E-learning Satisfaction Model Inventory

1. Abstract

The E-learning Satisfaction Model Inventory is a psychometric instrument developed by A. Devisakti and T. Ramayah (2023) to evaluate the multidimensional determinants of electronic learning (e-learning) portal adoption, continuous usage, and academic performance among higher education students. Conceptualized and validated within a population of postgraduate learners in Malaysia, the instrument integrates classical technology acceptance constructs with non-cognitive and affective drivers of learning persistence. Specifically, the instrument investigates eight distinct latent dimensions: Attitude, Instructor Influence, Peer Influence, Self-Efficacy, Sense of Belonging, Grit, Usage, and Academic Performance.

Comprising 30 structured items, the scale employs a split-format response architecture: the psychological, social, and non-cognitive subscales (Attitude, Instructor Influence, Peer Influence, Self-Efficacy, Sense of Belonging, and Grit) are evaluated along a 7-point Likert scale, whereas the behavioral and outcome metrics (Portal Usage and Academic Performance) utilize a 5-point Likert scale. Psychometric validation conducted via partial least squares structural equation modeling (PLS-SEM) demonstrated robust measurement properties. Convergent validity was established across all items, with standardized outer factor loadings exceeding the 0.70 threshold and Average Variance Extracted (AVE) values surpassing 0.50 for all latent constructs. Discriminant validity was rigorously corroborated through the Heterotrait-Monotrait (HTMT) ratio of correlations, yielding values consistently below the conservative 0.85 threshold, with corresponding bootstrap confidence intervals strictly excluding 1.0. Reliability evaluations confirmed high internal consistency, exhibiting Cronbach’s alpha coefficients ranging between .843 and .881 across all subscales. The inventory provides an empirical and diagnostic tool for educational researchers, institutional administrators, and instructional designers seeking to optimize digital learning environments by addressing both technological infrastructure and learner-centric psychological attributes.

2. Keywords

Attitude, E-Learning Portal Environment, Grit, Higher Education, Instructor Influence, Measurement Model, Peer Influence, Academic Performance, Self-Efficacy, Sense of Belonging, System Usage, Educational Assessment, Virtual Classrooms, Electronic Learning

3. Authors

The E-learning Satisfaction Model Inventory was developed and empirically validated by:

• A. Devisakti — National Higher Education Research Institute (Institut Penyelidikan Pendidikan Tinggi Negara, IPPTN), Universiti Sains Malaysia (USM), 11800 Gelugor, Penang, Malaysia. Email: [email protected].
• T. Ramayah — School of Management, Universiti Sains Malaysia (USM), 11800 Gelugor, Penang, Malaysia.

4. Purpose

The principal objective of the E-learning Satisfaction Model Inventory is to provide a granular, empirically validated diagnostic instrument to examine how technological, social, personal, and psychological variables influence students’ engagement patterns and subjective academic outcomes within higher education e-learning portals. The proliferation of learning management systems (LMS) such as Moodle, Blackboard, and proprietary university portals has transformed modern tertiary instruction. However, high system availability does not organically translate into sustained student engagement, deep learning, or enhanced academic execution. Prior frameworks have frequently confined their assessment to utilitarian and technical dimensions—such as system functionality or perceived ease of use—while failing to account for the emotional, relational, and non-cognitive traits that govern persistence in remote, virtual learning environments.

The inventory was formulated to overcome this empirical fragmentation. By uniting social influence vectors (lecturers and peer networks) with intrinsic cognitive evaluations (self-efficacy and attitudes) and long-term behavioral persistence markers (grit and sense of belonging), the instrument enables researchers and educational administrators to uncover why certain students thrive while others disengage in digital environments. In research settings, the inventory serves as an operational framework for structural equation modeling, permitting hypotheses testing on direct, mediated, and moderated pathways connecting affective states to technology utilization and educational performance.

From an applied institutional and instructional perspective, the scale functions as an evaluative diagnostic tool. Universities can deploy the instrument to benchmark institutional LMS health, evaluate pedagogical transitions (e.g., hybrid, blended, or fully asynchronous curricula), and detect at-risk cohorts experiencing digital alienation or low portal engagement. Instructional designers can utilize its granular subscales to ascertain whether poor learning performance stems from platform-specific technical intimidation (low self-efficacy), insufficient institutional social facilitation (weak instructor and peer influence), or broader deficits in community integration and learner tenacity.

5. Psychological Construct

The overarching construct evaluated by this instrument is multidimensional engagement and performance in electronic learning environments, structured across eight primary operational dimensions:

Attitude

Operationalized as an individual’s favorable or unfavorable cognitive and evaluative assessment regarding the deployment of an e-learning portal for scholarly activities. Attitude reflects the affective readiness and general disposition toward incorporating educational technology into daily study habits. Positive attitudes are characterized by perceptions of convenience, utility, and modern educational value, whereas negative attitudes embody technostress, resistance, and perceived cognitive friction.

Instructor Influence

Reflecting a core facet of the subjective norm, this dimension measures the perceived explicit and implicit cues, expectations, and instructional support provided by academic faculty. It captures the degree to which learners perceive that course lecturers and tutors view the platform as essential, actively encourage its adoption, and integrate it meaningfully into course delivery.

Peer Influence

A lateral social normative construct assessing the perceived social pressure, behavioral modeling, and shared beliefs originating from fellow students within the educational ecosystem. Peer influence reflects how a learner’s utilization of the portal is shaped, reinforced, or conditioned by observing the technological habits, values, and endorsements of their academic contemporaries.

Self-Efficacy

Derived from social cognitive principles, e-learning self-efficacy measures a student’s confidence in their capabilities to operate the digital learning platform autonomously. This construct encompasses operational task execution without supervision, the ability to troubleshoot minor technical hurdles independently, and navigating complex portal functionalities without sustained pedagogical scaffolding.

Sense of Belonging

An affective-relational dimension defining the learner’s psychological feeling of being an integral, accepted, and comfortable member within the virtual learning portal environment. Rather than perceiving the digital system as an impersonal, sterile repository, learners with a high sense of belonging report feeling “at home,” emotionally secure, and culturally embedded within the institution’s virtual campus.

Grit

Rooted in trait psychology, grit within this inventory captures a learner’s perseverance of effort and consistency of interest toward long-term academic goals despite technical delays, structural setbacks, and platform-related obstacles. It isolates personal work ethic, stamina in goal pursuit, and resistance to short-term motivational fluctuations during autonomous digital study.

Usage

A behavioral acceptance metric tracking the self-reported frequency, volume, and deliberate integration of the e-learning portal into routine learning regimens. It assesses how regularly and flexibly students activate the platform for self-directed study, resource retrieval, and coursework completion.

Performance

An educational outcome metric reflecting the student’s subjective assessment of academic efficiency, productivity, and efficacy gains attributed directly to the utilization of the e-learning portal. It captures self-reported performance enhancements, time management optimizations, and overall scholastic usefulness derived from platform operations.

6. Theoretical Framework

The E-learning Satisfaction Model Inventory is theoretically grounded at the intersection of information systems theory, educational psychology, and social-cognitive frameworks. The theoretical architecture draws heavily upon four seminal conceptual models:

Technology Acceptance Model (TAM) and Extended Normative Models

At its base, the inventory reflects the conceptual logic of the Technology Acceptance Model formulated by Fred Davis (1989), which posits that technological usage is dictated by behavioral intentions, shaped fundamentally by cognitive evaluations of utility and ease of use. Devisakti and Ramayah (2023) extended this baseline by embedding subjective normative structures from the Theory of Reasoned Action (TRA) and Theory of Planned Behavior (TPB) developed by Ajzen and Fishbein. Rather than viewing the student as an isolated rational actor, the framework bifurcates social normative pressures into vertical hierarchies (Instructor Influence) and horizontal peer environments (Peer Influence), capturing how institutional authority and social contagion collectively sculpt technology adoption.

Social Cognitive Theory (Self-Efficacy)

The operationalization of technical confidence relies upon Albert Bandura‘s (1986, 1997) Social Cognitive Theory. Bandura demonstrated that behavioral execution is heavily determined by an individual’s personal agency and perceived self-efficacy—the conviction that one can successfully execute the behavior required to produce desired outcomes. In an e-learning context, self-efficacy acts as a self-regulatory cognitive bridge: even when a platform is functionally robust, students with deficits in self-efficacy experience avoidance behaviors, higher cognitive load, and elevated system frustration, ultimately suppressing voluntary portal usage.

The Need to Belong Theory

Addressing the acute sense of isolation characteristic of remote and virtual education, the model incorporates Baumeister and Leary’s (1995) fundamental Need to Belong hypothesis. This framework asserts that human beings possess an innate drive to form and maintain stable, positive, and significant interpersonal relationships. In digital platforms, spatial dislocation often induces alienation, transforming the LMS into an alienating cognitive repository. By operationalizing Sense of Belonging (adapted from Asatryan et al., 2013; Lee & Suh, 2015), the inventory posits that psychological attachment and virtual platform assimilation are imperative prerequisites for turning passive compliance into sustained affective engagement.

The Theory of Grit

Recognizing that academic success requires stamina across prolonged developmental arcs, the inventory embeds Angela Duckworth’s (2007) construct of Grit. Duckworth delineated grit as passion and perseverance toward long-term objectives in the face of adversity. Within asynchronous and blended tertiary education, learners must continually navigate interface disruptions, autonomous time management demands, and reduced immediate supervision. Grit provides the non-cognitive self-regulatory resilience necessary to resist digital distractions, overcome instructional ambiguities, and sustain continuous portal interaction across semesters.

7. Validity

The psychometric validity of the E-learning Satisfaction Model Inventory was thoroughly analyzed within a sample of postgraduate students in higher education using variance-based structural equation modeling (PLS-SEM), following guidelines established by Hair, Hult, Ringle, and Sarstedt:

Content and Face Validity

To ensure high initial content and face validity, all 30 survey items were adapted from mature, peer-reviewed operationalizations in educational technology and psychometrics literature. Scale items capturing Attitude were derived from Alharbi and Drew (2014); Instructor Influence and Self-Efficacy were grounded in Teo and van Schaik’s (2012) technology acceptance inventories; Peer Influence was mapped from Wu and Zhang (2014); Sense of Belonging was synthesized from Asatryan et al. (2013) and Lee and Suh (2015); Grit was calibrated from Aparicio’s (2017) educational perseverance scales; Usage was integrated from Lu et al. (2009) and Hsieh et al. (2016); and Performance metrics were adapted from Mohammadyari and Singh (2015). Pre-testing and expert panels verified semantic clarity and contextual suitability for postgraduate higher education cohorts.

Convergent Validity

Convergent validity evaluates the degree to which individual survey indicators accurately reflect their intended underlying latent construct. In the measurement model evaluation:

• Standardized Outer Loadings: All 30 measurement items displayed outer factor loadings exceeding the widely recommended threshold of 0.70 (ranging upwards of .75 to .88), indicating that each individual item accounts for greater than 50% of its shared indicator variance.
• Average Variance Extracted (AVE): The calculated AVE for each latent dimension substantially surpassed the standard benchmark of 0.50. This confirms that, on average, each latent construct explains more than half of the variance observed among its corresponding item indicators, affirming robust convergent validity at both the item and construct levels.

Discriminant Validity

Discriminant validity—the requirement that each latent construct represents a distinct phenomenon not captured by other model constructs—was evaluated via the Heterotrait-Monotrait (HTMT) ratio of correlations, currently recognized as the gold standard over older cross-loading and Fornell-Larcker criteria:

• HTMT Threshold: All pairwise HTMT values among the eight constructs remained strictly below the conservative threshold of 0.85, eliminating risks of conceptual collinearity or construct redundancy.
• HTMT Inference (Bootstrapping): Complete non-parametric bootstrapping was executed to derive confidence intervals for all HTMT point estimates. The resulting upper boundaries of the 95% bias-corrected confidence intervals remained strictly below the critical value of 1.0. None of the intervals encompassed 1.0, providing definitive empirical evidence of discriminant validity across the entire instrument.

8. Reliability

The reliability of the E-learning Satisfaction Model Inventory was confirmed through comprehensive assessments of internal consistency:

• Cronbach’s Alpha ($lpha$): Across all eight subscales, Cronbach’s alpha coefficients fell within the narrow, highly dependable range of .843 to .881. Because all subscale coefficients comfortably exceeded the standard psychometric floor of .70 (and the stricter .80 threshold recommended for advanced structural research), the items within each construct demonstrate high mutual covariance and construct homogeneity.
• Composite Reliability ($
  ho_c$ / CR): To complement Cronbach’s alpha—which tends to underestimate internal consistency due to its assumption of tau-equivalence (equal factor loadings across all items)—composite reliability was evaluated. In alignment with PLS-SEM standards, all latent dimensions returned CR values well in excess of .85, confirming superior construct reliability without indicator bloat or redundancy.
• Item-Level Communalities: As verified during convergent validity analyses, item-level communalities (the square of standardized outer loadings) confirmed that residual measurement error remained minimal across indicators.

9. Factor Analysis

The dimensional architecture of the instrument was tested via a reflective measurement model framework utilizing Partial Least Squares Structural Equation Modeling (PLS-SEM) within SmartPLS. While traditional Exploratory Factor Analysis (EFA) serves early item discovery, the theoretical maturity of the underlying scales justified a confirmatory measurement model approach to verify structural integrity.

During the measurement model evaluation phase, all eight operational constructs were specified as first-order reflective factors:

• Indicator Loadings: Every indicator mapped unambiguously to its designated parent construct. No individual item registered cross-loadings approaching its primary loading, preserving unifactorial clarity across the 30 indicators. Standardized outer factor loadings were uniformly high, consistently exceeding .70.
• Multicollinearity Diagnostics: Variance Inflation Factor (VIF) values for all structural and indicator paths were evaluated. All inner and outer VIF values fell well below the conservative ceiling of 3.3, confirming the absence of lateral collinearity among the independent predictor dimensions (e.g., separating Instructor Influence from Peer Influence, and Attitude from Self-Efficacy).
• Model Fit: In accordance with modern composite-based modeling practices, the standardized root mean square residual (SRMR) was examined to establish overall measurement fit, displaying values well within acceptable thresholds (SRMR < .08).

10. Instrument / Measurement Tool

• Test Name: E-learning Satisfaction Model Inventory
• Instrument Classification: Diagnostic Self-Report Inventory / Educational Research Questionnaire
• Target Population: Adult learners enrolled in higher education programs (undergraduate, postgraduate, professional education) utilizing an institutionally hosted e-learning management portal.
• Administration Format: Self-administered digital/electronic questionnaire (web-based, LMS plug-in, or survey software).
• Total Item Count: 30 items structured across 8 distinct subscales:
• Attitude: 3 items
• Instructor Influence: 3 items
• Peer Influence: 3 items
• Self-Efficacy: 4 items
• Sense of Belonging: 3 items
• Grit: 6 items
• Usage: 4 items
• Academic Performance: 4 items
• Response Architecture: A differentiated Likert scale format is utilized:
  • Subscales 1 through 6 (Attitude, Instructor Influence, Peer Influence, Self-Efficacy, Sense of Belonging, and Grit): Evaluated using a 7-point Likert scale (e.g., 1 = Strongly Disagree to 7 = Strongly Agree).
  • Subscales 7 and 8 (Usage and Academic Performance): Evaluated using a 5-point Likert scale (e.g., 1 = Strongly Disagree / Never to 5 = Strongly Agree / Very Frequently).
• Scoring and Computational Procedures: Individual construct scores are computed by calculating the arithmetic mean or composite latent variable score of the items within each designated subscale. Higher mean values indicate elevated levels of the underlying attribute (e.g., higher grit, stronger sense of belonging, more extensive portal utilization, or superior perceived academic performance). In structural equation modeling studies, latent construct scores are derived directly via PLS-SEM weighting algorithms.

11. Permissions & Fee and Test Year

The E-learning Satisfaction Model Inventory was published and validated in 2023 by researchers at Universiti Sains Malaysia. The formal psychometric characteristics, structural pathways, and complete item battery appeared in the peer-reviewed journal Interactive Learning Environments.

• Cost and Royalties: The instrument is non-commercial. There is no fee associated with the utilization of the inventory for non-profit scholarly research, academic dissertations, and institutional quality improvement initiatives.
• Permissions and Inquiries: Researchers desiring to deploy, adapt, or translate the instrument are advised to contact the original corresponding author ([email protected]) or obtain permissions via the journal publisher (Taylor & Francis / Informa UK Limited). Commercial applications, digital packaging into commercial LMS software, or for-profit consulting deployments require explicit authorization from the copyright holders.

12. References

Alharbi, S., & Drew, S. (2014). Using the Technology Acceptance Model in understanding academics’ behavioural intention to use Learning Management Systems (LMS). International Journal of Advanced Computer Science and Applications, 5(1), 143–155. https://doi.org/10.14569/IJACSA.2014.050120

Aparicio, M., Bacao, F., & Oliveira, T. (2017). Grit in the path to e-learning success. Computers in Human Behavior, 66, 388–399. https://doi.org/10.1016/j.chb.2016.10.009

Asatryan, V. S., Oh, H., & Cavusoglu, M. (2013). The role of social presence and belongingness in online education. Journal of Hospitality & Tourism Education, 25(3), 115–124. https://doi.org/10.1080/10963758.2013.826950

Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory. Prentice-Hall.

Baumeister, R. F., & Leary, M. R. (1995). The need to belong: Desire for interpersonal attachments as a fundamental human motivation. Psychological Bulletin, 117(3), 497–529. https://doi.org/10.1037/0033-2909.117.3.497

Davis, F. D. (1989). Perceived usefulness, perceived ease of use, and user acceptance of information technology. MIS Quarterly, 13(3), 319–340. https://doi.org/10.2307/249008

Devisakti, A., & Ramayah, T. (2023). Sense of belonging and grit in e-learning portal usage in higher education. Interactive Learning Environments, 31(8), 4850–4864. https://doi.org/10.1080/10494820.2021.1983611

Duckworth, A. L., Peterson, C., Matthews, M. D., & Kelly, D. R. (2007). Grit: Perseverance and passion for long-term goals. Journal of Personality and Social Psychology, 92(6), 1087–1101. https://doi.org/10.1037/0022-3514.92.6.1087

Hsieh, P. J., Rai, A., & Xu, S. X. (2016). Extracting business value from IT: A sensemaking perspective of post-adoptive use. Management Science, 62(4), 1147–1168. https://doi.org/10.1287/mnsc.2015.2169

Lee, J., & Suh, A. (2015). How do virtual community members build social capital? The role of sense of belonging and cognitive social capital. Computers in Human Behavior, 52, 464–473. https://doi.org/10.1016/j.chb.2015.06.018

Lu, H. P., Lin, J. C. C., & Hsiao, K. L. (2009). Exploring the determinants of continuous usage of e-learning systems: An empirical study. International Journal of Information and Communication Technology Education, 5(3), 44–58. https://doi.org/10.4018/jicte.2009070104

Mohammadyari, S., & Singh, H. (2015). Understanding the effect of e-learning on individual performance: The role of digital literacy. Computers & Education, 82, 11–25. https://doi.org/10.1016/j.compedu.2014.10.025

Teo, T., & van Schaik, P. (2012). Understanding the intention to use technology by preservice teachers: An integration of technology acceptance model and theory of planned behavior. Interactive Learning Environments, 20(5), 425–439. https://doi.org/10.1080/10494820.2010.513876

Wu, B., & Zhang, C. (2014). Empirical study on student acceptance of mobile learning in higher education. British Journal of Educational Technology, 45(5), 874–886. https://doi.org/10.1111/bjet.12093

13. Items of the Scale