Has the COVID-19 pandemic disproportionately affected women relative to men possibly reversing some of the gains made in the labor market? We explore this question using CPS data covering the first several months of the pandemic. We find that the employment-to-population ratio for prime-age women with school-aged children declined substantially relative to comparable men beginning in April 2020 and continuing through subsequent months of the pandemic. The reductions among women with school-age children relative to similar men range from 3.7 to 4.8 percentage points from April to August 2020. Triple difference estimates suggest that 64 to 89 percent of this difference can be attributed to decreased work activity due to their children. With respect to hours of work, women with school-age children suffered major losses relative to men (13-33 percent). Based on triple difference estimates that control for different caregiving responsibilities, 50 to 97 percent of the reduction of hours of work for women with school-age children can be attributed to additional child care responsibilities. In contrast, we find using nonlinear decomposition techniques that women had favorable job and skill characteristics, specifically a greater likelihood to telework, higher education levels and a less-impacted occupational distribution, which all contributed to lessening the negative impacts of COVID-19 for women relative to men. We find some evidence, however, that women were more likely to work in “non-essential” industries contributing to higher relative unemployment in the pandemic.