DIGITAL LIBRARY SERVICES AND USER BEHAVIOR IN NIGERIAN UNIVERSITIES: AN EXPECTATION-CONFIRMATION MODEL ANALYSIS

7 International Journal of Social and Educational Innovation Vol. 12, Issue 24, 2025 ISSN (print): 2392 – 6252 eISSN (online): 2393 – 0373 DOI: 10.5281/zenodo.17278219 DIGITAL LIBRARY SERVICES AND USER BEHAVIOR IN NIGERIAN UNIVERSITIES: AN EXPECTATION-CONFIRMATION MODEL ANALYSIS Adedeji Daniel GBADEBO Department of Accounting Science Walter Sisulu University, Mthatha, South Africa [email protected] Abstract This study investigates the behavioral factors influencing university students’ continued use of digital library services in Nigeria, applying the Expectation Confirmation Model (ECM) as the theoretical framework. A structural equation modeling approach was employed to test a conceptual model developed from ECM constructs, such as the perceived usefulness, confirmation, satisfaction, and continuance intention, augmented by system quality and perceived ease of use. Primary data were collected via an online survey distributed across multiple universities in Lagos State, Nigeria, using random sampling techniques. The empirical findings demonstrate that confirmation significantly affects both perceived usefulness and satisfaction, which in turn influence students’ intention to continue using digital library services. Additionally, system quality and perceived ease of use emerged as significant predictors of satisfaction. The study contributes to the literature on digital service adoption in developing contexts by offering evidence-based insights that inform the design, implementation, and policy surrounding academic digital infrastructures. Recommendations are provided for enhancing system quality, managing user expectations, and ensuring equitable digital access in higher education. Keywords: Digital Libraries; User Behavior; Expectation Confirmation Model; Structural Equation Modeling; Higher Education; Nigeria. JEL Codes: C51; I23; O33; D83; L86. International Journal of Social and Educational Innovation (IJSEIro) Volume 12/ Issue 24/ 2025 8 1. Introduction The digital transformation of academic institutions has significantly redefined how information is accessed, processed, and utilized. Among the most affected are academic libraries, which are increasingly integrating smart technologies to enhance the efficiency, accessibility, and quality of library services. This evolution has brought about the proliferation of digital services such as online catalogues, remote access to e-resources, digital lending, and AI-powered information retrieval systems, all aimed at fostering user-centric experiences and improving student engagement with scholarly content (Zhou & Li, 2021). In the context of developing economies like Nigeria, where access to conventional learning resources remains uneven, digital library services present a critical opportunity to bridge gaps in academic resource provision and democratize knowledge access. However, despite the growing presence of these services in Nigerian universities, there remains a lack of empirical evidence concerning how students engage with them over time, especially beyond initial adoption. Understanding user behavior in digital environments necessitates an exploration of not only adoption patterns but also the determinants of continued use. Research in information systems has demonstrated that users’ post-adoption experiences are vital to the long-term success of any technology intervention (Bhattacherjee, 2001). The Expectation Confirmation Model (ECM), widely used in digital service research, offers a comprehensive lens through which to assess this phenomenon. Rooted in cognitive dissonance theory, the ECM suggests that users form expectations prior to using a system, which are then either confirmed or disconfirmed based on their experiences. These outcomes influence users’ perceptions of usefulness, satisfaction, and their subsequent intention to continue using the service (Roca et al., 2006; Mensah & Mi, 2020). Within the academic library context, the ECM serves as a valuable framework to understand the evolving dynamics of student interaction with digital services to satisfaction and loyalty. While numerous studies have applied ECM in areas such as mobile banking, e-learning, and cloud computing (Lin et al., 2019; Bawack et al., 2023), relatively few have examined its application in academic libraries, particularly within Sub-Saharan Africa. This is a notable gap given the unique infrastructural, pedagogical, and sociocultural dynamics that characterize digital service delivery in this region. In Nigeria, academic libraries are gradually transitioning from manual to digital and smart systems, but challenges such as inadequate digital literacy, intermittent internet connectivity, and limited institutional support continue to hinder full utilization (Ojedokun & Okewale, 2020). Consequently, while students may initially adopt International Journal of Social and Educational Innovation (IJSEIro) Volume 12/ Issue 24/ 2025 9 these technologies, their willingness to continue usage often depends on their subjective evaluations of system performance and service quality. To address these gaps, this study adopts the Expectation Confirmation Model to examine how students in Nigerian universities engage with smart library services. Specifically, it evaluates how expectation, confirmation, perceived usefulness, and satisfaction interact to influence continuance intention. A conceptual model with four hypotheses, grounded in ECM literature, was developed to guide the empirical analysis. The study employed Structural Equation Modeling (SEM) to test these hypotheses using survey data collected from a diverse sample of university students in Lagos State, Nigeria. A random sampling strategy was applied to ensure variability in demographic and institutional characteristics, thereby enhancing the generalizability of findings. The data collection instrument consisted of validated ECM measurement items adapted from previous empirical research, ensuring construct validity and reliability (Alraja et al., 2023; Roca et al., 2006). This study contributes to theory by extending the applicability of ECM to the underexplored domain of smart library services in a developing country context. It also provides a studentcentered perspective that illuminates the psychological and cognitive processes underpinning digital service engagement in higher education. Practically, the findings offer strategic insights for library administrators, academic policymakers, and technology developers aiming to optimize user satisfaction and ensure long-term viability of digital library systems. Key implications include the importance of aligning digital services with user expectations, enhancing perceived value through training and interface design, and establishing feedback mechanisms to facilitate continuous service improvement. As digital library ecosystems continue to evolve, particularly in resource-constrained educational settings, there is a pressing need to ground service design and implementation in empirical insights about user behavior. This study answers that call by exploring the cognitive antecedents of digital service continuance through the lens of the Expectation Confirmation Model. In doing so, it enriches the literature on academic library modernization and provides an evidence-based framework for promoting sustained user engagement in digital academic environments across Nigeria and comparable contexts. 2. Literature Review The evolution of digital library services in academic institutions has been extensively studied over the past decade, with numerous empirical investigations exploring factors influencing student adoption, usage, and satisfaction. The Expectation Confirmation Model (ECM) has International Journal of Social and Educational Innovation (IJSEIro) Volume 12/ Issue 24/ 2025 10 been a prevalent framework in these studies, emphasizing the role of user expectations and perceived performance in determining continued usage intentions. Complementing ECM, the Technology Acceptance Model (TAM) and the Unified Theory of Acceptance and Use of Technology (UTAUT) have been widely applied to understand user acceptance of digital library services. A meta-analysis by Ali and Warraich (2024) synthesized findings from multiple studies, highlighting that perceived usefulness and ease of use are significant predictors of digital library adoption among students. In the African context, several studies have examined digital library usage among university students. For instance, Kiana et al. (2023) investigated predictors of digital library usage among undergraduate students in Namibia, finding that library training significantly influences perceived usefulness and actual usage. Umukoro and Tiamiyu (2017) explored determinants of e-library service use among Nigerian university students, identifying factors such as system quality, service quality, and user satisfaction as critical to usage behavior. The impact of digital library services on user satisfaction has also been a focal point in recent studies. Azib et al. (2025) evaluated user satisfaction with digital library services in Malaysian higher education institutions, revealing that information quality, system quality, and service quality significantly affect overall satisfaction. Moreover, the integration of emerging technologies into library services has been explored to enhance user engagement. Wei et al. (2024) introduced an augmented reality system designed to enrich physical library experiences, demonstrating its potential to increase student engagement. In terms of accessibility, Paul and Chauhan (2024) examined the role of AI-powered assistive technologies in enhancing library access for patrons with disabilities, highlighting the importance of inclusive digital services. Studies have also addressed the challenges of digital library service utilization. Adetayo et al. (2024) investigated university students' library engagement, noting that factors such as reading habits and gender dynamics influence usage patterns. Furthermore, the role of social media in engaging library users has been analyzed. Zou et al. (2020) explored strategies for using social media to build participatory library services, emphasizing the need for libraries to adopt user-centered engagement approaches. In the Nigerian context, Isa et al. (2025) conducted a comparative study of digital reference service utilization between Nigerian and Malaysian university libraries, identifying differences in usage patterns and suggesting strategies for optimization. Al-Suqri and Al-Aufi (2019) highlighted that in addition to perceived usefulness, constructs such as digital literacy and cultural context significantly affect the intention to use digital library services among Gulf Cooperation Council (GCC) university students. This supports the growing recognition that International Journal of Social and Educational Innovation (IJSEIro) Volume 12/ Issue 24/ 2025 11 regional, sociocultural, and infrastructural conditions modulate technology acceptance patterns in academic settings, particularly in lowand middle-income countries. Such insights have prompted a reevaluation of one-size-fits-all adoption models, advocating for context-sensitive implementations that address localized barriers to usage. Further, several studies have explored how user training and orientation programs can enhance the utilization of digital resources. For example, Mwantimwa and Ndenje-Sichalwe (2021) demonstrated that structured digital literacy initiatives significantly improve the perceived ease of use and actual usage frequency of e-resources among Tanzanian university students. Their findings align with those of Abdul Karim and Darus (2020), who noted that students exposed to regular digital library training sessions reported higher satisfaction and continuance intention scores. These empirical insights suggest that beyond technological availability, user preparedness and competence play a central role in digital library success. In addition, system design and user interface functionality have emerged as critical determinants of digital service engagement. A study by Monfaredzadeh and Krishnan (2023) indicated that system quality had a significant positive impact on both perceived usefulness and satisfaction among postgraduate users in Iranian academic libraries. This resonates with findings from Asemi and Riyazi (2020), who showed that technical issues such as poor search algorithms or system lags can negatively influence continued usage, even when digital services are otherwise available and promoted. Thus, technical infrastructure, including back-end system reliability, must align with front-end usability to support sustained engagement. Moreover, the role of institutional support and policy frameworks cannot be understated. Empirical studies such as those by Onyegbule et al. (2022) have shown that universities with clearly defined digital library policies, adequate funding, and ongoing technical support exhibit higher student satisfaction and usage rates. This indicates that institutional commitment— manifested through budgetary allocation, continuous service improvement, and responsive user feedback mechanisms—significantly enhances the success of digital library services. Such findings are particularly pertinent for Nigerian universities, where infrastructural underfunding and policy fragmentation frequently undermine the sustainability of digital library projects. Finally, recent interdisciplinary research has highlighted the importance of emotional and psychological variables in determining students’ interaction with digital library environments. For example, a study by Chang and Fang (2022) integrated the concept of digital anxiety into the ECM framework, revealing that students experiencing high levels of anxiety were less likely to perceive digital libraries as useful or satisfying. Similarly, Nguyen and Ha (2021) found that positive emotional engagement, fostered through gamification and personalized International Journal of Social and Educational Innovation (IJSEIro) Volume 12/ Issue 24/ 2025 12 content delivery, led to stronger continuance intentions. These emerging perspectives urge scholars and practitioners to move beyond purely cognitive models and consider affective dimensions of digital service usage in educational settings. 3. Methodology This study adopts a quantitative research approach to evaluate the behavioral determinants of students’ continued usage of digital services in academic libraries, leveraging the theoretical foundation of the Expectation Confirmation Model (ECM). The ECM framework, initially developed by Bhattacherjee (2001), posits that users’ intention to continue using a system is influenced primarily by their satisfaction and perceived usefulness, both of which are conditioned by the degree to which initial expectations are confirmed after actual usage. To adapt this model to the context of smart digital library services in Nigerian universities, the current study integrates additional constructs such as perceived ease of use and system quality, drawing from established extensions in information systems research (DeLone & McLean, 2003; Venkatesh et al., 2012). To formally articulate the hypothesized relationships, a structural model is specified with multiple endogenous constructs. Expectation confirmation (EC) serves as a critical antecedent variable influencing perceived usefulness (PU) and user satisfaction (SAT). These, in turn, are hypothesized to determine continuance intention (CI) toward digital library technologies. In addition, perceived ease of use (PEOU) and system quality (SQ) are introduced as exogenous predictors of continuance intention, extending the ECM to capture system-related dimensions. The structural relationships among the constructs are represented by the following equations: 𝑃𝑈 = 𝛽1𝐸𝐶 + 𝜀1 (1) 𝑆𝐴𝑇 = 𝛽2𝐸𝐶 + 𝛽3𝑃𝑈 + 𝜀2 (2) 𝐶𝐼 = 𝛽4𝑃𝑈 + 𝛽5𝑆𝐴𝑇 + 𝛽6𝑃𝐸𝑂𝑈 + 𝛽7𝑆𝑄 + 𝜀3 (3) In these equations, 𝑃𝑈 denotes perceived usefulness, 𝑆𝐴𝑇 represents user satisfaction, and 𝐶𝐼 is the behavioral intention to continue using digital library services. 𝐸𝐶 stands for expectation confirmation, defined as the degree to which students’ experiences align with their initial expectations. 𝑃𝐸𝑂𝑈 captures perceived ease of use, referring to the extent to which students find the digital services user-friendly, while 𝑆𝑄 reflects system quality, encompassing factors such as system reliability, accessibility, and response time. The coefficients 𝛽1 through 𝛽7 denote the structural path coefficients, and 𝜀1, 𝜀2, and 𝜀3 represent the corresponding error terms. Figure 1 provides the flowchart of the expectation confirmation model. International Journal of Social and Educational Innovation (IJSEIro) Volume 12/ Issue 24/ 2025 13 Figure 1: Expectation Confirmation Model Flowchart All latent variables in the model are measured through multiple indicators adapted from prior validated instruments. Expectation confirmation and perceived usefulness are measured using scale items from Bhattacherjee (2001). Satisfaction is operationalized using items reflecting overall contentment with the digital service experience, while perceived ease of use is assessed using measures adapted from the Technology Acceptance Model (Venkatesh et al., 2003). System quality constructs are based on DeLone and McLean’s (2003) IS success model, incorporating items on reliability, system uptime, and user interface functionality. Continuance intention is measured by students’ stated likelihood of future use and recommendation of the digital services. The study sample consists of university students enrolled across five academic institutions in Lagos State, Nigeria. A stratified random sampling technique was adopted to ensure representative coverage across public and private universities, faculties, and academic levels. The data were collected via an online questionnaire disseminated through institutional email lists and student forums. Responses were screened for completeness, resulting in 452 usable observations. All items were rated on a five-point Likert scale ranging from 1 (“Strongly Disagree”) to 5 (“Strongly Agree”). Data analysis was conducted using SEM implemented in SmartPLS 4.0. The method was chosen due to its capacity to handle complex models involving latent variables and its robustness in analyzing reflective measurement models in small to medium samples. The reliability and validity of the measurement model were assessed using composite reliability, Cronbach’s alpha, and average variance extracted (AVE), all of which met the recommended thresholds. Path coefficients were evaluated through bootstrapping with 5,000 subsamples to test the statistical significance of the hypothesized relationships. The overall model fit was International Journal of Social and Educational Innovation (IJSEIro) Volume 12/ Issue 24/ 2025 14 assessed using the standardized root mean square residual (SRMR), with values below 0.08 considered acceptable, as well as the normed fit index (NFI) and R-squared values for endogenous constructs. The research design adheres strictly to ethical standards for studies involving human subjects. Informed consent was obtained electronically from all participants, and institutional ethical approval was secured prior to data collection. Participant anonymity and data confidentiality were maintained throughout, and the dataset was stored securely for academic purposes only. The methodological framework, therefore, provides a rigorous empirical foundation for evaluating how students' expectations, satisfaction, and perceptions of digital library services interact to influence their long-term usage behaviors. 4. Results and Implications 4.1. Results Table 1 shows the reliability and convergent validity of the measurement model. reliability and convergent validity were assessed using Cronbach’s alpha, composite reliability (CR), average variance extracted (AVE), and indicator loadings. All constructs demonstrated high internal consistency reliability, with Cronbach’s alpha values ranging from 0.817 to 0.880, exceeding the commonly recommended threshold of 0.70 (Nunnally & Bernstein, 1994). Composite reliability scores were all above the acceptable cutoff of 0.70, further confirming internal consistency (Hair et al., 2019). The AVE values ranged from 0.603 to 0.691, surpassing the 0.50 threshold, indicating that more than 50% of the variance in each construct is explained by its respective indicators. Furthermore, the indicator loadings for all items were above the recommended 0.70, signifying strong individual item reliability. Collectively, these results confirm that the measurement model exhibits adequate convergent validity and reliability for all latent constructs. Discriminant validity (in Table 2) was assessed using the Fornell-Larcker criterion. The square root of each construct’s AVE was greater than the corresponding inter-construct correlations, as evidenced by the bold diagonal values in Table 2. For example, the square root of AVE for Continuance Intention (0.832) was greater than its correlation with Perceived Usefulness (0.719) and Satisfaction (0.655), indicating sufficient discriminant validity (Fornell & Larcker, 1981). The construct Expectation Confirmation demonstrated discriminant validity with its highest correlation being 0.589 (with PU), which is lower than the square root of its AVE (0.808). This pattern was consistent across all constructs, confirming that each latent variable in the model is empirically distinct from the others. International Journal of Social and Educational Innovation (IJSEIro) Volume 12/ Issue 24/ 2025 15 Table 1: Measurement Model Assessment Construct Cronbach's Alpha Composite Reliability (CR) Average Variance Extracted (AVE) Indicator Loadings (Range) Expectation Confirmation (EC) 0.842 0.889 0.652 0.730 – 0.880 Perceived Usefulness (PU) 0.865 0.906 0.678 0.740 – 0.900 Satisfaction (SAT) 0.830 0.876 0.615 0.710 – 0.870 Perceived Ease of Use (PEOU) 0.817 0.868 0.603 0.700 – 0.860 System Quality (SQ) 0.849 0.890 0.642 0.730 – 0.885 Continuance Intention (CI) 0.880 0.914 0.691 0.760 – 0.910 Source: Author (2025) Table 2: Discriminant Validity (Fornell-Larcker Criterion) Construct EC PU SAT PEOU SQ CI EC 0.808 PU 0.589 0.823 SAT 0.532 0.678 0.784 PEOU 0.470 0.510 0.455 0.777 SQ 0.525 0.574 0.497 0.613 0.801 CI 0.564 0.719 0.655 0.498 0.551 0.832 Note: Bold values on diagonal are square roots of AVE. Source: Author (2025) The structural model results (in Table 3) indicate strong empirical support for the proposed hypotheses. Expectation Confirmation had a significant and substantial effect on Perceived International Journal of Social and Educational Innovation (IJSEIro) Volume 12/ Issue 24/ 2025 22 Isa, I. I., Nor, N. S. B. M., BT, S. N. A., & Asadullah, S. (2025). Optimizing digital reference services utilization: A comparative study of two university libraries in Nigeria and Malaysia. Open Access Library Journal, 12, e3243. https://doi.org/10.4236/oalib.1113243 Kiana, L., Ujakpa, M. M., Iyawa, G. E., Osakwe, J. O., & Iguna, K. (2023). Predictors of digital library usage by undergraduate students at a Namibian university: Perspectives based on technology acceptance model. African Journal of Library, Archives and Information Science. https://doi.org/10.4314/vsr3a145 Lin, W. S., Wang, C. H., & Huang, M. P. (2019). Investigating the continuance intention of elearning systems: An extension of the expectation-confirmation model. Computers & Education, 136, 118–129. https://doi.org/10.1016/j.compedu.2019.03.009 Mensah, I. K., & Mi, J. (2020). User satisfaction and system continuance: Applying the expectation confirmation theory to mobile government services. Government Information Quarterly, 37(1), 101409. https://doi.org/10.1016/j.giq.2019.101409 Monfaredzadeh, M., & Krishnan, S. (2023). System quality and user engagement in digital academic libraries: Evidence from postgraduate students in Iran. The Electronic Library, 41(4), 753–771. https://doi.org/10.1108/EL-03-2022-0073 Mutula, S. M., & Wamukoya, J. (2019). Bridging the digital divide in Africa: The role of academic libraries. The Electronic Library, 37(2), 248–267. https://doi.org/10.1108/EL-03-2018-0054 Mwantimwa, K., & Ndenje-Sichalwe, E. (2021). Digital literacy and e-resource use among postgraduate students: A case from Tanzanian universities. Journal of Documentation, 77(2), 471–488. https://doi.org/10.1108/JD-10-2019-0204 Ojedokun, A. A., & Okewale, S. A. (2020). Users’ attitudes towards digital library resources in South-Western Nigerian universities. The Journal of Academic Librarianship, 46(5), 102198. https://doi.org/10.1016/j.acalib.2020.102198 Onyegbule, O., Okoro, C., & Eze, S. C. (2022). Institutional support and students’ satisfaction with digital library services in South-Eastern Nigerian universities. Journal of Documentation, 78(4), 861–879. https://doi.org/10.1108/JD-07-2021-0140 Patel, D., Sharma, R., & Joshi, A. (2023). Exploring digital library adoption in Indian universities: A post-adoption analysis. Information and Learning Sciences, 124(1), 34– 53. https://doi.org/10.1108/ILS-08-2022-0092 Paul, S., & Chauhan, S. (2024). Enhancing accessibility in special libraries: A study on AIpowered assistive technologies for patrons with disabilities. arXiv preprint arXiv:2411.06970. https://arxiv.org/abs/2411.06970 Roca, J. C., Chiu, C. M., & Martínez, F. J. (2006). Understanding e-learning continuance intention: An extension of the technology acceptance model. International Journal of Human-Computer Studies, 64(8), 683–696. https://doi.org/10.1016/j.ijhcs.2006.01.003 Sife, A. S., Lwoga, E. T., & Sanga, C. (2020). Mobile technologies for teaching and learning in higher education institutions in Tanzania. Information Development, 36(4), 520–533. https://doi.org/10.1177/0266666919879422 Silva, R. B., de Souza, T. L., & Costa, F. M. (2018). Evaluating university digital library use in Brazil: A structural equation model approach. Online Information Review, 42(3), 337–354. https://doi.org/10.1108/OIR-11-2016-0323 International Journal of Social and Educational Innovation (IJSEIro) Volume 12/ Issue 24/ 2025 23 Umukoro, I. O., & Tiamiyu, M. A. (2017). Determinants of e-library services’ use among university students: A study of John Harris Library, University of Benin, Nigeria. Journal of Librarianship and Information Science, 49(4), 438–453. https://doi.org/10.1177/0961000616654959 Venkatesh, V., Thong, J. Y. L., & Xu, X. (2016). Unified theory of acceptance and use of technology: A synthesis and the roadmap ahead. Journal of the Association for Information Systems, 17(5), 328–376. https://doi.org/10.17705/1jais.00428 Wang, Y., Sun, Y., & Wang, Y. (2020). The impact of perceived value and trust on e-learning adoption in China. Computers & Education, 143, 103618. https://doi.org/10.1016/j.compedu.2019.103618 Waweru, J. N., & Mutula, S. M. (2022). Technology adoption and digital information service use among postgraduate students in Kenya. Library Hi Tech, 40(2), 409–428. https://doi.org/10.1108/LHT-04-2021-0125 Wei, Q., Zhang, J., Wang, P., Jin, X., & Fan, M. (2024). Augmented library: Toward enriching physical library experience using HMD-based augmented reality. arXiv preprint arXiv:2408.06107. https://arxiv.org/abs/2408.06107 Zhou, M. (2021). Expectation confirmation and digital learning continuance: A cross-cultural study of Chinese and European university students. Educational Technology Research and Development, 69(3), 1239–1260. https://doi.org/10.1007/s11423-021-09993-z Zhou, Y., & Li, M. (2021). Intelligent digital libraries: Current trends and future directions. The Electronic Library, 39(5), 910–929. https://doi.org/10.1108/EL-11-2020-0329 Zou, H., Chen, H. M., & Dey, S. (2020). Engaging users through social media in public libraries. arXiv preprint arXiv:2003.04204. https://arxiv.org/abs/2003.04204