Revised Mosher Guilt Inventory

Abstract

The Revised Mosher Guilt Inventory (RMGI) represents a landmark instrument in the psychometric assessment of dispositional guilt within personality, clinical, and human sexuality research. Originating from Donald L. Mosher’s foundational work with sentence completion techniques in the 1960s, the inventory transitioned across multiple psychometric paradigms—from projective weights and true-false formats to forced-choice dyads, and ultimately to its current 114-item limited-comparison format utilizing a 7-point Likert response continuum (ranging from 0 = not at all true of (for) me to 6 = extremely true of (for) me). Anchored theoretically in Silvan Tomkins’ Affect and Script Theory, the RMGI operationalizes guilt not merely as a fleeting emotional state, but as a stabilized cognitive-affective script—a systematic psychological apparatus comprising rules for predicting, interpreting, responding to, and evaluating scenes that trigger amplified negative affect. The scale yields three robust, discriminable subscales: Sex-Guilt (50 items), Hostility-Guilt (42 items), and Guilty-Conscience (22 items, formerly designated Morality-Conscience). Across decades of empirical replication, the RMGI and its antecedent forms have demonstrated exceptional construct, convergent, and discriminant validity, exhibiting internal consistency reliability coefficients (Cronbach’s alpha and split-half) consistently hovering around .90. Its subscales have proven uniquely predictive of sexual dysfunction, contraceptive compliance, sexual risk-taking, the cognitive inhibition of erotic fantasy, aggressive impulse control, interpersonal conflict styles, and moral scrupulosity. This article provides an exhaustive examination of the RMGI’s developmental trajectory, theoretical architecture, psychometric parameters, factor structure, scoring algorithms, and full item corpus.

Keywords

Revised Mosher Guilt Inventory, RMGI, Donald L. Mosher, Sex-Guilt, Hostility-Guilt, Guilty-Conscience, Script Theory, Affect Theory, Silvan Tomkins, Personality Assessment, Psychometrics, Moral Cognition, Sexual Inhibition, Trait Guilt

Authors

The RMGI was conceived and standardized by Donald L. Mosher, Ph.D., Emeritus Professor of Psychology at the University of Connecticut, Storrs, Connecticut, USA. Mosher stands as one of the preeminent figures in twentieth-century clinical psychology and personality assessment, particularly recognized for his pioneer work in psychosexual research, affect theory, and the measurement of moral and aggressive emotional scripts. His academic lineage includes doctoral training at The Ohio State University, where he completed his seminal 1961 dissertation on the measurement of guilt via projective sentence completion methods under the influence of early social learning and cognitive-personality paradigms. Throughout a prolific career spanning over four decades, Mosher served as an editor, methodological innovator, and frequent contributor to flagship behavioral science journals, including the Journal of Consulting and Clinical Psychology, the Journal of Personality and Social Psychology, and The Journal of Sex Research, fundamentally shaping how clinical psychologists operationalize personality dispositions in relationship to sexual and aggressive behaviors.

Purpose

The primary clinical and empirical purpose of the Revised Mosher Guilt Inventory is to quantify individual differences in generalized dispositional guilt, differentiated across three distinct behavioral-affective domains: sexual expression, hostile-aggressive actions, and generalized moral conscience. Historically, the conceptualization of guilt suffered from over-generalization, frequently conflated with state anxiety, generalized neuroticism, or broad superego strength. Mosher sought to build an empirically grounded, multidimensional inventory capable of capturing guilt as an active, self-regulating personality disposition that acts as an internalized barrier against socially prohibited impulses.

In empirical research, the RMGI was engineered to overcome major methodological limitations inherent in early forced-choice and true-false inventories. During the sexual revolution and changing socio-cultural mores of the late 1960s and 1970s, the baseline endorsement of traditional guilt items shifted radically; mean guilt scores declined, and response distributions became increasingly truncated. Furthermore, respondents frequently reported cognitive frustration with dichotomous forced-choice pairings, where neither or both options felt personally applicable. The RMGI resolved these constraints through its innovative limited-comparison format, allowing respondents to assign continuous ratings to paired completions derived from identical sentence stems, thereby restoring behavioral variance, capturing nuanced intermediate endorsements, and maintaining sensitivity across modern demographic cohorts.

Clinically, the inventory provides diagnostic insight into dysfunctional internalizations. Within sex therapy and clinical sexology, the Sex-Guilt subscale reliably predicts psychosexual distress, including hypoactive sexual desire, erectile dysfunction, vaginismus, anorgasmia, and acute distress surrounding masturbation or erotic fantasies. In behavioral medicine and public health, high Sex-Guilt scores consistently correlate with contraceptive negligence—paradoxically, individuals burdened with intense sex guilt are less likely to plan for, obtain, or consistently utilize contraception or barrier protection because acknowledging contraceptive preparedness requires acknowledging and accepting their sexual intentions. Within forensic and general clinical domains, the Hostility-Guilt subscale elucidates maladaptive patterns of emotional regulation, distinguishing between over-controlled individuals prone to somatic symptoms and explosive episodic outbursts versus under-controlled individuals who lack internalized hostility restraints. Finally, the Guilty-Conscience subscale informs the clinical assessment of obsessive-compulsive scrupulosity, depressive self-reproach, and generalized moral perfectionism.

Psychological Construct

Guilt, as operationalized within the RMGI, represents a generalized disposition to experience self-condemnation, subjective discomfort, remorse, and behavioral inhibition following the real or anticipated violation of internalized standards of conduct. Mosher explicitly conceptualized guilt not as a global undifferentiated monolith, but as a domain-specific personality construct comprised of three correlated yet functionally distinct dimensions:

1. Sex-Guilt

Sex-Guilt represents an internalized disposition to experience self-punishment, remorse, anxiety, and shame in response to sexual arousal, thoughts, desires, or explicit behaviors that violate internalized moral standards regarding sexuality. This construct is manifested across an array of cognitive and behavioral targets, including autoeroticism (masturbation), premarital intercourse, petting, exposure to erotic or “obscene” literature, sexual humor (“dirty jokes”), atypical sexual practices, and spontaneous sexual fantasies or nocturnal dreams. Individuals high in sex guilt perceive sexual impulses as intrinsically polluting, dangerous, or disruptive to moral integrity. Consequently, the construct involves powerful cognitive-affective defenses, such as perceptual defense against sexual stimuli, active suppression of erotic imagery, and avoidance of sexual communication with intimate partners. Behaviorally, sex guilt operates as an internal inhibitor that blunts hedonic pleasure, transforming interest-excitement into subjective revulsion, distress, or depressive self-blame.

2. Hostility-Guilt

Hostility-Guilt reflects an internalized standard against the experience, expression, or interpersonal manifestation of anger, rage, and aggressive behavior. The construct evaluates the psychological aftermath of interpersonal confrontations, arguments, physical fights, profanity, and internal hostile urges. Individuals possessing elevated hostility-guilt experience acute dysphoria, self-disgust, anxiety, and a compulsive urge to apologize or make restitution following even minor disagreements or assertive expressions of self-interest. Rather than experiencing cathartic tension reduction following an argument or competitive victory, individuals high in hostility-guilt report feeling “mean,” “jittery,” “depressed and disgusted,” or “ridiculous and sorry.” In extreme manifestations, hostility-guilt prompts the complete internalization of aggressive impulses, wherein anger that cannot be directed outward is retroflexed inward, leading to self-punitive cognitions, depressive rumination, and somatic distress.

3. Guilty-Conscience

Originally designated as Morality-Conscience in the early iterations of the instrument, the Guilty-Conscience subscale captures an overarching disposition toward self-monitoring, moral scrupulosity, and self-administered punishment for perceived ethical transgressions and general personal failures. This construct reflects an internalized standard of perfectionism, acute vulnerability to self-hatred for forbidden thoughts, and the subjective conviction that a guilty conscience is “worse than a sickness.” It assesses whether an individual responds to rule-breaking with genuine remorse and an internal expectation of retribution, as opposed to an instrumental, Machiavellian orientation characterized solely by the fear of external detection. Following rigorous psychometric revisions, items within this subscale that shared excessive empirical overlap with the sexual domain were eliminated, leaving a distilled measure of core moral scrupulosity, existential regret, and punitive self-appraisal.

Theoretical Framework

The Revised Mosher Guilt Inventory is rooted in the integrative theoretical architecture of Silvan S. Tomkins’ Affect Theory and Script Theory (1979). Tomkins posited that the primary human motivational system is biological affect, comprising nine primary, biologically hardwired discrete affects: positive affects (interest-excitement, enjoyment-joy), neutral affect (surprise-startle), and negative affects (distress-anguish, anger-rage, fear-terror, shame-humiliation, disgust, and dissmell). As an individual navigates the lifespan, affect is triggered by varying patterns of neural stimulation and subsequently synthesized with cognitive schemas, memories, and socio-cultural learning through repetitive life experiences termed scenes.

When multiple scenes sharing similar affective and cognitive characteristics recur across life, they undergo psychological magnification. Through magnification, the individual co-assembles these scenes into a unified script—defined by Tomkins as a formalized cognitive-affective program composed of rules for the interpretation, prediction, production, control, and evaluation of co-assembled scenes. In Mosher’s framework, guilt is precisely understood as an amplified, scripted affective system. Guilt is not a raw, standalone biological affect; rather, it is a complex, cognitively elaborated script predominantly rooted in the auxiliary magnification of the discrete affect of shame-humiliation, interacting dialectically with other primary affects:

• The Sex-Guilt Script: Psychologically magnified in developmental scenes wherein the child’s primary positive affects of interest-excitement and enjoyment-joy (aroused by tactile bodily exploration, sexual curiosity, or genital responsiveness) are abruptly interrupted by parental disapproval, societal prohibition, or moral condemnation. This sudden, painful interruption of positive affect automatically triggers the biological affect of shame. Over repeated developmental episodes, the child internalizes cognitive rules equating sexual arousal with moral corruption. In adulthood, the emergence of sexual excitement into consciousness automatically recruits the script, triggering shame, fear, and moral self-reproach, which collectively manifest as subjective guilt.
• The Hostility-Guilt Script: Psychologically magnified in childhood scenes where the biological affect of anger-rage and behavioral motor aggression are met with parental counter-aggression, withdrawal of love, or intense moral scolding. The script establishes rules stipulating that aggressive affect and competitive triumph are inherently sinful and dangerous. As a consequence, adult interpersonal friction or anger cues the immediate co-assembly of scenes of punishment, transforming physiological arousal into dysphoric self-loathing.
• The Conscience Script: Operates as an overarching nuclear script governing moral accountability, self-surveillance, and the regulation of behavioral temptations. When personal behavior deviates from internalized ego ideals, the conscience script generates anticipatory or post-hoc self-punitive sequences to restore moral equilibrium.

By framing the inventory around scripted affective sequences, Mosher successfully bridged dynamic psychoanalytic intuitions regarding the superego with the empirical, operational precision of cognitive-affective personality science.

Validity

The construct validity of the Mosher Guilt Inventories has been subjected to extensive empirical verification spanning hundreds of independent investigations. Early construct validation established convergent and discriminant validity utilizing Campbell and Fiske’s classic multitrait-multimethod (MTMM) matrix approach (Mosher, 1966, 1968). By contrasting the three guilt subscales across diverse measurement methodologies—including incomplete sentence blanks, forced-choice self-report inventories, and true-false formats—Mosher established that the subscales shared robust convergent validities across methods while demonstrating marked discriminant validity from one another and from generalized response sets, such as social desirability and acquiescence.

Construct and Convergent Validity in Sexuality Research

The Sex-Guilt subscale has generated an exceptionally cohesive nomological network within human sexuality research. In laboratory investigations evaluating subjective versus physiological sexual arousal (Mosher & O’Grady, 1979), high sex-guilt females and males exposed to explicit erotica exhibited significant attenuation of subjective sexual arousal, despite exhibiting measurable physiological signs of genital vasocongestion. High sex-guilt individuals routinely engaged in perceptual defense, demonstrating elevated recognition thresholds for tachistoscopically presented erotic words while exhibiting normal thresholds for neutral words. Green and Mosher (1985) demonstrated through path analysis and structural equation modeling that sex guilt exerts a direct, negative causal influence on subjective sexual arousal, mediated primarily by the conscious suppression of erotic fantasy and the intrusion of negative affective appraisals.

In behavioral and reproductive domains, Mosher and Vonderheide (1985) demonstrated that sex guilt and masturbatory guilt significantly impaired contraceptive preparedness among college women. Women with elevated sex guilt exhibited severe deficits in sexual knowledge, held highly negative attitudes toward obtaining contraception, and engaged disproportionately in unprotected intercourse. Because using contraception mandates an intentional, premeditated acknowledgment of impending sexual behavior, the high sex-guilt individual utilizes denial as a defense mechanism, ironically heightening real-world vulnerability to unplanned pregnancy and sexually transmitted infections. Furthermore, research by Kelley (1985) confirmed that high sex guilt correlates robustly with sexual authoritarianism, punitive attitudes toward sexual minorities, and intense revulsion toward visual depictions of autoeroticism.

Discriminant and Criterion Validity of Hostility and Conscience

The Hostility-Guilt subscale has consistently exhibited discriminant utility in behavioral conflict experiments. In laboratory aggression paradigms (e.g., the Buss teacher-learner aggression machine), individuals scoring high in hostility-guilt systematically administer lower shock intensities to provocative confederates, exhibit prolonged latency prior to retaliation, and express substantial post-experimental remorse. Conversely, when provoked individuals with high hostility-guilt are forced to administer punishment, they demonstrate elevated cardiovascular reactivity and acute elevations on state depression scales. The Guilty-Conscience subscale uniquely predicts resistance to cheating under conditions where detection is impossible, scrupulosity in self-administered tasks, and negative self-esteem indices characterized by moral self-deprecation.

Reliability

Across its developmental history, the Mosher Guilt Inventory series has displayed superior psychometric stability. In the original True-False and Forced-Choice iterations (Mosher, 1966, 1968), internal consistency was thoroughly documented across diverse collegiate and community populations. Split-half reliability coefficients, corrected via the Spearman-Brown prophecy formula, yielded estimates ranging from .88 to .94 across both male and female cohorts. Subsequent internal consistency evaluations utilizing Cronbach’s alpha in large psychosexual cohorts (e.g., Mosher & Vonderheide, 1985; Green & Mosher, 1985) repeatedly demonstrated alpha coefficients exceeding .90 for the Sex-Guilt subscale, and falling between .84 and .91 for Hostility-Guilt and Guilty-Conscience.

During the standardization of the 114-item Revised Mosher Guilt Inventory in its limited-comparison format, an exhaustive item analysis was performed across a sample of 187 male and 221 female undergraduates at the University of Connecticut. For the retained 114 items, corrected item-total correlations with their designated subscales ranged from .32 to .62, with a high median correlation of .46. Critically, to preserve strict discriminant validity between dimensions, 90% of the selected items demonstrated a statistically significant higher correlation with their parent subscale than with the alternative subscale totals. Test-retest reliability across a four-to-six-week temporal window has consistently yielded stability coefficients between .82 and .89, demonstrating that the RMGI effectively measures an enduring, trait-like affective script rather than momentary emotional fluctuations.

Factor Analysis

The factor structure of the Mosher guilt instruments has been rigorously explored through both exploratory factor analysis (EFA) and confirmatory factor analysis (CFA). In an influential structural investigation, O’Grady and Janda (1979) conducted a principal components factor analysis with varimax rotation on the forced-choice inventory. Their findings decisively corroborated Mosher’s tripartite conceptualization, revealing three clear, orthogonal-to-oblique primary factors accounting for the majority of the common variance, cleanly aligning with Sex-Guilt, Hostility-Guilt, and Morality-Conscience. Importantly, O’Grady and Janda also evaluated the scoring system, demonstrating that an unweighted binary scoring procedure (assigning 1 or 0 to guilty versus non-guilty alternatives) correlated .99 with Mosher’s original clinical weighting scheme, permanently establishing that differential item weights were empirically redundant.

In refining the instrument into the Revised Mosher Guilt Inventory, Mosher submitted the non-overlapping items from both gender-specific forms (233 true-false items and 151 forced-choice items) to factor-analytic and item-discrimination filtering. A central structural discovery during this revision was that several items on the original Morality-Conscience dimension exhibited unacceptable secondary factor cross-loadings onto the Sex-Guilt factor. Cultural shifts had fused general notions of “morality” almost indistinguishably with “sexual purity” for many respondents. To restore structural clarity and eliminate factorial contamination, all morality items demonstrating salient cross-loadings with Sex-Guilt were eliminated. The surviving 22 items cleanly defined an autonomous construct of existential moral scrupulosity, prompting Mosher to rename the subscale Guilty-Conscience. Confirmatory models of the resulting 114-item matrix support a three-factor oblique model where Sex-Guilt, Hostility-Guilt, and Guilty-Conscience are moderately intercorrelated (typically between .25 and .45), demonstrating shared participation in a higher-order generalized guilt propensity while maintaining distinct lower-order structural independence.

Instrument / Measurement Tool

The operational characteristics, administration parameters, and scoring framework of the Revised Mosher Guilt Inventory are structured as follows:

• Test Type: Multi-dimensional self-report personality inventory utilizing a limited-comparison paired-response format.
• Target Population: Adults and adolescents (approximate reading age of 14 years and older).
• Total Item Count: 114 items (organized as 57 paired completions stemming from shared sentence roots).
• Subscale Distribution:
  • Sex-Guilt: 50 items.
  • Hostility-Guilt: 42 items.
  • Guilty-Conscience: 22 items.
• Administration Time: Approximately 20 minutes for the complete 114-item battery. Subscales may be administered independently without compromising psychometric validity.
• Response Format: Continuous 7-point Likert-type trueness scale ranging from 0 to 6:
  • 0 = Not at all true of (for) me
  • 1 = Barely true of me
  • 2 = Slightly true of me
  • 3 = Moderately true of me
  • 4 = Substantially true of me
  • 5 = Very true of me
  • 6 = Extremely true of (for) me
• The Limited-Comparison Paradigm: Items are presented in sets of two distinct completions appended to a single incomplete sentence stem. Respondents rate both completions independently on the 0–6 scale. This format encourages respondents to compare the relative subjective intensity of the two potential resolutions, capturing fine-grained intra-individual variance.
• Scoring and Keying Procedures:
  • Each subscale score is computed by summing the direct ratings of “guilty” items and adding the reversed ratings of “non-guilty” items.
  • For non-guilty items, reverse scoring is executed via the formula: Reversed Score = 6 - Raw Rating.
  • Higher aggregate subscale scores reflect a more pervasive, rigid, and psychologically magnified guilty script.
  • Sex-Guilt Subscale Items: 5, 6, 7, 8, 11, 12, 13, 14, 15, 16, 17, 18, 25, 26, 31, 32, 35, 36, 41, 42, 51, 52, 53, 54, 61, 62, 63, 64, 67, 68, 71, 72, 75, 76, 81, 82, 83, 84, 87, 88, 93, 94, 101, 102, 103, 104, 107, 108, 111, 112.
  • Hostility-Guilt Subscale Items: 3, 4, 19, 20, 21, 22, 23, 24, 29, 30, 33, 34, 37, 38, 39, 40, 43, 44, 45, 46, 55, 56, 69, 70, 77, 78, 79, 80, 85, 86, 91, 92, 95, 96, 97, 98, 99, 100, 109, 110, 113, 114.
  • Guilty-Conscience Subscale Items: 1, 2, 9, 10, 27, 28, 47, 48, 49, 50, 57, 58, 59, 60, 65, 66, 73, 74, 89, 90, 105, 106.

Permissions & Fee and Test Year

The conceptual genesis of the Mosher Guilt Inventories dates to Donald L. Mosher’s doctoral research at The Ohio State University in 1961, followed by the formal publication of standardized male and female psychometric inventories in 1966 and 1968. The standardized Revised Mosher Guilt Inventory (RMGI), featuring the 114-item limited-comparison format, was consolidated during the 1980s and formally published in landmark handbooks of psychosexual measurement, most notably within Davis, Yarber, Bauserman, Schreer, and Davis’s foundational compendium, Handbook of Sexuality-Related Measures.

Dr. Donald L. Mosher placed the instrument into the public academic domain for non-commercial educational, clinical, and scholarly research purposes. Researchers and qualified behavioral healthcare professionals may reproduce, administer, and analyze the RMGI without paying licensing fees or royalties, provided that proper scholarly citation and attribution are maintained. For commercial software deployment, proprietary digital platforms, or profit-generating endeavors, licensing arrangements and intellectual property permissions must be clarified through the author’s estate or the copyright holding academic publishers.

References

Green, S. E., & Mosher, D. L. (1985). A causal model of sexual arousal to erotic fantasies. The Journal of Sex Research, 21(1), 1–23. https://doi.org/10.1080/00224498509551242

Kelley, K. (1985). Sex, sex guilt, and authoritarianism: Differences in responses to explicit heterosexual and masturbatory slides. The Journal of Sex Research, 21(1), 68–85. https://doi.org/10.1080/00224498509551246

Mosher, D. L. (1961). The development and validation of a sentence completion measure of guilt (Doctoral dissertation). The Ohio State University, Columbus, OH.

Mosher, D. L. (1966). The development and multitrait-multimethod matrix analysis of three measures of three aspects of guilt. Journal of Consulting Psychology, 30(1), 25–29. https://doi.org/10.1037/h0023030

Mosher, D. L. (1968). Measurement of guilt in females by self-report inventories. Journal of Consulting and Clinical Psychology, 32(6), 690–695. https://doi.org/10.1037/h0026577

Mosher, D. L. (1979). The meaning and measurement of guilt. In C. E. Izard (Ed.), Emotions in personality and psychopathology (pp. 305–329). Plenum Press. https://doi.org/10.1007/978-1-4613-3000-4_11

Mosher, D. L., & O’Grady, K. E. (1979). Sex guilt, trait anxiety, and females’ subjective sexual arousal to erotica. Motivation and Emotion, 3(3), 235–249. https://doi.org/10.1007/BF00994049

Mosher, D. L., & Vonderheide, S. G. (1985). Contributions of sex guilt and masturbation guilt to women’s contraceptive attitudes and use. The Journal of Sex Research, 21(1), 24–39. https://doi.org/10.1080/00224498509551243

O’Grady, K. E., & Janda, L. H. (1979). Factor analysis of the Mosher Forced-Choice Guilt Inventory. Journal of Consulting and Clinical Psychology, 47(6), 1131–1133. https://doi.org/10.1037/0022-006X.47.6.1131

Tomkins, S. S. (1979). Script theory: Differential magnification of affects. In H. E. Howe, Jr. & R. A. Dienstbier (Eds.), Nebraska Symposium on Motivation (Vol. 26, pp. 201–236). University of Nebraska Press.

Items of the Scale