Needs of digital transformation requires specific flexibility from modern universities to ensure the society demands implementation through innovative teaching and IC-technologies. Modern universities create a digital learning environment to support studying activities. This research presents an experts' estimate of the current condition and perspectives of universities digital studying environments in Ukraine. We verified the theoretical model structure of the university digital studying environments by means of the empirical data factor analysis. We studied the components of the existing learning environment and enabling environment and compared them to the results of our previous research. We proved the digital learning environment theoretical model was correct. We proved that visions of students and teachers correspond to the key trends accelerating higher education technology adoption. We assume the digital learning environment development benefits overcoming significant challenges impeding higher education technology adoption.
The biggest digital transformation ever occurs right now. Unfortunate countries and
enterprises those won’t be able to adapt are done for. The Global Competitiveness
Report 2018 claims that the promise of leveraging technology for economic
leapfrogging remains largely unfulfilled [1, p.9]. A number of organizations require help to
envision, structure, and sequence successful digital transformation efforts [
Needs of digital transformation requires the flexibility of modern universities to
ensure the implementation of society demands through innovative teaching and
ICtechnologies. Leveraging these technologies requires not only the creation of the
digital learning environment [
The tools to assess competitiveness, along with traditional concepts (such as ICT and physical infrastructure, macroeconomic stability, property rights, years of schooling) become crucially important concepts those go in a row with an entrepreneurial culture, multi-stakeholder collaboration, critical thinking, and social trust [1, p.7]. All these factors together influence the universities’ competitiveness. Under a condition of the education system digital transformation enabling environment is meant to become the university digital learning environment (DLEs) with its following integration to the global digital environment.
Digitalization of the educational environments will improve the university competitiveness, that is important both for the students who decide on what university to choose and for the universities interested in attracting potential students, best teachers and researches, investments and grants.
This research aims to prove the theoretical model of the university digital environment structure and evaluate its relevance and perspectives for universities in Ukraine. 2
Existing research studies in higher education proved that it’s easier to engage students to learn with when ICT [5, 6]. The universities’ key priority is improving their digital environment, that would support new academic policy, practices and technological landscape [7]. Accepting the digital learning environment in many ways depends on the educational trends and the most recent educational requirements. However, the technologies are also important for DLEs development. Digital learning environments include any set of digital tools and technology-based methods that can be applied to support learning and instruction [8]. We can claim that DLE is a next stage for the elearning environment and the virtual learning environment [9], however, some researchers use these terms as synonyms. Universities and non-commerce organizations research on designing and developing digital learning environments and their effectivity. The digital learning environment Manifesto from the Edutainme aims to proclaim the principles of how to create digital learning environments, where the student will be a performer of his own learning, entitled to influence his own growth [10]. DUCAUSE (e.g., https://library.educause.edu/topics/teaching-and-learning/nextgeneration-digital-learning-environment-ngdle) helps elevate the impact of IT, thus the next generation digital learning environment (NGDLE) concept seeks for a balance between the openness of learning and the need for coherence in the environment and emphasizes personalization, collaboration, and accessibility/universal design – all essential to learning.
The university digital learning environment on different levels can be indicated by electronic scientific and educational resources, communication in the scientific and educational environment, management of scientific and educational activities, the formation of new scientific and educational relations, competences. An international Project IRNet studied its participants' evaluation indicators of the digital environment in various universities and IC-competencies [11].
Herewith, the projects on improving the digital learning environments require both the teachers and students to participate in (https://www.plymouth.ac.uk/news/connect/spring15/digital-learning-environment). As soon as the students order educational service, and the teachers are responsible to provide these services at a great level, they become the categories to ask for an expert estimate of the higher education level and its components [12].
This assumption corresponds to the quality management principles of ISO quality management standards [13], namely QMP 1 – Customer focus and QMP 3 – Engagement of people. 3 3.1
There exist various approaches to define the university digital environment components [14]. This paper considers the university digital learning environment as a cluster of components, which structure was modeled and proved by Ukrainian researcher L. Panchenko [15]. The author distinguishes such components as available equipment and Internet access (space-semantic component), students and teachers information competency (competency component), communication, and organization of the learning process (technological). As far as the received results validity depends on the research reproducibility [16] we conduct the repeated expertise on the mentioned components, taking into account the changes occurred lately. The MC Horizon Report claims there exist consistent educational trends, new trends appear all the time, and some trends and issues reappear over time [3, рр. 4-5]. For example, the need in growing focus on measuring learning and redesigning learning space is still immediate. The requirements to the open educational resources (OER) and their proliferation change the requirements of cross-institution & cross-sector collaboration; rise of new trend. The new forms of interdisciplinary studies step forward. The modern universities react to the changing requirements. The technology development (open source software for scientific communication), wider access to the external resources (scientific platforms and databases), rising demands and educational requirements from the students and such objectives as academic mobility and scientific cooperation, including the international cooperation, lead to specifying the components of the suggested theoretical model. The common tendencies rely on the transformation of the educational and information environments into the digital one, information competency into the digital competency, communication in education, that is not limited to the university environment. Scientific researches in the field of advancing cultures of innovation, advancing digital equity plays an even more important part.
To understand the attitude of the Ukrainian teachers and students to the universities’ digital learning environment we put together a set of the theoretical model components.
- Space-semantic: Available Internet access, good traffic, equipped studying rooms, hostings, and educational platforms, particularly LMS, e-library, institutional repository, e-conference system, access to the wiki-portal and corporate accounts, etc.
- Technological: educational resources integration (e-library, OJS edition, repositories, etc.), content development and delivery, access to the external educational sources, scientific databases, well- organized consultation and expert estimation system, creating the educational program according to the educational requests from the students, monitoring and tweaking the processes of using the environments for individual work, applying e-learning, project-based learning, blended learning, collaborative learning, combined formal and informal learning, shared research work, etc.
- Communicative: scientific and educational communication through email, corporate resources (websites of departments professors, and conferences, corporate clouds, e-libraries, etc.), external resources (social networks and services, forums and communities, e-conferences, etc.), consulting, experts’ evaluations.
- Competency-based: the level of digital competencies through self-evaluation, peer-to-peer evaluation, e-portfolio, achievements recognition, motivation and training those who can improve the level of digital competency.
The authors of the article claim, that the defined challenges impeding higher education technology adoption can be solved by building and applying the digital learning environment. Thus, the digital learning environment contributes authentic learning experiences, improving digital literacy, adapting organizational designs to the future of work, advancing digital equity.
Research Tasks: 1. Provide a theoretical model of the university digital educational environment expertise and to build a statistical factor model of the university digital environment. 2. Analyze if the digital educational environment of the Ukrainian universities corresponds to the digital and educational trends.
Assumptions: 1. The digital environment model planned to build using the statistical methods and models corresponds to the suggested theoretical model. 2. The universities digital environments development reacts to modern technologies and educational trends. That is also one of the tools to overcome the challenges impeding higher education technology adoption and to improve on the higher education quality. 3.2
We performed the expert estimate of the university digital environment by means of online inquiry and in-depth interview (in case if we needed elaborateness). We distributed the survey (https://forms.gle/7h56MAxf5JAGQ9Eh6 ) with mailout and specific-purpose contacts with the educational institutions.
To perform the expertise of the university digital environment for our research we invited masters and teachers (professors) from the best universities of Ukraine (where 70% are research universities). Mostly, our respondents’ occupations lie in the field of Mathematics, Computer programming, IT (28%), Education (22%), those are considered to be the top-priorities in Ukraine. The age, gender, and positions of the sampled population represent the real situation in the educational institution: there are more students and teachers, the age of students and teachers corresponds to the age-grade in general, there are more women among the respondents that is natural gender correlation for the educational institutions in Ukraine. The research didn’t take into account the connections between the features, fields of occupation and the educational institutions, that is why it can’t be considered from that point of view. Mostly, our respondents had assessed to the computers and to the international scientific databases. The non-sampling error on the studied features didn’t exceed 9% (123 person). The full list of the estimated features that reflect personal data of respondents is provided in Table 1. Every feature has calculated beforehand descriptive statistics and constructed frequency distributions. Knowing the level of the respondents’ digital competency is essential to conduct an estimation of the digital educational environment. Mostly, our respondents evaluated heir levels as middle and advanced proficiency [17]. In addition to the questions on the research topic, they had to answer if they had registered profiles in the scientific databases such as Web of Science (WOS) or Scopus, personal profiles in the ResearchGate social network, publications in the online journals or experience in informal education. We added these questions to understand if our experts are ready to overcome such challenges as advancing digital equity and participating crossinstitution & cross-sector collaboration. The answers we received were mostly positive. Those respondents who had no profiles in scientific databases or experience with online conferences and courses claimed they wished to have that experience and believe in its importance. Herewith, we observe an obvious statistical connection between the estimated level of the respondents’ digital competencies and the answers to the mentioned questions. Table 2 contains answers of our respondents. Thus, we can claim that aggregated values on the selection that we received correspond to the goals of our research. The level of competencies allows teachers and students (masters) who participated in our survey to be the experts.
The questionnaire contained two question pools considering the students and teachers attitude to the educational information environment of the university, the need for the environment development, and the questions considering the respondents' personal data and competency level.
The questions consider the university digital learning environment, which our theoretical model consider as 4 interacting components. The questionnaire has 4 sections that correspond to 4 components: space-semantic, technological, communicative, and competency-based. Each section contains from 6 to 14 assertions. The respondents estimated if the mentioned components are available in their educational environments (1st group of questions) from 1 to 4 points (where 1 stand for the poor level of availability, 2 is for middle level of availability, 3 is the enough level and 4 stands for the expert level). The respondents estimated the importance and availability of improvement for the mentioned components (2nd group of questions) from 1 to 3 points (where 1 stands for low, 2 for the middle, 3 for high). For example, respondents have a request “Please rate the proposed components of the environment on a scale of 0-4” and several assertions such as: “Your university has access to broadband internet”, “You can access the internet in every lecture hall in your university”, etc. 3.3
The choice of methods is determined by the purpose of the study. We needed to process a rather large array of statistical data and identify the main patterns. During the research, we applied the methods of descriptive statistics to find the frequency distribution and to define the central tendency rates. To prove the hypothesis we stuck to the statistical inferences methods and models. The method selection based either on the type of the scale used for estimation or on the datatype of the features we had to estimate. To analyze connections between the features we applied the methods of correlation and regression analysis. The calculations were conducted based on the sampled population, and the statistical results were verified at the 95% integrity level.
During the study, it was necessary to consider a large number of variables that describe the digital environment of the universities. However, it is difficult to identify patterns in a large array of features without data reduction. With factor analysis, we managed the empirical data received in the survey, performed the data reduction and shortened the number of features, in order to study the received model structure of the educational environment of the university. The factor analysis was performed in accordance with the basic stages: defining the preliminary features to be reduce, building a correlation matrix to find the connection between the elements, defining the methods of data reduction, choice and explanation of the main factors, calculations and interpretation of the results we received.
The analysis is not reliable if the basic requirements for the reliability of data and measurement scales are not taken into account. To estimate the reliability of suggested scales we used the intraclass correlation coefficients so that later we could calculate the ‘intra respondents’ estimates of reliability. To find out the internal consistency in the survey we found the Cronbach's alpha and Spearman-Brown coefficient. The calculations were done mostly using SPSS software [18, 19].
At the first stage of our research, we estimated the reliability if the respondents’ answers and analyzed what different kind of analysis we can apply.
For the questions on the university digital learning environment, we performed separate analysis considering the availability of required components in their learning environments. Just the same we performed a separate analysis on the components’ level of development and on ways to improve some component.
For both these questions the Cronbach's alpha and Spearman-Brown coefficients were quite good: Cronbach's alpha — 0.981 and 0.978; Spearman-Brown — 0.902 and 0.861. According to the correlation matrix we built, the correlation of some points of the survey was equal to 0.78-0.79. That means, that the features used to build the theoretical model shared common agents that can be combined. The received numbers are reliable, that we proved with the Fisher coefficient equal to p<0.05. 4.2
At the next stage, we leveraged the factor analysis based on the method of main components [20, 21]. This method allows reducing the number of features that describe the university digital learning environment in accordance with a theoretical model build before. The factor analysis validity was proved both when evaluating the respondent’s answers and based on query obtained from the Kaiser-Meyer-Olkin measure of sampling adequacy and Bartlett’s test of sphericity [18].
Besides that, we calculated these criteria for the questions that consider the availability of the required components (group 1) and the importance of components development and improvement (group 2).
The Kaiser-Meyer-Olkin measure of sampling adequacy for the questions in the 1st group is equal to 0.9. High criteria value (from 0.5 to 1) proves that the factor analysis was viable in this case. Low values (less than 0.5) prove that the factor analysis is not beneficial for the specific situation. Thus, for our case, we can use the factorial analysis. The Bartlett’s value of sphericity is equal to 9132.97 at df=2080, that is large for the р<0.001 level, and also proves that factorial analysis is beneficial for this specific case.
The value of Kaiser-Meyer-Olkin criteria for the questions of the 2nd group is equal to 0.802. The Bartlett’s value of sphericity is equal to 8599.164 at df=2080, that is large for the р<0.001 level. This group is also appropriate to use the factor analysis.
At the next stage, we defined a number of factors. There are several methods to do so, such as to calculate the proper values, or to use a scree plot and Kaiser’s criterion [18]. However, for this research, we considered the worked through the information of the problem structure that we received from the previous stages of the research and confirmed that structure with statistics (namely, with the sampling variance percentage). As a rule, the researchers recommend selecting the number of factors that samples at least 60% of the variance.
Based on the environment expertise results and theoretical analysis we received before, we selected the 4 factor model of the university digital learning environment that has such components as special equipment and Internet access, educational websites and portals, teachers’ and students’ digital competencies and communication, and a well-organized process of education.
According to the sampling variance percentage criteria, we can claim that 4 factors for the 1st group of questions sample over 60% of the variance (63.51%), and almost 60% for the 2nd group 58.27% (see tables 3 and 4). In these tables, you can also find the sampling variance percentage after we turned the matrix of main components. The numbers in the ‘variance %’ column proves that the components we’ve built are quite informative.
We can see that in the 1st group of questions 3 groups of components sample the part of variance (20.85%, 17.07% , 16.2%), while for the 2nd group the distribution is more smooth between 2 first components (16.76%, 16.53%) and 2 following components (13.82% , 11.16%). Thus, some factors in the information environment correlate more, and so explain the percentage of the factors variation, while the importance of development is the same for all components. For the factor rotation, we utilized a common rotation method “varimax” that minimizes the number of variables with high values and increases the possibility for factor interpretation. At the following stage, we received rotated solutions of the factor matrix, that allowed us to combine the features according to the results of the factor values of the 4 separate components. In tables 5 and 6 you can see the fragments of the factor loadings, as they were quite a lot of features for every group. We should also mention that the features for the groups 1 and 2 were grouped with different approaches, so the main components were interpreted separately. Teachers’ usage of Internet social services Participation in scientific communities in the university Students’ participation in scientific social networks Teachers’ participation in scientific social networks Teachers’ participation in the professional Internetcommunities Students’ participation in the professional Internetcommunities Searching and inviting the experts (scientific consultants, mentors, etc.) Consultations and reviewing, in particular at the webinars and in Internet-communication Students’ publications in online journals Students’ participation in online conferences Students’ usage of Internet social services Teachers’ publications in online journals Teachers’ participation in online conferences Teachers’ usage of emails, in particular, the corporate accounts Students’ usage of emails, in particular, the corporate accounts Systematic publications of records about the completed plans, scientific activities, cooperations, etc.
The level of students’ digital competency Experience utilizing digital competencies in scientific work We defined the variables that have high load values on the same factor. Then, we analyzed this factor considering the mentioned variables. We also interpreted the variables’ graphics, those coordinates the factor loads (Figure 1.).
Fig. 1. The graph of the contribution of characteristic values to the main components: groups 1, 2 (Source: Own work) As a result, we received a confirmation for the university digital learning environment theoretical model we built, as with small deviations we managed to combine and group features into four components.
Here we suggest a data interpretation of receive four-factor model for the 1st and 2nd groups. We found a common factor that corresponds to the competency-based component that we defined both rot 1s and 2nd groups. To the 2nd group got an additional element “distant learning” (see Table 6), that we can explain as the readiness of teachers and students for self-education and online study mode. Today, (group 1) the face-to-face courses are used mostly for improving the level of competencies in the universities. The other factors for the 1st and 2nd groups differ.
The second factor (group 1) has high factor loads connected by technological and space-semantic component. The respondents in the available environment do not distinguish the space-semantic component, yet consider the technological component as an optimal combination of infrastructure, resources topology and educational technologies. The third factor (group 1) corresponds to the communicative component that has features of scientific communication by means of digital technologies (Table 5). This component also included the variables connected to the returns automation and systematic academic and scientific journals declaration, and students and teachers mobility (technological component). Thus we can make an assumption on the availability of communication management from the university. The fourth factor (group 1) we would explain as collaborative and research component, as it contains the variables of monitoring and correcting the process of environment usage for self-guided work, formatting the messages of education according to the student’s requests, learning in cooperation, applying inquiry-based learning, using e-library, wiki-portal, and availability of internet traffic. Thus, we can argue defining a component that combines separate features of space-semantic, technological, and communicative components and corresponds to trends in education.
Among the factors of university digital environment development (group 2) spacesemantic component corresponds a lot to the theoretical model. The respondents assume that building a modern infrastructure and resources topology is a basis to build the university digital environment. The third factor included the variables connected to the educational and scientific communication (communicative component) and organization of the process of education (technological component). In the improved environment (group 2), the respondents consider communication resources and pedagogic strategies to be a part of the technological process. For example, preparation, organization, and participation in the conferences must be conducted in terms of learning (self-conducted work), researching, and leveraging training projects. The fourth factor, that we can call communicative and dissemination, has such features are participating scientific societies, using the social services, wiki portals, creating and supporting websites of departments, participating research projects, etc. Thus, we can assume this component to mostly correspond to the communicative component. Though, at the same time, it includes some features of the competency-based component, connected to the presentation of the achievements and reports automatization. 4.3
Comparing to the environmental expertise of 2013 [15] we can claim the results
repeatability. The model of the university digital learning environment that we received
by leveraging the factor analysis corresponds to the theoretical model of the
information and educational environment both for 1st and 2nd-factor groups. However, we
observe the development that corresponds to modern requirements [
We used factor analysis to confirm the theoretical model. We used it to find 4 main components that group all the factors of the digital educational environment into such areas of focus as IT infrastructure and resources' provision, students’ and teachers’ digital competencies, scientific and educational communication between the students, teachers, and stakeholders, and educational process organization.
Comparing to the results of similar researches, even if we take into consideration the global development of the distance learning, online courses, open electronic resources, and redesigning learning spaces, we observed no significant changes on the main factors during the 2013-2018. However, messages, contents, and scopes change.
Both students and teachers claim that enabling digital learning environment as an improvement of the existing learning environment correlates to the mid- and longterm key trends accelerating higher education technology adoption: proliferation of open educational resources, the rise of new forms of interdisciplinary studies; advancing cultures of innovation, cross-institution & cross-sector collaboration. Effective implementation of the digital learning environment, both at the stage of designing and applying its methods, helps to overcome significant challenges impeding higher education technology adoption. Thus it empowers implementation of authentic learning experiences and improving digital literacy (solvable); adapting organizational designs to the future of work, advancing digital equity (difficult). However, having good enough IT-infrastructure, equipment, and level of digital competencies of the educational process participants, the solution of the problem depends more on the rethinking the roles of educators in the digital learning environment.
We consider the pedagogic design of the educational and scientific cooperation in the digital learning environment to be a prospective field for further research. We need to find out what factors influence the competencies of the digital environment participants most of all.
Repeating the digital learning environment expertise after a period of time for more respondents, and engaging participants (universities) from different countries will allow us to find out if the universities are ready for transformation to confirm the demands of the modern digital society. 5. Bryson, C., & Hand, L.: The role of engagement in inspiring teaching and learning. Innovations in Education & Teaching International, 44(4), 349-362 (2007). 6. Kuzminska, O., Morze, N., Smyrnova-Trybulska, E.: Flipped learning model: Tools and experience of its implementation in higher education, The New Educational Review Vol. 49, No. 3, pp. 189-200. DOI: 10.15804/tner.2017.49.3 (2017). 7. Smyrnova-Trybulska, E., Noskova, T., Pavlova, T., Yakovleva, O. and Morze, N.: New educational strategies in the contemporary digital environment, International Journal of Continuing Engineering Education and Life-Long Learning, Vol.26 No.1, pp.6 – 24, DOI: 10.1504/IJCEELL.2016.075036 (2016). 8. Wheeler, S.: E-Learning and digital learning. In N. M. Seel (Ed.), Encyclopedia of the sciences of learning, pp. 1109–1111, New York: Springer (2012). 9. Smyrnova-Trybulska, E., Morze, N., Glazunova, O.: Design of a University Learning Environment for SMART Education, Smart Technology Applications in Business Environments, DOI: 10.4018/978-1-5225-2492-2.ch011 (2017). 10. Digital Learning Environment Manifesto, http://manifesto.edutainme.ru/en, last accessed 2019/10/03. 11. Drlík, M., Švec, P., Kapusta, J., Munk, M., Noskova, T., Pavlova, T., Yakovleva, O., Morze, N., Smyrnova-Trybulska, E.: Identification of Differences in University Eenvironment between Selected EU and Non-EU Countries Using Knowledge Mining Methods: Project IRNet Case Study, International Journal of Web Based Communities (IJWBC), Vol. 13, No. 2, DOI: 10.1504/IJWBC.2017.10004116 (2017). 12. Center for Studies in Higher Education. The University of California, Berkeley, 2009, https://cshe.berkeley.edu/seru, last accessed 2019/10/03. 13. Quality management principles, https://www.iso.org/files/live/sites/isoorg/files/archive/pdf/en/pub100080.pdf, last accessed 2019/10/03. 14. Witl, M., Dompseler, H.: How to create a digital learning environment consisting of various components and acting as a whole?, http://www.eunis.org/download/2017/EUNIS_2017_paper_16.pdf, last accessed 2019/10/03. 15. Panchenko, L.: Modeling the Informational Educational Environment of University with Factor Analysis, Bulletin of the University of Luhansk, № 10 (269), P. ІII, pp. 6-17, http://nbuv.gov.ua/UJRN/vluf_2013_10(3)__3 (2013) (in Ukrainian). 16. Reproducibility gold standard, https://www.coursera.org/learn/dataresults/lecture/KOraT/reproducibility-gold-standard, last accessed 2019/10/03. 17. Kuzminska, O., Mazorchuk, M., Morze, N., Pavlenko, V., & Prokhorov, A.: Digital Competency of the Students and Teachers in Ukraine: Measurement, Analysis, Development Prospects. In: Information and Communication Technologies in Education, Research, and Industrial Applications, Communications in Computer and Information Science, vol. 2104, p. 366-379 (2018). 18. Levesque, R.: SPSS Programming and Data Management, 2nd Edition A Guide for SPSS and SAS Users. Chicago: SPSS Inc. (2005). 19. Carmines, E., Zeller, R.: Reliability and validity assessment. London, Sage Publications,
Quantitative applications in the social sciences series no. 17 (1979). 20. Kutner, M., Nachtsheim, C., Neter, J., and Li, W.: Applied Linear Statistical Models,
McGraw-Hill/Irwin, Homewood, IL. (2004). 21. Jolliffe, I.: Principal Component Analysis, 2nd Edition. Springer-Verlag New York, Secaucus, NJ. (2002).