Background: Previous population-based surveys to monitor sun protection behavior over time have relied on self-report, which can be subject to recall and misclassification bias and social desirability bias. The present study aimed to describe the prevalence and determinants of teenagers' and adults' observed sun protection behavior while engaged in outdoor leisure activities on summer weekends, over a decade of the SunSmart skin cancer prevention program, which involved public education and advocacy. Method: Serial cross-sectional observational field surveys of teenagers and adults at leisure were undertaken during summer weekends between 11 a.m. and 3 p.m., from 1992 to 2002 (N = 46,810). The four types of setting for observation were parks and gardens, golf courses, tennis courts, and pools and beaches, located within a 25-km radius of Melbourne city center, Australia. The main outcome measure was a binary clothes cover