Background:
Perceptions of neighborhood safety are important environmental factors associated with physical activity. Safety is a multi-layered concept that is perceived differently by different groups of people (e.g., adults vs. children, male vs. female, white vs. non-white Hispanic) and for different judgmental criteria (e.g., safety from crime, traffic or injury). Latinos often live in an environment with high rates of crimes and crashes, and they perceive safety issues to be the key barriers to their walking and physical activity.

Objectives:
To assess differences in the perception of neighborhood safety (specifically focusing on the factors related to walking and bicycling) between children and their mothers; and to examine the roles of neighborhood safety as perceived by children vs. by their mothers, and the congruence between the maternal and children’s perceptions, in explaining children’s physical activity.

Methods:
Participants included 136 Latino children (mean age 10.0 years; SD 0.8 years) and their mothers (mean age 36.2 years; SD 7.3 years). Most of the children (78%) were born in the United States; whereas, 74% of the mothers were foreign-born with an average of 16 years living in the United States. Approximately, 62% of mothers reported having a Hispanic orientation, 28% reported a bicultural orientation, and 10% a non-Hispanic orientation. Both mothers and children completed a survey on their perceptions about safety of their neighborhoods for cycling and walking related to traffic, car speed, crosswalks, traffic signals, lighting, stray dogs, gangs, and strangers. Maternal and children’s physical activity was measured using Actigraph accelerometers over a period of seven days. Paired t-tests were conducted to compare mothers’ and children’s safety perceptions. Moderate-to-vigorous physical activity average was regressed on children’s perceptions, mothers’ perceptions, and the congruency of perceptions between them, using the eight safety perception variables.

Results:
For all eight safety variables examined, maternal perceptions received lower ratings than their children’s. The differences between parental and children’s ratings were significant at the 0.01 level for six variables, including safety related to traffic, car speed, crosswalks, traffic signals, lighting and gangs. None of the multiple regression models predicting children’s physical activity levels were statistically significant (at the 0.05 level). However, the maternal model predicted higher percentage of the variance (26%) compared to the children’s model (22%) and the congruency of the score model (16%). This suggests a potentially stronger role of the maternal safety perception in predicting children’s physical activity.

Conclusions:
Findings from this study revealed disagreement between Latino mothers’ and children’s perceptions about the safety of their neighborhoods for cycling and walking. Maternal perception of safety appears to be more relevant to children’s physical activity, than the children’s perception and congruence between maternal and children’s perceptions. Thus, interventions and policies aiming at increasing children’s physical activity levels should involve parents, especially mothers, and target the specific safety barriers perceived important by parents.

Support:
This research was supported by Grant # 63755 from the Robert Wood Johnson Foundation, Active Living Research Program.