The use of Technology to facilitate better learning and training is gaining momentum worldwide, reducing the temporal and spatial problems associated with traditional learning. Despite its several benefits, retaining students in online platforms is challenging. Through a literature review of the factors affecting adoption, the continuation of technology use, and learning outcomes, this paper discusses an integration of online learning with virtual communities to foster student engagement for obtaining better learning outcomes. Future directions have been discussed, the feedback mechanism which is an antecedent of students’ continuation intention has a lot of scopes to be studied in the virtual community context. The use of Apps in m-learning and the use of cloud services can boost the ease and access of online learning to users and organizations.
1. Introduction Online learning and training are gaining popularity worldwide, reducing the temporal and spatial problems associated with the traditional form of education. The primary factors behind using online learning are not only to improve access to education and training, and quality of learning, but also to reduce the cost and improve the costeffectiveness of education (Bates, 1997). Online learning is mainly provided in two ways—in synchronous and asynchronous environments (Jolliffe, Ritter, & Stevens, 2012). The time lag attributes of asynchronous learning unlike synchronous learning in online platforms take the advantage of accessing materials anytime and anywhere, ability to reach a greater mass at the same time, and uniformity of content. Online learning along with face-to-face learning is successfully used in industry as well as academia with positive outcomes (Chang, 2016). The geographically distributed team in an organization can get their skill training through online platforms at the same time, gaining a greater level of competitiveness. Online learning is also beneficial for students as they can learn at their own pace with the availability of online materials. The e-learning market is becoming popular and widely adopted by the education sector and industry. The growth of the e-learning market can be demonstrated by the fact that the global elearning market is expected to reach 65.41 billion dollars by 2023 growing at a cumulative average growth rate of 7.07% (Research and Markets, 2018a). In addition to this, the global learning management system (LMS) is expected to increase from 5.05 billion USD in 2016 to 18.44 billion USD by 2025 growing at a rate of 15.52% (Research and Markets, 2018b). Despite several advantages of online learning such as improving access to education and training, improving the quality of learning, reducing the cost and improving the cost-effectiveness of education, retaining students in such platforms is a key challenge with a high attrition rate (Perna et al., 2014). Several strategies such as briefing, buddying, and providing feedback on the platform are proposed to retain and engage students (Nazir, Davis, & Harris, 2015). It is also noted that more self-discipline is required by students in online education, unlike traditional classroom education (Allen & Seaman, 2007). Keeping users enrolled and engaged is a challenging job as a personal touch by the instructor is missing or limited. The learning engagement which is an important antecedent for learning outcome is lower for technology-mediated learning than face-to-face learning (Hu & Hui, 2012). As a higher amount of money is spent on infrastructure, staff training, etc., organizations seek to take maximum benefit from online learning which requires an understanding of the factors that drive the adoption, continuation intention, and learning outcome of users on online learning platforms. Therefore, the primary focus of research remains on how to retain online learning users, and increase the efficiency of the online learning. Users may learn inside and outside the classroom; inside classroom learning is through instructors either from face-to-face, pure online or blended learning (combination of face-to-face and pure online learning) whereas outside classroom learning is conducted by users anytime and anywhere after the class. The exponential growth of the Internet has enabled individuals to share information, participate, and collaborate to learn from virtual communities (VC) anytime and anywhere (Rennie & Morrison, 2013). In a virtual community, people do everything that they do in real life but leaving their bodies behind (Rheingold, 2000). Virtual communities keep its users engaged based on familiarity, perceived similarity, and trust by creating a sense of belongingness (Zhao, Lu, Wang, Chau, & Zhang, 2012). It is essential to assess the role of a less constrained informal mode of learning (Davis & Fullerton, 2016) like virtual communities in the formal learning to engage and retain students. The paper is organized as: Section 2 presents the research methodology with a bibliographical database and the framework in which the review is conducted. Section 3 provides details of the literature review with categorizations –technology adoption, the continuation of technology use, and learning outcomes. In Section 4, a detailed discussion is presented, followed by future directions in Section 5. Section 6 sums up the paper with concluding remarks. 2. Research methodology The methodology used for the review of literature is presented in this section. The research methodology is divided into two phases which are Article selection, and Classification and Categorization as depicted in Fig. 1. 2.1. Article selection 2.1.1. Initial pooling The initial pool of sources or articles is obtained from recent IS (Information Systems) journals, education journals, books, and articles. The most recent research papers are searched in the databases by using advanced search options with keywords ‘Online learning’, ‘e-learning,’ ‘Virtual communities,’ ‘Technology adoption,’ ‘Continuation intention,’ ‘Technology use,’ ‘Virtual worlds,’ ‘Learning outcome,’ etc. Also, a backward and forward snowballing method is applied to the initial pool of journal articles. Snowballing method in literature review is a technique for identifying additional papers based on the reference list of the current paper and the citations to the paper (Wohlin, 2014). Here, the articles are searched from the reference list and the cited-by articles to ensure that the risk of missing relevant studies is reduced. The backward snowballing method is where the sources (reference section) of the journal articles with any contribution in the area of online learning are investigated. On the other hand, the forward snowballing method is where the articles citing the journal article under study are examined to discover the contribution in the area of online learning. Along with it, the seminal works related to technology adoption and use in the area of information systems are also selected for review. 2.1.2. Inclusion/exclusion criteria The inclusion/exclusion criteria are applied to the initial pool of research articles. All the research articles in the initial pool are examined. The abstracts and keywords are thoroughly studied and checked for any theoretical contribution in the area of online learning; virtual communities; mobile learning; seminal works on IS theories; the extension of theories; etc. However; the research papers with detailed technology architectures are excluded from the pool. For example; articles with a complete implementation focus or tools development are excluded. 2.1.3. Final pool The final pool of articles contains all the research articles in the initial pool minus the excluded articles based on the inclusion/exclusion criteria applied. The major journals considered in the final pool are provided in Table 1. The final pool of articles is now ready for classification and categorization. 2.2. Classification and categorization 2.2.1. Attribution identification As organizations are moving towards providing education and training with the help of technology with spending in the infrastructure and training, it is essential to understand the factors that affect adoption, continuation use of technology, and learning outcomes. Therefore, the attributes identified to map the final pool of articles are factors affecting adoption, continuation use, and learning outcomes in online learning. 2.2.2. Attribution mapping The final pool of research articles is mapped to the attributes identified in the previous step, technology adoption, continuation of technology use, and learning outcomes.
2.2.3. Categorization The socio-structural influences and psychological mechanisms together produce human behavioral effects (Bandura, 2001). To understand the human behavioral aspects of adoption, continuation, and learning outcomes, it is essential to understand the personal and environmental factors. This paper, categorizes the antecedents of adoption, continued use of technology, and the outcome of virtual learning platforms in the framework of Social Cognitive Theory (Bandura, 1986) and then proposes future directions of research based on the literature review. 3. Classification and categorization of literature This section provides a detailed analysis of the factors affecting technology adoption, continuation, and learning outcomes. Further, the literature is categorized into personal and environmental factors. Numerous effort and research have been carried out on adoption and continuation of technology use. The main theories studied for technology adoption are Diffusion of Innovation Theory (Rogers, 2010), Technology Acceptance Model (TAM) (Davis, 1989), Theory of Planned Behavior (TPB) (Ajzen, 1991; Fishbein & Ajzen, 1975), and Unified Theory of Acceptance and Use of Technology (UTAUT) (Venkatesh, Morris, Davis, & Davis, 2003). Theories which provides a base for technology continuation are the Expectation-Confirmation model for IS continuation (Bhattacherjee, 2001), DeLone and McLean IS success model (Delone & McLean, 2003), Flow theory (Csikszentmihalyi, 2000), Social Cognitive Theory (Bandura, 1986), etc. This paper looks at the factors of adoption, continuation use of the technology, learning outcomes, and categorizes them through the lens of Social Cognitive Theory. 3.1. Technology adoption The base theories of adoption have been used and extended by many researchers (Chen, Hsieh, Van de Vliert, & Huang, 2015; Cheng, 2012; Ho, Ke, & Liu, 2015; Hong, Lui, Hahn, Moon, & Kim, 2013; Kaba & OseiBryson, 2013; Kirs & Bagchi, 2012; Knight & Burn, 2011). The antecedents of technology adoption from various theories and their extensions are discussed in the following sub-sections in terms of personal and environmental factors. 3.1.1. Personal factors 3.1.1.1. Perceived usefulness and perceived ease of use:. According to the Technology Acceptance Model (TAM) (Davis, 1989), the Perceived Usefulness (PU) and the Perceived Ease of Use (PEoU) are the predominant antecedents of technology adoption and they determine the intention to use through attitude (Fishbein & Ajzen, 1977), which in turn leads to the use behavior (Knight & Burn, 2011; Ros et al., 2015; Saadé & Bahli, 2005). However, a study by Kim, Kim, and Han (2013) has found that PEoU has no significant effect on attitude in TAM in the development of learning widgets for the e-learning environment context. This indicates that ease of use of a widget alone does not imply the attitude, but the usefulness of it does. There are several factors explored by researchers which along with PU and PEoU affect the acceptance of the technology. These are: 3.1.1.1.1. Perception of interaction (PoI):. The TAM framework is extended with construct PoI to accommodate continuous system usage rather than users’ early adoption of IT (Knight & Burn, 2011). The PoI is the users’ perception of their ongoing interactions with their adopted technologies. Furthermore, PoI along with PU determines the use intentions of service-oriented e-learning management systems of students (Ros et al., 2015). 3.1.1.1.2. Cognitive absorption:. The three dimensions of cognitive absorption (the level of involvement), temporal dissociation, heightened enjoyment, and focused immersion is different for individuals, and affects both PU and PEoU to determine the technology adoption (Saadé & Bahli, 2005). 3.1.1.1.3. Self-distraction:. The self-distraction as an escape mechanism built on Escape theory (Heatherton & Baumeister, 1991) affects the attitude towards the virtual world. With higher selfdistraction (or escaping the negative confines of real life), playfulness would lead to greater attitude towards the virtual world, and social presence would lead to lesser attitudes towards the virtual world (Schwarz, Schwarz, Jung, Pérez, & Wiley-Patton, 2012). 3.1.1.1.4. Cognitive age:. PU, PEoU, and Perceived Enjoyment (PE) play a significant role in the decision of IT acceptance for individuals who are young at heart (cognitive age lesser than their actual age). On the other hand, PEoU and subjective norm (perceived social pressure) play a significant role for individuals who perceive themselves as old as they are (Hong et al., 2013). 3.1.1.1.5. Social networks:. The eigenvector centrality (the extent to which an individual is connected to influential others) and the closeness centrality (how close or distant the network users are with other users within the network) positively influence the technology use (Venkatesh & Sykes, 2013). 3.1.1.1.6. National culture (Hofstede, 1984):. The relationship between PU and attitude is more intense for masculinity, individualistic, low power distance and low uncertainty avoidance culture (Kaba & Osei-Bryson, 2013). 3.1.1.1.7. Surrounding conditions (thermal climate and national wealth):. Thermal climate (temperature conditions) and national wealth (national income) moderates the relationship from PEoU to PU and PU to continuation intention. The difference in the strength of the relationships is more for the poor (national wealth) − harsh (climate) and poor-temperate countries than that of rich-harsh and rich-temperate countries regarding knowledge seeking behavior (Chen et al., 2015). The acceptance of technology not only depends on the factors such as PU and PEoU but also on various internal and external/environmental factors. The personal factors such as perception of interaction, cognitive absorption, escape mechanism, cognitive age, etc. play vital roles in determining the use of technology. In addition to this, the environment of an individual such as culture, surrounding conditions, the social network, etc. are pivotal in determining the use of technology. Hence, organizations should carefully examine these factors along with perceived use and perceived ease of use while implementing any technology to understand their adoption. 3.1.1.2. Perceived behavior control:. Perceived Behavior Control (PBC) Table 1 Bibliographical databases included. Bibliographical Databases included Australasian Journal of Information Systems Behaviour & Information Technology. British Journal of Educational Technology Decision Support Systems European Journal of Information Systems Information & Management Information and Organization Information Systems Research Information Technology & People Information, Communication & Society International Journal of Information Management Internet Research Journal of Knowledge Management Journal of Management Information Systems Journal of the Association for Information Systems MIS Quarterly Personal and Ubiquitous Computing The Information Society R. Panigrahi et al. International Journal of Information Management 43 (2018) 1–14 3 is defined as the individuals’ perception of the ease/difficulty of performing the behavior. PBC is introduced in the Theory of Planned Behavior to deal with its limitation of behaviors over which individuals have incomplete volition control (Ajzen, 1991). The perceived behavior control positively affects the use intentions in a virtual world context (Mäntymäki, Merikivi, Verhagen, Feldberg, & Rajala, 2014). 3.1.1.3. Performance and effort expectancies:. The performance and effort expectancies from the Unified Theory of Acceptance and Use of Technology (UTAUT) affect the behavioral intention and behavior (Venkatesh et al., 2003). The performance expectancy is the degree to which an individual believes that using a system would help to attain the gain in the performance, and it captures the concept of constructs perceived usefulness, extrinsic motivation, job fit, relative advantage, and outcome expectations. The effort expectancy can be defined as the degree of ease associated with the system, and the construct captures the concept of perceived ease of use, ease of use, and complexity. In addition to the expectancies, the culture affects the technology adoption by individuals. The impact of effort expectancy on behavioral intention is stronger in an individualistic, and low power distance culture and the impact of behavioral intention on actual behavior are stronger in low uncertainty avoidance cultures (Im, Hong, & Kang, 2011). 3.1.1.4. User resistance:. User resistance which is a primary concern in IT implementation, negatively impacts the adoption. It has three perspectives, system (technology related factors), people (individuals or group factors), and interaction oriented (interactions between characteristics related to people and system). The main sources of resistance are distorted perception, low motivation for change, political and cultural deadlock, lack of creative response, and others (Ali, Zhou, Miller, & Ieromonachou, 2016). 3.1.2. Environmental factors 3.1.2.1. Perceived characteristics of innovation:. Innovation is defined as “an idea, practice, or object that is perceived as new by an individual or another unit of adoption” (Rogers, 2010). According to Diffusion of Innovation Theory, the perceived characteristics of innovation (relative advantage, compatibility, complexity, trialability, and observability) in the persuasion stage of diffusion plays a significant role in adoption (Rogers, 2010). The online learning has certain relative advantages such as flexibility in schedule and lower cost than the offline learning. The compatibility of the e-learning system is positively related to its adoption. Furthermore, the adoption rate depends on the simplicity (less complexity) of the technology to implement and use. The more the online learning is tried by users, the more the rate of adoption. The observability characteristic requires track records and statistics to make the result visible. Internet diffusion is vital for technological progress in a country, and the factors affecting diffusion in developing countries differ from those of developed countries. Users’ cognition and government policies can accelerate the diffusion process only after a certain threshold level of human rights has been attained in a developing country (McCoy, Cha, & Durcikova, 2012). The technology adoption is positively related to nations’ generalized trust (trust in members of the society), and the rate of change of adoption is associated with a rate of change of trust (Kirs & Bagchi, 2012). 3.1.2.2. Subjective norm:. Subjective norm is the perception of an individual about whether the others who are important to them believe that he/she should perform a particular behavior (Fishbein & Ajzen, 1975). It determines the intention to use, which further leads to use behavior. Although subjective norm positively impacts participation intention in several studies (Fishbein & Ajzen, 1975; Hong et al., 2013), Zhou (2011) has posited that subjective norm does not affect the participation intention in the online community. Social influence from the Unified Theory of Acceptance and Use of Technology capture the concepts of constructs –subjective norm, social factors, and image (Venkatesh et al., 2003). The social influence is positively affected by the interpersonal and external influence (Mäntymäki et al., 2014), and it positively impacts the use behavior through the use intentions (Venkatesh et al., 2003; Im et al., 2011). 3.1.2.3. Facilitating conditions:. The facilitating conditions are the degree to which an individual believes that the organizational and technical infrastructure exists to support the use of the system (Venkatesh et al., 2003). The facilitating conditions directly impact the actual use of the system rather than the intention to use (Venkatesh et al., 2003; Im et al., 2011). It is also found that the impact of the facilitating conditions on the actual use has no difference in the individualistic and collectivist cultures (Im et al., 2011). 3.1.2.4. Technology inhibitors:. Technology inhibitors work along with technology enablers; inhibitors are not just antipoles of technology enablers but have a separate existence by themselves (Cenfetelli & Schwarz, 2011). The study identified intrusiveness, effort redundancy, and process uncertainty as system inhibitors, and information overload, irrelevant requests for information, and deceptiveness as information inhibitors. These inhibitors negatively impact the usage intentions. 3.1.3. Technology adoption in organizations The organizational support and prior experience affect both PE and PEoU whereas computer self-efficacy predicts PEoU, and task equivocality (level of ambiguity occurs during the task) predicts PU for technology adoption by employees (Lee, Hsieh, & Chen, 2013). With simultaneous use of e-learning systems where an old system is not phased out, the management who makes the decision for adoption in an institution with a higher construal level likely to pay attention to PEoU whereas the management with a lower construal level pays attention to PU of the new system (Ho et al., 2015). High construal level is seeing the bigger picture, not the details and vice versa (Trope & Liberman, 2010). Organizational infrastructure and culture, staff attitude and skills, student expectations and experience along with the introduction of departmental e-learning champions are key factors determining the adoption of online learning in higher education (King & Boyatt, 2015). 3.1.4. Summary It is essential for the organizations to understand the factors that affect adoption both at the individual level as well as at the environment level before implementing online learning because once the factors are identified and understood, it becomes easier for the stakeholders to successfully implement it. At a broader level, the perceived characteristics of innovation (of online learning) such as relative advantage, compatibility, complexity, trialability, and observability play a significant role in adoption. Literature suggests that the primary factors of adoption are perceived usefulness and perceived ease of use. But, there are several other factors which work along with these two factors to explain technology adoption. They are the perception of interaction, cognitive absorption, self-distraction, cognitive age, social network characteristics, national culture, surrounding conditions, etc. In addition to this, the unified theory of acceptance and use of technology states that the performance expectancy, effort expectancy, social influence, and facilitating conditions affect the acceptance and use of technology. Apart from the technology enablers which positively affect the technology adoption, there are also certain technology inhibitors which have separate existence and negatively impact the technology adoption; they are intrusiveness, effort redundancy, process uncertainty, information overload, irrelevant requests for information, and deceptiveness which negatively impacts the usage intentions via system and information quality. User resistance is a major concern while implementing any technology in an organization which should be handled with different strategies to overcome it