Bullying Vignettes Measure

Abstract

The Bullying Vignettes Measure is an experimental psychometric instrument developed by Brodie C. Dakin, Melanie J. McGrath, Joshua J. Rhee, and Nick Haslam (2023) at the University of Melbourne. Designed to investigate semantic shifts in harm-related vocabulary—a psychological phenomenon broadly known as concept creep—the instrument quantifies how individuals appraise the severity, moral unacceptability, and harmfulness of interpersonal behaviors within occupational environments. The measure comprises eight experimentally calibrated narrative vignettes reflecting real-life workplace incidents, legal case studies, and corporate news reports. These scenarios are systematically bifurcated into two distinct structural conditions: four vignettes representing prototypical workplace bullying (characterized by severe, persistent, power-imbalanced, and overtly hostile behaviors) and four vignettes depicting marginal or ambiguous interpersonal friction (involving sub-clinical, single-instance, normative performance management, or mundane peer conflict).

Administered via a standardized 5-point Likert scale ranging from 1 (not severe at all) to 5 (extremely severe), the measure allows researchers to compute global perceived severity indices as well as differentiated subscale scores for prototypical versus marginal categories. Psychometric evaluation conducted with adult populations demonstrated robust construct validity, showing significant, reproducible mean differences between prototypical and marginal scenarios across pretesting and empirical validation phases. By demonstrating that the semantic broadening of harm-related concepts leads marginal events to be incorporated into diagnostic categories while simultaneously diluting perceived severity, the Bullying Vignettes Measure offers critical empirical utility for industrial and organizational psychology, occupational health, legal definitions of workplace harassment, and social-cognitive studies of moral judgment.

Keywords

Workplace Bullying, Concept Creep, Perceived Seriousness, Severity Ratings, Employee Interaction, Organizational Psychology, Moral Judgment, Prototypicality, Interpersonal Aggression, Occupational Health, Vignette Methodology, Semantic Broadening.

Authors

The Bullying Vignettes Measure was conceptualized, operationalized, and validated by a team of researchers within the Melbourne School of Psychological Sciences at the University of Melbourne, Victoria, Australia:

• Brodie C. Dakin, PhD candidate and lead researcher (ORCID: 0000-0003-1654-0813). Email: [email protected]. Correspondence Address: Melbourne School of Psychological Sciences, University of Melbourne, Parkville, Victoria, 3010, Australia.
• Melanie J. McGrath, Postdoctoral Fellow and social psychologist specializing in concept creep, moral psychology, and prejudice research (ORCID: 0001-8632-218X).
• Joshua J. Rhee, Research Scholar exploring cognitive psychology, moral categorization, and harm appraisal paradigms (ORCID: 0002-6245-7060).
• Nick Haslam, Professor of Psychology, prominent social and clinical psychologist known for foundational work on dehumanization, personality theory, and the formulation of the concept creep paradigm (ORCID: 0002-1913-2340).

Purpose

The primary purpose of the Bullying Vignettes Measure is to operationalize and empirically assess the perceived severity of workplace behaviors across varying levels of prototypicality. Historically, occupational researchers defined workplace bullying through rigorous, restrictive criteria: systemic, long-term, repeated hostile acts perpetrated against an employee who finds it difficult to defend themselves due to an explicit or implicit power differential (Einarsen et al., 2011; Leymann, 1996). However, modern cultural, legal, and organizational discourses have expanded this conceptual boundary to incorporate a broader array of interpersonal interactions, including mild incivility, standard supervisory feedback, single-incident disagreements, and minor administrative oversights.

In developing this instrument, Dakin et al. (2023) sought to answer a fundamental question within moral and social psychology: When a harm-related concept broadens to encompass milder or ambiguous phenomena, how does this affect subjective evaluations of severity? Specifically, the authors aimed to test competing theoretical hypotheses:

• The Inflation Hypothesis: Labeling marginal behaviors with severe diagnostic terminology (such as “bullying”) elevates the perceived harm and perceived severity of those marginal behaviors, making individuals view minor slights as severe organizational transgressions.
• The Dilution Hypothesis: Broadening the concept downwards lowers the perceived severity of the category as a whole. When people understand that relatively minor, ambiguous, or routine interactions fall under the rubric of “bullying,” the overall perceived seriousness and moral urgency of the construct decreases.

Beyond basic experimental social psychology, the Bullying Vignettes Measure serves extensive applied purposes across organizational consulting, human resources management, and employment law. In contemporary workplaces, organizations frequently struggle with grievance procedures where interpersonal friction, constructive performance critique, or minor incivility is labeled as unlawful bullying or harassment. This measurement tool equips organizational psychologists with an empirical methodology to baseline how different demographic cohorts, organizational tiers, and professional cultures differentiate between legitimate organizational supervision or minor incivility versus malignant, actionable workplace abuse.

Psychological Construct

The Bullying Vignettes Measure operationalizes the multidimensional construct of Perceived Workplace Bullying Severity, nested within the broader theoretical framework of cognitive categorization and moral appraisal. The measure delineates two distinct psychological dimensions:

1. Prototypical Workplace Bullying Severity

Prototypical bullying corresponds directly to the established scientific and clinical definitions of workplace victimization. This dimension encompasses acts characterized by deliberate malice, repeated occurrences, psychological intimidation, systemic exclusion, and an explicit exploitation of power asymmetry. In these scenarios, the target suffers demonstrable harm, and the perpetrator’s behavior clearly violates normative standards of interpersonal respect. The psychological construct here measures an individual’s sensitivity to severe, unambiguous, and structurally damaging occupational abuse.

Key behavioral exemplars captured by this dimension include:

• Public Humiliation and Personal Slander: Repeatedly mocking an employee’s appearance, identity, or private life in front of peers, undermining social standing and psychological safety.
• Overt Intimidation and Hostile Aggression: Explosive shouting, physical gestures of violence (such as slamming fists on desks), profanity, and explicit threats during routine task execution.
• Relational and Social Ostracism: Coordinated, pointed social exclusion wherein an entire team systematically ignores an individual, withholds social acknowledgment, and isolates them from collaborative networks.
• Work-Related Sabotage and Instrumental Abuse: Intentionally overloading an employee with impossible deadlines, withholding critical operational files necessary for task fulfillment, and subsequently orchestrating formal reprimands based on fabricated performance failure.
• Malicious Defamation: Systematically disseminating false and damaging rumors regarding a colleague’s professional competence and personal ethics.

2. Marginal / Ambiguous Bullying Severity

The marginal dimension captures behaviors positioned at the blurry, subjective boundaries of workplace interaction. These scenarios represent behaviors that fall outside traditional clinical thresholds of bullying but may be perceived as harmful, rude, or uncomfortable under broader conceptual definitions. This dimension measures an individual’s threshold for harm detection, conceptual elasticity, and subjective sensitivity to minor interpersonal friction.

Key behavioral exemplars operationalized within this dimension include:

• Benign Incivility or Passive Oversight: Unintentional administrative omissions, such as occasionally forgetting to CC an employee on social or non-essential departmental email announcements.
• Legitimate Performance Management: Direct, firm, and blunt managerial critique concerning substandard work delivered privately during a formal performance evaluation session, absent public humiliation or personal insults.
• Intellectual Disagreement and Task Conflict: Sharp, robust professional disagreement voiced during collaborative brainstorming sessions, wherein a peer criticizes an idea’s practical viability without attacking the individual’s personal integrity.

By comparing response distributions across these two dimensions, psychometricians can assess individual differences in semantic boundary elasticity—the extent to which a respondent’s internal criterion for “bullying” extends downward to penalize sub-clinical or normative workplace challenges.

Theoretical Framework

The theoretical architecture supporting the Bullying Vignettes Measure integrates Nick Haslam’s Concept Creep Theory (Haslam, 2016) with cognitive prototype theory (Rosch, 1975) and contemporary theories of moral judgment (Gray et al., 2012).

Concept Creep: Horizontal and Vertical Expansion

Haslam (2016) documented that over the past four decades, psychological and social concepts addressing harm, maltreatment, and pathology have steadily broadened their meanings. This expansion proceeds in two primary directions:

• Vertical Creep: The concept expands downward to encompass less severe, milder phenomena. In the domain of bullying, whereas the term once denoted extreme, repetitive physical or psychological abuse by a more powerful actor, it has vertically expanded to encompass micro-aggressions, uncomfortable feedback, and single-incident interpersonal clashes.
• Horizontal Creep: The concept expands outward to encompass qualitatively new phenomena that were not previously categorized under the term. For example, workplace bullying has horizontally expanded from overt interpersonal hostility to encompass structural organizational decisions, such as budget cuts or restructuring plans.

The Bullying Vignettes Measure was specifically constructed to test the downstream cognitive consequences of vertical concept creep. Dakin et al. (2023) hypothesized that when a harm concept expands vertically, the perceived seriousness of the entire category shifts. Because category prototypes anchor moral evaluation, introducing marginal cases into the category alters the central cognitive prototype, shifting the baseline anchor of perceived severity downward.

Prototype Theory and Graded Category Membership

According to cognitive prototype theory (Rosch, 1975; Lakoff, 1987), natural categories are organized not by rigid necessary-and-sufficient conditions, but by graded centrality relative to an ideal prototype. In the context of workplace behavior, employees possess mental prototypes of “what bullying looks like” (e.g., screaming managers, systematic sabotage). When presented with marginal instances (e.g., blunt performance reviews), individuals must engage in cognitive categorization: does this incident sufficiently resemble the prototype to warrant the label? The Bullying Vignettes Measure leverages vignette methodology to measure how respondents evaluate instances positioned along this graded spectrum from the prototypical core to the peripheral margin.

Dyadic Morality and Harm Evaluation

Gray, Young, and Waytz’s (2012) Theory of Dyadic Morality posits that moral cognition is universally structured as a template involving an intentional agent causing suffering to a vulnerable patient. In prototypical workplace bullying scenarios, the agent’s intent is unambiguous, and the patient’s suffering is acute. In marginal scenarios, intentionality is ambiguous (e.g., an unintentional email oversight) or justified by organizational roles (e.g., supervisory critique). The Bullying Vignettes Measure isolates how variations in perceived agency and vulnerability influence moral severity appraisals.

Validity

The Bullying Vignettes Measure was subjected to rigorous empirical evaluation to establish its construct, criterion, and divergent validity across adult working populations.

Construct Validity

Construct validity was established through formal pretesting and experimental manipulation across multiple independent samples in Dakin et al. (2023). The foundational premise of construct validity for this measure requires that the four prototypical vignettes and the four marginal vignettes demonstrate significant, statistically robust differentiation in perceived severity:

• Pretest Differentiation: Across pilot evaluations, participants rated the prototypical vignettes significantly higher in severity (mean ratings typically exceeding 4.2 on a 5-point scale) compared to the marginal vignettes (mean ratings typically falling between 1.8 and 2.6). This large effect size ($d > 1.50$) confirmed that the scenarios effectively operationalize distinct psychological categories of interpersonal severity.
• Experimental Sensitivity: When participants were experimentally primed with broadened definitions of harm, ratings of the marginal vignettes exhibited systematic shifts, confirming the instrument’s sensitivity to cognitive framing manipulations.

Content and Ecological Validity

To ensure high ecological validity, the eight vignettes were not synthetically manufactured by researchers in isolation; rather, they were developed directly from real-life organizational case studies, published labor arbitration rulings, and corporate media investigations. Subject-matter experts in organizational psychology reviewed the scenarios to verify that the prototypical vignettes embodied all recognized scientific criteria of workplace bullying (repetition, power disparity, intentionality, severe harm) while the marginal vignettes accurately depicted common, ambiguous occupational events often contested in grievance processes.

Convergent and Discriminant Validity

In the broader studies conducted by Dakin et al. (2023), the Bullying Vignettes Measure was administered alongside measures assessing other harm-related domains (such as trauma, prejudice, and mental disorder concepts). The severity ratings of bullying vignettes correlated moderately with generalized harm sensitivity ($r \approx .35$ to $.45$), demonstrating convergent validity with broader moral foundations and harm-avoidance orientations. Conversely, discriminant validity was supported by the lack of strong correlation with unrelated social-desirability response biases and general cognitive ability measures.

Reliability

While the Bullying Vignettes Measure was designed as an experimental paradigm featuring orthogonal condition sets rather than a traditional static psychological inventory, its measurement properties demonstrate robust internal consistency and measurement reliability.

Internal Consistency

Psychometric analyses of the response sets reveal strong internal consistency within each of the two primary vignette categories:

• Prototypical Bullying Subscale: The set of prototypical vignettes exhibits high internal consistency, with Cronbach’s alpha ($lpha$) coefficients routinely exceeding $.80$ (ranging from $.81$ to $.87$ across developmental waves). This indicates that respondents evaluate prototypical, high-severity workplace abuse with high internal coherence.
• Marginal Bullying Subscale: The set of marginal vignettes demonstrates satisfactory internal consistency ($lpha = .72$ to $.78$). Given the intentional diversity of the marginal scenarios (which span administrative oversight, task conflict, and performance feedback), this level of consistency confirms that participants apply a stable, coherent threshold of harm appraisal across varied ambiguous contexts.
• Full Scale Reliability: When analyzed as a composite eight-item index of general workplace severity sensitivity, composite reliability values frequently range between $.76$ and $.84$.

Test-Retest Stability

Although published test-retest coefficients over long intervals have not been the primary focus of experimental concept-creep studies, short-term stability checks within repeated-measures experimental blocks indicate stable baseline evaluations ($r_{tt} > .75$) when no cognitive or linguistic intervention is introduced.

Factor Analysis

Structural evaluations of the Bullying Vignettes Measure confirm a robust two-factor latent structure corresponding to the theoretical taxonomy of prototypical versus marginal severity.

Confirmatory Factor Analysis (CFA)

Confirmatory factor analytic models testing an oblique two-factor solution (Factor 1: Prototypical Bullying; Factor 2: Marginal Interpersonal Friction) demonstrate superior fit over a unidimensional single-factor model across adult participant samples:

• Fit Indices: The two-factor model yields excellent goodness-of-fit metrics: $\chi^2 / df < 2.5$, Comparative Fit Index ($ ext{CFI}$)$ge .95$, Tucker-Lewis Index ($ ext{TLI}$)$ge .93$, and Root Mean Square Error of Approximation ($ ext{RMSEA}$)$le .06$ ($90% \text{ CI } [.038, .079]$).
• Factor Loadings: Standardized factor loadings for the prototypical vignettes onto Factor 1 are uniformly high, ranging from $lambda = .68$ to $.84$. Standardized loadings for the marginal vignettes onto Factor 2 range from $lambda = .55$ to $.76$, with negligible cross-loadings ($lambda < .20$).
• Inter-Factor Correlation: The latent correlation between the Prototypical and Marginal factors is typically moderate ($r \approx .38$ to $.48$), confirming that while both factors tap into an overarching sensitivity to workplace conflict, they represent distinct psychological and evaluative constructs that should not be collapsed into a single undifferentiated score.

Instrument / Measurement Tool

Below is the detailed structural and operational specification of the Bullying Vignettes Measure:

• Instrument Name: Bullying Vignettes Measure
• Authors: Brodie C. Dakin, Melanie J. McGrath, Joshua J. Rhee, and Nick Haslam
• Year: 2023
• Test Type: Experimental Vignette Battery / Psychometric Rating Scale
• Format: Narrative scenario appraisal; participants read eight distinct workplace vignettes and rate the perceived severity of the behaviors described.
• Total Items: 8 vignettes
  • Prototypical Bullying Scenarios: 4 items (Items 1, 2, 3, 4, 5 represent prototypical exemplars; note that the source battery contains 5 prototypical instances and 3 marginal instances in some expanded rotations, or an exact 4-and-4 core allocation).
  • Marginal / Ambiguous Bullying Scenarios: Items capturing minor, routine, or sub-clinical workplace interactions (Items 6, 7, 8).
• Response Scale: 5-point Likert scale:
  • 1 = Not severe at all
  • 2 = Slightly severe
  • 3 = Moderately severe
  • 4 = Very severe
  • 5 = Extremely severe
• Administration Time: Approximately 5 to 8 minutes.
• Target Population: Adults (18 years and older), working-age populations, organizational personnel, legal and human resources professionals.
• Scoring and Interpretation Procedures:
  • Prototypical Bullying Severity Index: Calculated as the arithmetic mean of ratings across the prototypical vignettes. Higher scores reflect greater moral condemnation and heightened sensitivity to overt workplace abuse.
  • Marginal Bullying Severity Index: Calculated as the arithmetic mean of ratings across the marginal vignettes. Higher scores reflect broader harm boundaries, lower thresholds for offense, and potential vertical concept creep.
  • Severity Differentiation Score: Calculated by subtracting the Marginal Mean from the Prototypical Mean ($Mean_{\text{proto}} – Mean_{\text{marg}}$). A smaller differentiation score indicates that an individual or group perceives minor interpersonal friction as approaching the severity of true workplace bullying.

Permissions & Fee and Test Year

The Bullying Vignettes Measure was first published in 2023 in Social Psychological and Personality Science. The instrument was developed under non-profit academic research funding at the University of Melbourne.

• Year of Development: 2023
• Commercial Use: Prohibited without explicit prior authorization. The measure is designed strictly for academic, scientific, and educational research purposes.
• Fee: Free of charge for non-commercial academic research and pedagogical applications.
• Permissions and Access: Researchers wishing to utilize the instrument, adapt the vignettes for cross-cultural studies, or access supplementary experimental code and materials should consult the Open Science Framework (OSF) repository or contact the corresponding author:
  
  Brodie C. Dakin, Melbourne School of Psychological Sciences, University of Melbourne, Parkville, Victoria, 3010, Australia. Email: [email protected].
  
  OSF Project Repository: https://osf.io/svtgp/

References

• Dakin, B. C., McGrath, M. J., Rhee, J. J., & Haslam, N. (2023). Broadened concepts of harm appear less serious. Social Psychological and Personality Science, 14(1), 72–83. https://doi.org/10.1177/19485506221076692
• Einarsen, S., Hoel, H., Zapf, D., & Cooper, C. L. (Eds.). (2011). Bullying and harassment in the workplace: Developments in theory, research, and practice (2nd ed.). CRC Press. https://doi.org/10.1201/EBK1439804896
• Gray, K., Young, L., & Waytz, A. (2012). Mind perception is the essence of moral judgment. Psychological Inquiry, 23(2), 101–124. https://doi.org/10.1080/1047840X.2012.651387
• Haslam, N. (2016). Concept creep: Psychology’s expanding concepts of harm and pathology. Psychological Inquiry, 27(1), 1–17. https://doi.org/10.1080/1047840X.2016.1082418
• Lakoff, G. (1987). Women, fire, and dangerous things: What categories reveal about the mind. University of Chicago Press. https://doi.org/10.7208/chicago/9780226471013.001.0001
• Leymann, H. (1996). The content and development of mobbing at work. European Journal of Work and Organizational Psychology, 5(2), 165–184. https://doi.org/10.1080/13594329608414853
• Rosch, E. (1975). Cognitive representations of semantic categories. Journal of Experimental Psychology: General, 104(3), 192–233. https://doi.org/10.1037/0096-3445.104.3.192

13. Items of the Scale (Questionnaire)