Establishing Students’ Satisfaction with a Learning Management System Using a Modified DeLone and McLean Model: A South African Sample Perspective

Abstract

Students’ use of LMSs in higher education institutions can be severely hampered by several factors that could lower their satisfaction. Good LMS service quality will increase student LMS satisfaction. Student LMS dissatisfaction will increase if the expectations are not fulfilled. The purpose of this study was to establish the factors influencing students’ satisfaction with IMFUNDO, the pseudonym for the LMS used at the University of Technology. This study was motivated by the literature that recommended further research on students’ LMS satisfaction. The quantitative method was used to attain the objective of testing the variables affecting students’ LMS satisfaction and validating the scientific model and hypotheses. The participants comprised 595 students from the Faculty of Science at the University of Technology in South Africa, who were selected using stratified random sampling. Data were gathered through student LMS satisfaction surveys. Data were analyzed using SPSS and AMOS version 29 software inferential statistics for validating CFA and SEM. The results revealed that the alpha values for the entire questionnaire were 0.96. The results showed that the chi-square (χ²) was statistically significant [χ² = 743,52, df = 160, p < 0.0001]. The goodness of fit was TLI = 0.930, the CFI was 0.941, the RMSEA was 0.078, and all indicators were statistically significant (p < 0.0001). Using the conceptual framework that was grounded on the modified DeLone and McLean model was beneficial for the students at the University of Technology in South Africa. A mixed-method approach should be utilized to support the statistical findings with the participants’ opinions on this phenomenon.

1. Introduction

Higher education learning and teaching procedures are greatly impacted by learning management systems, which are widely acknowledged as a hub for innovation (Rulinawaty et al., 2024). Furthermore, higher education makes extensive use of online course management systems, and learning management systems (LMSs) have been the primary platforms for e-learning delivery (Simelane-Mnisi, 2023b). This is because LMSs provide an integrated system for organizing course materials, fostering collaboration, and monitoring student progress (Aldila et al., 2024). This substantial rise in the use of LMSs in higher education can be attributed to the variety of learning and teaching opportunities that are accessible to both students and academics (Pham et al., 2019). This implies that LMSs are crucial for enhancing learning activities and developing innovating learning services to produce the greatest teaching resources and learning content (Rulinawaty et al., 2024).

However, students’ use of LMSs in higher education institutions can be severely hampered by several factors that can lower their satisfaction with the technology. These challenges relate to technology (unstable internet connection, poor network connectivity, inadequate electrical supply, lack of technical skills and infrastructure support, inability to download learning materials on the LMS), lack of motivation, lack of face-to-face interaction, and lack of time to attend online classes (Alumona & Akinseinde, 2023; Mahama et al., 2024). In this case, (Nguyen, 2021; Sumi & Kabir, 2021) argued that good LMS service quality will increase student LMS satisfaction. This implies that student LMS satisfaction will be achieved if the expectations of the students are fulfilled. Student LMS dissatisfaction will increase if expectations are not fulfilled.

Regardless of the extensive usage of LMS systems in higher education, it is still necessary to ensure that these platforms satisfy the requirements and expectations of their users (Mizan et al., 2023). Student satisfaction is referred to as all of the benefits that users receive from utilizing an LMS, and it can be used to provide a measure of students’ overall behavioral beliefs and attitudes (Nguyen, 2021). To achieve this, institutions may customize their online offerings to leverage the information provided by identifying the factors that have the greatest impact on students’ satisfaction (George, 2024).

The LMS known as IMFUNDO (pseudonym for the LMS used at the University of Technology) for blended and online learning was used in this study (Simelane-Mnisi, 2022b). The University of Technology (UoT) has implemented IMFUNDO to enhance students’ academic competencies, cognitive attributes, preferences for learning, performance, outcomes, and skills. This was accomplished using the LMS template/structure, which emphasizes student engagement, and designing each learning unit’s content and activities, which were scaffolded and chunked (Simelane-Mnisi, 2022b). In this case, (Muhammad et al., 2022) argued that students require a well-planned, informative, and simple-to-understand online course with explicit assessment and evaluation procedures. Often, this is obtained from a well-designed and developed structured online course. Another study (Cho et al., 2021) supported the use of an LMS template or structure by demonstrating the effectiveness of the LMS common template/structure, with 100% teacher usage and student involvement. The IMFUNDO modules were designed based on the different types of modules offered in the Faculty of Science related to year/semester theory module, year/semester practical module, block, problem-based learning, and work-integrated learning. In this regard, the IMFUNDO modules were useful assistants to academics and students because of optimal functions, such as the course home page, which consisted of a welcome message and introduction, learning content, assessments, practical component widgets, and academic user profiles.

The welcome and introduction widgets comprised module information relating to the welcome statement in video and text format; academics’ images, details, and consultation times; study guides; prescribed and recommended resources; timetables; and general information about the institution, resources, and relevant policies. In this element, each academic could establish their personal identity and teaching presence in an online environment. In this section, the academics ensured that their personas and teaching identities were visible in the online courses (Simelane-Mnisi, 2022a; Simelane-Mnisi & Mokgala-Fleischmann, 2022).

Academics constructed learning content widgets; for example, they segmented the learning materials and activities into units in this way. It was anticipated that this would reduce the cognitive load and encourage student participation and engagement, especially during independent study. The learning units, topics, descriptions, and learning outcomes, as well as the assessment criteria, were utilized to promote constructive alignment in the online environment. The unit/chapter/topic short introduction and outline of the learning materials were created using HTML documents for learning materials, such as notes in PDF, PPT, PPT with audio, etc. Clear instructions were provided to the students. Links to live classes with MS Teams were also made available. The chunking and scaffolding of learning materials and activities supported student-centered learning, encouraged participation and engagement, and prevented distractions and boredom (Simelane-Mnisi & Mokgala-Fleischmann, 2022).

Communication and collaboration tools involved discussion forums that were also used for feedback and reflection, announcements, chats (WhatsApp class group), email, calendars (Simelane-Mnisi, 2023b), assessment, video conferencing, monitoring, tracking, and surveys. (Simelane-Mnisi, 2023b) argued that interactive learning tools for communication and collaboration could promote social engagement in the LMS and motivate students to learn actively.

Assessment widgets included IMFUNDO tools that fostered engagement with quizzes, assignments, discussions, gradebooks, and Microsoft forms. The assessments could be used in the LMS to help students demonstrate cognitive engagement and become more interested in the subject (Simelane-Mnisi, 2023b). Gradebooks contributed to an increase in student engagement because students needed to be constantly informed on the outcomes of their online behavior to monitor their academic success on the LMS (Simelane-Mnisi, 2022a).

Practical component widgets comprised a brief practical overview, HTML documents for experimental materials [notes in PDF, PPT, PPT with audio, etc.], gradebook, group, and quiz tools that included pre- and post-laboratory tests. Most Faculty of Science modules included a practical component requiring students to demonstrate laboratory skills (Simelane-Mnisi, 2022a).

The purpose of this study was to establish the factors influencing students’ satisfaction with the IMFUNDO LMS. This study is significant because it was motivated by the (Nguyen, 2021) recommendations that more research should be performed on students’ satisfaction with the quality of LMSs. Furthermore, (Pham et al., 2019) recommended that future studies should examine the relative significance of the characteristics that constitute the overall quality of e-learning services in developed and developing nations. The researchers did not come across a national study that investigated this phenomenon in the context of the UoT. For these reasons, this study was conducted and grounded in the modified DeLone and McLean information system (IS) success model (DeLone & McLean, 2016). This study aims to contribute to the body of knowledge about students’ satisfaction with using LMSs in higher education. (Elmunsyah et al., 2023) emphasized that LMSs may be assessed alongside information system success and quality. The researchers utilized this model because several empirical studies employed the D&M ISS model, which demonstrated the reliable and valid constructs of the relationship indicator of information systems success concerning LMSs in the higher education context (Asakdiyah et al., 2024; Muhammad et al., 2022; Nguyen, 2021; Rulinawaty et al., 2024; Sumi & Kabir, 2021). The modified DeLone and McLean (D&M) model is one option for measuring students’ LMS satisfaction. In this study, a conceptual framework was developed and hypothesized to attain thorough knowledge about students’ LMS satisfaction by defining, characterizing, and explaining the links between the five primary variables with the three main dimensions of the updated DeLone and McLean model (DeLone & McLean, 2003).

This study used the IS success model (system quality, information quality, and service quality). The five major dimensions include students’ LMS satisfaction, service quality, information quality, convenience, and system quality.

2. Hypotheses Development

2.1. Service Quality

Service quality is the level of assistance that users of the system receive from the IT support staff and the information systems organization (Adayemi et al., 2024). In the service sector, the outcome of the service delivery system is typically used to evaluate the quality of services (Sumi & Kabir, 2021). In fact, customers’ perceptions are connected to their attitudes and behavioral intentions. Furthermore, service quality in higher education refers to the gap between students’ expectations and experiences with higher education services (Pham et al., 2019), which implies that service quality affects satisfaction. Moreover, service quality in the context of online learning, where LMSs are often utilized, refers to the distinction between service expectations and students’ perceived experiences (Sumi & Kabir, 2021). This implies that higher education institutions must improve the quality of online services including the LMS provided to students as part of the blended learning process (Asakdiyah et al., 2024). Considering the current increase in e-learning and LMSs, a metric measurement of service quality is required to determine how students actually perceive it (Sumi & Kabir, 2021). Furthermore, analyzing how students perceive the quality of services they receive online allows for the evaluation of LMS service quality (Asakdiyah et al., 2024). This study determined students’ LMS satisfaction with service quality by evaluating aspects relating to training, the LMS support service desk, and the department LMS influencers. This study proposes the following hypothesis considering the above discussion.

H₁: 

Students’ LMS satisfaction is positively related to the service quality in South Africa.

2.2. Information Quality

Information quality is defined as the measure of output quality from the information system (Rasheed & Rashid, 2024). Information quality captures e-commerce content or web content that is essential for potential customers or suppliers to make purchases online and visit the website frequently (DeLone & McLean, 2003). This is attained by ensuring that the information on the web is personalized, comprehensive, pertinent, user-friendly, and secure. Measures of information quality included timeliness, accuracy, understandability, completeness, relevance, usability of the generated information, and consistency (Adayemi et al., 2024; DeLone & McLean, 2003). Information quality on the IMFUNDO LMS refers to the content and learning activities uploaded by the academics that permit the students to engage actively, interact, and participate in an online environment. In this study, information quality was measured by the provision of accurate information, sufficient information to perform tasks, precise information, and updated information. This study proposes the following hypothesis considering the discussion above.

H₂: 

Students’ LMS satisfaction is positively related to information quality in South Africa.

2.3. Convenience Quality

Convenience quality is determined by the amount of time needed to find and use any materials or information on an LMS. Shorter materials are preferable since they show that all LMS features are simple to use and save users time when searching or asking questions. In this case, a technology would be viewed as being extremely user-friendly if the less challenging the technology is, the more time the user could devote to learning (Venkatesh et al., 2003). This suggests that students spend more time learning when the IMFUNDO LMS is less challenging and easier to use. In this study, the convenience of the LMS was measured when searching for information or content on the LMS without time constraints, the effort made when performing activities, affording learners an opportunity to improve learning outcomes, and the convenience of accessing and using the LMS. This study proposes the following hypothesis considering the discussion above.

H₃: 

Students’ LMS satisfaction is positively related to convenience quality in South Africa.

2.4. System Quality

System quality is one of the most important factors determining users’ satisfaction with an LMS (Mizan et al., 2023). In this case, (DeLone & McLean, 2003) argued that to measure the quality of a system’s ease of use, usefulness, dependability, flexibility, data quality, portability, integration, and importance should all be utilized. Furthermore, system quality describes an LMS’s technical features, such as its responsiveness, reliability, and user-friendliness (Mizan et al., 2023). LMS system quality is referred to as a reliable, compatible, and stable system that integrates both software and hardware (Nguyen, 2021). This study focused on the IMFUNDO LMS as it was imperative that the LMS system quality be measured. In this case, the system quality measured usefulness, flexibility, interaction, and ease of use. The researchers argued that since LMSs are important for providing excellent services and establishing competitive advantages, understanding how these systems are used to solve various challenges and improve the quality of services effectively is crucial. Students’ positive behavior is greatly influenced by other e-learning factors, such as online facilitation, learning materials, and assessments (Simelane-Mnisi, 2023b). In this case, (Nguyen, 2021) emphasized that academics and IT personnel must ensure that students understand an LMS and are satisfied with it. In this regard, users are more likely to use an LMS’s features and functionalities when it is easier to access (Mizan et al., 2023). This study proposes the following hypothesis with regard to the above discussion.

H₄: 

Students’ LMS satisfaction is positively related to the system quality in South Africa.

2.5. Students’ LMS Satisfaction

(DeLone & McLean, 2003; DeLone & McLean, 2016) indicated that user satisfaction should include the full customer experience cycle from the information search to the purchase, payment, receipt, and service. Moreover, user satisfaction is the degree of the users’ level of satisfaction with reports, websites, and support services (Adayemi et al., 2024). Furthermore, user satisfaction is crucial when measuring consumers’ perceptions of e-commerce systems (DeLone & McLean, 2003; DeLone & McLean, 2016). Most universities use an LMS as their primary platform for delivering online courses (Simelane-Mnisi, 2023b). This is because LMSs could encourage active learning and foster student engagement in an online platform (Simelane-Mnisi, 2023b), contributing to students’ satisfaction. This might be accomplished by utilizing the constructivist, connectivist, and Ubuntu educational theories that emphasize collaboration, engagement, and communalism in the learning activities integrated into the technology (Bekele et al., 2023; Simelane-Mnisi, 2023b), such as an LMS. Satisfaction is the aim of LMS deployment because it measures how well the system satisfies students’ needs and expectations (Mizan et al., 2023). Convenience, usability, content quality, and system functionality all influence students’ satisfaction significantly with an LMS (Nguyen, 2021). Students are more likely to be satisfied with an LMS if it is easy to use, has high-quality and engaging course materials, and is functional (Mizan et al., 2023). (Mizan et al., 2023) argued that the ability of academics and instructional or learning designers to enhance the level of quality and efficiency of an LMS is essential for improving students’ overall learning experiences. In this study, students’ LMS satisfaction was measured by students’ LMS satisfaction, LMS high quality, and meeting students’ expectations. This study proposes the following hypothesis considering the discussion above.

H₅: 

Students’ LMS satisfaction is positively related to information quality, service quality, system quality, and convenience quality in South Africa.

2.6. Conceptual Framework

Figure 1 shows the proposed conceptual framework of the current study and the hypotheses, grounded in our discussions of the prior literature. This conceptual framework was developed based on (DeLone & McLean, 2003)’s suggestion that a specific study’s context should guide any hypotheses regarding the nature of these causal relationships.

3. Methods

The researchers relied on the prior literature to develop a conceptual model to establish students’ LMS satisfaction in the South African context. Consequently, the quantitative method was used to test the variables affecting students’ LMS satisfaction. Nguyen (2021) argued that a quantitative approach should be used when the objective of a study is to determine the relationship between the variables. This is accomplished because the main principle behind the quantitative method is the prediction of data assumptions. Furthermore, to validate the scientific model and hypotheses, a quantitative approach assists in explaining the relationship between them.

3.1. Participants

Stratified random sampling was used to select the participants in this study. Stratified random sampling, also called proportionate random sampling and quota random sampling, is a probability sampling approach that separates the total population into homogeneous groups (called strata) to complete the sampling process (Gandhi, 2022). The strata in this study were students from the Faculty of Science in the UoT. A random sample was drawn from the students who gave their consent and completed the online survey questionnaire across 14 departments in the Faculty of Science. Accordingly, 595 students from the Faculty of Science participated.

The results in

Table 1. Cross-tabulation of the participants’ biographical data.

Level of Study [N = %]
		First Year	Second Year	Third Year	Advance Diploma	Postgraduate	Total
Gender	Male	47 (7.7)	34 (5.4)	77 (12.9)	18 (3.0)	6 (1.0)	181 (30.4)
	Female	113 (19.0)	84 (14.1)	162 (27.2)	36 (6.1)	19 (3.2)	414 (69.6)
Total		159 (26.7)	118 (19.8)	339 (40.2)	54 (9.1)	25 (4.2)	595 (100)
Institution Data Bundles	Yes	147 (24.7)	115 (19.3)	227 (38.2)	51 (8.6)	25 (4.2)	565 (95.0)
	No	12 (2.0)	3 (0.5)	12 (2.0)	3 (0.5)	0 (0.0)	30 (5.0)
Total		159 (26.7)	118 (19.8)	239 (40.2)	54 (9.1)	25 (4.2)	595 (100)

show that based on the level of study, gender, institution data bundles, and the device used, less than three-quarters (69.6%) of the students were females. Of these female students, more than a quarter (27.2%) were in their third year of study. Furthermore, less than half (40.2%) of female and male students were also in their third year. The results revealed that the majority (95.0%) of students received institutional data on their cell phones according to the service provider of their choice. Of these students, more than a quarter (38.2%) were in their third year.

It may be observed in

Table 2. Cross-tabulation of the device used by the participants.

Level of Study [N = %]
		First Year	Second Year	Third Year	Advance Diploma	Postgraduate	Total
Device	Smartphone/Cellphone	146 (27.8)	105 (20.0)	210 (39.9)	41 (7.8)	24 (4.6)	526 (100)
	Laptop	95 (27.1)	77 (22.0)	131 (37.4)	29 (8.3)	18 (5.1)	350 (100)
	iPad/Tablet	4 (8.3)	15 (31.3)	25 (52.1)	4 (8.3)	0	48 (100)
	Desktop Computer	11 (21.6)	10 (19.6)	22 (43.1)	7 (13.7)	1 (2.0)	51 (100)
Total		256 (26.3)	207 (21.2)	388 (39.8)	81 (8.3)	43 (4.4)	975 (100)

that students accessed the IMFUNDO LMS using more than one device, with the majority (526) owning a cellphone or smartphone. Of these students, 210 were in their third year, followed by 146 in their first year. All the students in this study were taught using IMFUNDO-LMS utilizing different devices.

3.2. Instrument and Procedure

Part A of the survey comprised the gathering of demographic data about the respondents involved in this study. In Part B, students’ LMS satisfaction was measured quantitatively. The student LMS satisfaction survey questionnaire comprised 23 items with five subscales that measured student LMS satisfaction, service quality, information quality, convenience quality, and system quality. Four hypotheses were generated in this study based on the correlation between each conceptual framework construct. When responding to the questionnaire, the students were asked to rate their response on a 5-point Likert-type rating scale, with 1 denoting Disagree and 5 denoting Agree.

The first subscale (students’ LMS satisfaction) was about how well the system satisfied the students’ needs and expectations. Examples of typical items from this subscale were “The IMFUNDO has met my expectations” and “I like working with the LMS”. The second subscale (service quality) was about their higher education institution improving the quality of the online LMS provided to students as part of the blended learning process. An example of a typical item from this subscale was “In general; the university provides enough support to help using IMFUNDO”. The third subscale (information quality) was about the LMS content and learning activities uploaded by the academics that allowed the students to engage, interact, and participate actively in an online environment. An example of a typical item from this subscale was “IMFUNDO can provide the precise information I need”.

The fourth subscale (convenience quality) concerned the LMS being viewed as extremely user-friendly; the fewer challenges encountered, the more time the students could devote to learning. An example of a typical item from this subscale was “I can access and use IMFUNDO conveniently and quickly”. The fifth subscale (system quality) was about students being aware of the various ways the LMS could be applied to address various challenges successfully and enhance service quality based on usefulness, flexibility, interaction, and ease of use. An example of a typical item from this subscale was “IMFUNDO offers flexibility as to time and place of use”.

3.3. Data Collection and Analysis

The data originated directly from the responses of the participants who consented and completed the Likert scale online survey (score 1–5). All analyses were carried out using SPSS version 29 and Amos. The reliability of scores from this instrument was determined by computing Cronbach’s (1951) alpha coefficient (Cronbach, 1951; García-García et al., 2024). Regarding the validity of the scores from this instrument, a confirmatory factor analysis was carried out. Furthermore, to determine the underlying relationship among the variables, structural equation modeling (SEM) was applied in this study to analyze the data.

4. Results

4.1. Reliability

The results in

Table 3. The number of items, alpha values, and factor loading of the student LMS satisfaction questionnaire.

Variables	Alpha	Items	Cronbach’s Alpha If Item Deleted	Factor Loading
Student LMS Satisfaction	0.95	LMSUQ3	0.93	0.823
		LMSUQ2	0.93	0.817
		LMSUQ1	0.94	0.796
		SS1	0.94	0.660
		SS3	0.93	0.638
		SS2	0.94	0.607
Service Quality	0.90	SERQ1	0.88	0.777
		SERQ4	0.89	0.763
		SERQ2	0.88	0.758
		SERC3	0.87	0.754
		SERQ5	0.87	0.678
Information Quality	0.86	IQ3	0.80	0.765
		IQ2	0.81	0.753
		IQ4	0.81	0.733
System Quality	0.83	SYSQ3	0.83	0.756
		SYSQ2	0.72	0.704
		SYSQ1	0.72	0.639
Convenience Quality	0.74	C3	0.72	0.760
		C4	0.76	0.665

reveal the internal consistency scores for each of the 20 items in the questionnaire. The Cronbach’s alpha (Cronbach, 1951) values for the entire questionnaire were 0.96, while the alpha values of the items ranged from 0.72 to 0.94. This demonstrated that the items’ degree of internal consistency was relatively high.

4.2. Validity

In ensuring the validity of the student IMFUNDO-LMS satisfaction survey questionnaire, confirmatory factor analysis (CFA) was carried out first to check the correlation of each determinant. The sample size for this analysis was 595 with no missing data. When determining the factor structure from the data, the principal component with a Promax rotation and coefficient loading value of 0.05 was specified as the cut-off point. In the initial analysis, default settings in SPSS, such as the eigenvalue greater than unity, were specified.

In the initial analysis, a lower percentage of the variance in the SYSQ4 (I can easily access IMFUNDO-LMS anytime I need to use it), with 0.471, and C2 (Using IMFUNDO-LMS lessens my effort in performing my study and assignment activities), with 0.460, items is explained by the components. This could mean that the factors found in the analysis are less represented in these items and thus, they were removed. Secondly, a lower percentage of the variance in C1 (Using IMFUNDO-LMS enables me to search for the information or content for my study without time constraints), with 0.483, indicates the item is less significant with regard to the main factors that were extracted. Thirdly, the principal component analysis with updated extraction values was above 0.05. A decision to examine a six-factor solution was made to verify if the factor solution was appropriate.

The results in Table 3 show that a five-factor solution was produced because the first factor had six items. These items were consistent with the LMSQ and students’ satisfaction, which the researchers referred to as student LMS satisfaction. The highest loading item (0.823) was “I like working with IMFUNDO-LMS”. The second factor had five items. These items were consistent with service quality. The highest loading item (0.777) was “Training or student orientation on the operation of IMFUNDO-LMS is sufficient.” The third factor had five items. These items were consistent with information quality. The highest loading item (0.765) was “IMFUNDO-LMS can provide the precise information I need”. The fourth factor had five items. These items were consistent with system quality. The highest loading item (0.756) was “IMFUNDO-LMS language and means of communication are effective”. The fifth factor had two items. These items were consistent with convenience quality. The highest loading (0.760) item was “Using IMFUNDO-LMS allows me to improve my learning outcomes”.

4.3. Structural Equation Model (SEM) and Criteria

An SEM was computed to test the four hypotheses proposed in the model mentioned in Figure 1 with the same criteria as CFA to check for the model goodness of fit. The results show that the chi-square (χ²) was statistically significant [χ² = 743,52, df = 160, p < 0.0001]. The normed chi-square was 3.518, the goodness of fit statistics was TLI = 0.930, the comparative fit index (CFI) was 0.941, the RMSEA was 0.078, and all the indicators were statistically significant (p < 0.0001). All the values met the threshold level as indicated by (Hair et al., 2014). These results are comparable to those reported in the literature (cf. (Nguyen, 2021; Sumi & Kabir, 2021). The analysis in Figure 2 shows the estimated correlations between pairs of constructs associated with the LMS, with an emphasis on student LMS satisfaction, information quality, convenience quality, system quality, and service quality. The values show these strong associations (R² = 0.775, R² = 0.844).

The findings in Figure 2 suggest that system quality and student LMS satisfaction are most strongly correlated (R² = 0.871), with convenience quality coming in second (0.844). This implies that student satisfaction with the LMS is influenced significantly by these two elements. A comprehensive strategy for enhancing the LMS experience is suggested by the other correlations between the constructs (which range from R² = 0.721 to R² = 0.835), which demonstrate the interconnectedness of all quality factors: information, system, and convenience.

4.4. Hypotheses Measurement

Rulinawaty et al. (2024) highlighted that the best and most advised method for examining hypotheses in social and behavioral sciences is structural equation modeling or SEM. Overall,

Table 4. Cross-model goodness-of-fit indices analysis.

Fit Index	Research Obtained	Recommended Values
CMIN/DF	4.647	<5.00
GFI	0.879	<0.80
AGFI	0.841	<0.80
RMSEA	0.078	<0.05
CFI	0.941	<0.90
TLI	0.930	<0.90

indicates that the data and the model fit are acceptable. As evidenced by the RMSEA, CFI, and TLI all lying within or close to acceptable ranges, the data structure is sufficiently represented by the model. Thus, AMOS version 29 software for Windows was used to investigate the suggested hypotheses using the structural equation model. In this regard, the measurement results showed that all the hypotheses were accepted.

4.4.1. Student LMS Satisfaction

The results showed that Hypothesis H1 was validated by the student LMS satisfaction <--> service quality correlation of 0.736. This shows a significant positive relationship between satisfaction and the perceived quality of service in the LMS. Furthermore, Hypothesis H2 obtained acceptance with the student LMS satisfaction <--> information quality correlation of 0.775. This suggests a strong positive relationship between students’ satisfaction with the LMS and the quality of information provided. A subsequent analysis indicated the acceptance of Hypothesis H3 with the student LMS satisfaction <--> convenience quality correlation of 0.844. This revealed an even stronger positive relationship between student satisfaction and the convenience offered by the LMS. The results revealed that Hypothesis H4 was accepted with the student LMS satisfaction <--> system quality with a correlation of 0.871. This was the strongest relationship among the student satisfaction correlations, suggesting that system quality is highly influential on student satisfaction.

4.4.2. Service Quality

The results revealed a correlation of 0.727 for service quality <--> information quality. This suggests a moderate to strong positive relationship, meaning higher service quality is associated with better information quality. Subsequently, the results showed a correlation of 0.751 for service quality <--> convenience quality. This indicates a similar strength of the relationship, exhibiting that service quality is positively associated with convenience. Moreover, the results indicated a correlation of 0.734 for service quality <--> system quality. This also indicates a positive relationship, suggesting that higher service quality is related to better system quality.

4.4.3. Information Quality

The results showed a correlation of 0.721 for information quality <--> convenience quality. This indicates a moderate to strong relationship between information quality and convenience, implying that as information quality improves, convenience also tends to improve. Furthermore, the results indicated a correlation of 0.835 for information quality <--> system quality. This is one of the strongest relationships in the table, indicating a close association between information and system quality.

4.4.4. Convenience Quality

The results demonstrated a correlation of 0.797 for convenience quality <--> system quality. This shows a strong positive relationship, suggesting that better system quality is closely linked to increased convenience.

5. Discussion

The purpose of this study was to establish the factors influencing students’ satisfaction with the IMFUNDO LMS. In this study, it was found that most of the students access the IMFUNDO LMS using mobile devices. This is because this research revealed that most students in Africa have access to mobile devices and use them for learning (Simelane-Mnisi, 2023a) Furthermore, (Kaisara & Bwalya, 2023) contended that in countries, including those in Africa, that have a high number of mobile devices, students most commonly use mobile devices to access digital learning materials. It may be observed from the results that the students were satisfied with accessing IMFUNDO LMS information on their mobile devices since the LMS material was also provided on the IMFUNDO mobile app. This implies that the learning material and activities were automatically customized by the system for their appropriate display on smaller devices. Students accessed the material with ease and at their convenience.

In this study, three of the items from the questionnaire were removed because they were below the cut-off point of 0.5 for analysis CFA. (Nguyen, 2021) argued that to ensure consistency, the degree of data reliability should be considered using a standard ratio; to maintain a dependable data system, the variable that falls outside the permissible range should be removed. Regarding the reliability of scores from the questionnaire, it was found that the alpha values ranged from 0.72 to 0.94. Tavakol and Dennick (2011) argued that a reliability test score of 0.70 and above is considered to be good. The best scores were found to be above 0.90, which is even more trustworthy than 0.80 (Nguyen, 2021; Simelane-Mnisi & Mji, 2017; Tavakol & Dennick, 2011). One may argue that the items have a comparatively high level of internal consistency. In this regard, the alpha values were accepted for this study.

Validity was calculated using CFA to determine the factor structure using the principal component with Promax rotation. CFA is said to be adequate because it enables a researcher to specify the number of factors in a model based on what is reported in the literature (Simelane-Mnisi & Mji, 2017). The purpose of principal component analysis is to determine how well each item contributes to the component structure (Simelane-Mnisi & Mji, 2017). Items with lower commonalities may require further analysis to determine their relevance to the constructs being measured since they might not fit well with the overall factor solution.

In the initial analysis, items SYSQ4 and C2 were removed because they loaded below the specified cut-off point. In the second analysis, item C1 had a lower percentage of the variance, and it was also removed. It may be seen that in the third round of the analysis, all the items were above the stipulated cut-off point with a six-factor solution to verify their appropriateness. However, a five-factor solution was produced relating student LMS satisfaction, service quality, information quality, convenience quality, and system quality.

Based on factor 1, student LMS satisfaction, one could claim that the South African sample was unique because the results revealed that the six items of both LMSUQ and student satisfaction loaded in this factor. The highest loading item was “I like working with IMFUNDO-LMS”. This implies that students, in general, were extremely satisfied with the LMS. This finding is supported by (Mizan et al., 2023; Nguyen, 2021; Rulinawaty et al., 2024) as their results produced these factors separately.

All the valid items were categorized into a category with a significant correlation; the suggested study model was extremely valid, and all the survey results ensured alignment with the four previously mentioned hypotheses. These results suggest that if students’ expectations are met, they will be satisfied with the LMS. In this study, it may be argued that for the students to be satisfied with the use of the LMS, the university should ensure that factors relating to information, convenience, system, and service quality are considered. These results are supported by (Nguyen, 2021; Rulinawaty et al., 2024; Sarstedt et al., 2021), who indicated that student LMS satisfaction is expected to increase with high-quality LMS services.

It is evident from this study’s findings that information, convenience, and system quality were demonstrated to be significantly and positively correlated with service quality. This suggests that universities should enhance the quality of the LMS they offer to students (Asakdiyah et al., 2024). This is because students can spend more time learning if the LMS is easy to use and presents few obstacles (Venkatesh et al., 2003). It was found that students in the South African UoT expressed satisfaction with the service quality of the LMS based on their training on how to use the LMS. In this instance, student orientation was provided at all levels of study. In addition, although the LMS support service desk was functioning, it was unable to offer sufficient assistance and could not respond to students’ queries promptly. In this instance, the strategy of department LMS influencers was initiated in the Faculty of Science. Most of these students were senior tutors who helped other students with any LMS-related problems. In this study, it may be argued that the service quality of the LMS was determined by considering the advantages of technology support and responsiveness, which is based on the IS success model (DeLone & McLean, 2003; DeLone & McLean, 2016).

The results of this study revealed that information quality was demonstrated to be significantly and positively correlated with convenience and system quality. It may be argued that students in the South African UoT have a greater chance of interacting with the content on the LMS attained through the learning activities and content that encouraged and motivated them to participate, engage, and interact actively in an online environment. This finding is supported by Simelane-Mnisi (2023b), who contended that academics utilized a variety of multimedia tools in the LMS to develop module content that catered to different learner preferences.

It may also be seen from the results that convenience quality demonstrated a significant and positive correlation with system quality. In this case, students should be able to access and use any learning materials, activities, and other information on the LMS quickly and effortlessly, as this could increase the effectiveness (Mizan et al., 2023). DeLone and McLean (2016) asserted that better system quality is expected to increase user satisfaction and usage, which would improve individual productivity and enhance organizational productivity. This indicates that the South African UoT students were more likely to use the features and functionalities of the LMS to complete the tasks. This is because using their preferred devices and receiving data bundles from the institution made it simpler for them to access the LMS.

6. Limitations of This Study

This study made use of a model based on theory and examined which of four factors affect students’ LMS satisfaction. These factors were service, information, system, and convenience quality. Despite the significant findings, there are many limitations that should be taken into consideration. This study was conducted with one Faculty at a UoT in South Africa with only 595 responses from the electronic survey of 11,000 students. It is essential to increase the sample size. One further limitation is that this study was quantitative in nature, excluding participant voices. This study only concentrates on the viewpoint of the students. Since academics are users and play a significant part in using the IMFUNDO LMS, gathering data from them would lead to a more in-depth and comprehensive understanding.

7. Conclusions

In conclusion, LMSs are the main platforms for delivering quality e-learning and online courses in higher education, making certain that learning, teaching, and assessment become more effective and efficient. Accordingly, the purpose of this study was to establish the factors influencing students’ satisfaction regarding the IMFUNDO LMS. In this study, it was noted that students’ expectations were realized by ensuring their satisfaction with the IMFUNDO LMS in a blended approach. This was accomplished by using the conceptual framework that was grounded in the modified DeLone and McLean information system model, which was beneficial for the students in the UoT in South Africa. The factors that had a significant effect on the satisfaction of students were found to be those related to service, information, system, and convenience quality. In this study, the researchers found that service, information, system, and convenience quality were strongly correlated with students’ LMS satisfaction. These factors are necessary to ensure that an LMS is employed effectively. By providing innovative insight into the variables influencing students’ continuous use of the LMS, this study contributes to the body of knowledge about the adoption of LMSs and user satisfaction already in existence. This was attained by the strategies used by IT personnel, LMS administrators, academics, and instructional and learning designers to maintain the technical support, ease of use, and easy navigation of the learning activities and materials, thus increasing student satisfaction. The aim was to reduce the challenges that students might have had when using IMFUNDO to learn.

8. Recommendations

The findings in this study may assist higher education institution administrators, LMS developers, instructional and learning designers, and academics when designing, implementing, and supporting LMS platforms. It is essential that the use of LMSs by academics be promoted and supported with policy through the creation of frameworks and recommendations, and they should be included in the institution’s learning, teaching, and assessment strategies. To create an effective LMS utilizing the accumulated effort of all the stakeholders involved in higher education, the focus should fall on issues relating to technical, management, cultural, and financial support. To improve the student LMS experience, it is imperative that technological challenges be minimized at all costs. Learning activities and materials created on an LMS should encourage active learning, student-centered learning, and student engagement. To support the statistical findings with the opinions of the participants, a mixed-methods approach should be taken into consideration. Furthermore, a larger sample size could be used in a future investigation. To replicate the results of this study, another investigation could be carried out in a different context.