Cloud Computing Adoption Guidelines in a South African Higher Education Institution R Azwidowi orcid.org 0000-0002-6878-1267 Dissertation accepted in fulfilment of the requirements for the degree Master of Science in Computer Science at the North-West University Supervisor: Dr I Smit Co-supervisor: Dr M Thobejane Graduation: July/August 2024 i ABSTRACT Global technology is advancing rapidly, particularly within higher education institutions, where remaining competitive in teaching and learning is paramount. Embracing cloud computing offers numerous benefits that higher education institutions could leverage to enhance their operations. However, this requires a paradigm shift in one’s approach. While many higher education institutions aim to adopt cloud computing, their focus tends to centre on Software as a Service initially, overlooking the broader potential benefits. Cloud computing offers scalability solutions that could drive down costs and increase flexibility, allowing higher education institutions to concentrate on their core educational missions. Nevertheless, challenges persist, hindering widespread adoption, especially among smaller institutions. The primary objective of this study focused on cloud computing adoption guidelines in a South African higher education institution. Through comprehensive research, various barriers to adoption have been identified. Moreover, this study sheds light on the benefits of cloud computing, particularly evident during times such as the COVID-19 pandemic, where it enhanced teaching and learning experiences significantly. Some higher education institutions have already embraced certain cloud computing services, such as Microsoft applications through Software as a Service models. Employing an interpretative research approach, this study engaged information technology technical staff and information technology governance within a higher education institution. Data collection was primarily conducted through interviews, with content analysis serving as the primary research method. Utilising a case study approach, the interview questions were informed by an extensive literature review, structuring the investigation process. Qualitative methods were used to gather insights from information technology governance and information technology technical staff, with data being carefully analysed, coded, and categorised to identify emerging patterns and themes. Through this process, this study found barriers and challenges confirmed by extant literature on cloud computing. The development of cloud computing guidelines aimed at facilitating cloud computing adoption in South African higher education institution. The key findings of this study outline the guidelines for cloud computing adoption in higher education institutions. The guidelines incorporate participant insights on optimizing cloud computing adoption. Factors extracted including technological, environmental, governance, financial, organizational factors, and challenges, guide the guidelines' formulation. Recommendations include evaluating data protection, upskilling information technology staff, prioritizing security, aligning with higher education institution policies, awareness on cloud ii computing benefits, assessing financial impacts, fostering a learning culture, and addressing challenges such as trust, costs and support. It emphasizes the importance of service level agreement with service providers, vendor selection, intellectual property considerations, and staff training to ensure successful cloud computing adoption aligned with higher education institution digital migration strategies. Keywords: Cloud computing, cloud computing characteristics, cloud computing models, guidelines, higher education institution iii ACKNOWLEDGEMENT I would like to express my gratitude to God for providing me with the strength and energy to complete this project. Without His support, it would not have been possible. This journey has been arduous, but it has taught me invaluable lessons and strengthened me. Firstly, I extend my sincere appreciation to Dr I Smit, my supervisor, for her unwavering guidance and support throughout this endeavour. Additionally, I am indebted to my co- supervisor, Dr MC Thobejane, whose encouragement helped me persevere during moments of doubt. I dedicate this dissertation to my late father, Mr TL Azwidowi Nemukula, who instilled in me the value of education and continuously encouraged me to pursue my dreams. I am also grateful to my mother, a pillar of strength, whose prayers sustained me through this journey and enabled me to attain this degree. I extend my heartfelt thanks to my supportive family, especially my wife, Merium Azwidowi, and my children, for their understanding and patience during times when my studies required my full attention. Additionally, I am thankful to my colleagues who provided support and encouragement, offering a listening ear during moments of frustration. I am deeply thankful to several individuals who have played pivotal roles in helping me achieve my academic aspirations. Lastly, I express my appreciation to my colleagues at NWU Vaal for their support and assistance throughout this journey. Thank you all for your unwavering encouragement and belief in my abilities. iv DECLARATION I, Robert Azwidowi, declare that Cloud Computing Adoption Guidelines in a South African Higher Education Institution, is entirely my own work that I have submitted for the Master of Computer Science programme at North-West University and has not been presented to any other institution. Furthermore, I have appropriately cited and acknowledged all references used in the dissertation. Signature: ……………………………. Date: ………………………………… v ABBREVIATIONS API Application Programming Interface CC Cloud Computing CS Case study EO Empirical objectives FNAS-REC Faculty of Natural and Agricultural Science Research Ethics Committee HEI Higher Education Institution IAAS Infrastructure as a Service ICT Information Communication Technology IT Information Technology NHMRC National Health and Medical Research Council NIST National Institute of Standards and Technology NWU North-West University NWU-RDGC North-West University Research Data Gatekeeper Committee PAAS Platform as a Service SAAS Software as a Service SANReN South African National Education and Research Network SLA Service level agreement TO Theoretical objectives TVET Technical Vocational Education and Training colleges A note on abbreviations This list of abbreviations helps readers understand the content of this dissertation. When a concept is first mentioned in each chapter, its full definition is provided in brackets with the abbreviation, followed by the abbreviated form. vi TABLE OF CONTENTS ABSTRACT ........................................................................................................................... i ACKNOWLEDGEMENT ...................................................................................................... iii DECLARATION ................................................................................................................... iv ABBREVIATIONS ................................................................................................................ v TABLE OF CONTENTS ...................................................................................................... vi LIST OF TABLES ................................................................................................................. x LIST OF FIGURES .............................................................................................................. xi CHAPTER 1: INTRODUCTION AND BACKGROUND TO THE STUDY .............................. 2 1.1 INTRODUCTION ................................................................................................. 2 1.2 CONCEPTS CENTRAL TO THIS STUDY ........................................................... 3 1.2.1 Cloud computing and its characteristics ........................................................... 3 1.2.2 Cloud computing models ................................................................................. 3 1.2.3 Guidelines ........................................................................................................ 4 1.2.4 Higher education institutions ............................................................................ 6 1.3 PROBLEM STATEMENT .................................................................................... 7 1.4 OBJECTIVES OF THIS STUDY .......................................................................... 9 1.4.1 Primary objective ............................................................................................. 9 1.4.2 Theoretical objectives ...................................................................................... 9 1.4.3 Empirical objectives ......................................................................................... 9 1.5 RESEARCH DESIGN AND METHODOLOGY................................................... 10 1.5.1 Appropriate theory of this study...................................................................... 10 1.5.2 Research paradigm........................................................................................ 10 1.5.3 Empirical research ......................................................................................... 12 1.5.4 Delimitations to this study .............................................................................. 18 1.5.5 Expected contribution to knowledge ............................................................... 19 1.6 LITERATURE REVIEW ..................................................................................... 19 1.7 ETHICAL CONSIDERATIONS .......................................................................... 19 1.8 CHAPTER CLASSIFICATION ........................................................................... 20 1.9 SUMMARY ........................................................................................................ 21 CHAPTER 2: RESEARCH DESIGN AND METHODOLOGY ............................................. 23 2.1 INTRODUCTION ............................................................................................... 23 2.2 RESEARCH PARADIGMS ................................................................................ 23 2.2.1 Positivism ...................................................................................................... 24 2.2.2 Interpretivism ................................................................................................. 26 2.2.3 Critical social theory ....................................................................................... 26 vii 2.2.4 Critical realism ............................................................................................... 27 2.2.5 Paradigm appropriate for this study ............................................................... 27 2.3 RESEARCH METHODOLOGY ......................................................................... 28 2.3.1 Principle for Interpretative research ............................................................... 29 2.3.2 Theoretical grounding .................................................................................... 32 2.3.3 Research methods ......................................................................................... 34 2.3.4 Case study method ........................................................................................ 35 2.3.5 Data collection method .................................................................................. 37 2.4 RESEARCH DESIGN ........................................................................................ 39 2.5 SUMMARY ........................................................................................................ 40 CHAPTER 3: LITERATURE REVIEW ................................................................................ 43 3.1 INTRODUCTION ............................................................................................... 43 3.2 CLOUD COMPUTING AND ITS CHARACTERISTICS ...................................... 44 3.2.1 Defining cloud computing ............................................................................... 44 3.2.2 Characteristics of cloud computing ................................................................ 45 3.2.3 Cloud computing models ............................................................................... 47 3.2.4 Cloud deployment models.............................................................................. 52 3.3 BENEFITS OF CLOUD COMPUTING ............................................................... 54 3.4 CHALLENGES IN ADOPTING CLOUD COMPUTING ...................................... 56 3.5 DEVELOPMENT OF GUIDELINES ................................................................... 63 3.6 KEY PRINCIPLE OF GUIDELINES ................................................................... 64 3.7 HIGHER EDUCATION ...................................................................................... 66 3.8 GUIDELINES EXTRACTED FROM LITERATURE ............................................ 70 3.9 SUMMARY ........................................................................................................ 75 CHAPTER 4: EMPIRICAL STUDY ..................................................................................... 77 4.1 INTRODUCTION ............................................................................................... 77 4.2 DATA COLLECTION ......................................................................................... 77 4.3 PARTICIPANTS ................................................................................................ 78 4.4 DATA CODING ................................................................................................. 81 4.5 DATA ANALYSIS .............................................................................................. 81 4.5.1 Step 1: Preparation of data ............................................................................ 83 4.5.2 Step 2: Define the unit of analysis .................................................................. 83 4.5.3 Step 3: Developing categories and coding scheme ........................................ 83 4.5.4 Step 4: Test codes and categories ................................................................. 84 4.5.5 Step 5: Code all the text ................................................................................. 86 4.5.6 Step 6: Assess consistency of codes, categories and themes ....................... 86 4.6 DISCUSSION OF THEMES .............................................................................. 86 viii 4.6.1 Theme: Technological factors ........................................................................ 87 4.6.2 Theme: Environmental factors ....................................................................... 95 4.6.3 Theme: Governance factors ........................................................................ 100 4.6.4 Theme: Financial factors .............................................................................. 104 4.6.5 Theme: Organisational factors ..................................................................... 108 4.6.6 Theme: Challenges of CC in higher education institutions ........................... 111 4.7 EMERGING THEMES AS GUIDELINES ......................................................... 115 4.7.1 FG01: Ensures secure, scalable, and robust IT infrastructure for cloud adoption in HEIs 119 4.7.2 FG02: Emphasizes the need for HEI staff to develop technical skills for cloud initiatives .................................................................................................................... 119 4.7.3 FG03: Mitigates data privacy, uncertainty, and training challenges for smooth cloud integration in HEIs ............................................................................................ 120 4.7.4 FG04: Establishes frameworks for cloud strategy and governance, aligning with institutional goals and regulations ....................................................................... 120 4.7.5 FG05: Encourages HEIs to collaborate with reliable cloud providers for expert management and SLA adherence .............................................................................. 121 4.7.6 FG06: Optimizes budget management for cloud adoption in HEIs, focusing on cost reduction, effectiveness, and outsourcing ........................................................... 121 4.8 RESPONSE TO THE FINDINGS .................................................................... 121 4.9 SUMMARY ...................................................................................................... 122 CHAPTER 5: CONCLUSION AND RECOMMENDATIONS ............................................. 125 5.1 INTRODUCTION ............................................................................................. 125 5.2 CHAPTER SUMMARY .................................................................................... 125 5.3 STUDY REFLECTIONS .................................................................................. 126 5.3.1 Research objectives ..................................................................................... 126 5.3.2 Research questions ..................................................................................... 127 5.4 INTERPRETIVE RESEARCH PRINCIPLES .................................................... 128 5.5 GUIDELINES IN SETTING GUIDELINES ....................................................... 130 5.6 ETHICAL CONSIDERATIONS ........................................................................ 132 5.7 STUDY LIMITATIONS ..................................................................................... 132 5.8 RESEARCH FINDINGS ........................................................................................... 133 5.9 CONTRIBUTION OF STUDY .......................................................................... 136 5.10 RECOMMENDATIONS FOR FUTURE RESEARCH ....................................... 137 5.11 SUMMARY ...................................................................................................... 137 REFERENCES .............................................................................................................. 138 APPENDIX A: INTERVIEW QUESTIONS ..................................................................... 154 APPENDIX B: PARTICIPANT CONSENT FORM .......................................................... 156 APPENDIX C: LIST OF CODES FROM EMPIRICAL DATA .......................................... 158 ix APPENDIX D: FACULTY OF NATURAL AND AGRICULTURAL SCIENCE RESEARCH ETHICS COMMITTEE ................................................................................................... 162 APPENDIX E: EDITOR’S DECLARATION .................................................................... 164 x LIST OF TABLES Table 1.1: Four types of cloud deployment models ............................................................... 4 Table 2.1: Comparison of the paradigms within ontological, epistemological and methodological (Aliyu et al., 2014:81; Aliyu et al., 2015) ..................................................... 25 Table 2.2: Summary of the principles of hermeneutics and its application to this study ....... 29 Table 3.1: Comparison of cloud computing models (Bokhari et al., 2016:894) .................... 51 Table 3.2: Comparison of cloud development models ........................................................ 54 Table 3.3: Cloud computing adoption factors in HEIs .......................................................... 61 Table 3.4: Nine key principles of guidelines ........................................................................ 64 Table 4.1: Participants ........................................................................................................ 79 Table 4.2: Guiding questions ............................................................................................... 80 Table 4.3: An overview of unit testing .................................................................................. 85 Table 4.4: Codes related to technological factors ................................................................ 87 Table 4.5: Codes related to environmental factors .............................................................. 95 Table 4.6: Codes related to governance factors ................................................................ 100 Table 4.7: Codes related for financial factors .................................................................... 105 Table 4.8: Codes related for organisational factors ........................................................... 108 Table 4.9: Codes related challenges of CC in higher education institutions ....................... 111 Table 4.10: Emerged themes and guidelines from empirical study .................................... 116 Table 4.11: Emerged guidelines from Chapter 3 - literature review ................................... 117 Table 4.12: Guidelines comparison literature review and empirical results on the emerged final guidelines .................................................................................................................. 118 Table 4.13: Summarising benefits, challenges, and recommendations for each guideline 122 xi LIST OF FIGURES Figure 1.1: Education cloud stakeholders .............................................................................. 7 Figure 2. 1: Presentation of the research design process ................................................. 39 Figure 4.1: ATLAS.ti participant identification ...................................................................... 83 Figure 4.2: Provides a description of developing categories and coding scheme (Saldaña, 2021:12).............................................................................................................................. 84 Figure 5.1: Final list of guidelines ...................................................................................... 135 1 2 CHAPTER 1: INTRODUCTION AND BACKGROUND TO THE STUDY 1.1 INTRODUCTION Cloud computing (CC) has significantly transformed processes in higher education, playing a pivotal role in delivering services to higher education institution (HEI). Students, teachers, administrators, and other stakeholders in HEI stand to gain from the promising new features offered by CC (Ali, 2020:413). Cloud computing enhances the sustainability of HEI by reducing information technology (IT) expenses and enhancing operational efficiency, thereby bolstering their long-term viability (Qasem, Asadi, et al., 2020). A growing model of cloud computing enables institutions to access computing resources directly from the cloud as a service (Moloja & Ruhode, 2020:1262). McCrea (2009:54) mentions that CC provides universities with more possibilities to enhance teaching, learning and research, rather than focusing on complex information communication technology (ICT) configurations and software integration systems. Cloud computing has emerged as an innovative paradigm where companies and organisations enhance their ICT investment, making these resources quickly accessible (Singh & Bhisikar, 2013:37). Cloud computing refers to the services and the program applications that are delivered over the Internet (Armbrust et al., 2010:50). Kulkarni (2012:11) mentions that in simple terms, CC means Internet Computing, where the Internet is virtualised as a cloud; hence, the term cloud computing, which means that computing is done through the Internet. Through CC, users can access data anywhere and anytime, if they have an Internet connection. Universities and companies experience challenges related to managing big data and complex systems, because of a lack of experience in effectively using ICT. Hussein and Khalid (2016:52) state that CC enables cost-effective data sharing over the Internet, as well as big storage of data. According to Rajan (2013:38), CC changes the way in which recent ICT infrastructure manages and presents consumable services such as platforms, applications and infrastructure. Section 1.2 focuses on the concepts central to this study that include CC and its characteristics, CC models, guidelines, and HEIs. The problem statement is discussed in section 1.3. Section 1.4 focuses on the objectives of this study that include primary objectives, theoretical objectives and the empirical objectives. Section 1.5 focuses on the research design and methodology and includes the research paradigm, empirical research, data collection and 3 the evaluation methods. Section 1.6 refers briefly to the literature review, while Section 1.7 explains ethical considerations. The last section 1.8 outlines the chapter classification. 1.2 CONCEPTS CENTRAL TO THIS STUDY The key concepts of this study included the following: cloud computing (CC) and its characteristics, guidelines, and higher education institutions (HEIs). Each concept is introduced below and discussed in detail in the literature review of this study. 1.2.1 Cloud computing and its characteristics This section provides a description of CC, which includes definitions, cloud characteristics, cloud models and cloud deployment. The future of ICT application systems is determined by the characteristic of CC (Rashid & Chaturvedi, 2019:423). 1.2.1.1 Cloud computing Rashid and Chaturvedi (2019) define CC as “storing and accessing data and programs over the Internet instead of using our computer hard drive”. They mention that the cloud is a metaphor for Internet. CC is a way of delivering computing services over the Internet and because of its capabilities to be incorporated into education processes, it can bring wider benefits, such as the improvement of the quality of higher education across HEIs. 1.2.1.2 Characteristics of cloud computing A considerable number of scholars (Aldossary & Allen, 2016:485; Mell & Grance, 2011) describe the characteristics of CC in the same way. For instance, the National Institute Standard of Technology (NIST) identifies the following four characteristics of CC, namely on- demand self-service, broad network access, resource pooling, and rapid elasticity (Aldossary & Allen, 2016:485; Mell & Grance, 2011). Kaur (2020:920) adds one to this list, namely measured service. 1.2.2 Cloud computing models Businesses and organisations are able to benefit from the cloud models, which improve efficiency and reduce cost (Rashid & Chaturvedi, 2019). The CC models bring numerous benefits, for example Software as a Service (SaaS), which provides pay-per-use services to customers over the Internet; implying that you pay only for what you consume (Singh & Bhisikar, 2013:38). 4 1.2.2.1 The cloud deployment model Felter (2021) states that cloud deployment may be described as a way in which the cloud platform executes, how it is hosted and who can access it. He further explains that all CC deployment models work on the same proposition by virtualising the servers into components that result in enhanced data processing and storage capacity (Felter, 2021). Alam (2020:111) states that during CC implementations, application models are deployed. He further mentions that there are four deployment models, including public cloud, private cloud, hybrid cloud and community cloud. Tavbulatova et al. (2020:1) posit that a cloud deployment model represents four types of cloud environments, including private clouds, public clouds, hybrid clouds and community clouds, all of which may be differentiated by ownership, as illustrated in Table 1.1. Table 1.1: Four types of cloud deployment models Ownership Public Cloud Private Cloud Hybrid Cloud Community Cloud Performance Owned by customers Owned by single organisation Owned partially by both service provider and consumer Owned by two or more organisations having the same goal Used by Anyone can access Few people can access Medium accessibility Depend upon the number of cooperatives Reliability Medium Highest Moderato High Maintenance cost Lowest Highest Moderate High Security Less Highest Medium High Example Amazon Microsoft Azure Rackspace Hybrid cloud Microsoft government community cloud 1.2.2.2 The cloud service models According to Rashid and Chaturvedi (2019:423), there are three common CC service models, namely SaaS, Platform as a Service (PaaS), and Infrastructure as a Service (IaaS). 1.2.3 Guidelines Craig et al. (2011:1) mention that by applying theory and practice, guidelines can help intervention program implementers to ensure that their interventions are implemented and evaluated effectively. It is generally believed that guidelines can make it easier to give consistent and efficient care and that they can bridge the gap between what scientists do and what is supported by scientific evidence (Eccles & Mason, 2001:1). 5 Gopalakrishna et al. (2013:14) state that by using the best available evidence, guideline panels develop recommendations. The evidence provided by panels is often graded in this process. It can reduce bias and provide transparency in the process of developing guidelines when the evidence is evaluated systematically, and recommendations are developed in this way. Guideline developers and target audiences both benefit from this. Thus, evidence and recommendations have been graded according to their quality and strength (Gopalakrishna et al., 2013:14). Clifton and Petrosino (2016) state that across all fields of practice, policy, and resource allocation, guidelines have a significant impact. Therefore, guideline documents should include recommendations based on objective evidence assessments. Thorough literature reviews and using rigorous methods to evaluate evidence help to reduce scientific bias. In this way, they improve the chances that high-quality, relevant evidence will be considered. Clifton and Petrosino (2016) further stated that evaluating scientific evidence is only one aspect of guideline development. To achieve a desired outcome, guidelines define the procedures that need to be followed (Howard & Jenson, 1999:285). There are nine basic key principles that the researcher will adopt on the development of the guidelines (NHMRC, 2000:1). The National Health and Medical Research Council (NHMRC) of Australia recommends the following guideline principles: • Principle one: The focus should be on outcomes when developing guidelines. • Principle two: Resource constraints should be considered when developing guidelines. • Principle three: It is important that guidelines are based on the best scientific evidence and that their recommendations are described as strong. • Principle four: To synthesize the evidence, one of the most valid methods should be used. • Principle five: Consumers must be involved in the guideline development process. • Principle six: Flexibility and adaptability should be key characteristics of guidelines. • Principle seven: A target audience is considered when developing guidelines for dissemination and implementation. • Principle eight: Guidelines should be evaluated for their implementation and impact. • Principle nine: A regular review of guidelines is necessary. Therefore, the purpose of this study was to develop guidelines for the adoption of CC in HEIs in South Africa. The key principles identified above guided the development process. The main purpose of this study was to establish guidelines to provide a practical ethical framework for 6 the adoption of CC at HEIs. In this study guidelines were developed that will assist HEIs to adopt CC. 1.2.4 Higher education institutions Higher education institutions play a major role in the growth of the community (Alharthi et al., 2015:2). The South African higher education system has key challenges that they faced, as outlined in the White Paper (Jaffer et al., 2007:132): “to redress past inequalities and to transform the higher education system to serve a new social order, to meet pressing national needs, and to respond to new realities and opportunities”. There are many challenges that affect the adoption of CC in HEIs to allow its significant and fast growth (Attaran et al., 2017:27). Cloud computing is faced with common challenges which are: data protection, data recovery and reliability management, capability, and regulatory and compliance restrictions (Rashid & Chaturvedi, 2019:422). However, they add that HEIs are faced with other challenges, such as the rapid development of technology, political issues, and non-traditional demand. The higher education sector experiences the challenge of limited data storage, which can be easily solved by implementing CC (Yadav, 2014:3110). Higher education institutions should weigh the advantages and disadvantages associated with new technologies, because of limited budgets (Pardeshi, 2014b:591). These challenges make the higher education sector an attractive area of research (Elrehail et al., 2018:55). The adoption of CC in HEIs will maximise the benefits of faster deployment of software and reduce the demand for infrastructure and highly skilled ICT staff (Akande, 2018:6). Higher education institutions can benefit from using CC, as stakeholders will be able to access the service via the Internet, as has been proven during the COVID-19 pandemic-induced lockdown. To mitigate the disruptions to the academic year, cloud services enable students to register online and attend classes virtually (Chen et al., 2020:1). Higher education institutions use SaaS computing model applications, such as Microsoft Office 365 and Google Apps for e-mails and to create documents and spreadsheets (Attaran & Woods, 2019:499). In the context of HEIs, CC can be incorporated into the learning management system, thus simplifying the teaching and learning process (Attaran & Woods, 2019). Figure 1.1 presents all the HEIs-stakeholders that can be inter-connected using CC (Abdulkareem & Ismaila, 2017). 7 Figure 1.1: Education cloud stakeholders Mary and Rose (2020:5476) state that in the world of education, cloud–based learning is a new technology, which provides, creates, and distributes information at any time, at any location. Many HEIs across the world have already embraced CC. The authors further identify the following characteristics of CC in higher education: data storage, sharing, worldwide access and collaborative interaction. 1.3 PROBLEM STATEMENT Rapid changes in IT are affecting the teaching and learning process in the higher education context. Higher education institutions are experiencing problems regarding IT storage infrastructure and affordable education services due to a yearly increase in the number of students and staff. Although CC became an important platform to drive digital innovation in HEIs to improve the quality of education (Al-Shqeerat et al., 2017:22), the rate of cloud adoption in HEIs is very slow, particularly in developing countries (Qasem, Asadi, et al., 2020:1047). Arkorful (2019:5) cites that HEIs are no different when it comes to adoption considerations. From student registration to accessing research data, data traffic has increased exponentially in recent years. Educating, researching, and innovating require tremendous amounts of IT support. As a result, scalability challenges at HEIs can be easily addressed via CC by IT staff or system administrators (Arkorful, 2019:5). To handle the growing demand for data and control costs, it is necessary to find smarter ways. Mitrovic (2017:2) mentions that IT staff experience less stress because of the effective use of CC. In large companies and HEIs, IT systems are typically developed and maintained internally. It is the responsibility of IT 8 departments to manage the control and support of IT services. Therefore, there is a need to focus on adoption guidelines of CC in HEIs. Higher education institutions stand to gain significantly by embracing the CC model for specific tasks. Through resource sharing, including expensive hardware, software, and technical expertise among multiple institutions, IT costs can be minimized. This maximizes resource utilisation and distributes delivery costs effectively (Moloja & Ruhode, 2020:1262). Krauss and Van der Schyff (2014:41) suggest that to utilise CC effectively, there should be trust between the cloud provider and the organisation. According to this study conducted by Krauss and Van der Schyff (2014:41), the issue of adopting CC in HEIs in South Africa was raised with the South African National Education and Research Network (SANReN) by the Department of Arts, Culture, Science and Technology in 2003. SANReN is an organisation that provides the broadband connectivity in all public universities and Technical Vocational Education and Training colleges (TVETs) in South Africa. The final approval for implementation came in 2006, but the actual implementation only started in 2011. The main focus of adopting CC was to ensure that South African universities were abreast with other global institutions of higher education (Krauss & Van der Schyff, 2014:41). The adoption of CC in HEIs in South Africa has been insufficiently investigated although cloud adoption awareness in South Africa has seen some growth, yet actual adoption faces constraints, despite the operations of major global cloud providers like Microsoft, Amazon, and Google in the market (Adendorff & Smuts, 2019). Moloja and Ruhode (2020:1262) also attest to this by mentioning that South Africa currently has 26 public universities that are spread across the country. They further mention that only a few universities have adopted some of the CC services and models. Cloud computing has received significant attention from many industries as a promising paradigm. Globally, CC has been adopted by many business domains due to its rapid growth (Ali, 2020:413). The research on CC primarily focuses on security, since security concerns hinder the use of cloud computing in HEIs. Cloud adoption is slow in HEIs, especially in developing countries, likely due to a lack of clear guidelines and successful adoption examples. Research on cloud adoption in HEIs, particularly in developing nations is limited (Al-Sharafi et al., 2021). Therefore, CC adoption in Higher Education aims to fill these knowledge gaps. This clearly shows that there is still a need to investigate and conduct in-depth research into the guideline’s adoption of CC in higher education in South Africa. 9 Therefore, the research problem in this study was to develop adoption guidelines for CC in South African HEIs. To achieve the objective of this study, the overarching question need to be answered: • What would a set of CC adoption guidelines include? This will be supported by the following sub-questions: o What are the benefits of CC in HEI? o What are the challenges faced by HEI? o What are the recommendations to enhance the adoption of CC in HEI? 1.4 OBJECTIVES OF THIS STUDY The following were objectives of this study: 1.4.1 Primary objective The primary objective of this study was to develop adoption guidelines of cloud computing in South Africa higher education institutions. 1.4.2 Theoretical objectives • T01: To clarify interpretive research as an appropriate methodology for this study. • T02: To elucidate CC by focusing on the CC models and its characteristics. • T03: To pave the way in which HEIs can improve their process about the CC adoption. • T04: To create a mutual understanding regarding the development of guidelines. • T05: To extract a comprehensive list of guidelines from literature. 1.4.3 Empirical objectives • E01: To extract adoption guidelines from the literature review on how HEI can facilitate CC adoption. • E02: To extract adoption guidelines on how HEI can facilitate CC adoption from data gathered and analysed – to make sense of it. • E03: To incorporate adoption guidelines based on the perceptions and experiences of the participants into guidelines extracted from literature on how HEI can facilitate CC adoption. 10 1.5 RESEARCH DESIGN AND METHODOLOGY This study included literature reviews and an empirical study. The interpretive research paradigm was applied to the empirical portion of this study. 1.5.1 Appropriate theory of this study This study focused on the formal logic format. The deductive approach of this study involves utilising pre-established themes identified through a review of literature, along with a theoretical framework constructed through engagement with literature. Additionally, the inductive aspect entails extracting themes directly from the data gathered from the participants (Proudfoot, 2023:311). Research questions were developed based on the literature review that were answered through a hybrid inductive and deductive analysis of the qualitative data as it was gathered through the interviews. Therefore, the hybrid inductive and deductive theory was appropriate for this study. The researcher started with the premise, data that were collected from the IT governance and IT staff through the interviews and subsequently organizing this data into emerging themes. 1.5.2 Research paradigm An American philosopher, Kuhn (1962:158), was the first scholar to use the word paradigm, which in simple terms mean a philosophical way of thinking. He then mentions, “the philosophy and sociology of science cannot be practiced independently of each other”. A research paradigm is concerned with the quest for knowledge. This means that a researcher starts with positive assumptions on how to seek knowledge to be able to address a problem. Positivism is grounded in the philosophical stance of a natural scientist that is focused on the reality which is observable within the society (Alharahsheh & Pius, 2020:41). The positivist approach is focused on the natural sciences, where it seeks information that is based on experiments and observation (Abdulkareem & Ismaila, 2017:28; Roth & Mehta, 2002:133). Positivism relates to the accuracy of data without the influence or bias of humans – it mainly focuses on the facts. Positivism has its philosophy based on the natural sciences (Abdulkareem & Ismaila, 2017:28; Roth & Mehta, 2002:133). Interpretivism is subjective, and its concern is with the variables and factors that are related to context; it reflects humans as different from physical phenomena (Alharahsheh & Pius, 2020:41). Rashid et al. (2019:4) say that through the eyes of participants, the researcher has multiple views of a research problem. They further mention that an interpretivist does not view the world in an objective light. Interpretivism is dependent on a constructivist ontology 11 (Alharahsheh & Pius, 2020), which is concerned with human existence in the world, and how humans are able to acquire knowledge (Moon & Blackman, 2017). Klein and Myers (1999:265) describe critical social theory as a paradigm that starts with the premise that people engage in two different types of activities, which are social interaction and work. In work, everyone and everything can be seen as an object that needs to be manipulated, which means it can also be controlled and predicated. Social interaction is quite different from work, and this deals with mutual understanding in a communicative relation, which means people need to share their views and have others to agree to those views but that does not mean that manipulations does not occur, it alters the way in which the participants relate to each other. There are numerous explanations of critical social theory; however, one of the most cited is that of Myers and Avison (2002:7), and they define critical social theory in the following way: “Critical researchers assume that social reality is historically constituted and that it is produced and reproduced by people. Although people can consciously act to change their social and economic circumstances, critical researchers recognize that their ability to do so is constrained by various forms of social, cultural and political domination.” Vom Brocke et al. (2020:1) define design science research as a paradigm that seeks to enhance human knowledge by creating an innovative artefact in order to solve a particular problem. Design science research paradigm has its root focus in engineering and artificial science. Its aim is to produce knowledge on how things can be re-arranged or designed; usually by a human agency, to accomplish a desired set of goals or achievement (Vom Brocke et al., 2020:1). Hevner and Chatterjee (2010:9) explain design science research as a paradigm which is concerned with the design of an artefact that will be able to resolve its exact problem. This has been supported by Johannesson and Perjons (2014:7), who mention that design science research is a scientific study that is used by people to develop or create an artefact with the aim of solving a practical problem. The four paradigms, which are positivism, interpretivism, critical social theory and design science research, were explained which helps the researcher to choose the suitable paradigm for this study. The positivism paradigm was not an applicable for this study because there would be no testing of a hypothesis of this study by using quantitative methods. Critical social theory was not suitable to this study because it gathers knowledge in which participants express their own views, a self-understanding and self-reflecting process that provides criticism of the existing social knowledge. Design science research was not applicable for this 12 study because this study did not aim to design an artefact to solve the problem. The appropriate paradigm for this study was interpretivism as this study focused on the development of adoption guidelines of CC in HEIs in South Africa. To achieve this, the researcher interacted with the participants. Thus, the research approach to be used in this study was the interpretive paradigm because it is compatible with qualitative methods. 1.5.3 Empirical research According to Myers and Avison (2002:21), a research design is the plan of action of the whole qualitative research project that should involve various components of a research project, such as research method, data collection techniques, analysis techniques, and publishing your findings. As established in the previous section, the empirical study of this research was done in the interpretive research paradigm. The empirical objective was to develop adoption guidelines of CC in HEIs in South Africa. For this reason, IT governance and IT technical staff were interviewed in terms of their technological background knowledge and experience regarding CC. Interviews were used to gather qualitative data. The collected data were analysed and discussed. 1.5.3.1 Research Method The researcher applied case studies (CSs) as a research method in this study. According to Yin (2009:4), a case study (CS) is a research method that can be used in different situations, which helps to enhance our knowledge of organisational, individual, group, political and another related phenomena. He further explains that using the CS approach helps to understand the complexities of such social phenomena. According to Feagin et al. (1991), a CS is defined as “an in-depth, multifaceted investigation, using qualitative research methods, of a single social phenomenon”. They further mention that a CS allows the researcher to investigate not only the complexities of life in which the specific groups of people are implicated, but it also has an impact on the beliefs and complex decisions of social interaction. A CS does not intend to generalise its focus of this study on the entire organisation, but rather focuses on the particular issue to help the researcher to understand the complex real-life activities of a problem in depth (Noor, 2008:1602). A CS has been defined as a research method that focuses on: “… the complexity and uniqueness of a particular project, policy, institution, program or system in a real life context, it can be research based, inclusive of different methods and is evidence-led” (Thomas, 2021:10). 13 Rashid et al. (2019:5) explain that a CS often consists of empirical material collected from detailed investigations over a period of time in order to define a case that will provide the analysis of the context and processes that are involved in a phenomenon. Thomas (2021:3) continues by saying that a CS is a method that provides the complexity of life in the form of inquiry that elevates a view of life. The researcher focused on the development of the adoption guidelines of CC in higher HEIs in South Africa, therefore the focus was on the development of adoption guidelines. According to Yin (2003:360), in CSs are more relevant when there is little knowledge about the topic that has being studied, and this study will be answering the how and why questions. Answering the how and why questions allows the researcher to focus on human behavioural attributes, interactions and actions (Rashid et al., 2019). Hafiz (2008:546) mentions that while considering what the research question will be, it is very important to consider what the case will be. For instance, if this study focuses on the development of adoption guidelines of CC at HEIs, the researcher should be able to answer why and how questions that are related to the adoption of CC, and as such determine the relevant data to be collected during interviews. Many social scientists have the view that CSs may be implemented in three different ways, including descriptive CS for surveys and history, exploratory CS for investigations, and explanatory CS implemented through experiments (Yin, 2009:6). In addition, the CS focuses on primary research and should not be used to test or describe propositions. Typically, CSs are the favoured approach when exploring how or why questions, especially in situations where the investigator lacks control over events, and the emphasis is on examining a contemporary phenomenon within a real-life context (Yin, 2012:4). These explanatory CSs can also be supplemented by exploratory and descriptive variants. Regardless of the chosen type, researchers must meticulously plan and execute CSs to address conventional criticisms associated with this method (Yin, 2012:3). In descriptive research, questions can be answered such as who, what, when, where, and how. It is both creative and analytical (Holmes et al., 2024:51).The researcher used the descriptive CS, this study analysis methods focus more on describing rather than analysing data, since this study focuses on developing adoption guidelines of CC in HEIs. This study utilises descriptive analysis methods to investigate viewpoints of IT governance and IT technical staff on the adoption of CC within HEIs. Qualitative description requires researchers to clearly explain their disciplinary background, why they are interested in the topic, and their assumptions (Bradshaw et al., 2017). According to Myers and Avison (2002:4), research methods can be categorised into qualitative and quantitative. Positivistic study methods include experimental, survey and numerical 14 methods. It can be used in research where this study needs to quantify the problem by generating numerical data that can be transformed into statistics (DeFranzo, 2020). Qualitative research normally falls into the interpretative paradigm, whose key aim is to clarify the subjective details that may originate from social action (Abdulkareem & Ismaila, 2017). In a qualitative research study data have to be collected in order to allow a better understanding of the participant’s perspective (Gelo et al., 2008). Qualitative research methods include case studies, interviews, questionnaires and observations (Myers & Avison, 2002:4). The researcher focused on the CC adoption guidelines in HEIs. Selected IT staff and IT governance from one HEI were interviewed, therefore the CS methodology guided the researcher to investigate the problem in depth, and particularly focus on the adoption guidelines of CC in HEIs. 1.5.3.2 Data collection Paradis et al. (2016) mention that collection of data is very important because it explains how the information collected is used; they further mention that there are five types of data collection, which are surveys, interviews, focus groups, observations and content or textual analyses. According to Sargeant (2012:1), data collection is an approach which is frequently used by qualitative researchers. Qualitative research focuses on an individual or group of people who can be interviewed or observed. Interviews are commonly utilised in interpretive studies as a primary method for accessing the interpretations of informants within the field (Walsham, 2006:323). Interviews are used to collect data that can question a selection of participants from this study and can be analysed across the qualitative approaches used. The saturation of data needs to align with the research questions, theoretical orientation, and analytical framework employed. Nonetheless, it is important to establish a reasonable limit to its application to maintain the coherence and effectiveness of saturation without stretching its conceptualisation (Saunders et al., 2018). As mentioned earlier, qualitative research is synonymous with the interpretative paradigm and its key aim is to clarify the subjective details that may originate from social action (Abdulkareem & Ismaila, 2017). According to Alsaawi (2014:149-151), there are different types of interviews, which are: structured interviews, unstructured interviews, semi-structured interviews and focus group interviews. For this study, semi-structured interviews were selected as the method of choice to gather data. Semi-structured interviews allow researcher to ask detailed questions while ensuring important topics are covered. Semi-structured interviews encourage participants to freely 15 share their thoughts and experiences, leading to richer qualitative data (Adeoye‐Olatunde & Olenik, 2021:1362; Kallio et al., 2016:2960). 1.5.3.3 Participants This study was conducted in South Africa, with a particular focus on one HEI. The participants of this study included IT governance and technical IT staff from one HEI in South Africa. South African universities are categorised into three types: traditional universities, which are academic in focus, universities of technology (previously known as technikons), with a vocational focus, and comprehensive universities that offer a combination of both types of qualifications. The university that was the focus of this study has three distinct campuses aligned in offering academic courses. This study focused on one university campus. IT governance was selected because the adoption of CC will have an impact, which may be positive or negative, on teaching and learning. In addition, the IT staff corporately responsible for the actual implementation of CC was interviewed. The number of participants was determined by the saturation of data gathered through the interviews (Mason, 2010) . The participants were selected voluntarily, both male and female. Informed consent letters were distributed to participants, in time and for them to familiarise themselves with this study and understand the aim of study. 1.5.3.4 Data analysis The phenomenon of this study was facilitated and understood by using qualitative research to obtain results from participants and interpret the collected data (Sargeant, 2012:2). Data collected throughout this study would be meaningless if it was not analysed. Data collected through interviews were analysed using the interpretative approach. Medelyan (2020) mentions that coding is the process of organising qualitative data that help to identify relationships and different themes among them. Furthermore, he mentions that group response, the feedback you receive from a small group of selected participants who contributed to a discussion, is always based on the themes, not on wording. This means a group response with the same themes should be under the same code, but they might not use the exact same wording. There are five different types of qualitative data analysis, which are content analysis, narrative analysis, discourse analysis, framework analysis and grounded theory (Medelyan, 2020). Prasad (2008:1) defines content analysis as a study of communication content that can be described by a scientific study. Mayring (2004:173) mentions that the main objective of 16 qualitative content analysis is to record all communications that have been obtained. The researcher used qualitative content analysis. Qualitative content analysis has been defined as: “a research method for the subjective interpretation of the content of text data through the systematic classification process of coding and identifying themes or patterns” (Hsieh & Shannon, 2005:1278). Zhang and Wildemuth (2005:2) mention that qualitative content analysis starts on the early stages of data collections as it will be able to address the research questions. They further explain that qualitative content analysis should have different steps that the researcher should follow. The researcher followed the following steps to conduct content analysis, which have been mentioned by (Zhang & Wildemuth, 2005:3-5): 1. Prepare the data. 2. Define the unit of analysis. 3. Develop categories and a coding scheme. 4. Test coding scheme on a sample of text. 5. Code all the text. 6. Assess your coding consistency. 7. Draw conclusions from the coded data. 8. Report your methods and findings. 1.5.3.5 Evaluation of the method In a study conducted by Klein and Myers (1999a:72), they mention that in order to confirm the quality of this study, the following principles for an interpretive research study should be followed: The fundamental principle of hermeneutic circle This principle suggests that there is a fundamentals process of interpretation with the aim to determine the objective this study. This first principle gave guidance to the other six principles for a better understanding of this study material (Klein & Myers, 1999:79). This principle of human understanding is fundamental to all the other principles. The understanding is achieved when the researcher iterates between the participants, and as more participants are interviewed, the better the understanding about the aim of the research will be. 17 The principle of contextualisation The principle of contextualisation focuses on (Gadamer, 1976) insight cited by Klein and Myers (1999:73) and mentions that: “… there is an inevitable difference in understanding between the interpreter and the author of a text that is created by the historical distance between them”. Here the data must be understood in a context of why or how it is gathered and from whom, then in this study the findings may be understood, for example why guidelines can affect the adoption of CC in one HEI and why adoption affects its stakeholders. The principle of interaction between the researchers and the subjects This principle requires critical thinking on research materials that involves the participants’ interactions with the researcher. The understanding needs to be established between the researcher and the participants. It further states that in an interpretative research approach both the participants and the researcher could be seen as an analyst and interpreter (Klein & Myers, 1999:74). During the interviews, a clear understanding should be established that the aim of this study is to develop adoption guidelines of CC in HEIs, unlike using assumptions. In this study, the researcher had to be focused on one HEI and made sure that the empirical work was guided by the participants, not the researcher’s own views. The principle of abstraction and generalisation The researcher must be sensitive with the bias inputs, but only focus on the outcome of the data. Klein and Myers (1999:75) mentions that the principles of abstraction with its philosophy says that: “… the validity of the inferences drawn from one or more cases does not depend on the representativeness of cases in a statistical sense, but on the plausibility and cogency of the logical reasoning used in describing the results from the cases, and in drawing conclusions from them”. The researcher must be a fair judge and should refrain from being biased and using false inputs, but only focus on what emerges from the data. 18 The principle of dialogical reasoning The researcher needs to understand and analyse preferred the research approach, that will be able to identify the weakness and strength of the favourite philosophy of this study (Klein & Myers, 1999:79). The researcher needs to be sensitive to the underlying theories as per the research design, that means the theoretical and the actual findings would not be the same. It is the researcher’s responsibility to interpret the findings appropriately. The principle of multiple interpretations The researcher must be sensitive and able to examine the outcome of study that may be influenced by the social context from different perceptions of different participants. Data from participants should not be ignored because of their background; all information is important in order to identify possible supporting factors and contradictions that can be influenced (Klein & Myers, 1999:77). This principle is similar to multiple witness accounts, even if all tell it as they saw it. Participants may have different perceptions, for example every participant may have their own perception and experience about CC. The principle of suspicion The researcher must always be aware of the fact that some of the participants may be biased towards the research study. Some of the participants’ findings might be invalid and this may affect the findings if included (Klein & Myers, 1999:77). 1.5.4 Delimitations to this study This study intended to develop adoption guidelines of CC in HEIs in South Africa. South Africa has 26 public HEIs, but this study was conducted at only one HEI in South Africa, focusing on only one campus. South African universities are categorised into three types, which are traditional universities, comprehensive universities, and universities of technology. The university as subject of this study is a traditional university, with three diverse campuses. The outcome of this study can therefore not be generalised to all HEIs in South Africa and not to other countries. Especially newly launched universities, and universities that are located in rural areas may have different experiences regarding CC. In addition to the above, the lack of utilisation of CC in HEIs, and specifically the university as subject of this study, may limit this study. 19 1.5.5 Expected contribution to knowledge The focus of this study was to develop adoption guidelines of CC in HEIs in South Africa which the South African HEIs sector should take into consideration when adopting cloud services. The guidelines are essential in understanding of CC; to facilitate the adoption of CC. This will enhance teaching and learning through an online platform, for example file sharing, increased data storage and reduced cost. There is little research on the CC adoptions; therefore, the findings of this study will contribute towards the academic literature of CC adoption in South African HEIs. 1.6 LITERATURE REVIEW According to Ishtiaq (2019:40), a literature review must have a central position in this study and should find a gap in research done previously. The literature review chapter provides a description of what other scholars and researchers have published about a specific topic that the researcher focuses on (Taylor & Procter, 2008). The literature review was conducted using various publications, journal articles, relevant textbooks, and the online academic databases. Journals were used through the North-West University library system, such as EBSCOhost, Google Scholar and SA publishers. The keywords used were: • Cloud computing: “cloud computing in higher education institutions”, “cloud computing models”. • Technology in higher education: “technology in higher education”, and “benefits of cloud computing”. 1.7 ETHICAL CONSIDERATIONS Bell and Bryman (2007:71) identify the following most important principles related to ethical considerations: • Research participants should not be subjected to harm or threat. • The participants’ dignity needs to be respected. • Participants should provide a full consent before they take part in this study. • Confidentiality of the research data has to be ensured at all times. • Anonymity of individuals and organisations who are participating in this study must be ensured. • Bias, misleading information and misrepresentation of primary data findings must always be avoided. • Honesty and transparency in relation to the research should be observed accordingly. 20 • Exaggeration about the objective of the research should be avoided. The researcher informed the participants that their responses would be treated with the utmost confidentiality. In addition, the identities of the participants and organisations will remain anonymous. The respondents were informed of their right to withdraw from this study at any time. 1.8 CHAPTER CLASSIFICATION This study comprises the following chapters: Chapter 1 - Introduction and background to this study: The focus is on the background to this study, the problem and research questions and its objectives. The chapter provides an overall detailed background of this study on CC. Chapter 2 - Research design and methodology: The focus is on research design and the methodology used in this study. The chapter discusses the paradigms pertaining to Information Systems, applicable research methodologies, with the focus on interpretivism as the adopted approach to this study, as well as outline to the process of data collection from participants, which includes an HEI’s IT governance and technical IT staff, and the data analysis. Chapter 3 - Literature review: This chapter focus on the literature review and key concepts related to this study. The chapter investigate previous research on the adoption guidelines of CC in HEIs. By analysing previous studies, the researcher was able to extract relevant guidelines. This focus allowed the researcher to compare and establish the findings of other scholars in the field of CC adoption in HEIs. Chapter 4 - Results and findings: The focus is on the discussion of the results and findings of this study. It provides the process used in data gathering, its review and the presentation of the interview results. The researcher formulates guidelines from the empirical data. The results were obtained by analysing the collected data using an interpretative method with the intention to provide adoption guidelines of CC to HEI IT practitioners. Chapter 5 - Conclusion and recommendations: The focus is on the conclusion and recommendations of this study. This chapter presents a complete overview of this study and gives recommendations emanating from this study on how this research may be conducted in the future. 21 1.9 SUMMARY Chapter 1 provided a comprehensive overview of the research, including its background and research objectives. It explains the significant motivations driving the need for this study and outlines the research approach. The introduction also offers a clear definition of cloud computing, the problem statement, and the research questions of this study. Additionally, this chapter outlines the layout of the subsequent chapters in this study. The next chapter presents the research design and methodology of this study. 22 23 CHAPTER 2: RESEARCH DESIGN AND METHODOLOGY 2.1 INTRODUCTION The purpose of this study was to develop guidelines for the adoption of cloud computing (CC) in a South African higher education institution (HEI). The aim of this chapter is to present to the reader, in more detail, the research design used to guide this study. This chapter focuses on achieving the main goal by addressing the first theoretical objective. This chapter examines a variety of philosophical assumptions based on the interpretive paradigm with the aim to focus on the theoretical objective (T01). Researchers who use interpretive paradigms are theoretically able to view the world through the participants' eyes, perceptions and experiences by considering the views of various scholars. Interpretive paradigm investigators construct their understanding from gathered data after using the experiences they have accumulated in seeking answers to research questions (Thanh, 2015:24). Information systems paradigms are discussed in this chapter. This study identified research methodologies that are applicable, with interpretivism as the preferred method, as well as outline the steps involved in collecting data from participants. HEI participants include the information technology (IT) governance and IT technical staff. The case study research method, which was appropriate for this study, is discuss in detail in this chapter. Section 2.2 introduces the research paradigms and provides definitions; it also discusses the different types of paradigms and the suitable paradigm of this study. Section 2.3 explains the research methodology, elaborates the research methods and the case study as an appropriate method for this study. Section 2.4 describes the research design of this study, including the selection of participants, data collection and analysis methods. This section outlines the actual picture of what the researcher did. Section 2.5 is the overall summary of this chapter to summarise the main topic which has been discuss and introduce the reader to the next chapter. 2.2 RESEARCH PARADIGMS Göktürk (2005:1) mentions that the word paradigm has gained a lot of traction in recent years, it has become increasingly dominant in research. Yet the question remains unanswered: What exactly is paradigm and how did it originate? He elaborates that, in many ways, the meaning of the word has escaped philosophers' laboratories, mostly because it has a vague meaning. According to Saleh et al. (2021:22), the word paradigm has been used since the late 15th century, originating from the Greek word Paradigma, which means pattern or model. According to the Cambridge Dictionary as cited by (Saleh et al., 2021:22), the word paradigm 24 means “a model of something, or very clear and typical example of something”. The scientific dictionary also describes it as "the philosophical and theoretical basis of a school of thought or discipline in which theories, laws, and generalizations are formulated and experiments are proposed in support of them". As Guba and Lincoln (1982:233) mention, a paradigm differs greatly on fundamental issues, much more profound than where the inquiry is conducted. According to the researcher, paradigms are philosophical frameworks used to guide research based on deep understanding, beliefs, and ideas. Kroeze (2011:4) mentions that paradigm is used in philosophy of science to describe a set of theories that reflects a historical phase in the development of knowledge. Paradigm contains three main elements, namely ontological, epistemological, and methodological (Johannesson & Perjons, 2014:167; Kroeze, 2011:4). Scotland (2012:9) attests to this by saying these assumptions are based on ontology and epistemology for every paradigm. Rehman and Alharthi (2016:52) cite that positivism, interpretivism, and critical thinking are three types of educational research approaches. Johannesson and Perjons (2014:167) urge that for information systems research in terms of research paradigms, there are two main approaches which are positivism and interpretivism. In addition, they mention additional paradigms which are critical realism and critical theory. Table 2.1 below illustrates the comparison of paradigms within the three philosophical assumptions. The following paradigms will be discussed. 2.2.1 Positivism Kaboub (2008:343) says the positivist paradigm was developed in the 19th century with Auguste Comte's rejection of metaphysics as well as his claim that science alone reveals the truth about reality. Hiller (2016:105) posts that positivism is founded on the principle that knowledge and learning are based on objectivity. Positivism's ontological assumption is that reality has objective properties that can be quantified and that are independent of the observer (researcher) and their equipment (Myers, 1997). Positivism epistemology is that scientists believe that science is based on knowledge and is value-free (Carson et al., 2001:5). Methodology assumptions are that experimentation plays a major role in positivism methodology. In hypotheses, the causal relation between phenomena is proposed as propositions or questions (Rehman & Alharthi, 2016:54). Surveys and experiments are quantitative research methods applied to positivism philosophy (Guba & Lincoln, 1994:110). 25 Table 2.1: Comparison of the paradigms within ontological, epistemological and methodological (Aliyu et al., 2014:81; Aliyu et al., 2015) Analysing Paradigms Positivism Interpretivism Critical Theory Critical realism Ontological (Nature of reality) Established pattern or order based on pre- existing stable patterns. It does not matter when or where reality occurs. It is possible to generalise reality. Complexity and dynamicity characterise the world. In human interaction and social construction, it is constructed, interpreted, and experienced by people interacting with one another and with wider social systems. In a society governed by conflicting structures - political, economic, social, cultural, ethnic, and gender. Societal entities constantly influenced by their own internal forces are realities. A true understanding of reality is only possible through triangulation from numerous sources. (Sobh & Perry, 2006:1195) Epistemological (Nature of knowledge) (Role of theory) Systematic descriptions of knowledge are possible. The concept of knowledge consists of facts or laws that are based on verified hypotheses. Probabilistic – i.e. occurs in many situations and applies to a large group of people. Accuracy and certainty are key characteristics of knowledge. Observable phenomena are not the only basis for knowledge. Knowledge is constructed not just based on objective beliefs, values, reasons, or understandings People make meaning in their lives in different ways, and knowledge is about how they do this, not just that they do it. There is a dispersal and distribution of knowledge. Power is derived from knowledge. The lived experience and the social ties that organise it constitute knowledge. Social and economic contexts help us make sense of events. Results likely true - researcher triangulates perceptions to make sure they are valid. Findings probably true, but mediated by humans; knowledge used to construct theories regarding underlying reality (Wynn Jr & Williams, 2008:3) Methodology (Findings) Focus on experimental research to verify the hypotheses, e.g. survey research, uses the quantitative methods Focus on the hermeneutic principles, Uses the following methods: Grounded theory, ethnography, action research, case study Dialogic/dialectical Critical discourse analysis, critical ethnography action research Case studies and convergent interviews are the main qualitative methods used (Sobh & Perry, 2006:1195) 26 2.2.2 Interpretivism According to Kroeze (2011:2), philosophy of interpretivism emphasises the human construction of reality, which can be understood only subjectively. Humans are only able to perceive an independent, concrete reality through the filters of their sense organs in the absence of an independent and concrete reality. A foundational basis of knowledge is rejected in interpretive research, which casts doubt on its validity. There is no way to compare interpretive research to scientific paradigms (Scotland, 2012:12). Goldkuhl (2012:137) mentions that “Interpretivism is not unified and unequivocal tradition”. Ontological assumption is that interpretivism adopts a relativistic ontological position; our senses give us insight into our real world. Life is meaningless without consciousness (Scotland, 2012:11). Epistemology assumption is that according to interpretive epistemology, the real-world phenomena form the basis of the subjectivism (Scotland, 2012:11). Methodology assumptions are that an interpretive research framework is based on seven principles proposed by (Klein & Myers, 1999:70). It includes the “fundamental principle of the hermeneutic circle, principle of contextualisation, principle of interaction between the researcher(s) and the participants, principle of abstraction and generalisation, principle of dialogical reasoning, the principle of multiple interpretations and the principle of suspicion”. An interpretative research paradigm is typically used in qualitative research and aims to clarify the subjective details that may stem from social action (Abdulkareem & Ismaila, 2017). 2.2.3 Critical social theory According to Ngwenyama (1991:2), critical social theory focuses on improving the human condition as its primary objective. The goals of critical social theory, cited by Ngwenyama (1991:2) at the opening of the Institute for Social Research were as follows: “On the contrary, according to the critical theory of society (as opposed to the positivist view of social science), men are historical forms of life themselves. Assuming the conditions of reality from which science starts out, it does not rely solely on probabilistic laws for establishing and calculating its intentions. In each case, the outcome depends not only on nature, but also on what men choose to do with it.” Ontological assumption is that a critical paradigm assumes historical realism as its ontological position (Scotland, 2012:13). Reality is the product of constant internal influences that are constantly constructed by individuals. Critical epistemology is based on societal ideology; it is 27 a subjective view that references real-world phenomena. Knowledge is permeated by social relations and is socially constructed (Scotland, 2012:13). Methodology assumptions are that theoretically, critical scholars presume social reality as historically constituted and that it is manufactured and reproduced by humans (Myers, 1997). 2.2.4 Critical realism Archer et al. (2016) cite that the principles of critical realism include a set of philosophical positions on matters such as causality, structure, people, and modes of explanation. In addition, critical realism emerged in response to post-positivist crises in the social and natural sciences of the 1970s and 1980s. According to critical realism, the evidence around us can serve as an accurate glimpse of reality; however, it is subject to fallibility, social relationships, and subjective interpretations (Sturgiss & Clark, 2020:143). The goal of this alliance is to establish a proper postpositivist social science (Archer et al., 2016). Ontological assumption is that it observes the world as it is, regardless of how people perceive it. Cause-and-effect causal structures may not always be visible until they are activated in certain circumstances because the world is a multidimensional system (McEvoy & Richards, 2003). Epistemology assumption is based on observations and interpretations of participants' experiences according to critical realism, the world is described in a way that reflects it (Wynn Jr & Williams, 2012:793). Methodology assumptions, according to Wynn Jr and Williams (2012:795) state that critical realism is driven by the goal of developing explanations for how things act and behave. 2.2.5 Paradigm appropriate for this study The purpose of this study was to develop adoption guidelines of CC in a HEI in South Africa. The four paradigms, positivism, interpretivism, critical social theory, and critical realism, were discussed and explained according to epistemological, ontological assumptions and methodological considerations. The paradigm relevant to this study was chosen based on the epistemological and ontological assumptions. According to Klein and Myers (1999:67), in the field of information system, interpretive research has the potential to assist researchers in understanding the behaviour of individuals in social and organisational contexts; it can also help to develop new insights into information systems phenomena such as information system management and information system development. Through the meanings that individuals assign to phenomenon, it attempts to facilitate understanding of individuals (Klein & Myers, 1999:69). 28 This study did not fit the positivism paradigm because it was not objective, and there was no hypothesis testing. Positivism holds that the truth exists independent of external influences. In this study, critical social theory was not applicable as participants are empowered through the emancipation process. In critical research, reality is assumed to be incomplete, and its cultural studies focuses on action research. Although if applied both qualitative and quantitative, it mostly uses participants’ observation, or triangulation methods. Critical realism involves quantitative approach for data collection which was incompatible with the purpose of this study as critical realism treats the world as theory laden, but not as theory determined. Research paradigms are determined by their epistemological characteristics and ontological assumptions. The constructivist epistemology of qualitative research examines what it perceives to be a dynamic reality created by society through a holistic, descriptive, value- laden, and contextual perspective; in other words, from the people's perspective (Yilmaz, 2013:312). Conducting interpretive research within a contextual setting enhances the depth of understanding and knowledge of phenomena. This approach facilitates the interpretation of the meaning’s participants attribute to their expressed views, as the research occurs within their social environment (Fossey et al., 2002:727). A friendly environment enables participants to articulate their experiences more readily. Fostering such an environment strengthens the relationship between the researcher and the participant, emphasizing the socially constructed nature of reality (Fossey et al., 2002:720). This is in line with an interpretive approach to the main purpose of this study to develop adoption guidelines of CC in South Africa HEIs. The next section examines the research methodology that we used for this study, considering the research paradigms that have been discussed. 2.3 RESEARCH METHODOLOGY Research methodology relates to the general research strategy used for conducting research, which will identify what methods to employ and how those methods will match the general research strategy (Alharahsheh & Pius, 2020:40). Research methodology does not prescribe a particular method for use, but rather emphasises the significance of the process pursued to achieve the objective of this study (Alharahsheh & Pius, 2020:40). Singh (2006:79) cites that a research methodology describes how a researcher moves from identifying a problem to arriving at a conclusion. By defining a methodology, you ensure that research work is conducted in a valid and scientific way (Singh, 2006:79). He adds that methodology refers to the techniques and tools used in solving the research problem. For a study to be successful, there must be correct procedures, principles, and techniques. The perspectives and views of information technology (IT) governance and IT technical staff provided guidance in addressing the main objective of this study – to develop CC guidelines 29 in a South African HEI. A review of these perspectives and views led to the development of guidelines for CC adoption in HEIs in South Africa. As part of the research, the researcher evaluated the participants' narratives considering the literature's context. 2.3.1 Principle for Interpretative research For conducting and evaluating interpretive studies, researchers can use a set of seven principles, as proposed by Klein and Myers (1999:70) illustrated on Table 2.2. As part of this research, it provides support. Table 2.2: Summary of the principles of hermeneutics and its application to this study Principle Descriptions 1. The fundamental principle of the hermeneutic circle The principles advocates that all human understanding is attained by iteration of the interdependent meaning of parts and considering the whole meaning that the parts form. This principle of human understanding is fundamental to all the other principles. 2. The principle of contextualisation It requires critical reflection of the social and historical background of the research setting. This is to help the intended audience to see how the current situation under investigation emerged. 3. The principle of interaction between the researcher(s) and the participants The principle requires critical reflection on how the research data were socially constructed through the interaction between the researcher(s) and participants. 4. The principle of abstraction and generalisation It requires relating the idiographic details revealed by the data interpretation. This is done by applying principles of the hermeneutic circle and of contextualisation to theoretical, general concepts that describe the nature of human understanding and social action. 5. The principle of dialogical reasoning The principle requires sensitivity to possible contradictions between the theoretical preconceptions guiding the research design and actual findings with subsequent cycles of revision. 6. The principle of multiple interpretations It requires sensitivity to possible differences in interpretation among the participants and are typically expressed in multiple narratives or stories of the same sequence of events under study. This is similar to multiple witness accounts even if they all tell it as they saw it. 7. The principle of suspicion This principle requires sensitivity to possible prejudices and systematic alterations in the narratives collected from the participants. The following principles will be discussed. 2.3.1.1 The fundamental principle of the hermeneutic circle A key principle in the philosophy of language believes that all human understanding comes from iterating the interdependence of parts and evaluating the whole meaning that is formed from the parts. Every other principle is grounded in this principle of human understanding 30 (Klein and Myers (1999:71). The idea of a hermeneutic circle implies that we can understand a complex whole based on our preconceptions of its parts, as well as the relationships between them. The main aim of the researcher in this study was to develop guidelines of CC in HEIs in South Africa. The understanding was interpreted according to the participants’ views. Understanding the questions is developed based on each participant's response. After understanding is gained, each participant interprets that understanding from their own perspective (Gill et al., 2008:291). Every interview followed the same principle, which applied to each individual answer and to the entire interview. New interviews were handled in the same manner, while also seeing them as a part of the entire process. 2.3.1.2 The principle of contextualisation Gumperz (1992:39) mention that by integrating broad ethnographic perspectives with specific conversation-analytic insights, he developed the concept of contextualisation to elucidate how individuals construct meaning in their interactions. Using contextualisation, the intended audience can see how the current situation under investigation evolved by placing the subject matter in its social and historical context (Klein & Myers, 1999:73). By investigating and collecting additional data related to the background of the participants, the researcher obtains more historical understanding. Talking to participants in interviews, making sure they understand and can explain their answers, helps the researcher to understand things better. 2.3.1.3 The principle of interaction between the researchers and participants Researchers who adhere to this principle must place themselves and their subjects in a historical context. Social researchers do not just gather data like rocks at the shore. The data are all around. Interaction between the researcher and the participants, rather than fact, is the means by which the facts are produced (Klein & Myers, 1999:74). In the research process, the researcher must consider how social experiences are constructed through interactions with participants. Therefore both researchers and participants are considered important interpreters and analysts of data (Klein & Myers, 1999:74). This study involved interviews as the data collection method. Participants were given the opportunity to give full and open answers during an interview to get unbiased feedback. If the participants gave unclear responses, the researcher followed up on the same question. 31 2.3.1.4 The principle of abstraction and generalisation The previous two principles focus on those aspects that are specific to the situation under study (Klein & Myers, 1999:75) Interpretive research values documenting unique situations but questions the idea of universally applicable laws governing human behaviour across cultures. However, that is not the complete picture. Interpretive field studies have been abstracted and generalised by philosophical debates (Klein & Myers, 1999:75). In this study, to aid understanding of the IT staff and academics perspectives about the CC, we had to apply logic to the gathered data. Qualitative data drive interpretive research, but quantitative data can support correctness and greater clarity of what the researcher is trying to convey, i.e. to abstract/generalise what they are seeing. Both IT staff and academics had an opportunity to offer their perspectives of CC in HEI, but it was also significant to collect all the information for purposes of identifying a common challenge. 2.3.1.5 The principle of dialogical reasoning Using the data gathered from their research, researchers must confront the preconceptions (prejudices) which influenced their original research design (i.e. lenses through which they interpreted their research) (Klein & Myers, 1999:76). They add that as the researcher, it is essential to make the historical intellectual basis (i.e. the underlying philosophical assumptions) of this study as transparent as possible to the reader. The hermeneutical method assumes that prejudice is the basis of interpretive understanding, but researchers must differentiate true prejudices that enable understanding from false prejudices that bring misinterpretation (Klein & Myers, 1999:76). There are several voices of participants, and these voices can complement and contradict each other, as well as the literature. As part of the analysis and interpretation process, the researcher made sure that all views and expressions of the participants were considered. 2.3.1.6 The principle of multiple interpretations Klein and Myers (1999:77) mention that by seeking out and documenting multiple perspectives with their reasons, the researcher can examine the influence that the social context has on actions under study, according to the principle of multiple interpretations. The evaluation of reasons may be influenced by issues related to power, economy, and values. In addition, researchers must examine the contradictions inherent in multiple viewpoints themselves, in order to revise their analyses accordingly (Klein & Myers, 1999:77). For this study, participant interpretations could vary, even when participants relate the same information which is related to the adoption of CC in HEIs, therefore multiple participants will differ as well. Multiple perspectives of the same event should be considered. Research 32 findings will reflect this complexity of voices. In this case, the context can be important. It is quite common for different participants to answer the same question differently due to their subjective perception and observations of the situations. Views about one event may be offered from multiple angles. To interpret multiple concepts derived from their empirical findings and the literature, the researcher must condense the findings and theory into a unified picture. 2.3.1.7 The principle of suspicion Klein and Myers (1999:77) cite that in this approach, the analysis of the data is clearly not sufficient to explain its significance. In this way, the researcher can understand and read the society behind the words of the actors. In this social world, there is power structure, vested interests, and limited resources to meet the goals of various actors who engage in the construction and implementation of this social world. Critical thinking is used to collect and interpret data, interpret results, and write reports. This ensures a contextual assessment of the whole process. During data collection, the researcher is required to act with extreme understanding to avoid modifying the data collected from the subjects. Considering all contributions rather than focusing on one specific involvement alone, the hermeneutic circle principle helps in determining the understanding of each involvement in context with the broader discourse. 2.3.2 Theoretical grounding Research approaches are strategies and methodologies that cover everything from general hypotheses to specific techniques for gathering, analysing, and interpreting data (Gabriel, 2013). Hyde (2000:83) supports this by saying that inductive reasoning and deductive reasoning are two general approaches for reasoning that allow for new knowledge to be acquired. Johnson-Laird (1999:110) urges that in reasoning, perceptions, ideas, and assertions are used to draw conclusions. Goel et al. (1997) mention evaluating arguments as a way of reasoning. Arguments give some reason for accepting another proposition based on one or more propositions. New knowledge can be acquired by both inductive reasoning and deductive reasoning, two general approaches to reasoning. A theory is built by inductive reasoning by first observing specific examples and then establishing generalisations about them. Deductive reasoning is the process of determining whether a widely accepted theory can be applied to a specific situation based on an established theory or generalisation (Heit & Rotello, 2010). A comprehensive approach to qualitative data analysis can be achieved by 33 applying both deductive and inductive approaches. To make sense of data and understand what's happening, one must immerse themselves in it by reading and digesting (Kumar & Ujire, 2024). 2.3.2.1 Inductive Hyde (2000) defines inductive approach as a method of building theory starting by observing specific instances and then attempting to generalise the phenomena under investigation. DeCarlo (2018) explains that an inductive technique requires a researcher to start by gathering information that is pertinent to their area of interest. After gathering a sizable amount (Streefkerk, 2023) of data, the researcher will pause data collecting to stand back and gain a bird's eye view of their data. At this point, the researcher examines the data for patterns while trying to create a hypothesis that may account for those trends. Streefkerk (2023) adds by saying when there is little to no existing literature on a topic, it is common to perform inductive research, because there is no theory to test. The inductive approach consists of three stages: Observation, seeking patterns and developing a theory or general (preliminary) conclusion (Streefkerk, 2023). Kumar and Ujire (2024:60) agree by saying in an inductive research approach, the researcher gathers relevant data on the topic. After collecting a lot of data, they take a step back to see the big picture. At this point, they try to come up with a theory that explains the patterns they have found in the data. 2.3.2.2 Deductive Deductive research involves reading existing ideas about the subject being studied, reviewing what others have done, and then testing hypotheses that result from those theories (DeCarlo, 2018; Streefkerk, 2023). Essentially, deductive reasoning uses established theories to test whether they apply to a specific situation based on the generalisations and hypotheses (Hyde, 2000:83). A deductive approach may be a form of scientific investigation. In deductive reasoning, it is impossible to accept the premises and reject the conclusions. As a starting point, the deductive approach to accounting theory st