts.

3.3. Theme Definitions

With the six main themes identified, each was defined clearly to explain its meaning in the context of UAE higher education Facebook communication. Below are the six themes and their definitions, along with examples of the subthemes they encompass:

• Announcements: This theme covers official informational posts, including administrative messages such as messages from the university president or policy updates, besides messages related to student recruitment or admissions notices.
• Education: This theme covers academic offerings and learning opportunities and includes content such as promotion of academic programs or courses, information about new curricula, and encouragement of online learning or e-learning platforms.
• Events: This theme covers all posts related to events and activities reflecting campus life and engagement. It includes academic events such as conferences, seminars, workshops, guest lectures, and webinars hosted by the university, as well as non-academic or social events such as sports tournaments, cultural festivals, student club activities, graduation ceremonies, and national celebrations like UAE National Day.
• Faculty and Research: This theme covers all posts related to university faculty members and research output. The posts include faculty achievements, such as receiving an award or being featured in media, along with posts related to research activities such as studies conducted by the university, publications, innovations, or research collaborations.
• Reputation and Social Responsibility: This theme covers all content aimed at building and maintaining the university’s public image and societal role. It includes posts related to campus facilities and resources, such as opening a new building or lab; in addition, it covers institutional achievements such as rankings, accreditations, and awards earned by the university, and also includes alumni success stories and partnerships or industry relations.

4. Results

4.1. HEI Facebook Post Characteristics

Table 1. The characteristics of HEIs’ Facebook posts.

	Posts’ Reactions
Variable		N	Percentage	Min–Max
Emotes		177,022	87%	0–17,696
	Likes	154,638	87.4%	0–14,000
	Loves	19,060	10.8%	0–3600
	Wow	660	0.4%	0–96
	Cares	803	0.5%	0–129
	Sad	368	0.2%	0–106
	Angry	13	0.0%	0–4
	Haha	1480	0.8%	0–386
Shares		17,269	9%	0–2000
Comments		8374	4%	0–943
	Posts’ Format
Plain text		25	0.5%
Post including link		1305	23.9%
Post including video		913	16.7%
Post including image		3208	58.9%
	Public (PU) and Private (PV) HEIs
Private HEIs posts		2901	69.9%
Public HEIs posts		1247	30.1%
	Posts’ Language
English		1494	36%
Arabic		827	20%
English and Arabic		1825	44%

shows the post characteristics’ descriptive statistics. The highest percentage of reactions (N = 177,022; 87%) were in the emote category; shares represented 9% and comments 4%. “Like” was the most used emote (87.4%) across all HEIs. The largest percentage of HEI posts (58.9%) contained images, followed by links (23.9%), videos (16.7%), and text-only posts (0.5%). The majority of posts used both the Arabic and English languages (44%), while, on the other hand, 36% used only English and 20% used only Arabic.

4.2. Public (PU) and Private (PV) HEIs’ Post Reactions

All types of reactions (comments, emotes, and shares) were higher for private HEIs. Specifically, 76.97% of reactions were emotes for private HEIs, while 10.37% were emotes for public HEIs. The percentage of comments was low for both private and public HEIs (3.58% and 0.56%, respectively). Shares were also higher for private (7.65%) compared to public HEIs (0.87%). This result could be explained by the higher number of private compared to public HEIs. To address this concern, we performed an independent t-test to detect differences in the mean reactions per post.

Table 2. Independent t-test: Reaction differences between PV and PU HEIs.

	Emotes					Comments					Shares
	Mean	SD	T	Df	Sig	Mean	SD	T	Df	Sig	Mean	SD	t	Df	Sig
PV	53.77	366.66	4.86	3940.08	0.00	2.50	21.47	3.81	3433.29	0.00	5.34	50.03	4.18	3027.47	0.00
PU	16.86	118.59				0.90	4.47				1.42	4.90

shows that there were significant differences in reaction types between private and public HEIs.

A chi-squared test was performed to examine the association between reaction type (emotes, comments, and shares) and HEI type. It showed no significant association between HEI type and emote reactions, but significant associations between HEI type and comments, and HEI type and shares (see

Table 3. Chi-square: The association between posts’ reactions and the type of HEI.

	Emotes			Comments			Shares
	Sig	Did Not Receive Emotes	Received Emotes	Sig	Did not Receive Comments	Received Comments	Sig	Not Shared	Shared
PV	0.105	3.4%	96.6%	p < 0.001	63.3%	36.7%	p < 0.001	40.0%	60.0%
PU		2.5%	97.5%		71.1%	28.9%		54.1%	45.9%

).

Private HEIs showed significantly higher engagement across all types of Facebook reactions. Specifically, private institutions received an average of 53.77 emotes per post compared to 16.86 for public institutions (p < 0.001). This nearly threefold difference underscores the greater digital interaction that private HEIs achieve, likely due to more proactive and targeted communication strategies. Similarly, comment and share rates were significantly higher for private HEIs, indicating not only more frequent user reactions but also deeper forms of engagement, such as dialogue and content amplification. While statistical significance was observed, it is important to contextualize these differences as substantial and practically meaningful, not just numerically distinct.

4.3. Public (PU) and Private (PV) HEI Reactions Based on Post Language

Public HEIs’ preferred language was both English and Arabic at the same time (83%), followed by Arabic (12%) and English (5%). Private HEIs showed a slight preference for English over English and Arabic (48% and 43%, respectively), and a low preference for Arabic (9%).

The highest percentage of reactions were to posts written in English (emotes 71%, shares 69%, comments 66%). Reactions to posts written in English and Arabic were relatively low by comparison (26%, 25%, and 23% for shares, comments, and emotes, respectively). Arabic posts had the fewest reactions, ranging between 5% and 9%.

Private HEIs showed a higher percentage of emotes for posts written in English (70.91%), with public HEIs showing a low percentage of emotes for English posts (0.18%). Emote reactions to private HEIs’ English and Arabic posts and Arabic posts were low, at 13.19% and 4.02%, respectively. Emote reactions to public HEI posts in English only and Arabic only were 10.30% and 1.4%, respectively.

We originally divided language into three categories (English, Arabic, and English and Arabic). However, to conduct independent t-tests, it was necessary to convert the categories into two dummy variables: English/no English and Arabic/no Arabic.

As shown in

Table 4. Independent t-test: The differences in HEI post reactions based on the used language.

	Emotes					Comments					Shares
	Mean	SD	T	Df	Sig	Mean	SD	T	Df	Sig	Mean	SD	T	Df	Sig
No English	31.73	158.49	−1.12	4146.0	0.26	1.43	5.64	−1.05	4146.0	0.30	2.34	7.14	−2.68	3832.7	0.01
English	45.41	341.82				2.17	20.07				4.62	46.77
No Arabic	30.30	100.50	−2.45	3249.6	0.01	1.49	12.21	−1.60	4137.9	0.11	3.02	13.93	−1.69	3290.5	0.09
Arabic	49.65	384.96				2.31	20.73				4.81	51.40

, there were no significant differences in emote or comment reactions between English and no-English posts. However, share reactions showed a statistically significant difference in English versus no-English posts, where posts published in English generated a higher number of shares.

There were no significant differences in comment or share reactions between Arabic and no-Arabic posts. However, there was a statistically significant difference in the proportion of emotes for Arabic versus no-Arabic posts, where Arabic posts generated a higher proportion of emotes (see Table 4). A chi-squared test was performed to examine the association between reaction type (emotes, comments, and shares) and post language. The results showed a significant association between posts in English and emotes, with 97.1% of English text posts generating emotes and 95.7% of no-English text generating emotes. There was also a significant association between comment reactions and English text, with 35.1% of English posts generating comments and 31.4% of no-English posts generating comments (see

Table 5. Chi-square: The association between posts’ reactions and post language.

	Emotes			Comments			Shares
	Sig	Not Used	Used	Sig	Not Used	Used	Sig	Not Used	Used
No English	0.029	4.3%	95.7%	0.044	68.6%	31.4%	0.177	46.3%	53.7%
English		2.9%	97.1%		64.9%	35.1%		43.7%	56.3%
No Arabic	0.039	2.4%	97.6%	0.688	66.1%	33.9%	0.438	43.4%	56.6%
Arabic		3.6%	96.4%		65.4%	34.6%		44.7%	55.3%

).

A significant association was found between the percentage of emotes for Arabic posts (96.4%) versus no-Arabic posts (97.6%). There was no significant association between share reactions and using English, as well as between share reactions and using Arabic. Similarly, there was no significant association between comment reactions and using Arabic.

4.4. Public (PU) and Private (PV) HEI Reactions Based on Post Format

Post formats included text, link, video, or image. Posts containing images generated the highest percentage of comments, emotes, and shares (59.4%, 51.1%, and 50.9%, respectively), followed by posts containing videos (23.5%, 22.6%, and 22.1% of shares, comments, and emotes, respectively). Posts containing links generated 26.1%, 17.5%, and 16.1% emote, comment, and share reactions, respectively. Text posts generated a low percentage of reactions (below 1%).

Private HEIs showed a higher use of all formats than public HEIs, except in text format use. Image use was 34.2% in PV and 16.9% in PU; reactions to link posts were 20.6% in PV and 5.5% in PU; reactions to video posts were 14.7% in PV and 7.3% in PU; and reactions to text posts were 0.27% in PV and 0.34% in PU. There were also significant differences in PV and PU HEIs’ reactions to different post formats. Comment reactions to image-format posts were higher in PV than in PU HEIs (51% and 8.3%, respectively), followed by share reactions (45.7% and 5.2%, respectively) and emotes (34% and 17%, respectively).

Reactions to link-format posts were higher in PV than PU HEIs (emotes 20.6% in PV versus 5.5% in PU; comments 16% in PV versus 1.5% in PU; and shares 15% in PV versus 0.8% in PU). However, reactions to video-format posts showed a higher percentage in PV versus PU HEIs for share reactions (28.5% and 3.9%, respectively), comments (19.1% and 3.5%, respectively), and emotes (14.7% and 7.3%, respectively). Unexpectedly, text-format posts showed a similar low reaction in both PU and PV HEIs (0.34%).

We originally used four categories for post formats (text, link, video, and image). However, it was necessary to convert the formats into three dummy variables to conduct independent t-tests: image/no image, video/no video, and link/no link.

The independent t-test results showed no significant interactions between emotes, comments, and shares by the post format dummy variables (see

Table 6. Independent t-test: The difference in posts’ reactions based on posts’ format.

	Emotes					Comments					Shares
	Mean	SD	T	Df	Sig	Mean	SD	T	Df	Sig	Mean	SD	T	Df	Sig
No image	35.54	148.32	−0.79	4146.00	0.43	1.58	7.96	−0.84	4146.00	0.40	5.12	67.39	0.80	4146.00	0.43
Image	44.77	347.77				2.15	20.16				3.88	30.77
No video	44.69	346.23	0.78	4146.00	0.44	2.13	20.08	0.77	4146.00	0.44	3.87	30.64	−0.86	4146.00	0.39
Video	35.54	151.18				1.61	8.09				5.21	68.39
No link	38.86	143.60	−0.83	1396.83	0.41	2.08	20.83	0.34	4146.00	0.73	4.10	41.30	−0.15	4146.00	0.88
Link	51.00	517.97				1.88	9.98				4.30	43.38

).

A chi-squared test was performed to examine the association between post reactions (emotes, comments, and shares) and post format. The results showed a significant association between image posts and share reactions (see

Table 7. Chi-square: The association between posts’ reactions and post format.

	Emotes			Comments			Shares
	Sig	Not Used	Used	Sig	Not Used	Used	Sig	Not Used	Used
No image	0.884	3.1%	96.9%	0.124	67.8%	32.2%	0.003	48.5%	51.5%
Image		3.2%	96.8%		65.1%	34.9%		43.0%	57.0%
No video	0.803	3.1%	96.9%	0.146	65.1%	34.9%	0.001	42.8%	57.2%
Video		3.1%	96.9%		67.7%	32.3%		49.3%	50.7%
No link	0.816	3.2%	96.8%	0.231	66.3%	33.7%	0.234	43.6%	56.4%
Link		3.1%	96.9%		64.4%	35.6%		45.6%	54.4%

).

In total, 57% of posts with images were shared, compared to 51.5% of posts without images. However, there was no significant association between emote or comment reactions and the use of images in a post. There was also no significant association between posts with video and emote or comment reactions. However, a significant association was found between posts with video and share reactions. In all, 50% of posts with video were shared compared to 57.2% of posts without video. There was no significant association between posts with links and emote, comment, or share reactions.

4.5. Post Content Analysis

The focus of universities is sharing information about events (32%), followed by student-related posts (23%), followed by posts related to reputation and social responsibility (21%). Posts related to education hold a moderate percentage (12%). It seems that universities focus less on posts related to announcements (7%) and posts related to faculty and research (5%) (see

Table 8. Descriptive statistics of post themes.

Codes	Themes	N	Percentage
1	Announcements	285	7%
2	Education	500	12%
3	Events	1330	32%
4	Faculty and Research	208	5%
5	Reputation and Social Responsibility	861	21%
6	Student	955	23%
7	Others	9	0%
	Total	4148	100%

).

Student-related content generates the highest level of engagement (35.4%). The audience’s engagement is reflected mainly by sharing (40.7%), commenting (38.5%), and using emotes (34.7%). Posts related to reputation and social responsibility and to events also drive a relatively high level of engagement of 17.9% and 17.6%, respectively. Posts driving the lowest levels of engagement are related mainly to announcements (2.4%) and faculty and research (3.5%). Posts related to education and “other” generate a medium level of engagement (10.5% and 12.6%, respectively) (see

Table 9. Engagement per theme.

		Comments		Shares		Emotes		Total Reactions
	Themes	N	Perc	N	Perc	N	Perc	N	Perc
1	Announcements	348	4.2%	433	2.5%	4156	2.3%	4937	2.4%
2	Education	1168	13.9%	1761	10.2%	18,340	10.4%	21,269	10.5%
3	Events	1688	20.2%	2697	15.6%	31,245	17.7%	35,630	17.6%
4	Faculty and Research	295	3.5%	632	3.7%	6250	3.5%	7177	3.5%
5	Reputation and SR	1206	14.4%	2701	15.6%	32,442	18.3%	36,349	17.9%
6	Student	3223	38.5%	7024	40.7%	61,465	34.7%	71,712	35.4%
7	Others	446	5.3%	2020	11.7%	23,124	13.1%	25,591	12.6%
	Total	8374	100%	17,269	100%	177,022	100%	202,665	100%

).

The results from

Table 10. Engagement per theme for public and private HEIs.

				PUBLIC
		Comments		Shares		Emotes		Total Reactions
	Themes	N	Perc	N	Perc	N	Perc	N	Perc
1	Announcements	37	3%	38	2%	531	3%	606	3%
2	Education	114	10%	147	8%	4783	23%	5044	21%
3	Events	370	33%	730	41%	7116	34%	8216	34%
4	Faculty and Research	136	12%	105	6%	1400	7%	1641	7%
5	Reputation and SR	369	33%	578	33%	4915	23%	5862	25%
6	Student	101	9%	172	10%	2276	11%	2549	11%
7	Others	0	0%	0	0%	0	0%	0	0%
	Total	1127	100%	1770	100%	21,021	100%	23,918	100%
				PRIVATE
		Comments		Shares		Emotes		Total Reactions
	Themes	N	Perc	N	Perc	N	Perc	N	Perc
1	Announcements	311	4%	395	3%	3625	2%	4331	2%
2	Education	1054	15%	1614	10%	13,557	9%	16,225	9%
3	Events	1318	18%	1967	13%	24,129	15%	27,414	15%
4	Faculty and Research	159	2%	527	3%	4850	3%	5536	3%
5	Reputation and SR	837	12%	2123	14%	27,527	18%	30,487	17%
6	Student	3122	43%	6852	44%	59,189	38%	69,163	39%
7	Others	446	6%	2020	13%	23,116	15%	25,582	14%
	Total	7247	100%	15,498	100%	155,993	100%	178,738	100%

show that private HEIs receive significantly more engagement across all themes categories (23,918 reactions for public HEIs against 178,738 reactions). For public HEIs, engagement is highest in events (34%) and reputation and SR (25%), while for private HEIs, the highest reactions come from posts related to student content (39%) and reputation and SR (17%).

For public HEIs, the lowest-engagement post themes are those related to others, announcements, and faculty and research, with 0%, 3%, and 7%, respectively. For private HEIs, the themes with the lowest reactions are announcements (2%), faculty and research (3%), and education (9%) (Table 10).

The results from

Table 11. Engagement per post format and theme.

		Text Only		Link		Video		Image
	Themes	N	Rea	N	Rea	N	Rea	N	Rea
1	Announcements	1	2	131	2367	46	527	234	4523
2	Education	2	44	272	12,477	147	6445	352	14,807
3	Events	9	207	291	8081	314	9791	1008	25,768
4	Faculty and Research	1	22	89	3073	34	1259	173	5895
5	Reputation and SR	5	95	230	8282	172	6108	682	30,126
6	Student	7	210	289	11,576	199	12,673	751	58,911
7	Others	0	0	3	2038	1	159	8	25,432
	Total	25	580	1305	47,894	913	36,962	3208	165,462

show that image is the most used format, and it represents the media that generates the highest level of engagement (number of total reactions), specially for the student (58,911 reactions) and reputation and social responsibility (30,126) themes. Links demonstrate significant engagement, particularly in education- (12,477) and student-related content (11,576). Videos perform best in the student (12,673) and events (9791) themes. Posts with text only perform poorly across all themes, which shows the preference of the audience for visual and interactive content.

Table 12. Engagement per language and theme.

		Arabic Only		English and Arabic		English Only
	Themes	N	Rea	N	Rea	N	Rea
1	Announcements	85	831	125	1740	75	2366
2	Education	75	881	214	12,437	211	7951
3	Events	310	5033	646	13,482	373	17,053
4	Faculty and Research	9	189	61	2265	138	4723
5	Reputation and SR	178	2571	372	7868	311	25,910
6	Student	169	1618	404	10,283	381	59,700
7	Others	1	103	3	82	5	25,406
	Total	827	11,226	1825	48,157	1494	143,109

shows that posts written in English generate the highest engagement, particularly for posts related to students (41.75%). Even though HEIs tend to make the highest number of posts in both languages Arabic and English (44%), overall this produces a moderate engagement, with more concentration on event content (35.41%). Posts written in Arabic generate only 5.5% of the reactions, also having a strong focus on event content (44.83%).

5. Discussion

Previous studies have examined users’ reactions to Facebook posts, especially likes, shares, and comments (Ellison et al., 2015; Joo et al., 2018; Kaur et al., 2019; Kim & Yang, 2017; Krebs et al., 2018; Smoliarova et al., 2018; Sturm Wilkerson et al., 2021; Zell & Moeller, 2018). Our results suggest that emotes were the primary reaction, with likes representing the highest proportion (87%). The results also showed more positive (like, love, wow, care, haha) than negative (sad, angry) reactions. This result is consistent with previous studies (Frison & Eggermont, 2015; Kaur et al., 2019). The second most common reaction was sharing (9%), followed by commenting (4%). This result is in line with previous studies’ findings (Kaur et al., 2019; Khobzi et al., 2019; Kim & Yang, 2017). These studies found that sharing posts requires deep cognitive processing and more effort than merely hitting the like button. In addition, users tended to share more comments on the posts. The post itself was shared only if a user liked the information and felt it was worthy of another audience (Carah, 2014; Kim & Yang, 2017). The comparative analysis results showed that private HEIs are more active than public HEIs on the Facebook platform in terms of the number of posts and format diversification, which reflects private HEIs’ efforts to manage and attract more reactions and build their reputation in the market. In addition, private and public HEIs differed in the types of reactions gained. In contrast to (Alsufyan & Aloud, 2017) findings in Saudi HEIs, our results show that private HEIs had a higher overall percentage of comments, emotes, and shares per post than public HEIs. This may be because there are more private than public HEIs; however, our statistical method considered the average reactions per post. The results indicate that, overall, the majority of HEI posts are written in English, with very few using Arabic only. Public HEIs primarily posted in English and Arabic, whereas private HEIs mainly posted in English only. This may be related to cultural changes, with younger generations communicating primarily in English. English has become the dominant language on social media (Maresova et al., 2020), even for users who are not native English speakers (Lantz-Andersson, 2018; Mcgee et al., 2015). Additionally, English is the medium of instruction in the majority of UAE academic programs, with the International English Language Testing System as an admission requirement (Ashour, 2020).

The sociolinguistic context of the UAE plays a pivotal role in shaping online engagement patterns. The UAE is characterized by its cultural diversity and linguistic duality; Arabic is the official language, but English functions as the primary medium of instruction in most universities. In addition, high school graduates must take a federal English exam in order to apply to college or university (Siemund et al., 2021). Over 88% of the population consists of expatriates, many of whom are more comfortable with English communication (GMI, 2025). This demographic makeup may explain why English-language posts received significantly more shares and emotes, as they potentially reach a broader audience. Meanwhile, Arabic posts appeared to evoke more emotional reactions, perhaps resonating more deeply with Emirati users. These dynamics suggest that bilingual posting strategies can optimize reach and resonance among diverse audiences. With a diverse population composed of both Emiratis and a large community of expatriates, bilingualism (primarily Arabic and English) is common. Our results also showed a significant difference in share reactions between private and public posts using English only, and in emote reactions for posts in Arabic. We found that the image format was the most frequently used (58% of posts) and that it is a preferred format in terms of reactions. This result aligns with (Valerio et al., 2015), who showed that images were the most effective format for promoting engagement among Mexican users. Overall, Abu Dhabi HEIs already use the most effective format for engaging users: images are used most frequently and text is used the least. These results are in line with (Rus & Cameron, 2016) findings that images are processed more quickly than text, are easier to recall, and create stronger emotional reactions. However, unlike Valerio et al.’s (2015) findings, our study found that the text format, rather than links, were the least efficient for engaging users.

In private HEIs, images were most effective for increasing digital engagement, followed by videos. Images were also most effective for public HEIs, but their second most effective format was links. In both private and public HEIs, the image format generated the highest percentage of reactions, followed by videos, links, and text. This is consistent with (Rus & Cameron, 2016) findings that different formats predict different types of interaction; posts with images generated more like and share reactions.

As per the qualitative analysis, the results show that even though posts related to events hold the largest percentage, it seems that posts related to students generate the highest level of engagement across all types of reactions. This result suggests that HEIs should increase the number of posts around student content. More particularly, posts related to students and containing images had the highest reactions. HEIs should rely on visual content to enhance the level of engagement. As per the difference between public and private HEIs, public HEIs attract more engagement on academic-related topics while private HEIs receive stronger reactions with posts related to students.

6. Conclusions

With shifting digital media landscapes, educational organizations are developing new ways of communicating and interacting. To manage social media growth, HEIs must continuously improve their digital content to attract their audiences’ attention and engage them. This study contributes to previous research by exploring the social media context and examining the Facebook posts and associated reaction features of higher education institutions (HEIs). This paper provides several theoretical contributions. Initially, it addresses the differentiation between public and private higher education institutions in terms of post and reaction activities, which have received little attention in the literature. Additionally, it contributes to analyzing the digital engagement in higher education, especially during the COVID context, mainly to understand how Facebook post characteristics influence user interaction across different institutional types. Specifically, it highlights the role of content type, format, and language in driving audience engagement.

The findings reinforce the applicability of Engagement Theory and Media Richness Theory in the higher education social media context. Posts that combine rich media formats with emotionally engaging community-centered content are more likely to elicit strong reactions. The differential performance between public and private HEIs also highlights the impact of institutional strategy and audience targeting. By understanding these dynamics, HEIs can refine their content strategies to better meet the expectations and preferences of their followers, ultimately enhancing both visibility and reputation. Furthermore, in multilingual and multicultural contexts like the UAE, linguistic inclusivity should be considered a strategic imperative for maximizing user engagement.

Our findings indicate that reactions to Facebook posts are diverse across a wide range of factors, such as HEI type, post format, and language. Private HEIs are more active on the Facebook platform than public HEIs, reflecting private HEIs’ efforts to attract an audience and build their reputation and image in the market. An understanding of these factors may help HEIs predict user reactions and thereby guide their social media digital content. It is crucial to analyze each post’s impact to identify the most and least effective formats to improve HEIs’ Facebook content strategies and enhance users’ engagement. The finding that user reactions are affected by post format and language emphasizes the importance of carefully choosing appropriate formats and languages to boost specific user interaction. In general, HEIs should focus on the effective management and promotion of user engagement with their Facebook pages and increase interaction in order to enhance their popularity and image to attract students.

The results emphasize that HEIs primarily focus on posting content on their social media related to promoting events and student-related topics, with student posts generating the highest engagement. Visual formats like images and videos, especially those related to students and HEI reputation and social responsibility, generate more engagement than text-only posts. Hence, HEIs should target specific visual and student content written in English to better engage with their audience. It is noticeable that the posts driving the lowest levels of engagement are related mainly to announcements and faculty and research. The results of the study can also be used to develop a social media model for HEIs where social media marketers can better understand the social media strategy to increase users’ engagement and enhance their interaction. Therefore, HEIs should create engaging posts aligned with their users’ reactions regarding shares, comments, and emotes. They should also evaluate the performance of their social media strategy by continuously adjusting their content based on their users’ reactions. The results of this study can be used as a guide for HEIs currently using social media to develop content that includes more images and less plain text. The main contribution of this study is that it provides insight into how public and private HEIs in Abu Dhabi exploit the Facebook platform to engage with their audiences.

This study also has practical implications for HEIs that will help them improve the effectiveness of their digital media strategies. The findings also provide a general picture of HEI practices on the Facebook platform and essential user reactions, which can help them better understand and predict how users engage with HEI posts and what encourages their engagement. In addition, the study findings can encourage UAE HEIs to build an integrated social media plan as part of their communication strategy, by considering that posts provide both a tool for disseminating information and a space for constructive communication that supports coordinating institutional information and performance. Finally, the study of the reactions to Facebook posts in higher education has the potential to inform and improve various aspects of student engagement, communication, and marketing. It is essential for HEIs to understand the types of Facebook posts that generate the highest engagement from users in order to tailor their content to increase student interaction. The study can also provide valuable information about users’ trends and preferences by understanding how students react to different types of Facebook posts, helping HEIs to better serve their target audience.

Limitations and Future Research

Despite offering significant insights, this study has limitations. The reliance on correlational data prevents causal inference. Additionally, user characteristics such as demographics, motivation, and platform familiarity were not measured and may have influenced engagement metrics. Lastly, while Facebook remains a dominant platform in the UAE, future research should explore cross-platform dynamics to validate whether these findings hold across other social networks like Instagram, YouTube, or LinkedIn.

This study has limitations that should be considered in future research. First, it focused on Abu Dhabi HEIs and users’ reactions to Facebook posts in general; therefore, future studies should consider the reactions of other user categories, such as students, parents, and teachers. Second, the study used a quantitative approach; future studies should broaden their methods to incorporate qualitative approaches. Finally, the study covered a small period of one academic year that coincided with pre- and mid-COVID-19 pandemic periods; future studies should examine longer time horizons and different periods.

Future studies will include a thorough investigation of content from various social media networks, including Twitter, Instagram, and YouTube, and will involve a critical statistical analysis of their datasets to gain a deeper understanding of Abu Dhabi HEIs’ social media usage in the age of digital communication. Future work could explore the influence of other factors, such as post length, content type, timing, posting frequency, and number of followers.