Hierarchical Service Quality Scale

1. Abstract

The Hierarchical Service Quality Scale (HSQS), developed by Michael K. Brady and J. Joseph Cronin Jr. in 2001, represents a landmark psychometric and theoretical synthesis in the fields of consumer psychology, service marketing, and psychometrics. Prior to the formulation of the HSQS, the conceptualization and measurement of perceived service quality were sharply divided between two dominant paradigms: the Nordic school of thought led by Christian Grönroos (which emphasized technical and functional quality outcomes) and the American school led by A. Parasuraman, Valarie A. Zeithaml, and Leonard L. Berry (which framed quality through the five-dimensional SERVQUAL framework: reliability, responsiveness, empathy, assurance, and tangibles). Brady and Cronin unified these perspectives into a comprehensive, third-order hierarchical factor structure.

In this structural conceptualization, overall service quality operates as a third-order latent construct manifested through three primary, second-order dimensions: Interaction Quality, Physical Environment Quality, and Outcome Quality. Each primary dimension is in turn defined by three distinct, first-order subdimensions (comprising nine subdimensions in total): Interaction Quality consists of Attitude, Behavior, and Expertise; Physical Environment Quality comprises Ambient Conditions, Facility Design, and Social Factors; and Outcome Quality encompasses Waiting Time, Tangibles, and Valence. Each first-order subdimension is measured via three reflective survey items constructed using reliability, responsiveness, and empathy modifiers, yielding a baseline inventory of 27 subdimension items, complemented by global marker indicators for the primary dimensions and overall construct.

Administered primarily via a 7-point Likert or semantic differential scale, the instrument was rigorously validated using a structural equation modeling (SEM) framework across a diverse sample of 1,133 consumers across four distinct service sectors: fast-food restaurants, photographic development, amusement parks, and dry cleaning. Empirical evaluation demonstrated superior goodness-of-fit indices relative to alternative non-hierarchical, unidimensional, and first-order multidimensional models. The scale exhibits exceptional composite reliability coefficients (consistently exceeding .80 to .90 across subdimensions), strong convergent and discriminant validity, and high predictive power regarding customer satisfaction and behavioral intentions.

2. Keywords

Hierarchical Service Quality Scale, HSQS, perceived service quality, third-order factor model, interaction quality, physical environment quality, outcome quality, SERVQUAL, servicescape, psychometrics

3. Authors

The Hierarchical Service Quality Scale was conceptualized, operationalized, and psychometrically validated by:

• Michael K. Brady, Ph.D. — Department of Marketing, College of Business, Florida State University, Tallahassee, Florida, United States. Dr. Brady is the Bob Sasser Professor of Marketing and has published extensively on frontline service interactions, customer orientation, and structural models of consumer evaluations.
• J. Joseph Cronin Jr., Ph.D. — Department of Marketing, College of Business, Florida State University, Tallahassee, Florida, United States. Dr. Cronin is the John R. Kerr Eminent Scholar Chair in Marketing and a leading authority on service quality, customer satisfaction, value models, and consumer behavioral intentions.

Inquiries regarding the theoretical formulation of the original study can be directed to the Department of Marketing, College of Business, Florida State University, Tallahassee, FL 32306-1110, USA.

4. Purpose

The primary purpose of the Hierarchical Service Quality Scale is to offer an empirically validated, theoretically sound, and psychometrically robust measurement framework that captures the complex, multilevel nature of consumer perceptions of service quality. For decades, academic researchers and organizational practitioners struggled with notable inconsistencies in measuring service quality. The original SERVQUAL battery, while pioneering, faced persistent operational and statistical criticisms, including disputes over the conceptual appropriateness of difference scores (Expectations minus Perceptions), instability of its five hypothesized factors across divergent industry sectors, and an overemphasis on the functional delivery process at the expense of technical results.

Brady and Cronin (2001) addressed these fundamental deficiencies by operationalizing service quality as a multilevel, hierarchical construct. The purpose of the HSQS is threefold:

• Theoretical Integration: It resolves the long-standing debate between the Nordic paradigm (Grönroos, 1982, 1984), which asserts that service quality consists of what the consumer receives (technical quality) and how they receive it (functional quality), and the American paradigm (Parasuraman et al., 1988), which conceptualizes quality as process-driven evaluation attributes. The HSQS demonstrates that these paradigms are not mutually exclusive but rather hierarchically nested within a single cognitive assessment network.
• Diagnostic Granularity: From an applied diagnostic perspective, standard unidimensional or unilevel instruments fail to isolate the precise operational loci of consumer dissatisfaction. By evaluating three overarching primary dimensions through nine distinct subdimensions, the HSQS allows service organizations to determine whether customer dissatisfaction originates from employee social dynamics (interpersonal interaction), physical architecture or sensory ambient cues (servicescape), or the ultimate outcome of the core service exchange (tangibles, waiting time, or core valence).
• Predictive Efficacy in Consumer Research: The scale was engineered to serve as a robust predictor of critical consumer outcome variables, specifically cumulative customer satisfaction, brand equity, and future behavioral intentions (e.g., word-of-mouth referral, customer retention, willingness to pay a price premium). By measuring service quality at the highest level of abstraction while grounding it in concrete first-order experiences, researchers obtain superior predictive validity compared to legacy single-level instruments.

5. Psychological Construct

The construct measured by the HSQS is Perceived Service Quality, defined as a global, cognitive-affective consumer evaluation resulting from the comparison of expectations with actual service performance across multiple hierarchical tiers. Brady and Cronin structured this construct into a three-level hierarchy. At the apex sits the third-order construct of Overall Service Quality. This overarching abstraction is determined by three second-order primary dimensions, each manifested through three first-order subdimensions.

Primary Dimension 1: Interaction Quality

Interaction Quality captures the dynamic, interpersonal processes that occur between the customer and service personnel during the delivery phase. Recognizing that services are intrinsically relational and performative, this dimension assesses the human element of service exchange through three subdimensions:

• Attitude: The customer’s perception of the service provider’s disposition, emotional tone, courtesy, and willingness to assist. For example, an employee displaying authentic warmth, enthusiasm, and an attentive demeanor projects a positive attitude that directly elevates the customer’s perceived interaction quality.
• Behavior: The specific observable actions, operational competence, responsiveness, and procedural professionalism displayed by frontline staff. For example, an employee who anticipates customer requirements, resolves situational problems swiftly, and follows through on verbal commitments demonstrates exemplary behavioral performance.
• Expertise: The perceived knowledge, technical mastery, domain fluency, and professional capability of the service provider. For instance, a financial consultant or a technical repair specialist who clearly explains complex issues and diagnoses problems accurately instills consumer confidence through high perceived expertise.

Primary Dimension 2: Physical Environment Quality

Drawing on environmental psychology and the concept of the servicescape (Bitner, 1992), Physical Environment Quality reflects the ambient, architectural, and social surroundings within which the service is enacted. It comprises three subdimensions:

• Ambient Conditions: Non-visual, background sensory stimuli that affect human physiological and psychological comfort, including temperature, lighting, acoustics, background music, scent, and air quality. For instance, an environment that is excessively noisy, poorly illuminated, or uncomfortably warm degrades the perceived physical environment quality regardless of employee behavior.
• Facility Design: The architectural layout, aesthetic beauty, functional spatial arrangement, signage, and ergonomic furnishing of the service facility. For example, an intuitive floor plan, high-end finishing materials, comfortable seating, and clear navigation elevate the aesthetic and practical usability of the service site.
• Social Factors: The presence, visual appearance, density, and behavioral conduct of other customers sharing the service environment. As consumers frequently co-create the service experience, overcrowding, loud fellow patrons, or poorly dressed customers can negatively influence a customer’s evaluation of the overall atmosphere.

Primary Dimension 3: Outcome Quality

Outcome Quality captures the post-consumption appraisal of what the consumer actually takes away from the service encounter—representing the technical result of the transaction. It consists of three subdimensions:

• Waiting Time: The customer’s subjective evaluation of the duration, efficiency, and perceived fairness of the time spent queueing or waiting for service execution. Even if an outcome is satisfactory, excessive or unexplained delays degrade this subdimension.
• Tangibles: The physical, tangible evidence or artifacts left in the customer’s possession or observed after the service has been rendered. In a restaurant, this includes the food presentation and tableware; in dry cleaning, the pristine condition and protective packaging of the garments.
• Valence: The customer’s holistic evaluation of whether the outcome of the service encounter was inherently positive or negative, independent of the process itself. Valence captures unexpected attributes or situational eventualities that dictate the ultimate success or failure of the service goal (e.g., whether a sports team won a game or an event went smoothly despite minor logistical hiccups).

6. Theoretical Framework

The theoretical architecture of the Hierarchical Service Quality Scale is rooted in advanced cognitive appraisal theories, mental models of categorical hierarchy, and decades of conceptual development across Nordic and American services marketing scholarship.

The Nordic vs. American Paradigms

In the early formulation of service quality theory, Christian Grönroos (1982, 1984) proposed that service evaluations are bifurcated into technical quality (the “what”—the outcome that the consumer receives from the firm) and functional quality (the “how”—the expressive, relational manner in which the service is transferred). Grönroos argued that corporate image moderates and filters these perceptions.

Conversely, Parasuraman, Zeithaml, and Berry (1985, 1988) formulated the American school, postulating that consumers evaluate service quality across five generalized functional dimensions: Tangibles, Reliability, Responsiveness, Assurance, and Empathy. While SERVQUAL gained widespread empirical traction, critics noted that its five dimensions predominantly captured functional, process-oriented elements, largely omitting the technical outcome of the service exchange.

Multilevel and Hierarchical Mental Models

To resolve this fragmentation, Brady and Cronin drew inspiration from cognitive psychology models of categorization (e.g., Eleanor Rosch’s prototype theory and hierarchical categorization) and the multi-level retail service quality framework established by Pratibha A. Dabholkar, C. David Shepherd, and Dayle I. Thorpe (1996). Dabholkar et al. posited that retail customers form service quality evaluations across three cognitive tiers: overall retail quality, primary dimensions, and subdimensions.

Brady and Cronin integrated this hierarchical view with Rust and Oliver’s (1994) conceptual three-component model, which hypothesized that service quality comprises the service product (outcome), service delivery (interaction), and service environment. Brady and Cronin operationalized these three components as the primary, second-order dimensions of their model, as illustrated in the schematic below:

                [ Level 3: Overall Service Quality ]
                                 |
      +--------------------------+--------------------------+
      |                          |                          |
[ Level 2:                 [ Level 2:                 [ Level 2:
 Interaction Quality ]    Physical Env. Quality ]    Outcome Quality ]
      |                          |                          |
  +---+---+                  +---+---+                  +---+---+
  |   |   |                  |   |   |                  |   |   |
[Att][Beh][Exp]            [Amb][Des][Soc]            [Wai][Tan][Val]
 (Level 1: 9 Subdimensions, each measured by Reliability, Responsiveness, & Empathy modifiers)
  

Integration of SERVQUAL Descriptors as Modifiers

A crucial theoretical innovation introduced by Brady and Cronin is how they treated the classical SERVQUAL dimensions. Rather than treating Reliability, Responsiveness, and Empathy as independent, overarching primary dimensions, the authors demonstrated that these attributes act as modifiers or sub-attributes that consumers use to evaluate each of the concrete first-order subdimensions. For example, when a customer evaluates frontline employee Attitude or Behavior, they judge how reliably, responsively, and empathetically that attitude or behavior is manifested.

7. Validity

The Hierarchical Service Quality Scale underwent rigorous empirical validation by Brady and Cronin (2001) utilizing a multi-industry cross-sectional design. The psychometric evaluation confirmed robust construct, convergent, discriminant, nomological, and predictive validity.

Construct and Convergent Validity

Construct validity was evaluated using maximum likelihood structural equation modeling. In the validation dataset comprising 1,133 consumers across fast food (n = 290), photo developing (n = 277), amusement parks (n = 296), and dry cleaning (n = 270), each indicator loaded significantly on its intended first-order subdimension factor. Standardized factor loadings across all subdimensions were exceptionally high, typically falling between .70 and .95 (p < .001). The Average Variance Extracted (AVE) for each subdimension comfortably exceeded the conservative .50 benchmark established by Fornell and Larcker (1981), demonstrating that the variance explained by the underlying construct was greater than the variance attributable to measurement error.

Discriminant Validity

Discriminant validity was verified across all nine first-order subdimensions and the three second-order primary dimensions. To confirm that the dimensions represented distinct cognitive constructs rather than redundant markers, the square root of the AVE for each latent construct was evaluated against the inter-construct correlation coefficients. In every instance, the variance shared between any two constructs was systematically lower than the AVE of the individual constructs. Furthermore, competitive nested-model tests (wherein inter-factor correlations were fixed to unity) produced a statistically significant degradation in chi-square fit, confirming that each dimension provides unique explanatory variance.

Nomological and Predictive Validity

The nomological validity of the HSQS was established by situating overall service quality within an extended structural model incorporating customer satisfaction, service value, and future behavioral intentions (repurchase likelihood and positive word-of-mouth). Consistent with cognitive appraisal theories:

• Overall Service Quality exercised a strong, statistically significant direct effect on Customer Satisfaction (γ values ranging from .54 to .78 across industry samples, p < .001).
• Service Quality accounted for substantial portions of variance in customer behavioral intentions, operating both directly and indirectly via customer satisfaction.
• The three primary dimensions exhibited strong, statistically significant loadings onto the third-order Service Quality construct: Interaction Quality (γ ≈ .85 to .92), Physical Environment Quality (γ ≈ .62 to .81), and Outcome Quality (γ ≈ .78 to .89), confirming that all three primary facets are essential pillars of perceived quality.

8. Reliability

The scale demonstrates remarkable internal consistency across varied operational environments. Brady and Cronin (2001) evaluated scale reliability using both traditional Cronbach’s alpha (α) and Bagozzi and Yi’s Composite Reliability (CR) metrics.

Reliability Estimates Across Dimensions

• Interaction Quality Subdimensions:
  • Attitude: α values ranged from .86 to .92; CR > .88 across samples.
  • Behavior: α values ranged from .84 to .91; CR > .86 across samples.
  • Expertise: α values ranged from .87 to .93; CR > .89 across samples.
• Physical Environment Quality Subdimensions:
  • Ambient Conditions: α values ranged from .81 to .89; CR > .83 across samples.
  • Facility Design: α values ranged from .85 to .92; CR > .87 across samples.
  • Social Factors: α values ranged from .79 to .88; CR > .82 across samples.
• Outcome Quality Subdimensions:
  • Waiting Time: α values ranged from .83 to .91; CR > .85 across samples.
  • Tangibles: α values ranged from .80 to .88; CR > .82 across samples.
  • Valence: α values ranged from .82 to .90; CR > .84 across samples.

Overall, every individual subdimension demonstrated composite reliability values comfortably exceeding the standard .70 threshold recommended for basic research and the .80 benchmark desired for applied diagnostic measurement. Test-retest reliability across independent cross-validation subsamples verified the stability of the factor structure over repeated administrations.

9. Factor Analysis

To confirm the hierarchical structural architecture of perceived service quality, Brady and Cronin conducted extensive Confirmatory Factor Analyses (CFA) using maximum likelihood estimation in LISREL, testing multiple competing factor models against the proposed third-order model.

Model Comparisons and Goodness-of-Fit

The authors systematically compared the hypothesized third-order model against three distinct alternative structural specifications:

1. Model 1 (Single-Factor Model): All items loading directly onto a solitary latent service quality factor. This model exhibited exceptionally poor fit (χ² / df > 8.5; Comparative Fit Index [CFI] < .70; Root Mean Square Error of Approximation [RMSEA] > .14), disproving unidimensionality.
2. Model 2 (Nine Correlated First-Order Factors): Nine discrete, correlated factors representing the subdimensions without higher-order integration. While fit indices improved, the high inter-factor correlations (some exceeding .85) indicated structural redundancy and lack of parsimony.
3. Model 3 (Second-Order Model): Three correlated primary dimensions (Interaction, Environment, Outcome) directly explaining the observed items. This model suffered from substantial unexplained covariation among the underlying specific subdimensions.
4. Model 4 (Proposed Third-Order Factor Model): Nine first-order factors loading onto three second-order primary factors, which in turn load onto a single third-order overarching service quality construct.

The third-order hierarchical model demonstrated superior fit statistics across all four tested service sectors, satisfying established psychometric standards:

• Fit Indices (Combined Samples): χ² / df ratios consistently fell below 2.5 (ranging between 1.84 and 2.32 across industries).
• Incremental Fit: Comparative Fit Index (CFI) values ranged from .94 to .97; Non-Normed Fit Index (NNFI / TLI) values ranged from .93 to .96.
• Residual Fit: Root Mean Square Error of Approximation (RMSEA) ranged from .042 to .058, well below the conservative .06 threshold, with standardized root mean square residuals (SRMR) under .05.

First-order factor loadings on observed variables consistently exceeded .75. Second-order loadings of the nine subdimensions onto the three primary factors ranged between .70 and .94. Third-order loadings of the three primary dimensions onto the general service quality construct were: Interaction Quality (λ ≈ .88), Physical Environment Quality (λ ≈ .71), and Outcome Quality (λ ≈ .84).

10. Instrument / Measurement Tool

The standard research version of the Hierarchical Service Quality Scale is structured as follows:

• Test Type: Multi-item, multidimensional self-report psychometric rating inventory designed for consumer survey administration.
• Target Population: Consumers, clients, or service patrons across retail, hospitality, professional, entertainment, and commercial service domains.
• Item Count:
  • Subdimension Items: 27 primary reflective items (3 items × 9 subdimensions).
  • Primary Dimension Global Items: 3 items evaluating overall Interaction, Physical Environment, and Outcome quality.
  • Third-Order Global Quality Items: 3 overall service quality summary evaluation items.
  • Total Operational Scale: 33 items in full structural specification.
• Response Format: 7-point Likert scale (ranging from 1 = Strongly Disagree to 7 = Strongly Agree) or 7-point semantic differential scales (e.g., 1 = Poor to 7 = Excellent).
• Administration Mode: Paper-and-pencil questionnaire, online computerized survey, or mobile digital survey. Completion time is approximately 8 to 12 minutes.
• Scoring and Computational Procedures:
  • Subdimension scores are calculated as the unweighted arithmetic mean of the three corresponding indicator items (scores range from 1 to 7).
  • Primary dimension indices can be computed either by averaging the three corresponding subdimension scores or by calculating factor-score-weighted composites via SEM latent variable weights.
  • The Global Service Quality Index is derived by averaging all 27 subdimension items or by structural weighting of the three primary dimensions. Higher scores reflect superior perceived service quality.

11. Permissions & Fee and Test Year

The Hierarchical Service Quality Scale was formally published in 2001 in the peer-reviewed Journal of Marketing. The academic theoretical framework, structural equations, and methodological operationalization are documented in the original publication.

The copyright for the published article belongs to the American Marketing Association (AMA). For non-commercial academic research, pedagogical purposes, and scholarly thesis investigations, the conceptual model and scale items may typically be adapted and utilized under standard fair-use scholarly conventions, provided that full academic citation and attribution are given to Brady and Cronin (2001). Commercial deployment, corporate diagnostic licensing, inclusion in fee-for-service software platforms, or large-scale proprietary distribution requires formal copyright permission through the American Marketing Association or the Copyright Clearance Center (CCC).

12. References

• Bagozzi, R. P., & Yi, Y. (1988). On the evaluation of structural equation models. Journal of the Academy of Marketing Science, 16(1), 74–94. https://doi.org/10.1007/BF02723327
• Bitner, M. J. (1992). Servicescapes: The impact of physical surroundings on customers and employees. Journal of Marketing, 56(2), 57–71. https://doi.org/10.1177/002224299205600205
• Brady, M. K., & Cronin, J. J. (2001). Some new thoughts on conceptualizing perceived service quality: A hierarchical approach. Journal of Marketing, 65(3), 34–49. https://doi.org/10.1509/jmkg.65.3.34.18334
• Cronin, J. J., & Taylor, S. A. (1992). Measuring service quality: A reexamination and extension. Journal of Marketing, 56(3), 55–68. https://doi.org/10.1177/002224299205600304
• Dabholkar, P. A., Thorpe, D. I., & Rentz, J. O. (1996). A measure of service quality for retail stores: Scale development and validation. Journal of the Academy of Marketing Science, 24(1), 3–16. https://doi.org/10.1007/BF02893933
• 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
• Grönroos, C. (1982). An applied service marketing theory. European Journal of Marketing, 16(7), 30–41. https://doi.org/10.1108/EUM0000000004859
• Grönroos, C. (1984). A service quality model and its marketing implications. European Journal of Marketing, 18(4), 36–44. https://doi.org/10.1108/EUM0000000004784
• Parasuraman, A., Zeithaml, V. A., & Berry, L. L. (1985). A conceptual model of service quality and its implications for future research. Journal of Marketing, 49(4), 41–50. https://doi.org/10.1177/002224298504900403
• Parasuraman, A., Zeithaml, V. A., & Berry, L. L. (1988). SERVQUAL: A multiple-item scale for measuring consumer perceptions of service quality. Journal of Retailing, 64(1), 12–40.
• Rust, R. T., & Oliver, R. L. (1994). Service Quality: New Directions in Theory and Practice. SAGE Publications.

13. Items of the Scale