Product Failure Attribution (PFA)

Abstract

The Product Failure Attribution (PFA) scale—originally developed and operationalized within service and product failure contexts by Dong, Evans, and Zou (2008)—is a psychometric instrument designed to assess the degree to which a consumer assigns causal responsibility and blame to an external service provider or manufacturer following an unfavorable consumption outcome. Grounded in social cognitive frameworks and classical attribution theory, the instrument captures the psychological mechanisms through which individuals infer causality when experiencing service disruptions, product breakdowns, or unmet expectations. Comprising three unidimensional items measured on a 7-point Likert response format (ranging from 1 = Strongly disagree to 7 = Strongly agree), the scale isolates the external locus of causality, perceived culpability, and accountability assigned to the firm.

Empirical evaluations across consumer psychology, service-dominant logic, and marketing disciplines demonstrate robust psychometric properties. The scale consistently exhibits high internal consistency, with Cronbach’s alpha coefficients typically exceeding .88 and composite reliability estimates surpassing .90. Confirmatory factor analyses across experimental and field studies verify strict unidimensionality, with high factor loadings (generally exceeding .85) and optimal goodness-of-fit indices (e.g., Comparative Fit Index [CFI] > .98, Root Mean Square Error of Approximation [RMSEA] < .06). Convergent and discriminant validity are evidenced by expected directional relationships with constructs such as post-recovery customer satisfaction, negative emotional affect (e.g., anger, frustration), customer revenge intentions, and willingness to co-create service recovery solutions. This instrument serves as an empirical tool for researchers investigating consumer attributional processes, service recovery paradoxes, self-serving biases in co-production, and organizational accountability.

Keywords

Product Failure Attribution, Service Failure, Attribution Theory, Locus of Causality, Blame Attribution, Customer Participation, Co-Created Service Recovery, Consumer Psychology, Psychometrics, Service Marketing

Authors

The Product Failure Attribution scale was developed and validated by an academic research team specializing in marketing strategy, consumer behavior, and international service management:

• Beibei Dong, Ph.D. — Professor of Marketing, Department of Marketing, LeBow College of Business, Drexel University. Her research focuses on service marketing, customer participation in service co-creation, service recovery, and customer-firm interactions.
• Kenneth R. Evans, Ph.D. — President Emeritus and Professor of Marketing, Lamar University; formerly Dean and Fred E. Brown Chair in Business at the Trulaske College of Business, University of Missouri-Columbia. His scholarship centers on sales management, services marketing, and organizational boundary-spanning behavior.
• Shaoming Zou, Ph.D. — Emeritus Professor of Marketing, Robert J. Trulaske, Sr. College of Business, University of Missouri-Columbia. His research expertise encompasses international marketing strategy, global supply chain management, and quantitative modeling in marketing.

Purpose

The fundamental purpose of the Product Failure Attribution (PFA) scale is to provide a reliable, parsimonious, and theoretically grounded instrument to quantify consumer perceptions of firm culpability following a service failure or product malfunction. In both commercial and research settings, negative consumption events rarely elicit passive responses; rather, consumers act as intuitive psychologists who systematically search for the causes underlying unexpected disruptions. The PFA scale directly measures this subjective attributional conclusion—specifically, whether the consumer identifies the service provider as the primary causal agent and moral bearer of responsibility for the breakdown.

From an applied and experimental standpoint, the scale addresses critical questions regarding customer journey dynamics, service recovery design, and customer retention. When an offering fails, the extent to which the customer holds the firm liable dictates the trajectory of downstream cognitive, affective, and behavioral outcomes. High external failure attribution is empirically linked to intensified consumer anger, increased propensity for negative word-of-mouth (NWOM), third-party complaining behavior, retaliatory behavior, and catastrophic churn. Conversely, when failure attribution is mitigated—or shared, as observed in high-involvement co-creation settings—the negative psychological impact on brand equity is substantially attenuated.

Furthermore, the scale was engineered to investigate the complex psychological trade-offs inherent in contemporary service paradigms, such as customer co-production and self-service technologies (SSTs). As consumers increasingly participate in the creation of value (e.g., assembling products, using self-checkout kiosks, co-designing financial or healthcare plans), the boundary lines of operational failure blur. Dong, Evans, and Zou (2008) introduced this scale to demonstrate how customer participation during initial service delivery and subsequent recovery affects whether customers direct blame toward the provider or recognize shared responsibility. Consequently, the PFA scale enables behavioral researchers and marketing scientists to evaluate experimental interventions, recovery communication strategies (e.g., apologies versus explanations), and operational designs aimed at defusing unjustified external blame attributions.

Psychological Construct

The Product Failure Attribution scale operationalizes a specific domain within social cognition: the subjective allocation of causal agency, accountability, and blame for an adverse event to an external institutional actor. Within psychometrics and cognitive psychology, attribution is not treated as an objective assessment of mechanical or procedural breakdown; rather, it is a socio-cognitive construction influenced by cognitive biases, information asymmetries, and affective heuristics.

The construct measured by the PFA scale encompasses three tightly interrelated conceptual facets:

• Causal Locus (Primary Causation): Refers to the perceived origin of the event. In the PFA scale, item 1 (“The problem was primarily caused by the service provider”) captures the cognitive appraisal that the antecedent conditions generating the disruption resided within the firm’s sphere of operations, personnel, or systems, rather than stemming from consumer error, third-party interference, or uncontrollable environmental factors (force majeure).
• Responsibility Assignment: Distinct from mere physical or proximal causation, responsibility implies an operational duty or standard of care. Item 2 (“The service provider is responsible for the problem that occurred”) measures the moral and organizational expectation that the provider had the obligation to foresee, mitigate, or prevent the failure from transpiring.
• Blame Attribution: Blame represents the evaluative, sanction-oriented culmination of the attributional sequence. Item 3 (“The service provider is to blame for the service failure”) reflects the consumer’s emotional and normative judgment that the firm is culpable, blameworthy, and legitimately accountable for the negative utility inflicted upon the consumer.

Psychologists and consumer researchers draw vital distinctions between causality, responsibility, and blame. While causation concerns the empirical question of who or what set the failure in motion, responsibility assesses whether the actor had control and foresight over the outcome. Blame adds an emotional and moral indictment, frequently triggering psychological desires for retribution, restitution, or cognitive closure. The PFA scale unifies these facets into a coherent unidimensional latent construct reflecting the intensity of provider-directed fault.

The construct is particularly sensitive to psychological phenomenologies such as the self-serving bias (the tendency for individuals to attribute successes to their own actions and failures to external circumstances) and the fundamental attribution error (overemphasizing internal dispositions or provider incompetence while underestimating situational constraints faced by the provider). When measuring product or service failure attribution, the scale captures the net manifestation of these cognitive tendencies in consumer-firm exchanges.

Theoretical Framework

The PFA scale is anchored primarily in Bernard Weiner’s Attribution Theory of motivation and emotion (1985, 1986). Weiner proposed that individuals evaluate unexpected, negative, or significant life events along three core causal dimensions:

1. Locus of Causality: Whether the cause of an event is internal to the actor (the consumer) or external (the firm or environment).
2. Controllability: The degree to which the cause was perceived to be volitionally preventable by the responsible entity.
3. Stability: Whether the cause is permanent/recurrent or temporary/fluctuating across time.

Weiner’s cognitive-emotion-action paradigm dictates that causal ascriptions directly govern emotional reactions, which in turn drive behavioral decisions. In consumer contexts (Folkes, 1984; Bitner, 1990), when consumers ascribe the locus of causality to the firm (external to the consumer, internal to the organization) and perceive the failure as controllable by the firm, they experience elevated anger and moral outrage. These affective states mediate subsequent behavioral responses, such as boycotting, complaining to management, litigating, or spreading damaging word-of-mouth.

In addition to Weiner’s classical model, the PFA scale draws upon Mark Alicke’s Culpable Control Model of blame attribution (2000). Alicke posits that blame is not purely a rational, stepwise calculation of mechanical cause and effect; instead, it is often driven by spontaneous affective evaluations of the actor’s behavior. When a service or product fails, consumers rapidly form an initial negative impression, which then biases subsequent assessments of the firm’s intent, foresight, and causal involvement. The PFA scale captures this aggregated psychological perception of culpable control.

Finally, the scale interacts deeply with Service-Dominant (S-D) Logic and co-creation paradigms formulated by Vargo and Lusch (2004). Under S-D logic, value is always co-created through the combined inputs of the firm and the customer. When value destruction occurs instead of value creation (i.e., service failure), the attribution of causality becomes ambiguous. Dong et al. (2008) utilized the theoretical underpinning of attribution theory to demonstrate that greater customer participation during service recovery alters the customer’s cognitive calculus: by actively engaging in resolving the failure, customers perceive higher internal control, which moderates their attribution of sole blame to the service provider, thereby facilitating psychological reconciliation and restoring transactional satisfaction.

Validity

The psychometric validity of the Product Failure Attribution scale has been substantiated through rigorous quantitative validation across multiple empirical studies, experimental designs, and cross-sectional surveys in marketing and organizational psychology.

Construct and Convergent Validity

Construct validity—the extent to which the scale operationalizes the theoretical construct it purports to measure—was established by Dong et al. (2008) using structural equation modeling (SEM) and confirmatory factor analysis (CFA). All standardized factor loadings for the three items systematically exceed the conventional threshold of .70, typically loading between .85 and .95 on the single latent attribution factor. Furthermore, the Average Variance Extracted (AVE) consistently surpasses .75 (well above the conservative cutoff of .50 proposed by Fornell & Larcker, 1981), demonstrating that the latent construct accounts for the vast majority of the variance in its observed measurement items, confirming robust convergent validity.

Discriminant Validity

Discriminant validity has been demonstrated by verifying that the PFA scale remains statistically and conceptually distinct from related constructs within the nomological network, such as:

• Failure Severity: The perceived magnitude of the financial, physical, or psychological loss incurred by the consumer. Correlation analyses confirm that while failure severity moderately correlates with failure attribution (r ≈ .30 to .45), the shared variance does not exceed the AVE of the individual scales.
• Overall Service Dissatisfaction: The broad affective response toward the consumption experience. Discriminant validity tests using the Fornell-Larcker criterion and the Heterotrait-Monotrait ratio of correlations (HTMT < .85) verify that failure attribution functions as a cognitive antecedent rather than a synonymous manifestation of dissatisfaction.
• Perceived Justice (Distributive, Procedural, Interactional): Perceptions of fairness in the firm’s resolution attempts consistently exhibit negative relationships with provider failure attribution (r ranging from −.35 to −.55), establishing statistical divergence.

Predictive and Nomological Validity

The scale demonstrates exceptional predictive validity across experimental and longitudinal investigations. In service recovery experiments, scores on the PFA scale significantly predict:

• Post-Recovery Satisfaction: As predicted by attribution theory, higher PFA scores exert a robust negative direct effect on customer satisfaction with recovery (standardized path coefficients typically β = −.30 to −.48, p < .001).
• Repatronage Intentions: Attributing blame to the service provider significantly decreases the probability that a customer will repurchase or remain loyal to the firm.
• Consumer Retaliation and Negative Word-of-Mouth: High PFA scores strongly predict consumer retaliatory behaviors, public complaining on social media platforms, and third-party dispute filings.

Reliability

The reliability of the Product Failure Attribution scale has been thoroughly established across diverse customer samples, experimental scenarios, and real-world service environments. The instrument exhibits remarkable internal consistency and stability across diverse operational contexts.

Internal Consistency

Internal consistency evaluates the degree to which items within an instrument measure the same underlying construct. Across multiple published studies, the scale’s Cronbach’s alpha (α) consistently surpasses standard psychometric benchmarks (α ≥ .80 for basic research, α ≥ .90 for applied diagnostics):

• In the foundational study by Dong, Evans, and Zou (2008), the scale yielded a Cronbach’s alpha of .92, indicating that 92% of the variance observed in the composite score is attributable to true score variance rather than measurement error.
• Subsequent replications in varied service environments (e.g., hospitality, airlines, telecommunications, financial services) have reported Cronbach’s alpha values consistently ranging between .88 and .94.
• Composite Reliability (CR): Structural equation modeling evaluations report CR values routinely exceeding .90, demonstrating that the latent construct is reliably indicated by its manifest variables without excessive item redundancy.

Test-Retest Stability and Standard Error

In experimental longitudinal designs assessing pre- and post-intervention evaluations where no recovery intervention occurs, the scale demonstrates substantial temporal stability (test-retest correlation coefficients r > .80 across short intervals). The Standard Error of Measurement (SEM) remains minimal, indicating precise measurement properties across the full range of the 7-point continuum.

Factor Analysis

Extensive factor-analytic testing confirms that the Product Failure Attribution scale is strictly unidimensional. Both exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) confirm that the three items load cleanly onto a single underlying factor representing external provider blame.

Exploratory Factor Analysis (EFA)

When subjected to EFA using principal axis factoring or principal component analysis without rotation (given the unidimensional structure):

• A single factor consistently emerges with an eigenvalue well above Kaiser’s criterion of 1.0 (typically ranging from 2.40 to 2.75).
• The primary factor accounts for 80% to 91% of the total variance across the items.
• Scree plot examinations show a sharp elbow drop immediately following the first factor, with no secondary factors demonstrating an eigenvalue greater than 0.35.

Confirmatory Factor Analysis (CFA)

Confirmatory factor analytic evaluations conducted in AMOS, LISREL, and Mplus report exceptional model fit across independent calibration and validation samples. Standard fit indices for the single-factor measurement model regularly satisfy the most stringent cutoff criteria established by Hu and Bentler (1999):

Individual completely standardized factor loadings for the three items are exceptionally uniform and strong:

• Item 1 (Causation): λ = .86 – .92
• Item 2 (Responsibility): λ = .89 – .95
• Item 3 (Blame): λ = .88 – .94

These empirical findings confirm that the three items function as interchangeable, highly reliable manifest indicators of a unified cognitive construct.

Instrument / Measurement Tool

The Product Failure Attribution scale is structured as a brief self-report psychometric test designed for easy embedding within consumer surveys, experimental post-manipulation checks, and customer experience tracking systems.

• Instrument Name: Product Failure Attribution (PFA) Scale
• Construct Assessed: Consumer attribution of cause, responsibility, and blame to an external service provider or product manufacturer.
• Format: Self-administered questionnaire (paper-and-pencil or computerized/online survey platforms such as Qualtrics, SurveyMonkey, or Prolific).
• Item Count: 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).
• Administration Time: Approximately 1 to 2 minutes.
• Target Population: Adult consumers, service clients, and experimental participants who have experienced or evaluated a simulated or real service/product disruption.
• Scoring and Computational Rules:
  • All three items are positively worded (scored directly from 1 to 7).
  • There are no reverse-scored items.
  • Individual respondent scores are calculated by averaging the three items (yielding a composite mean score ranging from 1.00 to 7.00) or by summing them (yielding a sum score ranging from 3 to 21).
  • Interpretation: Higher aggregate scores indicate stronger attribution of failure, culpability, and blame to the service provider. Lower scores reflect that the customer attributes the problem to self-inflicted mistakes, third parties, or unavoidable environmental circumstances.

Permissions & Fee and Test Year

The Product Failure Attribution scale was formally published in 2008 in the Journal of the Academy of Marketing Science (Volume 36, Issue 1, pp. 123–137), published by Springer. Under academic standard conventions:

• Academic Research Use: The scale items may be utilized without monetary charge for non-commercial academic research, pedagogical purposes, theses, dissertations, and non-profit scientific scholarship, provided proper bibliographic citation is given to Dong, Evans, and Zou (2008).
• Commercial and Proprietary Use: Commercial market research, corporate customer intelligence auditing, or incorporation into proprietary commercial SaaS analytics may require formal copyright clearance and permission from the copyright holder (Springer Nature / Academy of Marketing Science) via the Copyright Clearance Center (CCC) or the publisher’s RightsLink service.
• Fee: Free for non-commercial academic use. Commercial licensing fees are subject to publisher terms.

References

• Alicke, M. D. (2000). Culpable control and the psychology of blame. Psychological Bulletin, 126(4), 556–574. https://doi.org/10.1037/0033-2909.126.4.556
• Bitner, M. J. (1990). Evaluating service encounters: The effects of physical surroundings and employee efforts. Journal of Marketing, 54(2), 69–82. https://doi.org/10.1177/002224299005400106
• Dong, B., Evans, K. R., & Zou, S. (2008). The effects of customer participation in co-created service recovery. Journal of the Academy of Marketing Science, 36(1), 123–137. https://doi.org/10.1007/s11747-007-0059-8
• Folkes, V. S. (1984). Consumer reactions to product failure: An attributional approach. Journal of Consumer Research, 10(4), 398–409. https://doi.org/10.1086/208978
• 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
• Oliver, R. L. (1997). Satisfaction: A behavioral perspective on the consumer. McGraw-Hill.
• Vargo, S. L., & Lusch, R. F. (2004). Evolving to a new dominant logic for marketing. Journal of Marketing, 68(1), 1–17. https://doi.org/10.1509/jmkg.68.1.1.24036
• 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

Items of the Scale