African higher Education institution’s response to COVID-19: A bibliometric analysis and visualisation study

Abstract The coronavirus outbreak significantly disrupted human life, including education, where in-person classes were cancelled, with online-only instruction the only feasible option. The crisis offered previously disadvantaged communities opportunities to leap into a new digital trajectory. The study aimed to analyse higher education institutions’ responses (HEI) in Africa during COVID-19 and evaluate their navigation into remote-based learning alternatives through a bibliometric and visualisation analysis. We used 238 articles published in the Scopus database between 2020 and 2022. VOSviewer extracted bibliometric indicators showing network diagrams and heat maps for keywords, affiliated institutions, and countries. African-based scholars contributed almost two-thirds of the publications. South Africa contributed almost half of the publications, with Egypt, Nigeria, and Ghana making other notable contributions. South Africa contributed to seven of the top ten leading universities in publications. Most African students lost learning time and failed to write examinations, thus disrupting semesters. The most common strategies employed were online learning and emergency remote learning. African educational institutions faced barriers such as a lack of access to electricity, the Internet, technological devices, technology-driven pedagogies, and deficient curricula. Across Africa, the crisis widened the digital divide and exposed inherent inadequacies and inequalities. These findings offer reflection points for academics, practitioners, and governments and advocate for policy reviews to support online learning. The study proposes a research agenda for African HEIs that incorporates artificial intelligence.


PUBLIC INTEREST STATEMENT
The COVID-19 outbreak significantly disrupted all facets of human life.Educational institutions closed their doors and moved to online-only instruction.Over 80% of African higher education institutions (HEI) struggled to go online, disrupting teaching and learning.The study analyses African HEIs' response and navigation into online learning.Disadvantaged communities can take this opportunity to leap into the digital trajectory.We conducted a bibliometric analysis using 238 articles published in Scopus between 2020 and 2022.We extracted bibliometric indicators showing network diagrams and heat maps for keywords, affiliated institutions, and countries.Results show that South Africa contributed half of the publications, followed by Egypt, Nigeria, and Ghana.African-based scholars contributed two-thirds of the publications.The strategies employed were online and emergency remote learning.Barriers include a lack of access to electricity, the Internet, devices, and deficient curricula.These findings offer reflection points for stakeholders for policy reviews to support online learning.

Introduction
The outbreak of the COVID-19 pandemic transformed every facet of human life with catastrophic consequences for higher education systems globally.Educational institutions worldwide were forced to close and suspend face-to-face activities within their campuses to curtail the spread of the disease.The highest percentage of higher education institutions (HEIs) that closed during the COVID-19 pandemic are in Africa, where 9.8 million African students were affected (Tamrat & Teferra, 2020).Although access to education is a constitutional and fundamental right for every citizen, COVID-19 disrupted traditional learning.It made learning a preserve for those with access to technological resources while disfranchising the poor (Jachna, 2021).
Governments across the world acted responsibly by curtailing the spread of the pandemic, and institutions abruptly migrated to the online environment with no time to train lecturers and students on how to use online platforms (Maphosa, 2022b;Ndebele & Mlambo, 2021).About 75% of learners who could not access online learning platforms came from rural and poor households; therefore, digital networks should be expanded to reduce such vulnerabilities (Demuyakor, 2021).This threatens equitable access to learning resources during the COVID-19 pandemic (Tamrat & Teferra, 2020).Simui et al. (2018) contended that African governments should focus on bridging the digital divide to enable students to acquire digital skills that would enable them to participate in the knowledge economy.
The significant challenges affecting online learning adoption include accessibility, affordability, curriculum review, flexibility, and inadequate educational policies (Xie et al., 2020). Du Plessis et al. (2022) summarised the critical factors universities require for successful online learning implementation, including connectivity, affordability, and student support.A study across five African countries showed that learners had embraced online learning after the pandemic, but 70% of the learners and lecturers considered it ineffective as they faced the reality of unreliable electricity, lack of computers, poor and expensive bandwidth, lack of technological and pedagogical skills (Paschal & Mkulu, 2020).This is worrying, considering Dhawan (2020:7) said, "Online teaching is no longer an option; it is a necessity."Maphosa (2022b) also found that students' lack of access to laptops and computers in their homes significantly impedes e-learning adoption in developing countries.El-Azar and Nelson (2020) noted that the current disruption to face-to-face teaching would result in the metamorphoses of higher education and that institutions that want to revert to the old status quo will not survive.COVID-19 is redefining teaching, learning and assisting developing countries in leapfrogging into the Fourth Industrial Revolution (4IR) (Onwuegbuzie & Ojo, 2021).Educational institutions are attracting learners across the globe; using artificial intelligence (AI) tools can assist instructors in translating content into the learner's native language, and voice recognition systems can assist in delivering lectures (Pranam, 2019).AI tools such as chatbot tutors are ideal during COVID-19 when face-to-face teaching is restricted and can replace human tutors and reduce human resources costs (Treve, 2021).
Bibliometric analysis has become useful for assessing and measuring research impact and productivity in the academic and scientific literature (Maphosa, 2023).Bibliometric analysis studies can be seen as analytical tools that have shaped the landscape of scholarly publishing and research management, ultimately enriching our understanding of the dynamic evolution of various fields of study (Donthu et al., 2021).Ye et al. (2012) used a bibliometric analysis to identify research themes and hotspots in tourism and hospitality research.Merigo et al. (2017) used bibliometrics to uncover prominent trends that shape the trajectory of academic domains, such as publishing trends in a journal.
The study was motivated by the learning disruptions caused by COVID-19 after many African institutions failed to migrate to the online environment, affecting the region's socioeconomic development.The study reviews scholarship on African higher education institutions' response to COVID-19, productive countries, co-authorship, productive institutions and H-index analysis.Our study contributes to the literature by identifying and summarising the main affordances and challenges brought by online learning.A bibliometric analysis was used to provide a thematic and visual overview of the response to COVID-19 by higher education institutions.A total of 238 articles were selected from the Scopus database for text mining and visualisation to reveal the strategies used by HEIs in response to COVID-19 and the emerging themes.The remainder of the paper covers connectivity challenges in Africa, the experiences of teachers and learners, methodology, and results.The results delve into productive countries and institutions, country coauthorship analysis, h-index and keyword analysis.The study ends with a discussion and conclusions.

Connectivity challenges in Africa
About two-thirds (60%) of the world's population is online, with 95% of North America connected compared to 39.3% of the African continent (internet world stats, 2022).In sub-Saharan Africa, 18% of learners have Internet access, and 11% have access to household computers (UNESCO, 2020).Reports on Internet connectivity reveal that Africa has an Internet penetration rate of 24%, with some rural communities experiencing poor connectivity.
Despite the challenges students and educators face, online learning is now accepted as a paradigm for now and the future without face-to-face teaching.Africa had the highest percentage of institutions affected by COVID-19, as a global survey reported that 77% of African campuses closed down at the pandemic's peak (International Association of Universities, 2020).Marinoni et al. (2020) estimate that 95% of African institutions closed due to the low uptake of educational technologies.
The cost of data in sub-Saharan Africa is the highest globally, as research showed that lecturers and students struggle to go online (Maphosa, 2021b).The cost of data forced some governments to declare radio and television the official channels for supporting learning during the pandemic (Maphosa, 2021b(Maphosa, , 2022a;;Mengistie, 2021).A survey across 52 African countries revealed that 71% of the participants agreed that the educational outcomes between urban and rural communities would continue to widen unless universal access to the Internet is prioritised (Edtech Hub, 2020).After the COVID-19 outbreak, only 29% of HEIs quickly migrated to the online environment, compared to 85% for European institutions, 72% for Americans, and 60% for Asian institutions (Tamrat & Teferra, 2020).

Lecturer's experiences
As universities worldwide turn to remote teaching and learning platforms, academics face challenges acquiring the competencies that enable them to teach effectively.Migrating to the online environment was abrupt, and academics were ill-prepared as most did not have the skills and had little to no experience using online learning platforms before the pandemic.Müller et al. (2021) note that lecturers usually use pre-recorded material from the previous semester with thematic forums for seamless integration, but during COVID-19, the transition was abrupt.In most African institutions, students enter university without knowledge of educational technology and tools such as laptops and smartphones due to a lack of electricity and access to broadband (Mengistie, 2021).Due to the lack of access to gadgets, the high cost of data, and inadequate training by academics and students, it is sad to note that teaching is conducted through WhatsApp, a social media tool with several limitations (Maphosa, 2021a).An Africa-wide survey revealed that 53 per cent of academics noted that lack of training was the primary obstacle affecting the adoption of remote learning during COVID-19 (Edtech Hub, 2020).Teymori and Fardin (2020) affirmed that academics experienced challenges in migrating to online learning due to unfamiliarity with the technology and the new teaching methods.Similarly, faculty at HEIs in developing countries have limited opportunities to learn how to teach with technology to enrich the learning process.Most educators were unprepared to design remote learning experiences with technology as there was no time to learn how to use technology to enrich learning as institutions suddenly closed (Trust, 2020).There was also resistance from academics with experience in traditional pedagogies who did not want to shift to new technologies (Duraku & Hoxha, 2020).

Learner's experiences
In West Africa, only 5% of learners have access to the Internet at home, compared to a 33% global average; 87% of learners in high-income countries have access to the Internet compared to 6% in low-income countries (Demuyakor, 2021).At the same time, the poor learners failed to access learning content and pursue their studies (United Nations Development Programme, 2020).During COVID-19, learners experienced several challenges, from substance abuse to financial difficulties and learning access.O'Regan (2021) reported that 65% of the participants suffered psychological distress, 57.9% lost study time, and 55.8% did not have enough money to procure learning materials.Dashtestani and Hojatpanah (2020) state that as learners returned home during the pandemic, digital gaps in the education sector became apparent as most learners from rural areas failed to attend lessons (Dashtestani & Hojatpanah, 2020).Although South Africa has the highest digital literacy in Africa, reports show that 68.4 per cent of South African students have difficulties traversing the online environment (Hanekom, 2020).Oyedemi and Mogano (2018) reported that almost half (47 per cent) of first-year students use a computer and the Internet for the first time when they start university.
The pandemic completely transformed students' lives as they migrated back to their homes and struggled to quickly learn new technologies in their content (Govindarajan & Srivastava, 2020).Students experienced challenges after moving home as there was minimum support compared to on-campus support services, which they usually get when they experience learning difficulties.Most students from public universities did not afford devices that could allow them to take online examinations, showing a lack of preparedness (Osabwa, 2022).Student bodies in South Africa, Malawi, and Zimbabwe rejected online learning, citing that it was not affordable and elitist (Maphosa, 2020;Mukeredzi et al., 2020).Some universities saw students adopting calls for #EndSemesterNow, #MassPromotionNow, and #NoStudentLeftBehind to push institutions to be sympathetic to challenges that learners were experiencing (Bozkurt et al., 2020).The learners' motivation decreases online as teacher feedback is not readily available as teachers may be offline, making learners feel isolated.
Governments worldwide have been left exposed by the COVID-19 pandemic, and this is an opportunity for investments in technology and curriculum development to ensure that future plagues are not very disruptive.Muftahu (2020) contends that African universities should fortify their digital capabilities as contingency plans to tackle future disruptions such as COVID-19.National Research and Education Networks (NRENs) mediate between institutions and service providers to ensure HEIs can access bandwidth, software, and cloud services at concessionary rates.
The visualisation not only uses data mining technology to excavate useful information from data but also displays the information obtained by data mining technology to users intuitively.Therefore, we must comprehensively overview this research direction and find some basic patterns of African HEIs' response to COVID-19.This study uses bibliometric analysis to understand how African HEIs responded to teaching and learning during COVID-19.The network analysis and heat maps help visually define this topic's hotspots.

Methodology
The Preferred Reporting Items for Systematic Review and Meta-Analyses (PRISMA) guidelines provide a comprehensive contextual understanding of previous studies published as a transparent and critically assessed report.PRISMA consists of three phases −1) identification, 2) screening, and 3) inclusion (Page et al., 2021).On 15 August 2022, we searched and collected data from the Scopus database, a comprehensive and widely used repository for bibliometric analysis.
The initial search string involved selecting articles focused on remote-based teaching/online systems by searching "COVID-19" AND "higher education" in the titles, abstracts, and keywords.This search yielded 5 106 articles.In the next step, we excluded articles where the geographical location of the study was undefined.A total of 159 articles were undefined, leaving 4 947 articles.Since the study was interested in the response of African HEIs, 4 607 studies conducted in non-African countries were excluded, leaving 340 articles.The authors exported the remaining 340 articles and the abstracts to MS Excel for analysis.Publications not related to COVID-19 and African HEIs' response to COVID-19 were removed, leaving 238 articles that were analysed.Figure Figure 1 shows the article selection process.We downloaded data from the 238 articles from the Scopus database in text format.We then exported the complete records for these articles to VOSviewer, a free bibliometric tool that offers visualisation and text-mining abilities.

Results
We evaluated bibliometric data from the Scopus database to better understand how African HEIs responded to COVID-19.The study results are presented, showing the publication type, the authors' geographic distribution, and the top 20 universities with which the authors are affiliated.Finally, we analyse bibliometric data through a heat map and a network diagram based on the keywords.Most of the documents retrieved were journal articles, accounting for over three-quarters (77.45 per cent) of all publications, as shown in Table 1.Conference proceeding papers followed with 15.32 per cent, book chapters accounted for 3.83 per cent, and book reviews with 3.4 per cent.The small number of conference papers indicates that most articles were published during the pandemic's peak when face-to-face interactions were restricted.
Table 2 shows the geographical distribution of the articles.Results show that South Africanbased authors contributed more than half (53.62 per cent) of the publications on African HEIs' response to COVID-19.This dominance is corroborated by the African university rankings, where South Africa contributes 8 of the top 10 universities (UniRank, 2022).Other countries with significant contributions include Egypt (8.08 per cent), Nigeria (7.23 per cent), and the United Kingdom (6.81 per cent), with Ghana and Morocco contributing 6.83 per cent each.Close to half (41.79 per cent) of the countries contributed a single publication, while 8.96 per cent contributed double-figure publications.Authors based in the African continent dominated publications, with 135 articles, while those from non-African countries contributed 100 articles.

The country co-authorship analysis
Country co-authorship analysis is an essential form of co-authorship analysis.It can reflect the degree of collaboration between countries and influential countries in this field.The country's coauthorship network of publications on African HEIs' response to COVID-19 is shown in Figure Figure 3.The map has many colours, showing the diversification of research directions.The significant nodes represent the influential countries.The links between nodes define the cooperative relationships among institutions.The distance between the nodes and the thickness of the links represent the level of cooperation among countries.
The 238 articles are related to 67 countries.Figure Figure 3 demonstrates the closest cooperation network, including 65 countries.The nodes denote countries, and the sizes of the nodes mean the number of articles.The 65 countries are classified into 5 clusters, and the most cooperative countries are South Africa, Egypt, Nigeria, the United Kingdom, Ghana, and Morocco.Additionally, the connections between two countries represent the degree of cooperation; the thicker the connection, the closer the cooperation.The map analysis indicates that geographical proximity is not the primary factor influencing the cooperative relationship.

H-Index analysis
The retrieved research papers have an h-index of 12.The h-index of 12 means that of the 238 research articles, 12 have received at least 12 citations.This low h-index can be attributed to the fact that the articles have not been published for a long time, with the first paper published in 2020.

Keyword analysis
Figure Figure 4 shows the keyword density visualisation for the study.Chawla and Davis (2013) posit that density maps visually represent keywords, drawing readers to the map's most critical areas.VOSviewer uses colours to reveal the density of the terms, with green having the lowest density while red having the highest density (van Eck & Waltman, 2017).From Figure Figure 4, (source: generated using VOSviewer on data).
e-learning, online learning, teaching, students, and South Africa are the most dominant and frequently used keywords.The prominently referenced words are learning, human, and blended learning.Words like distance learning, remote learning, face-to-face, the emergence of remote teaching, developing countries, online teaching, learning systems, engineering education, decolonisation, and educational computing follow this.Anxiety, Egypt, and universities are the least referenced set of words in green.
In a visualisation map, concepts and keywords are joined into clusters that interconnect with other clusters.A strong link between two words is depicted by a thick line joining the two concepts (van Eck & Waltman, 2017).Figure Figure 5 shows that the blue cluster covers e-learning and related concepts, such as learning systems, blended learning, and teaching and learning methods deployed in developing countries.The orange cluster relates to online learning and  (source: generated using VOSviewer on data).covers concepts related to teaching, learning, and student engagement concepts.The green cluster represents the human elements related to learning, such as anxiety, fatigue, boredom, and university education.The purple cluster represents the most dominant country regarding articles related to the response to COVID-19 by HEIs.As South Africa dominated, themes such as decolonisation of education, learner support through the provision of laptops, and comparison between white-dominated and black-dominated universities were covered.The last cluster is the red cluster, which is the biggest.This cluster relates to students and the remote-based learning methods available to them.The network diagram shows that the significant learning methodologies available to students include distance learning, remote learning, online teaching, emergency remote teaching, face-to-face and computing, and engineering education.This shows the different forms of remote teaching that many African institutions adopted.

Discussion
This study provided a bibliometric analysis of the response to COVID-19 by African HEIs, as face-toface teaching and learning were banned.This scientific study analysed 238 publications from the Scopus database covering teaching and learning strategies adopted by African HEIs during COVID-19 from 2020 to 2022.Literature shows that over 85 per cent of African countries struggled to migrate to online teaching and learning (Tamrat & Teferra, 2020).The COVID-19 outbreak devastated teachers and learners in most African HEIs.Students experienced negative emotional and socio-economic challenges as institutions closed campuses.Despite the challenges, online learning has been accepted as a current and future paradigm.
African countries experienced several challenges during the pandemic outbreak, and the urgent challenge was how universities could transform their traditional face-to-face courses for online delivery.The bibliometric analysis covered critical factors that hindered the migration to remote-based learning, which include connectivity, affordability, and student support.Analysis of results shows that affordability and connectivity were the most critical factors.In most African countries, learners experience erratic power supply, lack of devices, poor Internet services, and pedagogical challenges (Mengistie, 2021).Internet penetration in Africa is 24%.In most rural areas where most of the population resides, it is lower, with some with no electricity and Internet connectivity, making online-based learning impractical (Tamrat & Teferra, 2020).This resulted in many learners failing to attend online lectures.This forced many lecturers to adopt asynchronous learning platforms such as WhatsApp as their preferred tool, as learners and teachers could not afford to use videoconferencing, which can improve information assimilation and understanding.
As learners moved to their homes, most home environments were not conducive to learning.During lockdowns, where movement was restricted, learners further experienced isolation as they could not visit information centres to access the Internet.Many student bodies took their education ministries to court as they felt that adopting online learning platforms would widen the digital divide further (Bozkurt et al., 2020).Some of the themes from the study relate to human factors regarding adopting remote-based teaching and learning by HEIs in Africa.Many learners and teachers experience stress, anxiety, and depression due to the abrupt migration to remote-based teaching and other online applications.The on-campus support services students usually experienced became unavailable as learners moved to their homes, raising stress levels.It takes years for institutions to migrate to online learning platforms; during COVID-19, African institutions were forced to shift overnight without any preparation or training.The learners' motivation decreases due to erratic feedback from the teacher who may not be online, making learners feel isolated and frustrated.Most teachers were not trained in online pedagogies, making them anxious, and most failed to deliver their lectures effectively.Due to a lack of appropriate training, most teachers failed to adapt and integrate their teaching material into the online environment, making learners face challenges in assimilating the material.
From the bibliometric analysis, countries responded differently, but some remote teaching was adopted.Regarding publications and citations on how African higher education countries responded to COVID-19, South Africa is the leading country, contributing almost 50% of the articles.Most South African universities procured laptops and data for students to support remotebased teaching.COVID-19 was seen to increase pre-existing disparities.There was an outcry that former white-based universities were more prepared for the abrupt face-to-face teaching disruptions due to COVID-19 than black-based universities.The pandemic demonstrated the glaring inequalities in higher education in South Africa as the wealthy easily accessed technologies supporting the new teaching mode while the poor were left behind.
In other countries, governments declared that remote-based teaching and learning should be adopted through online, radio, and television applications.Access to educational devices, the high cost of the Internet, and the lack of a curriculum that supports emerging technologies was a barrier to online learning adoption.Teaching and learning time was lost, and most public institutions failed to administer examinations, resulting in students failing to conclude their semesters.Emerging research fields of remote-based teaching include artificial intelligence-driven fields such as deep learning, machine learning, virtual and augmented reality, and recommender systems.The findings of this study can assist governments, policymakers, and educational technologists in implementing remote-based teaching and learning for developing and emerging economies.

Conclusion
Using bibliometric analysis is a new methodology by authors in developing countries.Our study contributes to the literature by using bibliometrics to show the current landscape of African HEIs' response to adopting remote-based teaching methods during the COVID-19 pandemic, a unique and currently explosive research area.This study will assist new researchers who want to use bibliometrics to establish research gaps for future work.The identified significant themes and emerging research areas provide a gap for future researchers who may want to further research Africa's response to other disruptions to face-to-face teaching.This research contributes to practice by raising debate and discussions on how countries responded to the COVID-19 disruptions in teaching and learning.African countries should improve their technological infrastructure and be responsive in supporting the implementation of curricula that support remote-based teaching and learning.Teachers should be capacitated in digital pedagogies and support rural institutions in acquiring educational technologies and infrastructure investments to improve connectivity.Future studies may focus on the lived experiences of the learners and teachers during the migration from faceto-face teaching to remote-based teaching in African countries.Online migration is irreversible; institutions should calibrate their curriculum, upgrade their infrastructure, and train academic staff.We recommend the integration of AI into online learning platforms to increase efficiency and responsiveness and take advantage of the many affordances brought by AI.AI can supplement online learning through personalised learning, intelligent tutoring systems, chatbots, robots, and virtual laboratories.Future research should evaluate the integration of AI by African HEIs.
Our study has a few limitations.We used the Scopus database for the bibliometric analysis, which means that other articles that could provide good insight were excluded, leaving out some valuable insights.Since Scopus is a leading database, our findings reflect the status quo.The high subscription for open-access journals published under Scopus may have hindered researchers from developing countries from contributing articles.The high volumes of articles from South Africa show the country's positive outlook on innovation and research by assisting researchers in paying for open-access publication fees.The exclusion of non-openaccess journals may have excluded authors from other countries who do not get assistance to publish in top open-access journals which demand payment.

Figure
Figure 2. Top 20 universities based on publications.

Figure
Figure 4. Keyword density visualisation map.(source: generated using VOSviewer on data).

Figure
Figure 5. Keyword cooccurrence network visualisation of high-frequency terms in titles and abstracts.

Table 1 . Document types on the articles retrieved Document Type Count Percentage
Statistics showing the top 20 productive institutions in African HEIs' response to COVID-19 are presented in Figure Figure 2. The dominance of South African institutions reflects the country's

Table 2 . Geographical distribution of the articles Rank Country Publications
position in academic and scientific research.The top three largest economies in Africa in gross domestic product contributed to the top 20 universities.This shows researchers' enthusiasm for tackling problems affecting the continent.All the top 10 contributing universities were South African, indicating that the published literature is highly skewed towards the South African context and narrative.Out of the top 20-ranked universities in article contributions, 17 (85 per cent) were South African-based, while Egypt, Ghana, and Nigeria have one university each.This aligns with rankings of African universities, which show that Egyptian, Nigerian and Ghanaian universities are ranked highly after the South African universities (UniRank, 2022).