The current study involved multilevel analysis of high-stakes examination results (i.e., GCSE) in Northern Ireland to investigate gender differentials in mathematical achievement, whereas most previous research in the area used results from low-stakes tests (i.e., PISA, TIMSS). The analysis supported the gender similarities hypothesis with respect to both overall and content domain-specific mathematical attainment. Similar conclusions were drawn from the current study as have been reported in studies into gender differentials using data from low-stakes assessments in the respective jurisdiction. This suggests that previously expressed concerns in the literature about the viability of using data derived from low-stakes assessments to accurately assess gender differentials in achievement may be unfounded. Furthermore, the context for the current study permitted an investigation into the effects of school type (grammar versus non-grammar) and gender on overall and domain-specific mathematical achievement, an area that has received scant attention in the literature. School type was not found to have an effect on the applicability of the gender similarities hypothesis with respect to mathematical achievement. The study findings are likely to prove useful to researchers and policymakers who are interested in gender equity issues in mathematics.