Many questionnaires ascertain physical activity (PA) frequency, duration, and intensity to benchmark achievement of PA recommendations. However, most scoring algorithms utilize absolute intensity estimates when exertion may be influenced by age or health characteristics. This study quantified PA estimates with and without adjustments for perceived exertion and evaluated if differences were associated with individual-level characteristics. Women (n = 2,711) in the United States from the Study of Women's Health Across the Nation who completed ≥ 3 Kaiser Physical Activity Surveys (KPAS) across 8 biennial visits were included (baseline age: 46.4 ± 2.7 years). KPAS responses about activity mode and exertion were converted to metabolic equivalent of a task (METs) using the 2011 Compendium of Physical Activities to estimate absolute and perceived intensity-adjusted METs. Repeated measures (linear mixed effects) regression models were used to examine associations of sociodemographic and health-related characteristics with change in the difference between absolute MET estimates and perceived intensity-adjusted MET estimates. Older age (p < 0.001), Chinese (p < 0.001) and Japanese (p = 0.01) ethnicity, and current smoking (p = 0.001) were associated with positive differences between absolute and perceived intensity-adjusted MET estimates, which is suggestive of lower perceived-intensity physical activity. However, for most participants, absolute intensity-based estimates closely approximated perceived intensity-adjusted estimates over time. Traditional PA scoring techniques using absolute intensity estimates only may provide sufficient estimates of PA in longitudinal cohort studies of mid-life and older adult women.