The HMI measures the "macho personality constellation": an exaggerated masculine style expressed as calloused sexual attitudes toward women, the belief that violence is manly, and the experience of danger as exciting. Respondents choose between paired statements rather than rating agreement.
Published by Mosher and Sirkin in 1984 with US college men, it is the earliest widely used hypermasculinity measure and remains the historical benchmark in this literature.
Its evidence base links scores to self-reported sexual aggression, laboratory aggression toward women, and risk behaviors such as dangerous driving and substance use.
The item content reflects its era and is considered offensive by later researchers; newer value-based alternatives exist (notably the Hypermasculine Values Questionnaire), and the HMI is a research instrument with no norms or clinical use.
At a Glance
Items
30 forced-choice pairs (choose one of two statements per item)
Administration time
Approx. 10 minutes (no published estimate)
Response format
Binary forced choice between a hypermasculine and a contrasting statement
Subscales
Calloused Sex Attitudes toward Women (10), Violence as Manly (10), Danger as Exciting (10)
Validated populations
Adult men; derived with US college samples
License
No published license terms; recorded in PsycTESTS (DOI 10.1037/t08609-000); verify permissions before use
Original citation
Mosher & Sirkin (1984), Journal of Research in Personality
Introduction
The Hypermasculinity Inventory (HMI) is a 30-item forced-choice measure of what Donald Mosher and Mark Sirkin (1984) called the “macho personality constellation”: a personality-level pattern of exaggerated masculinity in men. Each item presents two statements, one expressing a hypermasculine position and one a contrasting position, and the respondent chooses whichever is closer to his own view. The instrument was developed and validated with US college men and became the standard hypermasculinity measure for the following two decades.
Understanding the Macho Personality Constellation
Mosher and Sirkin (1984) conceptualized hypermasculinity as three interlocking components, each measured by a 10-item subscale: calloused sex attitudes toward women, treating sex as conquest and women’s consent as unimportant; violence as manly, valuing physical aggression as an expression of manhood; and danger as exciting, seeking risk as validation. The items are forced-choice pairs built from statements heard in all-male peer groups, reduced from an initial pool of 221 items in two item-analysis stages (samples of 60, then 135 college men); the final inventory keeps the 10 most discriminating items per subscale (Mosher & Sirkin, 1984).
Theoretical Foundation
The HMI treats hypermasculinity as a personality constellation: a stable, trait-like style expressed across sexuality, aggression, and risk. This differs from later approaches that measure hypermasculinity as a set of endorsed values; the Hypermasculine Values Questionnaire (Archer, 2010) was explicitly derived as a broader, psychometrically updated alternative, and its author noted the advantage of avoiding the HMI’s more offensive item content. The two instruments correlate substantially but are not interchangeable, and findings from one should not be attributed to the other (Archer, 2010).
📜 Key insight: The HMI is the field’s original hypermasculinity measure; its three-part structure (calloused sex, violence, danger) defined how the construct was studied for a generation, even as its 1980s item content aged poorly.
Key Features
Assessment Characteristics
30 forced-choice item pairs, scored 0 or 1 per item
Three 10-item subscales plus a total score
Adult male populations; derived with US college samples
Forced-choice format, an older design intended to structure the choice between contrasting positions
Dimensions Assessed
Calloused Sex Attitudes toward Women – sexual conquest orientation and disregard for consent
Violence as Manly – physical aggression valued as masculine
Danger as Exciting – risk-seeking as validation of manhood
Versions & Adaptations
No short forms or revisions are reported in the development paper (Mosher & Sirkin, 1984), and none were identified during verification (July 2026).
Research Applications
Sexual aggression research – the calloused-sex subscale’s association with self-reported sexual aggression (Mosher & Anderson, 1986)
Aggression research – laboratory aggression paradigms (Parrott & Zeichner, 2003)
Each of the 30 items presents two paired statements; the respondent selects the one closer to his view. Choosing the hypermasculine alternative scores 1, the contrasting alternative 0.
Illustrative Item Pairs
The full item set is printed in Mosher & Sirkin (1984). Two pairs illustrate the format:
“I like to drive fast, right on the edge of danger” vs “I like to drive safely avoiding all possible risks”
“Any man who is a man needs to have sex regularly” vs “Any man who is a man can do without sex”
Scoring Procedure
Subscale scores = count of hypermasculine choices within each 10-item subscale (range 0-10).
Total score = count across all 30 items (range 0-30).
Higher scores indicate a stronger hypermasculine orientation.
Interpretation
There are no published norms, bands, or cutoffs. For descriptive context only: the derivation sample of 135 college men averaged 11.03 of 30 hypermasculine choices (SD = 6.79) (Mosher & Sirkin, 1984). Interpretation should be relative, within study samples or between groups. No score on the HMI carries clinical meaning.
Research Evidence and Psychometric Properties
Reliability Evidence
Internal consistency (total): α = .89 in the derivation sample of 135 college men, where the total score averaged 11.03 (SD = 6.79) (Mosher & Sirkin, 1984)
Subscale internal consistency: Calloused Sex α = .79, Violence α = .79, Danger α = .71 in the original sample (Mosher & Sirkin, 1984); .64, .78, and .68 respectively in an independent UK sample of 200 men (Archer, 2010, Study 3); the binary item format constrains alpha
Subscale intercorrelations: r = .58-.63 in the original sample (Mosher & Sirkin, 1984); r = .42-.53 in Archer’s (2010) UK sample
Factor structure: principal-axes analysis of the 30 items showed one predominant factor (24% of variance, all loadings .33-.60), interpreted by the authors as a single macho personality pattern underlying the three content areas (Mosher & Sirkin, 1984)
Validity Evidence
Original validity correlates (N = 133-134 college men): aggressive behavior after drinking r = .68; dangerous driving after drinking r = .47; self-reported high-school delinquency total r = .38, with the fighting items at r = .47; drug-use frequency r = .20-.31 (alcohol, stimulants, depressants, marijuana, hashish), driven mostly by the Danger subscale (Mosher & Sirkin, 1984)
Sexual aggression: the calloused-sex subscale was associated with self-reported sexual aggression history in 175 college men (Mosher & Anderson, 1986)
Laboratory aggression: men high in hypermasculinity delivered electric shocks to a female opponent earlier and at higher intensities (Parrott & Zeichner, 2003)
Convergence with newer measures: subscale correlations with the Hypermasculine Values Questionnaire were .51 (calloused sex), .56 (violence), and .34 (danger); with trait physical aggression, .48, .56, and .45 respectively (Archer, 2010, Study 3)
Usage Guidelines and Applications
Primary Research Applications
Sexual aggression and violence research where continuity with the classic literature matters
Benchmark or comparison measure alongside newer hypermasculinity instruments
Research Design Considerations
Male samples: derived and validated with men; no evidence base for other populations
Dated, explicit item content: several items are sexually explicit or, by later standards, offensive (Archer, 2010, noted the absence of such items as an advantage of the HVQ); plan ethics review and participant debriefing accordingly
Forced-choice format: yields coarser (binary) item information than Likert formats and constrains reliability estimates
Era effects: item idioms and social assumptions are those of early-1980s US college culture; interpret cross-era comparisons with care
Limitations and Cautions
Not a clinical instrument: no cutoffs, no norms, no diagnostic meaning
Superseded for many purposes: value-based measures with modern derivations exist (Archer, 2010)
Attitude-behavior gap: scores measure a self-described style, not observed behavior
Copyright and Usage Responsibility: Check that you have the proper rights and permissions to use this assessment tool in your research. This may include purchasing appropriate licenses, obtaining permissions from authors/copyright holders, or ensuring your usage falls within fair use guidelines.
Proper Attribution: When using or referencing this scale, cite the original development:
Mosher, D. L., & Sirkin, M. (1984). Measuring a macho personality constellation. Journal of Research in Personality, 18(2), 150-163. https://doi.org/10.1016/0092-6566(84)90026-6
Mosher, D. L., & Sirkin, M. (1984). Measuring a macho personality constellation. Journal of Research in Personality, 18(2), 150-163. https://doi.org/10.1016/0092-6566(84)90026-6
Validation and Applications:
Mosher, D. L., & Anderson, R. D. (1986). Macho personality, sexual aggression, and reactions to guided imagery of realistic rape. Journal of Research in Personality, 20(1), 77-94. https://doi.org/10.1016/0092-6566(86)90111-X
Parrott, D. J., & Zeichner, A. (2003). Effects of hypermasculinity on physical aggression against women. Psychology of Men & Masculinity, 4(1), 70-78. https://doi.org/10.1037/1524-9220.4.1.70
Successor Instrument:
Archer, J. (2010). Derivation and assessment of a hypermasculine values questionnaire. British Journal of Social Psychology, 49(3), 525-551. https://doi.org/10.1348/014466609X471525
Related Assessments:BSRI: Bem Sex Role Inventory | HVQ: Hypermasculine Values Questionnaire (link when its page goes live)
Frequently Asked Questions
What does the HMI measure?
A "macho" personality style in men, expressed as calloused sexual attitudes toward women, valuing violence as manly, and finding danger exciting. It yields three subscale scores and a total.
Is the HMI free to use?
No license terms or fees are published; the items appear in the original journal article and the instrument is recorded in PsycTESTS. Verify permissions with the rights holder for anything beyond ordinary cited research use.
How is it scored?
Each of 30 items is a forced choice between two statements; hypermasculine choices count 1. Subscales range 0-10, the total 0-30. There are no norms or cutoffs.
How does the HMI differ from the HVQ and SHVQ?
The HMI (1984) is a forced-choice personality-style measure with three subscales. The HVQ and its short form (Archer, 2010) are Likert-rated, unidimensional measures of endorsed values, derived from a broader UK sample. They correlate substantially (r = .34-.56 across HMI subscales) but are distinct instruments.
How reliable is it?
Total-score internal consistency was α = .89, with subscale alphas of .71-.79, in the original college sample (Mosher & Sirkin, 1984); in a later independent UK sample subscale alphas were .64-.78 (Archer, 2010). The binary item format constrains alpha.
Can it diagnose anything?
No. It is a research measure with no clinical cutoffs, norms, or diagnostic validity.
