BACKGROUND
Lack of physical activity in the US generally and among low income and minority populations particularly remains a significant health concern. National studies show that 36% of non-Hispanic Whites report engaging in no leisure-time physical activity (LTPA), while over half of African Americans (52%) and Hispanic Americans (54%) reported no LTPA (Healthy People, 2000). Racial and ethnic minority children get less physical activity than children of European American backgrounds (Gordon-Larsen et al., 2002). Active living studies, guided by ecologic models, have begun to investigate how the built environment facilitates or constrains physical activity. For example, lower SES neighborhoods tend to have fewer resources for LTPA such as parks, sport facilities, and walking/biking trails than higher SES neighborhoods (Gordon-Larsen et al., 2006). Communities with high concentrations of African Americans or Hispanics have been shown to have fewer LTPA resources (Powell et al., 2004). Few studies have examined the association between neighborhood factors and LTPA in public parks.

OBJECTIVES
If neighborhood parks are to help reverse the physical inactivity trend in diverse and disadvantaged communities, research that describes how parks are used and identifies what settings support LTPA is needed. This paper addresses the following objectives: 1) Examine levels of LTPA in selected neighborhood parks in diverse communities; 2) Measure the association between levels of LTPA and planned activity spaces (e.g., playgrounds, sport fields, etc.); and 3) Examine the association between levels of LTPA and planned activity spaces across diverse neighborhood types.

METHODS
Data were obtained by direct observation in 11 neighborhood parks in Tampa FL, and 18 parks in Chicago, IL. Trained observers recorded LTPA between 10AM and 6PM from Friday through Sunday using a modified version of the System for Observing Play and Leisure among Youth (McKenzie, 2002) similar to the tool reported by McKenzie et al. (2006). Primary activity observation codes accounted for age (youth/adult), gender, and activity levels (sedentary, walking, and vigorous). Percent agreement between paired observers ranged from 0.79 to 0.97.

Parks were selected to represent predominantly White, African American and Hispanic neighborhoods. An attempt was made to select an upper/middle and low-income neighborhood for each race/ethnic group. Parks were similar in size, facilities, activity areas, and accessibility to residents. Using individuals as units of analysis, chi-square tests assessed associations between LTPA, age, gender, and activity areas; ANOVA was employed to test for significant differences in energy expenditure (EE) (Ainsworth 2002) associated with activity spaces across neighborhood types.

RESULTS
Over 9000 individuals were observed, 6,996 in Tampa and 2,413 in Chicago. In both cities significantly more adults than children were observed, especially in Tampa parks (56.4% vs 43.6%, Chicago; 66.3% vs. 33.7%, Tampa). Males were significantly more likely to be observed than females with the pattern more pronounced in Chicago parks (68.4% vs. 31.6, Chicago; 51.3% vs. 48.7%, Tampa).
Overall, 11% of park users engaged in vigorous activity; 23% were observed walking, and 65% were coded as sedentary. In Tampa, 70% of park users engaged in sedentary activity; 21% were observed walking, and 8% were coded as vigorous. For Chicago parks, 51% of park users were sedentary; 28% were observed walking, and 22% were coded as vigorous. Statistically significant associations were observed between LTPA and age, gender, and activity spaces in both cities. Males and youth were more likely to be observed walking or in vigorous activity.

The percent walking or vigorous physical activity (WVPA) observed in primary activity spaces in Tampa parks was as follows: tennis/racquetball courts (75%), basketball courts (74.4%), open space areas (45.5%), playgrounds (44.6%), softball/baseball fields (32.5%), fishing areas (19.3%), and picnic shelters (16.9%). In Chicago, the percent observed in WVPA in primary activity zones was as follows: tennis/racquetball courts (54.5%), basketball courts (58.3%), open space areas (46.8%), playgrounds (50.9%), and softball/baseball fields (37%). Overall, sport facilities were associated with greater WVPA. The association between EE and activity spaces did not vary by neighborhood race/ethnicity and income. However, EE was significantly greater for higher income area parks in Tampa and Chicago. Also, EE associated with parks in African American neighborhoods was significantly less than EE for parks in Hispanic and White neighborhoods in Tampa. In Chicago, EE was significantly greater in African American neighborhood parks than White and Hispanic area parks.

CONCLUSIONS
Our findings suggest that efforts to increase park use should target youth and females. Interventions to support WVPA in parks should target adults and females. Additionally, the capacity of non-sport activity areas to support LTPA should be further investigated. Inconsistent findings involving race/ethnicity and EE warrant additional attention as well.