Computers & Education 109 (2017) 109e121

Contents lists available at ScienceDirect

Computers & Education

journal homepage: www.elsevier .com/locate/compedu

How are e-leadership practices in implementing a schoolvirtual learning environment enhanced? A grounded modelstudy

Yan Piaw Chua a, *, Yee Pei Chua b

a University of Malaya, Institute of Educational Leadership, Level 11, Wisma R & D, UM, Jalan Pantai Baru, 59990 Kuala Lumpur, Malaysiab Universiti Putra Malaysia, Faculty of Computer Science and Information Technology, 43400 Serdang, Selangor, Malaysia

a r t i c l e i n f o

Article history:Received 15 June 2016Received in revised form 21 February 2017Accepted 22 February 2017Available online 24 February 2017

Keywords:Computer-mediated communicationInteractive learning environmentsLearning communitiesSecondary education

* Corresponding author.E-mail addresses: chuayp@um.edu.my (Y.P. Chua

http://dx.doi.org/10.1016/j.compedu.2017.02.0120360-1315/© 2017 Elsevier Ltd. All rights reserved.

a b s t r a c t

E-leadership is defined as a social influence process mediated by information andcommunication technology to produce change in behavior and performance with in-dividuals and groups in an organization. This study investigates e-leadership practicesamong users of a school virtual learning environment. It was performed in two stages.First, semi-structured interviews with school administrators, teachers, students, parentsand school software experts were conducted. The qualitative data collected from the in-terviews were coded and analyzed using open and axial coding procedures. As a result, ane-leadership model emerged from the data that consisted of eight themes: e-leadershipquality with seven core factors, namely, readiness, practices, strategies, support, culture,needs and obstacles. Second, the validity and reliability of the model were further ascer-tained with a quantitative survey study involving 320 school administrators. The findingsof this study established a grounded model for e-leadership practices in schools.

© 2017 Elsevier Ltd. All rights reserved.

1. Introduction

Leaders of educational institution in the information technology era face more challenges due to the dynamics of theworkplace and institutional culture. One of the new challenges is that the staff of the institution is physically dispersed awayfrom the leaders (Schultz, 2010) because school leaders are busy with meetings and commitments outside the institutions.School leaders cannot fully focus on school leadership and management (Albidewi, 2014; Mohd Yusri, 2014).

To overcome this problem, the use of information technologies such as the internet with personal computers, smartphones and mobile phone applications in monitoring school activities and networking among school communities hasincreased. Schools are also equipped with a virtual learning environment or e-learning platform to comprehend face-to-faceteaching and learning processes (Alvarez, Martín, Fernandez-Castro, & Urretavizcaya, 2013). To further enhance competi-tiveness and effectiveness, educational institutions are increasingly relying on ICT to optimize their e-teaching and learningprocesses and digitize their operations, innovations and services (Albidewi, 2014; Aral, Brynjolfsson, & Wu, 2012). Aseducational institutions rely more on technology, they are demanding new types of leaders: leaders who are e-skilled andable to integrate technology in leadership andwho can lead andmanage staff through technology and an e-learning platform.

), yeepei94@hotmail.com (Y.P. Chua).

Y.P. Chua, Y.P. Chua / Computers & Education 109 (2017) 109e121110

In this situation, school leaders need to change their role as e-leaders. E-leadership is a new leadership paradigm that requiresthe leader to achieve leadership objectives in a computer-mediated manner with virtual teams that are dispersed over spaceand time (Fonstad, 2013). However, the integration of technology into leadership practices requires the leader to transformhis role, vision and practices as face-to-face dependency on the leader decreases. It also causes changes in the strategies usedto lead the institution, especially in the fields of planning, monitoring and follow-up (Albidewi, 2014).

The problem that faces the application of e-leadership is sometimes not the shortage of facilities or the failure of the e-learning platform but the behavior of leaders, the organizational culture, the readiness of leaders and staff, and their un-willingness to adapt and change (Hung, 2016; Jameson, 2013). Some school leaders who have easy access to technology do notuse it because they lack confidence and e-skills and feel that the use of technology in leadership entails problems andcomplications (Albidewi, 2014). A careful review of the literature shows that there is no appropriate research-based model orframework of e-leadership that can be used as a reference in implementing an e-learning platform in schools.

To address these shortcomings, this study is conducted to understand e-leadership practices in a school virtual learningenvironment. In particular, it proposes a grounded model of e-leadership to understand the underlying mechanisms thataccount for e-leadership quality. In doing so, it provides a new framework for e-leadership practices in schools.

2. Literature review

E-leadership is defined as a social influence process mediated by ICT to produce a change in attitudes, feelings, thinking,behavior, and performance with individuals, groups and organizations (Avolio, Kahai & Dodge, 2001; Blasco-Arcas, Buil,Hernandez-Ortega& Sese, 2013; Dundar& Akcayir, 2014; Fonstad, 2013; Orus et al., 2016). Conducting an extensive review ofthe e-leadership literature, Jameson (2013) defines e-leadership as “virtual relationships of influence”. According to thisscholar, this new highly adaptive field of knowledge affects multiple daily interactions across professional education andtraining, spontaneously involving people who use social networking facilities regularly both at home and at work. E-lead-ership skill is also defined as “technology-leadership integration” of influence that consists of a T-shaped portfolio of skills,representing expertise in both “vertical set of skills”e the expertise in using ICTeand “horizontal set of skills”e the expertisein leading andmanaging the organization (Fonstad, 2013). The vertical skills include “deep knowledge” of the application andmaintenance of ICT (e.g., functional; technical; product-related; and customer experience-related); the horizontal skillsrepresent expertise in developing the organization, i.e., competence in leadership and management (e.g., envisioning;building and aligning relationships across boundaries; sense making; inventing). Adaptive Structuration Theory (Avolio et al.,2001) explains how technology and leadership impact each other in e-organizations and that the technology-leadershipintegration is heavily dependent on e-knowledge, e-skills and e-leadership quality (Avolio & Kahai, 2003; Purvanova &Bono, 2009).

However, the technology-leadership integration can be difficult, and the addition of a complex organization can increasethis difficulty due to multiple leadership structures and cultures (Evans, Ashbury, Hogue, Smith, & Pun, 2014). Although e-leadership is needed at all levels of e-learning implementation, the uptake of technology in education, on the whole, has notbeen accompanied by specific e-leadership training and professional development (Jameson, 2013; Kearsley & Lynch, 1994),and the common failure of the implementation is typically due to scarce e-leadership (Hanna, 2009). Hanna (2009) notes thatthe gap between the demand for e-leaders and the supply is large, and it not only is quantitative but also involves qualitativeimprovements in the capacities to think beyond the technology to include leadership and managerial skills.

A careful search for research related to e-leadership over the past ten years (2006e2015) finds only a small number ofdocuments (see Table 1). Most are at the early stage of conceptualization. The variables involved in these studies focus on thee-skills of the school virtual leader (Garcia, 2015), e-leadership, globalization and innovation in educational technology(Bowen et al., 2013), e-strategic planning methods and e-learning outcomes (Gomes, 2011), e-leadership styles (Hadjithoma-Garstka, 2011), the e-leaders’ roles and competencies (Tan, 2010), the impacts of schools' cultures and structures on e-learning (Tondeur, Devos, Van Houtte, van Braak, & Valcke, 2009), school principals' e-competence (Afshari, Bakar, Wong,Samah, & Fooi, 2009), the support of leaders and training for staff for the effective delivery of e-learning (McPherson &Nunes, 2008), technology leadership and interpersonal and communication skills (Chang, Chin, & Hsu, 2008), e-leadershipchallenges and opportunities (Barwick & Back, 2007), transformational and transactional leadership styles in face-to-facelearning and e-learning (Humbly, O'Neill, & Kline, 2007) and the collaborative leadership of e-learning and the role of e-leadership (Jameson, Ferrell, Kelly, Walker, & Ryan, 2006).

3. What is missing in e-leadership studies in education?

Scholars have the same opinion that e-leadership research in education is scarce. For example, reporting an extensivereview on educational leadership studies between 2000 and 2013, Jameson (2013) summarizes that, “although leadership as ageneral area of research study remains highly significant and is growing in quantity and impact, e-leadership research in education,by contrast, has barely emerged into public recognition as a research concept within the recognisable surface of scholarlyendeavour, judging by its still thin citation counts, on the whole research studies in e-leadership appear to be surprisingly limitedwithin the databases available” (pp. 901). Similarly, Mcleod and Richardson (2011) extensively review school e-leadershipstudies over the 1997e2009 period and summarize that “scholars, researchers, and practitioners in the field of educationalleadership are rarely exposed to issues of e-leadership. We simply do not have enough high-quality research to inform best practice.

Table 1Summary of selected e-leadership research and reviews (2006e2015).

Authors Methods and findings

Garcia (2015) A review of literature on e-leadership, the strengths and skills of school virtual leaders and their importance in themanagement of e-learning.

Bowen et al. (2013) A survey on academic institutions and industries in Peru, India, Qatar, Germany, the United Arab Emirates, Saudi Arabia, SriLanka, Ireland, Kenya, Turkey, China, Spain and the USA. A qualitative ethnographic study regarding e-leadership, globalizationand innovation in educational technology.

Gomes (2011) A corporate reflective report on strategic planning methods to use technology more effectively for improving e-learningoutcomes.

Hadjithoma-Garstka(2011)

A large-scale survey case study in schools in Cyprus on the role of school principals' e-leadership styles and the factors thatinfluence the implementation of ICT in schools.

Tan (2010) A review of empirical reports of secondary schools in the USA, Canada, Singapore, Belgium and Tehran regarding thecompetencies of technology leaders and their roles.

Tondeur et al. (2009) A survey on a group of Belgian teachers in primary schools regarding the effects of schools' cultures and structures, i.e.,leadership support, ICT planning and the underlying structure in e-learning.

Afshari et al. (2009). A survey on secondary school principals in Tehran regarding the relationship between principals' e-competence andtransformational leadership.

McPherson and Nunes(2008)

A survey on higher education in the UK regarding the relationship between the support of leaders and training for staff witheffective e-learning.

Chang et al. (2008) A study on elementary schools in seven cities in Taiwan regarding the relationship between technology leadership,interpersonal skill and communication skill.

Barwick and Back(2007)

A case study of a blog involving academic staff and senior teachers regarding their e-leadership challenges and opportunities inhigher education.

Humbly et al. (2007) A study on Canadian undergraduate students regarding transformational and transactional leadership styles in differentcommunication media, i.e., face-to-face, videoconferencing and text-based chatting.

Jameson et al. (2006) A study on higher education, schools and colleges in the UK regarding the collaborative leadership of e-learning and the role ofe-leaders within lifelong learning contexts.

Y.P. Chua, Y.P. Chua / Computers & Education 109 (2017) 109e121 111

We needmore researchers andmore research. Much of the work will be foundational” (pp. 236). Reviewing seventy-seven journalarticles regarding e-leadership, DasGupta (2011) summarizes that “there does not appear to be any serious disagreementamongst scholars on e-leadership; However, there is agreement that this is a new field and that more research needs to be con-ducted.” (pp. 30).

Because e-leadership research in education is scarce, several scholars have expressed their concern over the “what” andthe “how” on the application of e-leadership in schools. For example, van Welsum and Lanvin (2012) are concerned over themissing link in the approach to e-leadership, which lacks an appropriate model and framework for e-leadership practices andan action plan necessary for fostering effectiveness. Much research must be conducted to answer the questions of “what arethe factors that determine the quality of e-leadership practices in schools?” and “how is the quality of e-leadership practicesin schools enhanced?”

To address themissing link and gap, this study investigates school e-leadership practices in a virtual learning environment.Specifically, the objectives of this study are to develop a grounded model of e-leadership practices in schools and to furthervalidate the model.

This study is significant in two ways. First, it helps educators better understand the concept of effective e-leadershippractices and the constructive and destructive factors that are related to it. Second, it proposes a grounded model forimplementing e-leadership in education institutions. Educators and researchers can use this information to identify unan-swered issues or questions in the literature and define future research directions concerning e-leadership in education.

4. The study

The study was conducted in two stages. First, semi-structured interviews with 22 school administrators, teachers, stu-dents, parents and software experts were conducted. The data from the interviews were coded and analyzed using qualitativedata analysis software, Atlas.Ti version 7 (Ringmayr, 2012), to gather the themes of e-leadership practices in schools toestablish a grounded model for e-leadership. Second, the validity and reliability of the model were further ascertained with aquantitative survey study involving 320 school administrators. The quantitative data were analyzed using the AMOS version22 model testing software.

4.1. Stage 1: establishing a grounded model of E-Leadership practices in schools

4.1.1. MethodsThe interpretative research method was adopted to achieve the first objective, that is, the collection of a rich set of data

from a variety of sources. The interpretative method explains the characteristics of a small number of participants in detailand in depth (Chua, 2016). The results involve the creation of a model that fully and directly emerges from the data, not fromthe researcher (Conrad, 1995).

Y.P. Chua, Y.P. Chua / Computers & Education 109 (2017) 109e121112

The theoretical framework of the grounded theory research methodology is appropriate for this research because it givesdue consideration to the theoretical requirements and the interpretative method. The grounded theory methodologydeveloped by Glaser and Strauss (1967) is defined as a set of specified operational procedures and systematic data analysesinductively developed from a set of data to enhance a theory regarding a particular social phenomenon. In accordance withgrounded theory, the data for this study were collected through interviews, were then systematically analyzed and inter-preted, and were finally used to generate a theory regarding a phenomenon that is “grounded” in its systematic analysis. Inthis manner, grounded theory is particularly appropriate because the aim is to explain a process rather than to verify anexisting theory. According to Strauss and Corbin (1990), the research procedures associated with grounded theory can beadapted, implemented again, and reassessed according to the researcher's creativity. To achieve the objective, it is necessaryto check and recheck the data, to question them critically, to obtain new data that are more appropriate, and to maintain askeptical attitude, that is, to question every piece of original data given by the respondents. This strategy enables us to raisequestions that are both precise and appropriate (Strauss & Corbin, 1990).

4.1.2. Validity and reliabilityIn grounded theory research, four criteria determine the validity of the data: (a) a constant comparison of the research

findings with everyday experience, (b) the application of the research findings in the field, (c) the abstract nature of thetheory, and (d) the ability of the theory generated to be adapted for application in various contexts (Strauss& Corbin,1990). Toenhance the validity of this study, in the process of discovering the model implicit in the data, the four criteria should be met.In addition, we use the emerging theoretical framework in which questions relating to categories of open codification areproposed and the original data are re-examined for further evidence. Axial coding, which is the process of developing mainsubthemes and themes, is subsequently performed to trace the relationships between themes and to enhance the validity andreliability of the study (Creswell, 2005).

4.1.3. Participants of the studySampling in grounded theory research is generally used to select respondents with the potential to provide the researcher

with the information required to generate a theory/model implicit in the corpus data. In stage 1 of this study, we usedtheoretical sampling to select suitable respondents, whichmeans that the sampling was performedwith a specific purpose inmind and focused on the generation of a model (Creswell, 2005). School administrators (n ¼ 5), teachers (n ¼ 5), students(n ¼ 5), parents (n ¼ 5) and software experts (n ¼ 2) were selected as the main respondents because they were directlyinvolved in the implementation of the virtual learning environment (VLE) in schools.

The VLE, also known as an e-learning platform or learningmanagement system, is the most popular e-learning platform ineducation and has been adopted by almost all higher education institutions in the world (Kroner, 2014). The VLE allows usersto be organized into groups and roles; to present resources, activities and interactions within a course structure; to providefor the different stages of assessment; to report on the participation rate; and to have some level of integration with otherinstitutional systems (Martin, 2007, pp. 4e5). In this study, the interview informants were users of the school virtual learningenvironment, the VLE FROG, a nationwide e-teaching and learning program performed by the Malaysian Ministry of Edu-cation. Every parent, student, teacher and school administrator has been provided with an ID that enables access to a varietyof tools that create a more exciting and interesting learning experience. With high-speed 4G mobile internet, this e-learningplatform has 10 million users in all schools nationwide in Malaysia, including 5 million students, 4.5 million parents and500,000 teachers (1BestariNet, 2012).

4.1.4. Research instrumentThree interview inventories were constructed for the collection of relevant data from the five groups of respondents. The

School Administrators Inventory contains four items concerning their e-leadership roles and involvement in implementingthe VLE in schools. The Teacher, Student and Parent Inventory contains four items concerned with their commitment to andsatisfaction with the effective implementation of the VLE in schools. The Software Expert Inventory contains five itemsconcerned with the functions and requirements of the skills, knowledge and issues of using the e-learning platform inschools. All of the inventories also include items on problems with planning and implementation, the support given for theimplementation of the e-learning platform in schools and recommendations to enhance the effectiveness of e-leadershippractices in leading and monitoring the VLE in schools.

4.1.5. Data collection and analysisData collection and analysis were performed simultaneously (Glaser, 1978; Strauss, 1987). In view of the time-consuming

nature of interview data analysis, the transcribed data were analyzed using the Atlas.ti software, which is used to record andcodify qualitative data and facilitates the task of coding the data. According to Chua (2016) and Strauss and Corbin (1998),Atlas.ti is an ideal tool for analyzing data in connection with grounded theory research. Grounded theory research uses asystematic method of data collection and analysis with the aim of constructing an in-depth understanding of social andpsychological phenomena (Chenitz & Swanson, 1986). To achieve this aim, data analysis involves two types of coding: opencoding and axial coding.

Open coding. Open coding involves checking and rechecking data that have been collected and transcribed and in whichrelevant codes are given to statements that are both meaningful and important.

Y.P. Chua, Y.P. Chua / Computers & Education 109 (2017) 109e121 113

Axial coding. Axial coding is a synthesis of open coding as the codes for categories are connected to each other. To make thelink between codes, Spradley's (1980) Semantic Relationship Questions Technique, a domain analysis technique, wasemployed. The types of question asked include “How is this assertion linked to other assertions?”, “Are there similarities toother assertions?”, “Is it one of the steps in a process?”, “Is it caused by an action or strategy?”, “Is it themain cause?” and “Is itthe outcome of a process?” The use of this technique enables the researcher to arrange and place an idea with data with thesame theme in the same category, known as a domain. By studying each paragraph in the data, sentences containing the sameideawere coded, and finally, when all of the data had been examined, units containing the codes were identified and placed inthe same domain or category.

4.1.6. ResultsBased on the axial coding and analysis, eight core themes emerged from the data, namely, readiness, practices, strategies,

support, culture, needs, obstacles and e-leadership quality (Table 2). The story line and analytic story or explanation for thegrounded model were formed by taking e-leadership quality as the main category and tracing its relationship to the otherseven themes.

4.1.6.1. Story line. The research data show that e-leadership is needed in schools because schools are equipped with a VLE fore-teaching and learning and networking among staff and stakeholders, and because it plays an important role in school as itcan raise the quality of teaching and learning and help teachers overcome the deficiency of face-to-face practices. A positiveschool culture and strong support for the e-leaders and the readiness of the leaders in terms of e-knowledge, e-skills, CMCcompetences and their attitude towards the school VLE application affect e-leadership practices, strategies and effectivenessin the schools. A negative attitude and low commitment among users, the failure of the VLE and insufficient training areobstacles that reduce rate of usage of the VLE. In the effort to introduce a positive e-teaching and learning culture, it isimportant to create an appetite among users (teachers, students and parents), design user-friendly curricula, foster multi-disciplinary approaches to networking and promote a better and greater use of the e-learning platform. Nevertheless, theresearch also shows that what determines success in e-leadership practices is the attitude, commitment and satisfaction ofusers with regard to the VLE, knowledge of computer-mediated communication, competence networking, user-friendlycurricula, a good reward system, the infra-structure of the VLE, and the change management ability of the e-leaders.

In addition, to achieve the objective of implementing high quality e-leadership in schools, effective strategies and practicesare essential. These includes e-competence training, effective curricula, enhancing lifelong learning, building long-term re-lationships among users, developing positive mindsets, developing a compelling mission and vision for networking andfostering the maximum use of the VLE. Furthermore, school leaders need to adopt appropriate e-teaching and learningmodels to increase effectiveness, maximize value from the VLE, encourage an effectiveway of using ICT and act as role modelsfor e-teaching and learning and for active participation.

To improve the quality of e-leadership practices, schools must take action to embrace and reward self-e-learning initia-tives, stimulate e-cooperation between staff and clients (education department, parents and students), create a better e-teaching and learning environment for a better workplace, create pathways for networking among users, make networkingan incentive and a basic for the performance rating of staff and create a conducive infra-structure for e-teaching and learning.

4.1.6.2. Analytic story. The qualitative data shows that e-leadership quality is important for the implementation of the VLE inschool. Optimum practices and strategies ensure the quality of e-leadership in schools. However, the main challenges of e-leadership are the context of readiness (readiness), the personal context (obstacle) and the organizational contexts (e-learning culture and support), as shown in the grounded model of e-leadership in Fig. 1.

4.2. Stage 2: validity and reliability of the model

4.2.1. ParticipantsQuantitative survey data were collected from 320 school administrators at secondary schools. The participants were

school principals and senior assistants. They played the role of leaders in leading andmonitoring the use of the VLE in schools.The participants comprised 143 males and 177 females with an average age of 47.1.

4.2.2. InstrumentThe survey questionnaire used in this study consisted of two sections that correspond to the demographic variables and

eight main variables in the e-leadership model generated from stage 1 - the emerging data from the interview. There were atotal of 40 items. The items were created based on the data of each theme generated from the semi-structured interviewsindicated in Table 1. For example, the first item of needs (Ne1) is “The e-leader is needed because of more investment in ICT inschool”. The items of this questionnaire were validated by a panel of three educational expects. The items used a continuousscale of measurement ranging from 0 to 10, with “0” indicating “completely disagree” and 10 indicating “completely agree”with the item statement. The reason for using this numerical rating scale is it provides a high level of measurement precisionand permitsmathematical operations to be performed, which is thus relevant for structural equationmodeling (SEM) analysis(Hair, Black, Babin, & Anderson, 2010). According to Cohen, Manion, and Morrison (2011), the ideal measurement scale for

Table 2Domain analysis output e the eight themes that emerged from the interview transcripts.

Data (indicator) Label ofIndicators

Theme (number ofindicators)

1. Needing e-leaders because of more investment in ICT in schools. Ne1 Needs (7)2. Demand of the digital era. Ne23. Leading e-learning and teaching process/online instructional guides and coaching. Ne34. Needed for e-leaders for innovatively developing more uses of ICT. Ne45. Needed for global cooperation. Ne56. Leading for changes. Ne67. Enhancing cooperation of the school with external parties (education office, parents, staff and students). Ne7

1. Adapting new models to increase effectiveness. Pr1 Practices (6)2. Maximizing value from ICT spending by the institution. Pr23. Enhancing the use of ICT in teaching and learning. Pr34. Leading the proper way of using ICT in schools. Pr45. As role models for e-communication/encouraging active participation. Pr56. Determining proper institutional goals for implementing e-teaching and learning practices in school. Pr6

1. Reshaping objectives and curricula/co-curricula in line with e-teaching and learning. St1 Strategies (7)2. Organizing e-competence training among teachers, students and parents. St23. Promoting lifelong self-e-learning. St34. Building long-term relationships among all users across boundaries and monitoring progress. St45. Leading change for positive mindsets towards e-teaching and learning and network thinking. St56. Developing a compelling mission and vision for networking. St67. Managing the maximum use of ICT services (process, design, networking among users). St7

1. Failure of technology. Ob1 Obstacles (4)2. Less ICT knowledge and skills. Ob23. Insufficient ICT training. Ob34. Negative attitude/low commitment among users. Ob4

1. Creating an appetite among users (teachers, students and parents). Cu1 Culture (4)2. Designing user-friendly curricula to enhance the rate of usage. Cu23. Fostering multi-disciplinary approaches to networking among users. Cu34. Promoting better and greater use of e-teaching and learning. Cu4

1. Embracing and rewarding lifelong and self-e-learning. Su1 Support (6)2. Psychological support for e-cooperation between staff and clients (education department, parents and

students).Su2

3. Creating an e-teaching/learning environment/workplace. Su34. Creating pathways for networking among users. Su45. Making networking an incentive and a basic for the performance rating of staff. Su56. Providing infra-structure between e-teaching and learning. Su6

1. Attitude towards e-leadership. Re1 Readiness (4)2. E-teaching/learning knowledge and skills. Re23. CMC networking knowledge and skills. Re34. CMC competence competencies. Re4

1. Maximizing the rate of usage among users (teachers, students and parents). Ef1 E-leadership quality (4)2. High level of satisfaction/commitment among users through excellent networking among users. Ef23. Positive attitude towards the use of ICT in communication/teaching & learning, and networking among

users.Ef3

4. An excellent e-teaching and learning process in line with high academic and non-academic studentperformance.

Ef4

Y.P. Chua, Y.P. Chua / Computers & Education 109 (2017) 109e121114

performing a SEM analysis with AMOS should be in the continuous interval scale from one to ten so that the data are moreindependent andmeet the requirements of SEM for parametric analysis. This statement is further supported by Awang (2014),who asserts that, if a researcher wishes to calculate a more sophisticated level of statistics such as regressions, factor analysisand SEM, then a ratio scale must have a true zero (‘0’) and equal intervals. Thus, many rating scales use an 11-point scale thatranges from 0 to 10, with 0 being the lowest score (or something equivalent to this, depending on the item statement) and 10being the highest score, to meet the minimum requirements for the ratio measures.

4.2.3. Data analysisFor testing the validity and reliability of the model, SEM analysis was performed in three stages. First, a preliminary

analysis for the normality of data of the items and the variables in the model was performed because normality is the basicrequirement of SEM analysis. Secondly, the validity (construct validity and discriminant validity) and reliability (internalconsistency reliability) of the variables were examined to ensure that the items validly and reliably represented the concept ofthe eight variables in the model. Finally, model fit testing was conducted to examine whether the entire model is valid forgeneralization to the population of the study. The output of the SEMwould suggest model modification if the model does not

Fig. 1. Paradigm diagram of the grounded model of e-leadership.

Y.P. Chua, Y.P. Chua / Computers & Education 109 (2017) 109e121 115

fit the data, that is, modifying (adding or removing) relationships between some of the variables in the model to achieve abetter fit of the model with the data. Hence, the final model is grounded in the data and directly emerges from the data.

4.2.4. Results

4.2.4.1. Preliminary analysis of data. SEM is a parametric model testing analysis that requires that the data of the items involvedin the analysis are normally distributed. The benchmark of the univariate normality of each item in a measurement model fora latent variable is that the skewness and kurtosis values for each item are in the range of �1.96 to 1.96; and the benchmarkfor multivariate normality for all of the items as awhole in a proposed model is that the value of the multivariate critical ratiois less than 8.00 (Byrne, 2010). The preliminary analysis of the data indicated that the univariate normality (skewness andkurtosis) of the 40 items ranged from�1.04 to 1.13 and that the value of the critical ratio for multivariate normality was 3.71.Thus, the data of the 40 items met the basic normality requirements for SEM analysis.

4.2.4.2. Validity and reliability of the variables. Further analyses of the validity and reliability of the eight variables are presentedin Table 3. In SEM analysis, the eight variables are latent variables that are reflectively represented by their indicators (items).The construct validity of a variable is achieved when the items of each variable are (i) significant, with (ii) the factor loadingsof each individual item being greater than 0.50 and (iii) the average variance extracted (AVE) of the variable being greater than0.50 (Byrne, 2010; Kline, 2016). The results in Table 3 show that the eight variables achieved their construct validity. Inaddition, the variables were reliable in terms of Cronbach's alpha internal consistency reliability. The internal consistency of avariable is achieved when its Cronbach's alpha is �0.70 (Byrne, 2010).

The discriminant validity of a construct is achieved when the inter-correlations among the items in the model are < 0.90.The implication is that the variables are independent of one another and no overlapping of the items of the variables is found.Multi-collinearity causes overlapping of concepts among the items of the variables in amodel. This problem exists inmodel fittesting due to extremely strong correlations (r � 0.90) between items in the model (Byrne, 2010). The results in Table 4 showthat the inter-correlation coefficients among all of the variables were less than 0.90. The variables are free of multi-collinearity problems, and therefore, the discriminant validity of the items for the model is achieved.

4.2.4.3. Model fit testing. The SEM analysis using AMOSwas performed for model fit testing. AMOS is one of the latest softwareapplications that enables researchers to accurately and effectively model and analyze the inter-relationships among con-structs that have multiple indicators (Byrne, 2010). More specifically, the multiple equations of the correlational and causalrelationships in a model are computed simultaneously. It is powerful software that enables researchers to support theirtheories by extending the standard multivariate analytical methodology, which includes regression, factor analysis, corre-lation and analysis of covariance (Kline, 2016).

Model fit testing is conducted on the original e-leadership model generated from stage 1. For model fit testing, scholarssuggest referring to three reliable indexes, which are the RMSEA or the Root Mean Square of Error Approximation (a

Table 3Validity and reliability and of variables in the model.

Variables in the model Number of items Construct validity Cronbach's alpha internal consistency reliability

Factor loadings of items AVE

1. E-leadership quality 4 0.77 to 0.90*** 0.914 0.7982. Readiness 4 0.81 to 0.91*** 0.926 0.8183. Strategies 7 0.66 to 0.85*** 0.904 0.7694. Practices 6 0.78 to 0.89*** 0.938 0.7655. Support 6 0.80 to 0.92*** 0.947 0.7966. Culture 4 0.84 to 0.92*** 0.939 0.8477. Obstacles 4 0.84 to 0.96*** 0.943 0.8568. Needs 7 0.71 to 0.88*** 0.938 0.736

Note: *** significant at p < 0.001.

Table 4Inter-correlations among the variables in the model.

Pearson correlation 1 2 3 4 5 6 7 8

1. Culture 1.0002. E-leadership quality 0.768 1.0003. Needs 0.806 0.710 1.0004. Obstacles �0.813 �0.726 �0.804 1.0005. Practices 0.765 0.714 0.815 �0.736 1.0006. Readiness 0.829 0.775 0.768 �0.835 0.735 1.0007. Strategies 0.658 0.664 0.696 �0.641 0.712 0.642 1.0008. Support 0.792 0.862 0.746 �0.736 0.726 0.769 0.670 1.000

Table 5Model fit indexes and the results of model fit testing.

Model fit Index Benchmark for model fit Model testing result Conclusion

1. RMSEA <0.08 RMSEA ¼ 0.076 Model fit confirmed2. CFI CFI > 0.90 CFI ¼ 0.907 Model fit confirmed3. RATIO RATIO < 5.0 RATIO ¼ 3.356 Model fit confirmed

Note: Chi-square is sensitive to a sample size >200. When the sample size greater than 200, ignore Chi-square results for model fit testing (source: Awang,2014, p. 64).

Y.P. Chua, Y.P. Chua / Computers & Education 109 (2017) 109e121116

representative of Absolute Fit Indexes), CFI or the Comparative Fit Index (a representative of Incremental Fit Indexes) and theRATIO or Relative Chi-Square (a representative of Parsimonious Fit Indexes). The threemeasures determine howwell an a priorior proposed model fits the sample data (Byrne, 2010; Hu & Bentler, 1999; Kline, 2016).

The results indicate that all of the relationships between variables proposed in the model were significant. However, themodel did not fit the data, with the RMSEA and CFI values being slightly beyond the acceptable model fit benchmarks(RMSEA ¼ 0.086 and CFI¼ 0.847). The outputs of the modification indices of AMOS suggest that model modification needs tobe performed because, in the data, there were significant correlations between readiness ↔ needs (modificationindices ¼ 188.876, parameter change ¼ 2.518), readiness ↔ culture (modification indices ¼ 227.298, parameterchange ¼ 3.010) and culture ↔ needs (modification indices ¼ 209.998, parameter change ¼ 2.714). Therefore, by connectingthe three correlations in the model, the model fit is confirmed, with RMSEA¼ 0.076, CFI¼ 0.907 and RATIO¼ 3.356 (Table 5).

4.2.4.4. The final model. The final model depicted in Fig. 2 consists of the e-leadership quality outcome variable with its sevencore factors. E-Leadership quality is directly influenced by three core factors, namely, readiness, strategies and support; and itis indirectly influenced by culture, practices, needs and obstacles. The data in Table 6 indicate that the factors contribute 87.3%of e-leadership quality (R2 ¼ 0.873).

Among the three direct factors of e-leadership quality, support is the main factor (b¼ 68, p < 0.001), followed by readiness(b ¼ 23, p < 0.001) and strategies (b ¼ 0.09, p < 0.05). A one-unit input of readiness would cause a 0.68-unit increase in e-leadership quality. The implication is that, with full support, readiness and the right strategies, e-leadership quality would bemaximized to nearly 90% (R2 ¼ 0.87).

In addition, a total of 58.4% of the strategies used by the e-leaders are significantly influenced by practices (b ¼ 0.51,p < 0.001), culture (b ¼ 0.28, p < 0.001) and obstacles (b ¼ �0.10, p < 0.05). Thus, to improve strategies, e-leaders need tomaximize practices, stimulate a conducive culture and overcome the negative effect of obstacles.

Fig. 2. Output of model fit testing. Note: *significant at p < 0.05; ***significant at p < 0.001; the factors loadings of the indicators for all eight latent variables are significant.

Y.P.Chua,Y.P.Chua/Com

puters&

Education109

(2017)109

e121

117

Table 6Standardized regression weights (b), R2 and inter-correlations of the relationships among the variables in the model.

Regression Unstandardized regression weight (b) Std. error Critical ratio p Standardized regression weights (b) R2

Strategies / e-Leadership 0.118 0.054 2.183 0.029 0.088Support / e-Leadership 0.647 0.055 11.744 0.000 0.683Practices / Strategies 0.366 0.066 5.566 0.000 0.509 0.584Obstacles / Strategies �0.068 0.031 �2.207 0.027 �0.096Culture / Strategies 0.191 0.058 3.309 0.000 0.279Needs / Practices 0.650 0.087 7.498 0.000 0.610 0.763Culture / Practices 0.275 0.073 3.774 0.000 0.288Obstacles / Practices �0.071 0.033 �2.185 0.029 �0.072Culture / Support 0.830 0.054 15.337 0.000 0.855 0.731

Inter-correlation Covariance Std. error Critical ratio p Correlation (r) e

Readiness ↔ Needs 3.107 0.297 10.456 0.000 0.82 e

Readiness ↔ Culture 2.763 0.284 9.737 0.000 0.90 e

Culture ↔ Needs 2.542 0.259 9.805 0.000 0.86 e

Fig. 3. Sub-models of the e-leadership grounded model. Note: The effect size interpretation for social science data e Effect size for R2: small ¼ 0.04;moderate ¼ 0.25; strong ¼ 0.64 (source: Ferguson, 2009).

Y.P. Chua, Y.P. Chua / Computers & Education 109 (2017) 109e121118

A total of 73.1% of the support provided by e-leaders is due to the culture, whereas the three factors of e-leadershippractices are needs, culture and obstacles. Needs have a large and positive effect (b ¼ 0.61, p < 0.001) on practices, whereasobstacles have a negative effect (b ¼ �0.07, p < 0.05).

Four sub-models derived from the results are depicted in Fig. 3. These sub-models are the basics of the e-leadershipmodel.In addition to the main findings, an unpredicted finding that emerged from the data is that there were significant, positive

and strong inter-correlations among needs, culture and readiness and the three correlations are strong (readiness ↔ needs:r ¼ 0.82, p < 0.001; readiness ↔ culture: r ¼ 0.90, p < 0.001; culture ↔ needs: r ¼ 0.86).

4.2.4.5. Discussion. The analytic narrative presented in the qualitative study leads to a model of e-leadership quality, which isused as a reference for e-leadership practices in schools. According to Creswell (2005), a theory or model in grounded theoryresearch is an abstract explanation or understanding of a process concerning variables in some topic hidden in the researchdata. The implication is that the model is generated from the data themselves, which are directly collected from differentsources and individuals.

Because the model that is generated derives from the data, it does not have a wide scope or applications (Chua, Tie, &Zuraidah, 2013; Glaser & Strauss, 1967). In light of this statement, it should be borne in mind that the model that is pro-duced from this study does not aim to produce standards to be used in all schools or to comparewith existing standards in theimplementation of education research in schools by relevant parties. However, it can be used as a reference by educationinstitutions involved in the implementation of the e-learning platform in enhancing e-leadership practices. There are fourcriteria that emerged from the data and that define the model of e-leadership practices in schools:

Y.P. Chua, Y.P. Chua / Computers & Education 109 (2017) 109e121 119

1. School leaders and followers should be given full support, and theymust be ready for implementation beforewell-plannedstrategies can be maximized. In terms of support, school leaders need to embrace and reward lifelong learning and self-e-learning; give psychological support for networking between staff and clients (education department, parents and stu-dents); create an e-teaching and learning environment/workplace; create conducive pathways for networking amongusers; make networking an incentive and a basic for the performance rating of staff; and provide a relevant infra-structure.Readiness on the part of school leaders and followers includes their attitude towards e-leadership; leadership skills, e-teaching and learning knowledge and skills; networking knowledge and skills; and computer-mediated competencies. Inaddition, school leaders should practice well-designed strategies, including reshaping objectives and curricula/co-curricula in line with e-teaching and learning; organizing e-competence training among teachers, students and par-ents; building long-term relationships among all users across boundaries andmonitoring progress; stimulating change forpositive mindsets towards e-teaching and learning and network thinking; developing a compelling mission and vision fornetworking; and managing the maximum use of ICT services (process, design, networking among users) (refer to sub-model 1).

2. The strategies taken by school leaders must be appropriate, with good practices that include adapting new models toincrease effectiveness; maximizing value from ICT spending by the institution; enhancing the use of ICT in teaching andlearning; leading a proper way of using ICT in schools; leading as role models for e-communication and encouraging activeparticipation; and setting proper institutional goals for implementing e-teaching and learning in the school. On the otherhand, to maximize the output quality of the strategies, leaders need to minimize or remove the four main e-leadershipobstacles, which include the failure of technology; a negative attitude and low commitment among users; insufficient ICTknowledge and skills; and insufficient ICT training (refer to sub-model 2).

3. The practice of e-leadership in implementing a VLE should be in line with the needs (demand for and investment in ICT;online instructional guides and coaching; developing more uses of ICT; global cooperation; leading for changes andenhancing the cooperation of the school with external parties, i.e., education officers, parents, staff and students). Inaddition, creating a conducive e-leadership culture maximizes the e-leadership practices. The strategies for encouraging aresearch culture include creating an appetite among users (teachers, students and parents); designing user-friendlycurricula to enhance the rate of usage; fostering multi-disciplinary approaches to networking among users; promotinga better and greater use of e-teaching and learning. Nevertheless, the practices will be maximized if the four main ob-stacles noted in the second criterion are minimized (refer to sub-model 3).

4. To cultivate an e-leadership culture in the school, schools need to provide full support to their community. This supportincludes providing psychological support and implementing a positive reward system for e-learning; providing a relevantinfra-structure and workplace; and creating conducive pathways for networking (refer to sub-model 4).

In addition to the above findings, a non-hypothesized finding that emerged from the data is that there were significant,positive and strong inter-correlations among needs, culture and readiness and the correlation were strong. The relationshipsamong the three variables can possibly be explained by the Strategic Change Model proposed by Dunphy (1996). According tothis strategic approach, the organizational leader determines the change goals by analyzing the organizational needs andculture in an effort to create a competitive strategy. This model proposes that, once the leader experiences the pressure ofindividual and organization needs, he will acquire knowledge and skills, be ready for a change and the readiness in turn willindirectly change the culture in the organization.

The literature review shows that, to date, no grounded model of e-leadership has been developed through research. Thegrounded model of e-leadership quality of this study is supported by several researchers who state that e-learning quality isdirectly dependent on technology; thus, e-leaders must provide strong support for the implementation of the e-learningplatform (Fonstad, 2013), focus on the needs and readiness of users (Hung, 2016; Jameson, 2013; vanWelsum& Lanvin, 2012)and be able to manage the ever-evolving nature of technology (Garcia, 2015) while maintaining a conducive culture withinthe groups of users (Albidewi, 2014). In implementing e-learning in schools, leaders should always be aware and attempt toreduce the obstacles to maximize practices and strategies (Alwidi & Cooper, 2015; Bowen et al., 2013; Lilian, 2014; Weng &Tang, 2014) to achieve a high level of quality of e-leadership practices in schools.

5. Conclusion

This article proposes a model for the implementation of e-leadership in educational institution, especially in schools.Educators and researchers can use this information to identify unanswered issues or questions in the literature and definefuture research directions concerning e-leadership. It helps educators better understand the concept of effective e-leadershipand the factors that are related to it.

Most grounded theory research stops after the theory/model is generated. The strength of this study is that it suggests anapproach to further validating the grounded model (from stage 1) with model fit testing analysis (in stage 2) to improve thegeneralizability of the model.

Because the findings of this study are limited to the characteristics of the school sample and the VLE e-learning platform,further research in e-leadership must be conducted in other locations and areas to provide a greater picture of e-leadershippractices.

Y.P. Chua, Y.P. Chua / Computers & Education 109 (2017) 109e121120

Funding

This work was supported by the Institute of Research Management & Monitoring, University of Malaya, Malaysia [Uni-versity of Malaya Research Grant. Grant number: UMRG - RG345-15AFR].

Declaration of conflicting interests

The authors declare no potential conflicts of interest with respect to the research, authorship, and/or publication of thisarticle.

References

1BestariNet. (2012). Empowering the next generation through the internet. http://1bestarinet.net/?page_id¼15.Afshari, M., Bakar, K. A., Wong, S. L., Samah, B. A., & Fooi, F. S. (2009). Factors affecting teachers' use of information and communication technology. In-

ternational Journal of Instruction, 2(1), 76e104.Albidewi, I. (2014). E-leadership system: A futuristic vision. International Journal of Business and Management Review, 2(2), 91e101.Alvarez, A., Martín, M., Fernandez-Castro, I., & Urretavizcaya, M. (2013). Blending traditional teaching methods with learning environments: Experience,

cyclical evaluation process and impact with MAgAdI. Computers & Education, 68, 129e140.Alwidi, I. T., & Cooper, M. (2015). Using management procedure gaps to enhance e-learning implementation in Arica. Computers & Education, 90, 64e79.Aral, S., Brynjolfsson, E., & Wu, L. (2012). Three way complimentarities: Performance pay, HR analytics and information technology. Management Science,

58(5), 913e931.Avolio, B. J., & Kahai, S. S. (2003). Adding the “E” to e-leadership: How it may impact your leadership. Organizational Dynamics, 3(4), 325e338.Avolio, B. J., Kahai, S. S., & Dodge, G. E. (2001). E-leadership: Implications for theory, research, and practice. Leadership Quarterly, 11, 615e668.Awang, Z. (2014). A Handbook on structural equation modeling. Malaysia: MPWS Rich Resources.Barwick, D., & Back, K. (2007). High tech's double edge: Creating organizationally appropriate responses to emerging technologies. On the Horizon, 15(1),

28e36.Blasco-Arcas, L., Buil, I., Hernandez-Ortega, B., & Sese, F. J. (2013). Using clickers in class. the role of interactivity, active collaborative learning and

engagement in learning performance. Computers & Education, 62, 102e110.Bowen, E. E., Bertoline, G. R., Athinarayanan, R., Cox, R. F., Burbank, K. A., Buskirk, D. R., et al. (2013). Global technology leadership: A case for innovative

education. Praxis, Procedia Social and Behavioral Sciences, 75(3), 163e171.Byrne, B. M. (2010). Structural equation modeling with AMOS: Basic concepts, applications, and programming (2nd ed.). N.Y: Routledge, Taylor and Francis

Group.Chang, I. H., Chin, J. M., & Hsu, C. M. (2008). Teachers' perceptions of the dimensions and implementation of technology leadership of principals in

Taiwanese elementary schools. Educational Technology & Society, 11(4), 229e245.Chenitz, W. C., & Swanson, J. M. (1986). From practice to grounded theory. Menlo Park, CA: Addison-Wesley.Chua, Y. P. (2016). Mastering research methods (2nd ed.). Shah Alam: McGraw-Hill Education.Chua, Y. P., Tie, F. H., & Zuraidah, M. D. (2013). Creating an education research acculturation theory for research implementation in school. Education and

Urban Society, 45(4), 506e513. http://dx.doi.org/10.1177/0013124511413124.Cohen, L., Manion, L., & Morrison, K. (2011). Research methods in education (7th ed.). New York: Routledge.Conrad, C. F. (1995). A grounded theory of academic change. In B. G. Glaser (Ed.), Grounded theory 1984e1994 (pp. 699e718). Mill Valley, CA: Sociology Press.Creswell, J. W. (2005). Educational research: Planning, conducting, and evaluating quantitative and qualitative research. Upper Saddle River, NJ: Pearson.DasGupta, P. (2011). Literature review: E-leadership. Emerging Leadership Journeys, 4(1), 1e36.Dundar, H., & Akcayir, M. (2014). Implementing tablet PCs in schools: Students' attitudes and opinions. Computers in Human Behavior, 32, 40e46.Dunphy, D. (1996). Organizational change in corporate setting. Human Relations, 49(5), 541e552.Evans, W. K., Ashbury, F. D., Hogue, G. L., Smith, A., & Pun, J. (2014). Implementing a regional oncology information system: Approach and lessons learned.

Current Oncology, 21(5), 224e233. http://dx.doi.org/10.3747/co.21.1923.Ferguson, C. J. (2009). An effect size primer: A guide for clinicians and researchers. Professional Psychology: Research and Practice, 40(5), 532e538.Fonstad, N. (2013). E-leadership, e-skills for competitiveness and innovation vision, roadmap and foresight scenarios: Final report, e-skills vision. The

European Commission, 67e103.Garcia, I. (2015). Emergent leadership: Is e-leadership importance in the quality of virtual education? RIED: Revista Iberoamericana de educaci�on a Distancia,

18(1), 25e44.Glaser, B. G. (1978). Theoretical sensitivity: Advances in the methodology of grounded theory. Mill Valley, CA: Sociology Press.Glaser, B., & Strauss, A. (1967). The discovery of grounded theory. Chicago, IL: Aldine.Gomes, W. (2011). Leadership in educational technology: Insights from the corporate world. Journal of Leadership Studies, 4(4), 57e60.Hadjithoma-Garstka, C. (2011). The role of the principal's leadership style in the implementation of ICT policy. British Journal of Educational Technology,

42(2), 311e326.Hair, J. F., Black, W., Babin, B. J., & Anderson, R. (2010). Multivariate data analysis (7th ed.). Upper Saddle River, NJ: Prentice Hall.Hanna, N. K. (2009). E-Transformation: Enabling new development strategies. Bathesda, MD: Springer Science & Business Media.Hu, L. T., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural

Equation Modeling, 6(1), 1e55.Humbly, L., O'Neill, T., & Kline, T. (2007). Virtual team leadership: The effects of leadership style and communication medium on team interaction styles and

outcomes. Organizational Behavior & Human Decision Processes, 103(1), 1e20.Hung, M. L. (2016). Teacher readiness for online learning: Scale development and teacher perceptions. Computers & Education, 94, 120e133.Jameson, J. (2013). E-Leadership in higher education: The fifth “age” of educational technology research. British Journal of Educational Technology, 44(6),

889e915. http://dx.doi.org/10.1111/bjet.12103. Special Issue: e-Learning and Leadership.Jameson, J., Ferrell, G., Kelly, J., Walker, S., & Ryan, M. (2006). Building trust and shared knowledge in communities of e-learning practice: Collaborative

leadership in the JISC, eLISA and CAMEL lifelong learning projects. British Journal of Educational Technology, 37(6), 817e990.Kearsley, G., & Lynch, W. (1994). Educational technology: Leadership perspectives. Englewoood cliffs, NJ: Educational Technology Publications, Inc.Kline, R. B. (2016). Principles and practice of structural equation modeling (4th ed.). New York: The Guilford Press.Kroner, G. (September 23, 2014). LMS data e the first year update. Edutechnica. http://edutechnica.com/2014/09/23/lms-data-the-first-year-update/.Lilian, S. C. (2014). Virtual teams: Opportunities and challenges for e-leaders. Procedia Social and Behavioral Sciences, 110, 1251e1261.Martin, W. (2007). Virtual learning environments: Using, choosing and developing your VLE. London: Routledge.Mcleod, S., & Richardson, J. W. (2011). The dearth of technology leadership coverage. Journal of School Leadership, 21, 216e240.McPherson, M. A., & Nunes, J. M. (2008). Critical issues for e-learning delivery: What may seem obvious is not always put into practice. Journal of Computer

Assisted Learning, 24(5), 433e445.

Y.P. Chua, Y.P. Chua / Computers & Education 109 (2017) 109e121 121

Mohd Yusri, I. (2014). Model of e-leadership, intra-team communication and job satisfaction among school leaders in Malaysia. Mediterranean Journal ofSocial Sciences, 5(23), 1927e1931.

Orus, C., Barles, M. R., Belanche, D., Casalo, L., Fraj, E., & Gurrea, G. (2016). The effects of learner-generated videos for YouTube on learning outcomes andsatisfaction. Computers & Education, 95, 254e269.

Purvanova, R. K., & Bono, J. E. (2009). Transformational leadership in context: Face-to-face and virtual teams. The Leadership Quarterly, 20, 343e357.Ringmayr, T. G. (2012). Atlas.ti 7 quick tour: Revision 27. Berlin: ATLAS.ti Scientific Software Development. http://atlasti.com/wp-content/uploads/2014/05/

QuickTour_a7_en_07.pdf.Schultz, R. W. (2010). Exploring leadership within the modern organization: Understanding the dynamics of effective leadership of a virtual, multigenerational

workforce. PhD Disertation. Capella University.Spradley, J. P. (1980). Participant observation. Orlando, FL: Harcourt Brace Jovanovich College Publishers.Strauss, A. (1987). Qualitative analysis for social scientists. Cambridge, UK: Cambridge University Press.Strauss, A., & Corbin, J. (1990). Basics of qualitative research: Grounded theory procedures and techniques. Newbury Park: SAGE.Strauss, A., & Corbin, J. (1998). Basics of qualitative research: Techniques and procedures for developing grounded theory. Thousand Oaks, CA: SAGE.Tan, S.C. (2010). School technology leadership: Lessons from empirical research. In C.H. Steel, M.J. Keppell, P. Gerbic, S.Housego (Ed.), Curriculum, technology

and transformation for an unknown future. Proceedings of the 2010 Australasian Society for Computers in Learning in Tertiary Education (ASCILITE)2010 Conference. Sydney, Australia.

Tondeur, J., Devos, G., Van Houtte, M., van Braak, J., & Valcke, M. (2009). Understanding structural and cultural school characteristics in relation toeducational change: The case of ICT integration. Educational Studies, 35(2), 223e235.

van Welsum, D., & Lanvin, B. (2012). E-leadership skills: Vision report. prepared for the European Commission. DG Enterprise and Industry http://eskillsvision.eu/fileadmin/eSkillsVision/documents/Vision%20report.pdf.

Weng, C. H., & Tang, Y. (2014). The relationship between technology leadership strategies and effectiveness of school administration: An empirical study.Computers & Education, 76, 91e107. http://dx.doi.org/10.1016/j.compedu.2014.03.010.