Correlates of preschool children's physical activity across all domains of the social ecological model have previously been reported in the literature. However, there is a scarcity of reliable instruments designed to capture such correlates in preschool-aged children. Objectives: This study reports the development, and assessment of the test-retest and internal reliability of a comprehensive proxy survey instrument designed to measure correlates of preschool children's physical activity across the three domains of the social ecological model. Design: Study one included data from 47 parents of preschool aged (3-5 years) children who completed the survey on two occasions. Study two included data from 943 participants who completed the survey as part of the larger Healthy Active Preschool Years Study. Methods: A comprehensive search of the literature and a series of focus groups were used to identify potential correlates for inclusion in the survey. Study one investigated the test-retest reliability of the survey; study two investigated internal reliability. Test-retest reliability was assessed using Kappa statistics and percent agreement for categorical variables, and intraclass correlation for continuous variables. Internal reliability was assessed using Cronbach's alpha. Results: In total, 230 items, summed and composite scores were assessed. Of those items, 190 (83%) showed acceptable reliability. The majority of items in each domain of the social ecological model - individual (75%), social (81%) and physical environment (90%) - showed acceptable reliability. Conclusions: The Healthy Active Preschool Years parent proxy survey shows good reliability and is appropriate for use with the target population. Future studies should investigate validity of the constructs in the survey. © 2012 Sports Medicine Australia.