There have been numerous reports of changes in phenology, which are frequently attributed to environmental change. Age-dependent change in phenotypic traits, fledgling production, and the timing of events in the life cycle is also widespread. This means that changes in the age structure of a population could generate changes in phenology, which may be incorrectly attributed to environmental change or microevolution. Here, estimates of selection for arrival date, arrival mass, and laying date are compared when age is and is not corrected for. This is achieved using long-term individual-based data collected from a breeding colony of Common Terns (Sterna hirundo) and a novel fitness measure: individual contributions to population growth. The failure to correct for age generated deceptive estimates of selection in eight out of nine comparisons. In six out of nine comparisons, the direction of selection differed between age-corrected and uncorrected estimates. Persistent individual differences were detected: individuals remained within the same part of the phenotype distribution throughout life. The age-corrected estimates of selection were weak and explained little variation in fitness, suggesting that arrival date, arrival mass, and laying date are not under intense selection in this population. These results also demonstrate the importance of correcting for age when identifying factors associated with changes in seabird phenology.