Purpose: This study aims to better understand how students' academic strains and multilevel strengths relate to their math achievement, with a particular emphasis on underrepresented students of color and girls given the need to broaden science, technology, engineering and math (STEM) participation for these groups. Design/methodology/approach: National Education Longitudinal Study of 1988 data was used for a historical examination of the various student academic strains and multilevel strengths that relate to math achievement in high school. T-tests and chi-square tests were conducted to examine differences in strains and strengths across policy-relevant student subgroups. Ordinary least squares (OLS) regression was used to examine how students' strains and strengths related to their math achievement and the relative importance of each of these factors. Findings: The findings suggest that both the academic strains and multilevel strengths that students' experience in middle school are related to their high school math achievement and the prevalence of these factors varies across different policy-relevant student subgroups. Furthermore, the relative importance of these factors on achievement differs. Originality/value: Studies which focus on either students' academic challenges or their adaptive strengths fall short of a more nuanced discussion about how both factors relate to math outcomes. This study addresses this limitation and emphasizes that stakeholders who are interested in STEM diversity should consider holistic strategies for alleviating gender and racial/ethnic discrepancies in secondary math achievement.