Anger at Service Provider (AASP)

1. Abstract

The Anger at Service Provider (AASP) scale is a concise, psychometrically validated three-item self-report instrument developed by Carolyn Bonifield and Catherine A. Cole (2008). It is designed to quantify the intensity of customer emotional anger, annoyance, and irritation directed toward a firm following a service failure or recovery encounter. Operating as a unidimensional construct, the instrument captures high-arousal, other-attributed negative consumption emotions grounded in the Cognitive Appraisal Theory of emotions and Social Comparison Theory. Each item is rated on a 7-point Likert response scale ranging from 1 (Strongly disagree) to 7 (Strongly agree), with an overall score calculated by averaging the individual item ratings. Across empirical testing in service marketing and consumer psychology, the AASP demonstrates robust psychometric integrity, consistently yielding high internal consistency reliability coefficients (Cronbach’s alpha typically exceeding .90) and strong factor loadings in exploratory and confirmatory factor analytic evaluations. The measure exhibits robust convergent validity with constructs such as perceived injustice, dissatisfaction, and retaliatory intentions, as well as distinct discriminant validity from low-arousal negative affective states (e.g., sadness, regret) and general service dissatisfaction. This article provides an exhaustive examination of the AASP scale, detailing its theoretical foundation, psychometric parameters, structural validity, administrative procedures, and practical utility in consumer behavior and service recovery research.

2. Keywords

Anger at Service Provider, AASP, consumer anger, service failure, service recovery, cognitive appraisal theory, social comparison theory, consumer emotions, psychometrics, customer dissatisfaction

3. Authors

The Anger at Service Provider measure was formulated and validated by scholars in marketing and consumer psychology:

• Carolyn Bonifield, Ph.D. — Associate Professor of Marketing, Grossman School of Business, University of Vermont. Her research focuses on consumer behavior, service marketing, affect and decision-making, and customer responses to brand transgressions.
• Catherine A. Cole, Ph.D. — Professor Emerita of Marketing, Tippie College of Business, University of Iowa. Her scholarship investigates marketing communication, consumer information processing, aging and decision-making, and emotional dynamics in service interactions.

The scale was formally introduced in their peer-reviewed investigation examining social comparison mechanisms during service recovery encounters, published in the Journal of the Academy of Marketing Science (Bonifield & Cole, 2008).

4. Purpose

In modern service environments, service failures—ranging from minor delivery delays to severe operational lapses—are inevitable. When a service transgression occurs, consumers rarely react with passive indifference; instead, they experience complex affective and cognitive reactions that dictate subsequent behavior. The primary purpose of the Anger at Service Provider (AASP) scale is to isolate and quantify the specific, high-arousal negative emotion of customer anger directed squarely at an organizational service entity. Rather than assessing generalized dissatisfaction or negative affect globally, the instrument isolates the acute emotional hostility that emerges when customers perceive that an unfavorable service outcome was preventable, unfair, or caused by negligence on the part of the service organization.

From an applied and managerial standpoint, capturing specific emotions like anger is vastly more diagnostic than tracking overall customer satisfaction (CSAT). While low satisfaction indicates that expectations were unmet, anger is uniquely characterized by an impulse to retaliate, engage in negative word-of-mouth (NWOM), lodge formal complaints, or aggressively switch providers. By employing the AASP, organizations and service designers can gauge the emotional volatility of customer segments following specific failure typologies (e.g., core service breakdowns versus interactional slights) and evaluate the remedial efficacy of different recovery strategies, such as apologies, compensation, or structural explanations.

In academic research, the AASP serves as an essential measurement tool within structural equation models examining consumer psychology, behavioral economics, and relationship marketing. It enables researchers to evaluate mediating pathways connecting cognitive evaluations—such as perceived injustice, locus of causality, and unfavorable social comparisons—to severe behavioral outcomes, including vengeful complaining, third-party disputes, and customer boycott mobilization. The brevity of the instrument makes it particularly well-suited for laboratory experiments, scenario-based vignettes, longitudinal customer journey surveys, and ecological momentary assessments where participant fatigue must be minimized without sacrificing measurement reliability.

5. Psychological Construct

The psychological construct measured by the AASP is customer anger directed toward a service provider. In psychometric and emotional literature, anger is conceptualized as an acute, high-arousal, goal-incongruent emotion triggered by the perception of an intentional, controllable, or unjustified transgression committed by an external agent (Lazarus, 1991). Unlike inward-focused negative emotions such as guilt, embarrassment, or self-blame, anger is inherently social and other-directed, mobilizing cognitive and physiological resources toward confronting the offending party.

The AASP operationalizes this construct across three closely related affective indicators, capturing a spectrum of hostile arousal:

• Anger: The core affective state characterized by intense hostility, moral outrage, and a heightened physiological drive to retaliate or demand restitution. Within the consumption realm, customer anger implies that the consumer views the service provider as blameworthy, competent enough to have avoided the failure, yet negligent or dismissive of the customer's welfare.
• Annoyance: A lower-intensity manifestation of irritation or frustration that signals disruption to goal attainment. Annoyance represents the friction experienced when service protocols fail to operate smoothly, serving as an emotional precursor or companion to fully mobilized anger.
• Irritation: An agitated affective state characterized by psychological discomfort and lowered patience. In consumption encounters, irritation reflects cognitive overload and friction caused by bureaucratic obstacles, unhelpful frontline employees, or prolonged resolution times.

Psychometrically, these three facets coalesce into a single, tightly bound unidimensional construct. Consumption emotions research emphasizes that while annoyance and irritation can exist at moderate thresholds of dissatisfaction, their convergence with explicit anger signifies a critical threshold of perceived violation. The construct measured by the AASP must be distinguished from related emotional and cognitive states:

• Anger vs. Dissatisfaction: Dissatisfaction is primarily a cognitive-evaluative summary judgment reflecting the gap between prior expectations and perceived performance (Oliver, 1980). In contrast, anger is an affective state with distinct somatic arousal and action tendencies. A customer may be dissatisfied with a mediocre hotel room without feeling angry, provided the failure was not attributed to deliberate indifference.
• Anger vs. Disappointment / Sadness: Disappointment stems from situational attributions (e.g., bad weather canceling a flight) and typically induces resignation or withdrawal. Anger requires an external agency attribution (e.g., flight canceled due to airline understaffing) and stimulates approach-oriented confrontational behavior.
• Anger vs. Regret: Regret involves an internal attribution, wherein the consumer blames their own decision-making process for an undesirable outcome (e.g., choosing the wrong vendor). The AASP explicitly assesses other-directed affective hostility aimed outward at the service brand.

6. Theoretical Framework

The Anger at Service Provider scale is grounded in two primary theoretical paradigms: Cognitive Appraisal Theory and Social Comparison Theory.

Cognitive Appraisal Theory

Pioneered by theorists such as Richard Lazarus (1991) and Roseman (1991), cognitive appraisal theory posits that emotions do not emerge directly from objective events, but rather from an individual's subjective cognitive evaluation (appraisal) of those events along specific dimensions. These dimensions include:

1. Goal Incongruence: The event obstructs the individual's goals or needs (e.g., missing an important business meeting due to a delayed flight).
2. Ego Involvement: The extent to which the failure threatens the individual's identity, self-esteem, financial resources, or basic rights.
3. Agency / Locus of Causality: The determination of who caused the event. When a failure is attributed to circumstances beyond human control (situational locus), sadness or distress results. When attributed to an external agent (external locus)—specifically the service organization—anger is elicited.
4. Controllability and Blame: If the external agent is perceived to have possessed the capacity to prevent the failure but failed to exercise due care, the appraisal shifts from mere negative disconfirmation to blameworthiness, eliciting acute customer anger.

Bonifield and Cole (2008) utilized this framework to show that when customers evaluate a service breakdown, their appraisal of fairness and culpability directly feeds the affective activation captured by the AASP.

Social Comparison Theory

The foundational study introducing the AASP was situated within Leon Festinger's (1954) Social Comparison Theory. Bonifield and Cole (2008) examined how consumers evaluate service recovery outcomes when they witness or learn about the service recovery remedies provided to other consumers in identical situations. In service failure contexts, social comparison processes strongly influence emotional appraisals:

• Downward Social Comparison: Observing another customer receive an inferior remedy (or suffer a worse outcome) can mitigate personal distress, leading to feelings of relative advantage.
• Upward Social Comparison: Observing another consumer receive superior service recovery compensation or preferential treatment by service staff induces severe perceptions of unfairness and relative deprivation. This violation of equity activates substantial other-directed hostility, which the AASP directly captures.

Furthermore, the AASP interfaces with Justice Theory (taxonomized into distributive, procedural, and interactional justice; Tax et al., 1998). Violations of interactional justice—such as rude, dismissive, or deceptive behavior by frontline employees during the complaint handling process—are particularly potent drivers of high-arousal anger, elevating AASP scores and catalyzing punitive customer retaliation.

7. Validity

The psychometric validity of the Anger at Service Provider scale has been established across multiple experimental investigations, cross-sectional surveys, and replicated field studies in consumer psychology and service marketing.

Construct and Convergent Validity

Construct validity denotes the degree to which an operationalization accurately reflects its target theoretical construct. In Bonifield and Cole's (2008) initial validation studies across varied service scenarios (e.g., airline disruptions, retail transaction errors), the three items loaded onto a single common factor with uniform standardized factor loadings consistently exceeding .85. Average Variance Extracted (AVE) values routinely surpass the recommended .50 threshold established by Fornell and Larcker (1981), often reaching values above .75, indicating that the majority of variance captured by the indicators is shared common construct variance rather than measurement error.

Convergent validity is further evidenced by statistically robust correlations between the AASP and conceptually allied constructs. Scores on the AASP demonstrate strong positive correlations with measures of perceived injustice ($r \approx .60$ to $.75$), overall transaction-specific dissatisfaction ($r \approx .65$ to $.80$), and perceived brand culpability ($r \approx .55$ to $.70$).

Discriminant Validity

To confirm that the AASP does not merely duplicate general negative affect, researchers have conducted rigorous discriminant validity assessments using the Fornell-Larcker criterion and Confirmatory Factor Analysis (CFA) nested model comparisons. Bonifield and Cole (2008) evaluated whether anger directed at the provider could be separated from inward-focused or low-arousal negative emotions:

• The square root of the AVE for the AASP consistently exceeds the bivariate correlations between the AASP and adjacent affective states, such as customer sadness, regret, and disappointment.
• Chi-square difference tests between a baseline unconstrained multi-factor CFA model (separating anger, guilt, and disappointment) and constrained models (forcing items onto a unified general negative affect factor) reveal significant deteriorations in fit for the single-factor models, confirming that customer anger constitutes an empirically distinct emotional phenomenon.

Predictive and Nomological Validity

Nomological validity is verified by the AASP's demonstrated ability to predict theoretical downstream consequences within broader consumer behavior models. Across studies, high AASP scores significantly predict:

• Negative Word-of-Mouth (NWOM): High customer anger is one of the strongest statistical predictors of aggressive NWOM and online public review venting ($r \approx .50$ to $.68$).
• Retaliatory and Vengeful Behaviors: Unlike passive dissatisfaction, scores on the AASP uniquely predict retaliatory intentions, vindictive complaining, and intentions to damage the firm's reputation.
• Repatronage and Switching: Elevated AASP scores negatively predict repurchase intentions and long-term brand loyalty while positively predicting actual service brand defection.

8. Reliability

The Anger at Service Provider scale demonstrates exceptional internal consistency reliability across diverse sampling frames, languages, and service failure typologies.

Internal Consistency

In the original validation studies conducted by Bonifield and Cole (2008), the scale's Cronbach’s alpha ($lpha$) coefficients ranged between .91 and .95 across experimental conditions. Subsequent research replicating and adapting the scale in banking, hospitality, telecommunications, and e-commerce contexts has routinely reported alpha values well above the conventional academic benchmark of .70, typically clustering between .89 and .96. Furthermore, composite reliability ($CR$) indices computed in structural equation modeling environments consistently exceed .90, demonstrating high item redundancy in capturing the core construct while maintaining high statistical power.

Item-Total Statistics and Inter-Item Correlations

The three items comprising the AASP exhibit uniformly high corrected item-total correlations, typically ranging from .78 to .92. Inter-item correlations among the three indicators (angry, annoyed, irritated) routinely exceed .70, confirming that respondents perceive the terms as closely aligned markers of their affective hostility toward the firm. Removal of any single item fails to improve the overall scale reliability coefficient, affirming that the three-item configuration provides an optimal balance between parsimony and measurement breadth.

Test-Retest Stability and State Sensitivity

Because the AASP is predominantly employed as an emotional state measure—capturing acute reactions to a discrete service incident—traditional test-retest reliability across long temporal intervals is neither expected nor theoretically desirable. Emotional anger naturally decays over time due to psychological habituation, cognitive rationalization, or compensatory organizational interventions. However, short-term stability assessments (e.g., repeated evaluations within controlled experimental protocols prior to recovery interventions) show high intra-subject consistency ($r > .85$). When applied longitudinally to track emotional recovery, the instrument displays sensitive responsiveness to service recovery interventions, registering sharp declines following effective managerial resolution.

9. Factor Analysis

The structural dimensionality of the AASP has been evaluated through both Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA) across numerous empirical investigations.

Exploratory Factor Analysis (EFA)

In exploratory factor analyses utilizing principal axis factoring or maximum likelihood extraction with oblique or orthogonal rotations, the three items consistently load onto a single dominant factor. The initial unrotated eigenvalue for the first factor typically accounts for 80% to 88% of the total variance, far exceeding Kaiser's criterion of 1.0. Scree plots display a distinct single-factor elbow, with no secondary factors emerging above an eigenvalue of 0.40. Standardized factor loadings across the three items are exceptionally high and homogeneous:

• “I would feel angry toward [service provider]”: Loadings typically range from .86 to .94.
• “I would feel annoyed with [service provider]”: Loadings typically range from .84 to .91.
• “I would feel irritated with [service provider]”: Loadings typically range from .85 to .93.

Confirmatory Factor Analysis (CFA)

Because a three-item measurement model with no cross-loadings has zero degrees of freedom ($df = 0$), a standalone one-factor CFA model of the AASP is mathematically just-identified (saturated), yielding perfect structural fit ($CFI = 1.000$, $TLI = 1.000$, $RMSEA = .000$). Consequently, structural validity is formally tested by embedding the AASP within multi-construct measurement models alongside related constructs such as distributive justice, procedural justice, overall satisfaction, and behavioral intentions.

In such multi-factor CFA specifications, the measurement models demonstrate outstanding fit to empirical data. Standard model fit indices routinely satisfy stringent psychometric thresholds:

• Comparative Fit Index (CFI): $ge .96$ (often $ge .98$)
• Tucker-Lewis Index (TLI): $ge .95$ (often $ge .97$)
• Root Mean Square Error of Approximation (RMSEA): $le .05$ (with 90% confidence intervals bounded below .08)
• Standardized Root Mean Square Residual (SRMR): $le .04$

Standardized path loadings ($lambda$) for each of the three items remain robust ($> .80$, $p < .001$) within complex structural models, confirming that the scale retains its psychometric properties when contextualized alongside other consumption emotion and cognitive appraisal batteries.

10. Instrument / Measurement Tool

The formal specifications of the Anger at Service Provider (AASP) instrument are summarized below:

• Instrument Name: Anger at Service Provider (AASP)
• Original Authors: Carolyn Bonifield and Catherine A. Cole (2008)
• Construct Measured: Other-directed, high-arousal customer anger and emotional hostility toward a service firm following an operational or interpersonal service failure.
• Test Type: Self-report questionnaire / psychometric rating scale.
• Administration Format: Paper-and-pencil, computer-assisted personal interviewing (CAPI), or web-based survey questionnaire.
• Number of Items: 3 items.
• Response Scale: 7-point Likert scale:
  • 1 = Strongly disagree
  • 2 = Disagree
  • 3 = Somewhat disagree
  • 4 = Neither agree nor disagree
  • 5 = Somewhat agree
  • 6 = Agree
  • 7 = Strongly agree
• Scoring and Indexing Procedures:
  • There are no reverse-coded items; all three items are positively worded reflecting anger arousal.
  • The overall AASP score is derived by computing the mathematical mean across all three items: $\text{AASP Score} = \frac{\text{Item 1} + \text{Item 2} + \text{Item 3}}{3}$.
  • Composite scores range from 1.0 to 7.0, where higher numerical values indicate greater levels of anger directed toward the service organization.
  • Alternatively, for structural equation modeling (SEM) applications, the three items can be modeled directly as continuous reflective indicators of a single latent construct.
• Administration Time: Less than 1 minute (typically 30 to 45 seconds).
• Target Population: Adult consumers, retail shoppers, and service clients across varied commercial sectors.

11. Permissions & Fee and Test Year

The Anger at Service Provider scale was published in 2008 in the Journal of the Academy of Marketing Science (Volume 36, Issue 4). As an academic research instrument published in a peer-reviewed scholarly journal, the scale items are widely accessible for non-commercial academic, scientific, and educational research purposes under standard academic fair-use guidelines. Researchers utilizing the instrument in scholarly studies, theses, dissertations, or classroom instruction are expected to provide full bibliographic attribution to Bonifield and Cole (2008).

No formal licensing fees are required for standard academic investigation. However, commercial enterprises, market research corporations, or consulting agencies seeking to integrate the scale into proprietary customer feedback systems, commercial software platforms, or fee-for-service evaluation audits should consult the copyright policies of the journal's publisher (Springer Nature / Academy of Marketing Science) or contact the corresponding author to determine if commercial permissions or licensing agreements are necessary.

12. References

• Bonifield, C., & Cole, C. A. (2008). Better him than me: Social comparison theory and service recovery. Journal of the Academy of Marketing Science, 36(4), 565–577. https://doi.org/10.1007/s11747-008-0109-z
• Festinger, L. (1954). A theory of social comparison processes. Human Relations, 7(2), 117–140. https://doi.org/10.1177/001872675400700202
• 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
• Lazarus, R. S. (1991). Emotion and Adaptation. Oxford University Press.
• 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
• Roseman, I. J. (1991). Appraisal determinants of discrete emotions. Cognition & Emotion, 5(3), 161–200. https://doi.org/10.1080/02699939108411034
• Tax, S. S., Brown, S. W., & Chandrashekaran, M. (1998). Customer evaluations of service complaint experiences: Implications for relationship marketing. Journal of Marketing, 62(2), 60–76. https://doi.org/10.1177/002224299806200205

13. Items of the Scale