Factors Influencing the Success of Online Education during COVID-19: A Case Analysis of Odisha

Abstract

The COVID-19 pandemic caused by the coronavirus has dramatically changed the lives of students all around the world, with the virus’s effects profoundly impacting students’ physical and emotional well-being. Due to a series of shutdowns and lockdowns, social distancing, and further closure of schools, colleges, and institutions to ameliorate the pandemic crisis, the teaching and learning process shifted to an online form. As a result, students all over the world have been forced to deal with the problem as a last resort to accepting online education. This study looked at the efficiency of online education in the current situation and the student’s reactions. To enhance the online method of education for students, we examined the success characteristics of online education in the Indian state of Odisha. The study’s samples were collected from the faculty members of various graduate and post-graduate educational institutions in Odisha, who were recruited by questionnaire to get an expert opinion.

1. Introduction

The recent COVID-19 pandemic outbreak in Wuhan city in December 2019 not only disrupted the course of history but also altered the way we live today (Stratton et al. 2020). Over 170 nations have been impacted by it, including the US, the UK, Italy, Pakistan, and many others. Given the interconnected and linked nature of the modern world, the COVID-19 pandemic has impacted several vital sectors, with the education sector being one of the worst affected (McKibbin and Fernando 2020). Since the spread of this pandemic, its economic effects have significantly altered both professional routines and the operational working patterns of numerous sectors, causing great concern worldwide (Wang et al. 2020). COVID-19 has caused disruptions in practically all facets of existence. Both developed and developing countries are currently looking for novel approaches to manage the core business operations of each impacted industry. Further, estimation of the end of COVID-19 appears to be unachievable, which has thus caused colleges and secondary school institutions around the world to become concerned (Crawford et al. 2020).

The COVID-19 pandemic has created stress among students regarding their academic continuity. Thus, educational institutes have been compelled to adopt emergency remote teaching (ERT), a temporary and quick instructional response to the pandemic to allow students to continue their studies. However, educational institutions have opted for online teaching processes while considering the duration and frequency of the pandemic. Online teaching pedagogy is not a quick solution as it requires time, money, training, and infrastructure for the design and implementation. Hodges et al. (2020) highlighted the difference between “Emergency Remote Teaching’ and ‘Online Learning’ and suggestions to help institutions implement temporary solutions, such as emergency remote teaching or permanent online learning based on the strengths and weaknesses of their organizations. This has sparked a wave of creative solutions, most notably in the educational field, to mitigate or lessen the impact of this pandemic on students and their academic pursuits. The industrialized nations have evolved with numerous effective techniques to manage their educational setup, particularly when we speak about the effects of the COVID-19 pandemic on the education industry (Ukpokodu 2021). Whereas underdeveloped nations lag far behind in managing their education sectors effectively due to a lack of money, experience, and inadequate infrastructure (Liguori and Winkler 2020). Although many private universities now use technology to deliver education and management internet services, they are now aware of several problems with e-learning. In contrast, many public educational institutions around the world are still establishing their digital platforms and working to remove the obvious drawbacks of the online educational system. Technology adoption in the education sector is not a novel idea because it has previously been practised in numerous ways. Examples of modern computer educational technologies include web-based learning and other applications. Nonetheless, with the current global catastrophe, there is a significant rise in the demand for technology and digital networks for education. Technology not only thrives on new teaching methods but it also improves the caliber of educational experiences.

Today’s global educational climate is extremely competitive. The growth of Internet-based computerized learning, specifically online learning, is the primary emphasis of all educators in this period of widespread digitalization (Goodman et al. 2018). Only when virtual education is supported and tailored can it be optimized (Gell-Mann 1996). According to the definition, online learning is a combination of several technologies (Wu et al. 2010), with computers serving as the system’s most crucial and potentially life-saving component. Computers are excellent at keeping track of our steady development and alerting us to any errors in reasoning so that we can remedy them (Hayashi et al. 2004). Therefore, while adopting virtual or digital education, the accessibility of computer systems is essential.

Additionally, online education is a thorough approach to learning that gives students the freedom to choose their preferred learning style (Hasnat 2009) and can support the growth of both personal and social knowledge in any nation. Good education and active learning through online courses appear to be crucial success criteria in our digital age (Siddiqui et al. 2019; Vohra 2013). A learner can build all these abilities in a virtual setting. As digital training involves the transmission of both internal and external content via information systems to improve the quality of education and the cognitive ability of students, consequently, it can directly alter students’ work habits.

The education system, the backbone of our country, has been impacted by war, riots, natural disasters, weather conditions, terrorist operations, and now COVID-19, which has already spread over to 100 nations worldwide, making it a pandemic. COVID-19 is rapidly claiming lives, and one of the preventative measures to avoid the virus is social distancing, for which the government has declared lockdowns and shutdowns to prevent the virus from spreading. To avoid social gatherings on the educational platform, all schools and colleges have been forced to close remain for indefinite periods. All courses, tests, and academic activities have been halted, and the epidemic has also affected the students. The effects of the coronavirus pandemic can be seen all across the world, with 98.5 percent of the world’s student population affected by the disrupted educational system. As the circumstances demanded and to preserve continuity in education, online education has sprung forward in full force to help students advance their careers while also keeping up with the educational industry’s competitiveness. From the month of July or August, all educational establishments in India have begun providing online education to their respective students.

2. Literature Review

Volery and Lord (2000) conducted a survey of students enrolled in an online management course at an Australian university and identified three success factors for online education: technology, the instructor, and the previous use of technology from the perspective of students, with an emphasis on the instructor as the central role for online education success. Puri (2012) outlined the essential success elements for online distance learning in higher education institutions, highlighting five variables for improving online education efficiency: learning environment, institutional administration, course assessment, services sector, and instructional design. Rohayani (2015) discussed the idea of e-learning-ready elements, as well as the e-learning variables identified by earlier researchers in higher education. He concluded that the most significant elements influencing e-learning readiness are skills and attitudes. The influencing variables and their impact on interactive e-learning were emphasized by Wan (2016). In his research article “Impact of online education in India”, Zahoor (2017) discussed the fledgling state of digital education in India, as well as how the combination of learning services and technology will deliver high value in the Indian context. Ch and Popuri (2013) emphasized the introduction of a variety of new portals for student usage, particularly the advent of edX as a worldwide learning platform for free online education. Pham et al. (2019) conducted a study on the e-learning service quality by polling 1232 college students on three factors: e-learning system quality, e-learning administration and support service quality, and e-learning teacher and course materials quality. He emphasized that the quality of the e-learning system is the most significant aspect of student happiness.

Fazil and Kesavan (2014) and Rajesh et al. (2013), identified the elements that are causing the issue, establishing a contextual connection between the variables. The structural self-interaction matrix is being developed (SSIM), which is used to create the initial reachability matrix. The finalization of the reachability matrix is the calculation of the driving forces and its dependency is divided into several tiers: the formation of a digraph, or driving power-dependence diagram, as well as a multileveled hierarchy model based on the interpretive structural model (ISM). The three dimensions of the interpretive structural model (ISM) are described by Jacob and Pramod (2013) as “I” for interpretive, “S” for structural, and “M” for modeling. Interpretive is the judgement of specialists in the same area, structural is the creation of links between the factors, and modeling is the presentation of the structure based on the significance of factors. According to Gorvett and Liu (2007), the interpretive structural model (ISM) is a computer-based technique for users dealing with complicated circumstances to determine the link between the elements. The current use of ISM in research articles, such as supply chain management, risk control decision, and mobile banking, was discussed by Jayant et al. (2015). Sreejith (2012) used the ISM methodology to provide a solid foundation for implementing green supply chain management in the construction industry. After evaluating the inter-relationships between the elements, Ambika et al. (2013) presented the interpretive structural model (ISM) to determine the factors that impact cloud computing.

Gamage et al. (2020) emphasized academic integrity in terms of delivery practices and assessment practices in universities as a challenge for the primary education sector because students are overly reliant on teachers and secondary and higher secondary students face the challenge of completing curricula and preparing for final examinations in a short period. Many colleges have advanced their students to the following semesters without using an appropriate assessment method for student evaluation. Cassibba et al. (2021) emphasized the difficulties faced by university professors while teaching mathematics through distant or online modes as this style of instruction necessitates face-to-face contact in which the instructor must provide symbols and formulae to illustrate the issues. Osman (2020) emphasized COVID-19’s global influence on education systems, as well as its impact on Sultan Qaboos University. He discussed how the university used e-learning ‘Emergency Remote Teaching’ (ERT) to meet the problems of giving equitable access to all students in remote regions where the Internet connection may be a barrier to the implementation of Emergency Remote Teaching. As a consequence of the COVID-19 pandemic, Tarkar (2020) emphasized the issues encountered by students, instructors, and parents as a result of the shift in teaching style from the offline mode to online mode. According to Partlow and Gibbs (2003), online programs should be innovative, interactive, relevant, student-centered, and focused on group-based education, with teachers making continual efforts to enhance the program. The influence of the COVID-19 epidemic on entrepreneurial education has been emphasized by Liguori and Winkler (2020), leading to the development of new resources for the online method of teaching and learning. Martin (2020) offered five essential issues for educators to consider while teaching in an online manner. The first is ‘instruction’, which may refer to load reduction instruction in clear, well-structured and manageable deliveries to give an opportunity to both students and teachers; the second is ‘content’, which should be appropriate for the level of knowledge and competence of the students. The third is’ motivation’, which refers to the motivation of a learner’s energy and effort to engage positively in the learning process. The fourth is ‘connection’, in which the interpersonal interaction between instructor and student in a peer group is extremely important to the online mode learning process. The fifth issue is ‘mental health’, as it is difficult to give aid to students suffering from mental health issues over the internet; however, this may be solved by maintaining closer contact with students to ensure ongoing moral support. Singh and Thurman (2019) conducted a systematic assessment of the literature from 1988 to 2018 to look at the definitions of online learning and the content analysis of these concepts. “Learning experiences in synchronous or asynchronous contexts using various devices such as mobile phones and computers with internet access to engage with instructors to learn” is how online learning is described. Song et al. (2004) identified the problems of online learning from the perspective of the student, including how course design, learner motivation, time management, and user-friendly technology influence the learning experience. Students believe that technology issues and a lack of knowledge of instructional goals are stumbling blocks in the online learning process. Pitambar Paudel (2021) highlighted the benefits and challenges faced by the students of higher educational institutes in Nepal by comparing online education with traditional education. The author also discussed the strategies during and after COVID-19 and suggested that the blended mode of education would be more effective, as opposed to going only for online education or only education in the traditional approach.

COVID-19 taught us that the use of the Internet for the online classes should be the final option employed to maintain continuity in the education industry. Technology development and the Internet increased the easy access to the online session, though there are disadvantages to it. It also increased the likelihood of students’ being academically dishonest. Students have often been involved in unethical practices and witnessed academic fraudulence in this online era of teaching-learning pedagogy. Serkan et al. (2012) highlighted such e-dishonesty at Midwestern University, where they developed a relationship between the involvement of e-dishonesty and the rationale for e-dishonesty. Seife and Stockton (2020) suggested that academic dishonesty can be mitigated using the proctoring system as online academic dishonesty decreased after the implementation of the proctoring system as per their research analysis. Yazici et al. (2023) compared face-to-face cheating and online cheating among Turkish university students during the pandemic time concerning the pre-pandemic period.

3. Research Gap

Based on the research paper, idea, and subsequent analysis done on the basis of reviewing the related research papers on online education during the COVID-19 pandemic, it has been found that the issues like benefits, challenges, impact on students, student satisfaction and dissatisfaction, comparison between the online education with the traditional teaching pedagogy, and many more issues highlighted by a number of authors over the time. The research work, which is still ongoing, aims to find out the solutions to the problems in the teaching–learning process. In the State of Odisha in particular, it was found that the research work on the online education system during the pandemic in educational institutes was much less. Also, few research studies have aimed to identify the success factors for online education using interpretive structural modeling (ISM) in Odisha State. This analysis will help the researchers to further explore this area of concern to acquire the solutions and suggestive measures to improve the online teaching pedagogy.

4. Methodology

Based on the research gap identified from the literature review, the following objectives were decided.

• To identify the variables affecting online education.
• To develop the contextual relationship between the variables.
• To identify and rank the driving power and dependence power of the variables.
• To develop the multilevel hierarchy model of the variables based on their driving power and dependence power.

Based on the objectives, we have applied the interpretive structural model (ISM). In this approach, we need at least eight expert opinions as an essential part of this technique (Warfield 1974). The ISM approach developed the contextual relationship between the variables using management techniques including brain-storming and nominal group techniques (Kannan et al. 2010). To collect expert opinions to understand the success factors of online education and its importance, we designed one questionnaire (Appendix A), which was circulated to 150 faculty members of different higher educational institutes (7 universities and 13 standalone colleges). Out of these, 96 questionnaires were returned by the respondents and 45 questionnaires were found complete. We selected 12 responses (7 senior professors and 5 professors with additional administrative responsibilities) based on their knowledge and duration in the academic field. We analyzed all the 12 responses collected and a convergence in the kind of relationships between the variables were identified. Finally, the results were discussed with these 12 experts to draw the final judgment. The data was collected in the month of July 2020, when the educational institutes implemented online education as the last resort.

5. Interpretive Structural Modeling (ISM)

We used the interpretive structural model (ISM) for the identification of the factors influencing the success of online education as it is a powerful mathematical-based qualitative tool, introduced by John Warfield in 1973. ISM model can be used by both individuals and groups to solve problems in complex situations (Lin and Yeh 2013). ISM is used to show the inter-relationship between the factors and develop a hierarchy model and graph (Ravi and Shankar 2005). ISM is used to develop the specific relationship between the factors of interest and to develop the graphical model or diagraph (Sage 1997). ISM provides a direction, realistic decisions, and an ordered framework to the complex situation faced by the researchers by involving the variables in analysis (Rajesh et al. 2013). Authors have used the ISM model to analyze different complex problems; for example, Mandal and Deshmukh (1994) used this model to select the vendors based on some criteria and explained the inter-relationships between these criteria. Singh et al. (2003) have used the ISM in the engineering industry for the successful implementation of knowledge management. Ravi et al. (2005) used this ISM model in the reverse logistic supply chain management in the computer hardware section to improve their performance. Sharma et al. (1995) used the ISM model in waste management and the action plan was designed to develop the hierarchy to solve issues regarding the future of the waste management in India. Thakkar et al. (2008) used the ISM in supply chain and compared the relationships in small and medium-scale enterprise (SME) sectors. Meghraj and Sridhar (2016) used ISM to analyze the factors involved in the successful implementation of total productive maintenance in a machine shop.

In this analysis, we aim to identify the relevant factors of online education and will prepare the structural self-interaction matrix (SSIM), the initial reachability matrix, the final reachability matrix, the driving power and dependence power matrix, the level of partition, the driving power and dependence power diagraph and multilevel hierarchy model. These are the steps for this ISM approach.

6. Case Analysis

In our study, we focused on ten elements that influence the success of online learning in the educational institutions. The study is based on the educational establishments in Odisha and the criteria were considered after consulting with the professionals and the academics from the state. Some of these factors are listed as follows:

○

Effective team work (ETW): Effective teamwork in the workplace, with both teaching and non-teaching employees working together to enable online education, can provide positive results. There will be no coordination between the staffs if they do not work together, which will result in a futile attempt.

○

Infrastructure and technology (IT): The necessary infrastructure and up-to-date technology will aid in the seamless and effective delivery of online programs.

○

Cost of project (COP): The success of the program will be determined by the project’s cost, as providing these services to the students requires the necessary infrastructure, technology, and trained staff.

○

Internet connectivity (IC): For students to participate in online classes, they must have access to the internet. It is difficult to hold online lessons without any access to the internet.

○

Power supply (PS): To operate the online classes efficiently, power is necessary to support the process and technology. The internet connections and other electrical devices needed for online classes will be made more accessible with a reliable and consistent power supply.

○

Trained manpower (TM): A trained workforce will boost the organization’s production by delivering results that are based on their abilities and knowledge. The success of online education will be aided by well-trained personnel who will make the process and operations run smoothly.

○

Student participation (SP): Students will be encouraged to participate by providing all of the necessary amenities. Students will be interested in enrolling in the online programs if the authority and skilled personnel of the organization provide support in all the areas.

○

Method of teaching (MOT): Offline classes and online classes have distinct teaching methods. The teaching method in online classrooms should be developed to encourage students to participate, and this is dependent on the availability of skilled people and physical facilities, as well as the organization’s support.

○

Organization support (OS): It is a critical component for executing any type of program in the organization since we cannot achieve the success that we expected and planned without the support of the organization, which includes the support of the authority and top-level management. As a result, government funding can facilitate other variables such as project cost, skilled labour recruitment, internet services, and other physical infrastructure.

○

Effective study environment (ESE): To ensure the success of any program, we must establish a conducive learning atmosphere in which everyone is eager to share and receive the information. With the aid of modern technology, the authorities and experienced personnel, this atmosphere can be established.

Structural Self-Interaction Matrix (SSIM). We utilized statistical tools to rank the elements in order of relevance in our research. We utilized the interpretive structural model (ISM) approach to conduct our research, considering the ten criteria listed above. The structural self-interaction matrix (SSIM), shown in

Table 1. Structural self-interaction matrix (SSIM).

	1-ETW	2-IT	3-COP	4-IC	5-PS	6-TM	7-SP	8-MOT	9-OS	10-ESE
1-ETW	X	A	O	A	A	A	V	V	A	V
2-IT		X	X	V	A	V	V	V	A	V
3-COP			X	X	X	V	O	V	A	O
4-IC				X	A	O	V	V	A	V
5-PS					X	O	V	V	A	V
6-TM						X	V	V	A	V
7-SP							X	O	A	A
8-MOT								X	A	V
9-OS									X	V
10-ESE										X

(Source: Survey data and author’s calculation). The above matrix is a symmetrical matrix.

, is used to determine the contextual connection between the variables. To establish a link between the variables, the ISM model employs management approaches such as brainstorming and nominal group methodology. The direction of the relationship between the variables i and ‘j’ is represented by four symbols (Sudarshan and Ravi 2013). A structural self-interaction matrix (SSIM) is prepared to measure the contextual relationship among the variables/factors by comparing the variables by assigning one possible relationship (i.e, ‘V’, ‘A’, ‘X’ or ‘O’).

The symbols are as follows:

• V—Factor ‘i’ influences factor ‘j’ but factor ‘j’ does not influence factor ‘i’.
• A—Factor ‘i’ does not influence factor ‘j’ but factor ‘j’ influences factor ‘i’.
• X—Factor ‘i’ influences factor ‘j’ and factor ‘j’ influences factor ‘i’.
• O—There is no relation between ‘I’ and ‘j’.

7. Initial Reachability Matrix

The structural self-interaction matrix (SSIM) is transformed in the format of the reachability matrix by transforming the information in each entry of SSIM into ‘1s’ and ‘0s’ in the reachability. The initial reachability matrix is designed by using some rules, which can be obtained from the SSIM (Table 1), regarding the pair-wise comparison of different factors. The initial reachability matrix is shown in binary form i.e., 1 and 0 (

Table 2. Initial reachability matrix of SSIM.

	1-ETW	2-IT	3-COP	4-IC	5-PS	6-TM	7-SP	8-MOT	9-OS	10-ESE
1-ETW	1	0	0	0	0	0	1	1	0	1
2-IT	1	1	1	1	0	1	1	1	0	1
3-COP	0	1	1	1	1	1	0	1	0	0
4-IC	1	0	1	1	0	0	1	1	0	1
5-PS	1	0	1	1	1	0	1	1	0	1
6-TM	1	0	0	0	0	1	1	1	0	1
7-SP	0	0	0	0	0	0	1	0	0	0
8-MOT	0	0	0	0	0	0	0	1	0	1
9-OS	1	1	1	1	1	1	1	1	1	1
10-ESE	0	0	0	0	0	0	1	0	0	1

(Source: Survey data and author’s calculation).

). The rules can be:

If (i, j) entry in SSIM is V, then the (i, j) will be ‘1′ and (j, i) will be ‘0′ in the initial reachability matrix. (e_ij = 1 & e_ji = 0)

If (i, j) entry in SSIM is A, then the (i, j) will be ‘0′ and (j, i) will be ‘1′ in the initial reachability matrix. (e_ij = 0 & e_ji = 1)

If (i, j) entry in SSIM is X, then both (i, j) and (j, i) will be ‘1′ in the initial reachability matrix. (e_ij = 1 & e_ji = 1)

If (i, j) entry in SSIM is o, then both (i, j) and (j, i) will be ‘0′ in the initial reachability matrix. (e_ij = 0 & e_ji = 0)

8. Final Reachability Matrix

We may use the transitivity check to construct a consistent model from the variables when generating the Final reachability matrix. The transitivity check in ISM is based on the premise that if “A” is linked to “B”, and “B” is related to “C”, “A” is related to “C”. If the factor “I” is linked to “k”, and “k” is related to “j”, then “I” is related to “j”.

Here, the link between “i” to “k” is the direct link and is denoted as “1”, and the link between “i” and “j” is the transitive or indirect link, which will be denoted as “1 *”.

In our analysis, e.g., ‘IT’ influence ‘COP’ and ‘COP’ influence ‘PS’, so based on the transitivity relation, ‘IT’ will influence ‘PS’ and this relation is denoted as “1 *”. Final reachability matrix will be prepared based on all transitivity checks (

Table 3. The final reachability matrix of SSIM.

	1-ETW	2-IT	3-COP	4-IC	5-PS	6-TM	7-SP	8-MOT	9-OS	10-ESE
1-ETW	1	0	0	0	0	0	1	1	0	1
2-IT	1	1	1	1	1 *	1	1	1	0	1
3-COP	1 *	1	1	1	1	1	1 *	1	0	1 *
4-IC	1	1 *	1	1	1 *	1 *	1	1	0	1
5-PS	1	1 *	1	1	1	1 *	1	1	0	1
6-TM	1	0	0	0	0	1	1	1	0	1
7-SP	0	0	0	0	0	0	1	0	0	0
8-MOT	0	0	0	0	0	0	1 *	1	0	1
9-OS	1	1	1	1	1	1	1	1	1	1
10-ESE	0	0	0	0	0	0	1	0	0	1

(Source: Survey data and author’s calculation). *: It shows the relationship based on transitivity check.

shows the results).

9. Driving Power and Dependence Power Matrix

We can now determine the driving power and dependence power of each component after obtaining the final reachability matrix. The results show that each variable drives x number of variables and is dependent on x amount of variables. As a result, we may add the matrix horizontally and vertically to determine the driving power and factor dependencies (

Table 4. Calculation of driving power and dependence power of SSIM.

	1-ETW	2-IT	3-COP	4-IC	5-PS	6-TM	7-SP	8-MOT	9-OS	10-ESE	Driving Power
1-ETW	1	0	0	0	0	0	1	1	0	1	4
2-IT	1	1	1	1	1 *	1	1	1	0	1	9
3-COP	1 *	1	1	1	1	1	1 *	1	0	1 *	9
4-IC	1	1 *	1	1	1 *	1 *	1	1	0	1	9
5-PS	1	1 *	1	1	1	1 *	1	1	0	1	9
6-TM	1	0	0	0	0	1	1	1	0	1	5
7-SP	0	0	0	0	0	0	1	0	0	0	1
8-MOT	0	0	0	0	0	0	1 *	1	0	1	3
9-OS	1	1	1	1	1	1	1	1	1	1	10
10-ESE	0	0	0	0	0	0	1	0	0	1	2
Dependence	7	5	5	5	5	6	10	8	1	9

(Source: Survey data and author’s calculation). *: It shows the relationship based on transitivity check.

).

Calculate the sum of rows and columns of matrix ‘S’. The sum of rows is denoted as ‘R’ and the sum of columns is denoted as ‘C’.

• S = ${\left[{\mathrm{t}}_{\mathrm{ij}}\right]}_{{\mathrm{n}\times \mathrm{n}}^{,}}\mathrm{i},\mathrm{j}=\left\{1,2,3\dots \dots ,\mathrm{n}\right\},$
• R = ${\lceil {{\displaystyle \sum }}_{\mathrm{j}=1}^{\mathrm{n}}{\mathrm{t}}_{\mathrm{ij}}\rceil }_{\mathrm{n}\times 1={\left[{\mathrm{t}}_{\mathrm{ij}}\right]}_{\mathrm{n}\times 1}},$
• C = ${\lceil {{\displaystyle \sum }}_{\mathrm{i}=1}^{\mathrm{n}}{\mathrm{t}}_{\mathrm{ij}}\rceil }_{1\times \mathrm{n}={\left[{\mathrm{t}}_{\mathrm{ij}}\right]}_{1\times \mathrm{n}}},$

10. Level of Partition

The final reachability matrix for each factor may be used to calculate the partition level. Reachability set (R), antecedent set (C), and interaction set (I) are examples of levels (RC). The reachability set (R) includes the factor and any other factors that it may influence, whereas the antecedent set (C) includes the factor and any other elements that may affect it. The interaction set (RC) is derived from all of the elements in order to identify the levels that will be used to construct the ISM hierarchical structural model.

Table 5. Level of partition using ISM Analysis.

LEVEL-1
Factors	Reachability Set (R)	Antecedent Set (C)	Intersection Set (RC)
1	1,7,8,10	1,2,3,4,5,6,9	1
2	1,2,3,4,5,6,7,8,10	2,3,4,5,9	2,3,4,5
3	1,2,3,4,5,6,7,8,10	2,3,4,5,9	2,3,4,5
4	1,2,3,4,5,6,7,8,10	2,3,4,5,9	2,3,4,5
5	1,2,3,4,5,6,7,8,10	2,3,4,5,9	2,3,4,5
6	1,6,7,8,10	2,3,4,5,6,9	6
7	7	1,2,3,4,5,6,7,8,9,10	7
8	7,8,10	1,2,3,4,5,6,8,9	8
9	1,2,3,4,5,6,7,8,9,10	9	9
10	7,10	1,2,3,4,5,6,8,9,10	10
LEVEL-2
Factors	Reachability Set (R)	Antecedent Set (C)	Intersection Set (RC)
1	1,8,10	1,2,3,4,5,6,9	1
2	1,2,3,4,5,6,8,10	2,3,4,5,9	2,3,4,5
3	1,2,3,4,5,6,8,10	2,3,4,5,9	2,3,4,5
4	1,2,3,4,5,6,8,10	2,3,4,5,9	2,3,4,5
5	1,2,3,4,5,6,8,10	2,3,4,5,9	2,3,4,5
6	1,6,8,10	2,3,4,5,6,9	6
8	8,10	1,2,3,4,5,6,8,9	8
9	1,2,3,4,5,6,8,9,10	9	9
10	10	1,2,3,4,5,6,8,9,10	10
LEVEL-3
Factors	Reachability Set (R)	Antecedent Set (C)	Intersection Set (RC)
1	1,8	1,2,3,4,5,6,9	1
2	1,2,3,4,5,6,8	2,3,4,5,9	2,3,4,5
3	1,2,3,4,5,6,8	2,3,4,5,9	2,3,4,5
4	1,2,3,4,5,6,8	2,3,4,5,9	2,3,4,5
5	1,2,3,4,5,6,8	2,3,4,5,9	2,3,4,5
6	1,6,8	2,3,4,5,6,9	6
8	8	1,2,3,4,5,6,8,9	8
9	1,2,3,4,5,6,8,9	9	9
LEVEL-4
Factors	Reachability Set (R)	Antecedent Set (C)	Intersection Set (RC)
1	1	1,2,3,4,5,6,9	1
2	1,2,3,4,5,6	2,3,4,5,9	2,3,4,5
3	1,2,3,4,5,6	2,3,4,5,9	2,3,4,5
4	1,2,3,4,5,6	2,3,4,5,9	2,3,4,5
5	1,2,3,4,5,6	2,3,4,5,9	2,3,4,5
6	1,6	2,3,4,5,6,9	6
9	1,2,3,4,5,6,9	9	9
LEVEL-5
Factors	Reachability Set (R)	Antecedent Set (C)	Intersection Set (RC)
2	2,3,4,5,6	2,3,4,5,9	2,3,4,5
3	2,3,4,5,6	2,3,4,5,9	2,3,4,5
4	2,3,4,5,6	2,3,4,5,9	2,3,4,5
5	2,3,4,5,6	2,3,4,5,9	2,3,4,5
6	6	2,3,4,5,6,9	6
9	2,3,4,5,6,9	9	9
LEVEL-6
Factors	Reachability Set (R)	Antecedent Set (C)	Intersection Set (RC)
2	2,3,4,5	2,3,4,5,9	2,3,4,5
3	2,3,4,5	2,3,4,5,9	2,3,4,5
4	2,3,4,5	2,3,4,5,9	2,3,4,5
5	2,3,4,5	2,3,4,5,9	2,3,4,5
9	2,3,4,5,9	9	9
LEVEL-7
Factors	Reachability Set (R)	Antecedent Set (C)	Intersection Set (RC)
9	9	9	9

The discovered factors are partitioned into seven levels, as shown in Table 5: L1 = (7), L2 = (10), L3 = (8), L4 = (1), L5 = (6), L6 = (2,3,4,5), L7 = (9).

shows the partitioning level (includes Level-1, 2, 3, 4, 5, 6, and 7).

11. MICMAC Analysis

Matriced‘Impactscroises-multiplication appliqu’e an classment (cross-impact matrix multiplication applied to classification), i.e., MICMAC analysis is used to analyze the driving power and dependence power of the variables (Attri et al. 2015). MICMAC analysis is based on the multiplication properties of matrices (Sharma et al. 1995). Based on the driving power and dependence power, the variables are classified into four quadrants (Faisal 2010). Also prepared the driving power and dependence power diagraph (Figure 1).

All the variables are displayed in four quadrants, i.e.,

Quadrant I (autonomous factors)—Those factors with weak driving power and weak dependence power will come under this quadrant. There is no factor preserved in this region.

Quadrant II (dependent Factors)—Those factors with weak driving power and a strong dependence power will come under this quadrant. The factors TM (6), ETW (1), MOT (8), ESE (10) and SP (7) are preserved in this quadrant.

Quadrant III (Linked Factors)—Those factors with strong driving power and strong dependence power will come under this quadrant. There is no factor preserved in this region.

Quadrant IV (Independent factors)—Those factors with strong driving power and weak dependence power will come under this quadrant. The factors OS (9), IT (2), COP (3), IC (4), and PS (5) are preserved in this quadrant.

12. Multilevel Hierarchy Model

We can now design the interpretive structural model (ISM) hierarchy model (Figure 2) using the level partition table (Table 5), in which all of the levels of the hierarchy model will be occupied by those components that have the same reachability and interaction sets. The variables in our study are divided into five tiers, i.e., L1 = (7), L2 = (10), L3 = (8), L4 = (1), L5 = (6), L6 = (2, 3, 4, 5), L7 = (9).

Factor OS (9) is dependent on 1 factor, i.e., itself (OS) and driving 10 factors, including itself (IT, COP, IC, PS, TM, ETW, MOT, ESE, SP, OS). Factor IT (2) is dependent on 5 factors including itself (IT, COP, IC, PS, OS) and driving 9 factors, including itself (IT, COP, IC, PS, TM, ETW, MOT, ESE, SP).

Factor COP (3) is dependent on 5 factors, including itself (IT, COP, IC, PS, OS) and driving 9 factors, including itself (IT, COP, IC, PS, TM, ETW, MOT, ESE, SP).

Factor IC (4) is dependent on 5 factors, including itself (IT, COP, IC, PS, OS) and driving 9 factors, including itself (IT, COP, IC, PS, TM, ETW, MOT, ESE, SP).

Factor PS (5) is dependent on 5 factors, including itself (IT, COP, IC, PS, OS) and driving 9 factors, including itself (IT, COP, IC, PS, TM, ETW, MOT, ESE, SP).

Factor TM (6) is dependent on 6 factors, including itself (TM, IT, COP, IC, PS, OS) and driving 5 factors, including itself (TM, ETW, MOT, ESE, SP).

Factor ETW (1) is dependent on 7 factors, including itself (ETW, TM, IT, COP, IC, PS, OS) and driving 4 factors, including itself (ETW, MOT, ESE, PS).

Factor MOT (8) is dependent on 8 factors, including itself (MOT, ETW, TM, IT, COP, IC, PS, OS) and driving 3 factors, including itself (MOT, ESE, SP).

Factor ESE (10) is dependent on 9 factors, including itself (ESE, MOT, ETW, TM, IT, COP, IC, PS, OS) and driving 2 factors, including itself (ESE, SP).

Factor SP (7) is dependent on 10 factors, including itself (SP, ESE, MOT, ETW, TM, IT, COP, IC, PS, OS) and driving 1 factors, i.e., itself (PS).

13. Discussion and Conclusions

During COVID-19, it is critical to analyze the variables impacting the success of online education in Odisha. We created a link between the elements in our study to assign emphasis on a priority basis for the proper implementation of online education. Data collected from the experts were interpreted using the interpretive structural model (ISM) and MICMAC analysis.

The factors ‘Organization support” is at the bottom level (Figure 1) of the hierarchy and has the highest driving power, which means the factor is the most important factor for the success of online education as it drives all other nine factors. The authority should provide all the support in terms of resources and a conducive environment for the successful implementation of online education. This factor comes under Quadrant-IV (Figure 1), which is for the independent factors.

Factors such as “Infrastructure & Technology”, “Cost of Project”, “Internet connectivity”, and “Power supply” are in level-6. Each of these variables has a driving power of 9 and dependence power 5, including themselves. The factor “Organization support” is driving these factors, and subsequently, these factors will drive the factor “Trained manpower”, which is in level-5. The factor “Trained manpower” drives the factor “Effective team work”, and it will drive another factor, “Method of Teaching”. The management should understand that the four factors in level-6 ultimately create the “trained manpower”, and this will facilitate “Effective team work” as, if the manpower is well trained, then the better the teamwork will be in the workplace. “Effective team work” will facilitate the factor “Method of teaching” as the better team can arrange all the facilities required to run the online sessions. The authority should understand that, if the method of teaching will be improved, then the study environment will be created and students will be eager to participate in the online classes organized by the institution. For any online education, the organization’s or authority’s support will be critical. Without support, we will not be able to spend money on the project in order to acquire the necessary infrastructure and internet access, which will create a conducive environment and encourage everyone to contribute effectively to the program’s success. To improve the use of such systems, educational institutes should think about providing instructors and other support personnel with training in troubleshooting and emergency preparation. No matter a person’s gender or level of prior educational attainment, our conceptual approach and findings can be adapted to online learning tools and are effective for everyone.

14. Limitations and Scope for Future Research

The study has certain limitations. First, the research is confined only to the state of Odisha and covers its higher educational institutes. Second, the analysis is based on the expert opinion, and may thus have some sort of bias as 12 expert opinion were taken for the analysis. The researchers may include more expert opinions to reduce this bias. Third, we have identified 10 variables for our analysis, and furthermore, such variables may be identified in future research to make the analysis on online education a robust one. In order to analyze the variables, other statistical tools can be used to find solutions concerned to the education sector.