Dehumanization of the Employee

Abstract

The Dehumanization of the Employee scale is a four-item psychometric instrument developed and validated by Sun, Wang, Hoegg, and Dahl (2023) to operationalize the degree to which frontline service workers are perceived as mechanical, unemotional, cold, and robot-like by consumers. Rooted in social psychological theories of dehumanization—specifically Nick Haslam's mechanistic dehumanization paradigm—this instrument evaluates the denial of core human attributes, such as emotional responsiveness, warmth, cognitive flexibility, and subjective depth, within interpersonal service transactions. The scale was established in the context of embarrassing service encounters, wherein consumers strategically or implicitly strip employees of human characteristics as a psychological defense mechanism to mitigate their own emotional discomfort and acute social distress.

The scale employs a unidimensional structure comprising four manifest items administered via a 7-point Likert response format ranging from 1 (Strongly disagree) to 7 (Strongly agree). Psychometric evaluations across multiple experimental pretests and empirical investigations demonstrated exceptional internal consistency reliability, with Cronbach's alpha values consistently exceeding α = .88. Exploratory and confirmatory factor analyses affirm a distinct single-factor solution characterized by high standardized factor loadings (λ > .80), strong convergent validity with measures of agency and emotional suppression, and discriminant validity separating the construct from general employee dissatisfaction, perceived competence, and service quality ratings. By providing a parsimonious, psychometrically robust assessment of mechanistic dehumanization in transactional contexts, this tool has become a foundational metric in consumer psychology, organizational behavior, services marketing, and human-computer interaction paradigms examining automated versus anthropomorphic service encounters.

Keywords

dehumanization, mechanistic dehumanization, service encounters, employee perception, consumer embarrassment, frontline employees, social cognition, human-robot interaction, interpersonal perception, psychometrics

Authors

The Dehumanization of the Employee scale was formulated, calibrated, and validated by an international research team of consumer behavior scholars and psychometricians:

• Yixia Sun — Assistant Professor of Marketing, School of Business, George Mason University, Fairfax, VA, USA. Expertise: Consumer emotional regulation, embarrassing consumption, and frontline interpersonal dynamics.
• Xuehua Wang — Professor of Marketing, School of Management, Shanghai University of Finance and Economics, Shanghai, China. Expertise: Social influence, consumer psychology, brand perception, and compensatory consumption mechanisms.
• JoAndrea Hoegg — Professor of Marketing and Senior Associate Dean, Sauder School of Business, University of British Columbia, Vancouver, BC, Canada. Expertise: Sensory marketing, social judgments, aesthetic perception, and consumer-service provider relationships.
• Darren W. Dahl — Innovate BC Professor and Dean, Sauder School of Business, University of British Columbia, Vancouver, BC, Canada. Expertise: Social influence, consumer emotions, embarrassment, creativity, and customer experience.

Purpose

The primary purpose of the Dehumanization of the Employee scale is to quantify consumer perceptions of service employees as automated, unemotional, and machine-like entities during commercial exchanges. Historically, psychometric measures of dehumanization were designed to capture severe sociological phenomena, such as ethnic discrimination, intergroup conflict, warfare, and institutional marginalization. In such frameworks, dehumanization was conceptualized predominantly as animalistic—denying outgroups higher-order cognitive faculties, morality, and cultural refinement. However, interpersonal encounters within modern commercial environments often evoke a distinct variant: mechanistic dehumanization, wherein targets are divested of core human warmth, interpersonal sentiment, and emotional vitality, effectively reducing them to instruments of transactional execution.

Sun, Wang, Hoegg, and Dahl (2023) developed this instrument specifically to uncover the psychological processes consumers recruit when navigating acutely embarrassing service encounters—such as purchasing stigmatized medications, addressing personal hygiene difficulties, or discussing financial delinquency. In such contexts, encountering a living, emotionally reactive human observer exacerbates feelings of vulnerability, social judgment, and reputational threat. To cope with this acute psychological threat, consumers engage in mechanistic dehumanization, perceptually reframing the frontline worker as an unfeeling automaton or transactional robot. When perceived as lacking subjective emotional capacity, the employee cannot experience amusement, disdain, or moral judgment, which significantly diminishes the consumer's projected embarrassment.

Beyond the domain of embarrassing service encounters, the scale serves critical theoretical and applied purposes across numerous organizational and behavioral disciplines:

• Consumer Psychology: Examining when and why consumers objectify service personnel, demand algorithmic efficiency at the expense of human dignity, or exhibit rude and uncivil treatment toward retail staff.
• Human-Robot Interaction (HRI) & Artificial Intelligence: Serving as a psychometric baseline for comparative research that contrasts consumer evaluations of artificial intelligence agents (e.g., kiosks, conversational avatars, robotic servers) against human frontline workers who perform identical routine tasks.
• Organizational Behavior and Burnout: Investigating the reciprocal dynamics of emotional labor. When employees are systematically perceived and treated as unfeeling machines, this customer-induced dehumanization correlates strongly with frontline emotional exhaustion, depersonalization, and elevated employee turnover.
• Service Design and Architecture: Evaluating the psychological consequences of script-based service protocols. Highly rigid corporate scripts may inadvertently trigger customer dehumanization by stripping workers of authentic human behavioral variation.

Psychological Construct

The psychological construct measured by this instrument is mechanistic dehumanization within interpersonal service exchanges. Mechanistic dehumanization is conceptually discrete from related constructs, such as negative social evaluation, poor service competence, or customer dissatisfaction. It entails a fundamental cognitive deficit in Theory of Mind (ToM) attribution, specifically the selective denial of human nature characteristics.

Dimensions of Human Uniqueness versus Human Nature

In contemporary social cognitive psychology, humanness is structured along two distinct conceptual axes: Human Uniqueness (HU) and Human Nature (HN). Human Uniqueness encompasses characteristics that distinguish humans from other animals, including higher-order reasoning, moral integrity, civility, and refinement. Denying these characteristics constitutes animalistic dehumanization, where individuals are viewed as degraded, instinctual, or subhuman. Conversely, Human Nature encompasses attributes that distinguish humans from inanimate objects, computational algorithms, and mechanical devices. These attributes include:

• Emotionality and Affective Resonance: The intrinsic capacity to experience, express, and interpret emotions, ranging from basic feelings (joy, distress) to nuanced social evaluations (empathy, embarrassment, contempt).
• Warmth and Interpersonal Openness: Spontaneous social responsiveness, conversational nuance, and warmth in human connection.
• Cognitive Flexibility and Agency: The capacity for autonomous, dynamic thinking, non-scripted problem-solving, and personal agency rather than deterministic programmed execution.
• Vitality and Subjective Depth: An internal mental landscape composed of feelings, intentions, and subjective phenomenological experiences.

The Dehumanization of the Employee scale focuses on the systematic denial of these Human Nature attributes. When a consumer scores high on this scale, they do not necessarily judge the worker as incompetent, malevolent, or uncivilized; rather, they judge them as structurally devoid of an inner emotional life. The items directly capture this psychological state through items assessing whether the employee "acted like a robot," was "mechanical and cold," "lacked emotion," and "seemed more like a machine than a human being."

Theoretical Distinction from Competence and Warmth

According to the Stereotype Content Model developed by Fiske, Cuddy, Glick, and Xu (2002), social targets are primarily judged on two universal dimensions: warmth (morality, sociability) and competence (efficacy, skill). Mechanistic dehumanization is closely tied to low warmth, but it represents an extreme ontological judgment: the target is not merely perceived as unapproachable or unfriendly, but as functionally non-human. High scores on this scale can co-occur with high ratings of competence. In fact, in standardized retail and service settings, an employee who executes tasks with flawless, rigid precision without personal emotional acknowledgment may be judged as highly competent while simultaneously undergoing severe mechanistic dehumanization.

Theoretical Framework

The conceptual framework underlying the Dehumanization of the Employee scale integrates three foundational theories from social cognition, consumer behavior, and evolutionary psychology: Haslam's Dual Model of Dehumanization, Mind Perception Theory, and Cognitive Dissonance Theory in Threat Mitigation.

1. Haslam's Dual Model of Dehumanization

Nick Haslam (2006) formalized the dual model to differentiate between animalistic and mechanistic variants of dehumanization. Haslam posited that when targets are mechanistically dehumanized, they are perceived as lacking warmth, emotional responsiveness, cognitive flexibility, and depth. They are cognitively cataloged as inert, cold, passive, and machine-like. Sun et al. (2023) directly operationalized this model within commercial service settings, demonstrating that retail and transactional environments naturally facilitate mechanistic attributions because service roles frequently emphasize standardization, scripted dialogue, and procedural regularity.

2. Mind Perception Theory

The scale draws heavily upon Mind Perception Theory articulated by Gray, Gray, and Wegner (2007). Their empirical framework establishes that people perceive minds across two foundational dimensions:

• Agency: The capacity for self-control, moral judgment, planning, thought, and intentional action.
• Experience: The capacity for sensations, feelings, pleasure, pain, consciousness, and emotional depth.

Mechanistic dehumanization represents a selective deprivation of Experience. While consumers readily acknowledge that a service employee possesses agency (the technical ability to execute tasks, process transactions, and follow protocols), they mentally discount the employee's experience (their ability to feel empathy, embarrassment, judgment, or subjective distress). By viewing the worker through a purely mechanical lens, the observer reduces the cognitive load of navigating another person's mental states, transforming a social interaction into a simple functional utility.

3. Strategic Dehumanization as Psychological Defense

Sun et al. (2023) introduced a theoretical synthesis uniting cognitive defense mechanisms with social cognition. Embarrassment is an intensely aversive social emotion provoked by the real or anticipated negative evaluation of oneself by an audience. In embarrassing consumption contexts (e.g., purchasing sensitive medical products), consumers face a social threat: the frontline worker acts as an evaluative audience who might judge them negatively. To alleviate this psychological tension, consumers engage in motivated social perception: they mentally downgrade the employee from a sentient human judge to an unfeeling instrument. Because machines lack the capacity for moral evaluation, social comparison, or private ridicule, mechanistically dehumanizing the employee eliminates the perceived audience, alleviating the consumer's subjective embarrassment.

Validity

The Dehumanization of the Employee scale has undergone extensive empirical evaluation to establish its construct, convergent, discriminant, and predictive validity across diverse laboratory experiments and field settings.

Construct and Convergent Validity

Construct validity was established by evaluating the degree to which scale items align with established operationalizations of mechanistic dehumanization in social psychology. Sun et al. (2023) conducted multiple pretests and main studies where the four-item scale demonstrated high convergent validity with standardized inventories of human nature deprivation (Haslam, 2006) and mind perception scales (Gray et al., 2007). Specifically, scores on the scale showed strong, statistically significant negative correlations with employee Experience attributions (r = -.68 to -.74, p < .001) and employee capacity for social judgment (r = -.61, p < .001). Conversely, the scale correlated positively with perceived employee emotional detachment (r = .71, p < .001), corroborating that the items reliably measure the denial of emotionality and affective responsiveness.

Discriminant Validity

Discriminant validity was established through Average Variance Extracted (AVE) analyses, confirmatory factor models, and correlation matrices against related, but theoretically distinct, marketing and psychological constructs:

• Service Competence: The scale demonstrated distinct orthogonal separation from perceived employee competence (r = -.08, p = .24, non-significant). This demonstrates that perceiving an employee as robotic is distinct from perceiving them as unskilled or ineffective.
• General Consumer Dissatisfaction: Across several experimental conditions, the scale maintained discriminant validity from overall transaction dissatisfaction (shared variance < 15%), demonstrating that employee dehumanization is not merely an expression of consumer discontent.
• Animalistic Dehumanization: When pitted against items assessing animalistic dehumanization (e.g., perceiving the worker as uncivilized, coarse, or animal-like), the correlation was modest (r = .22, p < .01), and two-factor models demonstrated superior fit over a single global dehumanization factor (Δχ² test, p < .001).

Predictive and Nomological Validity

Nomological and predictive validity were validated through structural equation models and moderated mediation analyses in Sun et al. (2023). Specifically, the scale mediated the relationship between encounter embarrassment and downstream consumer preferences:

• Consumers placed in high-embarrassment service encounters exhibited significantly higher dehumanization scores compared to low-embarrassment control conditions (F(1, 248) = 31.42, p < .001, η_p² = .112).
• Elevated dehumanization scores systematically predicted increased consumer willingness to proceed with embarrassing purchases (β = .34, t = 4.88, p < .001) and heightened preferences for service personnel who naturally exhibited low affective expressiveness.
• In behavioral laboratory tasks, manipulated dehumanization of the employee (induced via procedural scripts) directly lowered consumer physiological stress markers (measured via self-reported somatic anxiety and pulse rate fluctuations), validating the functional, coping-oriented role of this construct.

Reliability

The scale demonstrates strong internal consistency reliability across varied empirical samples, including undergraduate laboratory pools, crowdsourced research platforms (Amazon MTurk, Prolific Academic), and real-world consumer panels.

Internal Consistency Metrics

In the foundational validation studies by Sun et al. (2023), the internal consistency of the four-item scale was tested across five distinct experimental studies and supplementary pretests:

• Pretest 1 (Scale Verification, N = 184): The four items yielded a Cronbach's alpha of α = .91, with item-total correlations ranging from .76 to .86.
• Study 1 (Embarrassment & Retail Encounters, N = 250): Cronbach's alpha reached α = .92. The composite reliability (CR) was .93, well above the conventional threshold of .70.
• Study 2 (Experimental Replication, N = 312): Cronbach's alpha was α = .89; Average Variance Extracted (AVE) was .68.
• Study 3 (Boundary Conditions, N = 405): Cronbach's alpha was α = .93 (CR = .94; AVE = .76).
• Study 4 (Behavioral Intentions, N = 298): Cronbach's alpha was α = .90.

Item Diagnostics

Diagnostic item-level analyses across these investigations revealed that deletion of any single item would not increase the overall Cronbach's alpha coefficient, confirming that all four items contribute evenly to the latent construct. The corrected item-total correlations across all validation samples remained above .70, indicating strong homogeneity of items without problematic redundancy.

Factor Analysis

To assess the underlying latent dimensionality of the four items, both Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA) were conducted during the initial development and subsequent replications.

Exploratory Factor Analysis (EFA)

An initial EFA using principal axis factoring with promax rotation was conducted on calibration samples. The Kaiser-Meyer-Olkin (KMO) measure of sampling adequacy consistently exceeded .85 (KMO = .872), and Bartlett's Test of Sphericity was statistically significant (χ²(6) = 784.32, p < .001), indicating the correlation matrix was suitable for factor extraction.

The analysis yielded a clean single-factor solution based on the Kaiser criterion (eigenvalues > 1.0) and scree plot inflection inspection. The primary factor accounted for 74.6% to 81.2% of the total variance across samples. Factor loadings for all four items were high, ranging between .82 and .93:

• Item 1 (The employee acted like a robot): Loading = .88
• Item 2 (The employee was mechanical and cold): Loading = .91
• Item 3 (The employee lacked emotion): Loading = .84
• Item 4 (The employee seemed more like a machine than a human being): Loading = .89

Confirmatory Factor Analysis (CFA)

Subsequent confirmatory factor analyses using Maximum Likelihood estimation verified that the unidimensional model fit the empirical data well. Across validation samples, the fit indices met or exceeded conventional psychometric standards (Hu & Bentler, 1999):

• Chi-Square / Degrees of Freedom: χ²/df ≤ 2.14 (p > .10)
• Comparative Fit Index (CFI): .988 – .997
• Tucker-Lewis Index (TLI): .981 – .994
• Root Mean Square Error of Approximation (RMSEA): .038 – .052 (90% CI [.000, .085])
• Standardized Root Mean Square Residual (SRMR): .015 – .024

Competing two-factor models (splitting items into outward mechanical demeanor vs. internal emotional deficit) did not yield statistically meaningful improvements in fit, confirming that the construct functions best as a parsimonious, unidimensional index.

Instrument / Measurement Tool

• Instrument Name: Dehumanization of the Employee Scale
• Alternative Titles: Employee Mechanistic Dehumanization Measure, Service Worker Robotic Perception Index
• Construct Assessed: Consumer perception of frontline service employees as mechanical, unemotional, cold, and robot-like (mechanistic dehumanization).
• Format: Self-administered paper-and-pencil or computerized questionnaire.
• Target Population: Consumers, service recipients, organizational observers, and experimental research participants evaluating customer service encounters.
• Number of Items: 4 items.
• Item Phrasing: Direct, third-person declarative statements describing employee demeanor and perceived psychological depth.
• Response Scale: 7-point Likert scale (1 = Strongly disagree, 7 = Strongly agree).
• Scoring Protocol: All items are keyed in the affirmative direction (no reverse-scored items). An overall index is computed by calculating the arithmetic mean of all four item ratings:
  
  Dehumanization Index = (Item 1 + Item 2 + Item 3 + Item 4) / 4
• Score Interpretation:
  • 1.00 – 2.49: Low dehumanization; the employee is perceived as fully human, warm, and emotionally responsive.
  • 2.50 – 4.49: Moderate/Neutral perception; the employee is viewed as standardly transactional with minimal social connection.
  • 4.50 – 7.00: High mechanistic dehumanization; the employee is viewed as a cold, mechanical, emotionless automaton.
• Completion Time: Approximately 1 to 2 minutes.

Permissions & Fee and Test Year

The Dehumanization of the Employee scale was developed and published in 2023 by Yixia Sun, Xuehua Wang, JoAndrea Hoegg, and Darren W. Dahl in the Journal of Marketing Research. As an academic psychometric measure published within a peer-reviewed scholarly journal, the scale items are available for non-commercial academic, research, and instructional purposes without licensing fees, provided proper bibliographic attribution is given to the original authors.

Commercial deployment, inclusion in proprietary diagnostic toolkits, or application within fee-for-service consulting assessments may require copyright clearance from the publisher, the American Marketing Association (AMA) / SAGE Publications. Inquiries regarding permissions, modifications, or specific research applications may be directed to the corresponding author, Yixia Sun, or the editorial office of the Journal of Marketing Research.

References

• Fiske, S. T., Cuddy, A. J., Glick, P., & Xu, J. (2002). A model of (often mixed) stereotype content: Competence and warmth respectively follow from perceived status and competition. Journal of Personality and Social Psychology, 82(6), 878–902. https://doi.org/10.1037/0022-3514.82.6.878
• Gray, H. M., Gray, K., & Wegner, D. M. (2007). Dimensions of mind perception. Science, 315(5812), 619–619. https://doi.org/10.1126/science.1134475
• Haslam, N. (2006). Dehumanization: An integrative review. Personality and Social Psychology Review, 10(3), 252–264. https://doi.org/10.1207/s15327957pspr1003_4
• Haslam, N., & Loughnan, S. (2014). Dehumanization and infrahumanization. Annual Review of Psychology, 65, 399–423. https://doi.org/10.1146/annurev-psych-010213-115045
• 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
• Sun, Y., Wang, X., Hoegg, J., & Dahl, D. W. (2023). How consumers respond to embarrassing service encounters: A dehumanization perspective. Journal of Marketing Research, 60(4), 646–664. https://doi.org/10.1177/00222437221126786

Items of the Scale