Augmented Reality Intention–Measurement Model Questionnaire

Abstract

The Augmented Reality Intention—Measurement Model Questionnaire (ARI-MMQ) is an empirically validated psychometric instrument developed by Muhammad Saleem, Suzilawati Kamarudin, Haneen Mohammad Shoaib, and Asim Nasar (2023). Designed in response to the acute pedagogical transformations accelerated by the COVID-19 pandemic, the ARI-MMQ provides a rigorous operational framework to evaluate higher education students’ behavioral intentions toward adopting augmented reality (AR) applications within electronic learning (e-learning) environments. Grounded in the Theory of Planned Behavior (TPB) pioneered by Icek Ajzen, the scale systematically integrates dual-faceted perceptual value paradigms—specifically utilitarian and hedonic value systems—adapted from classical consumer experience and technology acceptance models. The ARI-MMQ comprises 18 self-report items mapped across six distinct latent dimensions: Augmented Reality Utilitarian Value (3 items), Augmented Reality Hedonic Value (3 items), Attitude Toward Use (3 items), Subjective Norms (3 items), Perceived Behavioral Control (3 items), and Intention Toward E-Learning (3 items). Responses are captured using an authentic five-point Likert-type probability-expectation metric ranging from 1 (“extremely unlikely”) to 5 (“extremely likely”). Confirmatory factor analysis and structural equation modeling demonstrate exemplary psychometric properties: item loadings universally exceed the 0.70 benchmark, construct reliability coefficients reflect high internal consistency with Cronbach’s alpha values spanning 0.702 to 0.885, composite reliability (CR) indices ranging between 0.680 and 0.913, and convergent validity established through average variance extracted (AVE) estimates between 0.636 and 0.819. The instrument models augmented reality as a higher-order construct with robust hierarchical factor loadings of 0.981 for utilitarian value and 0.980 for hedonic value. By elucidating how immersive interactive learning technologies modulate cognitive appraisals, normative social beliefs, and perceived self-efficacy, the ARI-MMQ serves as a cornerstone diagnostic instrument for psychometricians, educational technologists, instructional designers, and behavioral researchers navigating virtual classrooms and human-computer interactions in tertiary education.

Keywords

Augmented Reality, Theory of Planned Behavior, Educational Technology, E-Learning, Utilitarian Value, Hedonic Value, Attitude Toward Technology, Subjective Norms, Perceived Behavioral Control, Behavioral Intention, Higher Education, Measurement Model, Psychometrics, Virtual Classrooms, Human-Computer Interaction

Authors

The Augmented Reality Intention—Measurement Model Questionnaire was formulated and validated by an international team of researchers specializing in educational technology, business administration, and human-computer interaction:

• Muhammad Saleem, Ph.D. (Corresponding Author)
  Affiliation: Azman Hashim International Business School, Universiti Teknologi Malaysia, Kuala Lumpur, 54100, Malaysia.
  ORCID: 0000-0002-0325-217X
  Email: [email protected]
• Suzilawati Kamarudin, Ph.D.
  Affiliation: Azman Hashim International Business School, Universiti Teknologi Malaysia, Kuala Lumpur, Malaysia.
• Haneen Mohammad Shoaib, Ph.D.
  Affiliation: College of Business Administration, University of Business and Technology, Jeddah, Saudi Arabia.
• Asim Nasar, Ph.D.
  Affiliation: Azman Hashim International Business School, Universiti Teknologi Malaysia, Kuala Lumpur, Malaysia.
  ORCID: 0000-0003-0053-2595

Purpose

The primary purpose of the Augmented Reality Intention—Measurement Model Questionnaire (ARI-MMQ) is to diagnose, quantify, and explain the cognitive, affective, and normative mechanisms through which tertiary students develop behavioral intentions to adopt and utilize augmented reality (AR) applications within digital learning ecosystems. The abrupt transition to distance education during global public health disruptions, such as the COVID-19 pandemic, brought about critical challenges in higher education, notably academic disengagement, zoom fatigue, spatial disconnect, and impaired collaborative dynamics. Traditional two-dimensional virtual classrooms often struggle to replicate authentic laboratory experimentation, anatomical spatial awareness, architectural design modeling, and interactive pedagogical tasks. Augmented reality bridges this instructional divide by overlaying computer-generated digital sensory assets—such as three-dimensional graphic projections, spatial audio, and interactive tactical visual cues—directly onto the learner’s physical environment in real time.

Despite the proliferation of augmented reality mobile applications, software deployment alone does not guarantee student engagement or sustained academic adoption. Behavioral intention is the single most reliable proximal determinant of actual technology integration. Therefore, Saleem et al. (2023) constructed the ARI-MMQ to capture both the functional efficiency (utilitarian value) and emotional immersion (hedonic value) delivered by augmented reality systems, mapping these experiential appraisals onto the validated psychosocial architecture of the Theory of Planned Behavior. The measurement tool elucidates the precise cognitive path running from sensory and functional interaction with immersive media to internal attitude formation, perceived social expectations, domain self-efficacy, and ultimate programmatic utilization.

In applied academic and research settings, the ARI-MMQ serves multiple critical functions:

• Instructional Systems Design & Technology Optimization: Enables instructional designers and academic software developers to test educational AR software before widespread institutional rollout, isolating whether low adoption traces to poor perceived utility (e.g., lack of curricular alignment) or inadequate hedonic engagement (e.g., cumbersome or sterile user interfaces).
• Institutional Policy & Digital Curriculum Planning: Equips university administrators, academic deans, and IT infrastructure committees with empirical metrics to determine student readiness, social acceptability, and resource hurdles across diverse demographic groups and academic disciplines.
• Pedagogical Intervention Evaluation: Operates as a pre- and post-intervention diagnostic tool to quantify shifts in student motivation, autonomous capability beliefs, and pedagogical engagement following experiential augmented reality laboratory or classroom modules.
• Psychometric & Sociotechnical Research: Provides behavioral researchers with an empirically stable, structurally sound baseline questionnaire to test moderated and mediated models involving individual learner differences, technological anxiety, cognitive load, and academic self-regulation.

Psychological Construct

The ARI-MMQ operates as a multidimensional assessment instrument comprising six interconnected latent variables. Each construct captures a discrete facet of the student’s cognitive appraisal, social environment, or behavioral predisposition regarding augmented reality in higher education:

1. Augmented Reality: Utilitarian Value (AR-UV)

Utilitarian value reflects an instrumental, task-oriented, cognitive appraisal of technology. Rooted in functional theory and the perceived usefulness paradigm of technology acceptance, utilitarian value evaluates whether students perceive an augmented reality application as an effective, goal-directed tool that demonstrably enhances academic productivity. It reflects cognitive outcomes such as the clarification of complex theoretical concepts, acceleration of assignment completion, improvement in academic performance evaluation, and superior mastery of coursework. When an AR interface allows an engineering student to visualize internal mechanical stress points or a medical student to inspect three-dimensional vascular pathways, the perceived utility generated constitutes utilitarian value.

2. Augmented Reality: Hedonic Value (AR-HV)

Hedonic value encompasses the non-instrumental, experiential, and affective dimensions of technology utilization. Drawing from theories of intrinsic motivation, playfulness, and flow states, hedonic value assesses the level of spontaneous pleasure, enjoyment, excitement, stimulation, and psychological well-being experienced by the student during digital interaction. In contrast to utilitarian value, hedonic value is not driven by task execution or academic grading milestones; rather, it is anchored in the intrinsic emotional delight and sensory immersion afforded by the augmented interface. It accounts for why learners may remain voluntarily engaged with complex educational materials when represented through dynamic augmented media.

3. Attitude Toward Use (ATU)

Attitude represents the student’s overarching positive or negative cognitive and evaluative judgment regarding the integration of augmented reality into their regular e-learning routine. In psychological theory, attitude serves as an internal predisposition that synthesizes subjective beliefs about the consequences of performing a behavior. Within the ARI-MMQ, attitude captures whether the learner considers incorporating AR apps into their virtual classroom studies to be a wise, intelligent, pleasant, and desirable educational choice.

4. Subjective Norms (SN)

Subjective norms quantify the perceived social and environmental pressures exerted on the learner to engage or not engage in e-learning via augmented reality apps. This construct reflects the student’s normative beliefs regarding the expectations of influential referents, such as course instructors, academic peers, university mentors, and family members. It also incorporates perceived peer adoption dynamics—specifically, whether classmates are adopting similar tools—creating a collective normative climate that reinforces or discourages digital educational behaviors.

5. Perceived Behavioral Control (PBC)

Perceived behavioral control assesses the learner’s internal sense of self-efficacy, personal agency, and access to necessary external resources required to successfully deploy augmented reality applications for academic purposes. Conceptualized along Albert Bandura’s self-efficacy paradigm and Ajzen’s control beliefs, PBC evaluates whether the student possesses sufficient technological literacy, requisite hardware/software affordances, cognitive self-confidence, and operational autonomy to overcome technical disruptions, navigation barriers, or usability friction without succumbing to computer anxiety.

6. Intention Toward E-Learning (INT)

Intention toward e-learning serves as the central dependent outcome construct of the measurement model. It represents the conscious, planned decision and motivational investment of the learner to actively employ, continue using, and integrate augmented reality applications into their prospective educational routines. In accordance with behavioral decision models, behavioral intention is the most immediate and decisive precursor to sustained actual usage, encapsulating the student’s planned effort and voluntary dedication of cognitive bandwidth toward AR-mediated virtual education.

Theoretical Framework

The theoretical architecture of the ARI-MMQ synthesizes two foundational paradigms in social psychology and behavioral economics: the Theory of Planned Behavior (TPB) formulated by Icek Ajzen (1985, 1991), and the dual-factor Hedonic and Utilitarian Consumption Framework advanced by Childers, Carr, Peck, and Carson (2001) alongside Babin, Darden, and Griffin (1994).

The Theory of Planned Behavior Foundation

The Theory of Planned Behavior posits that human action is guided by three structurally distinct categories of beliefs:

• Behavioral Beliefs: Beliefs concerning the likely consequences of the behavior, which produce a favorable or unfavorable Attitude Toward the Behavior.
• Normative Beliefs: Beliefs regarding the normative expectations of others, which crystallize into Subjective Norms.
• Control Beliefs: Beliefs regarding the presence of factors that may facilitate or impede performance of the behavior, giving rise to Perceived Behavioral Control.

In combination, attitude toward the behavior, subjective norms, and perceived behavioral control lead to the formation of a Behavioral Intention. As an overarching principle, the more favorable the attitude and subjective norm, and the greater the perceived control, the stronger should be the individual’s intention to execute the behavior under consideration. In an educational technology context, however, classical TPB treats behavioral beliefs primarily as cognitive evaluations of outcomes, frequently omitting the rich sensory and emotional interactions inherent in modern human-computer interfaces.

Integration of Utilitarian and Hedonic Value Constructs

To address this limitation, Saleem et al. (2023) synthesized the TPB with interactive technology paradigms. Extending the work of Childers et al. (2001)—who demonstrated that computer-mediated consumer environments are governed simultaneously by “the Web as a digital store” (utilitarian efficiency) and “the Web as a digital playground” (hedonic immersion)—the ARI-MMQ models augmented reality as an antecedent higher-order construct that exerts direct structural influences on the core elements of the TPB.

Under this integrated model:

• Dual Appraisals Fuel Attitude: Both utilitarian value (cognitive utility, task efficacy) and hedonic value (affective enjoyment, visual pleasure) feed directly into the formation of positive student attitudes toward AR-enabled e-learning. Learners do not adopt immersive technology solely because it improves exam preparation; they adopt it because the interactive visualization makes the learning experience intrinsically gratifying.
• Higher-Order Structural Conceptualization: Augmented reality is operationalized as a second-order reflective construct constituted by two primary first-order dimensions: Utilitarian Value and Hedonic Value. This architectural design permits researchers to capture the holistic cognitive-affective impact of augmented reality without suffering from structural collinearity or conceptual fragmentation.
• Synergy with Control and Norms: The resulting attitudes interact with peer-reinforced subjective norms and individual perceived behavioral control, collectively driving the final behavioral intention to adopt AR in ongoing virtual and blended classroom spaces.

Validity

The validity of the ARI-MMQ was thoroughly substantiated by Saleem et al. (2023) using multivariate analytical standards in partial least squares structural equation modeling (PLS-SEM) and confirmatory psychometric evaluations. The validation sample comprised university students enrolled in higher education institutions in Pakistan who utilized augmented reality tools during the COVID-19 remote learning phases.

Content and Face Validity

To ensure content and face validity, the initial item pool was directly derived and adapted from validated psychometric instruments in human-computer interaction, retail technology acceptance, and educational psychology literature, specifically drawing upon scales developed by Childers et al. (2001) and established operationalizations of the Theory of Planned Behavior (Ajzen, 1991). The preliminary questionnaire underwent review by academic experts in educational technology and psychometrics to confirm that all items adequately represented their target latent constructs without linguistic ambiguity or redundancy.

Convergent Validity

Convergent validity demonstrates the degree to which an operationalized measure correlates positively with alternative measures of the same construct. In the ARI-MMQ, convergent validity was established via two rigorous psychometric criteria across all six latent variables:

• Average Variance Extracted (AVE): The AVE reflects the grand mean value of the squared loadings of the indicators associated with a specific construct. Across all six subscales of the ARI-MMQ, AVE values ranged from 0.636 to 0.819. Because every construct exceeded the standard 0.50 threshold established by Fornell and Larcker (1981), it was established that each construct explains well over 50% of the variance of its corresponding indicators, rather than residual measurement error.
• Standardized Factor Loadings: All 18 individual item factor loadings surpassed the conservative cutoff threshold of 0.70, indicating that shared variance between each observed item and its designated latent factor was substantially higher than error variance.

Discriminant Validity

Discriminant validity assesses whether constructs in the measurement model are empirically unique and capture phenomena not represented by other constructs. Discriminant validity for the ARI-MMQ was confirmed using the following established methodologies:

• Fornell-Larcker Criterion: In compliance with the guidelines established by Fornell and Larcker (1981) and reinforced by Hair et al. (2010, 2014), the square root of the AVE for each construct was compared against the inter-construct correlation coefficients between that construct and all other latent variables. In every case, the square root of the AVE was substantially higher than any bivariate correlation involving that factor.
• Cross-Loading Inspection: Analysis of cross-loadings confirmed that all indicator items exhibited their highest loadings on their theoretically specified construct, with noticeably lower cross-loadings across alternative latent factors.

Reliability

The reliability of the Augmented Reality Intention—Measurement Model Questionnaire was evaluated using both traditional lower-bound internal consistency estimates and modern structural equation composite metrics, yielding robust evidence of stability across all dimensions:

Cronbach’s Alpha

Internal consistency was initially gauged via Cronbach’s alpha (α). Across all six latent dimensions, the observed coefficients ranged from 0.702 to 0.885. Specifically:

• Each individual subscale successfully exceeded the widely accepted psychometric benchmark of α ≥ 0.70 recommended by Nunnally and Bernstein for confirmatory behavioral research.
• No construct approached excessively elevated levels (e.g., > 0.95), confirming that the instrument exhibits strong inter-item consistency while avoiding narrow semantic redundancy across its indicators.

Composite Reliability (CR)

Recognizing that Cronbach’s alpha frequently underestimates scale reliability in structural modeling due to its assumption of tau-equivalence (equal factor loadings across all items), Saleem et al. (2023) computed composite reliability (CR). Composite reliability prioritizes items based on their individual factor loadings, providing a more precise metric of internal structural consistency:

• Composite reliability values across the six constructs spanned between 0.680 and 0.913.
• These coefficients fully surpass the benchmark cutoff of 0.60 to 0.70 advocated by Hair, Hult, Ringle, and Sarstedt (2014, 2017) for robust structural equation modeling, verifying that the scale demonstrates minimal measurement error variance.

Factor Analysis

The dimensional structure of the ARI-MMQ was empirically validated through rigorous Confirmatory Factor Analysis (CFA) within a covariance-based or variance-based structural equation modeling paradigm. Rather than forcing items into an unconstrained exploratory space, CFA was executed to test the hypothesized theoretical mapping of items onto their respective latent dimensions and the overarching second-order configuration.

First-Order Measurement Model

The first-order measurement model evaluated 18 reflective indicators assigned across six primary latent constructs. The confirmatory results revealed:

• Indicator Loadings (λ): Every single item loaded onto its designated latent variable with an empirical loading exceeding the strict 0.70 threshold (λ > 0.70), confirming high indicator reliability and negligible specification error.
• Error Variances: Indicator error variances remained low and statistically significant, confirming that each item reliably reflects its targeted cognitive, normative, or behavioral construct.

Higher-Order Measurement Structure

A central theoretical and empirical innovation of Saleem et al.’s (2023) model is the parameterization of Augmented Reality as a second-order (higher-order) reflective construct constituted by two distinct first-order constructs: Utilitarian Value and Hedonic Value. When subjected to higher-order confirmatory modeling, the empirical loadings of the first-order constructs onto the overarching Augmented Reality latent variable were exceptionally robust:

• Utilitarian Value Loading: β = 0.981
• Hedonic Value Loading: β = 0.980

These nearly balanced and exceptionally high loadings indicate that augmented reality applications in educational environments operate as a dual-engine phenomenon: academic utility and affective immersion are equally vital and co-essential in defining the overall augmented reality learning experience. Omitting either the hedonic or the utilitarian facet compromises the predictive integrity of the measurement model.

Instrument / Measurement Tool

The Augmented Reality Intention—Measurement Model Questionnaire is structured as an objective, self-report psychometric inventory designed for digital or paper-and-pencil administration. The formal operational characteristics of the instrument are detailed below:

• Test Type: Original multidimensional inventory / structured measurement model questionnaire.
• Target Population: Adult learners, undergraduate and postgraduate university students, and higher education academic cohorts engaged in remote, blended, or virtual classroom coursework.
• Language Available: English (standard international educational administration).
• Total Item Count: 18 standardized items.
• Subscale Architecture: 6 distinct latent subscales (3 items per subscale):
  • AR: Utilitarian Value (Items UVL1, UVL2, UVL3)
  • AR: Hedonic Value (Items HVL1, HVL2, HVL3)
  • Attitude Toward Use (Items ATU1, ATU2, ATU3)
  • Subjective Norms (Items SN1, SN2, SN3)
  • Perceived Behavioral Control (Items BC1, BC2, BC3)
  • Intention Toward E-Learning (Items INT1, INT2, INT3)
• Authentic Response Scale: Five-point Likert scales are used for all 18 items, ranging from 5 (extremely likely) to 1 (extremely unlikely):
  • 5 = Extremely likely
  • 4 = Likely
  • 3 = Neutral / Undecided
  • 2 = Unlikely
  • 1 = Extremely unlikely
• Administration Time: Approximately 5 to 8 minutes for complete self-administration.
• Scoring and Computational Procedures:
  • All 18 items are positively worded; therefore, no reverse-scoring is required.
  • Subscale Scores: Calculated either by summing the raw item scores within each subscale (yielding a range of 3 to 15 per subscale) or by computing the mean arithmetic score across the three items (yielding a range of 1.00 to 5.00).
  • Higher-Order AR Score: Generated by computing the grand mean across the six items constituting Utilitarian Value and Hedonic Value (UVL1–UVL3 + HVL1–HVL3 / 6).
  • Structural Equation Modeling Use: For advanced path modeling, latent variable scores should be extracted using structural equation modeling software (e.g., SmartPLS, AMOS, lavaan in R) utilizing the established composite weights and factor loadings.

Permissions & Fee and Test Year

Publication Year: 2023 (Originally published online ahead of print in 2021).

Copyright and Ownership: The psychometric measurement model and its foundational validation study are copyrighted by the authors (Muhammad Saleem, Suzilawati Kamarudin, Haneen Mohammad Shoaib, and Asim Nasar) and the publishing house Informa UK Limited / Taylor & Francis Group.

Permissions: The scale items were published in the peer-reviewed scholarly literature for academic dissemination. Researchers, educators, and institutional assessment teams intending to use the ARI-MMQ for non-commercial academic research, pedagogical evaluation, or university dissertations may adapt and administer the items provided that full formal scholarly citation is credited to the original authors and the publishing journal (Interactive Learning Environments). For commercial technology testing, proprietary software integration, or fee-based institutional consulting implementations, explicit written permission must be secured directly from Taylor & Francis or via the Copyright Clearance Center.

Assessment Fee: There is no fee required for academic, scholarly, or non-commercial research use.

References

• Ajzen, I. (1985). From intentions to actions: A theory of planned behavior. In J. Kuhl & J. Beckmann (Eds.), Action control: From cognition to behavior (pp. 11–39). Springer. https://doi.org/10.1007/978-3-642-69746-3_2
• Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50(2), 179–211. https://doi.org/10.1016/0749-5978(91)90020-T
• Babin, B. J., Darden, W. R., & Griffin, M. (1994). Work and/or fun: Measuring hedonic and utilitarian shopping value. Journal of Consumer Research, 20(4), 644–656. https://doi.org/10.1086/209376
• Childers, T. L., Carr, C. L., Peck, J., & Carson, S. (2001). Hedonic and utilitarian motivations for online retail shopping behavior. Journal of Retailing, 77(4), 511–535. https://doi.org/10.1016/S0022-4359(01)00056-2
• 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
• Fornell, C., & Larcker, D. F. (1981). Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research, 18(1), 39–50. https://doi.org/10.1177/002224378101800104
• Hair, J. F., Black, W. C., Babin, B. J., & Anderson, R. E. (2010). Multivariate data analysis (7th ed.). Prentice Hall.
• Hair, J. F., Hult, G. T. M., Ringle, C. M., & Sarstedt, M. (2014). A primer on partial least squares structural equation modeling (PLS-SEM). SAGE Publications.
• Hair, J. F., Hult, G. T. M., Ringle, C. M., & Sarstedt, M. (2017). A primer on partial least squares structural equation modeling (PLS-SEM) (2nd ed.). SAGE Publications. https://doi.org/10.1007/978-3-319-05542-8_15-1
• Saleem, M., Kamarudin, S., Shoaib, H. M., & Nasar, A. (2023). Influence of augmented reality app on intention towards e-learning amidst COVID-19 pandemic. Interactive Learning Environments, 31(5), 3083–3097. https://doi.org/10.1080/10494820.2021.1919147

Items of the Scale