Evaluating Digital Maturity in Higher Education Institutions: A Preliminary Empirical Study in the Western Balkans

Abstract

Digital transformation (DT) has become one of the most significant trends in higher education institutions (HEIs) in both EU and non-EU countries. Using Information and Communication Technologies (ICTs) to reinvent higher education is contingent upon several factors, including an institution’s development stage regarding the application and strategic integration of ICTs across its key activities and processes. In the extant literature, multiple frameworks of ICT development (maturity) paths have been developed. However, there is a lack of empirical studies on how well those models predict the DT success, and which of their dimensions are most relevant. In this paper, we use a research instrument, adapted from the HigherDecision research project, to capture the subjective assessments of academics and students at three public higher education institutions in Bosnia and Herzegovina and Croatia. Using seven dimensions of the DT construct, prescribed by the HigherDecision framework, we examine their contribution to the subjectively evaluated success of each HEI’s DT initiative and identify the most impactful dimension(s). Our results show that the digital infrastructure and academic teaching and learning are perceived as critical drivers of DT in the academic sector. Provided that the University of Mostar, as a mid-sized public university located in Bosnia and Herzegovina, currently represents one of the DT leaders in the Western Balkans (WB) region, we discuss implications for scaling its good practices in smaller HEIs across the region.

1. Introduction

There is an increasing pressure to reinvent higher education by using ICTs, due to the increasing competitive and regulatory pressures, digitally intensive expectations of the current and future university students (belonging to the so-called Generation Z of ‘digital natives’), and the flexibilization of the traditional teaching and learning patterns [1].

Going beyond the limited initiatives of digitizing analog data into digital formats and digitalizing HEI processes, contemporary HEIs are targeting a series of ICT-enabled and mutually coordinated organizational changes, aiming to radically transform their strategy, operations, and the creation and delivery of value to HEI stakeholders. Such an approach is usually referred to as the digital transformation (DT) [2]. Fueled by the COVID-19 global crisis and the need to move their entire operations to the digital world quickly, many European universities realized they are not ready to address the complex DT challenges. Thus, a more comprehensive analytical framework was needed to support the HEIs in a successful transition toward the digital university. The European University Association (EUA) has developed such a framework in its DIGI-HE (2020–2023) project, supported by the European Union (EU) Erasmus+ project funds [3].

In EUA’s Digital Transformation Map, the fundamental enablers of an HEI’s DT are identified in terms of: (a) Institutional Culture (also referring to leadership, staff skills, as well as institutional strategy and good practices); (b) Digital Architecture and (c) Framework Conditions (which concerns the institutional environment and its requirements) [4].

The complex DT framework and environment make it difficult to track and follow an HEI’s progress through its DT journey. Therefore, an extremely popular construct of digital maturity (DM) is often used to mark the stages of ICT adoption and ICT-enabled organizational change. DM can be described as a comprehensive indicator of an organization’s development toward the fully developed adoption of a technological approach or technology-enabled organizational changes. It presumes that an organization evolves through a series of evolutionary stages, involving the development of digital maturity, which follow several generic stages (Initial, Emerging, Defined, Managed, and Optimized) [5]. As an organization progresses through the prescribed stages, it advances from an ad hoc to a systematic and strategic approach of using and leveraging ICTs. Different aspects of complex, ICT-driven transformations, such as those described by the EUA for the academic sector, can be measured and improved using the DM approach.

DM is widely accepted as a tool for benchmarking and quality improvement, and measuring progress toward best practices [6]. Such a conceptualization of ICT-enabled initiatives makes it relatively easy for managers and practitioners to perceive organizational change as a series of consecutive stages, leading toward the ultimate DT objective. Along this way, DM can be used both as a diagnostic tool (for measuring the current position and performance) and as a prescriptive guide for further organizational development [7].

DM has been criticized for its simplistic approach, and researchers argue that multiple maturity models only capture the specific needs and realities of individual industries or even specific consulting clients. In addition, there are multiple and diverse opinions on the number of maturity stages, objectivity of the maturity self-assessment, the lack of reliable external maturity indicators, and the unreliable relationship between the maturity self-assessment and the organizational performance [8]. Still, DM models provide a pragmatic tool of choice for many change managers, as well as internal and external organizational stakeholders, especially in underdeveloped sectors and contexts, where a theoretically grounded DT assessment strategy proves to be challenging to implement.

This challenge is especially pertinent to the Western Balkans region, including EU member states, as illustrated by the recent OECD review of higher education (HE) in Croatia. Although being an EU member-state and an advanced country, in the regional context, Croatian HEIs have low engagement with stakeholders, inadequate public funding, and mixed levels of infrastructural readiness for DT. In addition, the rhetoric of DT readiness is not always accompanied by the digital competencies and change motivation of HEI staff [9].

Although focused on the general DT patterns, another OECD report [10] emphasizes that the WB region collectively lags behind the developed EU member states. At the same time, Croatia could be used as a national benchmark for other WB countries. Still, HEIs focused on DM and DT, such as the University of Mostar (Sveučilište u Mostaru—SUM, located in Mostar, Bosnia & Herzegovina), are regionally recognized for their integrated ICT infrastructure, strategic governance, EU project engagement, and global innovation rankings.

Thus, we aim to conceptually analyze how individual maturity dimensions explain stakeholders’ perceptions of DT progress. On the other hand, our practical objective is to draw actionable lessons for smaller HEIs in the WB region. Thus, we formulate our research questions (RQs) as follows:

1.

RQ1. Which dimensions of digital maturity best predict stakeholders’ perceptions of successful digital transformation?

2.

RQ2. How can the identified good practices be adapted and used in benchmarking by smaller HEIs in the WB region?

In this context, this paper makes a dual contribution to the existing literature on DT in higher education and innovation practices in HEIs located in the WB region. Concerning the theoretical contribution, we provide comparative and regionally indicative empirical evidence from three public HEIs (two in Croatia and one in Bosnia and Herzegovina), using a shortened, validated version of the already validated DM instrument, grounded in the Digital Maturity Framework for Higher Education [11], developed by the Faculty of Organization and Informatics at the University of Zagreb, Croatia, within the nationally funded HigherDecision research project.

In addition to providing the region-specific empirical evidence to the extant literature, this contribution also has a methodological component, since the original DM self-assessment instrument is highly complex and presents a challenge to the respondents from the WB HEIs. A shortened version of the instrument, highlighting the most relevant dimensions, may be helpful for practitioners and administrators of regional HEIs who wish to conduct self-assessment and benchmarking exercises.

By focusing on the good practices from the University of Mostar (SUM), which can be identified as one of the regional leaders in infrastructure modernization and digital pedagogy, this study offers an analysis of good practices, relevant for HEI administrators and stakeholders, aiming to close the digital transformation (DT) gap across the WB region. Our study further confirms that regional leadership in digital infrastructure and digitally enabled academic teaching and learning is associated with success in digital transformation, at least from the relevant stakeholders’ viewpoints.

2. Theoretical Background

To avoid potential conceptual ambiguity, we define the digital maturity (DM) of an HEI in terms of the existing level of resources and competences, demonstrated by the current level of Information and Communication Technology (ICT) application across multiple areas. Those areas are referred to as DM dimensions, which span all relevant fields of HEI activities, including academic teaching and learning, research and technology transfer, community outreach and social mission, as well as supporting managerial and administrative processes [7]. The use of DM as a predictor of the realized, ICT-based change in HEI processes and their outcomes is supported by the common maturity frameworks, implying that organizations generally need to gradually develop their ICT-based resources and competences in order to reach higher levels of effectiveness [5].

This implies that the digital transformation (DT) should be conceptualized in terms of the previously mentioned level of realized ICT-based organizational change in an HEI. Using the EDUCAUSE framework, comprehensive organizational changes are found in business processes, the HEI’s strategy and its implementation, and the way an HEI creates value for its stakeholders [2]. Thus, DT should be theoretically positioned as a perceived outcome of the process of a significant organizational change, i.e., organizational transformation, and measured in terms of stakeholders’ assessment of the change outcome(s).

Therefore, a successful DT initiative denotes a comprehensive organizational transformation, based on the strategic usage of ICT resources and capabilities [1], In practice, higher levels of DM are typically associated with greater DT success. This is explicitly noted by Đurek et al. [11], who designed the HigherDecision DM self-assessment instrument, usually referred to as the DM Framework for Higher-Education Institutions (DMFHEIs). These authors, whose instrument has also been adopted by the OECD policy report for Croatia [9], single out the Optimized/Organizational as the most advanced DM stage. At this point, the ICT infrastructure, governance, teaching practice, and culture become fully aligned with institutional strategy.

However, not all dimensions of DM models are equally relevant for the stakeholders’ viewpoint of DM success. This can be concluded from the previous empirical results, analyzing the correlations among the DT dimensions and actors (stakeholders). In a recent research of 336 HEI stakeholders, Bravo-Jaico et al. [12] concluded that digital governance is the central DT dimension, bridging the internal efficiency of an HEI to its stakeholder engagement. Otherwise, external stakeholders are more concerned about the HEI’s marketing when considering its environmental impact. However, the socio-cultural impact of HEIs is considered to be much less relevant than the technical change. Consequently, academic teaching and learning are not yet considered fully addressed by the DT, as all stakeholder groups, except students, may believe that it is targeted for change only at a prescriptive level.

These results can be linked to the global DT framework, developed by EDUCAUSE, which emphasizes institutional culture change, leadership, and workforce development as key aspects of DT [13]. Previous analysis [12] is closely tied to the EDUCAUSE approach, which advocates for the coordinated management of HEI changes, including culture, workforce, and technology, informed and guided by academic leadership. Namely, external stakeholders might equate the DT culture with marketing and outreach initiatives, while the internal cultural change could be neglected. Additionally, academic leadership may focus on ICT governance and the formal introduction of digitally enabled teaching. Nevertheless, these processes could be merely prescriptive, with faculty acceptance based on incentives for ICT upskilling and the formal implementation of digital initiatives.

A similar critical evaluation could be repeated across other HEI DM models, which calls for an empirical evaluation of relevant relationships among the key stakeholders’ perceptions, related to the critical DM dimensions, and overall DT success. While we conduct such an analysis in this paper, there is an open question of a referent DM model to be applied in our empirical research. While the described EDUCAUSE model, and its UK equivalent—JISC [14], are informative in the global HE environment, they are still focused on HEIs and their wider environment in the Anglo-Saxon countries.

The regional WB context has been successfully addressed by the Higher Decision research project, implemented by the Faculty of Organization and Informatics in Varaždin. This project, funded by the Croatian Science Foundation, was implemented from 2015 to 2019 and resulted in the development of a comprehensive framework for strategic decision-making in Croatian higher education [15]. The DMFHEI DM framework has been developed within the project, based on the input of regional and national experts, who identified the specific characteristics of the relevant local/regional HEI environment, including the accreditation standards and resource constraints. One of the project outputs is a complex DM self-assessment instrument, consisting of seven dimensions and 43 items [16]. Specifically, DMFHEI covers:

• Leadership, Planning, and Management—the extent to which institutional strategy and decision-making drive and support digital transformation.
• Quality Assurance—integration of digital considerations in quality standards and continuous improvement processes.
• Scientific Research—digitization of research activities, data management, collaboration platforms, and innovation in research practices.
• Technology Transfer and Service to Society—using ICT to support partnerships, community engagement, and the transfer of knowledge/technology to external stakeholders.
• Learning and Teaching—the adoption of digital pedagogies, e-learning tools, and curriculum innovation to enhance teaching and student learning experiences.
• Digital Culture—the development of a culture among staff and students that values digital literacy, openness to innovation, and ethical use of technology.
• Digital Resources and Infrastructure—availability and quality of hardware, software, network infrastructure, digital platforms, technical support, and information security at the institution.

Although developed in a comprehensive methodological process and covering all the relevant DM dimensions, the self-assessment instrument proved to be very complex and demanding in an HEI’s self-assessment process. This issue has been highlighted as one of the main limitations by the authors of the instrument [10]. It reflects a need for a simplified DM assessment tool, which will still cover the national and regional characteristics of the WB region. Our solution, similar to other short DM scales and instruments [17], will be further discussed in the Methods and Materials section. Such a simplified approach has high chances of practical HEI usage and acceptance by higher education administrators and stakeholders as a measurement and benchmarking tool, guiding systematic DT processes.

The four DT components, as identified by the EDUCAUSE model, corresponding to the seven DM areas in the DMFHEI self-assessment tool, are heavily reinforcing each other. The core component relates to the governance of ICT and change management processes, which includes the leadership of the HEI. In the change management literature, strong leadership has been identified as one of the key aspects for ICT-enabled transformation [18], dating back to the early 1990s, when Business Process Reengineering (BPR) was introduced [19]. Lessons from the early attempts of ICT-supported organizational change are relevant even after 30 years have passed. It is not enough to declaratively introduce the ‘digital revolution’ and refer to a ‘clean sheet of paper’. A clear digital strategy and an action plan are necessary to support the transformation, along with strong top leadership support and a precise allocation of organizational change-related tasks and responsibilities [20]. However, formalities are not enough: the leadership vacuum, or inadequate leadership styles, almost inevitably lead to low engagement from both internal and external stakeholders, making it difficult to align the technology effort with the focus on students and the required changes in pedagogical approaches [12,21].

These aspects of successful DT initiatives are directly tied to Quality Assurance (QA) processes, as they need to ensure that ICTs are perceived as a strategic partner to the three traditional HEI processes: academic teaching and learning, scientific research and technology transfer, and community outreach. Unfortunately, this is not the case in the WB region, either in the academic sector [9,22] or in a more general socio-economic context [10,23,24]. Without the administrative and ICT skills, it is difficult, or entirely impossible, to break the traditional functional silos and set up a relevant measurement framework (i.e., determine the Key Performance Indicators—KPIs, track their values, and feed the information to the change leadership team). The same outcome is reached if leadership underestimates the technical aspects of the contemporary HE competitive landscape and remains committed to the traditional patterns of HEI functioning [20,21].

In the context of HEIs’ digital transformation, researchers have primarily focused on academic teaching and learning [25,26], which involve rethinking pedagogy to effectively integrate technology and develop digital content and curricula throughout lifelong education. This implies innovation and flexibility of teaching and learning approaches, including flipped classrooms, Massive Open Online Courses (MOOCs), blended courses, virtual laboratories, gamified learning, and personalized instruction based on Artificial Intelligence (AI). These implications stem from the need to reskill and upskill educators for the digital age [25,26], as well as to enhance the student experience and align it with their increasingly digital lifestyles [27] and learning preferences [28]. This is especially important in the WB context, where academic teaching and learning still rely on traditional methods, and the introduction of student-centric pedagogies remains a challenge [29].

Concerning scientific research and technology transfer, ICTs facilitate international and inter-disciplinary collaborations, which are especially important in the WB region, where research funding is limited and physical research infrastructure remains undeveloped [30]. Funding, provided by the EU’s Widening participation and strengthening the European Research Area (WIDERA) initiative, enables participation in open-science platforms, such as the European Open Science Cloud, open data sharing, and other forms of digital collaboration, providing significant opportunities within the DT context [31]. Virtual forms of collaboration make it easier for HEIs and research institutions from peripheral regions to ‘plug’ into the mainstream research topics and participate in significant partnerships, without waiting for national funding and improvement in physical research infrastructure. This is acknowledged by the current Horizon Europe Work Programme for 2025, in its calls for the Horizon Widening area [32], prioritizing participation and research excellence among peripheral European research institutions and regions.

These apply both to the research and innovation activities of WB HEIs, as the Horizon funding funds different ICT-enabled paths of research excellence and participation in EU-wide scientific collaborations [32]:

• European Excellence Initiative (EEI) can be used to modernize research and development and allow HEI networks to participate in EU innovation ecosystems.
• EIC Pre-accelerator programs fund the ‘deep technology’ programs (i.e., innovation based on advanced scientific discoveries, involving high technical risks and capital needs) at the stage of developing ideas into minimal viable products.
• Hop-on Facility as a specific form of supporting WB organizations to join the existing Horizon Europe partnerships and obtain working experience in international collaboration.
• Action Plans for Connected Regional Innovation Valleys provides funding to partnerships involved in smart specialization in peripheral regions. They are expected to engage regional HEIs, companies, and governments in the research and innovation goals defined by the smart specialization strategies. The funded actions involve building shared digital research and innovation infrastructure, technology transfer offices, and entrepreneurship support systems [33].

Therefore, regional HEIs can use EU funding for projects, building upon the existing DT initiatives, to integrate into mainstream research and overcome local and regional infrastructural constraints.

The digital infrastructure, including hardware, networking equipment, software, and ICT know-how, is one of the Critical Success Factors (CSFs) for the implementation of DT in HEIs [34]. This CSF needs to be assessed from the viewpoint of multiple stakeholders, with diverse computing skills and needs, as demonstrated by a recent case study [35]. While students emphasize the need for a variety of teaching and learning tools (including Learning Management Systems and innovative solutions, such as Artificial Intelligence-based academic chatbots, etc.), researchers will be interested in full-text scientific databases and data sources. In contrast, administrators will need access to data analysis and decision-making tools. In any case, the integration and interoperability of new and legacy systems are needed, along with robust information security to protect the digitalized operations of the HEI.

An additional critical CSF, integrating the technological, organizational, and human aspects of DT in HEIs, is the organizational culture, which describes the shared values, attitudes, and behaviors [36]. A successful DT effort requires an adaptive and innovative culture, scoring high levels of digital maturity. Its outcomes need to be visible, especially in terms of preferences for working in interdisciplinary teams, celebrating faculty attempts to try (and, potentially, fail) new tools and approaches, and investing in staff training and development [37]. This is especially significant for HEIs in the WB region, which have traditionally been administered in a bureaucratic manner, with significant influence from political elites.

Therefore, to avoid the ‘NIMBY’ (Not-In-My-Backyard) syndrome [38], HEI administrators need to demonstrate commitment to staff psychological safety [39] and evidence-based decision-making, in order to ensure the perception of continuous top management support to the DT initiative [40].

DT requires an understanding of the interaction among its critical components in terms of an ecosystem. The four interlocking dimensions (infrastructure, pedagogy, leadership/governance, and culture) need to be reinforcing and mutually supporting. Such a proposition is supported by recent empirical evidence: a recent study shows that investments in digital infrastructure translate into higher student and staff satisfaction when paired with digital literacy program strategies and evidence-based ICT governance [41]. There is also empirical evidence of a ‘virtuous circle’ of success in successive rounds of digital transformation, as technological capacity, modern pedagogy, strategic leadership, and innovative culture feed each other [7].

3. Materials and Methods

In March–April 2025 we circulated an online questionnaire (using the free, open-source LimeSurvey data collection system, hosted by the Croatian Academic Computing Center SRCE, located in Zagreb, Croatia) to academic staff (assistants and faculty), administrative staff and students, as the main stakeholders of the digital transformation efforts, of the three public HEIs in the WB region. Those included: University of Applied Sciences “Marko Marulić” (UASMMK), located in Knin, Croatia; University North (UN), located in Varaždin, Croatia, and the Faculty of Economics, University of Mostar (EF SUM), located in Mostar, Bosnia and Herzegovina.

We used the public email directories of staff at the three involved HEIs to reach both faculty and the administrative staff and distribute the survey link by email. A link to the electronic survey has been distributed to students by prominently placing it on the Learning Management System (LMS) pages. The survey included information about the research project and a consent form, placed on the introductory page of the online questionnaire. Participation was voluntary and anonymous, and no personally identifiable data were collected. Full text of the survey instrument is available in the Supplementary Materials.

After listwise deletion of incomplete cases, we collected 137 questionnaires (71% from students and 29% from HEI staff). There were 75.2% female and 16.1% male respondents, while 8.8% of participants did not provide an answer. We acknowledge the over-representation of female respondents in the sample, although this is a relatively common limitation in social science surveys in the WB region [42].

The majority of respondents (97 participants, i.e., 70.8%) were students, 14 participants (10.2%) were academic staff, 12 participants (8.8%) were administrative staff, and 3 (2.2%) were teaching and research assistants. While 11 participants did not respond to the question about their academic status, we retained their answers, as the survey had been made available only to students and both academic and administrative staff at the three analyzed HEIs. We shall further analyze if there are significant statistical differences in received responses across the consolidated respondent categories, according to their academic status.

Average respondent age was 29.7 years: in the student category, the average age was 25.6 years, which could be expected, as we surveyed students across the three Bologna program cycles (Bachelor, Master, and PhD studies). The average age of junior staff (teaching and academic assistants) was 29 years, while the faculty members had an average age of 48.9 years. The average age of the surveyed administrative staff equaled 40.1 years.

Two research constructs were used in the study and measured using the following scales:

• Digital maturity (DM): Twenty-one items were adapted from the HigherDecision DMFHEI. Respondents rated agreement on a standard 5-point Likert scale, with values ranging from 1 (strongly disagree) to 5 (strongly agree). The original DMFHEI instrument [16] is an extensive questionnaire, consisting of seven dimensions and as many as 43 items. The inclusion of the original DMFHEI items into research instruments or practitioner surveys for DM assessment could lead to overwhelming respondents and higher non-response rates. Our short version maximizes content coverage, with the high internal consistency of the instrument confirmed by the Cronbach α value of 0.966.
• Perceived digital transformation (DT) success: Seven items measured the extent and success to which digital technologies had improved academic teaching and learning, research, administration, and stakeholder engagement (α = 0.911).

Primary data were downloaded from the LimeSurvey system and are available in the Supporting Materials to this paper. All statistical analyses were conducted using IBM SPSS 28.

4. Results

4.1. Dimensional Profile of Digital Maturity

Table 1. Means and standard deviations for DM (including DMFHEI dimensions) and DT constructs.

	N	Min.	Max.	Mean	Std. Dev.
Digital Maturity: Leadership, Planning, and Management	135	1.33	5.00	3.4988	0.84335
Digital Maturity: Quality Assurance	134	1.50	5.00	3.5709	0.88248
Digital Maturity: Scientific Research	137	1.00	5.00	3.7506	0.88772
Digital Maturity: Technology Transfer and Social Impact	135	1.00	5.00	3.5691	0.87327
Digital Maturity: Teaching and Learning	134	1.00	5.00	3.8657	0.85559
Digital Maturity: Digital Culture	134	2.00	5.00	3.7463	0.79701
Digital Maturity: Digital Resources and Infrastructure	134	1.75	5.00	3.7805	0.79160
Digital Maturity: Overall	137	2.00	5.00	3.6739	0.72096
Digital Transformation	137	1.89	5.00	3.8560	0.75649
Valid N (listwise)	133

shows that, across the digital maturity (DM) dimensions, Teaching and Learning (M = 3.87, SD = 0.86) and Digital Infrastructure (M = 3.78, SD = 0.79) are the highest rated. Simultaneously, Technology Transfer and Social Impact have the lowest perceived digital maturity (M = 3.57, SD = 0.87). The composite maturity index averages 3.67 (SD = 0.72), while the composite indicator of perceived digital transformation (DT) equals 3.86 (SD = 0.76). These descriptive patterns suggest that teaching and learning practice and digital infrastructure are the most significant drivers of DT from the surveyed stakeholders’ perspective.

DM dimensional profiles are visualized in Figure 1, which shows the means and the 95% confidence intervals (CIs) for the seven analyzed dimensions.

4.2. Comparisons Across Stakeholder Groups

Descriptive statistics for DT and DM constructs, for respondents grouped according to their academic status, are shown in

Table 2. Descriptive statistics for DT and DM by academic status.

		N	Mean	Std. Dev.	Std. Error	Min.	Max.
Digital Maturity: Overall	Students	97	3.6417	0.70886	0.07197	2.00	5.00
	Assistants	3	3.5159	0.66169	0.38203	3.00	4.26
	Faculty (HEI Instructors)	14	3.8070	0.77368	0.20677	2.30	4.95
	Administrative Staff	12	4.0010	0.65694	0.18964	2.77	4.90
	Total	126	3.6913	0.71169	0.06340	2.00	5.00
Digital transformation	Students	97	3.8195	0.72581	0.07370	1.89	5.00
	Assistants	3	3.5556	0.50918	0.29397	3.00	4.00
	Faculty (HEI Instructors)	14	4.1190	0.70470	0.18834	2.78	5.00
	Administrative Staff	12	4.1667	0.75582	0.21819	2.56	5.00
	Total	126	3.8796	0.72822	0.06488	1.89	5.00

. Normality checks (visual inspection of Q–Q plots; Kolmogorov–Smirnov and Shapiro–Wilk tests) indicated approximate normal distributions. Levene’s tests (see

Table 3. Levene’s test.

	Levene Statistic	df1	df2	N	Sig.
Digital Maturity: Overall	Based on Mean	0.205	3	122	0.893
	Based on Median	0.283	3	122	0.838
Digital transformation	Based on Mean	0.319	3	122	0.812
	Based on Median	0.402	3	122	0.752

) confirmed homogeneity of variances (all p > 0.05). One-way ANOVAs (see

Table 4. ANOVA analysis of group effects.

		Sum of Squares	df	Mean Square	F	Sig.
Digital Maturity: Overall	Between Groups	1.669	3	0.556	1.101	0.352
	Within Groups	61.643	122	0.505
	Total	63.312	125
Digital transformation	Between Groups	2.457	3	0.819	1.565	0.201
	Within Groups	63.831	122	0.523
	Total	66.288	125

) found no statistically detectable significant differences (given sample sizes) across status groups (students, assistants, faculty, and administrative staff) for either DM (F(3,122) = 1.101, p = 0.352) or DT (F(3,122) = 1.565, p = 0.201).

Therefore, pooling responses across the analyzed categories is considered acceptable for further correlation and regression analysis, which concerns RQ1.

4.3. Correlation Analysis

Pearson correlations between DT and each maturity dimension, as well as the composite maturity indicator (see

Table 5. Pearson correlation matrix.

	DM1	DM2	DM3	DM4	DM5	DM6	DM7	DM_ALL	DT
Digital Maturity: Leadership, Planning and Management (DM1)	1	0.750 **	0.655 **	0.745 **	0.682 **	0.692 **	0.739 **	0.882 **	0.750 **
Digital Maturity: Quality Assurance (DM2)			0.632 **	0.721 **	0.556 **	0.652 **	0.731 **	0.847 **	0.699 **
Digital Maturity: Scientific Research (DM3)			1	0.571 **	0.531 **	0.547 **	0.660 **	0.778 **	0.913 **
Digital Maturity: Technology Transfer and Social Impact (DM4)				1	0.676 **	0.743 **	0.755 **	0.872 **	0.685 **
Digital Maturity: Teaching and Learning (DM5)					1	0.706 **	0.724 **	0.817 **	0.669 **
Digital Maturity: Digital Culture (DM6)						1	0.835 **	0.863 **	0.680 **
Digital Maturity: Digital Resources and Infrastructure (DM7)							1	0.910 **	0.794 **
Digital Maturity: Overall (DM_ALL)								1	0.873 **
Digital Transformation (DT)									1

Note. ** Correlation is significant at the 0.01 level (2-tailed).

), were positive and statistically significant (p < 0.001). Therefore, it can be concluded that stakeholders’ perceptions of transformation align most closely with overall maturity and, within that, with the multiple DM dimensions.

4.4. Single-Predictor Regression Model Predicting Perceived Digital Transformation

An OLS regression with DM as the sole predictor (see

Table 6. Regression summary.

Model	R	R Square	Adjusted R Square	Std. Error of the Estimate	Durbin-Watson
1	0.873 a	0.761	0.760	0.37087	1.881

^a Predictors: (constant), digital maturity: overall. Dependent variable: digital transformation.

) explained 76.1% of the variance in DT (Adj. R² = 0.761).

The predicted linear regression model (see

Table 7. ANOVA for the single-predictor OLS regression model.

	Model	Sum of Squares	df	Mean Square	F	Sig.
1	Regression	59.262	1	59.262	430.861	0.000
	Residual	18.568	135	0.138
	Total	77.830	136

) was statistically significant (F(1,135) = 430.861; p < 0.001). The standardized coefficient was also significant and positive (β = 0.873, p < 0.001), demonstrating that overall digital maturity strongly predicts perceived digital transformation (see

Table 8. Evaluation of regression coefficients.

Model		Unstandardized Coefficients		Standardized Coefficients	t	Sig.	Correlations			Collinearity Statistics
		B	Std. Error	Beta			Zero-order	Partial	Part	Tolerance	VIF
1	(Constant)	0.492	0.165		2.980	0.003
	Digital Maturity (Overall)	0.916	0.044	0.873	20.757	0.000	0.873	0.873	0.873	1.000	1.000

and Figure 2). Model diagnostics indicated an acceptable residual structure (SE = 0.371; Durbin–Watson = 1.881).

4.5. Two-Predictor (Teaching/Learning and Infrastructure) Regression Model Predicting Perceived Digital Transformation

To assess the relative contributions of the highest-scoring domains, in the second OLS model, we regressed DT on Digital Infrastructure and Teaching and Learning dimensions of DM. The model was statistically significant, F(2,131) = 120.824; p < 0.001, and explained 64.8% of the variance (Adj. R² = 0.643), without autocorrelation of residuals (Durbin-Watson = 2.045), and a standard error of estimate of 0.45 dependent variable units. (see

Table 9. Two-predictor regression model summary.

Model	R	R Square	Adjusted R Square	Std. Error of the Estimate	Durbin-Watson
2	0.805 a	0.648	0.643	0.44987	2.045

^a Predictors: (constant), digital maturity: teaching and learning and digital maturity: digital resources and infrastructure. Dependent variable: digital transformation.

and

Table 10. ANOVA for the two-predictor OLS regression model.

	Model	Sum of Squares	df	Mean Square	F	Sig.
2	Regression	48.905	2	24.453	120.824	0.000
	Residual	26.512	131	0.202
	Total	75.417	133

). Together, these two dimensions alone account for nearly two-thirds of perceived transformation variance, thereby responding directly to RQ1.

To illustrate the pairwise relationships among predictors and the perceived level of DT, we visualized them using the scatterplot matrix (see Figure 3). Visual inspection of the scatterplots confirms positive relationships between predictors and the DT, with the Digital Infrastructure providing a higher predictive power of DT than Teaching and Learning.

While both coefficients are statistically significant, standardized coefficients confirm that the Digital Infrastructure (B = 0.650, p < 0.01) has a larger effect than Teaching and Learning (B = 0.198, p < 0.01). Collinearity diagnostics were acceptable (Tolerance = 0.476; VIF = 2.102 for both predictors) (see

Table 11. Two-predictor evaluation of regression coefficients.

Model		Unstandardized Coefficients		Standardized Coefficients	t	Sig.	Correlations			Collinearity Statistics
		B	Std. Error	Beta			Zero-order	Partial	Part	Tolerance	VIF
2	(Constant)	0.862	0.200		4.320	0.000
	Digital Maturity: Teaching and Learning	0.174	0.066	0.198	2.634	0.009	0.669	0.224	0.136	0.476	2.102
	Digital Maturity: Digital Resources and Infrastructure	0.619	0.071	0.650	8.661	0.000	0.794	0.603	0.449	0.476	2.102

).

4.6. Summary Concerning the Research Questions

Results of the empirical research can be summarized as follows:

• RQ1 (key maturity dimensions). Overall, digital maturity is a robust predictor of perceived transformation. Within this composite indicator, Digital Infrastructure and Teaching and Learning account for the most significant amount of variance in DT, with infrastructure showing the stronger independent contribution.
• Dominance of the Digital Infrastructure, as a DT predictor, is consistent with the socio-economic realities of the Western Balkans region, where the majority of HEIs are supposed to find themselves on the lower end of the DM scale. A perceived lack of fundamental ICT resources, necessary to ensure the stable and interoperable functioning of the underlying infrastructure, still appears to play a major role in stakeholders’ perceptions. In addition, the Teaching and Learning dimension shapes the perception of students, as a dominant stakeholder group, and serves as a major driver of the lived experiences of the benefits promised by the DT concept.
• As HEIs progress up the maturity ladder, infrastructure issues are expected to be addressed, and the focus should shift to governance, quality assurance (QA), and digital culture. Those dimensions of the DM construct are more complex and determine the effectiveness of the technological platforms and actual activities (including research, academic teaching and learning, and outreach toward the economy and the community). Therefore, one of the future research tasks is to verify if the focus of digitally maturing HEIs shifts from infrastructure and digitally enabled teaching to governance, QA, and digital culture, to achieve the high levels of impact in research and development, technology transfer, and social outreach.
• RQ2 (transferability to smaller HEIs). Given the effect pattern, institutions seeking fast, visible progress should focus on: (i) major infrastructure upgrades (reliable networks, interoperable platforms, cybersecurity) and (ii) digital pedagogy practices (LMS-centered course design, using learning analytics, based on LMS data, supporting the responsible usage of AI learning tools, etc.). These priorities align with the DM dimensional profile and the two-predictor model. They also provide a practical starting point for DT in HEIs with limited resources. In the following section, we discuss the opportunities for benchmarking and transfer of good practices across a population of smaller HEIs in the WB region.

5. Discussion

In the extant literature, digital maturity has been adopted as a predictor of perceived DT scope and success. Our empirical results support these previous findings, as the composite measure of digital maturity in our study explains approximately 76% of the variance in stakeholders’ perceived scope and success of digital transformation (DT). However, there are only two digital maturity dimensions that are statistically significant for predicting the perceived DT. Those are: digital infrastructure and academic teaching and learning maturity. Both DM aspects are theoretically supported as significant enablers of HEI transformation [1,12]. Digital infrastructure provides the technological foundations for DT, while digitally enabled teaching and learning represent the activities of primary interest for learners.

The perceived importance of both DM dimensions is further emphasized by the direct exposure of students and HEI staff to digital teaching and learning tools and platforms, including Learning Management Systems (LMSs) and novel tools, such as AI chatbots. As already demonstrated by empirical studies [35,43], the everyday use of these tools, along with the perceived contribution of digital pedagogy to student outcomes, raises confidence and positive evaluations of DT efforts and fosters a positive perception of the organizational transformation efforts. In addition, digitally enabled teaching and learning, and digital infrastructure development are the most tangible aspects of a digitally mature HEI. While their ‘public exposure’ raises attention and helps provide tangible support to the DT efforts, other aspects of DM are interconnected.

Mabić et al. [44] find that digital skills and resources contribute to HEI DT, which can be explained by further emphasizing usage experiences and strengthening the overall perception among stakeholders that the transformation is taking place.

Other aspects of DM, such as ICT and change governance, along with the top management support, are needed to sustain the strategic direction and funding of the transformation effort [44]. At the same time, the digital culture helps avoid resistance to change and encourages experimentation [21]. Once the transformation is ‘frozen’ into a new organizational form (according to Lewin’s model of organizational change [45]), quality insurance needs to provide information on opportunities for improvement, as well as impetus for additional digital innovation [11].

The ‘virtuous circle’ of mutually reinforcing relationships within the process of improving the HEI DM contributes to perceived quality via a closed-loop system of data-based quality improvement. As a digitally mature HEI identifies and addresses the actual quality problems, maturity becomes a path to higher quality outcomes, rather than an end in itself.

The entire dynamics of the interactions among the DM dimensions and their contribution to the perceived DT are illustrated by the case study of the University of Mostar (Sveučilište u Mostaru—SUM), which can be described as one of the leading DT institutions in the WB region. SUM has developed a DT strategy [46], officially adopted in April 2023. The strategy follows an integrated approach, where robust digital infrastructure and digital skills are expected to improve the user experience and foster all aspects of university functioning (teaching and learning, research, and innovation, as well as social development). In this case, the theoretical argument of mutually reinforcing digital infrastructure, culture, and leadership (governance) [47] has been followed. SUM Information Technology department (SUMIT) has built a robust, campus-wide infrastructure and established a new digital innovation hub [48], which coincided with initiatives of DT thought leadership and knowledge-sharing across the WB region. Currently, the SUM conference MoStart is regionally recognized as a significant forum for regional researchers and practitioners interested in the digital transformation of higher education and the application of digital and AI tools and platforms in teaching and learning processes [49]. By providing digital services for schools in Bosnia and Herzegovina and participating in public administration projects, SUM extends its DT efforts to the social environment and fulfills its ‘third mission’.

In addition, the SUM DT efforts are aligned with the OECD’s recommendations for EU member states in the WB region [9], which call for strengthening competencies in the digital delivery of academic teaching and learning. In this context, SUM has surpassed many of the regional HEIs, as it delivers many of its academic programs either entirely online or in a hybrid mode. In addition, SUMIT has addressed the issue of inadequate digital infrastructure and now reaches out to its social environment by providing cloud infrastructure and digital skills to educational and government institutions in Bosnia and Herzegovina. Simultaneously, SUM reskills and upskills its faculty and administrative staff, which aligns with recommendations from the regional and European SME sector, where staff development was evaluated as equally important as digital infrastructure development [50].

Concerning RQ2, i.e., transferability of the SUM digital experiences can be achieved by using:

• Regional academic networks and mobility/twinning initiatives, sharing the technological solutions, as well as the insights concerning ICT/change governance and staff training;
• Offering microcredentials in relevant digital skills for HEI stakeholders;
• Using EU funding opportunities (focusing on Horizon Widening “Hop-On” facilities and Regional Innovation Valley initiatives) to fund the infrastructure upgrades, develop the research and development, as well as the technology transfer capacity;
• Regularly employing the short DMFHEI instrument, developed in this study, as a benchmarking tool, both to track one’s own progress and to compare it to regional counterparts.

Therefore, the contribution of this study is both empirical and theoretical. From an empirical viewpoint, the research on DT and DM in the WB region offers new insights, as the region appears to be not only less developed but also empirically underrepresented in the existing literature [9,22]. In addition, our research verifies a short version of the DM instrument [16], which is suitable for stakeholder assessment of digital maturity and its benchmarking.

The practical value of our study lies in its contribution to understanding stakeholders’ DM priorities and the priority assigned to different DT interventions. Without such empirical grounding, many different routes to HEI DT could be pursued, although they might be misaligned with stakeholders’ needs and interests. As demonstrated in the Results section, WB HEI stakeholders focus on digital infrastructure and digitalization of academic teaching and learning, which should be treated as “first-level” enablers of digital transformation.

6. Conclusions

In this paper, we empirically evaluate the contribution of digital maturity (DM) and its dimensions to implementing transformational change, specifically digital transformation (DT), in smaller higher education institutions (HEIs). Our results are supported by the extant HEI digital transformation frameworks [4,14] and empirical results [12,43]. They are also supported by the regional Western Balkans (WB) benchmarks [9,10,23], although it should be noted that the WB stakeholders’ perceptions are shaped and influenced by significantly lower levels of digital transformation and culture.

Although the study is limited by its regional scope and sampling, it provides a relevant theoretical contribution:

• We validated a short version of the DM assessment instrument derived from a comprehensive DMFHEI framework.
• The most significant DM dimensions were identified, specifically concerning their contribution to digital transformation in the WB context.
• The WB empirical evidence is positioned within the extant European and global literature.

From the practical viewpoint, our study shows that DM serves both as a roadmap and a tool for implementing digital transformation, where digital teaching and learning, and digital infrastructure can be used to showcase quick wins and ensure stakeholder support throughout an HEI’s DT journey. The quick wins are especially significant for smaller regional HEIs with limited resources, which need to showcase the effectiveness of DT interventions and sustain the motivation for further organizational changes.

Although these imperatives are clearly seen from our empirical results, they still need to be positioned within a comprehensive DT strategy, coordinating these “first-level” enablers with the “second-level” ones. Those are more complex dimensions of DM, including ICT Governance and Digital Culture, which are becoming further emphasized as an institution reaches higher levels of digital maturity.

Additional practical implications and recommendations for smaller HEIs in the WB region, based on the experience of the University of Mostar (Sveučilište u Mostaru—SUM), located in Bosnia and Herzegovina, could be identified as follows:

• DT efforts need to rely on an integrated strategy, supported by the overall governance effort, which ensures that investments in digital infrastructure are followed by prominent (and celebrated) “quick wins”, focused on students’ learning outcomes and improved research and innovation impact.
• Staff up-skilling and reskilling need to be formally introduced and acknowledged, along with each major infrastructural project. Comprehensive digital skills, including AI literacy, digital pedagogy, and data-driven decision-making, can be developed through micro-credential programs, following the successful practices of regional adult education [51].
• Digital culture, focused on flexibility and the freedom to experiment with digital tools and platforms, should support the overall innovation efforts, provide public visibility of the DT’s ‘quick wins’, and celebrate the examples of good practices and the most dedicated individuals and departments.

This study has the following limitations, which should be addressed by future research:

• Our results are based on limited data from three HEIs from the Western Balkans region, with a somewhat unbalanced stakeholder distribution (with approximately 70% of student participants). Measures are self-reported, and the research design is cross-sectional, which may further limit the generalizability of the results. However, since this is a preliminary study, the current data serve to identify general trends and inform future research priorities. In addition, we verify the applicability of a short-version DM scale, based on the DMFHEI framework, for use in future research and benchmarking across the WB region.
• Future research should focus on full DMFHEI replication in multiple WB countries, as well as in other regions, using both stakeholder perceptions, based on a more comprehensive sample, and the actual administrative data and user logs (since they demonstrate the use of HEI digital tools and platforms).
• Replication should aim to reach beyond our initial intention, which focused on designing a short DMFHEI-compliant instrument, ready for empirical analysis and rapid benchmarking of HEIs in the regional setting. If appropriately tested and applied in different regions, replication might lead to insights related to the generalizability of initial results and the dynamics of the digital maturity in maturing HEIs. Namely, the proposed focus-shifting among different dimensions of DM, as an HEI matures, could be tested by using simple mediation or moderation quantitative research designs.

Our empirical results and practical recommendations, including the verified short-form DM assessment instrument, should provide both a theoretical DM roadmap and a benchmarking tool to regional HEIs, aiming to achieve digital transformation.