The Effects of Information Systems Quality on Nurses’ Acceptance of the Electronic Learning System

Abstract

Background: 

To ensure the quality of healthcare provision, nurses must continuously enhance their professional knowledge and competencies via continuing education. As compared with traditional learning, electronic learning (e-learning) is a more flexible method for nurses’ in-service learning. Hence, e-learning is expected to play a key role in providing continuing education for nurses.

Purpose: 

The main purpose of this study was to examine whether system quality, information quality, service quality, and user-interface design quality as the antecedents to nurse beliefs can affect nurses’ intention to use the e-learning system.

Methods: 

A cross-sectional design was used to investigate the effects of information systems quality on nurses’ acceptance of the e-learning system. This study gathered sample data from nurses at 3 hospitals in Taiwan. A total of 450 questionnaires were distributed, and 320 effective questionnaires were analyzed in this study, indicating an effective response rate of 71.1%. Collected data were analyzed using structural equation modeling.

Results: 

System quality, information quality, and user-interface design quality had significant effects on perceived usefulness (PU), perceived ease of use (PEOU), and perceived enjoyment (PE), and service quality had significant effects on PU and PEOU. Moreover, PEOU had significant effects on PU and PE, and the effects of PU, PEOU, and PE on intention to use were significant.

Conclusions/Implications for Practice: 

User-interface design quality is the most key antecedent that can make significant impacts on nurses’ PU and PE, and more efforts should be made to develop a friendlier user interface via designing useful and enjoyable features to induce nurses to use the e-learning system. Moreover, system quality can make the greatest impact on nurses’ PEOU; thus, medical institutions should effectively enhance system quality to deliver benefits and pleasure to boost nurses’ usage intention of the e-learning system via reducing the complexity.

Introduction

Research Background and Purpose

Due to the rising trend in patient-centered care, nurses must continuously enhance their professional competencies to ensure the quality of healthcare provision (^{Chen, Yang, Tang, Huang, & Yu, 2008; Wu, Hwang, Tsai, Chen, & Huang, 2011; Žvanut et al., 2011}). Hence, continuing education plays a vital role in strengthening the nursing professional development. Recently, with the application trend of information and communication technologies and Web-based technologies on education, electronic learning (e-learning) may facilitate nurses to engage in more learning (^{Chen, Chang, Hung, & Lin, 2009; Liang, Wu, & Tsai, 2011}), and it is expected to play an important part in providing continuing education for nurses (^{Chen et al., 2008; Sheen, Chang, Chen, Chao, & Tseng, 2008}). Essentially, e-learning is a powerful tool that uses the network technology via electronic media to help organizations deliver learning materials to users and utilizes Web-based communication to support users’ active learning anytime and anywhere (^{Baylari & Montazer, 2009; Lee, Yoon, & Lee, 2009}). Moreover, e-learning systems can provide e-learning platforms that use the electronic medium as a delivery mechanism to allow users from all over the world to access a number of learning tools for their e-learning courses (^{Ngai, Poon, & Chan, 2007}). In general, traditional learning method typically occurs in a real classroom/laboratory and usually involves nurses in the clinical skills demonstration and rehearsal (^{Bloomfield, Roberts, & While, 2010}), but it is difficult for instructors to provide personalized learning support to tailor their instructions to individual nurses’ needs (^{Bloomfield et al., 2010; Hwang, Yang, Tsai, & Yang, 2009}). As compared with traditional learning, e-learning is a more flexible method for nurses’ in-service learning without time, distance, and space barriers (^{Lin, Lin, Jiang, & Lee, 2007; Yu, Chen, Yang, Wang, & Yen, 2007}), and it is particularly suited to nurses with high self-control that allows them to learn at remote locations based on their needs and schedules (^{Yu et al., 2007}).

However, simply providing nurses with a Web-based learning system does not guarantee successful e-learning. Nowadays, the quality assurance, improvement, and enhancement of e-learning in institutions have increasingly become a challenge for e-learning providers (^{Jara & Mellar, 2010; Ozkan & Koseler, 2009}). Hence, quality assessment has become an essential requirement of evaluation for users’ e-learning acceptance (^{Ozkan & Koseler, 2009}). Up to now, ^{Ahn, Ryu, and Han (2007), Kim, Lee, and Law (2008)}, and ^{Lin (2007)} have addressed the relationships between quality factors and user acceptance of information systems (IS)/information technology (IT), but the extent to which quality factors affect nurses’ acceptance of the e-learning system has seldom been examined. Hence, the main purpose of this study was to examine whether IS quality factors as the antecedents to nurse beliefs can affect nurses’ intention to use the e-learning system.

Theoretical Background

Technology acceptance model and Van der Heijden’s model

To date, many models have been used to explain users’ IS/IT acceptance. Among them, the technology acceptance model (TAM), proposed by ^{Davis (1989)} and ^{Davis, Bagozzi, and Warshaw (1989)}, has been one of the most widely applied models in related IS/IT acceptance studies and has received extensive empirical support (^{Lee et al., 2009}). Fundamentally, TAM examines IS/IT acceptance primarily from the instrumental view, focusing on extrinsic motivators such as perceived usefulness (PU) and perceived ease of use (PEOU), and the two beliefs, PU and PEOU in TAM, are the primary drivers for explaining user acceptance of the specific type of system (^{Davis et al., 1989; Lee, Cheung, & Chen, 2005}). PU is defined as “the degree to which a person believes that using a particular system would enhance his/her job performance,” and PEOU is defined as “the degree to which a person believes that using a particular system would be free of physical and mental effort” (^{Davis, 1989}, p. 320). However, TAM provides little assistance in capturing the hedonic feature of an IS. Hence, ^{Van der Heijden (2004)} extended TAM by integrating perceived enjoyment (PE) to provide better explanations of hedonic motivations for IS acceptance in nonworking environments. PE refers to the degree to which the activity of using a particular system is perceived to be personally enjoyable in its own right apart from the instrumental value of the specific type of system (^{Davis, Bagozzi, & Warshaw, 1992; Lee et al., 2005}). According to Van der Heijden (2004), PU, PEOU, and PE can affect the usage intention of an IS, and PEOU determines PU and PE, which in turn lead to the usage intention of an IS.

IS quality factors and IS acceptance

The updated DeLone and McLean IS success model has received much attention from researchers in the IS domain; it indicates that the success in an IS can be evaluated in terms of system quality, information quality, and service quality, and these three types of IS quality factors can further affect subsequent use and user satisfaction (^{DeLone & McLean, 2003}). Essentially, these three types of IS quality factors, system quality, information quality, and service quality, are repeatedly mentioned as central to users’ IS acceptance (^{Kim et al., 2008; Roca, Chiu, & Martínez, 2006}). Hence, it is expected that a model can be proposed in this study by adopting the system quality, information quality, and service quality included in the updated IS success model proposed by ^{DeLone and McLean (2003)} as the external variables and integrating the four dimensions of PU, PEOU, PE, and intention to use referred in ^{Van der Heijden’s (2004)} model. Moreover, the quality of the user-interface design is also central in determining the level of users’ IS acceptance (^{Cyr, Head, & Ivanov, 2006}). Accordingly, four types of IS quality factors as external variables to PU, PEOU, and PE are respectively proposed and examined below.

Research Model and Hypotheses

Research model

Based on the updated DeLone and McLean IS success model and Van der Heijden’s model, this study’s research model presents four types of IS quality factors (including system quality, information quality, service quality, and user-interface design quality) that lead to nurses’ acceptance of the e-learning system. The research model used in this study is depicted in Figure 1. In an e-learning context, the specific elements of the research model and related hypotheses are further detailed below.

Figure 1: The research model.

User beliefs and intention to use

In the e-learning context, if nurses/nursing students believe that it is easy to use the e-learning system to learn online, they will be more likely to believe that advantages result from the e-learning system and further feel that the system is useful (^{Chen et al., 2008; Tung & Chang, 2008}). On the other hand, when learners perceive that the e-learning system is easier to use, the system may be less threatening to learners; thus, the e-learning system that is perceived as easier to use will be more likely to be perceived as enjoyable (^{Lee et al., 2005; Sánchez-Franco, Martínez-López, & Martín-Velicia, 2009}). As for user beliefs and usage intention, if nurses/nursing students feel that the e-learning system is useful and believe that using the system can improve their learning performance, they will be more likely to intend to use the system (^{Chen et al., 2008; Tung & Chang, 2008}). Further, nursing students who perceive that the e-learning system is easy to use for learning with the minimum waste of time and effort and their interaction with the system is relatively free of mental burden will be more likely to intend to use it (^{Tung & Chang, 2008}). Moreover, if learners who focus their attention on learning via the e-learning system can be in an enjoyable state, they will be intrinsically motivated to intend to adopt the system (^{Lee et al., 2009; Lee et al., 2005; Sánchez-Franco et al., 2009}). Hence, this study hypothesizes the following:

H1: PEOU will positively affect PU of the e-learning system.

H2: PEOU will positively affect PE of the e-learning system.

H3: PU will positively affect intention to use the e-learning system.

H4: PEOU will positively affect intention to use the e-learning system.

H5: PE will positively affect intention to use the e-learning system.

IS quality antecedents to user beliefs

To predict nurses’ acceptance of the e-learning system, system quality, information quality, service quality, and user-interface design quality as the antecedents to PU, PEOU, and PE are respectively proposed and inferred below.

System quality refers to the quality of the functionality of an IS itself (^{DeLone & McLean, 2003; Lin, 2007}), and it signifies the accuracy, convenience, efficiency, flexibility, reliability, and responsiveness in the function of an IS (^{DeLone & McLean, 2003; Kim et al., 2008; Lin, 2007}). ^{Pituch and Lee (2006)} found that system functionality, interactivity, and response could positively affect PU and PEOU of the e-learning system. ^{Cho, Cheng, and Lai (2009)} also showed that system functionality could positively affect PU of self-paced e-learning tools. Furthermore, if users are allowed to control access to Web site content and their needs can be responded to promptly, then these situations will be more likely to result in higher level of users’ enjoyment (^{Cyr, Head, & Ivanov, 2009}); thus, this study infers that system quality has a significantly positive impact on PE of the e-learning system. Hence, this study hypothesizes the following:

H6a: System quality will positively affect PU of the e-learning system.

H6b: System quality will positively affect PEOU of the e-learning system.

H6c: System quality will positively affect PE of the e-learning system.

Information quality refers to the quality of report contents and form that the IS generates; its measurement includes dimensions such as accuracy, completeness, currency, efficiency, relevance, scope, and timeliness of information (^{DeLone & McLean, 2003; Kim et al., 2008; Roca et al., 2006}). Essentially, if an e-learning system can provide learners with the new and updated course content and the design of online courses can meet the needs of learners at different levels, these will lead learners to feel that the e-learning system can be a useful means of learning (^{Lee et al., 2009; Lee, 2006; Liu, Chen, Sun, Wible, & Kuo, 2010}). Further, if the online course content can be arranged and integrated with good figures and clear text and the course design can provide learners with ease of access to the learning materials, it will be easy for learners to use the e-learning system (^{Liu et al., 2010}). Moreover, if learners consider the course content provided by the e-learning system as useful for fitting to their needs, this belief will facilitate their positive flow experience (i.e., an intrinsically enjoyable experience) about using the system (^{Choi, Kim, & Kim, 2007}). Hence, this study hypothesizes the following:

H7a: Information quality will positively affect PU of the e-learning system.

H7b: Information quality will positively affect PEOU of the e-learning system.

H7c: Information quality will positively affect PE of the e-learning system.

Service quality is defined as the degree to which a user perceives that the overall quality of services from an IS (^{Ahn et al., 2007; Kim et al., 2008}) and refers to the availability of multiple communication mechanisms for timely assisting users in solving the problems of IS usage (^{Ahn et al., 2007}). ^{Lee (2010)} proposed that students’ perception of online support service quality could be regarded as the key role in facilitating their behavioral intention toward e-learning acceptance and further indicated that perceived service quality had significantly positive impacts on PU and PEOU of e-learning systems. Furthermore, higher service quality can provide users with higher playfulness in online retailing (^{Ahn et al., 2007}); thus, this study infers that service quality has a significantly positive impact on PE of the e-learning system. Hence, this study hypothesizes the following:

H8a: Service quality will positively affect PU of the e-learning system.

H8b: Service quality will positively affect PEOU of the e-learning system.

H8c: Service quality will positively affect PE of the e-learning system.

User-interface design quality refers to the quality of the structural design of an interface that presents the features and instructional support of an IS (^{Cho et al., 2009}). Essentially, if the self-paced e-learning tool’s screen layouts have a good structure and its instructions are explicit, learners may perceive that such an e-learning tool is useful via its improved functionality, and they may navigate the content and find information in an easier way (^{Cho et al., 2009}). Practically, a user-friendly interface design of the online learning community will make users feel more comfortable and find it easier to use (^{Liu et al., 2010}). Moreover, when learners are comfortable with the use of learner interface of the e-learning system, they will be more likely to experience flow (^{Choi et al., 2007}). Hence, this study hypothesizes the following:

H9a: User-interface design quality will positively affect PU of the e-learning system.

H9b: User-interface design quality will positively affect PEOU of the e-learning system.

H9c: User-interface design quality will positively affect PE of the e-learning system.

Methods

Research Design and Sample

A cross-sectional design was used to investigate the effects of IS quality on nurses’ acceptance of the e-learning system. This study gathered sample data from nurses at hospitals in Taiwan. This study’s sampling frame was taken from among nurses working in hospitals with over 500 beds in Taiwan. Hospitals were selected by using two criteria: (a) the hospitals must have implemented the same e-learning system, the focus in this study is on the so-called learning management system (LMS); and (b) to ensure experience among nurses, the hospitals must have used the LMS for more than 1 year prior to the study period and nurses from these hospitals must have experience in using the LMS. Overall, two regional hospitals and one district hospital matched the selection criteria and agreed to participate in this study.

Measurement Tools

In this study, responses to the items in system quality, information quality, service quality, user-interface design quality, PU, PEOU, PE, and intention to use were measured on a 7-point Likert scale from 1 (strongly disagree) to 7 (strongly agree), with 4 labeled as neutral. Items chosen for the constructs in this study were adapted and revised from previous research. Further, the questionnaire was pretested on 39 nurses from one hospital in Taiwan in a voluntary and anonymous way. This selected hospital had implemented the e-learning system at least 1 year ago, and the nurses from this selected hospital were using the e-learning system in their learning. Nurses were asked to identify any ambiguities in the meanings, and the questionnaire was revised based on their comments. The instrument’s reliability was evaluated, and the Cronbach’s alpha values ranged from .70 to .96, indicating a satisfactory level of reliability exceeding that commonly required for exploratory research (^{Hair, Anderson, Tatham, & Black, 1998; Nunnally, 1978}). The nurses who had participated in the pretest were excluded from the final data collection and subsequent study. The final items are listed in Table 1 along with their sources.

TABLE 1: Construct Measurement and Sources

Data Collection and Ethical Considerations

The telephone calls were made to each hospital to explain the purpose of this study, and the three hospitals were asked to identify a contact person who could distribute the survey questionnaires to nurses who had experience in using the LMS. A total of 450 questionnaires were randomly distributed, and 340 nurses agreed to participate in this study. Before answering the questionnaire, these nurses were asked to read the questionnaire cover letter that explained this study’s purpose and informed the voluntary and anonymous nature for their responses and the right to withdraw from participation in this study. Essentially, return of the completed questionnaire implied consent to participate in this study. Finally, 340 (75.6%) questionnaires were returned. Of these received questionnaires, 20 were discarded due to large portions of missing answers. Consequently, 320 effective questionnaires were analyzed in this study, with an effective response rate of 71.1%.

Data Analysis

The data analysis process of this study followed a two-step approach for structural equation modeling method recommended by ^{Anderson and Gerbing (1988)}. In the first step, confirmatory factor analysis (CFA) was used to develop the measurement model. In the second step, to explore the causal relationships among all constructs, the structural model for the research model depicted in Figure 1 was tested by using structural equation modeling. The statistical analysis software packages used to perform these processes were AMOS 5.0 (SPSS, Inc., Chicago, IL, USA) and SPSS 8.0 (SPSS, Inc., Chicago, IL, USA). The two steps of the data analysis approach are described in more detail below.

Measurement model

To assess the measurement model, three types of analyses were conducted in this study. First, squared multiple correlation (SMC) for each item and composite reliability (CR) and average variance extracted (AVE) for each construct were used in this study to test the reliability of all constructs (^{Hair et al., 1998; Holmes-Smith, 2001; Nunnally, 1978}). Essentially, the SMC should be greater than .50 (^{Holmes-Smith, 2001}), and the CR and the AVE should be greater than .70 and .50, respectively (^{Hair et al., 1998; Holmes-Smith, 2001; Nunnally, 1978}). Moreover, the reliability coefficient for each construct assessed by the Cronbach’s alpha coefficient should be greater than .70 as recommended by ^{Hair et al. (1998)} and ^{Nunnally (1978)}. Second, according to the rule of ^{Anderson and Gerbing (1988)}, the t value of every item should be greater than 1.96 (p < .05), which indicates a good convergent validity level. Furthermore, based on the procedure described by ^{Fornell and Larcker (1981)}, the AVE of each construct should be greater than the squared correlation for each pair of constructs, indicating a good discriminant validity level. Third, the most common rules used in performing the CFA for measurement model include stipulating that the goodness-of-fit index (GFI) should be greater than .90, the adjusted GFI (AGFI) should be greater than .80, the incremental fix index (IFI) should be greater than .90, the Tucker-Lewis index (TLI) should be greater than .90, the comparative fit index (CFI) should be greater than .90, the root mean square error of approximation (RMSEA) should be less than .08, and the χ²/df should be less than 3 (^{Adams, Nelson, & Todd, 1992; Bagozzi & Yi, 1988; Hair et al., 1998}).

Structural model

The most common rules used in testing the structural model include stipulating that the GFI should be greater than .90, the AGFI should be greater than .80, the IFI should be greater than .90, the TLI should be greater than .90, the CFI should be greater than .90, the RMSEA should be less than .08, and the χ²/df should be less than 3 (^{Adams et al., 1992; Bagozzi & Yi, 1988; Hair et al., 1998}).

Results

Descriptive Characteristics of the Effective Respondents

A total of 320 effective questionnaires were analyzed in this study. All respondents in this study were women. Among them, 168 (52.5%) had less than 5 years of work experience, 86 (26.9%) had 5–10 years, 37 (11.6%) had 11–15 years, 17 (5.3%) had 16–20 years, and 12 (3.7%) had above 20 years. Most of the respondents (76.6%, n = 245) had graduated from university or above. In addition, the respondents identified themselves as licensed practical nurses (68.8%, n = 220), registered nurses (25.9%, n = 83), and head nurses (5.3%, n = 17). The descriptive characteristics of the effective respondents are depicted in Table 2.

TABLE 2: Descriptive Characteristics of the Effective Respondents (N = 320)

Results of Structural Modeling Analysis

Measurement model

To assess the measurement model, three types of analyses were conducted in this study. First, the results of the CFA show that the SMC values for all of the items were greater than .50, which indicates a good reliability level, and the values of CR and AVE for all of the constructs exceeded the minimum acceptable values of .70 and .50, respectively, indicating a good reliability level and subsequently yielding very consistent results. Hence, the results of the CFA indicate an acceptable level of reliability for all of the constructs. Furthermore, the reliability coefficients of all constructs assessed by the Cronbach’s alpha coefficient exceeded the .70 cutoff value. The results of the reliability test are shown in Table 3. Second, the results of the CFA show that the t value of every item exceeded the 1.96 value (p < .05), so the evidence of good convergent validity was obtained as the items represented their constructs significantly. The reports are listed in Table 3. Moreover, as to the test for discriminant validity, the results of the CFA show that the AVE of each construct was greater than the squared correlation for each pair of constructs, indicating that each construct is distinct (Tables 3 and 4). Third, the overall fit indices of the measurement model were χ² = 385.31, df = 296, χ²/df = 1.302, p < .001, GFI = .913, AGFI = .887, IFI = .996, TLI = .995, CFI = .996, and RMSEA = .038. The results of the CFA show that the indices were over their respective common acceptance levels. Thus, the proposed model generally fits the sample data reasonably well.

TABLE 3: Results of Confirmatory Factor Analysis, Validity Analysis, and Reliability Test

TABLE 4: Discriminant Validity of the Measurement Model

Structural model

The further step is to test the structural model for the research model depicted in Figure 1. The overall fit indices for the structural model were as follows: χ² = 420.60, df = 307, χ²/df = 1.370, p < .001, GFI = .902, AGFI = .878, IFI = .989, TLI = .987, CFI = .989, and RMSEA = .042. The results show that the overall fit indices for this structural model were quite acceptable. Essentially, the values of overall fit indices indicate that the model fits the data well.

Hypothesis testing

The properties of the causal paths, including standardized path coefficients (β) and t values, are shown in Figure 2. In regard to IS quality antecedents to nurse beliefs, system quality had significant effects on PU (β = .15, p < .05), PEOU (β = .32, p < .001), and PE (β = .22, p < .01); hence, H6a, H6b, and H6c are supported. Information quality had significant effects on PU (β = .13, p < .05), PEOU (β = .16, p < .05), and PE (β = .24, p < .001); hence, H7a, H7b, and H7c are supported. Service quality had significant effects on PU (β = .23, p < .001) and PEOU (β = .14, p < .05), but service quality had an insignificant effect on PE (β = .10, p > .05); hence, H8a and H8b are supported, but H8c is rejected. User-interface design quality had significant effects on PU (β = .30, p < .001), PEOU (β = .21, p < .01), and PE (β = .36, p < .001); hence, H9a, H9b, and H9c are supported. Moreover, as to the relationships between nurse beliefs and usage intention, first, PEOU had significant effects on PU (β = .39, p < .001) and PE (β = .17, p < .05); hence, H1 and H2 are supported. Next, the effects of PU (β = .43, p < .001), PEOU (β = .14, p < .05), and PE (β = .18, p < .01) on intention to use were significant; hence, H3, H4, and H5 are supported. Further, using the results above, the direct and indirect effects between the constructs are shown in Table 5. The results indicate that system quality, information quality, and user-interface design quality can indirectly make significant impacts on nurses’ usage intention of the e-learning system via their extrinsic motivators (i.e., PU and PEOU) and intrinsic motivator (i.e., PE), but service quality can indirectly make significant impacts on nurses’ usage intention of the system only via their extrinsic motivators (i.e., PU and PEOU).

TABLE 5: Direct and Indirect Effects Between the Constructs

Figure 2: Results of structural modeling analysis. Standardized path coefficients are reported (t values in parentheses). Absolute t value > 1.96, p < .05; absolute t value > 2.58, p < .01; absolute t value > 3.29, p < .001.

Discussion

The main purpose of this study was to examine whether system quality, information quality, service quality, and user-interface design quality as the antecedents to nurse beliefs can affect nurses’ intention to use the e-learning system. Obviously, this study can enhance the understanding of the impacts of IS quality antecedents to nurses’ beliefs on their usage intention of the e-learning system and thus provide medical institutions wishing to successfully facilitate nurses’ usage of the e-learning system with implications and suggestions.

IS Quality Antecedents to Nurses’ Beliefs

Four types of IS quality factors, system quality, information quality, service quality, and user-interface design quality, as antecedents to nurse beliefs have been examined in this study.

First, system quality was found to significantly affect PU, PEOU, and PE. These results are consistent with the findings of ^{Cho et al. (2009), Cyr et al. (2009)}, and ^{Pituch and Lee (2006)}. The results implicate that learners who are sensitive about system quality may usually think the e-learning system more useful, easier to use, and more enjoyable because of the functionality, such as controllability, flexibility, multimedia, interactivity, and responsiveness (^{Cyr et al., 2009; Pituch & Lee, 2006}). Hence, to make nurses’ learning via the e-learning system useful, easy to use, and enjoyable, medical institutions should be cautious with the compatibility between system features and nurses’ needs to enhance nurses’ acceptance of the e-learning system. Concretely speaking, instructors may make good use of the rich multimedia resources to facilitate learners’ understanding and absorption of the course contents (^{Lee et al., 2005; Lee, 2006}). Furthermore, system designers should install more interactivity-based mechanisms (e.g., a discussion room, instant messenger, or chat room) and responsive mechanisms (e.g., e-mail and message board) in the e-learning system, and instructors may further make use of these communication tools to respond timely to learners to make learners feel connected to others within an online environment (^{Lee et al., 2009; Lee et al., 2005; Pituch & Lee, 2006}). However, system designers should also take into account the bandwidth problems that deliver interactive services to ensure that access is not slowed down unacceptably by the increased interactivity (^{Chang & Wang, 2008}).

Second, information quality was found to significantly affect PU, PEOU, and PE. These results support the findings of ^{Choi et al. (2007), Lee et al. (2009), Lee (2006)}, and ^{Liu et al. (2010)}. The results implicate that when learners feel that the online course contents are more abundant and updated regularly and perceive that the design of the online courses is level appropriate and flexible, they will further find the e-learning system more useful and easier to use (^{Lee et al., 2009; Lee, 2006; Liu et al., 2010}), and such online course contents fitting to learners’ needs can also facilitate their enjoyable system usage (^{Choi et al., 2007}). That is, learner-centered programs should be developed to enhance learners’ interest in learning via the e-learning system (^{Lee et al., 2009; Liu et al., 2010}). Accordingly, instructors and course designers may design nurse-centered e-courses taking nurses’ learning needs and knowledge levels into account to make nurses’ learning via the e-learning system useful, easy to use, and enjoyable.

Third, service quality was found to significantly affect PU and PEOU. This result is consistent with the finding of ^{Lee (2010)}. The result implicates that it is important for learners to acquire adequate technical guidance and support regarding on how to take online courses effectively and efficiently (^{Lee, 2010}). Hence, medical institutions may provide help desk engineers and service administrators with adequate support services training for assisting nurses in taking online courses effectively and efficiently.

Fourth, user-interface design quality was found to significantly affect PU, PEOU, and PE. The results are consistent with the findings of ^{Cho et al. (2009), Choi et al. (2007)}, and ^{Liu et al. (2010)}. The results implicate that user-interface design quality is central in determining the level of PU, PEOU, and PE in relation to learners’ e-learning. Therefore, a friendlier user interface should be developed to provide nurses with a well-arranged, integrated, and clearly identified screen layout design to make them feel the e-learning system more effective navigation, easier to use, and more enjoyable. Specifically, system designers may design the user interface of the e-learning system in meaningful ways by having different contents separated with different image headers, decorative fonts, colors, and graphical buttons (^{Cho et al., 2009}); thus, such an elegant user interface will make nurses’ experience more useful and enjoyable. Moreover, a clear, straightforward, and accessible user interface should be designed in a way that it is easy for nurses to find the learning materials they require.

Synthetically speaking, this study’s results reveal that four types of quality factors, including system quality, information quality, service quality, and user-interface design quality, can make significant impacts on nurses’ beliefs in affecting their usage intention of the e-learning system. The results further implicate that, collectively, these four types of quality factors have greater explanatory power than does any single factor in describing the central quality antecedents of nurses’ acceptance of the e-learning system.

Nurses’ Beliefs and Their Intention to Use the E-Learning System

In this study, PU, PEOU, and PE were found to respectively exhibit positively strong impacts on nurses’ intention to use the e-learning system. The results further implicate that nurses intend to use the e-learning system mainly because they perceive the system to be useful to their learning and secondarily because the system is easy to use and enjoyable. Moreover, PEOU was also found to indirectly affect nurses’ intention to use the e-learning system via PU and PE. The result implicates that the e-learning system should be developed to deliver benefits and pleasure to learners without increasing the complexity within the e-learning process (^{Lee et al., 2005; Sánchez-Franco et al., 2009}). These findings are consistent with the views of ^{Chen et al. (2008), Lee et al. (2009}), ^{Lee et al. (2005}), ^{Sánchez-Franco et al. (2009)}, and ^{Tung and Chang (2008)}.

On the basis of TAM, ^{Chen et al. (2008)} and ^{Tung and Chang (2008)} have explained nurses’ acceptance of the e-learning system primarily via the extrinsic motivation, focusing on PU and PEOU; however, simply focusing on nurses’ extrinsic motivators of e-learning may not be enough. Virtually, it should be noted that whereas TAM is based on the assumption of individuals’ rationality in evaluating external stimuli, PE (i.e., intrinsic motivator) is focused more on individuals’ affective experience in explaining their nonbeneficial behavior of IS/IT acceptance and usage that is difficult to explain with only the view of TAM (^{Choi et al., 2007; Lee et al., 2005}). In this study, the empirical evidence on capturing both extrinsic motivators (i.e., PU and PEOU) and intrinsic motivator (i.e., PE) for completely explaining quality antecedents of nurses’ acceptance of the e-learning system is well documented. Specifically, the evidence implicates that nurses’ PE (i.e., intrinsic motivator) can generate affective arousal caused by external stimuli (i.e., IS quality factors) to facilitate their usage intention of the e-learning system.

Conclusions and Suggestions

This study proposed a well-rounded research model that is based on the updated IS success model proposed by ^{DeLone and McLean (2003)} and ^{Van der Heijden’s (2004)} model to investigate the impacts of IS quality antecedents on nurses’ acceptance of the e-learning system for a more robust analysis. Synthetically speaking, system quality, information quality, and user-interface design quality can indirectly make significant impacts on nurses’ usage intention of the e-learning system via their extrinsic motivators (i.e., PU and PEOU) and intrinsic motivator (i.e., PE), whereas service quality can indirectly make significant impacts on nurses’ usage intention of the e-learning system only via their extrinsic motivators (i.e., PU and PEOU). Essentially, user-interface design quality is the most key antecedent that can make significant impacts on nurses’ PU and PE. More efforts should be made to develop friendlier user interface via designing useful and enjoyable features to induce nurses to use the e-learning system. Moreover, system quality can make the greatest impact on nurses’ PEOU; thus, medical institutions should effectively enhance system quality to deliver benefits and pleasure to boost nurses’ usage intention of the e-learning system via reducing the complexity. Obviously, this study’s empirical evidence contributes significantly to the body of research on bridging the gap of limited evaluation for IS quality antecedents of nurses’ acceptance of the e-learning system, which is very rare in an intrinsic motivational view.

Several limitations should be noted in this study, and the following suggestions for further research will be worth future efforts in this field. First, this study’s data were collected from three hospitals in Taiwan only. Given this study’s limited scope, caution must be exercised in generalizing from this study’s sample to the respondents of other national cultural backgrounds. Further research may include the interactions of national culture as hypotheses in the research model. Next, respondents might usually display different serious reactions or relaxed feelings for the same e-learning system, depending on which situation they used the system. Future research in nurses’ acceptance of the e-learning system may test the various set of mechanisms that involve voluntary/mandatory usage settings. Finally, this study implemented a cross-sectional analysis and might not draw a complete picture of the course for nurses’ acceptance of the e-learning system over time. It may be desirable to understand the IS quality antecedents of nurses’ acceptance of the e-learning system and how these influences change over time with increasing experience in using the e-learning system. Further research may use a longitudinal analysis by taking into consideration the evolution for nurses’ acceptance of the e-learning system over time.