Beyond General Awareness: A Global Empirical Framework for Behavior-Specific Policy Development Against Academic Bullying

Beyond General Awareness: A Global Empirical Framework for Behavior-Specific Policy Development Against Academic Bullying | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Beyond General Awareness: A Global Empirical Framework for Behavior-Specific Policy Development Against Academic Bullying Sherry Moss, Hamidreza Modares, Morteza Mahmoudi This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9269024/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Academic bullying and harassment are historically treated as idiosyncratic interpersonal conflicts, leading to generalized institutional policies that often fail to address the nuances of supervisory abuse. Using a global dataset (N = 2,041), we analyzed 15 Tepper Scale abusive behaviors and 10 contextual academic bullying behaviors in relation to perpetrator demographics and hierarchical positions. To account for cross-behavior dependence, we employed an end-to-end machine learning pipeline utilizing a multiclass classifier chain. Our findings reveal that the "one-size-fits-all" approach to harassment policy is empirically inadequate and pragmatically ineffective. We identified high-risk "behavioral clusters" tied to specific roles, such as Department Chairs and Group Leaders, who utilize distinct bullying tactics—ranging from visa cancellation threats to intellectual property (IP) theft. We propose a Researcher Bill of Rights (RBoR) as a blueprint for dismantling this structural ecology. Social science/Criminology Biological sciences/Psychology Social science/Psychology Social science/Science technology and society Introduction Academic bullying is frequently characterized in institutional discourse as a "personality clash" or a "lapse in professional etiquette." 1, 2 This framing, however, is empirically reductive. It ignores the structural reality that the research enterprise is built upon extreme power asymmetries and a "star system" that often prioritizes grant acquisition and publication over ethical management. 3 This cultural fixation on "prestige metrics" creates a fertile ground for abusive supervision, as institutional survival often depends on the continued success of high-output labs, regardless of their internal climate. When scientific merit is decoupled from professional conduct, the resulting vacuum is filled by tactical exploitation. Existing anti-harassment policies—often vague and reactive—fail to address the specific, localized, and hierarchical ways power is weaponized. 4 – 7 To protect the integrity of the scientific workforce, we must shift the focus from general awareness to specific policy development. 8 The "Master-Apprentice" model of research training, while intended to facilitate deep mentorship, inherently lacks the checks and balances found in corporate or governmental sectors. 9 In this ecosystem, a single individual—the Principal Investigator (PI) or Department Chair—often holds absolute authority over a subordinate’s funding, visa status, publication record, and career trajectory. This concentration of power is unique to academia; unlike in other professional sectors, researchers cannot simply "transfer" their work-in-progress to a different supervisor without significant loss of time and intellectual property. This "sunk cost" creates a hostage-like dynamic where targets endure abuse to avoid total career collapse. This "institutional insulation" allows abusive behaviors to manifest not as singular outbursts, but as sustained, strategic control tactics. 10 Furthermore, the lack of external oversight means that the PI acts as the primary career gatekeeper - investigator, judge, and jury of their own laboratory culture. To bridge this gap, our study utilizes the Tepper Scale 11 , 12 and a set of 10 contextual behaviors 6 —such as visa threats, funding removal, and authorship violations—to map the perpetrator’s profile across a global sample of N = 2,041. Details regarding the survey, including survey questions and outcomes, are available in Supporting Information (SI) . By decomposing "bullying" into specific actions, we identify how variables including academic position, geography, and age range intersect to create specific risk profiles. The objective of this paper is to provide the empirical foundation for a Researcher Bill of Rights (RBoR). Our findings demonstrate that bullying styles are not uniform; for instance, the risk of "credit-withholding" is significantly higher in specific European contexts, while "displaced anger" and "public ridicule" are more prevalent in North American labs. By identifying these clusters, we argue that institutional policy must evolve from broad "Code of Conduct" statements to targeted, behavior-based interventions. Only by codifying these protections, and developing targeted interventions, can we dismantle the structural ecology of abuse and ensure that scientific excellence is not built upon human suffering. Methods Survey design and ethical considerations. The Institutional Review Boards (IRB) of Wake Forest University and Michigan State University approved the survey and participant consent procedures, and the declaration of informed consent to conduct this study. Great care was taken to ensure the anonymity of respondents, as the fear of retaliation remains the primary barrier to reporting in academic settings. All data were encrypted and stored on secure servers, with identifying metadata removed to protect the privacy of those in vulnerable positions. Data were collected from 2,041 individuals recruited through targeted advertisements in Science and Nature—including an advertorial piece and third-party emails—as well as the American Chemical Society’s online panel. The sample was predominantly female (65%) and White (66.5%), followed by Asian (11.8%), Middle Eastern (7%), Hispanic (6%), and other ethnicities (under 3% each). At the time of the reported bullying, 60% of respondents resided in the same country where they worked or studied. Professionally, the cohort consisted primarily of postdocs (22.8%) and graduate students (21.6%), with the remainder comprising junior faculty (17%), senior faculty (13%), and other staff or professionals (21%). Participants represented a broad range of disciplines, led by the life sciences (19%) and social sciences (13.8%). Other significant fields included chemistry (8.8%), engineering (8.7%), molecular biology (7.6%), neuroscience (7.4%), and physical sciences (5.4%). Additional representations (ranging from 1.9% to 3.4%) included biotech, clinical science, genetics, cancer research, immunology, earth sciences, and mathematics. Measures and data collection. Upon providing consent, respondents were presented with a formal definition of academic bullying: sustained hostile behavior from an academic superior—such as ridicule, threats, or privacy invasion—as well as interference with career milestones (e.g., removing funding or writing falsely negative recommendation letters). Targets were identified as those who answered "yes" to having experienced such behavior. These targets then provided demographic details about the perpetrator before reporting specific behaviors. Interpersonal and sustained hostile behaviors were assessed using a 15-item abusive supervision scale known as the Tepper scale. 15 Participants rated items like "my supervisor ridicules me" on a 5-point Likert scale, ranging from 1 ("I cannot remember him/her ever using this behavior") to 5 ("He/she uses this behavior very often"). Items were adapted for witnesses (e.g., "the perpetrator ridicules others"). While this scale was developed for general use – primarily in business organizations – and items are typically averaged to formulate an “abusive supervision” score, we disaggregated the items because they represent distinct tactics. Rather than using the scale, we analyzed patterns for each distinct item. Context-specific bullying behaviors were measured using a 10-item checklist developed for a global study of bullying in academic science. 6 While the Tepper items measure supervisory abuse in general, the context-specific items were developed, based on personal accounts of specific abusive tactics used by leaders in academic science. A comparison of the Tepper scale items and the context-specific items is included in the Supporting Information. To measure the characteristics of perpetrators of interpersonal hostility and context specific behaviors, study participants indicated the gender, racioethnicity, nationality, age and position (Principal Investigator, Lab Supervisor, Group Leader, Department Chair, Senior Colleague, or “other”) of their perpetrators. These characteristics, along with the two forms of bullying (interpersonal hostility and contextual behaviors) were then subjected to machine learning to detect patterns, as described below. Machine learning. The study employs an analytical framework designed to extract actionable, demographic-specific rules for predicting academic bullying and contextual sabotage by addressing the unique sociological and mathematical challenges of the dataset (see the SI for full details on the machine learning framework). To capture the nuances of sociological intersectionality, the independent feature space—comprising age, ethnicity, gender, country, and academic position—is transformed into a sparse binary matrix via one-hot encoding. This allows the models to move beyond isolated, additive variables and natively learn multi-dimensional categorical interactions, such as the joint probability of a specific age, gender, and academic position. This approach is grounded in the reality that behavioral drivers for a specific role, such as a "Postdoc," may shift fundamentally depending on geographic and age-related contexts. A primary obstacle in the analysis is the severe class imbalance and the resulting "accuracy paradox," where critical bullying behaviors like visa or position cancellation are present in only roughly 9% to 11% of cases. In such "hostile" mathematical environments, standard machine learning models tend to prioritize global accuracy by predicting the majority "No" class for every respondent, effectively treating rare "Yes" signals as statistical noise. To counteract this, the methodology utilizes cost-sensitive learning and re-sampling strategies, including Random Under-Sampling (RUS) to create balanced training distributions, exponential weight updates to penalize misclassified "Yes" instances, and high bipolar margin penalties in SVM classifiers to force the geometric boundary toward rare positive signals. Given that behaviors are highly correlated and non-mutually exclusive, the data is partitioned into two distinct analytical frameworks based on target dimensionality. For the 15-behavior set, which involves five ordinal or nominal classes per behavior, the study utilizes an independent One-vs-All (OVA) decomposition strategy with Support Vector Machines (SVM). This framework strictly employs a linear kernel, leveraging Cover’s Theorem to ensure linear separability in the high-dimensional sparse feature space while avoiding the overfitting risks associated with non-linear kernels. Final class assignments are determined through a max-margin decision rule. For the 10-behavior binary set, which exhibits more severe imbalance and strong pairwise correlations, the study employs an Ensemble of Classifier Chains (ECC). This method sequentially concatenates demographic features with the predicted outcomes of preceding behaviors in the chain, mathematically forcing the model to internalize conditional dependencies. These chains utilize RUSBoost mechanics, where base learners are Classification and Regression Trees (CART) that prioritize splits based on balanced subsets to prevent the majority class from dominating the Gini Impurity calculation. The model’s performance is evaluated through a sociological lens, where a global testing accuracy of 53% is considered highly significant because human decisions are influenced by massive unobserved latent variables. Notably, identifying a sub-population with a 55% probability of exhibiting a behavior—compared to a 5% baseline—represents an 11-fold increase in predictive power. To ensure the ECC successfully learned behavioral dependencies, the association between variables was verified using a Cramer’s V heatmap comparison. Finally, the framework concludes with an exhaustive extraction of demographic rules using strict empirical guardrails: a minimum support of at least 100 individuals per profile and a probability threshold of at least 50% for predicting a "Yes" response. Results The empirical mapping of academic bullying through predictive modeling reveals two distinct streams of abusive supervision (Table 1 ): Interpersonal Hostility (measured via the Tepper Scale) and Contextual Sabotage (measured via specific administrative actions). The model demonstrates that these are not overlapping inconveniences but structured outcomes of institutional hierarchy, with an overall test accuracy of 56.48% for contextual behaviors and an accuracy of 82.9% for the Tepper tactics. The high accuracy for Tepper tactics suggests that psychological abuse follows highly predictable demographic patterns, while contextual sabotage is more closely tied to specific administrative levers available to certain roles. Table 1 Details of behaviors and actions in Tepper scale and contextual-specific behaviors. Tepper Scale Items (The perpetrator….) Context-Specific Behaviors Ridicules me Gave me a bad/unfair recommendation Tells me my thoughts or feelings are stupid Cancelled or threatened to cancel my visa Puts me down in front of others Unnecessarily lengthened my stay in his/her lab Blames me to save him/herself from embarrassment Took away my funding or threatened to take away my funding Invades my privacy Encouraged others to mistreat me Doesn’t give me credit for jobs requiring a lot of effort Used my data in papers/patents without acknowledging my contribution Reminds me of my past failures or mistakes Violated authorship guidelines Tells me I’m incompetent Forced me to sign away my rights Expresses anger at me when he/she is mad for another reason Violated my intellectual property rights Gives me the silent treatment Cancelled or threatened to cancel my current appointment/position Does not allow me to interact with my coworkers Doesn’t give me credit for my work Lies to me Breaks promises he/she makes Makes negative comments about me to others Interpersonal Hostility (Tepper Scale) . As detailed in Table 2 , interpersonal abuse is highly concentrated in senior leadership roles. Department Chairs and Group Leaders, as compared to other surveyed positions (i.e., Principal Investigators, Lab Supervisors, Senior Colleagues, Heads of a Public Research Institute or Observatory, and Heads of Private company research group or lab), exhibit the highest probabilities for verbal and psychological hostility. Eleven of the 15 Tepper scale items were associated with specific demographic clusters. For example, Department Chairs show a 72.0% probability for public ridicule and a 74.9% probability for making negative comments to others about a target. A significant finding in this stream is the "Silent Treatment" and "Privacy Invasion," which peak among senior administrators in the USA and UK. Intersectional data suggests that Asian Department Chairs and Group Leaders (n = 155) are likely to tell followers that they are 'incompetent' (mean prediction of 4.17), particularly in US-based labs. Contextual Sabotage (Administrative Actions). Beyond interpersonal friction, analysis of the 10 contextual behaviors exposes a "Career Sabotage" cluster where administrative power is weaponized. As detailed in Table 2 , eight of the 10 contextual behaviors were associated with different types of leaders. For example, Group Leaders pose the highest risk for providing Unfair Recommendations (62.3% probability) and violating IP Rights (60.9%). Principal Investigators (PIs) are the primary drivers of financial and legal coercion, with a 62.6% probability of threatening funding removal. The most severe form of legal leverage—Visa Cancellation—reaches an accuracy of 67.03% and is most prevalent among Senior Male PIs in the USA. Additionally, a striking "Administrative Malpractice" cluster was identified among White Female PIs (Age 46–55), who demonstrated study-high probabilities for Authorship Violations (68.0%) and Forced Rights Waivers (76.1%). This suggests that different cohorts utilize the specific levers of power most accessible to their stage of career; while senior males may rely on structural/legal threats, mid-career cohorts may focus on "intellectual accumulation" to solidify their standing. It is noteworthy that the title of Group Leader is more prevalent and formalized in European organizational structures than in North America. In countries like Germany, Austria, and Switzerland, it represents a standard, specific tier of middle management—situated between a Team Lead and a Department Head—that often carries formal disciplinary authority over multiple teams. This differs from the United States, where the term is less standardized and is frequently replaced by titles like "Senior Manager" or "Director" in corporate settings, or reserved for frontline supervisory roles in manufacturing. While the title is a global standard within scientific research and academia, its high visibility in the European private sector is largely due to more granular, traditional hierarchies and labor regulations that require clearly defined management classifications. Table 2 Integrated risk matrix: interpersonal hostility and contextual sabotage Behavior Description Primary Risk Factor N Probability (%) Mean Prediction Key Perpetrator and/or Demographic Context Tepper Scale Ridicules me Department Chair 486 72.00% 3.7 Males in USA Reminds me of failures Group Leader 268 64.20% 3.88 High in Females Thoughts are stupid Lab Supervisor 124 58.80% 3.48 Age 56–65 Tells me I’m incompetent Asian (Ethnicity) 155 65.20% 4.17 US-based labs Displaced anger Lab Supervisor 124 68.10% 3.89 Female PIs Negative comments Department Chair 486 74.90% 4.39 Female Chairs Puts me down public Department Chair 486 68.30% 3.83 Senior Males (56–65) Silent treatment UK 120 58.10% 3.59 Senior Female cohorts No credit for work Group Leader 268 73.30% 4.19 Males (Age 46–55) Invades privacy Department Chair 486 72.50% 3.86 US-based Chairs Lies to me Department Chair 486 76.20% 4.19 Group Leaders in USA Contextual Behaviors Unfair Recommendation Group Leader 268 62.30% 1.75 Senior White Male PI Visa Cancellation PI 1097 51.10% 1.58 Senior Male PI in USA Lengthened Stay in Lab Lab Supervisor 124 56.40% 1.41 Asian PIs Funding Removal PI 1097 62.60% 1.83 Senior Male PIs Encourage Mistreatment Department Chair 486 68.40% 1.98 White Male Department Chair Authorship Violation PI 1097 61.40% 1.79 White Female PI (Age 46–55) Sign away rights Group Leader 268 71.40% 1.88 White Female PI (Age 46–55) IP Violation Group Leader 268 60.90% 1.81 White Female PI (Age 46–55) Discussion The empirical evidence presented in Table 2 exposes a profound "Policy-Evidence Gap". By separating interpersonal hostility from contextual sabotage, we observe that current institutional policies are fundamentally ill-equipped to handle the dual nature of academic abuse. Most institutional responses treat bullying as a psychological issue requiring "mediation" or "conflict resolution." Our data shows that bullying is often a rational—albeit unethical—strategy for negatively impacting career advancement and resource allocation. Addressing Interpersonal Hostility. The high reported incidence of Lying (76.2%) and Privacy Invasion (72.5%) among Department Chairs indicates that existing internal grievance procedures may be structurally inadequate rather than merely underutilized. When the individual charged with upholding departmental norms and ethical standards is also perceived as the primary source of misconduct, the integrity of the reporting hierarchy is fundamentally compromised. In such circumstances, formal channels for redress become functionally inaccessible, as subordinates may reasonably anticipate retaliation, dismissal of claims, or reputational harm. This dynamic effectively produces an accountability “dead zone,” within which patterns of verbal abuse, information manipulation, and psychologically exclusionary practices such as the “Silent Treatment” can persist with minimal oversight or intervention. Moreover, the normalization of these behaviors at the leadership level risks diffusing downward through the departmental climate, reinforcing a culture in which incivility is tacitly accepted or strategically ignored. Addressing these systemic vulnerabilities requires more than incremental policy adjustments; it necessitates the implementation of robust, multi-source evaluation systems, including 360-degree leadership audits that incorporate confidential upward feedback and third-party review. Complementary bias-aware reporting mechanisms are also critical, particularly in light of the elevated levels of incompetence-based targeting observed in Asian demographic contexts, which may reflect both implicit bias and culturally mediated misinterpretations of communication styles. Collectively, these interventions can help reestablish credible oversight, mitigate power asymmetries, and create safer pathways for reporting and accountability. Dismantling Contextual Sabotage. The heterogeneous forms of sabotage identified in our data necessitate correspondingly differentiated and targeted interventions. Broad, principle-based instruments such as generic “codes of conduct” are insufficient to address behaviors that are structurally enabled and materially consequential. For instance, such codes offer little deterrent against a PI who leverages institutional authority to threaten visa status or appropriate intellectual contributions through authorship theft. These patterns of misconduct are not merely interpersonal but are embedded in administrative and governance arrangements, and thus require formal, enforceable policy reforms rather than aspirational norms. In this regard, the elevated risk of credit misappropriation observed among Group Leaders underscores the need to institutionalize a “Right to Digital Attribution.” This concept reflects both a legal and ethical entitlement for researchers to be recognized as the originators of their work across digital platforms and collaborative environments. Establishing formalized attribution protocols—such as immutable contribution records, timestamped authorship claims, and transparent version control systems—would help preserve the linkage between individuals and their intellectual output. Such mechanisms create an auditable trail that not only deters misconduct but also provides early-career researchers with a measure of protection against reputational and professional harm stemming from credit theft. Additionally, the strong predictive relationship between Visa Threats and institutional context within the United States highlights a critical structural vulnerability. Specifically, it points to the need to decouple a researcher’s legal residency status from the discretionary authority of an individual PI. As long as a single supervisor retains the capacity to terminate a researcher’s legal standing—potentially with minimal procedural oversight—the power imbalance is exacerbated to a degree that undermines the foundational premise of mentorship. Under these conditions, the relationship is more accurately characterized by dependency and coercion than by professional development. Addressing this issue may require policy innovations such as institutionally held visa sponsorship, multi-party oversight of employment termination decisions, or grace periods that allow researchers to transition between positions without immediate legal jeopardy. Finally, the intersectional patterns evident in our findings point to the necessity of culturally competent and context-sensitive interventions. The disproportionately high levels of incompetence-based targeting reported in Asian demographic contexts suggest that bullying behaviors are often intertwined with stereotyping, implicit bias, and, in some cases, overt xenophobia. Effective mitigation strategies must therefore extend beyond formal policy changes to include bias training, inclusive leadership development, and accountability mechanisms that explicitly address how cultural misperceptions and power asymmetries shape the expression and impact of workplace hostility. Collectively, these approaches are essential for dismantling the structural and cultural conditions that enable contextual sabotage to persist. Policy Implementation: Operationalizing the Researcher Bill of Rights (RBoR). To meaningfully translate empirical insights into durable institutional reform, the Researcher Bill of Rights (RBoR) must be operationalized not as a symbolic framework, but as an integrated system of automated and structural “circuit breakers.” These mechanisms should be designed to activate in response to empirically identified high-risk behaviors, thereby reducing reliance on discretionary enforcement and minimizing opportunities for suppression or retaliation. In light of the observed 76.2% risk of administrative deception and the 71.4% likelihood of forced rights waivers, implementation must prioritize ex ante safeguards that interrupt harmful conduct before it escalates into entrenched patterns of abuse. This approach reframes governance from reactive adjudication to proactive risk mitigation, embedding accountability directly into institutional processes and decision architectures. Neutralizing Administrative Hostility. Given that Department Chairs exhibit a 76.2% probability of administrative deception and a 72.5% probability of privacy invasion, existing internal grievance systems—typically routed through hierarchical reporting lines—are structurally ill-equipped to ensure impartial review. When those tasked with oversight are themselves implicated in misconduct, procedural legitimacy erodes and reporting becomes both risky and ineffective. To address this, institutions should establish an Ombudsperson Matrix, wherein reporting channels are deliberately decoupled from departmental authority and redirected to independent, extra-departmental ethics bodies with investigatory autonomy. Such bodies should be empowered to initiate inquiries without prior departmental consent, thereby restoring credibility to the reporting process. Complementing this, all high-stakes administrative decisions—particularly those involving laboratory space allocation, funding distribution, and contract or appointment renewals—should be governed by robust audit infrastructures. Specifically, a “Dual-Authorization” digital trail should be required, ensuring that no single administrator can unilaterally enact decisions with material consequences for a researcher’s career trajectory. These audit trails should be time-stamped, immutable, and subject to periodic external review, thereby constraining the “Breaking of Promises” (75.5%) identified in our model and enhancing procedural transparency across administrative domains. Protecting Intellectual Integrity and Legal Residency. The high predictive accuracy associated with Visa Threats (67.03%) and intellectual property (IP) violations (60.9%) highlights the extent to which researchers’ professional and personal security can be contingent upon the discretionary authority of individual supervisors. This interdependence creates a structurally coercive environment in which compliance may be driven less by scholarly norms than by fear of material consequences. To mitigate this, institutions must systematically decouple core elements of researcher identity—namely, legal residency and intellectual ownership—from supervisory control. One critical intervention is the establishment of “Visa Safe Harbors,” whereby visa sponsorship is administered centrally through the university’s International Office rather than delegated to individual Principal Investigators. This redistribution of authority directly attenuates the “Legal Leverage” identified as a key mechanism of control among Senior Male PIs (54.7%), and introduces institutional buffering that allows researchers to report misconduct or transition between positions without immediate risk to their legal status. Additional safeguards, such as mandatory transition periods or institutional sponsorship continuity clauses, could further reduce vulnerability during periods of professional disruption. Parallel protections are required to safeguard intellectual contributions. To address the 76.1% risk of forced rights waivers among mid-career researchers, journals and funding agencies should mandate standardized authorship disclosure frameworks, such as the CRediT (Contributor Roles Taxonomy). Embedding these frameworks within submission and grant-reporting systems creates a verifiable, multi-party record of contribution that cannot be unilaterally altered. Such “Verified Attribution” systems function as both deterrent and recourse mechanism, preventing the retroactive erasure of credit (73.3%) and ensuring that scholarly contributions remain transparently and durably linked to their originators. Targeted Leadership Intervention. The findings further suggest that uniform, “one-size-fits-all” training interventions are insufficient to address the heterogeneity of risk profiles across supervisory populations. Instead, leadership interventions should be calibrated to reflect demographic, structural, and contextual variations in observed behavior patterns. While all individuals in supervisory roles should be subject to routine Lab Climate Audits, enhanced scrutiny may be warranted in environments led by individuals or cohorts associated with elevated risk indicators. These audits should integrate both quantitative climate metrics and qualitative feedback, enabling a more nuanced assessment of leadership effectiveness and interpersonal conduct. In addition, the implementation of 360-degree feedback systems is essential for rebalancing power dynamics within academic hierarchies. Under such systems, supervisory privileges—including personnel management authority and eligibility to serve as principal investigator on grants—would be contingent upon demonstrably positive evaluations from subordinates and peers. This effectively reconfigures authority as contingent and performance-based, rather than as an entrenched entitlement. When coupled with transparent reporting thresholds and consequences for persistently negative evaluations, these systems can incentivize prosocial leadership behaviors while systematically identifying and remediating dysfunctional supervisory practices. Collectively, these implementation strategies reposition the RBoR as an enforceable governance framework rather than an aspirational document. By embedding accountability into institutional infrastructure—through independent oversight, auditable decision-making, and structurally decoupled power mechanisms—organizations can more effectively disrupt the conditions under which academic bullying and exploitation are able to persist. Limitations and Future Directions. It is important to recognize several limitations associated with this study. Most notably, our examination of bullying was constrained to individuals occupying higher-status roles, as participants were prompted to consider interactions with “academic superiors.” Consequently, additional empirical work is needed to capture bullying enacted by peers and other members of the academic community, including collective forms such as mobbing. We also note the potential for sampling bias, as individuals who had directly experienced or observed bullying may have been more inclined to participate than those without such experiences. In addition, a substantial proportion of respondents were based in the United States, which may limit the generalizability of the findings across different national and institutional contexts. Finally, aspects of the survey design may have restricted the range of insights obtained. Participants who initially reported no experience with or exposure to academic bullying were routed away from key measures, including the Tepper abusive supervision scale and the contextual behavior checklist. It is possible that some of these individuals would have endorsed exposure to specific behaviors captured by these instruments had they been given the opportunity. Notwithstanding these limitations, the study contributes a more fine-grained understanding of the behaviors most commonly encountered by targets of academic bullying, along with their typical responses and the consequences that follow. Conclusion This study provides evidence that academic bullying is not an incidental interpersonal phenomenon, but rather a structural outcome of an institutional ecology that allocates substantial discretionary power to supervisory actors without commensurate mechanisms of oversight, constraint, or accountability. Within this configuration, abusive supervision and related forms of academic bullying emerge not randomly, but in discernible and statistically patterned clusters that map onto hierarchical positions and role-based authority gradients. In particular, individuals occupying formal leadership roles are positioned at structural nodes where decision rights, resource control, and evaluative authority converge, thereby increasing both the opportunity and the efficacy of coercive or exploitative behaviors when safeguards are absent. Accordingly, the shift from merely documenting these behaviors to implementing infrastructural accountability mechanisms is no longer a discretionary matter of institutional preference or cultural aspiration; it is an empirically grounded necessity implied by the observed distribution and predictability of misconduct. The model presented in this study indicates that, in the absence of structural intervention, specific cohorts will continue to deploy identifiable and repeatable levers of academic sabotage, including threats to legal status, manipulation of authorship credit, and control over material research resources. These mechanisms persist precisely because they are embedded within governance structures that permit unilateral or minimally constrained decision-making. In response, the codification of the Researcher Bill of Rights represents a transition from a governance paradigm characterized as “mentorship by permission”—in which access, opportunity, and professional survival are contingent on discretionary approval—to one grounded in “mentorship by accountability,” in which supervisory authority is explicitly bounded by enforceable rights, transparent processes, and independent oversight. This reframing is intended to reduce asymmetries of power that enable coercion while reinforcing the principle that scientific training should be governed by standards of fairness rather than personal discretion. Operationally, this transition requires a set of interlocking institutional reforms. These include decentralized and externally insulated oversight structures for administrative decision-making roles, ensuring that evaluative authority is not concentrated within single supervisory units. They also include the implementation of verified digital credit systems that preserve the integrity of intellectual attribution across collaborative environments, thereby reducing opportunities for authorship manipulation or retrospective credit reallocation. In addition, jurisdictional protections are necessary to prevent the instrumentalization of visa or residency status as a mechanism of compliance enforcement or professional coercion. Taken together, these reforms reflect a broader normative and empirical conclusion: the integrity of scientific systems cannot be meaningfully separated from the dignity, autonomy, and security of the individuals who operate within them. Sustained scientific excellence is therefore contingent not only on intellectual rigor, but also on the establishment of institutional environments in which power is systematically checked, accountability is structurally enforced, and participation is not conditioned on vulnerability. Declarations Competing Interests M. M. is co-founder and director of the Academic Parity Movement (www.paritymovement.org) (a non-profit organization dedicated to addressing academic discrimination, violence and incivility). S.M. is director of the Academic Parity Movement. Author Contribution S.M. and M.M. designed and ran the survey; S.M., H.M. and M.M. analyzed the results. S.M., H.M. and M.M. wrote the main manuscript. All authors reviewed the manuscript. Data Availability Survey data and the developed code for data analysis are available in the following link: https://github.com/Morimahmoudi/AcademicBullying.git References Zabrodska, K.; Kveton, P., Prevalence and forms of workplace bullying among university employees. Employee Responsibilities and Rights Journal 2013, 25 (2), 89-108. Averbuch, T.; Eliya, Y.; Van Spall, H. G. C., Systematic review of academic bullying in medical settings: dynamics and consequences. BMJ open 2021, 11 (7), e043256. Mahmoudi, M., Academic bullying slows the evolution of science. Nature Reviews Materials 2023, 8 (5), 301-303. Täuber, S.; Loyens, K.; Oertelt-Prigione, S.; Kubbe, I., Harassment as a consequence and cause of inequality in academia: A narrative review. EClinicalMedicine 2022, 49 . Moss, S.; Täuber, S.; Sharifi, S.; Mahmoudi, M., The need for the development of discipline-specific approaches to address academic bullying. eClinicalMedicine 2022, 50 , 101598. Moss, S. E.; Mahmoudi, M., STEM the bullying: An empirical investigation of abusive supervision in academic science. eClinicalMedicine 2021, 40 , 101121. Moss, S. E.; Mahmoudi, M., Examining Age-Dependent Patterns in Academic Bullying Behaviors. Journal of Academic Ethics 2025, 24 (1), 25. Mahmoudi, M.; Keashly, L., Filling the space: a framework for coordinated global actions to diminish academic bullying. Angewandte Chemie 2021, 133 (7), 3378-3384. Manathunga, C., Supervision as mentoring: The role of power and boundary crossing. Studies in Continuing education 2007, 29 (2), 207-221. Vogelaar, A. E.; Mahmoudi, M., Chronic silencing is a critical barrier to breaking the cycle of bullying in academia and industry. Nature Biotechnology 2025, 43 (9), 1577-1579. Tepper, B. J., Abusive Supervision Scale. Journal of Applied Psychology 2000 . Tepper, B. J., Consequences of Abusive Supervision. Academy of Management Journal 2000, 43 (2), 178-190. Additional Declarations Competing interest reported. M. M. is co-founder and director of the Academic Parity Movement ( www.paritymovement.org ) (a non-profit organization dedicated to addressing academic discrimination, violence and incivility). S.M. is director of the Academic Parity Movement. Supplementary Files SIPeerReview.pdf Surveyquestions.docx Surveyoutcomes.xlsx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9269024","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":623260103,"identity":"569e076f-8634-40c2-b9b2-3ccecb68a1af","order_by":0,"name":"Sherry Moss","email":"","orcid":"","institution":"Wake Forest University","correspondingAuthor":false,"prefix":"","firstName":"Sherry","middleName":"","lastName":"Moss","suffix":""},{"id":623260104,"identity":"e4b5cd35-2a57-4eb8-bd3f-64494775d863","order_by":1,"name":"Hamidreza Modares","email":"","orcid":"","institution":"Michigan State University","correspondingAuthor":false,"prefix":"","firstName":"Hamidreza","middleName":"","lastName":"Modares","suffix":""},{"id":623260105,"identity":"57b524df-4140-4708-b8d4-221a26dade4c","order_by":2,"name":"Morteza Mahmoudi","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA/UlEQVRIiWNgGAWjYHADHhDBzMMPohIK8CplbEDRIgniJhiQoIXB4ACIxqNFt733+YMPf+yAjLNHN3xss5YxPr868cMDAwZ5frEDWLWYnTlu2DiDJxnIyEu7ObMtncfsxtvNEkCHGc6cnYBdy400xmYeCWYGswM5Zrd52w4DtZzdANKSYHAbj5Y/BvUMZuffQLQYzzi7+QdBLQwJh4EMqC0G/L3b8Nty5hjjzJ4Dx4HueWN2c8a5dB6JG7zbLBIMJHD75Xgbw4cff6rlzM7nmN34UGZtz99/dvPNHxU28vzS2LXAAA+CKQFWKYFXORrgP0CK6lEwCkbBKBgBAABnWGG7uQgKRQAAAABJRU5ErkJggg==","orcid":"","institution":"Michigan State University","correspondingAuthor":true,"prefix":"","firstName":"Morteza","middleName":"","lastName":"Mahmoudi","suffix":""}],"badges":[],"createdAt":"2026-03-30 15:23:25","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9269024/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9269024/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":107480657,"identity":"1565a04d-4e27-458a-be7f-2d989143832b","added_by":"auto","created_at":"2026-04-22 02:12:58","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":532017,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9269024/v1/f3ca5c94-67fb-45b0-9a99-80d93288bbc7.pdf"},{"id":107033956,"identity":"ac3fdcec-45e6-4484-96b7-2615ffc7891d","added_by":"auto","created_at":"2026-04-16 03:50:08","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":191300,"visible":true,"origin":"","legend":"","description":"","filename":"SIPeerReview.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9269024/v1/d91fbc010b5ff9377b03e132.pdf"},{"id":107033959,"identity":"2af1e1c1-5ce8-4b1e-8443-e54d4d81c741","added_by":"auto","created_at":"2026-04-16 03:50:08","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":40387,"visible":true,"origin":"","legend":"","description":"","filename":"Surveyquestions.docx","url":"https://assets-eu.researchsquare.com/files/rs-9269024/v1/575db98b44661d2352a57180.docx"},{"id":107033957,"identity":"9aedabdb-7a7e-498e-9fcb-a5bc61774c6c","added_by":"auto","created_at":"2026-04-16 03:50:08","extension":"xlsx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":297472,"visible":true,"origin":"","legend":"","description":"","filename":"Surveyoutcomes.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-9269024/v1/271b52e034c6fa2bb7ee130d.xlsx"}],"financialInterests":"Competing interest reported. M. M. is co-founder and director of the Academic Parity Movement (www.paritymovement.org) (a non-profit organization dedicated to addressing academic discrimination, violence and incivility). S.M. is director of the Academic Parity Movement.","formattedTitle":"Beyond General Awareness: A Global Empirical Framework for Behavior-Specific Policy Development Against Academic Bullying","fulltext":[{"header":"Introduction","content":"\u003cp\u003eAcademic bullying is frequently characterized in institutional discourse as a \"personality clash\" or a \"lapse in professional etiquette.\"\u003csup\u003e1, 2\u003c/sup\u003e This framing, however, is empirically reductive. It ignores the structural reality that the research enterprise is built upon extreme power asymmetries and a \"star system\" that often prioritizes grant acquisition and publication over ethical management.\u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u003c/sup\u003e This cultural fixation on \"prestige metrics\" creates a fertile ground for abusive supervision, as institutional survival often depends on the continued success of high-output labs, regardless of their internal climate. When scientific merit is decoupled from professional conduct, the resulting vacuum is filled by tactical exploitation.\u003c/p\u003e \u003cp\u003eExisting anti-harassment policies\u0026mdash;often vague and reactive\u0026mdash;fail to address the specific, localized, and hierarchical ways power is weaponized.\u003csup\u003e\u003cspan additionalcitationids=\"CR5 CR6\" citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u003c/sup\u003e To protect the integrity of the scientific workforce, we must shift the focus from general awareness to specific policy development.\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003cp\u003eThe \"Master-Apprentice\" model of research training, while intended to facilitate deep mentorship, inherently lacks the checks and balances found in corporate or governmental sectors.\u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e In this ecosystem, a single individual\u0026mdash;the Principal Investigator (PI) or Department Chair\u0026mdash;often holds absolute authority over a subordinate\u0026rsquo;s funding, visa status, publication record, and career trajectory. This concentration of power is unique to academia; unlike in other professional sectors, researchers cannot simply \"transfer\" their work-in-progress to a different supervisor without significant loss of time and intellectual property. This \"sunk cost\" creates a hostage-like dynamic where targets endure abuse to avoid total career collapse.\u003c/p\u003e \u003cp\u003eThis \"institutional insulation\" allows abusive behaviors to manifest not as singular outbursts, but as sustained, strategic control tactics.\u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e Furthermore, the lack of external oversight means that the PI acts as the primary career gatekeeper - investigator, judge, and jury of their own laboratory culture. To bridge this gap, our study utilizes the Tepper Scale\u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e and a set of 10 contextual behaviors\u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e \u0026mdash;such as visa threats, funding removal, and authorship violations\u0026mdash;to map the perpetrator\u0026rsquo;s profile across a global sample of N\u0026thinsp;=\u0026thinsp;2,041. Details regarding the survey, including survey questions and outcomes, are available in \u003cb\u003eSupporting Information (SI)\u003c/b\u003e. By decomposing \"bullying\" into specific actions, we identify how variables including academic position, geography, and age range intersect to create specific risk profiles.\u003c/p\u003e \u003cp\u003eThe objective of this paper is to provide the empirical foundation for a Researcher Bill of Rights (RBoR). Our findings demonstrate that bullying styles are not uniform; for instance, the risk of \"credit-withholding\" is significantly higher in specific European contexts, while \"displaced anger\" and \"public ridicule\" are more prevalent in North American labs. By identifying these clusters, we argue that institutional policy must evolve from broad \"Code of Conduct\" statements to targeted, behavior-based interventions. Only by codifying these protections, and developing targeted interventions, can we dismantle the structural ecology of abuse and ensure that scientific excellence is not built upon human suffering.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003e \u003cem\u003eSurvey design and ethical considerations.\u003c/em\u003e The Institutional Review Boards (IRB) of Wake Forest University and Michigan State University approved the survey and participant consent procedures, and the declaration of informed consent to conduct this study. Great care was taken to ensure the anonymity of respondents, as the fear of retaliation remains the primary barrier to reporting in academic settings. All data were encrypted and stored on secure servers, with identifying metadata removed to protect the privacy of those in vulnerable positions.\u003c/p\u003e \u003cp\u003eData were collected from 2,041 individuals recruited through targeted advertisements in Science and Nature\u0026mdash;including an advertorial piece and third-party emails\u0026mdash;as well as the American Chemical Society\u0026rsquo;s online panel. The sample was predominantly female (65%) and White (66.5%), followed by Asian (11.8%), Middle Eastern (7%), Hispanic (6%), and other ethnicities (under 3% each). At the time of the reported bullying, 60% of respondents resided in the same country where they worked or studied. Professionally, the cohort consisted primarily of postdocs (22.8%) and graduate students (21.6%), with the remainder comprising junior faculty (17%), senior faculty (13%), and other staff or professionals (21%). Participants represented a broad range of disciplines, led by the life sciences (19%) and social sciences (13.8%). Other significant fields included chemistry (8.8%), engineering (8.7%), molecular biology (7.6%), neuroscience (7.4%), and physical sciences (5.4%). Additional representations (ranging from 1.9% to 3.4%) included biotech, clinical science, genetics, cancer research, immunology, earth sciences, and mathematics.\u003c/p\u003e \u003cp\u003e \u003cem\u003eMeasures and data collection.\u003c/em\u003e Upon providing consent, respondents were presented with a formal definition of academic bullying: sustained hostile behavior from an academic superior\u0026mdash;such as ridicule, threats, or privacy invasion\u0026mdash;as well as interference with career milestones (e.g., removing funding or writing falsely negative recommendation letters). Targets were identified as those who answered \"yes\" to having experienced such behavior. These targets then provided demographic details about the perpetrator before reporting specific behaviors.\u003c/p\u003e \u003cp\u003eInterpersonal and sustained hostile behaviors were assessed using a 15-item abusive supervision scale known as the Tepper scale.\u003csup\u003e15\u003c/sup\u003e Participants rated items like \"my supervisor ridicules me\" on a 5-point Likert scale, ranging from 1 (\"I cannot remember him/her ever using this behavior\") to 5 (\"He/she uses this behavior very often\"). Items were adapted for witnesses (e.g., \"the perpetrator ridicules others\"). While this scale was developed for general use \u0026ndash; primarily in business organizations \u0026ndash; and items are typically averaged to formulate an \u0026ldquo;abusive supervision\u0026rdquo; score, we disaggregated the items because they represent distinct tactics. Rather than using the scale, we analyzed patterns for each distinct item.\u003c/p\u003e \u003cp\u003eContext-specific bullying behaviors were measured using a 10-item checklist developed for a global study of bullying in academic science.\u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e While the Tepper items measure supervisory abuse in general, the context-specific items were developed, based on personal accounts of specific abusive tactics used by leaders in academic science. A comparison of the Tepper scale items and the context-specific items is included in the Supporting Information. To measure the characteristics of perpetrators of interpersonal hostility and context specific behaviors, study participants indicated the gender, racioethnicity, nationality, age and position (Principal Investigator, Lab Supervisor, Group Leader, Department Chair, Senior Colleague, or \u0026ldquo;other\u0026rdquo;) of their perpetrators. These characteristics, along with the two forms of bullying (interpersonal hostility and contextual behaviors) were then subjected to machine learning to detect patterns, as described below.\u003c/p\u003e \u003cp\u003e \u003cem\u003eMachine learning.\u003c/em\u003e The study employs an analytical framework designed to extract actionable, demographic-specific rules for predicting academic bullying and contextual sabotage by addressing the unique sociological and mathematical challenges of the dataset (see the \u003cb\u003eSI\u003c/b\u003e for full details on the machine learning framework). To capture the nuances of sociological intersectionality, the independent feature space\u0026mdash;comprising age, ethnicity, gender, country, and academic position\u0026mdash;is transformed into a sparse binary matrix via one-hot encoding. This allows the models to move beyond isolated, additive variables and natively learn multi-dimensional categorical interactions, such as the joint probability of a specific age, gender, and academic position. This approach is grounded in the reality that behavioral drivers for a specific role, such as a \"Postdoc,\" may shift fundamentally depending on geographic and age-related contexts.\u003c/p\u003e \u003cp\u003eA primary obstacle in the analysis is the severe class imbalance and the resulting \"accuracy paradox,\" where critical bullying behaviors like visa or position cancellation are present in only roughly 9% to 11% of cases. In such \"hostile\" mathematical environments, standard machine learning models tend to prioritize global accuracy by predicting the majority \"No\" class for every respondent, effectively treating rare \"Yes\" signals as statistical noise. To counteract this, the methodology utilizes cost-sensitive learning and re-sampling strategies, including Random Under-Sampling (RUS) to create balanced training distributions, exponential weight updates to penalize misclassified \"Yes\" instances, and high bipolar margin penalties in SVM classifiers to force the geometric boundary toward rare positive signals.\u003c/p\u003e \u003cp\u003eGiven that behaviors are highly correlated and non-mutually exclusive, the data is partitioned into two distinct analytical frameworks based on target dimensionality. For the 15-behavior set, which involves five ordinal or nominal classes per behavior, the study utilizes an independent One-vs-All (OVA) decomposition strategy with Support Vector Machines (SVM). This framework strictly employs a linear kernel, leveraging Cover\u0026rsquo;s Theorem to ensure linear separability in the high-dimensional sparse feature space while avoiding the overfitting risks associated with non-linear kernels. Final class assignments are determined through a max-margin decision rule. For the 10-behavior binary set, which exhibits more severe imbalance and strong pairwise correlations, the study employs an Ensemble of Classifier Chains (ECC). This method sequentially concatenates demographic features with the predicted outcomes of preceding behaviors in the chain, mathematically forcing the model to internalize conditional dependencies. These chains utilize RUSBoost mechanics, where base learners are Classification and Regression Trees (CART) that prioritize splits based on balanced subsets to prevent the majority class from dominating the Gini Impurity calculation.\u003c/p\u003e \u003cp\u003eThe model\u0026rsquo;s performance is evaluated through a sociological lens, where a global testing accuracy of 53% is considered highly significant because human decisions are influenced by massive unobserved latent variables. Notably, identifying a sub-population with a 55% probability of exhibiting a behavior\u0026mdash;compared to a 5% baseline\u0026mdash;represents an 11-fold increase in predictive power. To ensure the ECC successfully learned behavioral dependencies, the association between variables was verified using a Cramer\u0026rsquo;s V heatmap comparison. Finally, the framework concludes with an exhaustive extraction of demographic rules using strict empirical guardrails: a minimum support of at least 100 individuals per profile and a probability threshold of at least 50% for predicting a \"Yes\" response.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003eThe empirical mapping of academic bullying through predictive modeling reveals two distinct streams of abusive supervision (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e): Interpersonal Hostility (measured via the Tepper Scale) and Contextual Sabotage (measured via specific administrative actions). The model demonstrates that these are not overlapping inconveniences but structured outcomes of institutional hierarchy, with an overall test accuracy of 56.48% for contextual behaviors and an accuracy of 82.9% for the Tepper tactics. The high accuracy for Tepper tactics suggests that psychological abuse follows highly predictable demographic patterns, while contextual sabotage is more closely tied to specific administrative levers available to certain roles.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eDetails of behaviors and actions in Tepper scale and contextual-specific behaviors.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"2\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTepper Scale Items (The perpetrator\u0026hellip;.)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eContext-Specific Behaviors\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRidicules me\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGave me a bad/unfair recommendation\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTells me my thoughts or feelings are stupid\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCancelled or threatened to cancel my visa\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePuts me down in front of others\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eUnnecessarily lengthened my stay in his/her lab\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBlames me to save him/herself from embarrassment\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTook away my funding or threatened to take away my funding\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eInvades my privacy\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eEncouraged others to mistreat me\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDoesn\u0026rsquo;t give me credit for jobs requiring a lot of effort\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eUsed my data in papers/patents without acknowledging my contribution\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eReminds me of my past failures or mistakes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eViolated authorship guidelines\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTells me I\u0026rsquo;m incompetent\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eForced me to sign away my rights\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eExpresses anger at me when he/she is mad for another reason\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eViolated my intellectual property rights\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGives me the silent treatment\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCancelled or threatened to cancel my current appointment/position\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDoes not allow me to interact with my coworkers\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDoesn\u0026rsquo;t give me credit for my work\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLies to me\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBreaks promises he/she makes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMakes negative comments about me to others\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e\u003cb\u003eInterpersonal Hostility (Tepper Scale)\u003c/b\u003e. As detailed in Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e, interpersonal abuse is highly concentrated in senior leadership roles. Department Chairs and Group Leaders, as compared to other surveyed positions (i.e., Principal Investigators, Lab Supervisors, Senior Colleagues, Heads of a Public Research Institute or Observatory, and Heads of Private company research group or lab), exhibit the highest probabilities for verbal and psychological hostility. Eleven of the 15 Tepper scale items were associated with specific demographic clusters. For example, Department Chairs show a 72.0% probability for public ridicule and a 74.9% probability for making negative comments to others about a target. A significant finding in this stream is the \"Silent Treatment\" and \"Privacy Invasion,\" which peak among senior administrators in the USA and UK. Intersectional data suggests that Asian Department Chairs and Group Leaders (n\u0026thinsp;=\u0026thinsp;155) are likely to tell followers that they are 'incompetent' (mean prediction of 4.17), particularly in US-based labs.\u003c/p\u003e \u003cp\u003e \u003cb\u003eContextual Sabotage (Administrative Actions).\u003c/b\u003e Beyond interpersonal friction, analysis of the 10 contextual behaviors exposes a \"Career Sabotage\" cluster where administrative power is weaponized. As detailed in Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e, eight of the 10 contextual behaviors were associated with different types of leaders. For example, Group Leaders pose the highest risk for providing Unfair Recommendations (62.3% probability) and violating IP Rights (60.9%). Principal Investigators (PIs) are the primary drivers of financial and legal coercion, with a 62.6% probability of threatening funding removal. The most severe form of legal leverage\u0026mdash;Visa Cancellation\u0026mdash;reaches an accuracy of 67.03% and is most prevalent among Senior Male PIs in the USA. Additionally, a striking \"Administrative Malpractice\" cluster was identified among White Female PIs (Age 46\u0026ndash;55), who demonstrated study-high probabilities for Authorship Violations (68.0%) and Forced Rights Waivers (76.1%). This suggests that different cohorts utilize the specific levers of power most accessible to their stage of career; while senior males may rely on structural/legal threats, mid-career cohorts may focus on \"intellectual accumulation\" to solidify their standing.\u003c/p\u003e \u003cp\u003eIt is noteworthy that the title of Group Leader is more prevalent and formalized in European organizational structures than in North America. In countries like Germany, Austria, and Switzerland, it represents a standard, specific tier of middle management\u0026mdash;situated between a Team Lead and a Department Head\u0026mdash;that often carries formal disciplinary authority over multiple teams. This differs from the United States, where the term is less standardized and is frequently replaced by titles like \"Senior Manager\" or \"Director\" in corporate settings, or reserved for frontline supervisory roles in manufacturing. While the title is a global standard within scientific research and academia, its high visibility in the European private sector is largely due to more granular, traditional hierarchies and labor regulations that require clearly defined management classifications.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eIntegrated risk matrix: interpersonal hostility and contextual sabotage\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"6\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBehavior Description\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePrimary Risk Factor\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eN\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eProbability (%)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eMean Prediction\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eKey Perpetrator and/or Demographic Context\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colspan=\"6\" nameend=\"c6\" namest=\"c1\"\u003e \u003cp\u003eTepper Scale\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRidicules me\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDepartment Chair\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e486\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e72.00%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e3.7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eMales in USA\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eReminds me of failures\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGroup Leader\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e268\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e64.20%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e3.88\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eHigh in Females\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eThoughts are stupid\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLab Supervisor\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e124\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e58.80%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e3.48\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAge 56\u0026ndash;65\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTells me I\u0026rsquo;m incompetent\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAsian (Ethnicity)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e155\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e65.20%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e4.17\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eUS-based labs\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDisplaced anger\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLab Supervisor\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e124\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e68.10%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e3.89\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eFemale PIs\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNegative comments\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDepartment Chair\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e486\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e74.90%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e4.39\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eFemale Chairs\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePuts me down public\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDepartment Chair\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e486\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e68.30%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e3.83\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eSenior Males (56\u0026ndash;65)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSilent treatment\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eUK\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e120\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e58.10%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e3.59\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eSenior Female cohorts\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo credit for work\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGroup Leader\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e268\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e73.30%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e4.19\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eMales (Age 46\u0026ndash;55)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eInvades privacy\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDepartment Chair\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e486\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e72.50%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e3.86\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eUS-based Chairs\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLies to me\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDepartment Chair\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e486\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e76.20%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e4.19\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eGroup Leaders in USA\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"6\" nameend=\"c6\" namest=\"c1\"\u003e \u003cp\u003e\u003cb\u003eContextual Behaviors\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUnfair\u003c/p\u003e \u003cp\u003eRecommendation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGroup Leader\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e268\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e62.30%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1.75\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eSenior White Male PI\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVisa Cancellation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1097\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e51.10%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1.58\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eSenior Male PI in USA\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLengthened Stay in Lab\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLab Supervisor\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e124\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e56.40%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1.41\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAsian PIs\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFunding Removal\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1097\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e62.60%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1.83\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eSenior Male PIs\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEncourage Mistreatment\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDepartment Chair\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e486\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e68.40%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1.98\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eWhite Male Department Chair\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAuthorship Violation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1097\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e61.40%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1.79\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eWhite Female PI (Age 46\u0026ndash;55)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSign away rights\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGroup Leader\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e268\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e71.40%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1.88\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eWhite Female PI (Age 46\u0026ndash;55)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIP Violation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGroup Leader\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e268\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e60.90%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1.81\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eWhite Female PI (Age 46\u0026ndash;55)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThe empirical evidence presented in Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e exposes a profound \"Policy-Evidence Gap\". By separating interpersonal hostility from contextual sabotage, we observe that current institutional policies are fundamentally ill-equipped to handle the dual nature of academic abuse. Most institutional responses treat bullying as a psychological issue requiring \"mediation\" or \"conflict resolution.\" Our data shows that bullying is often a rational\u0026mdash;albeit unethical\u0026mdash;strategy for negatively impacting career advancement and resource allocation.\u003c/p\u003e \u003cp\u003e\u003cb\u003eAddressing Interpersonal Hostility.\u003c/b\u003e The high reported incidence of Lying (76.2%) and Privacy Invasion (72.5%) among Department Chairs indicates that existing internal grievance procedures may be structurally inadequate rather than merely underutilized. When the individual charged with upholding departmental norms and ethical standards is also perceived as the primary source of misconduct, the integrity of the reporting hierarchy is fundamentally compromised. In such circumstances, formal channels for redress become functionally inaccessible, as subordinates may reasonably anticipate retaliation, dismissal of claims, or reputational harm. This dynamic effectively produces an accountability \u0026ldquo;dead zone,\u0026rdquo; within which patterns of verbal abuse, information manipulation, and psychologically exclusionary practices such as the \u0026ldquo;Silent Treatment\u0026rdquo; can persist with minimal oversight or intervention.\u003c/p\u003e \u003cp\u003eMoreover, the normalization of these behaviors at the leadership level risks diffusing downward through the departmental climate, reinforcing a culture in which incivility is tacitly accepted or strategically ignored. Addressing these systemic vulnerabilities requires more than incremental policy adjustments; it necessitates the implementation of robust, multi-source evaluation systems, including 360-degree leadership audits that incorporate confidential upward feedback and third-party review. Complementary bias-aware reporting mechanisms are also critical, particularly in light of the elevated levels of incompetence-based targeting observed in Asian demographic contexts, which may reflect both implicit bias and culturally mediated misinterpretations of communication styles. Collectively, these interventions can help reestablish credible oversight, mitigate power asymmetries, and create safer pathways for reporting and accountability.\u003c/p\u003e \u003cp\u003e \u003cb\u003eDismantling Contextual Sabotage.\u003c/b\u003e The heterogeneous forms of sabotage identified in our data necessitate correspondingly differentiated and targeted interventions. Broad, principle-based instruments such as generic \u0026ldquo;codes of conduct\u0026rdquo; are insufficient to address behaviors that are structurally enabled and materially consequential. For instance, such codes offer little deterrent against a PI who leverages institutional authority to threaten visa status or appropriate intellectual contributions through authorship theft. These patterns of misconduct are not merely interpersonal but are embedded in administrative and governance arrangements, and thus require formal, enforceable policy reforms rather than aspirational norms.\u003c/p\u003e \u003cp\u003eIn this regard, the elevated risk of credit misappropriation observed among Group Leaders underscores the need to institutionalize a \u0026ldquo;Right to Digital Attribution.\u0026rdquo; This concept reflects both a legal and ethical entitlement for researchers to be recognized as the originators of their work across digital platforms and collaborative environments. Establishing formalized attribution protocols\u0026mdash;such as immutable contribution records, timestamped authorship claims, and transparent version control systems\u0026mdash;would help preserve the linkage between individuals and their intellectual output. Such mechanisms create an auditable trail that not only deters misconduct but also provides early-career researchers with a measure of protection against reputational and professional harm stemming from credit theft.\u003c/p\u003e \u003cp\u003eAdditionally, the strong predictive relationship between Visa Threats and institutional context within the United States highlights a critical structural vulnerability. Specifically, it points to the need to decouple a researcher\u0026rsquo;s legal residency status from the discretionary authority of an individual PI. As long as a single supervisor retains the capacity to terminate a researcher\u0026rsquo;s legal standing\u0026mdash;potentially with minimal procedural oversight\u0026mdash;the power imbalance is exacerbated to a degree that undermines the foundational premise of mentorship. Under these conditions, the relationship is more accurately characterized by dependency and coercion than by professional development. Addressing this issue may require policy innovations such as institutionally held visa sponsorship, multi-party oversight of employment termination decisions, or grace periods that allow researchers to transition between positions without immediate legal jeopardy.\u003c/p\u003e \u003cp\u003eFinally, the intersectional patterns evident in our findings point to the necessity of culturally competent and context-sensitive interventions. The disproportionately high levels of incompetence-based targeting reported in Asian demographic contexts suggest that bullying behaviors are often intertwined with stereotyping, implicit bias, and, in some cases, overt xenophobia. Effective mitigation strategies must therefore extend beyond formal policy changes to include bias training, inclusive leadership development, and accountability mechanisms that explicitly address how cultural misperceptions and power asymmetries shape the expression and impact of workplace hostility. Collectively, these approaches are essential for dismantling the structural and cultural conditions that enable contextual sabotage to persist.\u003c/p\u003e \u003cp\u003e \u003cb\u003ePolicy Implementation: Operationalizing the Researcher Bill of Rights (RBoR).\u003c/b\u003e To meaningfully translate empirical insights into durable institutional reform, the Researcher Bill of Rights (RBoR) must be operationalized not as a symbolic framework, but as an integrated system of automated and structural \u0026ldquo;circuit breakers.\u0026rdquo; These mechanisms should be designed to activate in response to empirically identified high-risk behaviors, thereby reducing reliance on discretionary enforcement and minimizing opportunities for suppression or retaliation. In light of the observed 76.2% risk of administrative deception and the 71.4% likelihood of forced rights waivers, implementation must prioritize ex ante safeguards that interrupt harmful conduct before it escalates into entrenched patterns of abuse. This approach reframes governance from reactive adjudication to proactive risk mitigation, embedding accountability directly into institutional processes and decision architectures.\u003c/p\u003e \u003cp\u003e \u003cb\u003eNeutralizing Administrative Hostility.\u003c/b\u003e Given that Department Chairs exhibit a 76.2% probability of administrative deception and a 72.5% probability of privacy invasion, existing internal grievance systems\u0026mdash;typically routed through hierarchical reporting lines\u0026mdash;are structurally ill-equipped to ensure impartial review. When those tasked with oversight are themselves implicated in misconduct, procedural legitimacy erodes and reporting becomes both risky and ineffective. To address this, institutions should establish an Ombudsperson Matrix, wherein reporting channels are deliberately decoupled from departmental authority and redirected to independent, extra-departmental ethics bodies with investigatory autonomy. Such bodies should be empowered to initiate inquiries without prior departmental consent, thereby restoring credibility to the reporting process.\u003c/p\u003e \u003cp\u003eComplementing this, all high-stakes administrative decisions\u0026mdash;particularly those involving laboratory space allocation, funding distribution, and contract or appointment renewals\u0026mdash;should be governed by robust audit infrastructures. Specifically, a \u0026ldquo;Dual-Authorization\u0026rdquo; digital trail should be required, ensuring that no single administrator can unilaterally enact decisions with material consequences for a researcher\u0026rsquo;s career trajectory. These audit trails should be time-stamped, immutable, and subject to periodic external review, thereby constraining the \u0026ldquo;Breaking of Promises\u0026rdquo; (75.5%) identified in our model and enhancing procedural transparency across administrative domains.\u003c/p\u003e \u003cp\u003e \u003cb\u003eProtecting Intellectual Integrity and Legal Residency.\u003c/b\u003e The high predictive accuracy associated with Visa Threats (67.03%) and intellectual property (IP) violations (60.9%) highlights the extent to which researchers\u0026rsquo; professional and personal security can be contingent upon the discretionary authority of individual supervisors. This interdependence creates a structurally coercive environment in which compliance may be driven less by scholarly norms than by fear of material consequences. To mitigate this, institutions must systematically decouple core elements of researcher identity\u0026mdash;namely, legal residency and intellectual ownership\u0026mdash;from supervisory control.\u003c/p\u003e \u003cp\u003eOne critical intervention is the establishment of \u0026ldquo;Visa Safe Harbors,\u0026rdquo; whereby visa sponsorship is administered centrally through the university\u0026rsquo;s International Office rather than delegated to individual Principal Investigators. This redistribution of authority directly attenuates the \u0026ldquo;Legal Leverage\u0026rdquo; identified as a key mechanism of control among Senior Male PIs (54.7%), and introduces institutional buffering that allows researchers to report misconduct or transition between positions without immediate risk to their legal status. Additional safeguards, such as mandatory transition periods or institutional sponsorship continuity clauses, could further reduce vulnerability during periods of professional disruption.\u003c/p\u003e \u003cp\u003eParallel protections are required to safeguard intellectual contributions. To address the 76.1% risk of forced rights waivers among mid-career researchers, journals and funding agencies should mandate standardized authorship disclosure frameworks, such as the CRediT (Contributor Roles Taxonomy). Embedding these frameworks within submission and grant-reporting systems creates a verifiable, multi-party record of contribution that cannot be unilaterally altered. Such \u0026ldquo;Verified Attribution\u0026rdquo; systems function as both deterrent and recourse mechanism, preventing the retroactive erasure of credit (73.3%) and ensuring that scholarly contributions remain transparently and durably linked to their originators.\u003c/p\u003e \u003cp\u003e \u003cb\u003eTargeted Leadership Intervention.\u003c/b\u003e The findings further suggest that uniform, \u0026ldquo;one-size-fits-all\u0026rdquo; training interventions are insufficient to address the heterogeneity of risk profiles across supervisory populations. Instead, leadership interventions should be calibrated to reflect demographic, structural, and contextual variations in observed behavior patterns. While all individuals in supervisory roles should be subject to routine Lab Climate Audits, enhanced scrutiny may be warranted in environments led by individuals or cohorts associated with elevated risk indicators. These audits should integrate both quantitative climate metrics and qualitative feedback, enabling a more nuanced assessment of leadership effectiveness and interpersonal conduct.\u003c/p\u003e \u003cp\u003eIn addition, the implementation of 360-degree feedback systems is essential for rebalancing power dynamics within academic hierarchies. Under such systems, supervisory privileges\u0026mdash;including personnel management authority and eligibility to serve as principal investigator on grants\u0026mdash;would be contingent upon demonstrably positive evaluations from subordinates and peers. This effectively reconfigures authority as contingent and performance-based, rather than as an entrenched entitlement. When coupled with transparent reporting thresholds and consequences for persistently negative evaluations, these systems can incentivize prosocial leadership behaviors while systematically identifying and remediating dysfunctional supervisory practices.\u003c/p\u003e \u003cp\u003eCollectively, these implementation strategies reposition the RBoR as an enforceable governance framework rather than an aspirational document. By embedding accountability into institutional infrastructure\u0026mdash;through independent oversight, auditable decision-making, and structurally decoupled power mechanisms\u0026mdash;organizations can more effectively disrupt the conditions under which academic bullying and exploitation are able to persist.\u003c/p\u003e \u003cp\u003e \u003cb\u003eLimitations and Future Directions.\u003c/b\u003e It is important to recognize several limitations associated with this study. Most notably, our examination of bullying was constrained to individuals occupying higher-status roles, as participants were prompted to consider interactions with \u0026ldquo;academic superiors.\u0026rdquo; Consequently, additional empirical work is needed to capture bullying enacted by peers and other members of the academic community, including collective forms such as mobbing.\u003c/p\u003e \u003cp\u003eWe also note the potential for sampling bias, as individuals who had directly experienced or observed bullying may have been more inclined to participate than those without such experiences. In addition, a substantial proportion of respondents were based in the United States, which may limit the generalizability of the findings across different national and institutional contexts.\u003c/p\u003e \u003cp\u003eFinally, aspects of the survey design may have restricted the range of insights obtained. Participants who initially reported no experience with or exposure to academic bullying were routed away from key measures, including the Tepper abusive supervision scale and the contextual behavior checklist. It is possible that some of these individuals would have endorsed exposure to specific behaviors captured by these instruments had they been given the opportunity.\u003c/p\u003e \u003cp\u003eNotwithstanding these limitations, the study contributes a more fine-grained understanding of the behaviors most commonly encountered by targets of academic bullying, along with their typical responses and the consequences that follow.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eThis study provides evidence that academic bullying is not an incidental interpersonal phenomenon, but rather a structural outcome of an institutional ecology that allocates substantial discretionary power to supervisory actors without commensurate mechanisms of oversight, constraint, or accountability. Within this configuration, abusive supervision and related forms of academic bullying emerge not randomly, but in discernible and statistically patterned clusters that map onto hierarchical positions and role-based authority gradients. In particular, individuals occupying formal leadership roles are positioned at structural nodes where decision rights, resource control, and evaluative authority converge, thereby increasing both the opportunity and the efficacy of coercive or exploitative behaviors when safeguards are absent.\u003c/p\u003e \u003cp\u003eAccordingly, the shift from merely documenting these behaviors to implementing infrastructural accountability mechanisms is no longer a discretionary matter of institutional preference or cultural aspiration; it is an empirically grounded necessity implied by the observed distribution and predictability of misconduct. The model presented in this study indicates that, in the absence of structural intervention, specific cohorts will continue to deploy identifiable and repeatable levers of academic sabotage, including threats to legal status, manipulation of authorship credit, and control over material research resources. These mechanisms persist precisely because they are embedded within governance structures that permit unilateral or minimally constrained decision-making.\u003c/p\u003e \u003cp\u003eIn response, the codification of the Researcher Bill of Rights represents a transition from a governance paradigm characterized as \u0026ldquo;mentorship by permission\u0026rdquo;\u0026mdash;in which access, opportunity, and professional survival are contingent on discretionary approval\u0026mdash;to one grounded in \u0026ldquo;mentorship by accountability,\u0026rdquo; in which supervisory authority is explicitly bounded by enforceable rights, transparent processes, and independent oversight. This reframing is intended to reduce asymmetries of power that enable coercion while reinforcing the principle that scientific training should be governed by standards of fairness rather than personal discretion.\u003c/p\u003e \u003cp\u003eOperationally, this transition requires a set of interlocking institutional reforms. These include decentralized and externally insulated oversight structures for administrative decision-making roles, ensuring that evaluative authority is not concentrated within single supervisory units. They also include the implementation of verified digital credit systems that preserve the integrity of intellectual attribution across collaborative environments, thereby reducing opportunities for authorship manipulation or retrospective credit reallocation. In addition, jurisdictional protections are necessary to prevent the instrumentalization of visa or residency status as a mechanism of compliance enforcement or professional coercion.\u003c/p\u003e \u003cp\u003eTaken together, these reforms reflect a broader normative and empirical conclusion: the integrity of scientific systems cannot be meaningfully separated from the dignity, autonomy, and security of the individuals who operate within them. Sustained scientific excellence is therefore contingent not only on intellectual rigor, but also on the establishment of institutional environments in which power is systematically checked, accountability is structurally enforced, and participation is not conditioned on vulnerability.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eCompeting Interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eM. M. is co-founder and director of the Academic Parity Movement (www.paritymovement.org) (a non-profit organization dedicated to addressing academic discrimination, violence and incivility). S.M. is director of the Academic Parity Movement.\u003c/p\u003e\n\u003cp\u003eAuthor Contribution\u003c/p\u003e\n\u003cp\u003eS.M. and M.M. designed and ran the survey; S.M., H.M. and M.M. analyzed the results. S.M., H.M. and M.M. wrote the main manuscript. All authors reviewed the manuscript.\u003c/p\u003e\n\u003cp\u003eData Availability\u003c/p\u003e\n\u003cp\u003eSurvey data and the developed code for data analysis are available in the following link: https://github.com/Morimahmoudi/AcademicBullying.git\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n \u003cli\u003eZabrodska, K.; Kveton, P., Prevalence and forms of workplace bullying among university employees. \u003cem\u003eEmployee Responsibilities and Rights Journal\u0026nbsp;\u003c/em\u003e\u003cstrong\u003e2013,\u003c/strong\u003e \u003cem\u003e25\u003c/em\u003e (2), 89-108.\u003c/li\u003e\n \u003cli\u003eAverbuch, T.; Eliya, Y.; Van Spall, H. G. C., Systematic review of academic bullying in medical settings: dynamics and consequences. \u003cem\u003eBMJ open\u0026nbsp;\u003c/em\u003e\u003cstrong\u003e2021,\u003c/strong\u003e \u003cem\u003e11\u003c/em\u003e (7), e043256.\u003c/li\u003e\n \u003cli\u003eMahmoudi, M., Academic bullying slows the evolution of science. \u003cem\u003eNature Reviews Materials\u0026nbsp;\u003c/em\u003e\u003cstrong\u003e2023,\u003c/strong\u003e \u003cem\u003e8\u003c/em\u003e (5), 301-303.\u003c/li\u003e\n \u003cli\u003eT\u0026auml;uber, S.; Loyens, K.; Oertelt-Prigione, S.; Kubbe, I., Harassment as a consequence and cause of inequality in academia: A narrative review. \u003cem\u003eEClinicalMedicine\u0026nbsp;\u003c/em\u003e\u003cstrong\u003e2022,\u003c/strong\u003e \u003cem\u003e49\u003c/em\u003e.\u003c/li\u003e\n \u003cli\u003eMoss, S.; T\u0026auml;uber, S.; Sharifi, S.; Mahmoudi, M., The need for the development of discipline-specific approaches to address academic bullying. \u003cem\u003eeClinicalMedicine\u0026nbsp;\u003c/em\u003e\u003cstrong\u003e2022,\u003c/strong\u003e \u003cem\u003e50\u003c/em\u003e, 101598.\u003c/li\u003e\n \u003cli\u003eMoss, S. E.; Mahmoudi, M., STEM the bullying: An empirical investigation of abusive supervision in academic science. \u003cem\u003eeClinicalMedicine\u0026nbsp;\u003c/em\u003e\u003cstrong\u003e2021,\u003c/strong\u003e \u003cem\u003e40\u003c/em\u003e, 101121.\u003c/li\u003e\n \u003cli\u003eMoss, S. E.; Mahmoudi, M., Examining Age-Dependent Patterns in Academic Bullying Behaviors. \u003cem\u003eJournal of Academic Ethics\u0026nbsp;\u003c/em\u003e\u003cstrong\u003e2025,\u003c/strong\u003e \u003cem\u003e24\u003c/em\u003e (1), 25.\u003c/li\u003e\n \u003cli\u003eMahmoudi, M.; Keashly, L., Filling the space: a framework for coordinated global actions to diminish academic bullying. \u003cem\u003eAngewandte Chemie\u0026nbsp;\u003c/em\u003e\u003cstrong\u003e2021,\u003c/strong\u003e \u003cem\u003e133\u003c/em\u003e (7), 3378-3384.\u003c/li\u003e\n \u003cli\u003eManathunga, C., Supervision as mentoring: The role of power and boundary crossing. \u003cem\u003eStudies in Continuing education\u0026nbsp;\u003c/em\u003e\u003cstrong\u003e2007,\u003c/strong\u003e \u003cem\u003e29\u003c/em\u003e (2), 207-221.\u003c/li\u003e\n \u003cli\u003eVogelaar, A. E.; Mahmoudi, M., Chronic silencing is a critical barrier to breaking the cycle of bullying in academia and industry. \u003cem\u003eNature Biotechnology\u0026nbsp;\u003c/em\u003e\u003cstrong\u003e2025,\u003c/strong\u003e \u003cem\u003e43\u003c/em\u003e (9), 1577-1579.\u003c/li\u003e\n \u003cli\u003eTepper, B. J., Abusive Supervision Scale. \u003cem\u003eJournal of Applied Psychology\u0026nbsp;\u003c/em\u003e\u003cstrong\u003e2000\u003c/strong\u003e.\u003c/li\u003e\n \u003cli\u003eTepper, B. J., Consequences of Abusive Supervision. \u003cem\u003eAcademy of Management Journal\u0026nbsp;\u003c/em\u003e\u003cstrong\u003e2000,\u003c/strong\u003e \u003cem\u003e43\u003c/em\u003e (2), 178-190.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-9269024/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9269024/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eAcademic bullying and harassment are historically treated as idiosyncratic interpersonal conflicts, leading to generalized institutional policies that often fail to address the nuances of supervisory abuse. Using a global dataset (N = 2,041), we analyzed 15 Tepper Scale abusive behaviors and 10 contextual academic bullying behaviors in relation to perpetrator demographics and hierarchical positions. To account for cross-behavior dependence, we employed an end-to-end machine learning pipeline utilizing a multiclass classifier chain. Our findings reveal that the \"one-size-fits-all\" approach to harassment policy is empirically inadequate and pragmatically ineffective. We identified high-risk \"behavioral clusters\" tied to specific roles, such as Department Chairs and Group Leaders, who utilize distinct bullying tactics—ranging from visa cancellation threats to intellectual property (IP) theft. We propose a Researcher Bill of Rights (RBoR) as a blueprint for dismantling this structural ecology.\u003c/p\u003e","manuscriptTitle":"Beyond General Awareness: A Global Empirical Framework for Behavior-Specific Policy Development Against Academic Bullying","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-04-16 03:50:04","doi":"10.21203/rs.3.rs-9269024/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"7f277185-3c67-433b-8fce-ba370f06d4ff","owner":[],"postedDate":"April 16th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":66327986,"name":"Social science/Criminology"},{"id":66327987,"name":"Biological sciences/Psychology"},{"id":66327988,"name":"Social science/Psychology"},{"id":66327989,"name":"Social science/Science technology and society"}],"tags":[],"updatedAt":"2026-04-16T03:50:04+00:00","versionOfRecord":[],"versionCreatedAt":"2026-04-16 03:50:04","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9269024","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9269024","identity":"rs-9269024","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}