The objective of this study was to examine, across 22 countries, the association between community size and individual sport, team sport, and exercise participation. Hierarchal non-linear Bernoulli modeling is used to examine the association between community size (100,000-10,000; <10,000) and (a) individual sport, (b) team sport, and (c) exercise participation. After controlling for country-level clustering and demographic variables, those residing a community with between 100,000 and 10,000 residents are more likely to participate in individual sport [odds ratio (OR) = 1.14; 95% confidence interval (CI) = 1.05-1.23] while residing in a community with less than 10,000 residents is unrelated (OR = 1.06; 95% CI = 0.96-1.19). Those residing in communities with between 100,000 and 10,000 residents were more likely to participate in team sport (OR = 1.21; 95% CI = 1.01-1.45) while residing in a community with less than 10,000 residents is unrelated (OR = 1.02; 95% CI = 0.88-1.18). Residing in a community with between 100,000 and 10,000 residents is unrelated to exercise participation (OR = 0.97; 95% CI = 0.89-1.7), while residing in a community with less than 10,000 residents is negatively related to exercise participation (OR = 0.86; 95% CI = 0.79-0.93). These findings provide novel evidence that communities between 100,000 and 10,000 residents are related to increased sport participation, particularly team sport participation.
Keywords: Bernoulli models; Birthplace effect; city size; cross-country.
© 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd.