2.1. Theoretical Background

US stands as a paramount metric when gauging the success of IT implementations (Keikhosrokiani et al., 2020; Zaitul et al., 2018; Zviran & Erlich, 2003). Furthermore, the value of foundational skills training cannot be understated; such training amplifies US and promotes increased voluntary engagement with the system, although not necessarily more intensive usage (Zaitul et al., 2018). The quality of communication between users and technical personnel, as underscored by previous information system (IS) research, holds significant sway over the acceptance, evaluation, and ultimate utilization of an IS (Gefen & Keil, 1998; Naeem et al., 2023). In the ever-evolving digital landscape, librarians are called upon to be adaptive, embracing new technologies to optimize user service experiences (Igbinovia, 2021). This pivot toward new technologies is not just a matter of modernization; it is grounded in research that scrutinizes technology acceptance factors, especially in developing nations, shedding light on concepts such as the Internet of Things. Evaluating technology acceptance is integral to gauging the success of an IS. Esteemed IS researchers have posited that a user’s acceptance of technology is intrinsically linked to their intention to use and the actual utilization of said technology (Marangunić & Granić, 2015). According to the McLean and DeLone model of IS success, US and utilisation are significantly impacted by the quality of the system, the information, and the services (Chen, 2011; Swe & Yang, 2021). The research pinpoints perceived usefulness (PU) and perceived ease of use (PEOU) as key determinants in IS implementation (Gefen & Keil, 1998). Elaborating on this, Davis et al. (1989) highlighted challenges related to user accessibility of computer systems and the ensuing repercussions on organizational performance. This intricate dance of acceptance and rejection of computer based IS is influenced by a diverse array of beliefs, attitudes, and satisfaction levels. Often, these determinants are deployed as a robust theoretical or psychometric foundation. A critical distinction is made between the TAM and the Theory of Reasoned Action (TRA), with the former emphasizing the pivotal roles of PU and PEOU in shaping users’ attitudes, intentions, and actual computing behaviors. In essence, PEOU underscores the notion that individuals can achieve more with the same effort, thereby directly impacting PU. Models such as the one proposed by Saeed and Abdinnour-Helm (2008) lean on the principles of TAM and the Information System Success Model (ISSM), revealing the weight of information quality (IQ) and system integration on IS utility.

Further studies, such as findings by Wixom and Todd (2005), highlight the academic proclivity to merge TAM with US theories. Pioneering works by researchers, for example Chen et al. (2016), Chen and Tsai (2019), and Soltani-Nejad et al. (2020), have sought to integrate the TAM with other models, creating robust frameworks to understand user intentions and satisfaction in various digital platforms. Ali et al. (2019) integrated TAM and ISSM to examine e-learning success, specifically hypothesizing among undergraduate students in Pakistan. Jin and Xu (2021), leveraging both ISSM and structural equation modeling (SEM), found that a number of factors significantly influenced US, including the quality of the system, the information it contains, and the services it provides. This in turn impacted loyalty toward paid knowledge platforms, as gauged by perceived value. Abdul Rahman and Mohezar (2020) underscored influential factors on DL, including perceived quality, instrumental support, PEOU, user expectations, satisfaction, net benefits, and PU. Lan and Chen (2022) combined TAM and ISSM to investigate the impact of attitudes, satisfaction, and continuing usage of telehealth services on middle-aged and older individuals living in remote places. Oktal et al. (2016) evaluated the National Judiciary Information System, revealing that IQ was the pivotal latent variable in the SEM. When one considers the vast landscape of IT/IS adoption theories, several stand out, notably: the Unified Theory of Acceptance and Use of Technology, the Theory of Planned Behavior, and the TAM by Davis et al. (1989). Within this theoretical milieu, IT/IS evaluation models emerge as essential tools, helping to elucidate user requirements and pinpoint crucial dimensions and factors for system development, ultimately fostering user acceptance and satisfaction. This study, in its quest for depth and clarity, leans on two central models.

2.2. Information Quality

The caliber of information can be assessed through parameters such as usefulness, accuracy, reliability, and timeliness. Furthermore, a high quality of information indicates that it is pertinent, dependable, accurate, and presented in an easily understood way. Such high-quality information is typically viewed favorably by users, who find it helpful and valuable, which enhances their entire system experience. IQ was required to determine which library resources were most popular in order to enhance access to them (Shevchenko, 2020). Users are more inclined to trust the system and find it helpful for their particular needs when they believe the information is accurate and pertinent. Consequently, there may be a rise in user happiness, engagement, and system utilization. Adeyemi and Issa (2020) emphasized that the quality of information is pivotal when devising and implementing university web portals to ensure US. They advocate for university stakeholders – from management to IT experts – to prioritize assessing the quality of content on their web portals. Sharma et al. (2011) observed that users who are aware of the plethora of e-resources in their domain are more inclined to exploit online resources, bolster infrastructure, and participate in structured programs. This awareness, according to Sharma et al. (2011), increases the level of user pleasure. According to Soltani-Nejad et al. (2020), there is a strong correlation between US in DL and characteristics like information, system, and service quality. Zorba (2013) noted the dual roles of library websites, which can either serve informational purposes or act as marketing tools. The content requirements vary from departmental outlets to broader university platforms. In summary, high-quality information that is accurate, relevant, reliable, and well-presented positively affects opinions of users regarding practicality and convenience of usage. By ensuring the provision of high-quality information, systems and technologies can enhance US, engagement, and overall user experience. Given these insights, we put out the following hypotheses:

H1. Improving IQ positively affects users’ perceptions of its utility.

H2. Improving IQ positively affects users’ PEOU.

2.3. System Quality

Higher quality, dependability, and usefulness generally raise the possibility that consumers will find a system to be easier to use. It is important to remember, nevertheless, that users’ subjective assessments of usability can be impacted by a range of factors other than SQ. For instance, even a sophisticated system of the highest caliber could be viewed as challenging to use if the user interface is poorly made or if the user is not equipped with the required knowledge or abilities to operate it. Similar to this, if a system has an easy-to-use interface, even a lower-quality system could be seen as user-friendly. Superior SQ and user interface design are crucial in ensuring the delivery of high-quality information (Tun, 2021a). Academic libraries must invest in refining SQ, reconfiguring resource structures, overhauling operational processes, and ensuring seamless maintenance. This emphasis stems from the fact that SQ is crucial for optimizing web environments and addressing user demands (Zhai & Wang, 2016). Service quality stands as a paramount determinant of library website performance. Therefore, educational institutions must prioritize SQ to positively impact US with E-Library resources (Chang et al., 2015). Zhou (2014) contended that the perceived utility of a system largely hinges on its quality. Concurrently, the caliber of information has a significant bearing on US. To keep users engaged and contented, it is imperative for service providers to dispense reliable, up-to-date, and pertinent data. Satisfaction, as Zhou (2014) elucidates, is a metric reflecting the discrepancy between service quality and user expectations. Drawing from this discourse, SQ amplifies PEOU via PU, trust bolsters PU, and in turn, PU augments user contentment (Oktal et al., 2016). In conclusion, SQ has a significant impact on how users perceive a system’s usefulness and effortlessness of use. A high-quality system that performs well, provides the desired functionality, and is user-friendly can enhance the overall user experience, increase US, and favorably affect how beneficial and simple consumers perceive something to be. In light of these discoveries, we propose these hypotheses:

H3. PEOU benefits from high SQ.

H4. PU benefits from high SQ.

2.4. Librarian Service

To effectively advocate for the importance of libraries in today’s digital age, librarians must arm themselves with modern technological skills. By focusing on user-centric terminology, librarians can align their services with common language and terms, ensuring accessibility and promoting engagement within academic environments (Benedetti, 2017). Additionally, librarians can offer training sessions or workshops to educate users on information literacy skills, efficient search strategies, and proper citation practices. By empowering users with knowledge and skills, librarians enhance both PU and PEOU, making the library services more valuable and user-friendly. Besides this, PU greatly influences librarians’ attitudes toward IT. Jabeen et al. (2017) discovered that users expressed dissatisfaction with the assistance they received from librarians in navigating DL resources. Interestingly, these users were adept at utilizing various search engines and had gained substantial familiarity with library databases via librarian-provided services, such as accessing journals, articles, and full-text databases crucial for their courses. Emphasizing the significance of library orientation, Kiana et al. (2021) found a notable correlation between library training and DL usage. In the context of university DL, librarians play a pivotal role, particularly in offering specialized training sessions and facilitating research endeavors (Soltani-Nejad et al., 2020). In conclusion, librarians have a significant impact on how helpful and easy users believe library services to be. Their expertise, assistance, and guidance contribute to users’ perceptions that the service is valuable, relevant, and easy to use, ultimately enhancing the overall user experience. Therefore, the following hypotheses are formulated:

H5. There is a positive correlation between PU and librarian service.

H6. In terms of PEOU, librarian service is beneficial.

2.5. Perceived Ease of Use

Across the globe, libraries provide services mostly through websites and the Internet. Web designers have a responsibility to create websites that are visually appealing, engaging, and easy to use (Brahma & Verma, 2018). Ease of use makes it easier for users to extract value from the system, which in turn increases their perception of its usefulness. For example, a user-friendly project management tool that simplifies task management, collaboration, and reporting can be perceived as more useful by its users. Users are more satisfied with a system overall when they find it straightforward to use. Easy-to-use systems reduce frustration, minimize errors, and increase user confidence and control. Users appreciate systems that require less effort to accomplish tasks and allow them to achieve their goals efficiently. A positive user experience resulting from a high level of ease of use contributes to US. For instance, a seamless and intuitive e-commerce website that enables users to easily find and purchase products, that provides secure payment options, and offers clear and helpful product information can lead to higher US (Mamakou et al., 2024). For a system to be truly user-friendly, it should be intuitive, allowing users to seamlessly access information and services. A DL’s accessibility directly motivates its users (Masrek & Gaskin, 2016). These scholars posit that PEOU—a core tenet of the TAM—strongly resonates with US. Similarly, Thong et al. (2002) posited that PEOU indirectly enhances PU, mediated by the intention to use. To summarize, there is a positive correlation between PEOU and PU in addition to US. Users are more likely to find and accept systems that are convenient to use, which increases satisfaction with the system. Designing systems with a focus on usability and providing a smooth user experience can enhance both PU and US, leading to improved user engagement and overall system success.

Hence, we propose:

H7. PU benefits from PEOU.

H8. PEOU positively affects US.

2.6. Perceived Usefulness

Perception of usefulness is the term for the user’s perception of how beneficial a system or technology is in facilitating their tasks or achieving their goals. When users perceive a system as useful, they believe that it can effectively meet their needs, provide value, and help them accomplish their objectives. For example, if an e-commerce platform offers a wide range of products, competitive prices, and convenient shopping features, users are more likely to perceive it as useful. Masrek and Gaskin (2016) found that PU enhances the service of DL or job performance. Madle et al. (2003) emphasized the importance of usability testing for DL, especially in assessing the graphical interface, navigational ease, and user acceptability. Thong et al. (2002) suggested that PU has more influence over a DL than does PEOU. Additionally, PEOU indirectly sways intention to use, mediated through PU. US refers to the overall positive evaluation and contentment that users experience when interacting with a system or technology. When users perceive a system as useful in fulfilling their needs and goals, it contributes to their satisfaction. If a system effectively provides the features, functionalities, and benefits that users expect, it increases their satisfaction. For instance, if a user service software enables quick issue resolution, responsive communication, and personalized support, it enhances US. In summary, when users perceive a system as useful, it positively influences their satisfaction. Users are more likely to be satisfied when a system meets their expectations, helps them achieve their goals, and provides value. Designing systems with a focus on delivering PU can contribute to enhanced US, leading to increased user engagement, loyalty, and overall system success. Consequently, we propose the following:

H9. PU positively affects US.

To sum up, the main goal of this research is to develop and build a novel model that investigates the factors that influence users’ satisfaction with the websites of Myanmar universities’ libraries. Drawing inspiration from the preceding discourse, this research endeavor seeks to sculpt a high-quality framework tailored for Myanmar’s ULWs. The objective is not just a theoretical exercise but a practical overhaul—reimagining university libraries to underscore the role of their digital portals, meeting user needs, and ensuring satisfaction. This research model aims to elevate the standards of ULWs in Myanmar by diving deep into user feedback, as seen in Fig. 1.

Fig. 1

Model of research.

5.1. Demographic Profile of the Respondents

Given these contexts, the gender skew in the present study concurred with the prevailing societal dynamics. Age wise, the respondents were distributed as follows: 24.5% below 24 years, 23.1% between 25-29, 19.8% between 46-55, 17.6% between 30-35, and 15.0% between 36-45. Academically, a vast majority (79.5%) were graduate students, followed by postgraduates (14.7%) and undergraduates (5.9%). Disciplinary distribution revealed 57.1% from arts and 42.9% from sciences. Analyzing specific ULW engagement, 20.1% of participants used the universities’ central library (UCL), 17.2% Dagon University, 16.8% University of Yangon, 16.1% University of Computer Studies Yangon, 15.8% East Yangon University, and 13.9% engaged with Yangon University of Foreign Languages. Further details are located in Table 3.

5.2. Preliminary Descriptive Analysis of Data

According to Kline (2011), a Z score (standard deviation) that falls between -0.3 and 0.3 should be deleted. For data to be considered normally distributed, kurtosis should lie between -7 and +7, and skewness should lie between -2 and +2. These criteria can be viewed in Table 4. The results indicate that all questionnaire items possess standard deviation. Table 4 displays skewness and kurtosis values ranging from -2 to 2. Based on this data normality, respondents appear to have a strong conviction that IQ can meet their requirements for ULW usage, as seen in the standard deviations for IQ 2, 3, 4, and 1. Additionally, respondents also expressed positive beliefs regarding US 4 and SQ 5.

5.3. Results of Factor Loading and Cronbach’s Alpha

The Kaiser-Meyer-Olkin value for factor analysis stands at 0.912, with a chi-square value of 4,613.195. In line with Fornell and Larcker (1981), the Cronbach’s alpha lies between 0.875 and 0.93. All composite reliability values exceed 0.8, surpassing the recommended threshold of 0.7 (Hair et al., 2019). The factor loadings for each item in the measurement model, as shown in Table 5, range from 0.652 to 0.904. Hair et al. (2019) suggest that factor loadings below 0.50 indicate items that may need removal. However, with factor loadings above 0.50, as in this case, all criteria for convergent validity are met, as illustrated in Table 5.

5.4. Result of Pearson’s Correlation

The findings indicate that all variables exhibit a strong positive correlation at the 0.01 significance level.

The correlation coefficients range from 0.536 for IQ and PEOU to a peak of 0.731 for SQ and IQ. Shaded cells within Table 6 denote the nine causal relationships illustrated in the proposed research model (Fig. 2). It is important to note that while these correlations are substantial, they do not necessarily confirm the presence of strong causal relationships, as highlighted by Kline (2011).

Fig. 2

Research model with hypothesis results. *p<0.05, ***p<0.001.

5.5. Results of Convergent Validity, AVE, and Composite Reliability

The average variance extracted analysis (AVE) values ranged between 0.709 and 0.836, which are well above the guideline of 0.5 (or 50%) as suggested by Hair et al. (2019) (refer to Table 7). All the indicators were found to be significant, with the standardized regression weights for all indicators surpassing 0.768. Consequently, the dataset demonstrates robust convergent validity and a high level of internal consistency reliability.

5.6. Results of Discriminant Validity

Hair et al. (2019) emphasized that assessing discriminant validity, which determines differences among correlated factors, is crucial for validating constructs. In Table 8, all the square root values of AVE exceed the correlation estimates for each construct, and these values are highlighted in bold.

Based on the analysis, the proposed research model demonstrates robust discriminant validity, making it suitable for subsequent SEM analysis. This can be referenced in Table 8.

5.7. Result of Model Fit Indices

The fit indices for the measurement model are as follows: goodness-of-fit index (GFI)=0.886, adjusted goodness-of-fit index (AGFI)=0.842, normed fit index (NFI)=0.928, incremental fit index (IFI)=0.956, comparative fit index (CFI)=0.956, and the root mean square error of approximation (RMSEA)=0.074 with x²/df=2.479. In the subsequent analysis of the structural model’s fit indices, the results yielded x²/df=2.836, GFI=0.872, AGFI=0.827, NFI=0.916, IFI=0.944, CFI=0.944, and RMSEA=0.082. Given that GFI and AGFI values exceed 0.8, NFI, IFI, and CFI values surpass 0.9, and with RMSEA exceeding 0.08 for both the measurement and structural models, these indicators point toward a good fit. These findings can be cross-referenced in Table 9. Consequently, this suggests that the research model for this study offers an acceptable fit to the data at hand.

5.8. Results of Hypotheses and Research Mode

Table 10 presents the outcomes of the hypothesis testing. H1 and H5 indicate that librarian services significantly influence both PU (β=0.164, p<0.001) and IQ (β=0.195, p<0.05). Additionally, SQ considerably affects perceived usability (β=0.657, p<0.001). There is also a marked influence of PEOU on PU (β= 0.470, p<0.001), and PU notably impacts US (β=0.708, p<0.008), leading to the approval of H3, H7, and H9. Furthermore, both librarian services (β=0.037, p<0.05) and IQ (β=0.020, p<0.05) positively impact PEOU, while SQ has a notable influence on PU (β=0.135, p<0.05). However, H2, H4, and H6 were not supported. The data also contradict H8, indicating that PEOU does not significantly influence US. All these insights from the hypothesis testing can be viewed in Fig. 2.

6.1. Contribution of Theory

This research provides numerous noteworthy theoretical advances to the literature on ULWs, particularly in relation to US. First, this research aims to bridge the existing gaps between theory and practice specific to ULW services in Myanmar. Additionally, this study illuminates the multifaceted nature of US, revealing how it hinges on the prevailing quality of the ULW system, information, librarian service, PEOU, and utility. Several theories underscore the study’s findings. For instance, our results echo Gefen and Keil (1998), Davis et al. (1989), and others, suggesting that IQ significantly influences PU (Lan & Chen, 2022). Similarly, research by Mahmood et al. (2000), Zviran and Erlich (2003), Xu and Du (2018), Izuagbe et al. (2019), and then more recent studies resonates with our discovery that librarian services contribute to this PU. The relationship between service quality and PEOU has been substantiated by Senaratne and Samarasinghe (2019). Notably, the influence of PU on US was also observed in the findings of Zviran and Erlich (2003), who found that numerous ISs failed or were underutilized over the past five years. A plausible explanation for this could be the inactivity of the websites when users access ULW (Firdaus et al., 2019).

However, some of our findings diverge from established theories. For instance, contrary to expectations, IQ did not influence the PEOU of ULW services. This might be attributed to ULWs’ services. Similarly, librarian services did not have a significant impact on PEOU. This aligns with Wolfe et al. (2010), who found that users often relied more on search engines such as Google rather than ULWs. They also observed that users lacked the patience to await librarians’ responses, often feeling more knowledgeable than the librarians. Furthermore, SQ was not a significant influencer of the PU for ULW US. Cho et al. (2019) made a similar observation in their study on e-government services. The possible reason could be the enhanced user productivity offered by the current systems, ensuring seamless access. Last, while we anticipated a significant effect of PEOU on US, the results revealed no such influence. Zviran and Erlich (2003) also reported a similar result in their study of IS in the US literature: “Users will not be as willing to use a system if they believe that it is not satisfactory from their perspective.” Bandura (1982) and Lepper (1985) also reported similar findings in their studies on self-efficacy and motivational issues, both from psychological perspectives, and stating that efficacy is a major factor underlying intrinsic motivation.

Besides this, nothing observed on ULW is comprehensive or the end-all when it comes to tools or methods for ensuring US (Lim, 2023). In conclusion, our findings demonstrate that PU directly correlates with heightened US. If universities aspire to amplify US on their websites, these platforms should not only be inherently useful but should also offer seamless technical support, demand minimal setup, and be intuitively easy to learn (Zaitul et al., 2018). Interestingly, in the context of DL, PU emerged as a more reliable indicator of US than usability, aligning with studies by Madle et al. (2003), Oktal et al. (2016), and Chaloeyprach and Ketmuni (2021). This reinforces previously established patterns in research focusing on user acceptance and IT adoption (Guner & Acarturk, 2020). Importantly, PEOU and usefulness act as intervening variables bridging SQ, IQ, and librarian services to US concerning ULW service usage.

6.2. Managerial Contributions

Given that a ULW epitomizes a university’s academic standing, one can gauge the institution’s reputation through its ULW. Achieving optimal US is a continual endeavor, requiring sustained research to understand how user feedback evolves over time (Mahmood et al., 2000). In light of this, university heads should actively encourage website administrators and library specialists to elevate both the information and SQ of the ULW.

A collaborative approach involving university rectors, website administrators, librarians, and other library professionals is vital to ensure high-caliber ULW services in Myanmar. As we look forward, there is an evident need to enhance the information and SQ of ULW services in Myanmar. Addressing this requires a comprehensive strategy encompassing adequate budgets, skilled technicians, and effective dissemination of ULW services by librarians and library professionals. Furthermore, it is imperative for university librarians and academic staff in the Department of Information Studies to continually assess their library websites. They should gauge user engagement and satisfaction levels and provide annual reviews of ULW advancements to the Department of Higher Education (under the aegis of the Ministry of Education). By implementing such practices, users will become more familiar with ULW services through insights from research surveys. This will eventually foster increased awareness and satisfaction levels, propelling the stature of ULW services in Myanmar.

6.3. Research Limitations and Prospects for the Future

The present study’s scope is confined to a cross-sectional examination, focusing solely on factors influencing US. Notably absent is a comprehensive understanding of international ULW users’ perceptions and a ranking of which ULW offers the highest US in Myanmar. Given the rapid evolution of technical support, there is a pressing need for future studies to adopt qualitative and longitudinal approaches. Another limitation of this research pertains to its focus on only six universities. Expanding the study to encompass additional universities in Myanmar could provide a more holistic view of US in relation to the country’s ULW ecosystem. Furthermore, a potential avenue for exploration is the integration of mobile apps tailored for ULWs, emphasizing information and SQ designs. This proposition not only underscores the relevance of mobile technologies in today’s digital landscape but also suggests a forward-thinking approach to further embed IT within Myanmar’s academic sector.

Moreover, US is influenced by a wide range of factors from many perspectives. Among them, the concepts based on library service fields are mentioned for further direction. According to Sureshchandar et al. (2002), there are differences between US and service quality, but they are also closely related, thus improving one will probably improve the other. Other scholars have employed single-item scales to operationalize US, while others have employed multiple-item scales. Oh (2013) proposed a conceptual model consisting of fifteen items to calculate the public library’s satisfaction index. This model considers three factors: service quality (50%), customer satisfaction (40%), and loyalty (10%).

Using LibQUAL+TM, Twum et al. (2021) investigated the concept of service quality and offered factual data on variables impacting library patron loyalty. The image of a library brand is influenced by service effect, information access, personal control, and user loyalty. For the library to cultivate a favorable brand reputation and patronage, management must always raise the calibre of its services. According to Oh (2023)’s recommendation, users’ perceptions of service quality in public libraries focus on three dimensions: library personnel, resources and user services, and facilities and equipment. The results indicated that users were generally satisfied not only with the library services and expressed loyalty towards the libraries, but also that service quality positively influenced customer satisfaction, which impacted user loyalty. Similarly, Oh (2020) also analyzed data from 1,015 public library users in South Korea to investigate their perceptions of service quality, satisfaction, and loyalty. The results showed high satisfaction levels, loyalty, and perceived service quality, with significant impacts observed between these variables. Therefore, Oh (2020; 2023) observed that user loyalty was found across different demographic groups.