Four studies are reported on the derivation and assessment of a hypermasculinity scale. In Study 1, a questionnaire measure of hypermasculine values was derived from an initial 122 items, rated on a seven-point scale by 600 men from eight categories, based on occupation or sport interest. Factor analysis and item reduction produced 26-and 16-item scales (Hypermasculine Values Questionnaire, HVQ and Short Hypermasculine Values Questionnaire) with high internal consistencies. There were substantial differences between categories, consistent with predictions based on their gender-stereotypic connotations. Study 2 involved the scales being administered to another similarly composed sample: again high internal consistency and unidimensionality (in a confirmatory factor analysis) were found, and a similar association with category membership. Test-retest reliability was high. In Study 3, the concurrent and discriminative validity of the HVQ was studied, by comparing it with an existing measure of hypermasculinity, male role norms, attitudes to women's rights, gender-related traits, and trait aggression. Associations were found with other gender scales, and there was a moderate association with trait physical aggression. The range of associations reflected the items on the scale, which involve toughness, the need to avoid femininity, and control of women's sexuality, themes familiar from ethnographic accounts of masculinity. Study 4 showed that the HVQ was associated with hostile but not benevolent sexism, and replicated its association with trait aggression.