Buss-Durkee Hostility Inventory (BDHI)

1. Abstract

The Buss-Durkee Hostility Inventory (BDHI) is a foundational, multidimensional psychological assessment instrument developed by Arnold H. Buss and Ann Durkee in 1957 to measure various manifest and latent components of hostility and aggression. Spanning 75 self-report items administered via a dichotomous True/False response format, the inventory was constructed to overcome the unifactorial conceptualizations of aggression that dominated early twentieth-century psychometrics. The BDHI delineates hostility into eight specialized subscales reflecting behavioral, emotional, and cognitive manifestations: Assault (10 items), Indirect Hostility (9 items), Irritability (11 items), Negativism (5 items), Resentment (8 items), Suspicion (10 items), Verbal Hostility (13 items), and an independent evaluative subscale, Guilt (9 items). Psychometric investigations demonstrate robust internal consistency across subscales (Kuder-Richardson Formula 20 coefficients typically ranging between .60 and .84), satisfactory test-retest reliability across multiple temporal intervals, and convergent validity with observational ratings, laboratory provocation paradigms, and collateral personality inventories including the Minnesota Multiphasic Personality Inventory (MMPI). Structural and exploratory factor analytic investigations historically revealed two dominant higher-order dimensions: an Overt/Expressive Hostility (motor/behavioral) factor and a Covert/Neurotic Hostility (cognitive/attitudinal) factor. Widely utilized across five decades of clinical psychology, forensic criminology, behavioral medicine, and cardiovascular epidemiology, the BDHI established the empirical framework for modern anger-hostility-aggression research and served as the direct conceptual progenitor to the Buss-Perry Aggression Questionnaire (BPAQ).

2. Keywords

Buss-Durkee Hostility Inventory, BDHI, Aggression Assessment, Hostile Attitudes, Motor Aggression, Psychometrics, Factor Structure, Behavioral Medicine, Overt Hostility, Covert Hostility

3. Authors

The Buss-Durkee Hostility Inventory was co-authored by Arnold H. Buss (Ph.D.) and Ann Durkee (M.A.) in 1957 during their tenure at the Carter Memorial Hospital and Indiana University Medical Center in Indianapolis, Indiana.

Arnold H. Buss (1924–2021) subsequently served as a prominent Professor of Psychology at the University of Texas at Austin, after academic appointments at the University of Pittsburgh and Rutgers University. Dr. Buss was a pioneer in experimental personality research, behavioral genetics, human aggression, and childhood temperament (co-developing the EAS Temperament Model with Robert Plomin). Ann Durkee contributed extensively to clinical psychometric construction, psychological testing, and psychiatric data collection in hospitalized psychiatric cohorts and normative non-clinical samples throughout the mid-twentieth century.

4. Purpose and Clinical Utility

The primary purpose of the Buss-Durkee Hostility Inventory is to capture the complex, multidimensional nature of human hostility by explicitly distinguishing between behavioral acts of aggression, affective experiences of irritability, and cognitive or attitudinal orientations of distrust and cynicism. Prior to the construction of the BDHI in 1957, psychological testing routinely conflated subjective emotional anger, external behavioral destruction, and covert resentment into generalized, undifferentiated indices of “aggressiveness.” Buss and Durkee recognized that an individual might harbor intense cognitive resentment or chronic cynicism without ever initiating physical violence; conversely, another individual might display explosive verbal or physical aggression under provocation without enduring persistent persecutory suspiciousness.

Research Applications

The BDHI has served as a cornerstone instrument across several diverse academic disciplines:

• Epidemiology and Behavioral Medicine: The BDHI was instrumental in demonstrating the pathogenic links between chronic hostility, autonomic hyperreactivity, and coronary heart disease (CAD). When health psychologists investigated the toxic component of the Type A behavior pattern, subscales of the BDHI (specifically Resentment and Suspicion) proved superior to general activity scales in predicting myocardial infarction and endothelial dysfunction.
• Forensic and Correctional Psychology: Criminologists and institutional psychologists utilized the BDHI to evaluate violent offenders, assess institutional infraction risk, and differentiate instrumental, premeditated aggression from emotionally dysregulated, reactive aggression.
• Social and Experimental Psychology: The instrument provided an objective baseline covariate for laboratory experiments investigating frustration-aggression dynamics, media violence exposure, provocation paradigms (such as the Taylor Competitive Reaction Time Task), and social cognitive processing.

Clinical Assessment and Treatment Planning

Clinically, the BDHI assists psychotherapists, psychiatrists, and clinical neuropsychologists in deconstructing diffuse presenting complaints of “anger management issues.” By administering the 75 items, clinicians obtain a differentiated profile across eight distinct axes:

• Differential Diagnosis: High scores on Assault paired with elevated Irritability often suggest impulse-control dysregulation, such as Intermittent Explosive Disorder or Cluster B personality pathology (e.g., Antisocial or Borderline Personality Disorder). In contrast, marked elevations on Suspicion and Resentment in the absence of motor violence may indicate paranoid or persecutory ideation, depressive grievance processing, or somatic distress.
• Treatment Customization: Patients displaying heightened Verbal Hostility and Negativism benefit from targeted assertiveness training and communication restructuring, whereas patients scoring high on Guilt and Resentment typically require cognitive restructuring directed at moral rumination, shame processing, and cognitive reappraisal.
• Treatment Tracking: Administering the BDHI longitudinally enables clinical teams to track pharmacotherapeutic interventions (e.g., mood stabilizers, SSRIs) and psychotherapeutic regimens (e.g., Cognitive Behavioral Therapy for Anger) by monitoring objective reductions across behavioral vs. cognitive hostility clusters over time.

5. Psychological Construct

The BDHI conceptualizes hostility as a multifaceted affective-behavioral-cognitive matrix consisting of eight systematically operationalized subscales. Rather than viewing hostility as a singular drive, Buss and Durkee (1957) postulated that aggression diverges along distinct expressive modalities (physical vs. verbal), structural pathways (direct vs. indirect), and psychological domains (behavioral, affective, attitudinal, and conscience-related).

1. Assault (Physical Hostility)

Comprising 10 items, Assault measures physical violence directed against other persons. It assesses both initiation of bodily harm and physical retaliation when provoked. Manifestations include fist fights, slapping, using physical force to defend personal rights, and coming to physical blows during interpersonal disputes (e.g., Item 25: “If somebody hits me first, I let him have it”; Item 49: “When I really lose my temper, I am capable of slapping”).

2. Indirect Hostility

Represented by 9 items, Indirect Hostility captures covert, roundabout aggression where the aggressive behavior is either not directed toward the target person directly or is mediated through non-confrontational channels. This subscale encompasses both undirected emotional discharges (such as slamming doors, throwing or breaking objects, stomping feet) and covert interpersonal sabotage (such as malicious gossip, spreading damaging rumors, or passive-aggressive spiteful behavior).

3. Irritability

Containing 11 items, Irritability taps affective volatility and the threshold of emotional provocation. It measures a respondent’s readiness to explode into anger upon encountering minor environmental or interpersonal frustrations. Characteristics include losing one’s temper easily, feeling like a “powder keg ready to explode,” persistent grouchiness, and chronic low frustration tolerance (e.g., Item 4: “I lose my temper easily but get over it quickly”; Item 44: “I often feel like a powder keg ready to explode”).

4. Negativism

Formulated through 5 items, Negativism evaluates oppositional behavior directed against authority, structural rules, or interpersonal demands. Rooted in passive or active defiance, negativistic aggression manifests as deliberate non-compliance, violating unwelcome regulations, doing the opposite of what an authoritarian figure demands, or deliberately dragging one’s feet to express resistance (e.g., Item 19: “When someone is bossy, I do the opposite of what he asks”).

5. Resentment

Spanning 8 items, Resentment operationalizes internal cognitive hostility characterized by jealousy, chronic grievance, envy, and the pervasive perception of being mistreated or denied fairness by life or other individuals. Resentment represents a smoldering affective-cognitive state wherein individuals ruminate over perceived injustices, harbor long-standing grudges, and believe that “other people always seem to get the breaks” (Item 13).

6. Suspicion

Measured via 10 items, Suspicion constitutes the projective, cognitive dimension of hostility. It taps interpersonal distrust, paranoia, and the chronic belief that others possess malicious intent, are talking behind one’s back, or are attempting to harm, mock, or exploit the individual (e.g., Item 6: “I know that people tend to talk about me behind my back”; Item 54: “I commonly wonder what hidden reason another person may have for doing something nice for me”).

7. Verbal Hostility

Composed of 13 items, Verbal Hostility assesses overt negative affect expressed through spoken language. This dimension includes raising one’s voice, shouting, screaming, arguing aggressively, using profanity or nasty remarks, making verbal threats, and engaging in sharp, cutting sarcasm during disagreements (e.g., Item 47: “When people yell at me, I yell back”; Item 51: “When I get mad, I say nasty things”).

8. Guilt

Consisting of 9 items, Guilt was designed by Buss and Durkee not as an aggressive behavior per se, but as an index of self-directed negative affect, remorse, and conscience activation following perceived moral failure or harm inflicted upon others. Buss and Durkee included Guilt because strong internal remorse acts as a vital psychological brake or regulator of aggressive action. The items examine feelings of shame, conscience punishment, remorse over cheating or failure, and distress regarding moral conduct (e.g., Item 64: “When I do wrong, my conscience punishes me”).

6. Theoretical Framework

The Buss-Durkee Hostility Inventory was constructed during an era of fundamental conceptual realignments in psychological science, navigating between drive-reduction psychoanalytic frameworks, early behaviorism, and emerging social-cognitive paradigms.

The Frustration-Aggression Hypothesis and Buss’s Critique

In 1939, John Dollard, Leonard Doob, Neal Miller, O. H. Mowrer, and Robert Sears published the celebrated Frustration-Aggression Hypothesis, proposing that aggression is always a consequence of frustration, and that frustration invariably leads to some form of aggression. Arnold H. Buss, in his landmark theoretical text The Psychology of Aggression (1961), fundamentally reconceptualized this paradigm.

Buss argued that frustration is only an antecedent condition that produces emotional arousal, and that learning, reinforcement history, and instrumental conditioning dictate whether an individual mounts an aggressive response. He decoupled the internal emotional state from the observable behavior, defining aggression strictly as an instrumental response that delivers noxious stimuli to another organism. Hostility, in contrast, was defined by Buss as an enduring cognitive attitude involving negative evaluations of people and events, sustained by an underlying emotional substrate of anger.

The Tripartite Architecture: Affect, Behavior, and Cognition

Buss formalized a tripartite model that served as the theoretical infrastructure of the BDHI:

1. Affective Component (Anger / Irritability): Physiological arousal, autonomic preparation for flight-or-fight, and visceral sensations of displeasure, tension, and emotional volatility.
2. Behavioral Component (Motor Aggression / Assault / Verbal Hostility / Indirect Hostility): Instrumental actions executed to inflict injury, pain, or damage upon persons or physical property, either directly or circuitously.
3. Cognitive Component (Attitudinal Hostility / Resentment / Suspicion): Belief systems, schema-driven interpretations, cynicism, attribution of hostile intent to others, and persistent feelings of bitterness and persecution.

By mapping these three interrelated yet empirically separable components onto discrete subscales, the BDHI allowed psychometricians to assess whether individuals suffered primarily from emotional instability (elevated Irritability), behavioral behavioral disinhibition (elevated Assault and Verbal Hostility), or cognitive-perceptual paranoia and cynicism (elevated Suspicion and Resentment).

7. Validity

Over six decades of psychometric scrutiny have evaluated the validity of the BDHI across clinical, forensic, normal, and medical populations.

Construct and Convergent Validity

Convergent validity of the BDHI has been extensively verified against external behavioral markers, observational metrics, and established psychometric scales:

• Comparison with the MMPI: The BDHI total score and its constituent subscales correlate strongly with the Anger, Cynicism, and Paranoia scales of the Minnesota Multiphasic Personality Inventory (MMPI-2). Specifically, the BDHI Suspicion and Resentment subscales exhibit Pearson correlations ranging from .55 to .72 with the Cook-Medley Hostility Scale (derived from the MMPI pool).
• Convergence with the STAXI: Parallel administration with Charles Spielberger’s State-Trait Anger Expression Inventory (STAXI) demonstrates robust convergent alignment. BDHI Irritability correlates significantly with Trait Anger (.62 to .71), whereas BDHI Assault and Verbal Hostility correlate highly with Anger Expression-Out (AX-Out; .58 to .68).
• Laboratory Provocation: In experimental laboratory settings involving competitive reaction-time tasks (the Taylor aggression paradigm) or simulated shock administration, participants scoring higher on the BDHI Assault subscale delivered significantly higher shock intensities and sustained noxious noise blasts against opponents, demonstrating convergent criterion validity between self-report items and real-time physical aggression.

Discriminant Validity

Discriminant validity research confirms that the primary hostility subscales (Assault, Verbal Hostility, Indirect Hostility) can be statistically differentiated from generalized anxiety, neurosis, and somatic depression. While Resentment and Guilt show moderate overlap with dysphoric mood states (correlating .35 to .48 with Beck Depression Inventory indices), the behavioral motor subscales (Assault, Verbal Hostility) demonstrate low correlations (< .20) with generalized trait anxiety inventories (such as the Taylor Manifest Anxiety Scale), confirming that the BDHI does not simply reflect diffuse psychological distress.

Predictive and Ecological Validity

The predictive validity of the BDHI is notable across forensic and medical contexts:

• Criminological Outcomes: Longitudinal and cross-sectional studies in penal institutions demonstrated that incarcerated individuals with violent offense histories scored significantly higher on the BDHI Assault (mean differences reaching p < .001, Cohen's d > 0.85) than non-violent property offenders.
• Cardiovascular Morbidity: Prospective epidemiological studies (e.g., Barefoot et al., Williams et al.) monitoring large adult cohorts over 10-to-25-year intervals revealed that elevated scores on the cognitive hostility dimensions (Suspicion and Resentment) independently predicted coronary artery calcification, recurrent ischemic events, and all-cause cardiovascular mortality, even after controlling for standard biomedical risk factors such as hypertension, smoking, and cholesterol.

8. Reliability

The psychometric reliability of the Buss-Durkee Hostility Inventory has been demonstrated across diverse language adaptations, normative clinical cohorts, and experimental investigations.

Internal Consistency

Because the original BDHI uses a dichotomous (True/False) scoring format, internal consistency was initially evaluated using the Kuder-Richardson Formula 20 (KR-20), followed by Cronbach’s alpha in later modern psychometric evaluations:

• Total Scale: Across normative collegiate and adult cohorts, the composite BDHI total score (often calculated across the 7 hostility subscales, 66 items, excluding Guilt) displays internal consistency coefficients ranging from .81 to .91, indicating strong overall scale homogeneity.
• Subscale Reliabilities: Individual subscales reflect varied alpha levels, partially constrained by brief item lengths:
  • Assault (10 items): α = .78 to .84
  • Verbal Hostility (13 items): α = .74 to .82
  • Irritability (11 items): α = .72 to .80
  • Suspicion (10 items): α = .70 to .79
  • Resentment (8 items): α = .68 to .76
  • Indirect Hostility (9 items): α = .62 to .71
  • Guilt (9 items): α = .65 to .75
  • Negativism (5 items): α = .48 to .62 (the lower coefficient being directly attributable to its short 5-item brevity).

Test-Retest Stability

Temporal stability assessments reveal that the BDHI measures stable trait-like dispositions rather than transient situational mood fluctuations:

• Over a 2- to 3-week test-retest window, stability coefficients for the total BDHI score typically range between r = .82 and .88 in university and community samples.
• Longer test-retest intervals spanning 6 to 12 months in non-clinical adult populations demonstrate stability coefficients between .68 and .76, confirming that the underlying constructs—particularly Assault, Suspicion, and Resentment—represent enduring personality orientations.

9. Factor Analysis

The underlying factor structure of the BDHI has been the subject of extensive empirical debate and exploratory/confirmatory factor analytic (EFA/CFA) investigation since its 1957 debut.

Original Exploratory Factor Analysis (Buss & Durkee, 1957)

In their initial standardization sample of 270 college students, Arnold Buss and Ann Durkee conducted principal components analysis with orthogonal rotation. Their findings indicated that while the 8 subscales possessed face validity and theoretical coherence, they did not represent 8 completely independent latent constructs. Instead, two dominant higher-order factors emerged, accounting for the vast majority of the common variance:

1. Factor I: Motor/Overt Aggression (Expressive Factor): Heavily loaded by Assault, Indirect Hostility, Verbal Hostility, and Irritability. This factor captures outward behavioral energy, physical destructiveness, verbal confrontation, and acute emotional reactivity.
2. Factor II: Attitudinal/Covert Hostility (Cognitive/Neurotic Factor): Heavily loaded by Resentment and Suspicion, with secondary cross-loadings from Guilt and Negativism. This factor taps internal persecutory feelings, social distrust, envy, grievance processing, and moral distress.

Subsequent Replications and Models

Subsequent psychometricians conducted secondary analyses to refine the BDHI’s latent architecture:

• Bendig’s Two-Factor Solution (1962): Bendig administered the BDHI alongside social desirability scales and identified two robust dimensions: “Covert Hostility” (Resentment, Suspicion) and “Overt Hostility” (Assault, Verbal, Irritability). He demonstrated that Covert Hostility shared significant variance with neuroticism and introversion, whereas Overt Hostility aligned with extraversion and behavioral impulsivity.
• Sarason’s Factorial Investigations (1961): Sarason confirmed that when analyzing the 75 individual items directly (rather than subscale totals), the dichotomous format often produced difficulty-based artifact factors, yet the fundamental dichotomy between physical/verbal behavioral aggression and cognitive alienation consistently emerged.
• The Evolution into the BPAQ (Buss & Perry, 1992): By the late 1980s, psychometricians noted several structural limitations of the BDHI: the dichotomous True/False format restricted response variance, several items were culturally dated, and item-level factor analyses failed to cleanly replicate the 8 original subscales. In 1992, Arnold Buss teamed with Mark Perry to revise the BDHI into the 29-item Buss-Perry Aggression Questionnaire (BPAQ). Using modern confirmatory factor analysis and a 5-point Likert response scale, Buss and Perry refined the BDHI’s components into four robust latent factors: Physical Aggression (derived from BDHI Assault), Verbal Aggression (derived from BDHI Verbal Hostility), Anger (derived from BDHI Irritability), and Hostility (derived from BDHI Suspicion and Resentment). Thus, the structural factor analysis of the BDHI directly established the empirical model used in contemporary personality and clinical psychology.

10. Instrument / Measurement Tool

• Instrument Name: Buss-Durkee Hostility Inventory (BDHI)
• Alternative Titles: Buss-Durkee Hostility-Guilt Inventory (BDHGI), Buss-Durkee Inventory
• Original Authors: Arnold H. Buss, Ph.D., and Ann Durkee, M.A.
• Publication Year: 1957
• Format / Administration Method: Paper-and-pencil or computerized self-report questionnaire; individual or group administration.
• Administration Time: Approximately 15 to 25 minutes.
• Item Count: 75 items total.
• Response Scale: Dichotomous: True or False.
• Target Population: Adolescents (aged 14+) and adults across clinical, forensic, psychiatric, and community populations.
• Scale Structure (8 Subscales):
  • Assault (10 items): 1, 9, 17, 25, 33, 41, 49, 57, 65, 70
  • Indirect Hostility (9 items): 2, 10, 18, 26, 34, 42, 50, 58, 75
  • Irritability (11 items): 4, 11, 20, 27, 35, 44, 52, 60, 66, 71, 73
  • Negativism (5 items): 3, 12, 19, 28, 36
  • Resentment (8 items): 5, 13, 21, 29, 37, 45, 53, 61
  • Suspicion (10 items): 6, 14, 22, 30, 38, 46, 54, 62, 67, 72
  • Verbal Hostility (13 items): 7, 15, 23, 31, 39, 43, 47, 51, 55, 59, 63, 68, 74
  • Guilt (9 items): 8, 16, 24, 32, 40, 48, 56, 64, 69
• Scoring Instructions:
  • Each item is scored 1 point when answered in the keyed direction and 0 points when answered in the non-keyed direction.
  • Standard (Keyed “True” for 1 point): Items 2, 3, 4, 5, 6, 7, 8, 9, 11, 12, 13, 14, 15, 16, 18, 19, 20, 22, 23, 24, 25, 26, 28, 29, 30, 31, 32, 33, 35, 36, 37, 38, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 51, 52, 53, 54, 56, 57, 58, 59, 60, 61, 62, 64, 65, 68, 69, 70, 73, 75.
  • Reverse Scored (Keyed “False” for 1 point): Items 1, 10, 17, 21, 27, 34, 39, 50, 55, 63, 66, 67, 71, 72, 74.
  • Composite Hostility Score: Calculated by summing the scores of the 7 hostility subscales (Assault, Indirect, Irritability, Negativism, Resentment, Suspicion, Verbal Hostility), yielding a maximum score of 66. The Guilt subscale (9 items) is analyzed independently or combined to form the Total Hostility-Guilt Index (maximum score 75).

11. Permissions, Fee, and Test Year

The Buss-Durkee Hostility Inventory was originally developed in 1957 and published in the Journal of Consulting Psychology (Buss & Durkee, 1957). The full 75-item inventory was published directly in the academic literature for scientific, educational, and clinical utilization.

As an archival instrument published over six decades ago in a primary academic journal, the BDHI is widely regarded as entering the public domain for empirical scientific investigation and clinical educational use. No commercial licensing fees or royalty payments are required to administer or score the standard 75-item instrument for research purposes. Clinicians and researchers must cite the original 1957 validation manuscript when presenting or publishing empirical findings derived from the inventory. For institutional and high-stakes diagnostic applications, practitioners frequently utilize its modernized successor, the Buss-Perry Aggression Questionnaire (BPAQ), or standardized computer-scored editions distributed by clinical test publishers.

12. References

• Barefoot, J. C., Dahlstrom, W. G., & Williams, R. B. (1983). Hostility, CHD incidence, and total mortality: A 25-year follow-up study of 255 physicians. Psychosomatic Medicine, 45(1), 59–63. https://doi.org/10.1097/00006842-198302000-00008
• Bendig, A. W. (1962). Factor analytic scales of covert and overt hostility. Journal of Consulting Psychology, 26(2), 200. https://doi.org/10.1037/h0040776
• Bushman, B. J., & Anderson, C. A. (2002). Does the media make people more violent? Science, 291(5512), 2377–2379. https://doi.org/10.1126/science.291.5512.2377
• Buss, A. H. (1961). The Psychology of Aggression. John Wiley & Sons. https://doi.org/10.1037/11160-000
• Buss, A. H., & Durkee, A. (1957). An inventory for assessing different kinds of hostility. Journal of Consulting Psychology, 21(4), 343–349. https://doi.org/10.1037/h0046900
• Buss, A. H., & Perry, M. (1992). The Aggression Questionnaire. Journal of Personality and Social Psychology, 63(3), 452–459. https://doi.org/10.1037/0022-3514.63.3.452
• Cook, W. W., & Medley, D. M. (1954). Proposed hostility and pharisaic-virtue scales for the MMPI. Journal of Applied Psychology, 38(6), 414–418. https://doi.org/10.1037/h0060667
• Dembroski, T. M., MacDougall, J. M., Williams, R. B., Haney, T. L., & Blumenthal, J. A. (1985). Components of Type A, hostility, and anger-in: Relationship to angiographic findings. Psychosomatic Medicine, 47(3), 219–233. https://doi.org/10.1097/00006842-198505000-00001
• Dollard, J., Doob, L. W., Miller, N. E., Mowrer, O. H., & Sears, R. R. (1939). Frustration and Aggression. Yale University Press. https://doi.org/10.1037/10022-000
• Lemerise, E. A., & Dodge, K. A. (2008). The development of anger and aggression. In M. W. Vasey & M. R. Brendgen (Eds.), Handbook of Developmental Psychology: Social, Emotional, and Personality Development (pp. 395–420). John Wiley & Sons.
• Sarason, I. G. (1961). Intercorrelations among measures of hostility. Journal of Clinical Psychology, 17(2), 192–195. Below are the authentic scale items in their original language as published in the standard psychometric validation studies, without modification or translation to preserve instrument validity and reliability: