Does sociability quality of web-based collaborative learning information system influence students’ satisfaction and system usage?

Abstract

The use of collaborative learning technologies is a stimulating element of collaborative learning process, where social interaction and collaboration are key factors. This research examines the impact of sociability quality on the usage of web-based collaborative learning information system (WBCLIS) and user satisfaction. We propose a theoretical model by integrating the construct of ‘sociability quality’ in the DeLone and McLean’s (Journal of Management Information Systems 19:9–30, 2003) updated information system success model. Proposed theoretical model was empirically validated, in a service-learning course with undergraduate students, where data was collected using an online questionnaire and evaluated through partial least square, structural equation modelling (PLS-SEM) statistical approach. Results suggest that, sociability quality has a direct positive impact on the system use and overall user satisfaction, along with a strong indirect impact on the net benefits of the WBCLIS. Findings also confirmed that, system use and user satisfaction are strong predictors of the net benefits. These results about sociability quality, contribute significantly in the domain of IS success literature, by identifying a novel and critical IS success dimension. Further, theoretical contribution in the context of sociability quality for IS success, and practical implications entailing the use of WBCLIS in the domain of service learning are also discussed.

Introduction

Online / web-based collaborative learning information system (WBCLIS), offers a web-based social learning environment, which facilitates collaboration between various stakeholders with technological support and pedagogical processes (Al-Samarraie & Saeed, 2018; Cidral, Oliveira, Di Felice, & Aparicio, 2018). With recent overwhelming wide range expansion and promotion of the information and communication technologies (ICTs) coupled with better and fast speed access to World Wide Web, internet and advance computing devices (i.e. tablets, laptops, mobile devices), the concept of social and collaborative learning, has expanded very promptly in higher education worldwide (Molinillo et al., 2018a, b). Prominent examples of advance computer supported learning systems include WebCT, Blackboard and Moodle (Deng & Tavares, 2013; Salam, Awang Iskandar, Ibrahim, & Farooq, 2019a, b). Some of them are computer supported collaborative learning (CSCL) systems (e.g. WebCT and other similar knowledge building forums), which are specifically designed for collaborative learning (Hoppe, Soller, & Ogata, 2007). Whereas, other learning management systems (e.g. Blackboard and Moodle etc.) offer limited inbuilt features and tools for collaboration and social interaction between users (Deng & Tavares, 2013).

Therefore, various scholars (e.g. Al-Rahmi & Zeki, 2017; Smith, 2016) have also reported the use of social networking websites (e.g. Twitter, BlogSpot and Facebook etc.), as a platform for academic social interaction and collaborative learning in higher education. Collaborative learning improves students’ participation and contribution, by allowing them to interact with each other, in a group learning situation, which provides various opportunities to learn from their social interaction (Liaw, Chen, & Huang, 2008). The implementation of collaborative learning, especially project-based learning, has been extensively promoted in higher education (Gress, Fior, Hadwin, & Winne, 2010; Wang, Fang, & Gu, 2020; Wasko & Faraj, 2005). For this purpose, specific collaborative learning technologies have been implemented, which have emerged as an important communication tool for students’ academic social interaction with each other (Al-Samarraie & Saeed, 2018; Liao, Huang, Chen, & Huang, 2015). The effective use of CSCL systems in higher education is well documented, for a number of widespread positive outcomes (e.g. providing spot-on social learning environment, making learning fun, interactive and meaningful etc.), where students can vigorously intricate in the learning process (Kurilovas & Kubilinskiene, 2020; Martins et al. 2019; Radianti, Majchrzak, Fromm, & Wohlgenannt, 2020; Rajab & Eydgahi, 2019; Shen & Ho, 2020).

Moreover, it enhances students’ motivation to learn from social interaction in a peer learning environment, in which collaboration, interaction and active engagement are key factors (Arendt, 2019; Bulotsky-Shearer et al., 2020; Burdelski, 2019; Chen, Park, & Breazeal, 2020; Kim, Wang, & Ketenci, 2020; Lowton-Smith et al., 2019). A latest study by Molinillo et al. (2018b) have stated that, social interaction (i.e. students’ interaction with other students, and students’ interaction with teachers), engagement and social presence are strong determinants of an active learning environment. In this regard, Kreijns and Kirschner (2004) have also asserted that, CSCL systems should provide sociable learning environment for students, by focusing on the social aspects of collaborative learning process. Therefore, programmers and system developers should not undermine the importance of social functionalities, in the collaborative learning process, and their CSCL system should also provide an interactive learning environment (Farooq et al., 2017; Kreijns & Kirschner, 2004; Salam et al., 2019b). Literature review has shown that, sociability is one of the important elements, that influence the acceptance and use of collaborative technologies (Brown, Dennis, & Venkatesh, 2010; Gao, Dai, Fan, & Kang, 2010; Junglas, Goel, Abraham, & Ives, 2013).

When a system provides all-encompassing social space and tools for interaction and collaboration, it can inevitably increase users’ intent to employ that system, which ultimately affects the users’ satisfaction and net benefits derived from the performance of a CSCL system (Junglas et al., 2013). In a collaborative learning environment, it's social functionalities or sociability quality, allows students to interact, not only to absorb information in a deliberate manner, but it also enables them to attentively ponder, participate, discuss and share their ideas with others (Doumanis, Economou, Sim, & Porter, 2019; Jiang & Zhang, 2020; Rosen, Wolf, & Stoeffler, 2019; Wang et al., 2020; Zhou et al., 2020). Kreijns, Kirschner, Jochems, and van Buuren (2007) have described three major factors, “i.e. sociability, social presence and pedagogical structure”, that influence social interaction, and also contribute to the design of a sociable collaborative learning environment. Among these aforementioned factors, sociability is a strong determinant, which defines the propensity of overall system use and user satisfaction towards a CSCL system (Salam et al., 2019b). In this regard, Junglas et al. (2013) have asserted that, along with the technical and functional qualities of an information system, it's social components or sociability also contributes to the acceptance and actual use of an information system.

Regardless of the recognition of this fact, that social component plays an imperative role in the success of a CSCL system, there are very few studies (e.g. Al-Rahmi & Zeki, 2017; Brown et al., 2010; Gao & Bai, 2014), that have examined the impact of social factors in a collaborative learning environment. More importantly, analyzing the impact of sociability quality, in combination with DeLone and McLean’s (2003) updated IS success model, will not only allow us to evaluate it’s influence on system use and user satisfaction, but it will also help in enhancing our understanding, about it’s relative importance and contribution towards overall success of a WBCLIS (Salam et al., 2019a, 2019b). For that reason, this study aims to examine the direct impact of sociability quality, on system use and user satisfaction in the context of WBCLIS. For this purpose, a theoretical model is developed and evaluated, by conducting an in-depth empirical study with undergraduate students, who had used a WBCLIS for participating in a group based service learning project. Variance based structural equation modeling (SEM) statistical approach was employed, and data collected through an online self-administered survey, was analyzed, with the help of partial least square (PLS) method.

Next section outlines the detail of WBCLIS system used by our respondents, along with a thorough review of literature on DeLone and McLean’s (2003) updated IS success model, and it also defines the concept of sociability and social interaction theory (Gunawardena, 1995; Kreijns et al., 2007; Mills & Durepos, 2010). Then a detailed discussion on the proposed theoretical model, hypothetical relations and description of constructs is also provided. Research methodology section presents details regarding the design of this study, participants, research setting, methods and approaches used in this study, along with brief overview of data collection and evaluation process. Results section presents the description of empirical findings for measurement and structural models proposed in the theoretical model of this study. At last, discussion section outlines the theoretical contribution and implications of this research, along with limitations and future research directions, followed by a conclusion section for summarizing the outcomes of this study.

Literature review

Web-based collaborative learning information system (WBCLIS)

Collaborative learning has become very popular in higher education, for inculcating critical thinking skills and promoting active learning environment, in which group of students work together to attain a shared common learning objective (Al-Samarraie & Saeed, 2018). Theoretically, collaborative learning is associated with the social constructivism learning paradigm (Elia, Solazzo, Lorenzo, & Passiante, 2019; Mamun, Lawrie, & Wright, 2020; Shuell, 2001; Tennyson & Volk, 2015); in which, information sharing and knowledge building is regarded to be significantly influenced, by collaboration and interaction between group members (Curşeu, Rusu, Maricuţoiu, Vîrgă, & Măgurean, 2020; Hernández-Sellés, Muñoz-Carril, & González-Sanmamed, 2019; Molinillo et al. 2018a; Tan, 2019; Williamson et al., 2020). In the context of collaborative learning, in-group collaboration is often regarded as an affirmative forte, that aids students in the achievement of their shared common goals, while simultaneously upholding their own individual level learning goals (Lee, 2014).

In this modern age of artificial intelligence and advance technology enabled devices, access to fast speed internet has led to a widespread use and acceptance of collaborative learning, by facilitating the formation of online dialogical interaction between teachers and students (Molinillo et al., 2018a). More particularly, WBCLIS have been especially developed in recent past, to enhance the overall connotation and meaningfulness of the social interaction, for fostering the presence of an active learning environment (Molinillo et al., 2018a; Salam et al., 2019b). WBCLIS refers to a set of tools, that are specifically designed for promoting collaborative learning, through meaningful interaction, expressive knowledge sharing and frequent inter-linked goal oriented group activities (Cheung & Vogel, 2013). Most of the literature (e.g. Cheung & Vogel, 2013; Farrokhnia, Pijeira-Díaz, Noroozi, & Hatami, 2019; Garcia-Sanjuan, Jurdi, Jaen, & Nacher, 2018; Hernández-Sellés et al., 2019; Liaw et al., 2008; Reis et al., 2018), which reviewed the significance and use of CSCL systems, has been built on the fundamentals of cognitive science and social constructivist theory only (Holenko Dlab, Boticki, Hoic-Bozic, & Looi, 2020; Jeong, Hmelo-Silver, & Jo, 2019; Weimer, Parault Dowds, Fabricius, Schwanenflugel, & Suh, 2017; Zheng, Li, Zhang, & Sun, 2019). Thereby, previous studies (e.g. Cheung & Vogel, 2013; Farrokhnia et al., 2019; Hernández-Sellés et al., 2019; Liaw et al., 2008; Shen, 2012) have overlooked various dimensions, important to the functionalities of a WBCLIS system (Garcia-Sanjuan et al., 2018; Reis et al., 2018; Salam, 2020).

Specifically, WBCLIS are designed in such a way, that enables them to provide optimum computing functionalities, quality information and internet technologies, that can foster the interaction between students and teachers, for enhancing a seamless flow of information and sharing of knowledge (Liaw et al., 2008). According to Molinillo et al. (2018b), WBCLIS provides opportunities for students, to become more responsible and active learner, not only to absorb and exchange information, but also to easily reflect back, and straightforwardly link their previous knowledge, with the newly comprehended information on the subject matter. In recent few years, a rising trend of widespread application of collaborative learning, by means of advance technologies (i.e. WBCLIS and other similar virtual learning systems), has been witnessed, in the world over higher education / academic industry. Thereby, WBCLIS and other similar technologies have become an integral part of collaborative learning, which has been increasingly implemented, as an essential element of service learning pedagogy, in the top ranked institutions of higher education, around the globe (Al-Samarraie & Saeed, 2018; Cheung & Vogel, 2013; Molinillo et al., 2018a).

With reference to the above discussion, it is pertinent to note that, collaborative learning is mediated by technology, that facilitates a collaborative learning environment, in which necessary pedagogical and social functionalities are provided (Chan & Pow, 2020; Isohätälä, Järvenoja, & Järvelä, 2017; Lin, 2020; Molinillo et al., 2018b). In a collaborative learning tool, it is crucial to provide social features, that can facilitate and encourage the social interaction among teachers and students (Molinillo et al., 2018a). So that, students can gain deeper insights, for comprehending the multidimensional aspects of new knowledge, exchange their views and develop their independent opinion, along with cognitive skills, by participating in a socially supportive collaborative learning process (Kreijns et al., 2007; Molinillo et al., 2018a). Where many proponents (e.g. Atan, Rahman, Majid, & Dahlan, 2012; Doumanis et al., 2019; Jan, Chen, & Huang, 2016; Mata-Rivera, Torres-Ruiz, Guzmán, Moreno-Ibarra, & Quintero, 2015; Molinillo et al., 2018a; Molinillo et al., 2018b) support the use of WBCLIS, and acknowledge it’s significance, for facilitating the collaborative learning process. Still, some opponents (e.g. Fleaca & Stanciu, 2019; Krishnakumar & Nogales, 2020; Njenga, Garg, Bhardwaj, Prakash, & Bawa, 2019; Pinho, Franco, & Mendes, 2019) argue that, collaboration through WBCLIS, presents some operational difficulties, that must be considered. For instance, with respect to the face-to-face collaboration, collaborative learning through WBCLIS requires certain IT skills and computer knowledge (Doumanis et al., 2019; Jeong et al., 2019).

Moreover, students as well as teachers, also have to spend substantial amount of their time, for understanding the operational functionalities of a given WBCLIS (Atan et al., 2012; Jan et al., 2016). Further, successful implementation of a WBCLIS do require dedication and candid efforts from both, teachers and students, to actively participate in the collaborative learning activities, for achieving their shared goals, while simultaneously maintaining their individual beliefs through self-learning, comprehension and with a reflection of comparative analysis of their previous acquaintance, with newly acquired information (ChanLin, 2012). Moreover, teachers should clearly define the collaborative learning activities, along with outlining their own role in the planning, execution and evaluation of whole collaborative learning process (Churchill, 2011). Additionally, it is important to enrich a WBCLIS system, with various social features and tools, to facilitate in-group collaboration, social interaction and social presence, which are rationally very important, for the success of whole collaborative learning process (Kreijns et al., 2007; Kreijns, Van Acker, Vermeulen, & Van Buuren, 2014; Kreijns, Kirschner, & Vermeulen, 2013; Yamada, Goda, Matsukawa, Hata, & Yasunami, 2016).

Information system success

In this era of fast pace technological advancements, majority of the modern businesses are heavily relying on modern IS systems; thereby, becoming more dependent on IT, for accomplishing their daily routine operations (Farooq 2018; Salam 2020). Moreover, world over various departments of large size corporations, also count on IS, for attaining success in their different organizational goals (Grabowski & Roberts, 2019; Modaresnezhad, Iyer, Palvia, & Taras, 2020; Prys, Krysińska, Janaszkiewicz, Winiecki, & Różewski, 2018; Wang & Zhao, 2020). Therefore, acceptance and use of technology has also remained a subject of interest for many academicians (e.g. Baishya & Samalia, 2019; Farooq et al., 2017; Oviedo-Trespalacios, Briant, Kaye, & King, 2020; Tao et al., 2020; Tsertsidis, Kolkowska, & Hedström, 2019). Moreover, previous studies (e.g. Al-Fraihat, Joy, Masa’deh, & Sinclair, 2020; Cheung & Vogel, 2013; Pinho et al., 2019; Shim & Jo, 2020; Yuan, Chu, Lai, & Wu, 2020) have suggested several factors (i.e. technological, behavioral and organizational etc.), that determine the effective use, acceptance, adoption and success of an information system. Various technology related models (e.g. Davis, 1989; Farooq et al., 2017; Venkatesh, Morris, Davis, & Davis, 2003; Venkatesh, Thong, & Xu, 2012) have only focused on the pre-adoption stage of technology, by putting more emphasis on the acceptance of technology only; thereby, downplaying on the actual outcomes, success and benefits of an information system.

In this context, DeLone and McLean’s (1992) IS success model and DeLone and McLean’s (2003) updated IS success model, are commonly used frameworks for explaining the cognitive and behavioral facets, of the system's post-adoption (i.e. after acceptance of system) stage, system use and system success (Aparicio, Bacao, & Oliveira, 2017; Tam & Oliveira, 2017; Veeramootoo, Nunkoo, & Dwivedi, 2018). IS success model, first introduced by DeLone and McLean (1992) encompasses six major constructs, “i.e. information quality, system quality, user satisfaction, system use, individual impact and organizational impact”. DeLone and McLean’s (1992) IS success model is often braced for two main reasons, i.e. (1) it presents an in-depth description of many IS success elements and indicators, in a solo, yet comprehensive theoretical framework, (2) this model suggests a time-based and causal interdependence among different categories of indicators of IS success (Veeramootoo et al., 2018; Wang & Liao, 2008). In response to the modification in advance IS environment, DeLone and McLean’s (1992) IS success model was updated, with the inclusion of a new construct, “i.e. service quality” (DeLone & McLean, 2003).

By introducing the concept of service quality, DeLone and McLean’s (2003) updated IS success model, has enhanced the researchers’ propensity to measure and analyze the quality of service, provided by IT department, for a particular information system (Tam & Oliveira, 2017). Literature review has revealed that, DeLone and McLean’s (1992) IS success model and DeLone and McLean’s (2003) updated IS success model, have been frequently applied, for the evaluation of usage and success of various information systems, in several different contexts. For instance, success of e-learning systems (e.g. Aldholay, Isaac, Abdullah, Abdulsalam, & Al-Shibami, 2018; Al-Samarraie, Teng, Alzahrani, & Alalwan, 2017; Aparicio et al., 2017; Cidral et al., 2018; Wang & Chiu, 2011), success of e-commerce systems (e.g. DeLone & McLean, 2004; Tam & Oliveira, 2017) and success of e-government systems (e.g. Finney & Corbett, 2007; Mukred & Yusof, 2018; Rizal, Yussof, Amin, & Chen-Jung, 2018; Veeramootoo et al., 2018; Wang & Liao, 2008). However, these aforementioned scholars (i.e. Al-Samarraie et al., 2017; Aparicio et al., 2017; Cidral et al., 2018; DeLone & McLean, 2004; Finney & Corbett, 2007; Mukred & Yusof, 2018; Rizal et al., 2018; Tam & Oliveira, 2017; Veeramootoo et al., 2018; Wang & Liao, 2008) do not have a unanimous and concerted take, on the dimensions and definition of the success of an information system (Salam, 2020).

The concept of measuring success in the field of IT, is a multidimensional facet, and it’s various dimensions, can be considered, as distinct indicators of success of an information system (Aldholay et al., 2018; Aparicio et al., 2017; Mukred & Yusof, 2018; Wang & Chiu, 2011). According to Martins et al. (2018) the perception of IS success can vary, depending on the different type, purpose (i.e. operational functionality of an information system) and nature of an information system. For example, there are several studies (e.g. Al-Samarraie et al., 2017; Kim & Malhotra, 2005; Mohammadi, 2015) which have regarded the continuous usage, of a particular system, as a measure of IS success. However, some academicians (e.g. Al-Samarraie et al., 2017; Martins et al., 2018; Zolotov, Oliveira, & Casteleyn, 2018) assert that, continuous usage of a system can be a potential outcome of not having any other choice (e.g. a situation in which students have no option, but to register and enroll through a less user-friendly online registration portal). Therefore, continuous usage is not a real indication of the success of an information system (Al-Samarraie et al., 2017; Martins et al., 2018; Zolotov et al., 2018).

According to Zolotov et al. (2018), real success of an e-participation system, can be measured with the citizens’ overall satisfaction and intention to adopt the particular system, for a longer period of time. Further, Martins et al. (2018) and Al-Samarraie et al. (2017) have taken users’ continued satisfaction, as a fomented indication of the success of an e-learning system. While DeLone and McLean’s (2003) updated IS success model has received much support, and it has been efficaciously applied, for the evaluation of various information systems, since many decades. Still, various scholars (e.g. Aldholay et al., 2018; Farooq et al., 2017; Mukred & Yusof, 2018; Salam et al., 2019b) have called for further research, in the domain of success of information systems. More in-depth studies are warranted, to answer the call for research in this domain (Aldholay et al., 2018; Farooq et al., 2017; Mukred & Yusof, 2018; Salam et al., 2019b). Therefore, this study is aimed to extend the existing DeLone and McLean’s (2003) updated IS success model, to assess different new aspects, specific to the success of an information system, by incorporating the concept of “sociability quality”, which is an important element and indicator for assessing the success of a WBCLIS.

This study is grounded on the concept of rationality, and therefore, it is submitted that, different constructs, which are relevant to different situations, should be assessed along with the concept of “sociability quality” for assessing the significance of a WBCLIS. While convening the assessment of education related information systems, it is essential to consider the concept of sociability and collaboration, along with other factors, that can be effective, for facilitating the teaching and learning processes (Martins et al. 2018, 2019; Salam 2020). Collaborative learning technologies have emerged as a wide set of tools, that offer a flexible and eloquent kind of socially interactive platform for collaborative learning (Chan and Pow 2020; Tan 2019). Although, some social networking websites and purpose built e-learning applications, also provide functional support and tools for online collaboration, still the debate about the importance of WBCLIS, between it’s proponents and opponents cannot be to put to rest, without further in-depth investigation of the subject matter (Atan et al. 2012; Doumanis et al. 2019; Mata-Rivera et al. 2015).

Previous studies have also evaluated different types of education related information systems; for instance, Martins et al. (2018) have evaluated the success of education management system, and Cidral et al. (2018) reported the success factors of e-learning system. However, still the significance and success factors of sociability features, offered by WBCLIS, are not completely known to date (Salam, 2020). Therefore, in an effort to address this gap in the IS literature, this study incorporates a new construct of “sociability quality” (i.e. adapted from Kreijns et al.’s (2004, 2007, 2002) concept of sociability for CSCL systems) in the DeLone and McLean’s (2003) updated IS success model, to explore the success factors of a WBCLIS. In this context, this study argues that, sociability component (i.e. sociability quality) is as much important, for a WBCLIS, as are it’s other components, related to system quality or information quality etc. Following section presents more details, regarding major constructs involved in this study, along with hypotheses development and proposed theoretical model, for assessing the sociability quality of a WBCLIS.

Constructs, hypotheses development and proposed theoretical model

System quality

The functional eminence and quality of CSCL systems is considered their imperative chip, for offering a smooth collaborative learning experience, in a project based collaborative learning context (Dado and Bodemer 2017; Jeong et al. 2019; Lin et al. 2016; Zheng et al. 2019). Therefore, system quality has a significant impact on the functionality, performance characteristics and usability of a CSCL system (Bhuasiri et al. 2012; Cidral et al. 2018). System quality, of a CSCL system, is described as it’s functional and technical characteristic (Garcia-Sanjuan et al. 2018; Suebnukarn and Haddawy 2006; Ward 1998). Information systems are expected to have a momentous level of system quality, because it has a significant impact on the overall acceptance and adoption of a system (von Hellens 1997; Rangraz Jeddi et al. 2020; Salmela 1997). Moreover, it is an indispensable dimension, of the usefulness, of any information system (von Hellens 1997; Rangraz Jeddi et al. 2020). System quality defines the extent, to which an information system is successful, in real world environment, for performing multiple simultaneous functions, while achieving the agreed upon deliverables, by it’s all stakeholders (von Hellens 1997; Martins et al. 2018; Rangraz Jeddi et al. 2020). Previous studies (e.g. Aparicio et al. 2017; Balaban et al., 2013; Cidral et al. 2018; Martins et al. 2018) have also acknowledged that, system quality has a direct positive influence on the system use and user satisfaction.

System quality reflects the attainment of basic functionalities, in an efficient and evocative manner, for what a specific system was developed (von Hellens 1997; Martins et al. 2018; Rangraz Jeddi et al. 2020; Salmela 1997). More precisely, for education related information systems, their pedagogical functionalities are essential features, that enable users to effectively and efficiently interconnect with the system, for teaching and learning activities (Apostolou et al. 2014; Forster et al. 2020; Hamidi and Jahanshaheefard 2019; Martins et al. 2019). Moreover, CSCL systems are helpful in accessing online academic resources (e.g. course outlines and course contents etc.), along with information sharing, regarding other allied learning opportunities (e.g. upcoming events / milestones achieved during the course etc.), while promoting a dynamic interaction between teachers and students, throughout their whole learning process (Aparicio et al., 2016; Martins et al. 2018). In addition to this, Urbach et al. (2010) and Tarhini et al. (2017) have also emphasized on the importance of system structure, graphical logic design, accessibility, navigational suppleness, sustainability and reliability of online learning systems, as the indicators of a system’s operational quality. System quality is important to ensure the overall eminence, of collaborative learning experience, offered by a WBCLIS (Rangraz Jeddi et al. 2020; Salam 2020; Shim and Jo 2020).

While assessing the acceptance and usefulness of CSCL systems or similar educational technologies, users’ satisfaction towards these systems, coupled with their functional and technical quality, is always considered as an imperative factor (Apostolou et al. 2014; Atan et al. 2012; Forster et al. 2020; Hamidi and Jahanshaheefard 2019). In this context, Martins et al. (2018) and Cidral et al. (2018) have also argued that, overall quality of education-related information systems (i.e. also referred as e-learning systems) should be assessed on the base of systems’ ability to be used easily, along with its’ ability to persistently and appropriately, offer all necessary features, in a standardized and sustainable manner. Attainment of various modulated aspects, along with mutually predefined features, required by all users and stakeholders, is a constituent prerequisite of a CSCL system’s quality (Aslan and Hamurcu 2015; Dado and Bodemer 2017; Gress et al. 2010; Reis et al. 2018; Zheng et al. 2019). Though, previous studies on system quality (e.g. Aparicio et al. 2017; Balaban et al. 2013; Cidral et al. 2018; Martins et al. 2018) have explored it’s significance, for various other systems (e.g. e-learning / education management information systems, and e-portfolio systems etc.); however, less is known about the impact of system quality, on the system use and user satisfaction, in the case of WBCLIS. Therefore, further studies are required, to address this research gap, in IS literature (Balaban et al. 2013; Martins et al. 2018; Salam et al. 2019a). Developing on the aforementioned coherent arguments, this study submits following propositions:

• H1a: System quality has a positive impact on system use

• H1b: System quality has a positive impact on user satisfaction

Information quality

Information quality of a system, represents the characteristics of an information, generated as an output of a particular system (DeLone and McLean 2003; Riesener et al. 2019; Timmerman and Bronselaer 2019; Zhang et al. 2020). Information quality has been measured by evaluating the reliability, accuracy and trustworthiness of the information and results, produced by a specific system (Gerber et al., 2004; Michel-Verkerke 2012; Ojo 2017; Rangraz Jeddi et al. 2020; Shim and Jo 2020). In the context of online learning and CSCL, information quality particularly refers to the extent, to which system users perceive, that learning or education related academic information, available through their CSCL system, is adequate, accurate, reliable, comprehensive, relevant, organized and up-to-date for a rewardingly decent learning experience (Aldholay et al. 2018; Cidral et al. 2018; El-Masri and Tarhini 2017). Information quality has a strong influence on students’ satisfaction, system success and net benefits (Balaban et al. 2013; Cidral et al. 2018; Costa et al. 2016). Moreover, information quality, determines the tendency towards system use, and users’ overall satisfaction; therefore, various studies (e.g. Aparicio et al. 2017; Balaban et al. 2013; Bhuasiri et al. 2012; Cidral et al. 2018; Wang and Chiu 2011) have described that, information quality is an important factor, for the success of any e-leaning system.

In various distance learning programs (i.e. also known as online education system), students spend their considerable time and efforts to acquire education related academic information, for a productive e-learning experience (Al-Samarraie et al. 2017; Churchill 2011; Martins et al. 2018). Same is the case, with blended education system, offering service learning, through a collaborative environment (Mukred and Yusof 2018; Rangraz Jeddi et al. 2020; Salam et al. 2019a). According to Martins et al. (2018), academic institutions are responsible to ensure that, all information shared on the platform of their CSCL system is accurate, reliable and organized, in a way, that facilitates the conferring of whole learning processes. Previous studies on e-learning and CSCL (e.g. Aparicio et al. 2017; Balaban et al. 2013; Bhuasiri et al. 2012; Cidral et al. 2018; Wang and Chiu 2011) have found that, standard and quality of information characteristics, are positively linked to system use and users’ satisfaction. However, significance of information quality, in the context of WBCLIS, has remained largely unaddressed, by most of the aforementioned studies (Atan et al. 2012; Liaw et al. 2008; Shen 2012). Therefore, in order to address this gap, in the IS literature, regarding the evaluation of WBCLIS, this study submits that:

• H2a: Information quality has a positive impact on system use.

• H2b: Information quality has a positive impact on user satisfaction.

Service quality

Service quality is another important dimension of IS success (Li and Shang 2019; Martins et al. 2018; Ojo 2017; Yang et al. 2017). DeLone and McLean’s (2003) updated IS success model postulates that, quality of service, has a significant impact on the system use and subsequent user satisfaction. Service quality denotes to the quality of professional support and after deal service, provided by the system administrators and developers (Al-Fraihat et al. 2020; Bharati and Berg 2005; Li and Shang 2019; Mohammadi 2015; Shim and Jo 2020). Service quality of a CSCL system is assessed through various attributes, e.g. responsiveness, readiness, reliability, assurance, suppleness, interactivity and facility of tutorial / training for users, to help them in utilizing all features of a system (Aldholay et al. 2018; Ojo 2017; Urbach et al. 2010; Wang and Teo 2020). In the perspective of success, of online education / e-learning systems, findings reported by previous studies (e.g. Aparicio et al. 2017; Balaban et al. 2013; Wang and Chiu 2011) have demonstrated a positive association, between users’ perceived service quality and system use. A recent study by Yang et al. (2017), reported a positive relationship, among service quality and users’ long term intention to use open source online academic systems. Moreover, service quality is also a strong determinant, of users’ satisfaction towards e-learning systems (ChanLin 2012; Farooq et al. 2018; Martins et al. 2018; Veeramootoo et al. 2018).

When considering education-related information systems, the perceived service quality, has a high impact on users’ overall satisfaction level (Bhuasiri et al. 2012; Mohammadi 2015; Urbach et al. 2010). If service quality is high, it increases users’ satisfaction; thereby, discouraging them from opting any other alternatives (Balaban et al. 2013; Bharati and Berg 2005; El-Masri and Tarhini 2017; Shim and Jo 2020). Moreover, recently Cidral et al. (2018) and Martins et al. (2018) have also validated a positive association, among service quality and system use. However, this study argues that, WBCLIS have a unique nature, as they offer a platform for social collaboration, between teachers and students. Therefore, more studies are warranted, to explore the phenomenon of service quality, for WBCLIS, to better understand all the facets, of their use and users’ satisfaction towards WBCLIS. For that reason, based on the above discussion, this study submits following hypotheses in the context of WBCLIS:

• H3a: Service quality has a positive impact on system use.

• H3b: Service quality has a positive impact on user satisfaction.

Sociability quality

According to Kreijns et al. (2007:176) sociability of a CSCL system refers to “the extent to which a CSCL environment is perceived to be able to facilitate the emergence of a sound social space with attributes as trust and belonging, a strong sense of community, and good working relationships” (Kreijns et al. 2007:176). Sociability of a CSCL system, can be measured with it’s ability, to serve as a reliable mediated environment, to assist the development, of a strong social and collaborative relationship, between it’s all users (e.g. students and teachers etc.) and stakeholders (Gao et al. 2010; Junglas et al. 2013; Kreijns et al. 2004, 2007, 2002; Salam et al. 2019b, Salam et al., 2019a; Yamada et al. 2016). Further, Gao et al. (2010) have examined the relation, between sociability of a social software and it’s impact on users’ attitude and intention, towards long term use of a particular software. Findings of Gao et al. (2010) have confirmed that, sociability has a strong and significant influence on both, users’ attitude towards initial application of the social software, and their intention to use, that particular software, for a longer period of time.

Likewise, another study by Junglas et al. (2013) has also confirmed that, sociability (i.e. taken as a social component) has a positive impact on system use, and it has a significant role, in determining users’ overall tendency, towards technology acceptance. Moreover, various other studies (e.g. Gao et al. 2010; Junglas et al. 2013; Kreijns et al. 2007; Kreijns and Kirschner 2004; Oksanen and Hämäläinen 2013) have also acknowledged the significance of sociability component, and have attributed the presence of sociability, for it’s contributions, towards maximizing the usage of collaborative technologies. Although, aforementioned studies (e.g. Gao et al. 2010; Junglas et al. 2013; Kreijns et al. 2007; Kreijns and Kirschner 2004; Oksanen and Hämäläinen 2013) have explored the concept of sociability, for it’s significance, in enhancing system use. Yet, absence of a comprehensive framework for assessing sociability of WBCLIS, is still a major gap, in the IS literature (Salam 2020; Salam et al. 2019b).

As mentioned earlier, DeLone and McLean’s (2003) updated IS success model is most acclaimed and commended comprehensive framework, for explaining and assessing the success of information systems. Since, aforementioned studies on the sociability component (e.g. Gao et al. 2010; Junglas et al. 2013; Kreijns et al. 2007; Kreijns and Kirschner 2004; Oksanen and Hämäläinen 2013) have not investigated the concept of sociability, in the presence of other IS success factors, identified by DeLone and McLean’s (2003) updated IS success model. Therefore, this study argues that, previous work on the component of sociability (e.g. Gao et al. 2010; Junglas et al. 2013; Kreijns et al. 2007; Kreijns and Kirschner 2004; Oksanen and Hämäläinen 2013), has failed to offer a complete picture, of all the facets of sociability, in the context of WBCLIS. For that purpose, more studies are required, to explore the element of sociability, in the context of WBCLIS.

Building on the aforementioned, Kreijns et al.’s (2004, 2007, 2002) concept of sociability for CSCL systems, this study suggests that, WBCLIS have more to offer, in terms of a collaborative and social platform for higher education (ChanLin 2012; Jan et al. 2016; Salam 2020; Shen 2012). Further, this study argues that, mere presence of sociability features is not enough, for collaboration between teachers and students, rather high quality of sociability, is what makes a WBCLIS successful. Therefore, inspired by Kreijns et al.’s (2007) concept of sociability for CSCL, this study proposes a new terminology / dimension of quality (i.e. sociability quality), to be assessed along with three existing quality dimensions, “i.e. system quality, information quality and service quality” of DeLone and McLean’s (2003) updated IS success model. This study defines sociability quality, as the degree, to which a WBCLIS is capable, of offering eminent support, for online social interaction, as a dependable and trustworthy platform, for collaboration between teachers and students. Moreover, this study asserts that, similar to the three major quality components of DeLone and McLean’s (2003) updated IS success model, “i.e. system quality, information quality and service quality”, sociability quality is also an important facet, which has been overlooked in previous IS literature (Al-Fraihat et al. 2020; Vitale and Udell 2019; White et al. 2017; Yoder et al., 2019). Therefore, with respect to the preceding discussion, on the concept of sociability, sociability quality and importance of social collaboration in WBCLIS, this study submits following hypotheses:

• H4a: Sociability quality has a positive impact on system use.

• H4b: Sociability quality has a positive impact on user satisfaction.

User satisfaction

User satisfaction is known as one of the imperative factors, for evaluating the success of an information system (Balaban et al. 2013; Forster et al. 2020; Ojo 2017; Salam 2020; Santa et al., 2019; Xu and Du 2019). Various studies (e.g. Aldholay et al. 2018; DeLone and McLean 2003) have confirmed that, user satisfaction has a direct and significant relationship, with overall system use. User satisfaction is described as, the extent to which, a user, perceives a particular CSCL system, to be useful and effective, for achieving his / her objectives (DeLone and McLean 2004). In the context of online education / e-learning, various scholars (e.g. Aparicio et al. 2017; Cidral et al. 2018; Hassanzadeh et al., 2012; Wang and Liao 2008) have emphasized on the importance of user satisfaction. User satisfaction of an academic information system, is determined by the extent, to which it’s users (i.e. teachers and students etc.) are contented with the system functionalities, for a productive learning experience, and how well is it’s performance, to meet the expectations of all the stakeholders (Aparicio et al. 2017; Cidral et al. 2018; Hassanzadeh et al. 2012; Wang and Liao 2008).

User satisfaction towards a WBCLIS, is also reflected by users’ intention and willingness, to use it, again and again, for a longer period of time (DeLone and McLean 2004; Gluck 1996; Salam 2020). Among all the internal and external stakeholders; students, who are using an e-learning system, are the real legatees, of a CSCL system (Hassanzadeh et al. 2012; Salam 2020). The relationship among system use and user satisfaction, has been validated, in a number of studies (e.g. Al-Samarraie et al. 2017; Balaban et al. 2013; Martins et al. 2018), conducted in the domain of e-learning information systems. Specifically, in the context of WBCLIS, Salam (2020) has acknowledged that, if a user is more satisfied with a particular system, his / her intention to use that particular system, is proportionately enhanced. Further, Cidral et al. (2018) while performing a review of online education / e-learning systems’ effectiveness, argued that, measurement of user satisfaction, is crucial in the context of e-learning, to assess the long term adoption of an information system, used for academic purposes.

Moreover, according to Balaban et al. (2013) and Martins et al. (2018), user satisfaction is also a strong determinant, of the net benefits of an information system. Although, aforementioned studies (e.g. Al-Samarraie et al. 2017; Aparicio et al. 2017; Balaban et al. 2013; Cidral et al. 2018; Hassanzadeh et al. 2012; Martins et al. 2018; Wang and Liao 2008), have thoroughly explained the concept of user satisfaction (Mellikeche et al. 2020; Salam 2020). However, this study argues that, WBCLIS itself is unique, in it's nature; and therefore, it’s success is determined by it’s sociability quality (Salam 2020; Santa et al. 2019). For that reason, impact of user satisfaction, on overall system use and net benefits, warrants more research, in the domain of modern academic information systems, specifically with the reference of WBCLIS (Mellikeche et al. 2020; Salam 2020; Xu and Du 2019). Therefore, developing on the base of above mentioned logical arguments, and in order to extend our understanding, regarding the role of user satisfaction, in the context of WBCLIS, this study hypothesizes that:

• H5: User satisfaction has a positive impact on system use.

• H6: User satisfaction has a positive impact on net benefits.

System use

System use denotes to the extent to which a user uses, the entire spectrum, of the available features, in a particular system, to fulfill his / her needs (Handayani et al. 2019; Li et al. 2020; Muczyński et al., 2019; Salam 2020). More precisely, system use is concerned with evaluating, the overall use of an information system, by its’ users (El-Masri and Tarhini 2017; Farooq et al. 2017; Liaw et al. 2008). Various previous studies (e.g. Alzahrani et al. 2017; Balaban et al. 2013; El-Masri and Tarhini 2017; Farooq et al. 2017; Salam 2020) have measured the concept of system use, with a number of different aspects, e.g. frequency of system use, nature and duration of system use, and users’ perception towards perceived usefulness and effectiveness of a system, for fulfilling their needs. Additionally, aforementioned studies (e.g. Alzahrani et al. 2017; Balaban et al. 2013; El-Masri and Tarhini 2017; Farooq et al. 2017; Salam 2020) have also demonstrated that, various factors influence the long term use of an e-learning system, e.g. ease of use, and the provision of quality information etc. Moreover, systems’ reliability and functional structure, for facilitating the collaborative learning process, also effects the use of a CSCL system, in higher education (Gress et al. 2010; Reis et al. 2018; Salam 2020; Zheng et al. 2019).

Various allied features, e.g. start-up guidelines, tutorials and instruction manuals, also facilitate the use and acceptance of a CSCL system (Handayani et al. 2019; Li et al. 2020; Muczyński et al. 2019; Salam 2020). Moreover, a fair level of backup support, also offers a peace of mind, which usually leads to increase in system use, followed by an upsurge, in the overall net benefits (Gelderman 1998; Gluck 1996; Martins et al. 2019; Muczyński et al. 2019; Rangraz Jeddi et al. 2020; Salam 2020). Recently, Martins et al. (2018) have also confirmed, a strong and positive connection, among the use of education related information systems / e-learning systems, user satisfaction and apparent net benefits. Moreover, with reference to education related information systems, both Balaban et al. (2013) and Martins et al. (2018) have also reported a positive association, between system use and it’s inherent net benefits. However, this study argues that, as a social platform, outcomes of collaborative learning process, through WBCLIS can only be materialized in terms of it’s net benefits, when all users (i.e. teachers and students etc.) have a relationship of mutual trust (Handayani et al. 2019; Li et al. 2020; Salam 2020). A relationship of trust and confidence, between group members, is developed by continuous use of a WBCLIS, which leads to an increase in the net benefits (Muczyński et al. 2019; Salam 2020; Santa et al. 2019).

WBCLIS offers a convenient way, to implement collaborative learning, in the service learning pedagogy (Muczyński et al. 2019; Salam 2020; Salam et al. 2019a). As a popular mode of experiential learning, in-group collaboration, is a hallmark of service learning projects, which is made possible, with the right and timely, implementation of WBCLIS (Li et al. 2020; Salam 2020). However, not much is known, about the net benefits, derived from the use of a WBCLIS to date (Handayani et al. 2019; Salam 2020; Salam et al. 2019c; Santa et al. 2019). Therefore, extending the work of aforementioned studies (e.g. Balaban et al. 2013; Martins et al. 2018; Salam 2020), this study suggests that, system use of WBCLIS can offer more net benefits to users, as a platform of social information sharing and collaborative learning. In order to explore the proposition of positive connection, among system use and increase in net benefits, in the context of WBCLIS, this study submits following hypothesis:

• H7: System use has a positive impact on net benefits.

Net benefits

The concept of net benefits is defined, as to what extent, an information system, adds to the accomplishments of an individual, group, or an organization (Alzahrani et al. 2017; DeLone and McLean 2003; Hassanzadeh et al. 2012; Makhni 2020). The construct of net benefits is designed to encompass, all expected benefits of an information system, for its’ different stakeholders (Bhuasiri et al. 2012; Urbach et al. 2010; Zhang and Thompson 2019). In the initial version of DeLone and McLean’s (1992) IS success model, outcomes and contributions of an information system, were measured as a manifestation, of it’s overall individual impact or organizational impact (Hendra, S. Kom., and Arifin, S. Kom., 2018; Love et al., 2020; Salam 2020). However, developing on DeLone and McLean’s (2003) updated IS success model, this study has examined the contributions, of a WBCLIS as it’s net benefits. According to Balaban et al. (2013) the connotations of net benefits, are elusive in nature and therefore, net benefits are case specific. Moreover, Balaban et al. (2013) have noted that, the concept of net benefits may differ, according to the needs of each user, and it is totally dependent on the type and objectives of an information system.

Various scholars (e.g. Alzahrani et al. 2017; Anaya 2013; Balaban et al. 2013; DeLone and McLean 2003; Kurkalova and Carter 2017; Marnewick 2016; Martins et al. 2018; Salam 2020) have acknowledged that, the determinants of net benefits, are not fully known to date. Therefore, Balaban et al. (2013) and Martins et al. (2018) have asserted, on the need for more studies, to explore the determinants of net benefits, in the domain of education related information systems (Peña-Miguel and De La Peña 2018; Salam 2020; Yang and Lam 2019). Answering their call for more studies, in the domain of IS, for exploring new IS success dimensions, this study has proposed “sociability quality” as a new proponent of the net benefits, in the DeLone and McLean’s (2003) updated IS success model.

Theoretical model

Above-mentioned sections have presented an in-depth discussion, regarding a number of success factors, that can practically add to the success, of an information system. On the base of previously mentioned, detailed discussion on the subject matter, and developing on the logical propositions submitted earlier, a proposed theoretical model is presented in Fig. 1. Considering the unique nature of sociability, for collaborative learning, in a WBCLIS, this proposed theoretical model, integrates sociability quality, as an exogenous construct, and as a success dimension, for WBCLIS system use and user satisfaction. Hence, proposed theoretical model comprises of four exogenous independent variables, “i.e. system quality, information quality, service quality and sociability quality”, and three endogenous dependent variables, “i.e. user satisfaction, system use and net benefits”.

Fig. 1

Proposed Theoretical Model

As described earlier, this study is aimed to assess the role of sociability quality, in determining the success of WBCLIS users’ satisfaction and system usage. Therefore, in order to assess user satisfaction, system use and other IS success factors, in the context of WBCLIS, a comprehensive multi-construct theoretical model is proposed in this study (see Fig. 1). By proposing a new construct of “sociability quality”, this study has extended the DeLone and McLean’s (2003) updated IS success model, for assessing the sociability component of WBCLIS. It is expected that, sociability quality will help in explaining the hidden facets, of new IS success dimensions (Salam 2020; Vitale and Udell 2019; White et al. 2017). Proposed theoretical model of this study, is assessed by evaluating the success of a newly developed WBCLIS, implemented for facilitating the teachers and students, working on a collaborative service learning project, in a public sector university, in Malaysia. Further discussion regarding the assessment approach and research methodology, used for assessing the proposed theoretical model of this study, is described in next section.

Research methodology

Design of the Study

Proposed theoretical model of this study, integrates sociability quality, in the DeLone and McLean’s (2003) updated IS success model. Therefore, this study is exploratory in nature, and is aimed to explore the relationship, of sociability quality, with the system use, user satisfaction and net benefits of a WBCLIS. Moreover, this study is an effort of it’s own kind, which has assessed, the success dimensions of a WBCLIS, with a practical approach, in the project based collaborative and service learning context. As a practical part of this study, a flagship WBCLIS (i.e. the Co-Reflective), with a concentrated focus on the sociability quality, was developed, for testing the proposed theoretical model of this study. Further, for empirical part of this study, a variance based PLS-SEM statistical approach was employed. More details, regarding the recruitment of participants, research setting and data analysis, are presented in the upcoming sections.

Participants

Data was collected from students who have been actively involved in a collaborative learning project, using a WBCLIS (i.e. the Co-Reflective) system. Right selection of participants and correct sample size, is necessary, to determine the sanctity of an exploratory study (Farooq and Salam 2020; Hair et al. 2014, Hair et al., 2017). In this context, Barclay et al. (1995) have proposed the use of “10 times rule” as a basic guideline, for determining a minimum sample size, required for a PLS-SEM analysis. According to Hair et al. (2017:20) “minimum sample size should be 10 times the maximum number of arrowheads pointing at a latent variable anywhere in the PLS path model” (Hair et al. 2017:20). The proposed theoretical model of this study (see Fig. 1) includes seven constructs (i.e. four independent, two second order dependent and one third order dependent variable), and maximum four arrowheads are pointing towards system use and user satisfaction. Therefore, based on the aforementioned 10 times rule, minimum sample size for this study, should be 40 respondents. In view of the exploratory nature, and experimental design of this study, all 120 students, who were enrolled in the service learning course, and had worked on a collaborative project, using WBCLIS, were considered as potential participants of this study.

Research setting

This experimental research has been conducted in a service learning course, at a public sector university in Malaysia, using a newly developed WBCLIS, named as the Co-Reflective. As part of their service learning course, students were divided in small groups (i.e. 3 to 5 students in each group), to perform their service learning project (i.e. involving collaboration with students and teachers), at different locations. Being a web-based system, the Co-Reflective was accessible to all users (i.e. teachers and students etc.), all the time and without any constraints of their geographical location. Through the use of WBCLIS, students could perform their service learning project-related assignments, while taking lessons, as per their own convenience. Users were able to operate this WBCLIS, on their mobile devices and laptops, after a simple login / registration process, via any web-browser (e.g. Google Chrome and Internet Explorer etc.). WBCLIS enabled students to collaborate, during their service learning project. Service learning is a kind of experiential learning, in which real life problem based projects are assigned to students, working in collaboration with their teachers (Salam 2020; Salam et al. 2019a).

As part of their service learning class project, students were required to perform their collaborative learning assignments, using the platform of WBCLIS. Moreover, reporting their collaborative efforts for group assignments, and recording their participation in group reflection, through WBCLIS, after every field visit on the service site, was also a part of their service learning project. Therefore, in order to complete their group based service learning project, students were committed to work in collaboration with their teachers. WBCLIS offered a separate interface for teachers, where teachers could design different collaborative assignments, for assessing students’ in-group collaboration, learning outcomes and overall group reflection. Students used this WBCLIS, throughout their whole semester, as a part of their service learning group project. At the end of the semester, students were voluntarily invited, to participate in this study, by completing an online self-administered survey based questionnaire, designed for assessing the sociability quality of WBCLIS. Details regarding the preparation of questionnaire, and data collection process, are presented in the next section.

Questionnaire design, pilot study and data collection

As mentioned earlier, after the use of WBCLIS (i.e. the Co-Reflective) for whole semester, students were invited, with their informed prior consent, to voluntarily participate, in a self-administered online questionnaire based survey. An online questionnaire was particularly designed, to measure all the constructs involved in the proposed theoretical model of this study (see Fig. 1). The survey instrument comprised of 48 items, used to measure all seven constructs “i.e. system quality, information quality, service quality, sociability quality, user satisfaction, system use and net benefits”, of the proposed theoretical model. In order to finalize the items involved in the survey instrument, various relevant studies (e.g. Balaban et al. 2013; Brown et al. 2010; DeLone and McLean 2003; Kreijns et al. 2007; Martins et al. 2018; Mohammadi 2015), were consulted thoroughly, for ensuring the validity and reliability of this study. More precisely, a nine items scale was adapted from relevant previous studies (e.g. Balaban et al. 2013; DeLone and McLean 2003; Mohammadi 2015), for assessing system quality. Further, five items scale was adapted from Martins et al. (2018), Balaban et al. (2013), and DeLone and McLean (2003) for measuring information quality, of the under study WBCLIS.

Furthermore, service quality was assessed with a five items scale, adapted from Balaban et al. (2013), DeLone and McLean (2003), and Wang and Liao (2008). Likewise, an eight items scale was adapted from previously published, sociability related studies (e.g. Gao et al. 2010; Junglas et al. 2013; Kreijns et al. 2007; Yamada et al. 2016), for assessing the newly proposed construct of sociability quality. Further, a four items scale was adapted from Alzahrani et al. (2017) and Balaban et al. (2013) for evaluating the system use. Similarly, another four items scale, was borrowed from Balaban et al. (2013) and DeLone and McLean (2003) for assessing user satisfaction. Moreover, acknowledging this fact, that net benefits are case sensitive and they depend on the domain, nature, and characteristics of an information system, a single item scale, is inadequate, for assessing all the facets, of a particular information system (Gress et al. 2010; Martins et al. 2019; Rehak 2020; Salam 2020). Therefore, various levels, contexts, sub-categories and referred system features, were also deliberately considered, while measuring the net benefits of a WBCLIS in this study. Finally, a thirteen items comprehensive scale was adapted, from previous relevant studies (e.g. Balaban et al. 2013; Martins et al. 2018; Wang and Wang 2009), for measuring the perceived net benefits, derived from the use of WBCLIS. Complete list of measurement items, used in this study, is attached in Table 7 in Appendix.

A seven points, Likert type scale was used, to record the responses from all participants. Furthermore, in addition to the aforementioned 48 items, two more questions were also added, in the final questionnaire, for recording the demographic attributes (i.e. gender and age), of all participants. Moreover, before final data collection, a pilot study was also conducted, with a random sample of 10 students (i.e. selected from students’ enrolment list), from service learning course. Based on the results of that pilot study, it was found that, all measurement scales, used in this study, have an acceptable level of composite reliability and Cronbach alpha. After establishing the reliability and validity of all measurement scales, (i.e. through pilot study) final data collection started with a convenience sampling approach. As mentioned earlier, all 120 students, enrolled in the service learning class, were considered as potential participants of this study. Therefore, all 120 students, were invited to participate in the final data collection. At the end of survey completion deadline, total 95 responses were recorded; however, 15 inappropriate / incomplete responses were discarded, and only 80 complete responses, were used for final data analysis. Hence, a 66.6% response rate was recorded, which adds to the parsimony of this study.

Data analysis

In this study, partial least square based structural equation modelling (PLS-SEM) statistical approach was applied, to assess the explanatory and predictive power of the proposed theoretical model. Data analysis was performed, using latest available version 3.2.7 of the SmartPLS software. The PLS-SEM method is suitable, when one cannot guarantee the normality of data (Chin 2010; Hair et al. 2014). Further, choice of PLS-SEM approach, was also motivated by it’s ability to work with the formative variables, in exploratory studies, even with small sample size (Hair et al., 2011; Henseler et al., 2016). Particularly, in this study, PLS-SEM based statistical approach, was used for two major reasons, i.e. (1) data collected for this study is not normally distributed (Hair et al. 2014; Henseler et al. 2016), and (2) proposed theoretical model is exploratory in nature, and is still in the development phase (Chin 1998; Haque et al., 2009; Shanmugapriya and Subramanian 2015). Demographic information revealed that, 87.7% of our participants, were between the age group of 20–25 years. Moreover, it was found that, 56.8% of our participants, were female and 43.2% were male. Further, results derived from the data analysis of this study, are described in next section.

Results

Considering the best practices for implementing PLS-SEM, results were thoroughly assessed, for evaluating the measurement models, along with structural model, involved in the proposed theoretical model of this study. Measurement model refers to the scales, used to measure the underlying constructs (Hair et al. 2014). However, structural model, refers to the inner part of the proposed theoretical model (Hair et al. 2014). Observing the guidelines suggested by Hair et al. (2014), results derived from the data analysis, are categorized in following two sections, i.e. (1) evaluation of measurement models (i.e. aimed to assess the reliability and validity of our measurement models), and (2) evaluation of structural model (i.e. aimed to assess the significance of our proposed research hypotheses, along with explanatory and predictive power, of the proposed theoretical model). Now, following sub-sections present these results, starting with an exhaustive assessment of the measurement models, followed by a full-scale assessment of the structural model.

Evaluation of measurement models

Evaluation of measurement models begins with the assessment of constructs’ reliability and validity (Hair et al. 2014). Table 1 presents the results, of constructs’ reliability and validity. As per the guidelines of Henseler et al. (2009), and as an indicator of reliability, items’ loading values, should be greater than 0.70 threshold. As depicted in the Table 1, all items in this study, have a greater than 0.70 loading value, except for three items (i.e. SRQ_5, NB_5 and NB_9). Although, these aforesaid items, obtained a loading value, which is less than 0.70 threshold. However, removal of these items (i.e. SRQ_5, NB_5 and NB_9), with lower loading values, did not had a major influence on the overall reliability. Therefore, observing the suggestions of Hair et al. (2014), no items were eliminated from the measurement models. Moreover, as shown in the Table 1, all Cronbach alpha values and composite reliability values, are well above the threshold level, which is another indication of reliability of our measurement models (Hair et al. 2014; Henseler et al. 2009). Hence, overall these results, shown in the Table 1, have confirmed that, all items involved, in the measurement instrument of this study, have a good reliability.

Table 1 Results of Measurement Model, Construct Reliability and Validity

Further, various other tests were also performed, to assess the validity of all measurement models. Average variance extracted (AVE) is commonly used, for the assessment of convergent validity (Hair et al. 2014). As presented in the Table 1, all AVE values are greater than 0.50 threshold (i.e. ranging between 0.631 to 0.755), which shows that, the variance of all constructs, is larger than the variation triggered by the individual measurement errors (Hair et al. 2014; Segars 1997). Hence, these results confirm that, all measurement models have an acceptable level of convergent validity (Hair et al. 2011).

Further, various tests (i.e. cross loading values, Fornell and Larcker criterion, and Heterotrait-Monotrait ratio analysis) for checking the discriminant validity, of measurement models, were also performed. Meeting the criterion of cross loading, Table 2 shows that, all loading values, are greater, than their corresponding cross loadings, on other constructs. Conditions of Fornell and Larcker’s (1981) criterion are also met, as depicted in the Table 3, all correlation values, between the constructs, are less than the square root values, of their estimated average variance extracted (AVE). Further, results presented in the Table 4 have also fulfilled the conditions of Heterotrait-Monotrait (HTMT) ratio analysis. As depicted in the Table 4 all HTMT values, are less than 0.9 threshold (Henseler et al., 2015). These results have confirmed the discriminant validity, of all measurement models. Overall, results presented in this section, have thoroughly evaluated the measurement models. Results of measurement models assessment have revealed that, all measurement models are reliable and valid. After the assessment of measurement models, this study continues with the analysis of structural model. In order to assess the significance of proposed hypothetical relations, next section presents the results of structural model.

Table 2 Cross Loading Analysis

Table 3 Fornell-Larcker Criterion

Table 4 Heterotrait-Monotrait (HTMT)

Evaluation of structural model

In order to evaluate the structural model, first of all multicollinearity of all constructs was tested, by variance inflation factor (VIF) criterion. Results confirmed that, all VIF values range between 1.043 to 2.948, and are well below, than the threshold value of 5, as suggested by Hair et al. (2014:126, 2017). These results have established the absence of multicollinearity, in the proposed structural model (Hair et al. 2014, Hair et al., 2017). After assessing the proposed structural model, for potential multicollinearity, it was thoroughly assessed for path coefficient values (i.e. β values), significance of relationships (i.e. t-values) between constructs, along with it’s explanatory power (i.e. R² values). For this purpose, PLS algorithm was performed, followed by a bootstrapping procedure. Observing the suggestions of Hair et al. (2014, 2017), bootstrapping process was executed with 5000 resamples, to measure the significance of path coefficient values. The results of structural model are described in the Fig. 2, and Table 5.

Fig. 2

Findings of Structural Model

Table 5 Hypotheses Assessment

As depicted in the Fig. 2, proposed theoretical model, explains a substantial variation, in the dependent endogenous constructs of this study. More precisely, with a 70.9% explanatory power (i.e. R² = 0.709) for net benefits, followed by 66.2% explanatory power (i.e. R² = 0.662) for system use, and 61.5% explanatory power (i.e. R² = 0.615) for user satisfaction towards WBCLIS, this proposed structural model, has evidently advanced the DeLone and McLean’s (2003) updated IS success model. Further, results of path coefficients and bootstrapping procedure, have confirmed that, most of the hypotheses in the proposed structural model were supported, except for H1a, H2a, H2b and H3a. Since, majority of the proposed hypotheses are significant, and endogenous constructs have a high explanatory power; therefore, we suggest that, proposed theoretical model has imparted substantial theoretical contribution, by assessing the construct of sociability quality, along with other existing dimensions of the DeLone and McLean’s (2003) updated IS success model.

As depicted in the Table 5, endogenous construct ‘system use’ is significantly explained by sociability quality (β = 0.399, t-value = 3.368, p < 0.01) and user satisfaction (β = 0.354, t-value = 2.709, p < 0.01), consequently confirming that H4a and H5 are significant. Moreover, endogenous construct ‘user satisfaction’ is significantly explained by system quality (β = 0.404, t-value = 3.353, p < 0.01), service quality (β = 0.187, t-value = 2.350, p < 0.05), and sociability quality (β = 0.298, t-value = 2.211, p < 0.05), thus H1b, H3b and H4b are also supported. Further, as shown in the Fig. 2, proposed theoretical model, explains 70.9% variation (i.e. R² = 0.709) in the third order endogenous construct, i.e. ‘net benefits’, where both, user satisfaction (β = 0.446, t-value = 4.651, p < 0.01) and system use (β = 0.458, t-value = 4.640, p < 0.01) have a significant impact, on the net benefits derived from WBCLIS. Hence, H6 and H7 are also supported.

In addition to the aforementioned assessment, predictive relevance of the proposed theoretical model was also evaluated, by employing Stone (1974) and Geisser’s (1974) Q² test. As per rule of thumb, a higher Q² value, indicates higher predictive relevance, of a proposed theoretical model (Chin 2010; Henseler et al. 2009). The results confirm that, Q² values of all endogenous constructs, were greater than zero. Starting with system use (Q² = 0.445), followed by, user satisfaction (Q² = 0.417), and net benefits (Q² = 0.405), all Q² values indicate that, proposed theoretical model has a good predictive relevance, for assessing WBCLIS, for it’s sociability quality and other success dimensions of an information system.

Finally, after determining the explanatory power and predictive relevance, overall model fit is evaluated using Goodness of Fit (GoF = (√(AVE × R²))) criterion, proposed by Tenenhaus et al. (2005). According to Wetzels et al. (2009) and Henseler et al. (2016), a high GoF value shows that, proposed theoretical model is parsimonious. Based on the guideline of Tenenhaus et al. (2005), Wetzels et al. (2009) and Henseler et al. (2016), GoF calculation for our proposed theoretical model is presented in the Table 6. The results have confirmed that, proposed theoretical model has obtained a substantial GoF_large value of 0.6699, which shows a good model fit.

Table 6 Goodness of Fit (GoF) Index

Discussion

As mentioned earlier, most of the hypotheses in the proposed theoretical model are significant, except for H1a, H2a, H2b and H3a (see Table 5). Based on the statistical results, it is found that, the system use of WBCLIS is explained by sociability quality and perceived user satisfaction. Moreover, it is found that, social functionalities, which are represented by sociability quality of a CSCL system, are also crucial success factor for an information system. By investigating the link between sociability aspect and system use, upshots of this study, have also verified the outcomes reported by Junglas et al. (2013) and Gao et al. (2010), who initially proposed the connection, between sociability and system use. Further, with respect to the system quality, H1a was not significant, which shows that, the use of WBCLIS is not largely influenced by system quality. However, contrary to this, Mohammadi (2015) has reported that, technical system quality has an indirect positive impact, on actual system use. Further, Mohammadi (2015) has also reported a direct and positive connection, between system quality and users’ intention to use an e-learning system. Further, H1b shows that, the system quality has a significant influence on user satisfaction.

Similar results regarding system quality and user satisfaction, have been reported by various preceding studies (e.g. Aparicio et al. 2017; Cidral et al. 2018; Costa et al. 2016; Hassanzadeh et al. 2012) on the success of e-learning information systems. Therefore, outcomes of this study put forward that, if a WBCLIS has a high system quality, it can enhance users’ overall satisfaction, derived from the system use. Another important success factor is information quality, which is found non-significant, in the case of WBCLIS; results indicate that, information quality (see H2a and H2b) has no major influence on system use and user satisfaction. These outcomes are consonant and inline with the findings of Balaban et al. (2013), as they also reported that, information quality has a non-significant impact on system use and user satisfaction. However, these findings of H2a and H2b are contrary to Cidral et al. (2018), as they reported a statistically significant impact of information quality, on system use and perceived user satisfaction, towards an e-learning information system. It is important to note that, Cidral et al. (2018) conducted their study in the Brazilian context only; therefore, this contradiction of results, can be because of difference in the cultural background and origin of respondents, making it an entirely different sample.

Moreover, it is vital to note that, in this study, participants were undergraduate students, enrolled in a project based service learning course, at a public sector university in Malaysia; therefore, all respondents of this study had a similar period of experience on WBCLIS. Whereas, in other studies (e.g., Cidral et al. 2018; Martins et al. 2018) respondents were recruited, from various different levels and programs, making their sample a blend of respondents, having a varying period of experience. In addition, this study argues that, maturity of information system, also effects system use and user satisfaction. Studies, which have reported contradictory findings (e.g., Cidral et al. 2018; Martins et al. 2018), were conducted on very established and mature CSCL systems. However, in our case, this study was conducted on the success factors, of a newly developed / pre-mature WBCLIS (i.e. the Co-Reflective). Further, findings regarding a non-significant relationship, among service quality and system use (see H3a), have validated similar results, reported by some other studies (e.g., Aparicio et al. 2017; Cidral et al. 2018; Hassanzadeh et al. 2012; Tam and Oliveira 2016); however, these findings are contrary to the results reported by Martins et al. (2018).

Reporting a statistically significant relationship among service quality and system use, Martins et al. (2018) have argued that, if students experience any problem, while using education management information systems, access to online support and good service quality can motivate them towards system use; however, poor service quality may adversely affect, students’ intention to use an education management information system. Furthermore, results of H3b has confirmed that, service quality has a direct and significant impact on user satisfaction. Thus, the notion that, service quality leads to a higher user satisfaction towards WBCLIS, is also strengthened (Aldholay et al. 2018; Alzahrani et al. 2017; Mukred and Yusof 2018). Further, it is found that, sociability quality positively influences system use (see H4a) and user satisfaction (see H4b). As a unique contribution of the findings of this study, sociability quality of WBCLIS, has emerged as an important success factor. Therefore, this study argues that, if a CSCL system provides effective and flexible social functionalities, then these functionalities, will not only enhance the quality of it’s collaborative learning, but they will also add up, to the overall increase in user satisfaction and system use. For that reason, outcomes of this study put forward that, sociability, as it is reported by prior studies (e.g., Gao et al. 2010; Junglas et al. 2013) is not enough; rather, high ‘sociability quality’ is what makes a WBCLIS more successful.

Moreover, results have also confirmed that, user satisfaction has a direct and statistically significant influence on system use (see H5). Similar results were reported by Martins et al. (2018), regarding the use of education management information systems. Further, these results have validated the findings of Aparicio et al. (2016) and Mohammadi (2015), concerning the use of e-learning systems. Furthermore, findings of this study have established that, ultimately success of an information system is determined, with the net benefits, derived by its’ users (DeLone and McLean 2003). Therefore, as a determinant of success of WBCLIS, it’s net benefits for all users, were deeply analyzed. Results confirmed that, user satisfaction and system use, both hold a significantly cogent sway on the net benefits (see H6 and H7) of WBCLIS. Therefore, it is suggested that, user satisfaction and system use, can play a crucial part, in defining the success of an academic / education related information system (Balaban et al. 2013; Martins et al. 2018). Specifically, in the context of WBCLIS, if students are satisfied with the features and functionalities, offered by that particular system, then their performance, in collaborative learning activities will be enhanced; thereby, maximizing their learning outcomes and adding to the net benefits of WBCLIS.

Theoretical contribution and practical implications

This study has proposed and validated a theoretical model, for assessing the sociability quality of WBCLIS. The concept of ‘Sociability quality’ is proposed as a new success dimension, in the DeLone and McLean’s (2003) updated IS success model. A significant theoretical contribution is imparted, by extending the DeLone and McLean’s (2003) updated IS success model. Moreover, aforementioned discussion on the sociability aspects, in the findings of this study, has also enhanced our understanding, regarding social interaction theory (Gunawardena 1995; Kreijns et al. 2007; Mills et al. 2010). Findings of this study have established that, sociability quality is an imperative factor, for predicting the success of a CSCL system, and it has a salutary and favourable impact on system use and user satisfaction. Furthermore, results also confirm that, user satisfaction and system use are interdependent, and both have a favourable, positive influence on the net benefits of a WBCLIS. Further, detailed analysis of the proposed hypotheses has confirmed that, system quality, service quality and sociability quality, significantly explain the user satisfaction towards WBCLIS. Moreover, it is found that, sociability quality and user satisfaction, are significant drivers of system use, in the context of WBCLIS.

While incorporating a relatively new concept of ‘sociability quality’, proposed theoretical model has demonstrated a notable explanatory power, for assessing the net benefits of WBCLIS. Moreover, outcomes of this study, offer a noteworthy contextual contribution, by validating the success of a WBCLIS, using proposed theoretical model of this study, in the context of service learning based collaborative project performance. In addition, findings of this study, offer a number of practical contributions for designers / developers of CSCL systems and higher education institutions. As described previously, this study was conducted on the implementation, of a newly developed WBCLIS (i.e. the Co-Reflective). This WBCLIS enabled teachers, to design collaborative learning activities, in which students could work, in collaboration with their fellow students. The successful implementation of WBCLIS, can facilitate students to act as a team, share information, practice collaborative critical reflection, which is not otherwise possible in the traditional class-room settings. Moreover, it is suggested that, the use of WBCLIS can encourage students, to share their service learning experiences, exchange ideas, share their personal perspectives, deliberate on others’ opinions and choose a suitable solution for their real life problems.

Further, use of WBCLIS is thoroughly described, by the assessment of proposed theoretical model, which has integrated “system quality, information quality, service quality” and sociability quality, in the relationship with system use and user satisfaction, for assessing the net benefits of a WBCLIS. The results indicate that, along with functional qualities of a CSCL system, developers should also focus on the social components of CSCL system, for enhancing user satisfaction, through sociability quality of their system. By focusing on social and technological features of a CSCL system, institutions can be successful, in developing a sound social environment, for collaborative learning and social interaction between teachers and students. Therefore, understanding of sociability quality, is a critical aspect of CSCL system. Further, as another practical contribution of this study, it is suggested that, a CSCL system should allow articulation of collaboration and communication, among teachers and students. Moreover, as mentioned in the results section, sociability quality has a significant salutary impact, on the use of WBCLIS and user satisfaction. Therefore, online collaborative learning environment should be interactive, organized and sociable; so that, students can collaborate with teachers and other fellow students, simultaneously, for on time completion of their collaborative learning tasks.

Limitations and future research directions

Outcomes of this study have validated the significance of a new IS success dimension (i.e. sociability quality) in the context of WBCLIS. Although, this study contributes well in the advancement of theory, and also opens up various future research directions; yet, some limitations are identified, to serve as future research directions. As stated earlier, this study was aimed to assess ‘sociability quality’ in the context of WBCLIS; for that reason, this study is not passable to entirely capture all the factors of IS success. Therefore, more studies on new success dimensions, are warranted in future also, to have an exhaustive comprehension, of all IS success dimensions. During this study, it was observed that, existing CSCL systems have downplayed, on the provision of sociability features; therefore, it is suggested that, more sociable WBCLIS should be designed, with advanced technological support, to provide a sociable collaborative learning environment. Further, proposed framework of this study, can be expanded, to validate more novel IS success dimensions, in other cultural and contextual settings. Taking new IS success dimensions into consideration, future studies can employ our proposed, sociability quality based IS success model, as their underpinning framework. Moreover, future research work can be performed, to evaluate the proposed theoretical model, with different CSCL systems and with a larger sample size.

Conclusion

As described formerly, this study is about the sociability quality of WBCLIS. The construct of sociability quality is introduced, as a new success dimension in the DeLone and McLean’s (2003) updated IS success model. Considering the experimental nature of this study, a WBCLIS (i.e. the Co-Reflective) was designed, to facilitate teachers and students, with a collaborative learning environment. This WBCLIS was developed in a way, in which teachers and students could interact, collaborate and actively participate, throughout their service learning project. Under study, WBCLIS was aimed to offer, several sociability features, that can facilitate users, throughout their collaborative learning process. For instance, every group of students had a self-regulated reflection environment, which allowed them to update and monitor their daily project progress, share information with their group members, collaborate and reflect on their fieldwork experience. Moreover, WBCLIS also enabled teachers, to monitor students’ progress, social behavior and collaboration approach, throughout their service learning project.

As mentioned in the methodology section, this study involved students, who were enrolled in the service learning course. For this purpose, students were divided in small groups, of 3 to 5 students working in each group. As part of their group project, assigned to them, during their service learning course, at a public sector university in Malaysia, students used this WBCLIS, throughout their whole semester, to actively participate, in various collaborative learning activities, such as sharing experiences, and discussing project related issues. At the end of the semester, all students were invited to voluntarily participate in an online self-administered survey based questionnaire, which was designed to assess the ‘sociability quality’ of WBCLIS, through a proposed theoretical model (see Fig. 1). It is pertinent to mention that, all participants had an equal period of access and uniform experience with WBCLIS, which also adds to the standing of the sample and data used for analyzing the proposed theoretical model of this study.

PLS-SEM based statistical approach was employed, for a thorough assessment of measurement models and structural model of this study. Findings of GoF index has shown that, proposed theoretical model has a good fit. Further, on the base of the findings of this study, it is righteous to suggest that, a WBCLIS should offer a variety of ways, for collaboration and social interaction between users. Easy and flexible features, for interaction between users, and sociability quality of a WBCLIS, contribute much, towards user satisfaction, which in turn leads to a significant growth in the learning outcomes, productivity and performance of all the stakeholders. Therefore, this study confirms that, with good sociability quality, WBCLIS can offer a flexible collaborative platform, to all users, for interacting and interrelating with each other, to design solutions for their real-world problems. Further, empirical validation of proposed theoretical model, in the service learning course, has also confirmed that, the sociability quality of WBCLIS, is a strong determinant and precursor, of the system use and user satisfaction, along with other IS success dimensions.

Appendix

Table 7 Measurement Scale Items