AR Shopping App Usefulness (ARSAU)

1. Abstract

The AR Shopping App Usefulness (ARSAU) scale is a specialized, psychometrically validated three-item measurement instrument designed to capture consumer assessments of instrumental utility, cognitive efficiency, and task facilitation when employing augmented reality (AR) applications in retail and e-commerce environments. Developed and validated by Tim Hilken, Ko de Ruyter, Mathew Chylinski, Dominik Mahr, and Debbie I. Keeling (2017) across a series of empirical investigations published in the Journal of the Academy of Marketing Science, the ARSAU scale adapts foundational utilitarian technology evaluation principles originally introduced in Fred Davis’s (1989) Technology Acceptance Model (TAM) and refined for electronic commerce by Childers, Carr, Peck, and Carson (2001). By isolating the utilitarian mechanism through which spatial computer-generated augmentations integrate into the consumer’s real-time physical perception, the scale assesses the degree to which an interactive AR interface streamlines product inspection, mitigates diagnostic uncertainty, and enhances decision-making performance.

Operated as a strictly unidimensional instrument, the ARSAU utilizes a standard 7-point Likert-type response format ranging from 1 (“Strongly Disagree”) to 7 (“Strongly Agree”). Extensive empirical testing demonstrates exceptional psychometric properties, consistently exhibiting internal consistency coefficients (Cronbach’s alpha and composite reliability) exceeding 0.88, robust average variance extracted (AVE) values surpassing 0.70, and pronounced factor loadings above 0.85 across multiple experimental product categories. Confirmatory factor analysis confirms clean structural validity, demonstrating strict unidimensionality, absence of cross-loadings, and distinct discriminant validity against related constructs such as AR enjoyment, spatial presence, user comfort, and behavioral purchase intentions. The ARSAU provides researchers and human-computer interaction (HCI) specialists with an agile, high-precision diagnostic metric suitable for both mobile e-commerce platforms and contextual in-store physical retail deployments.

2. Keywords

Augmented Reality, Perceived Usefulness, Technology Acceptance Model, Shopping Performance, E-Commerce Evaluation, Mobile AR, Human-Computer Interaction, Product Inspection, Retail Technology, Psychometrics, Consumer Decision-Making, Spatial Computing, Instrumental Value, Cognitive Fit

3. Authors

The AR Shopping App Usefulness scale was adapted, validated, and operationalized within modern augmented reality retail contexts by an international research team comprising:

• Tim Hilken — Department of Marketing and Supply Chain Management, School of Business and Economics, Maastricht University, Maastricht, The Netherlands. Expertise in service design, digital marketing innovation, customer experience, and virtual/augmented reality interfaces.
• Ko de Ruyter — School of Business and Management, King’s College London, London, United Kingdom; and Department of Marketing, Maastricht University. Leading scholar in customer experience management, technology-mediated service encounters, and frontline marketing strategy.
• Mathew Chylinski — School of Marketing, UNSW Business School, University of New South Wales, Sydney, Australia. Specialist in decision neuroscience, behavioral economics, pricing mechanisms, and digital product simulation.
• Dominik Mahr — Department of Marketing and Supply Chain Management, Maastricht University, Maastricht, The Netherlands. Focuses on service innovation, digital interfaces, collaborative customer technologies, and healthcare service platforms.
• Debbie I. Keeling — School of Business, Management and Economics, University of Sussex, Brighton, United Kingdom. Researches digital consumer psychology, behavioral change interventions, and interactive interface adoption.

4. Purpose

The primary purpose of the AR Shopping App Usefulness (ARSAU) scale is to quantify consumer perceptions of utilitarian benefit, operational efficiency, and performance enhancement derived from employing augmented reality applications during commercial product evaluations. In traditional e-commerce paradigms, consumers routinely encounter cognitive constraints stemming from spatial and physical detachment from the target object. Static two-dimensional images and pre-rendered video representations often fail to convey critical tangible attributes, such as true physical scale, spatial volume, color harmony within specific domestic environments, and ergonomic compatibility. These informational deficits induce cognitive strain, elevate perceived transactional risk, and impede purchase finalization.

Augmented reality shopping applications remediate these constraints by superimposing virtual three-dimensional product representations into the user’s immediate real-world environment via mobile device cameras or head-mounted displays. Consequently, an urgent psychometric requirement emerged within marketing psychology and human-computer interaction to measure whether consumers recognize this technological affordance as an actionable, value-adding utility rather than a novel, transient gimmick. The ARSAU directly isolates this utilitarian mechanism. It captures the degree to which an AR tool successfully reduces cognitive load, clarifies product diagnostics, and optimizes the overall task of selecting, assessing, and purchasing goods.

In academic and applied research settings, the ARSAU serves multiple analytical functions:

• Comparative Interface Benchmarking: Enables software developers and usability engineers to assess whether modifications in AR user interfaces (e.g., markerless tracking accuracy, realistic photorealistic ray tracing, ambient lighting estimation) produce tangible improvements in perceived utility compared to traditional 2D web interfaces or alternative AR modalities.
• Structural Equation Modeling (SEM) Mediational Analysis: Functions as a key instrumental mediator bridging technical interface affordances (such as environmental embedding, real-time simulated interactivity, and visual immersion) with end-stage commercial criteria (including willingness to pay, cart abandonment mitigation, brand attitude, and post-purchase decision satisfaction).
• Cross-Context Diagnostic Implementation: The scale was deliberately constructed with contextual flexibility. While initially deployed in mobile e-commerce studies, deleting the single term “online” allows the instrument to evaluate in-store interactive digital signage, spatial mirror displays, and physical retail smart glasses, offering unified psychometric comparability across omnichannel touchpoints.
• Diagnostic Optimization in Retail Management: Provides commercial retail managers with a concise, low-friction diagnostic diagnostic tool to audit customer touchpoints, identifying whether low adoption rates stem from usability/utility deficits versus technical instability or lack of hedonic engagement.

5. Psychological Construct

The AR Shopping App Usefulness (ARSAU) scale measures the unidimensional psychological construct of perceived usefulness contextualized within computer-mediated spatial commercial interactions. Perceived usefulness is theoretically defined as the prospective user’s subjective probability that employing a specific application system will boost job, decision, or task performance within an organizational or behavioral context. When translated to consumer shopping psychology, this construct captures the instrumental, outcome-oriented appraisal of a technology’s capacity to facilitate shopping goals with minimum cognitive friction and maximum diagnostic clarity.

Within cognitive and consumer psychology, this construct embodies several core psychological facets:

1. Instrumental Utility and Goal Attainment

Shopping behavior is fundamentally bifurcated into hedonic (experiential, entertainment-oriented) and utilitarian (instrumental, task-focused) modalities. ARSAU isolates the utilitarian dimension. Instrumental utility reflects the cognitive appraisal that the technology functions as an effective, goal-directed tool. When a consumer evaluates furniture, fashion accessories, or home appliances, their primary task-oriented goal is to determine whether the product fulfills functional, aesthetic, and spatial constraints. ARSAU captures whether the AR system functionally expedites the achievement of this explicit goal, validating the application as an efficient shopping vehicle rather than a distracting novelty.

2. Cognitive Load Reduction and Diagnostic Processing

Evaluating physical products through flat computer screens forces consumers to engage in extensive mental rotation, hypothetical spatial extrapolation, and cognitive estimation. This high cognitive workload frequently depletes working memory resources and introduces epistemic uncertainty. ARSAU measures the subjective psychological relief experienced when the AR application offloads this mental computation. By projecting an interactive, accurately scaled, 3D visual proxy into the user’s physical setting, the application automates visual-spatial alignment. The construct reflects the user’s perception that evaluation has become easier, frictionless, and cognitively economical.

3. Decisional Confidence and Performance Optimization

Shopping performance is not merely evaluated by speed; it is fundamentally determined by decision accuracy, error avoidance, and subjective confidence. An AR shopping application is perceived as useful when it sharpens discrimination between acceptable and unacceptable product alternatives. By enabling real-time visual inspection from 360-degree viewing angles, the tool mitigates pre-purchase ambiguity. The ARSAU construct specifically encompasses this performance optimization facet—quantifying the individual’s subjective appraisal that their evaluation was more thorough, their product choice was superior, and their overall shopping execution was substantively enhanced.

6. Theoretical Framework

The conceptual architecture of the ARSAU scale is rooted in an integration of information systems theory, cognitive psychology, and visual human-computer interaction. Specifically, the instrument synthesizes the Technology Acceptance Model, Cognitive Fit Theory, and the principles of Situated Cognition.

The Technology Acceptance Model (TAM)

The foundational bedrock of ARSAU originates from Davis’s (1989) seminal formulation of the Technology Acceptance Model, which itself evolved from the Theory of Reasoned Action (TRA). Davis postulated that behavioral intention to utilize any technological innovation is predominantly dictated by two fundamental cognitive assessments: Perceived Usefulness (PU) and Perceived Ease of Use (PEOU). While PEOU captures the perceived intrinsic operational difficulty of the system, PU captures the extrinsic instrumental reward—the expectation that technology usage directly improves performance outcomes. Childers et al. (2001) systematically tailored Davis’s generalized workplace PU scale to retail commerce (the “Web@Work” framework), confirming that utilitarian usefulness remains a powerful, non-negotiable determinant of online shopping attitudes even when hedonic motivations are operative. Hilken et al. (2017) adapted this established lineage, refining the scale items to capture the distinct technological capabilities of augmented reality.

Cognitive Fit Theory (CFT)

Formulated by Iris Vessey (1991), Cognitive Fit Theory posits that task performance improves when the format of information presentation directly matches the cognitive processes required by the task at hand. Spatial product assessment (e.g., verifying if an armchair harmonizes with living room decor or fits into a specific corner) is fundamentally a visual-spatial diagnostic problem. Traditional textual descriptions or decoupled 2D photographs force an intellectual mismatch: consumers must convert symbolic or flat representations into dynamic mental simulations. Augmented reality provides pristine cognitive fit by presenting spatial visual information in a natively spatial, interactive visual format. Under CFT, the ARSAU scale measures the consumer’s subjective realization of this cognitive fit, capturing the transition from laborious mental synthesis to effortless perceptual confirmation.

Situated Cognition and Spatial Visual Epistemology

Situated Cognition theory asserts that human cognitive processing cannot be fully understood in isolation from the physical and environmental contexts in which it takes place. Learning, evaluation, and decision-making are intimately tethered to situated environmental interactions. In standard e-commerce, the product is evaluated out of context. Augmented reality enacts situated cognition by conceptually embedding the prospective item directly into the immediate environmental ecosystem of the consumer. This contextual integration grounds the diagnostic evaluation. By eliminating the disconnect between the target object and the situational environment, AR transforms abstract technological evaluation into direct, situated perception. The ARSAU captures this epistemic shift, reflecting the user’s perception that situated inspection yields superior diagnostic information.

7. Validity

The psychometric validity of the AR Shopping App Usefulness scale has been comprehensively established across multiple empirical studies conducted by Hilken et al. (2017), as well as independent subsequent investigations in consumer psychology, spatial computing, and digital marketing. Validity evaluations have encompassed construct, convergent, discriminant, and predictive/criterion validities.

Construct and Convergent Validity

Convergent validity evaluates whether the scale items that theoretically represent perceived usefulness correlate strongly with one another, demonstrating that they successfully converge on the underlying target construct. Across the three distinct experimental studies conducted by Hilken et al. (2017)—which involved varied product domains ranging from high-involvement home furnishings to personal aesthetic items (e.g., sunglasses, cosmetics):

• Factor Loadings: All standardized item factor loadings on the latent AR usefulness construct consistently exceeded the rigorous academic threshold of 0.80, with empirical values ranging between 0.84 and 0.94 (all statistically significant at p < .001).
• Average Variance Extracted (AVE): The AVE metrics across all experimental test environments substantially surpassed the conventional Fornell-Larcker benchmark of 0.50. Reported AVE figures for the ARSAU typically ranged between 0.72 and 0.83, verifying that the latent construct accounts for well over 70% of the variance observed across its individual measurement items.

Discriminant Validity

Discriminant validity confirms that AR usefulness represents an empirically discrete phenomenon, readily distinguishable from other concurrent cognitive and affective states elicited by augmented reality environments:

• Fornell-Larcker Criterion: In all structural model assessments, the square root of the AVE for ARSAU exceeded the highest inter-construct correlation observed between AR usefulness and any other latent variable in the model—including perceived enjoyment, spatial presence, perceived environmental embedding, and decision comfort.
• Heterotrait-Monotrait (HTMT) Ratio: Subsequent replications employing contemporary PLS-SEM guidelines have demonstrated HTMT ratios involving ARSAU consistently below 0.85, establishing strong empirical divergence from purely hedonic or immersive affective variables. Users clearly differentiate between finding an AR application fun/novel versus evaluating it as a functionally useful shopping aid.

Predictive and Nomological Validity

Nomological and predictive validities reflect the degree to which scale scores behave in theoretical alignment with established network hypotheses. In the empirical frameworks evaluated by Hilken et al. (2017), AR usefulness demonstrated robust, statistically significant predictive pathways to vital behavioral and psychological endpoints:

• Decision Comfort: ARSAU exhibited a significant positive direct effect on consumers’ decision comfort, demonstrating that perceived utility alleviates pre-decisional distress and choice uncertainty.
• Patronage and Purchase Intentions: Path analyses confirmed that AR usefulness significantly drives downstream behavioral intentions, mediating the relationship between AR interface quality (e.g., real-time tracking, spatial realism) and commercial transactions (e.g., willingness to buy, willingness to revisit the digital storefront).
• Mitigation of Return Propensity: Extended research programs validating the scale have confirmed that higher ARSAU evaluations predict lower post-purchase regret and reduced anticipated product return rates, solidifying its predictive power in commercial retail science.

8. Reliability

The AR Shopping App Usefulness scale demonstrates exceptional empirical reliability across diverse experimental paradigms, sample populations, and retail categories. Reliability has been verified through multiple classical test theory metrics and modern structural equation modeling indices.

Internal Consistency

Internal consistency reflects the degree to which all items within the test measure the same underlying construct with minimal measurement error:

• Cronbach’s Alpha (α): Across the empirical investigations reported by Hilken et al. (2017), the Cronbach’s alpha coefficients for the ARSAU scale consistently demonstrated remarkable internal consistency. In Study 1, the scale achieved an α of 0.89; in Study 2, α was recorded at 0.91; and in Study 3, α reached 0.93. All reported values comfortably exceed Nunnally and Bernstein’s stringent 0.80 benchmark for established research instruments, confirming negligible random error.
• Composite Reliability (CR): Because Cronbach’s alpha can underestimate reliability due to its assumption of tau-equivalence (equal factor loadings), Composite Reliability was evaluated. Across the empirical studies, CR values ranged from 0.90 to 0.94, surpassing the standard recommended threshold of 0.70 by an extensive margin.

Test-Retest Stability and Cross-Sample Robustness

The scale possesses stable operational performance across distinct experimental contexts, demographic segments, and shopping modes:

• Cross-Category Stability: The internal consistency metrics maintain stability regardless of whether the target product category is characterized by utilitarian features (e.g., kitchen appliances, modular office desks) or aesthetic-experiential attributes (e.g., designer eyewear, decorative home accessories).
• Platform Equivalence: Evaluations across mobile smartphone interfaces (iOS ARKit, Android ARCore) and browser-based WebAR frameworks demonstrate comparable reliability coefficients, confirming that the scale’s measurement precision is invariant to client-side hardware variations.

9. Factor Analysis

Both Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA) have been conducted to delineate and verify the internal latent architecture of the AR Shopping App Usefulness scale.

Exploratory Factor Analysis (EFA)

During preliminary instrument adaptation, principal axis factoring with promax and varimax rotations was conducted to verify item dimensionality:

• Eigenvalue Evaluation: Extraction yielded a single unambiguous factor with an eigenvalue exceeding 2.45, accounting for between 78% and 86% of the total shared item variance across experimental sets.
• Scree Plot Analysis: Visual inspection of the scree plot revealed a distinct, sharp elbow following the initial factor, confirming the absence of secondary or residual multidimensional sub-factors.
• Item Loadings: All three items loaded exclusively onto the single perceived usefulness component, with unrotated factor loadings exceeding 0.85 and negligible communality residuals.

Confirmatory Factor Analysis (CFA)

To rigorously test the theoretical single-factor model against empirical data, CFA was implemented utilizing maximum likelihood estimation techniques within structural equation modeling software (e.g., AMOS, Mplus, R package lavaan):

• Standardized Factor Loadings (λ):
  • Item 1 (Performance Enhancement): λ = 0.86 – 0.91
  • Item 2 (Evaluation Ease / Process Simplification): λ = 0.88 – 0.94
  • Item 3 (Overall Shopping Utility): λ = 0.85 – 0.92
• Model Fit Indices: In fully saturated single-factor three-item measurement models (which yield zero degrees of freedom, df = 0), the construct functions as just-identified. When integrated into comprehensive multi-construct measurement models comprising perceived enjoyment, spatial presence, cognitive load, and purchase intentions, the global measurement models yielded exemplary goodness-of-fit statistics across studies:
  • Comparative Fit Index (CFI): 0.97 – 0.99 (Standard benchmark ≥ 0.95)
  • Tucker-Lewis Index (TLI): 0.96 – 0.98 (Standard benchmark ≥ 0.95)
  • Root Mean Square Error of Approximation (RMSEA): 0.038 – 0.052 (90% CI [0.021, 0.068], standard benchmark ≤ 0.06)
  • Standardized Root Mean Square Residual (SRMR): 0.024 – 0.035 (Standard benchmark ≤ 0.08)
  • Chi-Square / Degrees of Freedom Ratio (χ²/df): Consistently between 1.25 and 1.85, well below the conservative maximum cutoff of 2.0.

10. Instrument / Measurement Tool

The AR Shopping App Usefulness scale is an ultra-brief, standardized self-report psychometric test designed for rapid, low-burden administration in laboratory experiments, field surveys, and live e-commerce mobile intercept evaluations.

Structural Specifications

• Instrument Type: Standardized Self-Report Psychometric Questionnaire.
• Target Respondent: Consumers, study participants, or interface testers engaging with an augmented reality shopping application or virtual product display system.
• Administration Duration: Approximately 30 to 60 seconds (minimizing survey fatigue and dropout rates).
• Item Count: 3 items (unidimensional structure).
• Response Format: 7-point Likert-type scale:
  • 1 = Strongly Disagree
  • 2 = Disagree
  • 3 = Somewhat Disagree
  • 4 = Neutral (Neither Agree nor Disagree)
  • 5 = Somewhat Agree
  • 6 = Agree
  • 7 = Strongly Agree

Scoring and Aggregation Protocols

• Item Polarities: All three items are positively phrased; no reverse-coded items are present.
• Mean Composite Score Calculation: Compute the unweighted arithmetic mean across all three answered items:
  
  ARSAU_Score = (Item_1 + Item_2 + Item_3) / 3
  
  This yields a continuous scale score ranging from 1.00 to 7.00, where higher scores indicate greater perceived usefulness.
• Sum Score Calculation (Optional): Alternatively, researchers may compute a cumulative raw score ranging from 3 to 21.
• Latent Variable Modeling: In structural equation modeling (SEM) applications, items should be specified as observed reflective indicators loading onto a single latent endogenous/mediational variable (AR Useful).
• Contextual Adaptability: For studies testing AR within physical brick-and-mortar retail settings (e.g., smart mirrors, in-store tablet kiosks, smart glasses), researchers can remove the word “online” from the prompt and statements without compromising the structural or nomological validity of the scale.

11. Permissions & Fee and Test Year

• Test Year: Originally adapted and empirically published in 2017 (building on Childers et al., 2001, and Davis, 1989).
• Copyright & Ownership: The intellectual structure of the adapted scale is detailed in the research publication by Hilken et al. (2017), published in the Journal of the Academy of Marketing Science by Springer Nature / Academy of Marketing Science.
• Permissions & Academic Usage: The scale items are publicly documented within the academic literature for non-commercial scholarly research, educational purposes, and scientific evaluation. In accordance with standard fair-use academic conventions, researchers may reproduce and employ the scale provided full bibliographic attribution is granted to the original authors (Hilken et al., 2017) and foundational sources (Childers et al., 2001; Davis, 1989).
• Fee: There are no licensing fees or royalties required for academic, university, or scientific use. Commercial organizations, private market research firms, or proprietary software developers seeking to integrate the scale into commercial benchmarking toolkits should consult the publisher’s copyright clearance center and the original authors.

12. References

• 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
• Hilken, T., de Ruyter, K., Chylinski, M., Mahr, D., & Keeling, D. I. (2017). Augmenting the eye of the beholder: Exploring the strategic potential of augmented reality to enhance online service experiences. Journal of the Academy of Marketing Science, 45(6), 884–905. https://doi.org/10.1007/s11747-017-0541-x
• Vessey, I. (1991). Cognitive fit: A theoretical framework for evaluating cognitive processes in information systems. Decision Sciences, 22(2), 219–240. https://doi.org/10.1111/j.1540-5915.1991.tb00344.x

13. Items of the Scale