Background: Strand, Deary, and Smith (2006) reported an analysis of sex differences on the Cognitive Abilities Test (CAT) for over 320,000 UK students 11-12 years old. Although mean differences were small, males were overrepresented at the upper and lower extremes of the score distributions on the quantitative and non-verbal batteries and at the lower extreme of the verbal battery. Aims: We investigate whether these results were unique to the UK or whether they would be seen in other countries, at other grades, cohorts, or forms of the test. Sample: The sample consisted of three nationally representative cohorts of US students in grades 3 through 11 (total N=318,599) for the 1984, 1992, and 2000 standardizations of the US version of the CAT. Methods: We replicated and extended the Strand "et al." (2006) results by comparing the proportions of males and females at each score level across countries (UK vs. US), grades (3-11), and cohorts/test forms (Forms 4, 5, and 6 standardized in 1984, 1992, and 2000, respectively). Results: The results showed an astonishing consistency in sex differences across countries, grades, cohorts, and test forms. Conclusions: Implications for the current debate about sex differences in quantitative reasoning abilities are discussed.