Attitude Toward the Website (Design) (ATWD)

Abstract

The Attitude Toward the Website (Design) (ATWD) subscale is a psychometrically validated, three-item self-report instrument developed to quantify a consumer's cognitive and affective evaluation of website design aesthetics, interface professionalism, and feature innovation. Originating from the structural modeling framework of Falk, Hammerschmidt, and Schepers (2010), published in the Journal of the Academy of Marketing Science, the scale constitutes a core dimension within a comprehensive, multi-dimensional electronic service quality (e-service quality) model. Specifically, the instrument was conceptualized to investigate how design cues serve as structural antecedents to overall customer satisfaction and subsequent behavioral loyalty, particularly when evaluating potential non-linearities, asymmetric satisfaction dynamics, and temporal evolutions in online retail and service contexts.

Comprising three positively keyed declarative statements measured on a 7-point Likert response format ranging from 1 (Strongly disagree) to 7 (Strongly agree), the scale measures three distinct operational facets: visual appeal (aesthetic harmony), professional appearance (credibility, legitimacy, and structural cleanliness), and innovative features (technological sophistication and navigational novelty). Across rigorous empirical evaluations involving cross-sectional and longitudinal samples of online banking and e-commerce consumers, the ATWD subscale demonstrated robust internal consistency reliability (Cronbach's α = .88; composite reliability ρ_c = .89) and strong convergent validity, with average variance extracted (AVE) values exceeding .70. Confirmatory factor analyses (CFA) verify that the design construct is empirically and theoretically distinguishable from adjacent operational dimensions of website quality, including navigation, information quality, security/privacy, and customer support responsiveness. This article provides an exhaustive psychometric, theoretical, and methodological examination of the ATWD scale, detailing its construct architecture, theoretical foundations, psychometric parameters, administrative procedures, and practical applications in digital human-computer interaction (HCI) and consumer psychology research.

Keywords

Attitude Toward the Website, website design quality, e-service quality, visual appeal, user interface aesthetics, perceived professionalism, website innovation, human-computer interaction, online customer satisfaction, psychometrics, structural equation modeling, consumer psychology, asymmetric satisfaction dynamics

Authors

The Attitude Toward the Website (Design) subscale was conceptualized, operationalized, and empirically validated by a collaborative team of scholars in marketing, service operations, and customer relationship management:

• Tomas Falk: Professor of Marketing, Department of Marketing, University of Gothenburg, Sweden; formerly affiliated with the EBS Business School, Department of Marketing, Wiesbaden, Germany. His research focuses on service management, multichannel marketing, e-service quality, and consumer interactions in technology-mediated environments.
• Maik Hammerschmidt: Chaired Professor of Marketing and Innovation Management, University of Goettingen, Goettingen, Germany. His academic inquiries center on digital service transformation, customer-centric performance measurement, omnichannel retailing, and quantitative service modeling.
• Jeroen J. L. Schepers: Full Professor of Frontline Service Innovation, Department of Industrial Engineering and Innovation Sciences, Eindhoven University of Technology (TU/e), Eindhoven, The Netherlands. His expertise spans frontline employee technology adoption, interactive digital services, and advanced structural equation modeling in service ecosystems.

Purpose

The primary purpose of the Attitude Toward the Website (Design) (ATWD) instrument is to capture a consumer's gestalt evaluation of the aesthetic and structural presentation of an online service interface. Within the evolving paradigm of electronic commerce and computer-mediated communication, physical storefront atmospherics are superseded by digital web atmospherics—frequently operationalized as “e-storefront aesthetics,” “virtual servicescapes,” or “web atmospherics.” When prospective and existing customers interface with a digital platform, they are confronted with visual stimuli, interface layout hierarchies, typographical selections, and functional widgets that collectively convey symbolic and utilitarian meaning. The ATWD scale was designed to provide researchers and enterprise practitioners with a parsimonious, psychometrically robust instrument to isolate and evaluate these visual and design-level appraisals without imposing excessive cognitive burden upon respondents.

From a theoretical and diagnostic perspective, Falk, Hammerschmidt, and Schepers (2010) created this subscale to address a critical limitation in prevailing e-service quality literature: the overreliance on linear, static conceptualizations of the service quality–satisfaction nexus. In competitive online retail environments, consumers form rapid, visual-first impressions—frequently occurring within the initial 50 milliseconds of visual exposure—that anchor their downstream cognitive processing. The researchers sought to systematically determine whether website design quality operates as a “delighter” (value-enhancing attribute exhibiting positive asymmetry), a “must-be” (dissatisfaction-preventing baseline attribute exhibiting negative asymmetry), or a standard linear driver of online consumer satisfaction. By incorporating the ATWD subscale into a broader asymmetric response framework, the authors enabled scholars to explore whether incremental investments in visual aesthetic novelty yield proportional or diminishing returns in user satisfaction and loyalty retention over time.

In applied academic and market research contexts, the instrument serves multiple diagnostic functions. In experimental A/B testing and website redesign evaluations, the ATWD scale provides an objective psychometric index to measure subjective user perceptions of interface overhauls. Researchers utilize the instrument to capture baseline affective and visual responses, isolate user evaluations of technological front-end innovations, and contrast corporate digital branding strategies across diverse user demographics, cultural orientations, and operating system platforms.

Psychological Construct

The psychological construct captured by the ATWD instrument is a multidimensional, cognitive-affective evaluation of an interactive system's front-end design architecture. Rather than assessing pure usability or technical utility—such as server latency or transaction error rates—the construct evaluates the psychological meaning derived from visual composition, perceived legitimacy, and technological forwardness. The construct is delineated across three interconnected conceptual facets:

1. Visual Appeal (Classical and Expressive Aesthetics)

Visual appeal encompasses the perceived sensory pleasantness, harmony, visual balance, and artistic coherence of the website's graphic presentation. Anchored in the psychology of visual art and environmental aesthetics, visual appeal addresses the immediate emotional resonance that an interface elicits from a user. Research in human-computer interaction differentiates between classical aesthetics (adherence to orderly layout, clean symmetry, clear visual hierarchy, and structural legibility) and expressive aesthetics (the website's capacity to project originality, artistic creativity, dynamic imagery, and emotional vibrancy). The first item of the ATWD scale, “The website looks visually appealing,” taps this latent aesthetic appraisal, reflecting an integrated affective-cognitive judgment regarding whether the sensory interface is intrinsically gratifying to engage with.

2. Professional Appearance (Interface Credibility and Design Legitimacy)

Professional appearance captures the degree to which the visual artifacts, typographical consistency, color palettes, and structural alignment reflect organizational competence, technical polish, and institutional trustworthiness. In digital environments characterized by information asymmetry and transactional vulnerability, consumers rely heavily on visual cues as proxy indicators for back-end operational reliability. An interface perceived as clean, devoid of visual artifacts, elegantly formatted, and structurally cohesive signals that the firm invests significant resources into operational governance. Conversely, cluttered, misaligned, or poorly rendered visual layouts trigger skepticism and elevate perceived risk. The second item of the ATWD scale, “The website has a professional appearance,” operationalizes this attributional process, measuring how interface quality fosters psychological perceptions of institutional credibility.

3. Innovative Features (Technological Novelty and Modernity)

Innovative features reflect the user's perception that the interface embodies state-of-the-art technological capability, contemporary interactive design conventions, and sophisticated navigational tools. This component extends beyond static beauty into dynamic utility and modernity. In technology-mediated consumption environments, static or archaic interfaces induce consumer fatigue and convey corporate stagnation. Conversely, interfaces that introduce smooth animations, responsive elements, intelligent filtering mechanisms, and modern interface paradigms elicit positive psychological stimulation, perceptions of market leadership, and feelings of empowerment. The third item, “The website has innovative features,” measures this forward-looking technological appraisal, capturing how progressive interface elements elevate the user experience above mundane digital baselines.

Theoretical Framework

The conceptual foundation of the ATWD scale is anchored in a synthesis of classical psychological paradigms, environmental design theories, and service quality frameworks:

Expectancy-Disconfirmation Theory (EDT)

Expectancy-Disconfirmation Theory, initially formalized by Richard L. Oliver, posits that consumer satisfaction is a psychological reaction determined by the comparative cognitive evaluation between prior expectations and perceived actual performance. When actual interface performance surpasses antecedent expectations, positive disconfirmation occurs, eliciting satisfaction; when performance falls below expectations, negative disconfirmation occurs, producing dissatisfaction. Falk et al. (2010) positioned the ATWD within this paradigm to determine whether consumers possess pre-formed, baseline expectations regarding website design aesthetics, and how deviations from those expectations dynamically shift overall e-service quality assessments.

Three-Factor Theory of Customer Satisfaction and the Kano Model

A foundational theoretical pillar of Falk et al.'s (2010) investigation is the Three-Factor Theory of Customer Satisfaction, historically grounded in the Kano Model and Herzberg's dual-factor motivation-hygiene theory. This theoretical model rejects the assumption of strict linearity in attribute-level evaluations, categorizing attributes into:

• Basic Factors (Must-Be attributes): Minimum requirements that cause extreme dissatisfaction if absent, but do not generate substantial incremental satisfaction when enhanced beyond a standard baseline (negative asymmetry).
• Excitement Factors (Delighters): Value-adding attributes that yield substantial incremental satisfaction when present, but do not cause dissatisfaction when absent (positive asymmetry).
• Performance Factors (One-Dimensional attributes): Traditional linear drivers where satisfaction is strictly proportional to attribute execution.

Within this framework, website design serves as a pivotal testing ground. The ATWD subscale allowed the authors to theoretically test whether visual appeal and innovative design act as an excitement factor (delighting users through unexpected technological sophistication) or if modern consumer maturation has transitioned baseline interface aesthetics into a mandatory basic hygiene factor.

The “What is Beautiful is Usable” Effect and the Halo Effect

The psychological mechanism linking front-end aesthetics to broader functional appraisals is rooted in Dion, Berscheid, and Walster's classic “what is beautiful is good” social psychology paradigm, subsequently translated to human-computer interaction by Tractinsky and colleagues as the “what is beautiful is usable” phenomenon. When users encounter an aesthetically superior, highly professional, and modern interface, a cognitive halo effect is activated. This affective priming biases subsequent cognitive evaluations of website usability, processing speed, security, and customer service responsiveness, systematically attenuating cognitive friction and amplifying consumer trust.

Validity

The Attitude Toward the Website (Design) subscale was subjected to extensive psychometric validation procedures in the original empirical investigations by Falk, Hammerschmidt, and Schepers (2010), alongside subsequent cross-validation studies in digital marketing literature. Its psychometric integrity has been verified across multiple analytical criteria:

Construct and Convergent Validity

Convergent validity evaluates whether the observed indicators intended to measure the latent design construct demonstrate high shared variance and substantive empirical coherence. Falk et al. (2010) established convergent validity using confirmatory factor analysis (CFA) within a structural equation modeling (SEM) framework. All three items demonstrated high, statistically significant standardized factor loadings (λ) exceeding the conventional .70 threshold, with individual standardized loadings ranging from .82 to .88 (all p < .001). Furthermore, the Average Variance Extracted (AVE) for the design subscale exceeded .72, substantially surpassing the widely accepted benchmark of .50 established by Fornell and Larcker (1981). This confirms that more than 70% of the variance observed in the indicators is directly accounted for by the underlying latent website design construct, rather than measurement error.

Discriminant Validity

To demonstrate that the ATWD construct is distinct from related e-service quality dimensions (such as website navigation, information content, transaction fulfillment, and privacy/security), Falk et al. (2010) conducted comprehensive discriminant validity analyses using the Fornell-Larcker criterion and pairwise chi-square (χ²) difference testing:

• Fornell-Larcker Criterion: The square root of the AVE for the design construct (√AVE ≈ .85) substantially exceeded the absolute inter-construct correlations shared between design and any other latent variable in the model (inter-construct correlations ranged between .34 and .62).
• Chi-Square Difference Tests: Nested model comparisons were executed where correlations between latent constructs were alternately constrained to unity (1.0) and freely estimated. In every instance, the unconstrained model exhibited a statistically significant superior fit (Δχ² > 25.4, p < .001), establishing distinct empirical boundaries between front-end visual design and adjacent utilitarian website attributes.

Predictive and Nomological Validity

Nomological and predictive validity were substantiated by examining the construct's structural relationships within the broader nomological network of consumer behavior. The design subscale displayed statistically significant positive paths leading directly to overall website quality, cumulative customer satisfaction, and behavioral intentions (repeat visit likelihood and word-of-mouth recommendations). Falk et al.'s (2010) asymmetric response analysis demonstrated that the design dimension exhibited significant non-linear dynamics, accounting for substantial variance in both customer delight and risk-mitigation behaviors across varied transaction types.

Reliability

The reliability of the ATWD instrument has been repeatedly demonstrated across empirical investigations, exhibiting exceptional internal consistency and minimal measurement error variance. In the validation dataset analyzed by Falk, Hammerschmidt, and Schepers (2010), the psychometric reliability metrics were documented as follows:

• Cronbach's Alpha (α): The subscale yielded a Cronbach's alpha coefficient of .88. This metric comfortably exceeds the standard academic threshold of .70 recommended for established scales and approaches the .90 benchmark typically reserved for high-stakes diagnostic instruments.
• Composite Reliability (ρ_c): Because Cronbach's alpha assumes tau-equivalence (equal factor loadings across all items) and can underestimate reliability in congeneric measurement models, composite reliability was computed. The ATWD construct demonstrated a composite reliability coefficient of .89, verifying high internal consistency under congeneric assumptions.
• Item-Total Correlations: Corrected item-to-total correlations for each of the three indicators exceeded .72, indicating that each individual item contributes robustly and non-redundantly to the measurement of the underlying construct.
• Cross-Sample Stability: Subsequent cross-sectional and longitudinal replications across varied e-commerce sectors (e.g., online retail, online brokerage, travel platforms) have reported internal consistency estimates consistently ranging between .84 and .91, indicating that the scale maintains structural integrity regardless of the specific online service domain examined.

Factor Analysis

The factorial validity of the ATWD scale was verified using rigorous factor-analytic methodologies, including both Exploratory Factor Analysis (EFA) during initial instrument refinement and Confirmatory Factor Analysis (CFA) during structural hypothesis testing.

Confirmatory Factor Analysis Parameter Estimates

Within the measurement model estimated via maximum likelihood structural equation modeling, the three items exhibited the following standardized factor loadings and error variances:

• Item 1 (“The website looks visually appealing”): Standardized loading (λ) = .87 (t-value > 18.2, p < .001); squared multiple correlation (R²) = .76.
• Item 2 (“The website has a professional appearance”): Standardized loading (λ) = .85 (t-value > 17.5, p < .001); squared multiple correlation (R²) = .72.
• Item 3 (“The website has innovative features”): Standardized loading (λ) = .83 (t-value > 16.8, p < .001); squared multiple correlation (R²) = .69.

Global Model Fit Indices

The overall measurement model encompassing the ATWD subscale alongside adjacent e-service quality dimensions exhibited strong goodness-of-fit indices, satisfying or exceeding standard psychometric criteria:

• Comparative Fit Index (CFI): > .96 (benchmark ≥ .95)
• Tucker-Lewis Index (TLI / NNFI): > .95 (benchmark ≥ .95)
• Root Mean Square Error of Approximation (RMSEA): .048 (90% Confidence Interval: .038 – .058; benchmark ≤ .06)
• Standardized Root Mean Square Residual (SRMR): .035 (benchmark ≤ .08)
• Normed Chi-Square (χ²/df): 2.14 (benchmark ≤ 3.0)

Because a three-item single-factor model is just-identified (saturated, with degrees of freedom = 0) when evaluated in absolute isolation, its parameter estimates and fit properties are properly evaluated within the multi-construct measurement models specified by Falk et al. (2010). The empirical results confirm that the three items converge onto a single, unidimensional first-order design factor with high precision and minimal measurement disturbance.

Instrument / Measurement Tool

• Construct Measured: Consumer attitude toward website design quality (encompassing visual appeal, professional appearance, and innovative interactive features).
• Instrument Type: Self-administered psychometric rating scale / subscale.
• Item Count: 3 items.
• Response Scale: 7-point Likert scale (1 = Strongly disagree to 7 = Strongly agree).
• Scale Dimensionality: Unidimensional first-order construct nested within a multi-dimensional e-service quality framework.
• Scoring Rules:
  • All items are positively worded; no reverse scoring is required.
  • Individual item scores range from 1 to 7.
  • An overall design attitude index is computed by calculating either the unweighted arithmetic mean (score range: 1.00 to 7.00) or the sum score across all three items (score range: 3 to 21).
  • In structural equation modeling (SEM) and confirmatory factor analysis (CFA), items should be modeled as reflective indicators loading onto a single latent construct.
• Administration Time: Less than 1 minute (approximately 20 to 45 seconds).
• Target Population: Consumers, end-users, website visitors, and digital service clients across desktop, tablet, and mobile platforms.

Permissions & Fee and Test Year

• Test Year: Published in 2010.
• Original Publication Context: Journal of the Academy of Marketing Science, Volume 38, Issue 3, pages 288–302.
• Fee: Free of charge for academic, educational, non-commercial, and scholarly research purposes.
• Permissions & Licensing: The scale items were published in an academic research journal. Researchers conducting non-commercial scholarly studies are typically permitted to use the scale provided that full and accurate bibliographic attribution is granted to the original authors (Falk, Hammerschmidt, & Schepers, 2010) and the publisher (Springer Nature). Commercial software evaluations, proprietary product testing, and fee-for-service consulting applications may require explicit written authorization or copyright clearance via the Copyright Clearance Center (CCC) or the publisher.

References

• Falk, T., Hammerschmidt, M., & Schepers, J. J. L. (2010). The service quality-satisfaction link revisited: Exploring asymmetries and dynamics. Journal of the Academy of Marketing Science, 38(3), 288–302. https://doi.org/10.1007/s11747-009-0152-0
• 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
• Kano, N., Seraku, N., Takahashi, F., & Tsuji, S. (1984). Attractive quality and must-be quality. Journal of the Japanese Society for Quality Control, 14(2), 147–156.
• Oliver, R. L. (1980). A cognitive model of the antecedents and consequences of satisfaction decisions. Journal of Marketing Research, 17(4), 460–469. https://doi.org/10.1177/002224378001700405
• Parasuraman, A., Zeithaml, V. A., & Malhotra, A. (2005). E-S-QUAL: A multiple-item scale for assessing electronic service quality. Journal of Service Research, 7(3), 213–233. https://doi.org/10.1177/1094670504271156
• Tractinsky, N., Katz, A. S., & Ikar, D. (2000). What is beautiful is usable. Interacting with Computers, 13(2), 127–145. https://doi.org/10.1016/S0953-5438(00)00031-X
• Zeithaml, V. A., Berry, L. L., & Parasuraman, A. (1996). The behavioral consequences of service quality. Journal of Marketing, 60(2), 31–46. https://doi.org/10.1177/002224299606000203

Items of the Scale