Exploring the nexus between digital competencies and digital citizenship of higher education students: a PLS-SEM approach

Abstract

In the ever-evolving digital landscape, higher education (HE) students find themselves at the crossroads of two critical aspects: their digital competencies and their digital citizenship. Given the ubiquitous integration of digital technologies in educational settings, it is essential to investigate the relationship between students’ digital competencies and their ability to engage as responsible digital citizens. This study examined the nexus between digital competencies and citizenship among HE students. This study involved 247 students from a Ghanaian HE institution. Digital competency and citizenship scales were used as data collection instruments. The findings revealed that problem-solving, communication and collaboration, and digital content creation competencies had a significant positive influence on HE students’ digital citizenship. Conversely, safety competence has no significant influence on digital citizenship. Interestingly, data and information literacy competence have a negative effect on digital citizenship. The implications of these findings provide valuable insights for educational institutions, policymakers, and educators to foster digital competencies and promote responsible digital citizenship among HE students.

Keywords:

• Communication and collaboration
• data and information literacy
• digital competencies
• digital citizenship
• digital content creation
• higher education students

REVIEWING EDITOR:

• Debra Laier Chapman, School of Computing, University of South Alabama, Mobile, AL, USA

SUBJECTS:

• Communication Ethics
• ICT
• Information Technology
• Education - Social Sciences

Introduction

The recent expansion of digital infrastructure, coupled with the COVID-19 pandemic, has precipitated digital technology, leading to increased Internet use globally (Amankwah-Amoah et al., 2021). Nevertheless, the pervasive use of the internet and technology has a profound impact on citizens’ civic participation (McCarthy et al., 2023; Shi et al., 2023). Embedded in society and everyday activities, digital technology has changed citizens’ interactions with social, political, economic, and environmental factors (Hintz et al., 2019; McCarthy et al., 2023).

Consequently, digital citizenship has become an essential dimension of responsible citizenship (Chen et al., 2021; Shi et al., 2023). Digital citizenship refers to the responsible, ethical, and mindful use of technology and the internet (Kocoglu et al., 2023). Other studies (Kara, 2018; Kim & Choi, 2018) define digital citizenship as norms, behaviours and ethical choices with regard to the use of technology. In an era in which digital technology is rapidly shaping the world, digital citizenship is of paramount importance.

Many countries have recently developed national and international policies to improve and support digital citizenship in the context of higher education (HE). Higher education institutions (HEIs) such as universities have put much effort into using Information and Computer Technology (ICT). Undoubtedly, Ghana, like many other nations, is experiencing the transformative impact of the digital age. However, the surge in Internet usage coupled with a lack of proper guidance raises concerns about online privacy, security and responsible digital behaviour.

As studies have repeatedly emphasised, students engaged in random AI systems often lack the ethical use of these systems, which poses a significant risk of harmful behaviour while attempting to complete a task (Al-Husseiny et al., 2023; Sanusi et al., 2022; Siddiq et al., 2023). In particular, HEIs are at a crossroads where digital citizenship has become an essential life skill because of the efforts to produce skilled professionals in various fields. Therefore, students need to be well versed in ethical online practices, digital professionalism and responsible social media engagement because of their frequent use of digital platforms.

As Mossberger et al. (2007) noted, digital citizenship is very important because of the following: (1) the dual nature of the Internet, which offers both prospects and challenges to social interactions; (2) the need to empower individuals to actively participate in global society through Internet access, regardless of their location; (3) the need to enhance democratic processes by leveraging online activities; (4) raising responsible and competent digital technology users; and (5) efforts to enhance equal opportunities for learning and teaching through Internet access to positively impact educational dynamics.

Nevertheless, it is worth noting that digital citizenship is characterised by digital literacy (Imer & Kaya, 2020; Kaya & Imer, 2020), which refers to the capacity of any technology user to comprehend and utilise media and information. This emphasises the need to develop digital competencies, as noted in recent studies (Örtegren, 2023; Vuorikari et al., 2022). Undoubtedly, Ribble (2012, 2015) and Ribble and Bailey (2007) have significantly contributed to the discourse on digital citizenship by emphasising that digital competencies are required to fulfil the rights and responsibilities of the digital world.

Digital competence in education encompasses a repertoire of skills, attitudes and capabilities for critically and creatively utilising technologies in both personal and professional contexts (Ala-Mutka, 2011; Aristeidou & Herodotou, 2020; Kim & Choi, 2018; Mattar et al., 2022; Rodríguez-García et al., 2022; Zhao, et al., 2021). Slavković et al. (2023) concur that digital competency is a set of abilities necessary for individuals to effectively conduct activities in a digital environment. According to Awaludin et al. (2022), there is a need for HE students’ digital literacy to enhance their ability to use digital technologies effectively in the learning-teaching process.

Effective use of digital technologies for learning, work and social participation necessitates a blend of digital competencies. The literature suggests that digital competencies encompass information and data literacy, communication and collaboration, digital content creation (including programming), safety (including well-being and cyber security) and problem-solving (i.e. critical thinking and understanding of intellectual property rights; Vuorikari et al., 2022). These competencies are of significant value to policymakers in the development of educational programs aimed at comprehending digital competencies and initiatives to foster and enhance digital citizenship (Hidayat et al., 2023; Zhao et al., 2021).

Following the inspiration of Hidayat et al. (2023) and Zhao et al. (2021), these dimensions of digital competencies from the perspective of HE students in Ghana can help contribute to the development of effective digital citizenship strategies. However, to the best of our knowledge, the relationship between digital competency and citizenship in Ghana has not been thoroughly examined using the interplay of these key dimensions. This notable research gap may hinder a nuanced understanding of the potential disparities in digital citizenship and the specific requirements of HE students.

In this context, it is imperative to investigate Ghanaian HE students’ digital competencies as digital citizens, specifically their perceptions of data and information literacy, communication and collaboration, digital content creation, safety and problem-solving, as these are essential dimensions of digital competencies for nurturing students’ digital citizenship. To bridge this gap, the present study aligns with the dataset from a recent study by Hidayat et al. (2023) to examine the nexus between digital competencies and citizenship among Ghanaian HE students.

Following Hidayat et al. (2023), this study uses the Digital Competence Framework for Citizens (DigComp), focusing on five key dimensions: (1) data and information literacy, (2) communication and collaboration, (3) digital content creation, (4) safety and (5) problem-solving. We believe that, for instance, data and information literacy as key attributes of digital competences should be a skill every student, not only for ICT scientists, as noted by Twidale et al. (2013). This is especially important for HE students to be able to navigate the digital world driven by data and develop the skills necessary to analyse, interpret and ethically use information.

To elaborate further, this study uses structural equation modelling to provide novel and contextually relevant insights into the intricate dynamics between digital competencies and digital citizenship while taking into account the moderating effect of gender among HE students. This would help to inform proper educational strategies and policies tailored to enhance digital citizenship and foster inclusivity and equity in the digital realm.

The subsequent section delineates the literature and develops our hypotheses. The third section describes the study’s research methodology. Subsequently, the findings are presented in the fourth section, while a discussion of the results that emerged from the fieldwork is presented in the fifth section. In the penultimate section, deductions are drawn and recommendations are made based on the findings of the study. Finally, this paper presents limitations and suggestions for future endeavours in the domain of digital competencies and citizenship in Ghana and beyond.

Literature review and hypotheses development

Digital technology refers to any electronic device, such as computers and cell phones that facilitates communication and interaction. HEIs have always been a strategic domain in which innovative technologies can be applied to increase the effectiveness of teaching and learning for quality education outcomes (Zhao et al., 2021). Due to recent advancements in the digital infrastructures coupled with the reflections from the COVID-19 destructions of the traditional teaching and learning methods, HEIs have experienced surge usage in this field.

The exponential growth of digital technologies has led to significant transformations, allowing students, in particular, to participate online as digital citizens in ways that were previously impossible (Kammer et al., 2021; Zhao et al., 2021). However, research reveals that while the level of digital infrastructure has increased, society has grappled with rapid digital transformation, creating a digital citizenship gap (Murawski & Bick, 2017; Murawski et al., 2020).

It is posited that digital citizenship entails the obligation to facilitate students in cultivating positive digital footprints during the creation and dissemination of online content through social media (Brown, 2023). Brown (2023) delineated digital citizenship as a societal nomenclature that signifies judicious and appropriate utilisation of technology. However, Ribble and Park (2019) proposed an encompassing definition of digital citizenship as an umbrella that addresses a broad spectrum of ethical concerns and opportunities associated with living in a digitally engaged society.

The rise of the Internet and ubiquity of social media and online platforms make it imperative to educate students about responsible online conduct. This highlights the importance of emphasising digital citizenship, especially digital competence, to equip people to traverse the digital landscape while they are engaged ethically. Digital competence is very critical because it is projected that by 2025, 85 million jobs will be displaced due to advances in technology, while 97 million new positions will emerge, signifying a shift in the distribution of work among humans, machines and algorithms (Sánchez-Canut et al., 2023).

This means those with high digital competencies would tend to engage in active and conscious consumption, whereas those with lower skills are prone to passive consumption and fall behind in technological advancement (Budai et al., 2023). That is why, to keep up with these digital transformation challenges, students need to increase their access to technology and raise their technological capabilities to acquire and enhance their digital competence. That highlights the burgeoning interest of digital competence. Given the interconnected nature of digital skills, tool utilisation and their consequences, it is essential to improve digital skills.

Thus, the digital success of future citizens depends on their competencies. To successfully navigate the dynamic and evolving job market, individuals need to acquire new knowledge as well as acquire the necessary digital competencies that would enable them to adapt to new positions and enhance their current roles in the context of digital transformation (Ellingrud et al., 2020). As the digital environment evolves, the skills necessary to thrive in it also change and encompass the understanding of the connection between digital competencies and digital citizenship.

The different current needs presented above demonstrate the essence of guidance on which abilities and skills need to be considered within HE digital competence (Budai et al., 2023). That is why in this study, we considered it important to give attention to the assessment and development of students’ digital skills in the context of Ghana. Based on the lessons learned from this review, we have decided to situate this study within the European Framework of Digital Skills for Citizens, known as DigComp.

The DigComp offers common guidance on what it means for citizens to have adequate digital skills (Carretero et al., 2017; Hidayat et al., 2023; Vuorikari et al., 2022). Thus, the DigComp framework integrates an essential digital competencies dimension such as: (1) data and information literacy, (2) communication and collaboration, (3) digital content creation, (4) safety and (5) problem-solving. Based on the lessons learned from this review, we have decided to situate this study within the DigComp framework.

The aim is to analyse how the diverse existing constructs within the DigComp framework to describe and measure digital competence relates to digital citizenship from the perspective of Ghanaian HE students. This can help to better offer comprehensive assessment of digital competencies and facilitate in predicting their digital citizenship. This next subsection of the study presents details of these digital competencies dimension and focuses on the hypotheses that were formulated to guide the study.

Problem-solving competence and digital citizenship

Our world becomes increasingly digital, embedded in the way we live, work, learn and participate in public life. However, merely awareness of technology and possessing knowledge of its functioning do not automatically guarantee the ability to utilise it responsibly (Bombardelli, 2021; Soares & Lopes, 2020). In particular, the notion that individuals born and raised in a highly technological environment are inherently digital natives is not always correct (Bennett et al., 2009). In this respect, problem-solving competence defined as encompassing the ability to tackle real-world problems and transfer problem-solving strategies from domain-specific to domain-general contexts is of paramount importance (Al-Hassawi et al., 2020; Scherer & Beckmann, 2014).

Digital citizenship has taken on great importance in modern educational settings, both inside and outside the classroom. High problem-solving competence among HE students is crucial for community and future success (Mahanal et al., 2022). Aligns with the rise in digital devices, with a shift from physical school classrooms to virtual ones, as well as an emphasis on digital citizenship and lifelong learning (Bombardelli, 2021), we contend that HE students’ problem-solving skills have become indispensable. When students possess the ability to address problems using digital technology, they are better equipped to assume their role as digital citizens. Consequently, the following hypothesis was proposed:

H1: Problem-solving competence will positively influence higher education students’ digital citizenship.

Data and information literacy competence and digital citizenship

Undoubtedly, the rapid diffusion of digital technologies has put pressure on users, highlighting the importance of data and information competence in articulating information needs, locating and retrieving data, and determining the relevance and content of a source (Vuorikari et al., 2022). In today’s data-rich environment, discernment is especially important, as such the ability to analyse, interpret and critically evaluate the substance of acquired data, information and digital content is crucial for nuanced comprehension and application of knowledge. Data and information literacy are recognised as essential skills for the 21st century, as it is now as important to collect and share relevant information as it was to read and write earlier (Katz, 2007; Simić et al., 2020). Similarly, the promotion of data and information literacy among students is paramount in the context of digital citizenship. However, the central question that arises is whether achieving this objective requires the development of data and information literacy competence. Undoubtedly, positive step towards enhancing students’ digital citizenship involves equipping them with the necessary data and information literacy skills to effectively navigate and utilise contemporary digital technologies. As previous studies (Carmi et al., 2020; Hidayat et al., 2023) have contended, without data and information literacy competence, HE students cannot act as responsible citizens, whether digital or otherwise, across the various types of digital tools now available. In particular, HE students’ ability to analyse data, handle information, solve problems and engage in critical thinking plays a substantial role in shaping their ability to engage meaningfully with data and information through digital technologies, which also shapes future success. Based on this, the following hypothesis was formulated:

H2: Data and information literacy competence will positively influence higher education students’ digital citizenship.

Communication and collaboration competence and digital citizenship

The rapid proliferation of emerging technologies in the 21st century, particularly within the ambit of education requires digital citizenship, has engendered a pressing need for effective communication modes and collaborative strategies (Xu et al., 2019a; Xu et al., 2019b). Consequently, there is a need for students to become proficient in collaboration and communication. Specifically, communication competencies constitute a contemporary skill set that enables individuals to interact effectively across a diverse array of digital platforms and discern suitable digital communication methods within specific (Vuorikari et al., 2022), while collaborative competence refers to the capability to disseminate data, information and digital content to others through appropriate digital technologies (Vuorikari et al., 2022). Communication and collaboration, that is, skills related to communication and collaboration in digital environments, cooperation and sharing documents and resources through online tools, collaborating with others through these media, as well as participating in scientific communities, forums and spaces for interaction (Pérez-Escoda et al., 2016; Rodríguez-García et al., 2022). Previous research has emphasised the importance of communication and collaboration skills in predicting digital citizenship, as evidenced by studies such as those conducted by Al-Abdullatif and Gameil (2020), Assante et al. (2022), Hidayat et al. (2023), Xu, H. H. Yang, et al. (2019) and Xu, H. Yang, et al. (2019). Similarly, in this study we believe that given the increasingly interconnected world and inexorable utilisation of new devices connected to the internet, HE students’ communication and collaborative competence would have a positive relation to their digital citizenship. Based on this we hypothesised that:

H3: Communication and collaboration competence will positively influence higher education students’ digital citizenship.

Digital content creation competence and digital citizenship

Digital citizenship is then only possible on the basis of a particular kind of conduct, with empowered citizens taking responsibility for their own life outcomes, self-governing behaviour which requires them to invest time and energy in enterprising their own lives. To assess general digital competences, the DigComp framework includes content creation that is also relevant for digital citizenship (Findeisen & Wild, 2022). Digital content creation encompasses the process of generating and refining digital content, incorporating and integrating information and knowledge into an existing body of work, and simultaneously grappling with the enforcement of copyright and licencing, as well as providing clear and comprehensible instructions to computer systems. Digital content creation has emerged over the course of the last decade where users are encouraged to generate create their own content. The debate around citizenship and everyday creativity is more prominent in academic, as a glut of academic writing practices are associated with the creation. Consequently, students are encouraged to participate as digital content-creators and to design, create, make, remix and share their creative content using a range of digital tools and technologies (McGillivray et al., 2016; Thorpe & Gordon, 2012; Sevillano-García & Vázquez-Cano, 2015). In this study, we conceptualised digital content creation as HE students’ ability to create and edit new content, integrate and re-elaborate previous knowledge and content. Based on this, the following hypothesis was formulated:

H4: Digital content creation competence will positively influence higher education students’ digital citizenship.

Safety competence and digital citizenship

Developing students’ safety competence is crucial for effectively infusing their digital citizenship experience into the emerging technological learning process. In today’s digital era, safety competence is essential as a vital aspect of digital citizenship. With digital safety competence, one can become a responsible, informed and ethical digital citizen. According to Budai et al. (2023), the ability to protect the privacy of information from digital devices and being aware of the environmental impact of digital technologies and their use constitute one digital safety competence. Generally, safety in the context of general digital competence refers to as protecting devices; protecting data and digital identity; protecting health; as well as protecting the digital environment (Findeisen & Wild, 2022). It is imperative to view digital citizenship and internet safety as indispensable components of quality education. It is crucial to equip students with the necessary knowledge and skills to navigate a virtual world in a responsible and secure manner. In addition, cultivating a culture of positive safety competence can contribute to a safer and more productive digital environment. This entails educating students about potential online dangers such as cyberbullying, identity theft and exposure to inappropriate content. Based on this, the following hypothesis was formulated:

H5: Safety competence will positively influence higher education students’ digital citizenship.

Conceptual framework

Figure 1 depicts the study’s conceptual framework, which was developed in accordance with the formulated hypotheses.

Figure 1. Conceptual framework. Source: Authors’ construct.

Research method

Design and participant

This research focuses on students enrolled in HEI in Ghana. The study involved current university students at the University of Cape Coast, Ghana. The sampling method employed was convenience sampling, wherein a paper-based questionnaire was administered to students at the university over a span of two months, from June to July 2023. This method encourages individuals to complete the questionnaire based on their availability and willingness to participate (Gravetter & Forzano, 2018). The adoption of the accidental (convenience) sampling method in this research was driven by its convenience and efficiency, as it proved to be easier and quicker in survey administration compared to alternative methods (Al-Adwan et al., 2021). In order to establish the sample size for the research, the investigators utilised the 10-time guideline as outlined by Hair et al. (2017). Following this guideline, the researchers ensured that the minimum number of respondents in the survey equalled or exceeded the total count of paths leading to a latent variable, which in this instance amounted to 50 (for instance, 5 paths multiplied by 10). However, to ensure robustness and validity, 400 questionnaires were administered to the participants. Out of 400 questionnaires distributed, 319 responses were received. Notably, 72 of these responses were deemed incomplete, as a majority of the questions (>70%) were left unanswered. Subsequently, these 72 incomplete questionnaires were excluded from any further analysis. Consequently, a total of 247 questionnaires remained valid for statistical analysis.

Measures

In this study, the researchers employed two key measures to assess the digital competency and citizenship of respondents. The first measure, the Digital Competency Scale developed by Hidayat et al. (2023), comprises 36 items distributed across five dimensions: data and information literacy (5 items), communication and collaboration (7 items), digital content creation (6 items), safety (9 items) and problem-solving (9 items). Respondents’ knowledgeability in these aspects of digital competency was evaluated using a five-point Likert-type scale. The second measure, adapted from Choi et al. (2017), is the Digital Citizenship Scale, which consists of 26 items.

Data analysis

In the present study, partial least squares structural equation modelling (PLS-SEM) was employed to assess the proposed research model. This choice is justified by its capacity to concurrently assess complex relationships among multidimensional constructs (Arthur et al., 2023; Hair et al., 2021). In contrast, alternative statistical techniques like multiple regression or multivariate analysis of variance are constrained to examining the connections between individual constructs separately. Also, it is important to highlight that PLS-SEM mitigates challenges related to small sample sizes and imposes less stringent assumptions regarding the normality distribution and error terms (Valerie, 2012). Given the aforementioned justifications, we contend that the use of the partial least squares (PLS) approach was apt for achieving accurate predictions in our study and may prove valuable in specific conditions where alternative methods fall short (Al-Okaily et al., 2020). Consequently, PLS-SEM was conducted using SmartPLS 3.0 software (version 3.2.9; Ringle et al., 2015) to test the hypotheses in the current study.

Results

PLS-SEM measurement model

The initial step in appraising the reliability and validity of measurement items in PLS involves scrutinising the measurement model, known as the outer model. The evaluation of the measurement model within the framework of PLS-SEM varies depending on whether the model incorporates formative or reflective measures and the inherent nature of the measurement model itself (Davcik, 2014). Therefore, prior to subjecting a proposed model to hypothesis testing, it is imperative to first verify the reliability and validity of its measurement model (Al-Okaily et al., 2022). To fulfil this objective, assessments of convergent validity and discriminant validity must be conducted for the evaluation of the measurement models before progressing to the structural model (Hair et al., 2021).

Reflective indicator reliability

Regarding the reliability assessments, the factor loadings of each indicator (item) were scrutinised. The suggested loadings are expected to exceed 0.708 (Hair et al., 2019), although a threshold of 0.50 is also considered acceptable. As indicated in Table 1, the standardised factor loading estimates for all items were deemed acceptable, ranging from 0.547 to 0.916.

Table 1. Constructs, items, IL, CA (α), rho_A, CR and AVE.

Constructs	No. of items	Item	IL	CA (α)	rho_A	CR	AVE	VIF	p-Value
CC	Six (6)	CC1	0.875	0.890	0.952	0.912	0.639	3.430	***
		CC2	0.894					3.022	***
		CC3	0.908					3.733	***
		CC4	0.602					1.754	***
		CC5	0.809					2.878	***
		CC7	0.653					1.593	***
DCC	Six (6)	DCC1	0.858	0.872	0.932	0.902	0.609	2.379	***
		DCC2	0.877					4.802	***
		DCC3	0.806					2.979	***
		DCC4	0.672					1.636	***
		DCC5	0.654					1.816	***
		DCC6	0.789					2.521	***
DIL	Four (4)	DIL1	0.869	0.805	0.927	0.849	0.592	2.038	***
		DIL2	0.898					1.647	***
		DIL3	0.653					1.679	***
		DIL4	0.616					1.437	***
DC	Seven (7)	CP3	0.599	0.888	0.903	0.913	0.606	1.518	***
		IPA1	0.840					2.933	***
		IPA2	0.856					3.912	***
		IPA3	0.844					6.305	***
		IPA4	0.842					4.561	***
		IPA5	0.817					4.175	***
		IPA6	0.596					2.580	***
PS	Eight (8)	PS1	0.629	0.901	0.922	0.921	0.596	2.526	***
		PS3	0.766					2.232	***
		PS4	0.916					4.706	***
		PS5	0.808					2.719	***
		PS6	0.764					2.328	***
		PS7	0.682					2.708	***
		PS8	0.822					2.961	***
		PS9	0.757					2.597	***
SF	Six (6)	SF1	0.733	0.831	0.883	0.871	0.533	2.177	***
		SF2	0.760					2.373	***
		SF4	0.855					2.335	***
		SF5	0.766					2.095	***
		SF7	0.683					1.714	***
		SF8	0.547					1.619	***

Note: CC, communication and collaboration; DIL, data and information literacy; DC, digital citizenship; DCC, digital content creation; PS, problem-solving; SF, safety; IL, item loadings; CA (α), Cronbach’s alpha; CR, composite reliability; AVE, average variance extracted.

***p < .001.

Internal consistency reliability

In this study, internal consistency reliability was assessed through the examination of composite reliability (CR) and Cronbach’s alpha (α). Consistent with the recommendation by Hair et al. (2019), both Cronbach’s α and CR estimates should be $\ge$0.7. As illustrated in Table 1, all constructs exhibited a CR (ranging from 0.849 to 0.921) exceeding the recommended threshold of 0.7, signifying robust internal consistency. Additionally, all constructs demonstrated acceptable Cronbach’s α coefficients, ranging from 0.805 to 0.901.

Convergent validity

Convergent validity refers to the extent to which multiple items measuring the same concept show agreement (Ramayah et al., 2011). In their research, Hair et al. (2014) proposed the use of average vain assessing convergent validity. The average variance extracted (AVE) gauges the variance captured by the indicators in relation to measurement error, and it should exceed 0.50 to validate the use of the construct (Hair et al., 2014). Conversely, a VIF approaching a value higher than 5 signals potential collinearity issues (Hair et al., 2014). Therefore, it must be less than 5 to ensure the absence of multicollinearity. As presented in Table 1, all the results were deemed acceptable with values aligning with conventional standards and falling within the recommended range. Moreover, an analysis of cross loadings for items across all constructs was conducted to validate convergent validity. Each item within a construct exhibited higher loadings on its intended construct compared to any other construct (see Appendix). Consequently, it can be asserted that convergent validity was established for all constructs. Figure 2 depicts the results of the PLS-SEM algorithm.

Figure 2. PLS-SEM algorithm results.

Discriminant validity

Discriminant validity was assessed using two criteria. Initially, the Fornell and Larcker (1981) criterion was applied, indicating that the square root of the AVE for each construct should exceed its correlation with any other construct in the model. As shown in Table 2, this condition was met, affirming the existence of discriminant validity. Subsequently, the heterotrait–monotrait ratio (HTMT) approach was employed (Henseler et al., 2015). In Table 3, all values are ≤0.85, confirming the results of the Fornell–Larcker criterion and establishing the presence of discriminant validity. In conclusion, all measures utilised in this study exhibited satisfactory validity and reliability. The outcomes for discriminant validity based on the Fornell–Larcker criterion are presented in Table 2.

Table 2. Fornell–Larcker.

	CC	DIL	DC	DCC	PS	SF
CC	0.799
DIL	0.772	0.769
DC	0.490	0.289	0.778
DCC	0.634	0.540	0.597	0.781
PS	0.741	0.513	0.594	0.696	0.772
SF	0.703	0.371	0.479	0.589	0.767	0.730

Note: Values (in bold) on the leading diagonal are the $\sqrt{\mathit{\text{AVE}}}.$

Table 3. HTMT correlation ratios.

	CC	DIL	DC	DCC	PS	SF
CC
DIL	0.895 [0.830, 0.954]
DC	0.476 [0.372, 0.590]	0.268 [0.206, 0.345]
DCC	0.697 [0.603, 0.764]	0.634 [0.513, 0.723]	0.593 [0.498, 0.683]
PS	0.805 [0.723, 0.870]	0.598 [0.477, 0.703]	0.607 [0.486, 0.720]	0.768 [0.683, 0.845]
SF	0.808 [0.722, 0.883]	0.444 [0.347, 0.533]	0.483 [0.374, 0.588]	0.676 [0.601, 0.745]	0.853 [0.789, 0.908]

Table 3 shows the results for discriminant validity using HTMT ratio criterion.

Goodness-of-fit

The assessment of the measurement model aimed for a satisfactory goodness-of-fit (GoF). The comprehensive model fit was appraised by considering five principal indices (see Table 4). Following the guidelines of Henseler et al. (2016) and Benitez et al. (2020), the observed values of all GoF indices fall within and meet the recommended values/conditions. The GoF indices were deemed satisfactory, adhering to recommended guidelines and suggesting a well-fitted model for the data (Henseler et al., 2016).

Table 4. Model fit indices.

Index	Acceptable value/condition	Actual value
SRMR	<0.08	0.054
d_ULS	d_ULS < bootstrapped HI 95% of d_ULS and d_G < bootstrapped HI 95% of d_G	10.894
d_G		8.914
Chi-Square		7098.911
NFI	>0.90	0.915

Note: SRMR, standardised root mean square residual; d_ULS, unweighted least squares; d_G, geodesic discrepancies; NFI, normed fit index.

Assessment of structural model

The evaluation of the structural model involved three primary stages, namely the coefficient of determination ($R^2$), predictive relevancy ($Q^2$) and the significance of path coefficients, as outlined by Hair et al. (2019). The $R^2$ assesses the explanatory capability of the structural model. Furthermore, every path underwent testing through a procedure involving 5000 bootstrap re-samples (Hair et al., 2013). Additionally, the blindfolding procedure was conducted to compute $Q^2$ estimates. The results from the PLS-SEM bootstrapping are depicted in Figure 3 and summarised in Table 5.

Figure 3. PLS-SEM bootstrapping results.

Table 5. Structural model.

Structural path	β	SD	T-Values	p-Values	$R^2$	$f^2$	2.5% LLCI	97.5% ULCI	$Q^2$
PS → DC	0.330	0.078	4.259	<.001		0.056	0.173	0.464
DIL → DC	−0.262	0.110	2.387	0.017		0.040	−0.493	−0.089
CC → DC	0.261	0.121	2.161	0.031	0.441	0.023	0.041	0.496	0.235
DCC → DC	0.399	0.055	7.287	<.001		0.132	0.289	0.506
SF → DC	−0.095	0.100	0.949	0.343		0.005	−0.304	0.084

Note: R² adjusted = 0.430.

Results from Table 5 reveal that problem-solving competence had a statistically significant positive influence on digital citizenship of HE students (PS → DC) [β = 0.330, t = 4.259, p < .001], supporting H1. Interestingly, digital and information literacy had a significant negative effect on digital citizenship (DIL → DC) [β = −0.262, t = 2.387, p = .017 < .05], hence, H2 is not supported. Additionally, the results in Table 5 indicate that communication and collaboration (CC) had a significant positive influence on digital citizenship (CC → DC) [β = 0.261, t = 2.161, p = .031 < .05]. Consequently, H3 was confirmed. The results also revealed that digital content creation competence significantly influenced digital citizenship positively (DCC → DC) [β = 0.399, t = 7.287, p < .001], thereby upholding H4. However, hypothesis five (H5) was not supported as safety competence showed no effect on digital citizenship (SF → DC) [β = −0.095, t = 0.949, p = .343 > .05]. Table 5 displays a coefficient of determination ($R^2$) value of 0.441, suggesting that PS, DIL, CC and DCC competencies collectively explain 44.1% of the variability in DC. This implies that 55.9% of the variance in DC is attributed to other variables not considered in the model. In addition, the results showed that the effect sizes for the significant paths varied between $f^2$ = 0.023 and $f^2$ = 0.132, all of which were characterised as small (Cohen, 2013). Lastly, as the $Q^2$ value (0.235) exceeds zero, it indicates that the model demonstrates predictive relevance (Kock, 2021).

Importance-performance map analysis

To enhance the PLS-SEM analysis, an importance-performance map analysis (IPMA) was carried out to evaluate the significance and effectiveness of factors influencing digital citizenship. Through the use of IPMA, variables with high importance yet low performance can be pinpointed for specific enhancement efforts. The standardised total effects (importance) and the standardised latent variable scores (performance) are detailed in Table 6 and illustrated in Figure 4, respectively.

Figure 4. Importance-performance map.

Table 6. Total effects and performance index values for digital citizenship.

Digital citizenship (DC)	Total effects (importance)	Index value (performance)
CC	0.261	63.800
DIL	−0.262	59.567
DCC	0.399	45.068
PS	0.330	60.094
SF	−0.095	57.161

In terms of their respective levels of significance, the results presented in Table 6 indicate that digital content creation (DCC) emerged as the most crucial factor with the highest ranking (0.399), closely followed by PS (0.330), DIL −0.262) and CC (0.261). Concerning performance, CC recorded the highest score (63.800), followed by PS (60.094), DIL (59.567) and finally DCC (45.068). These findings underscore the importance of higher institutions in Ghana directing their focus towards enhancing digital content creation, as it displayed relatively lower performance compared to other factors, despite being the most critical aspect in nurturing students’ digital citizenship. Figure 4 illustrates the importance-performance map within the context of our model, while Figure 5 depicts the performances of the latent variables.

Figure 5. Latent variable performances.

Revised conceptual framework

Figure 6 depicts a revised conceptual framework that encapsulates the research hypotheses.

Figure 6. Revised conceptual framework. ***p < .001, *p < .05, NS = not significant. Source: Authors’ construct.

Significant path.

Non-significant path.

Summary of results

Table 7 shows the summary of results for the hypotheses that were formulated to guide the study.

Table 7. Summary of results for the hypotheses.

Hypotheses	Statement	Decision
H1	Problem-solving competence will positively influence higher education students’ digital citizenship (PS → DC)	Supported
H2	Data and information literacy competence will positively influence higher education students’ digital citizenship (DIL → DC)	Not supported
H3	Communication and collaboration competence will positively influence higher education students’ digital citizenship (CC → DC)	Supported
H4	Digital content creation competence will positively influence higher education students’ digital citizenship (DCC → DC)	Supported
H5	Safety competence will positively influence higher education students’ digital citizenship (SF → DC)	Not supported

Discussion

This study thrived on the need to understand the factors that influence students’ confidence, critical and responsible use of and engagement with digital technologies for learning and participation in society in the context of Ghana. Drawing on evidence from HE students in Ghana, this study examined the link between digital competency and digital citizenship. The study draws inferences from the digital competence framework (DigComp) with five (5) key components of digital competence: (1) data and information literacy, (2) communication and collaboration, (3) digital content creation, (4) safety and (5) problem-solving to support the hypotheses set. The obtained results revealed that students’ digital competencies, encompassing problem-solving, communication and collaboration, as well as digital content creation had significant positive influence on their digital citizenship. Surprisingly, hypotheses two (H2) was not supported because data and information literacy competence had a significant negative influence on digital citizenship. Also, safety had no significant influence on digital citizenship.

The results of problem-solving emanating from this study imply that an improvement in HE students’ problem-solving skills could result in corresponding improvements in digital citizenship Thus, problem-solving as one key component of digital competency is crucial for, HE students to excel amid the digital paradigm shift to emerging technologies in education with its dual prospects and challenges. The importance of problem-solving competence was supported by previous studies (Bombardelli, 2021; Mahanal et al., 2022) who underscored the significance of problem-solving competence, linking it to enhanced digital competence and citizenship in students. This connection implies that adept problem solvers are likely to possess a well-defined understanding of digital skills and responsible online behaviour, emphasising the importance of problem-solving in digital education.

Unexpectedly, this study found that data and information literacy competence negatively influenced digital citizenship. The unexpected finding that data and information literacy competence negatively influenced digital citizenship may stem from several interconnected factors. Firstly, heightened awareness of online risks and privacy concerns, intrinsic to data and information literacy, might lead individuals to adopt an overly cautious approach in their digital interactions. This apprehension may result in reluctance to engage fully in digital citizenship activities, hindering the development of a sense of community and shared responsibility within online spaces. While a critical perspective is essential, excessive focus on potential risks may overshadow the collaborative and participatory aspects of digital citizenship. Secondly, the abundance of information accessible to those with strong data and information literacy skills could potentially lead to information overload. Additionally, the study utilised a self-rated questionnaire that emphasised competence itself rather than how effectively students applied this competence in the context of responsible digital citizenship. Hence, mere possession of digital and information literacy skills may not automatically translate into responsible digital behaviour. Furthermore, intervening variables such as social attitudes, ethical considerations, and the presence of supportive educational resources could mediate the relationship between digital and information literacy, and digital citizenship.

Negotiating vast amounts of data can be overwhelming, leading individuals to disengage from or withdraw online platforms. This disengagement can impede the development of digital citizenship, which depends on active participation, collaboration and responsible digital behaviour. The tension between managing information abundance and fostering a positive digital presence may contribute to the observed negative influence on digital citizenship. Overemphasising the potential dangers of the digital world in educational programs may unintentionally create a climate of fear and scepticism. If educational programs disproportionately focus on the risks associated with online activities, individuals with heightened data and information literacy may internalise a sense of trepidation, which hampers their willingness to meaningfully contribute to digital communities. It is essential to strike a balance between raising awareness of digital risks and fostering a positive participatory mindset to nurture the next generation of digitally literate and responsible citizens who can navigate the online landscape with confidence and integrity. This finding is contrary to Simić et al. (2020), who argued that data and information literacy are essential skills in the 21st century, as it is now as important to collect and share relevant information as it was to read and write earlier. This finding is also at odds with that of Hidayat et al. (2023), who maintained that without data and information literacy competence, students cannot act as responsible citizens, whether digital or otherwise, across the various types of digital tools currently available.

Furthermore, this study found a significant positive association between students’ digital competence in terms of communication and collaboration, and digital citizenship. This suggests that university students who possess the ability to effectively utilise digital devices and applications for communication and collaboration are inclined to exhibit critical and responsible behaviours when using technology and interacting in a digital environment. This finding is consistent with previous research that has emphasised the importance of responsible online communication and collaboration in fostering digital citizenship, which encompasses guiding students on how to navigate the digital world in their personal and academic lives (Kim & Choi, 2018; Öztürk, 2021; Ranchordás, 2020; Xu, H. H. Yang, et al., 2019; Xu, H. Yang, et al., 2019). Finally, the study established that digital content creation significantly has a positive influence on relationship between competence and digital citizenship. The finding is an indicative of how other factors like customer focus within organisations can be emphasised to yield the performance. When firms adequately direct part of their learning capacity to seeking and satisfying the expectations of customers, it will in turn enhance hotel performance.

However, this study found that safety from the perspective of HE students was not a significant factor associated with digital citizenship. This finding is contrary to existing evidence that safety is one of the key components of digital competence, as captured in the DigComp framework linked to digital citizenship (Hidayat et al., 2023). We believe that the observed inadequate awareness among HE students regarding the substantial positive influence of digital safety concerns on digital citizenship might be a contributing factor to the insignificance of safety competence. This possibly undervalued the influence of safety in the context of digital competence and digital citizenship linkage, highlighting the need to increase awareness creation and education on the crucial interplay between safety and digital citizenship. Moreover, cultural, demographic and institutional variations shape students’ perspectives on safety in the realm of digital citizenship. Cultural backgrounds may impact individuals’ attitudes towards online interactions and influence their perceptions of safety measures. Demographic factors such as age, gender and socioeconomic status can further moderate students’ digital experiences, while institutional settings such as diverse educational environments may introduce varying priorities and norms related to digital safety.

Implications for policy and practice

This study contributes to the literature on technology use in HE by identifying the mechanisms through which students can build their digital competencies in the Internet age and become digital citizens in the ever-evolving dual prospects and challenges of the digital landscape. This study proposed that the digital competence (DigComp) framework will serve as a reference for future researchers in the field. Thus, this study sheds light on the digital competence (DigComp) framework by emphasising the need for students as well as educational stakeholders to adapt to improved digital competencies and citizenship. The findings have confirmed the existing evidence that DigComp, which has identified problem-solving, data and information literacy, communication and collaboration, and digital content creation as key tenets of digital competence, is linked to digital citizenship among HE students in the Ghanaian setting. This means that when these digital competence tenets are deployed, HE students’ digital competencies enhance their digital citizenship towards long life quality and inclusive education expectations.

Correspondingly, Hidayat et al. (2023) reported that dataset from the DigComp framework can help teacher-training institutions, or HE policymakers design effective programmes to improve pre-service teachers’ digital competencies. Similarly, we believe that our dataset analysis from the DigComp framework can expose students to modern digital skill needs and also help policymakers devise strategic measures to meet digital competency demands. Besides the DigComp framework contributions to policy and practice, this study has useful managerial implications for promoting the digital citizenship of students and the HEIs at large. The study revealed that focusing on data and information literacy, communication and collaboration, digital content creation and problem-solving are significant in building HE students’ digital competencies and also their ability to practice digital citizenship.

Accordingly, specific guidelines must be developed by the management of the HEIs and directed towards enhancing students’ digital citizenship demands. Some of the more attractive ways of learning that management of HEIs can do are effective training activities on interactive online modules, cyber security workshops, ethical digital behaviour campaigns and implementing inclusive tech resources but with ethical checks to manage the emerging technological unethical issues. These will help to assimilate and modernise their digital context which can arrest any unforeseen rise in challenging situations that require rigorous problem-solving. Secondly, the management of the facilities should create a work environment and conditions where students will be encouraged to display innovative digital competence and responsible behaviours to meet the citizens’ expectations.

Conclusions

The findings established from the study have enhanced our understanding of factors linked to HE students’ viewpoints of digital competencies that can spur digital citizenship. In light of the prominent findings, it is essential that policy makers and managers of HEIs in Ghanaian setting and beyond draw their attention to the digital competence (DigComp) framework and focus strategies to derive the expected digital citizenship among HE students. Specifically, the DigComp framework, focusing on data and information literacy, communication and collaboration, digital content creation, safety and, problem-solving would unravel the digital skills and ideas that can polish students’ use of emerging digital technologies to make them stand out and appealing to the global digital demands. Also, given the critical role of these digital competencies focus in guiding students towards the digital citizenship, the HEIs will benefit from the recommendations offered by this study. This means that management of HEIs should pay much attention to the DigComp tenets because of their contribution to digital citizenship. By so doing, HE students would develop confidence, critical and responsible use of and engagement with digital technologies for learning in the context of Ghana.

Limitations and future research directions

The study used a cross-sectional and self-reported data and also focused on only one public university in Ghana. These limitations might present issues of generalisation and social desirability effects. Future scholars could further explore HE students’ perceptions on the relationships between digital competence and digital citizenship based on the DigComp framework employing an experimental design and also extend to cover different public universities either in Ghana or beyond. The DigComp model proposed in this study is also open for expansion. Thus, there may be other additional potential external variables explaining the relationship digital competence and digital citizenship. Hence, future studies can extend the widely DigComp framework, by also including attitudes from the study by Adov (2022) and focused more specifically on digital competence in learning settings. Future studies should concentrate on the relationships between various dimensions of digital competence. Moreover, future studies should focus on examining the influence of other factors such as social activity in offline spaces, sense of belonging and level of social capital on digital citizenship. Finally, by identifying several moderators through an exhaustive literature search, future studies can extend the model into a moderated model.

Disclosure statement

No potential conflict of interest was reported by the authors.

Data availability statement

The data on which the findings and conclusions of the study are derived will be available upon request from the corresponding author

Additional information

Funding

The authors did not receive any financial support. Thus, the work was not supported by grants from both internal and external sources.

Notes on contributors

Valentina Arkorful

Valentina Arkorful is a senior lecturer with a PhD in Educational Technology and a Master’s degree in Information Technology Education from the University of Cape Coast. Professionally, she holds a certificate in Faculty Online Teaching from Ghana Technology University and a Certificate in Designing and Facilitating eLearning, Open Polytechnic University of New Zealand. Valentina’s research interests focuses on ICT integration in Teaching and Learning, Gender Adoption of Technologies in Teaching and Learning, E-learning and Assessment Strategies in Education using ICT. She is currently at the College of Distance Education (CoDE), University of Cape Coast, where she facilitates courses on Technology Integration.

Iddrisu Salifu

Iddrisu Salifu is a social scientist with a Master’s degree in Economics from the School of Economics and another Master’s degree in Integrated Coastal Zone Management from the World Bank’s Africa Centre of Excellence in Coastal Resilience (ACECoR), both at University of Cape Coast (UCC), Ghana. His research interests include applied microeconomics and interdisciplinary studies focused on human behaviours in the adoption of emerging technologies, health, and issues related to coastal and environmental management. He is currently a research assistant at the Cardiometabolic Epidemiology Research Laboratory (CERL), HPER UCC, and teaches various courses in economics at the Institute of Education (IoE) Sandwich programme, University of Cape Coast.

Francis Arthur

Francis Arthur holds a Master of Philosophy degree in Economics Education. He is currently a Ph.D candidate in Economics Education in the Department of Business and Social Sciences Education (DoBSSE) at the University of Cape Coast (UCC) who is passionate about research on issues concerning teaching and learning of Economics with much interest in emerging issues such as Preservice and In-service Teachers’ Self-Efficacy, Application of Multiple Intelligences Approach, Artificial Intelligence and Emerging Technologies in Teaching and Learning.

Sharon Abam Nortey

Sharon Abam Nortey is a current Master of Philosophy candidate in Economics Education in the Department of Business and Social Sciences Education (DoBSSE) at the University of Cape Coast (UCC). She has a strong interest in researching issues related to the teaching and learning of economics, with a particular focus on emerging topics such as preservice and in-service teachers’ self-efficacy and the integration of Artificial Intelligence and Emerging Technologies in education.

Appendix

Cross loadings

Table

	CC	DIL	DC	DCC	PS	SF
CC1	0.875	0.637	0.415	0.450	0.598	0.614
CC2	0.894	0.719	0.525	0.628	0.746	0.648
CC3	0.908	0.673	0.482	0.572	0.616	0.691
CC4	0.602	0.488	0.128	0.348	0.509	0.366
CC5	0.809	0.682	0.367	0.568	0.596	0.528
CC7	0.653	0.431	0.139	0.398	0.472	0.363
DCC1	0.498	0.443	0.693	0.858	0.572	0.507
DCC2	0.668	0.487	0.493	0.877	0.666	0.583
DCC3	0.584	0.388	0.405	0.806	0.658	0.529
DCC4	0.371	0.382	0.238	0.672	0.376	0.172
DCC5	0.265	0.369	0.383	0.654	0.316	0.300
DCC6	0.548	0.468	0.373	0.789	0.617	0.543
DIL1	0.714	0.869	0.240	0.517	0.473	0.360
DIL2	0.641	0.898	0.309	0.457	0.407	0.320
DIL3	0.475	0.653	0.028	0.359	0.285	0.049
DIL4	0.554	0.616	0.080	0.292	0.456	0.233
CP3	0.587	0.387	0.599	0.488	0.578	0.497
IPA1	0.392	0.188	0.840	0.463	0.580	0.504
IPA2	0.447	0.350	0.856	0.672	0.514	0.360
IPA3	0.214	0.120	0.844	0.403	0.291	0.191
IPA4	0.263	0.180	0.842	0.445	0.257	0.181
IPA5	0.382	0.170	0.817	0.386	0.519	0.449
IPA6	0.198	0.011	0.596	0.221	0.305	0.274
PS1	0.544	0.388	0.373	0.296	0.629	0.492
PS3	0.552	0.387	0.524	0.708	0.766	0.564
PS4	0.777	0.498	0.615	0.627	0.916	0.745
PS5	0.541	0.373	0.367	0.583	0.808	0.542
PS6	0.559	0.331	0.481	0.471	0.764	0.642
PS7	0.337	0.225	0.279	0.440	0.682	0.349
PS8	0.668	0.553	0.477	0.632	0.822	0.668
PS9	0.474	0.339	0.427	0.458	0.757	0.613
SF1	0.522	0.234	0.262	0.341	0.545	0.733
SF2	0.579	0.232	0.344	0.417	0.618	0.760
SF4	0.530	0.243	0.521	0.522	0.643	0.855
SF5	0.588	0.397	0.341	0.563	0.599	0.766
SF7	0.456	0.294	0.307	0.323	0.503	0.683
SF8	0.528	0.421	0.066	0.478	0.428	0.547

Note: CC, communication and collaboration; DIL, data and information literacy, DC, digital citizenship; DCC, digital content creation; PS, problem-solving; SF, safety. Values (in bold) are the indicator loadings obtained in PLS-SEM and the other values are the cross loadings.