Looking Inside and Outside the System: Examining the Factors Influencing Distance Learners Satisfaction in Learning Management System

1College of Applied Studies and Community Service, Imam Abdurrahman Bin Faisal University, Al-Dammam, Saudi Arabia 2Department of Software Engineering, Faculty of Science and Information Technology, Al-Zaytoonah University of Jordan, Amman, Jordan Department of Computer Science, Faculty of Information Technology, Jerash University, 26150-311 Jerash, Jordan Department of Business Administration, College of Business Administration, Imam Abdurrahman Bin Faisal University, Al-Dammam, Saudi Arabia College of Science –Zulfi, Majmaah University, KSA Department of Mathematics, Faculty of Science, Cairo University, Egypt


Introduction
In the higher education, the use of learning management system in distance learning or online courses is widespread. Incorporation of LMS into teaching and learning practices was increased in the sectors of higher education (Gautreau, 2011). The successful implementation of LMS can be seen by the utilization of LMS by the end users (Rafi et al., 2015). End user satisfaction nowadays is an essential topic because of the increasing number of institutions that utilize LMS in online courses which resulted in a need for an evaluation method to measure the effectiveness of LMS. Every organization aims to satisfy their customers or users, especially the universities. Using LMS in delivering courses is considered as an important goal for the universities to manage and assess the work of their students. The concept of user satisfaction is known as the extent to which the stakeholders believe that the technology they are using has met their needs (Ives et al., 1983). It is considered as the main variable that shows the different in success in the 454 marketplace (Gitman and McDaniel, 2008). Furthermore, analysing user satisfaction is beneficial for improvement of the product (Li et al., 2010).
A number of researchers believe that if the Information System (IS) meets the needs of the students, the students' satisfaction will increase (Cyert and March, 1992). Conversely, if the Information System (IS) does not provide the required information, that will lead to dissatisfaction among users (Bergersen, 2004). Moreover, specifically relating to LMS, most previous studies evaluated only one type of LMS; researchers recommended to conduct further researches on multiple LMS programs to gain more generalized and valid results (Hilmi et al., 2012;Almarashdeh et al., 2013;Khalid et al., 2015). Hence, a comparative analysis of students' perceptions in using LMS is needed, as several researchers have recommended that more research should be conducted in order to better understand the distance learning courses and its impact on students' satisfaction (Sher, 2009).
The target of this study is the biggest shareholders of the educational system, the students. The level of students' satisfaction with the learning system needs to be investigated, in the other hand the satisfaction level depends on several factors that affects the users such as what service does students' needs? Or what the LMS provides? Is what students needs available in the LMS? Is the LMS user friendly or complicated? Therefore, this study aims to examine the factors influencing students' satisfaction with learning management systems in distance learning courses.

Related Work
Until today, there has been limited complete evaluation model to assess the effectiveness of LMS. The latest published information in user satisfaction considered various factors such as: The content currency, content completeness, ease of access, ease of navigation, as well as course staff responsiveness as being critical success factors of student satisfaction recorded when using LMS (Naveh et al., 2012;Almarashdeh, 2016). The authors did not take into account the perceived usefulness, system quality and service quality as the main factors that might affect a student's level of satisfaction. On the other hand, other researchers claimed that student satisfaction is influenced by frequency of use and service quality (Green et al., 2012). Another study claimed that accessibility problems, poor technical support and less familiarity with technology might frustrate the students' satisfaction and cause high levels of frustration with online courses (Naaj et al., 2012). The loyalty of user to service enterprises is directly affected by User satisfaction. In a service context, quality and value are proposed to be satisfaction antecedents; satisfaction mediates the influence of quality and value on loyalty (Llewellyn, 2011;Almarashdeh et al., 2010).
A more integrated view of IS and a more formulated IS success model was initiated by DeLone and McLean (1992), in which information quality and system quality are depicted as influencing factors for IS use and user satisfaction. A study of IS success dimensions suggests that technology qualities such as system, service and information quality have influenced perceived usefulness, user satisfaction and system usage (Sabherwal et al., 2006). Furthermore, a model developed by Wixom and Todd (2005) (based on user satisfaction and technology acceptance literature) was tested on a sample of 465 participants from seven organizations while considering their utilization of software for data warehousing. The finding of the study showed a significant effect of information and system quality on perceived ease of use and perceived usefulness. In addition, the study suggested more investigation to be carried out on the effects of the information technology design and development as a predictor to usefulness, ease of use and related variables (Wixom and Todd, 2005). In LMS, a many researchers choose different variables to expect satisfaction of user in terms of: Culture influence, capacity of use and technology adoption.
A recent study in user satisfaction of LMS used independent variables such as: Course discipline, course size, course year, staff size, instructor status, content posted on the online course and support of surveys and forums in online courses to predict student satisfaction (Naveh et al., 2010). Furthermore, recent researchers on the learning management system perception did not indicate the significance of system abilities; like: functionality and availability of the learning management system; the quality of service; like: A follow up service, empathy and help desk, in addition to quality of information that is important for students for obtaining valued information from the learning management system by guaranteeing: Accuracy of information, relevancy, legibility, reliability, availability, consistency, timeliness and completeness. These issues are important for the distance learning programs to be successful. The factors of technology can give sureness to students for online studying if the learning management system has suitable study environment, having ease of use and having better features will facilitate the learning process.

Method of this Research
The research model and hypotheses are outlined in this section.

Research Model
We built the proposed model based the information system success model and Technology Acceptance Model (TAM) (Almarashdeh et al., 2010a;2011a). Perceived ease of use and usefulness has been used as the mean variable in technology acceptance. However, the TAM did not measure overall service and information quality even though it has a strong effect on behaviour intention based on literature studies (Wixom and Todd, 2005). The IS success model did not provide a sufficiently clear intention regarding usefulness and ease of use of the new technology, even though prior studies found significant results (such as of usefulness and ease of use regarding user acceptance of new technology and user satisfaction) (Almarashdeh et al., 2010b;Rai et al., 2002).
This work shows the importance of the learning management system success measuring based on the perspective of IS. Particularly, the preceding factors are collected from the persistent development and improvement in, developing, designing and delivering the essential facilities for distance learning users. Figure 1 describes the characteristic of the User Satisfaction Evaluation Model (USEM) that is proposed in this research.

Research Hypothesis and Model Constructs
According to (Chen et al., 2009), educational technology quality is an important factor influencing usage and satisfaction of the learning system. Further, they have suggested further studies to be conducted in order to explore the impact of educational technology quality on learners' behavior. User satisfaction is seen to be influenced by different factors such as: User expectation; attitude and system output; perceived usefulness and ease of use; task difficulty and system type. Moreover, system effort and effectiveness influence user satisfaction (AlMaskari and Sanderson, 2010). Also, technological issues (System quality) have a major impact on user satisfaction (Almarashdeh and Alsmadi, 2017;MOHE, 2006). Some other researchers suggested that perceived ease of use, perceived service quality and perceived usefulness affect user satisfaction (Adewumi, 2013;Almarashdeh et al., 2010c). At the same time, others used quality dimensions such as: service quality, system quality and information quality to measure user satisfaction (Delone and McLean, 2003). On the other hand, developers were concerned with: content accuracy, ease of use, format and timeliness, speed and efficiency, language and learnability, documentation, motivation and job satisfaction, as well as aesthetics and enjoyment by which to measure user satisfaction (CSU, 2005). Although satisfaction of user was highly affected by many factors such as perceived benefit and expectations, but the ease of use hasn't been considered as one of the most significant factors (Adam, 2000). In different contexts, many researchers utilized diverse factors to measure user satisfaction of the LMS. These included: ease of use, content and systems integration, ease of ownership and support and vendor service (Ibrahim and Silong, 1997). In this study, we used: Service quality, system quality and information quality, perceived ease of use and perceived usefulness as the main measurements of student satisfaction when using LMS in distance learning courses. Figure 1 above represents the theoretical model for this study. The model indicates the relationship between: Perceived ease of use, perceived usefulness, system quality, service quality and quality of information and student satisfaction. The model that is proposed in this work is called User Satisfaction Evaluation Model. The following describes the model constructs and related hypothesis.

Service Quality (SVQ)
There is still no published information on the purpose of measuring quality of services in the online distancelearning context (Shaik et al., 2006). Generally, service quality represents the quality of support services provided to end users. The service quality measurements include: Reliability, tangibles, responsiveness, assurance and empathy of the system (Kettinger and Lee, 1997;Al-Busaidi and Al-Shihi, 2010). In terms of online courses, common measures of service quality are responsiveness, reliability and empathy, with all having significant influence on student satisfaction (Roca et al., 2006). We believe that the better service provided to the students the higher satisfaction level. To test this assumption, the following hypothesis is proposed.
H1: Service quality is positively linked to user satisfaction

Information Quality (InQ)
In general, information quality has been associated with measurements such as: accuracy, currency, precision, output timeliness, conciseness, completeness, format, relevance and reliability (Bailey and Pearson, 1983). In the context of online learning, information quality represents the perceived output produced by LMS. Commonly used indicators including: Sufficiency, completeness, accuracy, timeliness, relevance, format, accessibility and understandability are utilized to predict information quality (Saba, 2012;Al-Busaidi and Al-Shihi, 2010). Literature studies on user satisfaction have claimed that InQ has significant impact on satisfaction of user (Delone and McLean, 2003;Seddon, 1997).

H2: Information quality is positively related to user satisfaction
System Quality (SyQ) Researchers in Information Technology (IT) have commonly used measurements such as: Accessibility, language, timeliness, efficiency, flexibility and integration factors in order to predict system quality (Wixom and Todd, 2005). In online learning systems, such as LMS, system quality equates to a student's observation of a system's performance. This perception can be measured by some factors which are: Availability, usability, ease of learning, realization of user expectations and response time (Freeze et al., 2010). Previous researchers have indicated that system quality directly influences student satisfaction (Almarashdeh, 2016;Ramayah and Leeb, 2012).
H3: System quality is positively related to user satisfaction

Perceived Ease of Use (PEU)
The Technology Acceptance Model (TAM) proposed PEU, which is concerned with the expectation of user that utilizing a target system is effort-free (Davis et al., 1989;Almarashdeh and Alsmadi, 2016). Perceived Ease of Use refers to qualities such as: Being user friendly, ease of use and ease of learning (Wixom and Todd, 2005). According to Ba and Johansson (2008) study on user satisfaction, PEU has effect the overall user satisfaction (Almarashdeh et al., 2010b).

Perceived Usefulness (PU)
Perceived usefulness is the expectation of users that utilizing the proposed system can increase her or his work performance (Davis et al., 1989). The usefulness and relevancy factors are the main measurements of the students' perception (Wixom and Todd, 2005). Several studies suggest that PU has an important influence on student satisfaction (Ong et al., 2009).

User Satisfaction (SAT)
SAT is the potential influence of the LMS and the total assessment of the experience of users when utilizing the LMS. It is defined by two different variables; one of which is expectation from LMS and the other one is confirmation of these expectations (Ozkan et al., 2008). The user satisfaction influences the benefit or the net outcomes of the LM (Saba, 2012;Almarashdeh et al., 2011b).

Net Benefit (NB)
NB shows the balance of positive and negative effects upon a user. Therefore, NB can be measure by: Efficiency, job effectiveness, effects, as well as decision quality and error reduction. The net benefit, influenced by many factors, resulted from a number of studies which used user satisfaction as the main factor to predict system outcomes (Delone and McLean, 2003;Almarashdeh et al., 2010a;Lee-Post, 2009).

Data Collection
The population and sample were recognized and then the study questionnaire was designed depending on the requirements of study and from literature reviews. The questionnaire has been lunched online in April 2017 for 2 weeks after the pre-testing phase, were the pilot data was collected and then analyzed, with findings showing that the instrument was reliable to use. Furthermore, the suggestions and comments of the participant were taken into consideration. Hence, two experts in the field reviewed and revised the questionnaire. The final questionnaire consisted of 33 questions related to the used measurements including: System quality (9 questions), information quality (6 questions), service quality (5 questions (Hair et al., 2006). A 5 point-Likert-scale, which is ranging from answers of strongly agree to strongly disagree, was used to measure the items in the questionnaire.

Data Analysis
The Statistical Package for the Social Sciences (SPSS) and Structural Equation Model (SEM) were used to analyze the data. The first step of analysis of data is the reliability testing. The reliability testing (coefficient alpha) range of 0.80 or 0.90 is an indication for a well-constructed scale (Sekaran, 2003). The conducted questionnaire reliability testing value is 0.971, which is an indication that the scale is well-constructed. Figure 2 shows the values of each construct reliability testing result.

Demographic Information
The background information of the samples indicates that the majority of the students are aged between 20 to 40 years old. The majority of the students hold bachelor degrees (52.9%) and 23.8% are diploma holders. The highest percentage of the students have used LMS for a period of 1 to 3 years (34.8 %), while 30.1 % have used the LMS for between 3 and 7 years. Figures 3, 4 and 5 represent the demographic information obtained.

Descriptive Statistics
In data analysis, the reason of using descriptive statistics is to use a simple summary instead of the large amount of data. Descriptive statistics refer to a way of transforming data into graphic and numerical procedure, it will provide easy and clear method for person who reads to understand and interpret (Podsakoff et al., 2003). Frequency distributions and descriptive statistics were utilized to provide total view and represent the properties of the data that was collected. Descriptive statistics comprising standard deviation and means were obtained for the scaled variables, as presented in Table 1.

459
The result of descriptive statistics indicates that the mean is above 3.68 for all factors, with the respective means for all factors being the same. Therefore, the results were close around the mean, which indicates that opinions of students were similar. In this work, the values of standard deviations are between 0.79 and 1.00, which means that there were small variations in the opinions of students.

Measure of Fit
The evaluation of model is the most challenging problem regarding SEM (Arbuckle, 2005). Before analyzing the structural model, it is important to first know how to evaluate it. The first step of building a structural model is achieved by using experimental knowledge and research of the theory to determine the association between the perceived variables and after that utilizing a statistical method to assess the theory. The CFA testing is affected by the requirements of: multivariate normality, enough sample size, measurement instruments, model fit indices interpretation, the tested research hypothesis, outliers, parameter identification, data that are missing (Schumacker and Lomax, 2004).
Since previous research studies suggested that minimum 10 participants for each free parameter should be estimated as the recommended number (Schreiber et al., 2006), this study used 23 parameters, with the sample size being 425. The chi-square statistic (x 2 ) is the most commonly described fit index in the field of structural equation modeling (Davey and Savla, 2009). The chisquare value determines how much is the data mismatched with the theories. Therefore, the CMIN connected with the higher probability value (P), the nearer the fitting between the perfect fit and hypothesis (Byrne, 2009;Arbuckle, 2009). In the proposed model, the chi-square value is 11.017 and the P value 0.051, which indicates that the proposed model is right and does not differ from the other models. By measuring the total fit of the model, the results indicate that the Chisq/df is 2.203 at 5 degree of freedom (df), which shows that the model had a perfect fit. The User Satisfaction Evaluation Model fit measures can be seen as summarized in Table 2.
The results in the Table 2 conclude the fit measures of the research model which indicate that the proposed model has a clear fit with the data. The advantages of fit measures illustrated in this study mostly obey the earlier research recommendations. The test results indicate that the hypothesis of the model is accepted and predicts the proposed model very well. Figure 6 describes the standardized regression weights of the USEM hypothesis and the correlation between the independent variables as the SEM modification index recommended those variables to be correlated.

Hypotheses Testing
As shown in Fig. 6, all independent variables are associated with each other. The highest correlation is between PEU and PU (0.79), while the lowest is the correlation between PEU and SyQ (0.49). Furthermore, the figure represents the relationship between independent and dependent variables. All model hypotheses are accepted and show significance influence. Table 3 provides more details on the tested hypotheses.
In the proposed model, all of the pathways were significant. The C.R (critical ratio) illustrates the lowest C.R were between SAT and PEU with value of 2.089 for H1, with the highest being H6 between SAT and NB (12.3). The C.R indicates that the user satisfaction is more affected by system quality than the service quality.

Discussion
Measuring the importance of factors contributing to the success of LMS in distance learning programs is the main concern of this study. The implications of the study indicated that there are strong significant influences among all independent and dependent variables. Using SEM to validate the USEM shows a valid result with CMIN/DF value 2.203, which indicates a very good fit (Byrne, 2009;Arbuckle, 2009). Hence, SEM has been used in this study for testing the hypothesis of the study using regression analysis. Based on the data analysis results, the highest impact on student satisfaction was system quality with value C.R 6.68 (indicating a strongly significant effect). This finding is consistent with blogbased learning systems success study of Yi-Shun et al. (2014).This indication represents the importance of: Availability, usability, ease of learning, realization of user expectations and response time factors (Freeze et al., 2010). These system quality factors represent the main predictor of student satisfaction.
Several studies suggest that Perceived usefulness has a major effect on student satisfaction (Ong et al., 2009). Perceived usefulness takes second place (C.R 5.74) in influencing student satisfaction. If the LMS is found to be useful, student satisfaction will increase. The third value affecting student satisfaction is based on service quality, which has a strong effect on student satisfaction (C.R 5.61). This implies that if the LMS provides good quality service (such as available support on a "24/7" basis), useful services will be ready for students' use, as well as a high level of training and thus, student satisfaction will be increased (Hilmi et al., 2012). Furthermore, the information quality seems to have the same effect on student satisfaction as service quality. The result shows that information quality has a strong influence on student satisfaction (C.R 5.46). This indication confirmed that quality of information including: Relevance, accuracy, timeliness, completeness, sufficiency, understandability, accessibility and format (Saba, 2012;Al-Busaidi and Al-Shihi, 2010) are significantly important to the successful use of LMS in online courses. This, however, is not in line with the findings of a study by Green et al. (2012) who found that the delivery format has no significant influence on student satisfaction (Green et al., 2012).
Four hypotheses (H1, H2, H3 and H4) have a similar value (C.R more than 5.4) in affecting students' satisfaction. However, H5 (which measures the influence of PEU on student satisfaction) is lowest among the others. The ease of use factor represents the simplicity of LMS and it has a significant effect on student satisfaction with C.R 2.09. This indication comes as a response from which previous researchers claim there is a need for understanding the PEU of LMS from the students' perception. However, the indication of the strong significant influence between PEU and student satisfaction is not in line with Adam (2000), who claims that PEU was not deemed to be among the Students will tend to use LMS more if it is easy to use and consequently will boost learning achievements (Hilmi et al., 2012).
User satisfaction plays an important role in predicting system outcomes (Delone and McLean, 2003;Lee-Post, 2009). Related studies suggest that student satisfaction influences the outcomes or the net benefit of using the LMS (Saba, 2012). By testing H6 (which represents the relationship between student satisfaction and net benefit), the results (C.R 12.3) indicate that student satisfaction has a very strong effect on the outcomes or net benefit acquired from using the LMS. This effect shows that the more students are satisfied with using LMS in their learning process, the greater the benefit that will be gained from that use. It is also anticipated that there would also be an increase in their proclivity.
As this implies, a lesser effort in using and understanding LMS affects student satisfaction, but not as LMS quality, which affects student satisfaction three times (C.R 6.68) more than the perceived ease of use (CR 2.09). Thus, universities should consider looking into the LMS quality, services quality and information quality more than the simplicity or the price of such system. The students satisfaction can be effected negatively if the LMS breakdown or if there is no support when they need it or no value of the available services or information the students. In general, the current LMS analysis considers the success dimensions as being related to: Information, service and system quality; usefulness and perceived ease of use; user satisfaction and net benefit. The factors of SYQ, INQ and SVQ, PU and PEU are considered to be factors that influence user satisfaction, which is the mediating factor for net benefit. The contribution of this work to the studies of user satisfaction is adoption of the most significant factors which have been seen to affect user satisfactions from past researches. Analyzing the research model by the SEM indicates that the relationships within the proposed model are assumed to measure the effect and cause within success factors, in addition to success measures. The User Satisfaction Evaluation Model is assumed to influence (negatively or positively) the LMS quality and so, to affect net benefit and user satisfaction.
As a direction for further research, this study was limited to students in distance learning courses; a larger sample size would have been better to validate the research model. Furthermore, as different users have different perceptions, measuring students' satisfaction in a different context might provide valid results. As the learning process is a complete sharing process between two players (instructors and students), future research in instructor satisfaction with using LMS might provide valuable results related to the success of distance learning courses. This research is limited to a few factors that affect student satisfaction. As to other factors (such as training and user experience of using LMS), we believe it might influence user satisfaction as a mediator between factors such as: Perceived usefulness, perceived ease of use, service quality, information quality, system quality and particularly, end user satisfaction in online learning.

Conclusion
This LMS evaluation is vital to ensure the positive impact and effective implementation on distance learning courses. Measuring the satisfaction of distance learners is very important for construction of a perfect distance learning platform for educational purposes. The USEM has been proposed based on a comparison between the IS success model and TAM, by choosing important factors related to user satisfaction in online courses. The USEM identified critical factors influencing the satisfaction of distance learners as: Five independent variables (information quality, service quality, perceived usefulness (and perceived ease of use), system quality as well as two dependent variables (net benefit and user satisfaction). In this research, a survey was carried out to obtain relevant data from distance learners related to their perceptions of the impact of the LMS in terms of their benefits and satisfaction level. The results indicate that all hypotheses are accepted, while the regression analysis shows the significant influence on user satisfaction by system quality, service quality, perceived usefulness, perceived ease of use and information quality. However, the strongest significant effect on students' satisfaction comes from system quality, while a lower significant effect comes from perceived ease of use. Based on these findings, it can be concluded while an LMS need not be too easy to use, but the LMS nevertheless needs to provide better services, high availability, usability, accessibility, better response time and providing useful features. As using LMS in 462 universities becomes more and more widespread, the quality of LMS needs to be monitored. In the end, we hope future researches and developments can take into account lecturers' opinion when developing the LMS.

Acknowledgement
We are grateful to all participants whom shared their thoughts with us. This study did not have any financial support. All the work was done and all the expenses were paid for by the researcher.