Overweight among preschool age children has increased at a rapid rate and is becoming a worldwide epidemic. Previous research suggests that overweight children are less active and have poorer motor skills than children of normal weight. The objective of this study was to assess motor skill and physical activity in low-income overweight and normal weight preschool children (ages 3--5 years) and determine the influence, if any, preschool activity policy had on children's activity levels.;Data were collected at low-income preschool centers (n = 5) in the state of Iowa. Children (n = 77) ages three to four years were included in the sample. A sub sample was created of normal weight (n = 21) and overweight (n = 25) children based on CDC BMI percentile. Assessment of body composition (skinfold measurements), physical activity levels (accelerometer counts and observed intensity) and motor skill competency (TGMD-2 process assessment and outcome measures) were completed with the sub sample. The preschool center's physical activity policy was also reviewed using a policy checklist and teacher interviews.;One-way ANOVA's for BMI classified weight groups and gender failed to produce significant results for physical activity and motor skill variables. Trends were found in accelerometer counts when groups were based on sum of skinfold measurements rather than BMI. The correlation between BMI percentile and sum of skinfolds was low (r(46) = .42, p = .004) and suggested that over one-third of the children were placed in the wrong weight group (inconsistently classified). Trends were also found in accelerometer counts, observed intensity levels and locomotor TGMD-2 scores when groups were based on agreement between BMI percentile and sum of skinfold. These trends showed small shifts in confidence intervals between groups and small effect sizes. Generally, overweight preschool children appear to have the same activity levels and motor skill proficiency as their normal weight peers as determined by small effect sizes and confidence intervals. Of the inconsistently classified children, those who had high BMI's and low sum of skinfolds were better at locomotor skills that involved power (i.e. run, leap, throw, kick) when compare to children with low BMI's and high sum of skinfolds. There was little variability in the preschool activity policies, and the strength of preschool policy had no effect on children's activity levels or motor skills. Based on these finding, BMI may lack validity as a means of classifying obesity in young children.