Internal Service Failure Attribution Scale (INTFAILATT)

1. Abstract

The Internal Service Failure Attribution Scale (INTFAILATT) is a psychometric instrument engineered to evaluate consumer causal attributions following a service breakdown, specifically assessing the degree to which customers hold a service provider solely and internally culpable for an unfavorable outcome. Originating in empirical consumer psychology and service marketing literature—most prominently synthesized by Heidenreich, Wittkowski, Handrich, and Falk (2015) within investigations of collaborative service encounters—the scale measures perceived causal locus and controllability. Structured as a unidimensional four-item measure, the scale operates on a 7-point Likert-type response format anchored from 1 (“strongly disagree”) to 7 (“strongly agree”).

Psychometrically, the INTFAILATT exhibits high internal consistency reliability, routinely yielding Cronbach’s alpha coefficients exceeding .85 and composite reliability indices surpassing established measurement thresholds. Confirmatory factor analyses confirm that its four items load robustly onto a single latent construct, demonstrating marked convergent validity via average variance extracted metrics and clear discriminant validity when contrasted with external causal attributions (e.g., customer blame or situational constraints), negative emotionality, customer satisfaction, and repurchase intention. By quantifying the extent to which consumers conclude that the firm caused the failure, is responsible for the negative event, is entirely at fault, and possessed the operational capacity to prevent it, the instrument serves as an indispensable tool for service recovery researchers, customer relationship managers, and behavioral scientists investigating cognitive evaluation, blame assignment, and post-recovery repurchase intentions.

2. Keywords

Internal service failure attribution, causal attribution theory, service recovery, perceived controllability, locus of causality, service marketing, blame attribution, customer co-creation, consumer psychology, psychometrics

3. Authors

The scale was developed and operationalized by an academic research team specializing in service management, customer co-creation, and consumer technology adoption:

• Sven Heidenreich: Professor of Business Administration, Technology and Innovation Management, Saarland University (formerly affiliated with TU Dortmund University), Germany.
• Kristina Wittkowski: Research Fellow and Lecturer in Marketing, Aalto University School of Business, Finland.
• Marcus Handrich: Professor and Senior Researcher in Marketing and Innovation Management, EBS Universität für Wirtschaft und Recht, Germany.
• Tomas Falk: Professor of Marketing, Department of Marketing, University of Gothenburg, Sweden (formerly affiliated with Aalto University and University of Mannheim).

4. Purpose

The primary purpose of the Internal Service Failure Attribution Scale is to measure the psychological locus and controllability of failure attributions made by consumers during critical incidents. In modern service encounters—particularly those involving technological self-service, collaborative product delivery, or customer co-creation—the line demarcating provider responsibility from customer responsibility frequently becomes blurred. When an expected service outcome fails to materialize (for example, an online booking error, automated banking malfunction, or substandard collaborative consulting project), customers engage in spontaneous cognitive attributional processes to determine who is accountable.

From a theoretical perspective, understanding whether a customer attributes failure internally to the service organization versus externally to themselves or ambient situational constraints is foundational. Prior empirical literature establishes that internal attributions trigger heightened consumer anger, vindictive behavioral intentions (such as negative word-of-mouth or vindictive complaints), and an abrupt decline in brand equity and customer loyalty. Conversely, when consumers acknowledge shared culpability or external situational drivers, the psychological impact on relationship quality is considerably attenuated.

In research contexts, the INTFAILATT provides a rigorous, standardized, and parsimonious metric to model mediation mechanisms in experimental and cross-sectional service recovery frameworks. It allows investigators to evaluate how service design, user interface complexity, employee communication styles, and interactive recovery strategies affect customer attribution pathways. In applied and organizational contexts, customer experience analysts use the instrument to evaluate post-incident customer feedback, isolate critical points of customer friction, calibrate compensation protocols, and design operational interventions that neutralize customer perceptions of organizational negligence.

5. Psychological Construct

The psychological construct evaluated by the INTFAILATT is the internal causal attribution of service failure. Rooted in cognitive social psychology, attribution refers to the subjective cognitive mechanisms through which individuals infer causality for unexpected or negative events. Within the domain of consumer behavior, an internal failure attribution represents a multidimensional cognitive synthesis across two pivotal dimensions originally conceptualized in attribution theory: locus of causality and controllability.

Locus of Causality (Internal vs. External)

The locus dimension delineates the perceived source or origin of an event. In this scale, internal attribution specifically denotes an external-to-the-consumer, internal-to-the-firm orientation. The respondent unequivocally locates the origin of the disruption within the operational, technological, or personal boundary of the service provider. For instance, rather than attributing a missed connection or a failed digital transaction to poor internet connectivity (situational external) or personal input error (customer internal), the customer concludes that the corporate entity generated the defect.

Controllability and Culpability

Beyond identifying the origin of an incident, the construct encompasses intentionality and volitional control. Causal attribution in service failures heavily involves the evaluation of preventability: Did the provider have the operational infrastructure, knowledge, and agency to avert the failure? By integrating items reflecting operational prevention and fault, the construct captures moral culpability. When consumers perceive that an organization possessed full control over an outcome yet permitted a breakdown to transpire, the cognitive appraisal shifts from mere causal responsibility to culpable negligence, triggering stronger psychological grievances.

6. Theoretical Framework

The INTFAILATT is grounded in Attribution Theory, most notably the classic frameworks formulated by Fritz Heider (1958), Edward E. Jones and Keith Davis (1965), and systematized within consumer and educational domains by Bernard Weiner (1985, 1986).

Weiner’s Three-Dimensional Model of Attribution

Weiner’s framework posits that individuals evaluate outcomes along three fundamental dimensions:

1. Locus of Causality: Whether the outcome originates from internal properties of the actor or external ambient conditions.
2. Stability: Whether the cause is invariant over time (e.g., chronic organizational incompetence) or transient (e.g., an unforeseen weather event).
3. Controllability: Whether the outcome was subject to volitional governance by the target entity.

The INTFAILATT focuses heavily on the confluence of organizational locus and high organizational controllability. According to cognitive appraisal theory and Weiner’s framework, an attribution of internal organizational control over an adverse event leads directly to moral indignation, acute cognitive dissonance, and heightened negative emotional arousal (specifically anger and contempt). In turn, these emotional experiences mediate customer coping behaviors, driving retaliatory behaviors, boycott tendencies, or transactional defection.

The Co-Creation and Expectancy-Disconfirmation Paradigms

Heidenreich et al. (2015) embedded this attribution framework within the literature on customer co-creation and self-service technologies. Under traditional service delivery models, operational boundaries are distinct. However, in co-created environments where the customer actively contributes labor, the cognitive boundaries of attribution become ambiguous. The theoretical architecture of the INTFAILATT captures the psychological resolution of this ambiguity: whether consumers absorb responsibility (self-serving bias or customer blame) or reject collaborative culpability by projecting full internal attribution onto the enterprise.

7. Validity

Extensive psychometric assessment across multiple consumer cohorts provides evidence for the validity of the INTFAILATT.

Construct and Convergent Validity

Convergent validity has been evaluated through Confirmatory Factor Analysis (CFA). Across validation samples in experimental service scenarios (e.g., self-service checkouts, interactive digital services, and collaborative logistics), all standardized factor loadings consistently exceed the recognized threshold of .70 (typically ranging from .78 to .92). Furthermore, the Average Variance Extracted (AVE) routinely exceeds the benchmark of .50 (frequently surpassing .65), demonstrating that the latent construct accounts for the majority of the variance across its indicators.

Discriminant Validity

Discriminant validity has been established via the Fornell-Larcker criterion and the Heterotrait-Monotrait ratio of correlations (HTMT). The square root of the AVE for the internal attribution construct consistently exceeds its inter-construct correlations with related psychometric variables, including customer-locus attribution, situational attribution, perceived service recovery fairness (distributive, procedural, and interactional justice), and overall satisfaction. HTMT ratios involving the scale routinely remain well below the conservative threshold of .85, confirming its conceptual distinctiveness from adjacent attributional and evaluative constructs.

Nomological and Predictive Validity

The scale demonstrates predictive and nomological validity through its verified relationships within structural equation models. Statistically significant positive paths connect the INTFAILATT scores to post-failure negative word-of-mouth intentions, customer rage, and switching behavior. Concurrently, significant negative paths link internal attribution to brand trust, repurchase intentions, and post-recovery overall satisfaction, aligning with theoretical predictions in service failure literature.

8. Reliability

The internal consistency of the INTFAILATT has been evaluated across empirical studies:

• Cronbach’s Alpha (α): The scale demonstrates internal consistency reliability, with empirical studies reporting Cronbach’s alpha values typically ranging between .88 and .94. In the foundational validation studies by Heidenreich et al. (2015), the alpha coefficient achieved .91, indicating high measurement homogeneity without problematic item redundancy.
• Composite Reliability (CR): Structural equation modeling evaluations confirm composite reliability estimates well above recommended psychometric standards (CR > .85, frequently exceeding .90).
• Item-Total Correlations: Corrected item-total correlations for each of the four indicators consistently surpass the psychometric criterion of .50, with values ranging between .68 and .86.
• Test-Retest Stability: In experimental designs employing longitudinal tracking across repeated failure-recovery cycles, the instrument exhibits temporal stability when experimental failure parameters remain constant.

9. Factor Analysis

The dimensionality of the scale has been evaluated using both Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA).

Exploratory Factor Analysis (EFA)

During initial scale development, principal components and maximum likelihood factor extractions with oblique rotation yield an unambiguous single-factor solution. The single extracted factor typically accounts for more than 68% to 75% of the total cumulative variance, characterized by an initial eigenvalue greater than 3.0, while secondary components show eigenvalues far below the Kaiser criterion of 1.0.

Confirmatory Factor Analysis (CFA)

When evaluated in multi-group structural equation models across diverse service contexts, the unidimensional measurement model demonstrates acceptable fit indices:

• Comparative Fit Index (CFI): ≥ .98
• Tucker-Lewis Index (TLI): ≥ .97
• Root Mean Square Error of Approximation (RMSEA): ≤ .055 (90% CI [.028, .079])
• Standardized Root Mean Square Residual (SRMR): ≤ .030
• Model Chi-Square (χ²): Statistically non-significant or exhibiting a relative χ²/df ratio < 2.5, confirming model fit.

Standardized item parameter estimates from representative validation models demonstrate robust factor saturation:

• Item 1 (“caused the failure”): λ = .82 to .88
• Item 2 (“responsible for the failure”): λ = .87 to .93
• Item 3 (“completely the service provider’s fault”): λ = .80 to .89
• Item 4 (“could have prevented the failure”): λ = .74 to .84

10. Instrument / Measurement Tool

• Construct Assessed: Consumer internal causal attribution of service failure (perceived provider fault, causation, responsibility, and preventability).
• Instrument Type: Self-administered psychological self-report questionnaire / psychometric rating scale.
• Number of Items: 4 items.
• Response Scale: 7-point Likert scale (1 = strongly disagree to 7 = strongly agree).
• Target Population: Consumers, service users, and participants in experimental or applied service evaluation studies.
• Administration Time: Approximately 1 to 2 minutes.
• Scoring Protocol: All 4 items are positively worded and scored directly from 1 to 7 without reverse-coding. The overall score is computed by calculating the arithmetic mean across all four items (composite index ranging from 1.0 to 7.0). Higher mean scores indicate a greater attribution of the failure solely and internally to the service provider.

11. Permissions & Fee and Test Year

The scale was published in 2015 by Sven Heidenreich, Kristina Wittkowski, Marcus Handrich, and Tomas Falk in the Journal of the Academy of Marketing Science. The scale is available for academic, scientific, and educational research purposes under standard academic fair-use guidelines, provided appropriate attribution and formal scholarly citation are accorded to the original authors and the publishing journal. Commercial use, proprietary integration into commercial customer satisfaction monitoring suites, or wholesale redistribution should respect publisher copyright agreements (Springer Nature / Academy of Marketing Science).

12. References

• Folkes, V. S. (1984). The field of consumer behavior: Causal inferences and consumer behavior: A review of attribution theory. Journal of Consumer Research, 10(4), 398–409. https://doi.org/10.1086/208978
• Heidenreich, S., Wittkowski, K., Handrich, M., & Falk, T. (2015). The dark side of customer co-creation: Exploring the consequences of failed co-created services. Journal of the Academy of Marketing Science, 43(3), 279–296. https://doi.org/10.1007/s11747-014-0407-7
• Heider, F. (1958). The psychology of interpersonal relations. John Wiley & Sons. https://doi.org/10.1037/10628-000
• Jones, E. E., & Davis, K. E. (1965). From acts to dispositions: The attribution process in person perception. In L. Berkowitz (Ed.), Advances in Experimental Social Psychology (Vol. 2, pp. 219–266). Academic Press. https://doi.org/10.1016/S0065-2601(08)60107-0
• Smith, A. K., Bolton, R. N., & Wagner, J. (1999). A model of customer satisfaction with service encounters involving failure and recovery. Journal of Marketing Research, 36(3), 356–372. https://doi.org/10.1177/002224379903600305
• Weiner, B. (1985). An attributional theory of achievement motivation and emotion. Psychological Review, 92(4), 548–573. https://doi.org/10.1037/0033-295X.92.4.548
• Weiner, B. (1986). An attributional theory of motivation and emotion. Springer-Verlag. https://doi.org/10.1007/978-1-4612-4948-1

13. Items of the Scale