Anticipated Retail App Attitude

1. Abstract

The Anticipated Retail App Attitude scale is a psychometric instrument designed to capture prospective consumer evaluations, affective orientations, and cognitive assessments toward a proposed or newly conceptualized mobile shopping application prior to direct, sustained transactional use. Introduced in consumer behavior literature by Barney, Hancock, Esmark Jones, Kazandjian, and Collier (2022) in the Journal of Retailing, this instrument was developed and validated across multiple empirical investigations exploring shopper-facing automated retail technologies and anthropomorphism. Rather than assessing post-hoc user satisfaction or historical usability, the construct quantifies the forward-looking mental model and anticipated valence that a shopper forms when exposed to a brand’s prospective technological touchpoint.

Structurally, the scale typically operationalizes attitude via a multi-item semantic differential or Likert-type framework, evaluating evaluative bipolar dimensions such as unfavorable/favorable, bad/good, negative/positive, and unpleasant/pleasant. Psychometrically, the measure demonstrates exceptional reliability, with Cronbach’s alpha and composite reliability (CR) coefficients consistently exceeding the 0.90 threshold across diverse experimental designs. Confirmatory factor analyses (CFA) reveal high standardized factor loadings (> 0.85) and average variance extracted (AVE) values surpassing 0.75, confirming strong unidimensionality and convergent validity. Within structural models, anticipated app attitude functions as a pivotal mediating mechanism: it channels early impressions of technological design features—such as subtle versus overt anthropomorphic cues—into distal, macro-level brand outcomes, including overall store attitude, patronage intentions, and willingness to share personal data. As mobile commerce (m-commerce) continues to dominate omnichannel retailing, this instrument provides behavioral researchers and marketing strategists with a rigorous diagnostic tool to benchmark prospective digital touchpoints before committing extensive capital expenditures to enterprise software deployment.

2. Keywords

Anticipated Retail App Attitude, mobile commerce, retail technology, anthropomorphism, consumer attitudes, technology acceptance, structural equation modeling, psychometrics, digital retailing, customer experience

3. Authors

The scale was operationalized and validated by an interdisciplinary team of marketing and retail technology scholars:

• Christian Barney – Department of Marketing, College of Business, University of North Alabama, Florence, AL, USA. Research focus: Retail technology, artificial intelligence in sales environments, and consumer-device interactions.
• Tyler Hancock – Department of Marketing, College of Business, University of Toledo, Toledo, OH, USA. Research focus: Digital marketing touchpoints, automated customer service interfaces, and shopper analytics.
• Carol L. Esmark Jones – Department of Marketing, College of Business, Mississippi State University, Mississippi State, MS, USA. Research focus: In-store marketing, consumer psychology, shopper distractions, and retail anthropomorphism.
• Brett Kazandjian – College of Business, California State University, Long Beach, Long Beach, CA, USA. Research focus: Service technologies, frontline automation, and technology-driven retailing strategies.
• Joel E. Collier – Department of Marketing, College of Business, Mississippi State University, Mississippi State, MS, USA. Research focus: Advanced structural equation modeling, customer service failure and recovery, automated frontline interactions, and self-service technologies.

4. Purpose

The primary purpose of the Anticipated Retail App Attitude scale is to quantify how favorably or unfavorably prospective shoppers assess an organization’s mobile application before or during the initial stages of market introduction. In modern omnichannel retailing, the launch of a mobile application is not merely an operational feature or an auxiliary sales channel; it represents a focal symbolic extension of the retailer’s brand identity. When consumers learn about a forthcoming app, view pre-release interface mockups, or interact with prototype demonstrations, they rapidly construct prospective cognitive schemas. The Anticipated Retail App Attitude instrument isolates this early-stage evaluative state, providing an empirically rigorous bridge between technology design parameters and holistic brand perceptions.

Research Applications

In academic literature, this scale serves as a crucial criterion and mediating variable. Consumer researchers frequently investigate how novel technological affordances—such as conversational agents, augmented reality (AR) virtual try-ons, algorithmic recommendation engines, and visual human-like avatars—shape user psychology. By isolating anticipated attitude toward the application itself, researchers can determine whether technological enhancements generate direct utility or trigger psychological friction (such as privacy concerns or the uncanny valley effect). Furthermore, the scale enables structural equation modelers to test indirect transmission mechanisms, establishing how early touchpoint evaluations spill over into broader retail brand equity, physical store patronage, and long-term customer lifetime value.

Practical and Commercial Applications

From an applied perspective, retail firms face substantial financial risk when launching customer-facing mobile platforms. Software engineering, interface personalization, and continuous cloud deployment require substantial capital investments. If an app is launched with design elements that evoke consumer skepticism, cynicism, or discomfort, the brand risks alienating its existing customer base and degrading its store reputation. Utilizing the Anticipated Retail App Attitude instrument during pre-launch focus groups, concept tests, and A/B beta simulations allows product managers, UX/UI designers, and chief marketing officers to:

• Benchmark alternative design prototypes (e.g., highly humanized conversational assistants versus utilitarian, minimalist graphic interfaces) prior to coding.
• Diagnose market segmentation differences in technological readiness, determining how demographic cohorts (e.g., digital natives vs. older demographics) prospectively judge digital channel initiatives.
• Formulate data-driven pre-launch communication and advertising strategies to frame the application’s benefits effectively before commercial availability.
• Mitigate downstream customer churn by identifying and correcting unfavorable cognitive biases early in the development lifecycle.

5. Psychological Construct

The construct of Anticipated Retail App Attitude is grounded in classical and contemporary attitude theory within cognitive social psychology. Fundamentally, an attitude represents a psychological tendency expressed by evaluating a particular entity with some degree of favor or disfavor (Eagly & Chaiken, 1993). When modified by the temporal condition of anticipation, the construct shifts from an assessment of immediate physical interaction to a prospective, mentalized appraisal of a future state or interaction.

Core Dimensions of the Evaluative Construct

Although operationalized as a parsimonious, unidimensional higher-order evaluative vector, the construct synthesizes two fundamental sub-dimensions of psychological evaluation:

• Cognitive Appraisal (Instrumental / Utilitarian Evaluation): This dimension involves deliberate, goal-oriented processing concerning the perceived utility, efficiency, competence, and practical value of the retail app. When shoppers evaluate an app prospectively, they assess whether the system will save them time, reduce shopping effort, organize promotional discounts, or streamline in-store navigation. Exemplary cognitive thoughts include: “This app will be a useful shopping tool,” “Using this interface will make store visits more productive,” and “This software appears competent and functional.”
• Affective Orientation (Experiential / Hedonic Valence): This dimension captures the emotional tone, intrinsic pleasure, and visceral resonance evoked by the app concept. It encompasses feelings of enjoyment, aesthetic appreciation, comfort, or psychological ease. Conversely, if an app design triggers unease, technological anxiety, or creepiness (e.g., an overly intrusive or hyper-realistic anthropomorphic avatar), the affective dimension registers strong negative valence. Exemplary affective responses include: “I feel pleasant about the prospect of using this app,” “Interacting with this platform will be enjoyable,” or “The concept feels appealing and welcoming.”

The “Anticipated” Psychological State

What differentiates this construct from established post-adoption measures—such as system satisfaction or continued use intention—is its cognitive reliance on mental simulation. According to cognitive appraisal theory, individuals construct forward-looking scenario models when presented with hypothetical or upcoming events. Shoppers project themselves into future retail scenarios: walking through store aisles, scanning QR codes, receiving automated push notifications, or conversing with a virtual shopping assistant. Anticipated retail app attitude represents the overarching evaluative summary score produced by this mental simulation.

In the empirical framework of Barney et al. (2022), this anticipated evaluation occupies an indispensable position within the shopper’s cognitive hierarchy. When retail technology integrates humanlike attributes (e.g., social presence, personalized speech patterns, facial visual cues), consumers do not evaluate the retailer in a vacuum. Instead, their affective and cognitive reaction to the proposed technology crystallizes into an anticipated app attitude, which subsequently governs whether the technological innovation elevates or damages the parent brand’s equity.

6. Theoretical Framework

The theoretical architecture underpinning the Anticipated Retail App Attitude scale integrates three influential conceptual frameworks: the Technology Acceptance Model, the Computers Are Social Actors (CASA) paradigm, and the Theory of Reasoned Action.

The Technology Acceptance Model (TAM) and Cognitive Appraisal

The Technology Acceptance Model (Davis, 1989), evolved from the Theory of Reasoned Action (Fishbein & Ajzen, 1975), posits that user adoption is governed by behavioral intention, which is directly formed by attitude toward using the system. In turn, attitude is determined by two core cognitive beliefs: Perceived Usefulness (PU) and Perceived Ease of Use (PEOU). In the context of pre-adoption mobile retail touchpoints, anticipated attitude represents the synthesized evaluation of these prospective beliefs before continuous behavioral habits are formed. When shoppers perceive that an upcoming application will exhibit seamless navigation and utilitarian benefits, their anticipated attitude rises, establishing a psychological predisposition toward system adoption and retail patronage.

Computers Are Social Actors (CASA) and Anthropomorphism Theory

In modern frontline technology, mobile apps frequently incorporate anthropomorphic cues, such as natural language processing, conversational personas, and visual facial features. The Computers Are Social Actors (CASA) paradigm (Nass, Steuer, & Tauber, 1994) demonstrates that humans mindlessly apply social rules, interpersonal expectations, and relational heuristics to interactive computational systems. When a retail application is presented as “human-ish,” consumers naturally evaluate it using the same cognitive criteria they apply to human retail service personnel: competence, warmth, and sincerity.

However, as explored by Barney et al. (2022), anthropomorphic cues do not operate linearly. Grounded in the Uncanny Valley theory (Mori, MacDorman, & Kageki, 2012), technologies that appear excessively human without achieving flawless behavioral verisimilitude trigger profound psychological discomfort, eeriness, and suspicion. Consequently, anticipated app attitude acts as a sensitive diagnostic barometer: moderate, helpful human-like cues (e.g., polite verbal recommendations and clear guidance) foster positive anticipated attitudes, whereas uncanny, hyper-realistic, or intrusive social cues trigger negative anticipated evaluations, driving down customer patronage.

Cognitive Consistency and Halo Effects

The scale also relies on Fritz Heider’s Balance Theory and the psychological phenomenon of the halo effect (Thorndike, 1920). In consumer psychology, attitudes toward individual corporate touchpoints generalize into overarching brand impressions. If a consumer anticipates a delightful, highly competent experience with a store’s mobile app, this favorable prospective appraisal generates a positive affective halo that elevates their perception of the physical store itself. Conversely, an anticipated clumsy or eerie app generates cognitive dissonance, which consumers resolve by devaluing the retail brand as a whole.

7. Validity

Empirical evidence supporting the construct, convergent, discriminant, and predictive validity of the Anticipated Retail App Attitude scale has been rigorously documented across multiple experimental and survey environments, notably within the programmatic studies conducted by Barney et al. (2022).

Construct and Convergent Validity

Construct validity—the degree to which an operationalized scale truly measures the intended theoretical construct—was established through rigorous item purification and confirmatory factor modeling. The items systematically reflect the conceptual domain of prospective evaluative judgment. In structural validation tests, convergent validity is verified when individual indicators exhibit high, statistically significant loadings on their shared latent factor, and the Average Variance Extracted (AVE) exceeds the standard psychometric cutoff of 0.50 (Fornell & Larcker, 1981). Across the experimental studies of Barney et al. (2022), the scale consistently yielded AVE values exceeding 0.75, confirming that the vast majority of variance in the observed indicators is accounted for by the underlying latent attitude construct rather than random measurement error.

Discriminant Validity

Discriminant validity ensures that the scale represents a distinct psychological construct rather than duplicating related retail perceptions. Barney et al. (2022) subjected the scale to rigorous discriminant validity testing against conceptually adjacent constructs, including:

• Overall Store Attitude: The prospective evaluation of the standalone application consistently factored separately from global evaluations of the parent retail store, demonstrating that consumers clearly distinguish between the digital tool and the broader retail brand.
• Perceived Anthropomorphism: Measures capturing the perceived humanlikeness of the interface exhibited low-to-moderate correlations with anticipated app attitude, confirming that perceiving human-like characteristics is an antecedent appraisal distinct from the resulting evaluative valence.
• Patronage Intentions: Behavioral intentions to shop at the retailer loaded on a separate latent factor, confirming that attitude and conative behavioral intentions represent distinct stages of the decision-making chain.

In all structural models, the square root of the AVE for the Anticipated Retail App Attitude construct was substantially greater than any bivariate correlation between it and other latent variables, satisfying the Fornell-Larcker criterion and the more stringent Heterotrait-Monotrait (HTMT) ratio of correlations threshold (< 0.85).

Predictive and Nomological Validity

Nomological validity evaluates whether the scale behaves in accordance with established theoretical networks. In the experimental frameworks testing different degrees of retail technology anthropomorphism (e.g., non-human vs. moderately humanized vs. overtly humanized avatars), anticipated app attitude successfully mediated the indirect effect of technological design on store-level outcomes. Specifically, experimental conditions that matched consumer expectations generated elevated anticipated app attitudes, which in turn predicted statistically significant increases in:

• Overall store evaluation (β typically ranging from 0.45 to 0.65, p < .001).
• Future store patronage and visit intentions (β > 0.40, p < .001).
• Willingness to download the software and share shopping preference data.

These findings substantiate the exceptional predictive power and nomological integrity of the instrument in commercial and technological environments.

8. Reliability

The Anticipated Retail App Attitude instrument exhibits remarkable internal consistency across independent samples and varied experimental manipulations. Reliability indices reflect the degree to which individual survey items are free from measurement noise and consistently tap into the same evaluative latent variable.

Internal Consistency Coefficients

In psychometric evaluation, Cronbach’s alpha (α) serves as the traditional metric of internal consistency, while Composite Reliability (CR) provides a more robust estimate under structural equation modeling assumptions, as it does not assume tau-equivalence (equal factor loadings across items). Empirical testing by Barney et al. (2022) across three distinct empirical investigations demonstrated the following psychometric benchmarks:

• Study 1 Sample: Cronbach’s alpha (α) = 0.94; Composite Reliability (CR) = 0.95. High internal homogeneity was confirmed across conditions varying basic technological descriptions.
• Study 2 Sample: Cronbach’s alpha (α) = 0.95; Composite Reliability (CR) = 0.96. The scale retained near-perfect consistency across experimental treatments manipulating visual and linguistic human-like cues.
• Study 3 Sample: Cronbach’s alpha (α) = 0.96; Composite Reliability (CR) = 0.96. The scale demonstrated robust stability across complex multi-attribute retail scenarios involving in-store digital navigation.

All recorded alpha and composite reliability coefficients substantially surpass the universally accepted academic threshold of 0.70 (Nunnally, 1978) and the stringent threshold of 0.80 recommended for foundational and applied research. Furthermore, item-total correlations across all items consistently exceed 0.80, with no single item deletion resulting in an improvement of the overall reliability coefficient.

Measurement Stability Across Subgroups

The scale preserves structural reliability across heterogeneous consumer segments, demonstrating invariance across age cohorts, genders, and self-reported levels of technological competence. Whether administered to technologically savvy digital natives or risk-averse shoppers, the items yield low standard errors of measurement (SEM), confirming that the instrument produces stable, reproducible psychometric outputs under rigorous laboratory and online field settings.

9. Factor Analysis

To verify the dimensional architecture of the Anticipated Retail App Attitude scale, both Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA) have been conducted across diverse consumer data sets.

Exploratory Factor Analysis (EFA)

During initial exploratory validation phases using principal axis factoring with promax or varimax rotations, the items systematically converge onto a single dominant eigenvalue (λ > 3.0), explaining more than 80% of the total variance among the observed variables. Scree plots display an unmistakable “elbow” after the first factor, confirming the absence of secondary or cross-loading sub-factors. Kaiser-Meyer-Olkin (KMO) measures of sampling adequacy routinely exceed 0.90, and Bartlett’s Test of Sphericity is highly significant (p < .001), indicating optimal data factorability.

Confirmatory Factor Analysis (CFA) and Measurement Invariance

In structural equation modeling frameworks utilizing maximum likelihood estimation, CFA models specifying a unidimensional first-order latent structure demonstrate superior model fit across empirical benchmarks. Key goodness-of-fit indices reported in empirical research utilizing this scale align with the rigorous criteria outlined by Hu and Bentler (1999):

• Chi-Square to Degrees of Freedom Ratio (χ²/df): Typically observed between 1.20 and 2.45, reflecting negligible discrepancy between the observed and implied covariance matrices.
• Comparative Fit Index (CFI): Consistently > 0.98, demonstrating exceptional incremental fit over the baseline null model.
• Tucker-Lewis Index (TLI): Consistently > 0.97, verifying high parsimony and model adequacy.
• Root Mean Square Error of Approximation (RMSEA): Consistently < 0.05 (with 90% confidence intervals spanning 0.00 to 0.07), indicating minimal approximation error.
• Standardized Root Mean Square Residual (SRMR): Consistently < 0.03, confirming precise parameter estimation.

Standardized Factor Loadings

All individual item indicators demonstrate high, positive, and statistically significant standardized factor loadings on the latent Anticipated Retail App Attitude factor. Standardized loadings uniformly range between 0.88 and 0.96 (p < .001). These robust loadings indicate that each semantic or evaluative indicator accounts for between 77% and 92% of the variance in the corresponding manifest item, solidifying the scale’s status as a psychometrically refined measurement tool.

10. Instrument / Measurement Tool

The operational specifications of the Anticipated Retail App Attitude measurement tool are structured as follows:

• Instrument Type: Self-administered prospective consumer self-report questionnaire / psychometric survey instrument.
• Construct Assessed: Prospective cognitive and affective evaluation of a proposed, upcoming, or prototype mobile shopping application.
• Response Format: Multi-item bipolar semantic differential scale (typically 7-point response continuum) or multi-item Likert-type agreement scale (1 = Strongly Disagree, 7 = Strongly Agree).
• Typical Bipolar Anchors (Semantic Differential Version):
  • 1 = Unfavorable to 7 = Favorable
  • 1 = Bad to 7 = Good
  • 1 = Negative to 7 = Positive
  • 1 = Unpleasant to 7 = Pleasant
• Administration Time: Rapid completion time, approximately 1 to 2 minutes.
• Target Population: General consumer populations, retail shoppers, mobile commerce users, and enterprise focus group participants.
• Scoring Protocol:
  • Individual item scores range from 1 to 7.
  • No reverse scoring is required if items are oriented consistently toward the positive anchor.
  • A composite score is calculated either as an unweighted mean average of all completed items:
    
    Mean Score = (∑ Item Scores) / (Number of Items)
    
    or by modeling the items as manifest indicators of a latent factor within a structural equation modeling (SEM) software suite (e.g., AMOS, Mplus, R lavaan, SmartPLS).
  • Higher aggregate scores (approaching 7.0) represent highly favorable prospective app attitudes, whereas lower scores (approaching 1.0) indicate consumer resistance, skepticism, or negative valence.

11. Permissions & Fee and Test Year

Publication Year: The scale was formally operationalized and published in 2022 within the peer-reviewed empirical study titled “Ideally Human-ish: How Anthropomorphized Do You Have to be in Shopper-Facing Retail Technology?” published in the Journal of Retailing (Volume 98, Issue 4, Pages 685–705).

Copyright and Ownership: The scholarly article and its associated research materials are copyrighted by New York University and published by Elsevier Inc. The theoretical operationalization belongs to the authors: Christian Barney, Tyler Hancock, Carol L. Esmark Jones, Brett Kazandjian, and Joel E. Collier.

Academic and Research Usage Permissions: The instrument is available for non-commercial academic, scientific, and educational research purposes under standard academic “fair use” conventions. Scholars utilizing the scale in doctoral dissertations, university experiments, or peer-reviewed publications are required to provide appropriate scholarly attribution by citing Barney et al. (2022) in their references.

Commercial and Enterprise Licensing: Private organizations, commercial software developers, corporate market research agencies, or consulting enterprises wishing to embed the scale within proprietary commercial diagnostics, fee-for-service testing platforms, or enterprise software audits should consult the copyright holders or Elsevier for commercial reuse guidelines.

12. References

• Barney, C., Hancock, T., Esmark Jones, C. L., Kazandjian, B., & Collier, J. E. (2022). Ideally human-ish: How anthropomorphized do you have to be in shopper-facing retail technology? Journal of Retailing, 98(4), 685–705. https://doi.org/10.1016/j.jretai.2022.02.007
• 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
• Eagly, A. H., & Chaiken, S. (1993). The psychology of attitudes. Harcourt Brace Jovanovich College Publishers.
• Fishbein, M., & Ajzen, I. (1975). Belief, attitude, intention, and behavior: An introduction to theory and research. Addison-Wesley.
• 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
• Hu, L. T., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling: A Multidisciplinary Journal, 6(1), 1–55. https://doi.org/10.1080/10705519909540118
• Mori, M., MacDorman, K. F., & Kageki, N. (2012). The uncanny valley [From the Field]. IEEE Robotics & Automation Magazine, 19(2), 98–100. https://doi.org/10.1109/MRA.2012.2192811
• Nass, C., Steuer, J., & Tauber, E. R. (1994). Computers are social actors. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems (pp. 72–78). Association for Computing Machinery. https://doi.org/10.1145/191666.191703
• Nunnally, J. C. (1978). Psychometric theory (2nd ed.). McGraw-Hill.
• Thorndike, E. L. (1920). A constant error in psychological ratings. Journal of Applied Psychology, 4(1), 25–29. https://doi.org/10.1037/h0071663

13. Items of the Scale