Attitudes About Marriage scale

1. Abstract

The Attitudes About Marriage scale—specifically recognized in clinical psychology and marital discord literature as the Attitudes About Marriage Index – Aggression Scale (AMI-AS)—is a specialized psychometric assessment instrument initially conceptualized by Gayla Margolin and Laurence Foo (1992) and empirically detailed by Holly Garcia O’Hearn and Gayla Margolin (2000). Designed to evaluate the cognitive boundaries of interpersonal violence within intimate relationships, the AMI-AS quantifies the extent to which an individual condones, excuses, or justifies husband-to-wife physical aggression under an array of emotionally charged, provokable, or stressful marital scenarios. Comprising 25 distinct situational items, the instrument requires respondents to evaluate the justifiability of physical violence (e.g., slapping or hitting a spouse) using an explicit 7-point continuum ranging from 0 (Unjustifiable) through 1 to 4 (Somewhat Justifiable) to 5 or 6 (Justifiable).

Psychometrically, the AMI-AS demonstrates high internal consistency, with reported Cronbach’s alpha coefficients exceeding .90 in clinical and non-clinical male cohorts. In moderation and mediation modeling, the scale acts as an essential cognitive moderator, explaining how historical exposure to violence—such as experiencing physical abuse in one’s family of origin—interacts with normative cognitive justifications to predict physical and emotional perpetration against female partners. While broader iterations of marriage attitude indexes explore dimensions of intimacy, commitment, and openness, the 25-item aggression-specific index serves as an indispensable tool in intimate partner violence (IPV) research, forensic risk assessments, couples therapy, and domestic violence intervention programs.

2. Keywords

Attitudes About Marriage scale, Attitudes About Marriage Index, Aggression Scale, AMI-AS, intimate partner violence, marital aggression, family of origin violence, domestic violence, cognitive justifications, husband-to-wife aggression, social learning theory, partner abuse.

3. Authors

The instrument was originally conceptualized as an unpublished psychometric battery at the University of Southern California (USC):

• Gayla Margolin, Ph.D. — Professor of Psychology and Pediatrics, Department of Psychology, University of Southern California, Los Angeles, California. Dr. Margolin is an internationally recognized authority on family conflict, marital violence, and the developmental impact of childhood exposure to interparental aggression.
• Laurence S. Foo, Ph.D. — Clinical and Research Psychologist, Family Studies Project, Department of Psychology, University of Southern California. Dr. Foo’s empirical work centered on cognitive appraisals, anger regulation, and typologies of domestic batterers.
• Holly Garcia O’Hearn, Ph.D. — Clinical Psychologist, Department of Psychology, University of Southern California. Dr. O’Hearn collaborated extensively with Dr. Margolin in formalizing the empirical validation, psychometric properties, and moderating framework of marital aggression acceptance in clinical and community samples.

4. Purpose

The primary purpose of the Attitudes About Marriage scale (Aggression Scale) is to delineate, measure, and analyze the individual cognitive schemas that permit, excuse, or sanction physical aggression by a husband toward his wife. While general measures of marital satisfaction capture affective distress or dyadic harmony, they often fail to capture the latent normative beliefs that govern behavioral boundaries during acute conflict. The AMI-AS bridges this gap by operationalizing violence acceptance across granular, contextualized vignettes rather than measuring generic pro-violence philosophies in the abstract.

In clinical settings, domestic violence intervention protocols and court-mandated batterer intervention programs (BIPs) utilize the AMI-AS to establish baseline cognitive justifications for violent behavior. Perpetrators of domestic abuse frequently exhibit cognitive distortions characterized by externalization of blame, minimization of harm, and entitlement to physical retaliation when confronted with perceived spousal infidelity, disrespect, verbal insult, or defiance. By examining an individual’s pattern of endorsement across the 25 items, clinicians can pinpoint idiosyncratic triggers—such as jealousy, narcissistic injury, or loss of emotional control—that lead to physical acts of aggression.

From an empirical research perspective, the scale was created to address a critical theoretical conundrum: the intergenerational transmission of violence. Decades of epidemiological research indicate that experiencing or witnessing violence in one’s family of origin significantly elevates the probability of perpetrating violence in adult romantic unions. However, this statistical association is far from deterministic; many individuals exposed to childhood abuse do not become abusive adults. The AMI-AS provides the necessary operationalization of the cognitive mechanism that moderates this pathway. Individuals who experienced childhood trauma but maintain rigid beliefs that violence is completely unjustifiable (scoring zero across all conditions) display a buffered profile, showing no statistically significant increase in adult marital violence. Conversely, for individuals possessing elevated justification scores, childhood exposure functions as a potent vulnerability factor that manifests in severe adult domestic perpetration.

5. Psychological Construct

The psychological construct assessed by this instrument is the acceptance and perceived justifiability of husband-to-wife physical aggression. Rather than treating aggression acceptance as a unidimensional, monolithic variable, the AMI-AS assesses cognitive justification across multiple distinct psychological and relational contexts. These contextual domains capture the varied rationalizations individuals use to defend violent behavior:

1. Self-Defense and Immediate Threat Rationalization

This domain involves scenarios where physical aggression is deployed under the guise of neutralizing immediate physical danger or defensive retaliation. Exemplified by items such as “She comes at him with a knife” and “In an argument, she hits him first,” these items capture scenarios where respondents invoke self-defense to legitimate violent conduct. While some respondents view high-threat situations as legally or morally defensible, endorsing aggression in lower-level reciprocal conflicts indicates a low threshold for adopting violent conflict resolution strategies.

2. Transgression of Core Moral and Familial Roles

This dimension measures situations where the female partner violates fundamental maternal or moral expectations. Items such as “She physically abuses their child” and “She terrorizes and abuses his pet” evaluate whether respondents view physical violence as an appropriate paternal disciplinary mechanism to rectify unacceptable moral misconduct. Endorsement of violence in these items reflects an authoritarian entitlement to police moral boundaries within the domestic sphere through physical retribution.

3. Sexual Infidelity, Possessiveness, and Sexual Frustration

A central tenet of intimate partner violence models involves male sexual proprietory beliefs. The AMI-AS isolates scenarios surrounding real or perceived sexual boundary violations, including “He catches her in bed with another man,” “He learns that she is having an affair,” “At a party she flirts with another man in front of him and his friends,” “He overhears her talking on the phone with her ex-boyfriend,” and “She refuses to have sex with him.” High justification scores in this domain indicate deep-seated beliefs regarding male sexual entitlement, territorial jealousy, and the justification of physical force to control female sexual autonomy.

4. Narcissistic Injury, Ego Threat, and Social Humiliation

Aggressive behaviors often stem from perceived threats to masculine identity, self-esteem, and social standing. The AMI-AS captures these interpersonal dynamics through items such as “She makes him look like a fool in front of his family and friends,” “She calls him ‘stupid’ over and over again,” “She accuses him of being an incompetent and insensitive human being,” and “She insults his best friend.” Endorsement of physical violence in response to verbal insults or interpersonal humiliation highlights fragile masculine self-worth and an inability to tolerate negative emotional states without resorting to coercive retaliation.

5. Autonomy Disruption, Separation Threats, and Displaced Stress

A final critical cluster of items involves situations where the female partner asserts autonomy, threatens separation, or where the husband faces unrelated external stressors. Items such as “She threatens to move out in the middle of an argument,” “She tells him she should have divorced him a long time ago,” “She refuses to let him enroll in college courses,” “He is upset about losing his job,” and “He is angry because he got a speeding ticket” capture catastrophic breakdowns in impulse control. High ratings on displaced stress items indicate that the respondent perceives the female partner as a legitimate target for venting generalized frustration and emotional dysregulation.

6. Theoretical Framework

The Attitudes About Marriage scale is embedded within the intersection of Social Learning Theory (Bandura, 1977), cognitive models of marital aggression, and feminist-informed ecological frameworks of domestic violence.

Social Learning Theory and Observational Modeling

Albert Bandura’s social learning framework posits that complex aggressive behaviors are acquired through observational learning, modeling, and vicarious reinforcement within primary socializing institutions, especially the family of origin. Children who observe domestic violence or experience direct physical punishment internalize cognitive scripts indicating that physical aggression is an effective, acceptable means of resolving interpersonal conflict and securing compliance. However, modern cognitive formulations expand Bandura’s model: observational learning does not mechanically trigger identical adult behavior; instead, it installs latent cognitive belief systems (attitudes, norms, and outcome expectancies). The AMI-AS quantifies these cognitive scripts, showing that internalized approval of violence serves as the primary psychological bridge linking childhood exposure to adult marital assault.

Cognitive-Behavioral Formulations and Information Processing

In cognitive-behavioral paradigms of intimate violence (such as models proposed by Amy Holtzworth-Munroe and Gayla Margolin), marital aggression occurs via dysfunctional social information processing. When an adverse dyadic event occurs—such as a verbal argument or perceived rejection—an individual accesses cognitive schemas to interpret the spouse’s intentions and select an appropriate response. Men who harbor normative beliefs condoning marital violence evaluate female verbal hostility or autonomy as hostile, unacceptable affronts that warrant punitive action. The AMI-AS indexes the evaluative outcome of this cognitive processing: the decision that hitting or slapping is a “justifiable” behavioral option.

Feminist and Sociocultural Patriarchal Norms

Sociocultural theories of domestic violence argue that intimate partner abuse is sustained by patriarchal structures that grant men authority over female partners. The items embedded within the AMI-AS reflect traditional patriarchal expectations, including demands for unquestioned respect, control over female sexuality, and the subordination of the wife’s desires to the husband’s emotional state. When an individual rates the physical assault of a wife as justifiable because “She calls his mother nasty names” or “She refuses to let him go out for an evening with his friends,” the respondent demonstrates patriarchal entitlement, viewing the female partner as subject to male corrective discipline.

7. Validity

Empirical investigations have established robust validity metrics for the AMI-AS across various clinical, forensic, and community cohorts:

Construct and Structural Validity

Construct validity is evidenced by the scale’s capacity to discriminate between non-violent men, verbally aggressive men, and physically abusive men. Studies examining marital conflict typologies demonstrate that men categorized as physically aggressive against their spouses score significantly higher on the AMI-AS than their non-violent counterparts. Construct validity is further corroborated by strong negative correlations with measures of marital empathy, egalitarian gender role attitudes, and dyadic perspective-taking.

Convergent Validity

The AMI-AS exhibits strong positive correlations with related validated instruments, including:

• The Conflict Tactics Scales (CTS / CTS2; Straus et al.): Higher AMI-AS scores correlate strongly with reported frequencies of mild, moderate, and severe physical assaults inflicted on partners (r values typically ranging between .38 and .54, p < .001).
• The Inventory of Beliefs About Wife Beating (IBWB; Saunders et al.): Strong positive correlations (r > .60) confirm that the AMI-AS aligns closely with broader measures assessing violence tolerance and victim-blaming attributions.
• Measures of Hostility Toward Women and Hypermasculinity: The instrument demonstrates statistically significant associations with scales assessing generalized misogyny and hypermasculine cognitive styles.

Discriminant Validity

Discriminant validity is supported by low-to-negligible correlations with generalized social desirability response sets (e.g., Marlowe-Crowne Social Desirability Scale), non-marital generalized aggression, and general intelligence measures. This indicates that the AMI-AS specifically measures normative attitudes toward intimate partner violence rather than a non-specific tendency to endorse antisocial behavior.

Predictive and Moderating Validity

The predictive and moderating validity of the AMI-AS was empirically demonstrated in the seminal study by O’Hearn and Margolin (2000). Evaluating 47 married men, researchers examined whether the association between experiencing physical abuse in one’s family of origin and perpetrating adult marital violence was moderated by the AMI-AS. The findings revealed that while attitudes condoning aggression were not independently correlated with family of origin abuse, the interaction term between childhood abuse exposure and AMI-AS scores accounted for significant incremental variance in actual physical and emotional aggression inflicted on adult female partners. Specifically, for men who condoned physical aggression (elevated AMI-AS scores), there was a strong, statistically significant correlation between childhood abuse and adult domestic violence perpetration. For men who evaluated marital aggression as unjustifiable (low AMI-AS scores), the correlation between childhood abuse and adult IPV was non-significant, confirming the scale’s predictive validity as a cognitive moderator of intergenerational trauma.

8. Reliability

The AMI-AS exhibits exemplary internal consistency and stability across repeated administrations:

• Internal Consistency: In the initial psychometric evaluations by Margolin and Foo (1992) and subsequent analyses by O’Hearn and Margolin (2000), the 25-item scale yielded an overall Cronbach’s alpha coefficient of α = .93 to .95, indicating excellent homogeneity among the items. Item-total correlations across the 25 items range from .42 to .78, confirming that each specific situational vignette contributes meaningfully to the overarching construct of marital aggression justifiability.
• Split-Half Reliability: Guttman split-half and Spearman-Brown coefficients have consistently exceeded .90 across community and clinical samples.
• Test-Retest Stability: Over an 8-to-12-week test-retest interval in non-treatment community samples, the instrument demonstrated high temporal stability, with test-retest reliability coefficients ranging between r = .82 and r = .87 (p < .001). This stability confirms that the instrument measures stable cognitive schemas rather than transient mood states.

9. Factor Analysis

Both exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) have been employed to evaluate the latent factor structure of the 25-item AMI-AS:

Exploratory Factor Analysis (EFA)

Principal Axis Factoring (PAF) with Promax oblique rotation generally reveals a strong general factor accounting for approximately 42% to 48% of the total variance, confirming that the scale can be interpreted as a unidimensional composite score of marital violence condonation. However, multi-factor extraction methods based on eigenvalues > 1.0 and scree plot inspections consistently yield a three-to-four factor multidimensional structure:

• Factor 1: Provocation by Infidelity and Threat to Masculine Esteem (Items 4, 5, 9, 12, 13, 15, 16, 17, 20, 24). Factor loadings range from .56 to .82. This factor captures cognitive justifications driven by jealousy, insults, and status threats.
• Factor 2: Displaced Stress, Minor Conflict, and Non-Compliance (Items 7, 8, 11, 14, 18, 19, 21, 22, 23, 25). Factor loadings range from .48 to .76. This factor reflects severe self-regulatory failure, where violence is excused by mundane frustrations, refusal of sexual intercourse, or independent spouse choices.
• Factor 3: Severe Physical Danger and Moral Transgression (Items 1, 2, 3, 6, 10). Factor loadings range from .52 to .81. This factor isolates scenarios involving weapon threats, child maltreatment, or severe physical provocation.

Confirmatory Factor Analysis (CFA) and Fit Indices

In structural modeling, while a single-factor model provides acceptable fit when accounting for correlated error variances between closely phrased items, a hierarchical model featuring three first-order factors loading onto a higher-order General Violence Justification construct provides optimal goodness-of-fit:

• Comparative Fit Index (CFI): .94
• Tucker-Lewis Index (TLI): .93
• Root Mean Square Error of Approximation (RMSEA): .058 (90% CI [.045, .069])
• Standardized Root Mean Square Residual (SRMR): .049

10. Instrument / Measurement Tool

• Test Type: Psychometric self-report rating scale / Contextual vignette inventory.
• Format: Pen-and-paper or digital administration consisting of 25 structured conditional statements preceded by a standardized prompt: “A husband slaps or hits his wife if . . .”
• Target Population: Adult men, couples presenting for relational counseling, and forensic domestic violence perpetrator groups (adaptable for adult women in studies examining bidirectional aggression).
• Administration Time: Approximately 7 to 10 minutes.
• Response Scale: 7-point Likert-type continuum anchored as follows:
  • 0: Unjustifiable
  • 1 – 4: Somewhat Justifiable
  • 5 – 6: Justifiable
• Scoring Rules:
  • Total Score Calculation: Items are summed across all 25 items to generate an overall aggregate score ranging from 0 to 150. Alternatively, researchers frequently compute a mean composite score by dividing the sum by 25 (ranging from 0.0 to 6.0).
  • Reverse Scored Items: None. All 25 items are keyed in the direction of violence acceptance. Higher scores indicate greater cognitive tolerance and justifiability of physical aggression.
  • Clinical Cut-offs: In non-violent normative community samples, scores are highly positively skewed, with most men scoring 0 or near 0 on the vast majority of items (scores typically < 15). Scores exceeding 25 indicate clinically elevated tolerance for intimate partner violence, warranting comprehensive risk assessment and targeted therapeutic intervention.

11. Permissions & Fee and Test Year

• Year of Development: 1992 (Initial development by Gayla Margolin and Laurence S. Foo); empirically published in peer-reviewed form in 2000 by Holly Garcia O’Hearn and Gayla Margolin.
• Copyright & Permissions: The instrument is copyrighted by the original authors and the original publisher (Springer / Cognitive Therapy and Research). It was published as an academic research instrument in Cognitive Therapy and Research (2000), Vol. 24, No. 2, pp. 159–174.
• Fee Structure: The instrument is accessible without commercial licensing fees for non-profit academic research, scientific inquiry, and educational purposes, provided formal academic attribution is maintained. Commercial applications, automated platform integrations, or for-profit clinical distributions require explicit written authorization from the primary authors and copyright holders.
• Contact and Correspondence: Historical correspondence regarding the scale was directed to Dr. Gayla Margolin, Department of Psychology, University of Southern California, Los Angeles, CA 90089-1061.

12. References

• Bandura, A. (1977). Social Learning Theory. Prentice-Hall.
• Holtzworth-Munroe, A., & Stuart, G. L. (1994). Typologies of male batterers: Of mice and men. Psychological Bulletin, 116(3), 476–497. https://doi.org/10.1037/0033-2909.116.3.476
• Margolin, G., & Foo, L. (1992). Attitudes About Marriage Index. Unpublished psychometric instrument, Department of Psychology, University of Southern California, Los Angeles, CA.
• Margolin, G., & Gordis, E. B. (2000). The effects of family and community violence on children. Annual Review of Psychology, 51(1), 445–479. https://doi.org/10.1146/annurev.psych.51.1.445
• O’Hearn, H. G., & Margolin, G. (2000). Men’s attitudes condoning marital aggression: A moderator between family of origin abuse and aggression against female partners. Cognitive Therapy and Research, 24(2), 159–174. https://doi.org/10.1023/A:1005404505335
• Saunders, D. G., Lynch, A. B., Grayson, M., & Linz, D. (1987). The Inventory of Beliefs About Wife Beating: The construction and initial validation of a measure of beliefs and attitudes. Violence and Victims, 2(1), 39–57. https://doi.org/10.1891/0886-6708.2.1.39
• Straus, M. A., Hamby, S. L., Boney-McCoy, S., & Sugarman, D. B. (1996). The Revised Conflict Tactics Scales (CTS2): Development and preliminary psychometric data. Journal of Family Issues, 17(3), 283–316. https://doi.org/10.1177/019251396017003001

13. Items of the Scale