Ethnic minority women report poorer health outcomes and attitudes and are more vulnerable to overweight and/or obesity compared to Caucasian women. Epidemiological studies and ecologic models of health behavior suggest that built environmental factors are associated with health behaviors, like physical activity (PA), that can help to prevent obesity and its many comorbidities. Despite growth and development in this field of research, many questions remain about the relationship between the built environment and perceptions about the built environment, and whether accurate perceptions are important for PA adoption. The objectives of the study were (1) to measure the concordance of directly measured and indirectly measured neighborhood attributes and (2) to determine the correlates of the concordance between directly and indirectly measured built environment attributes among separate samples of African American and Hispanic or Latina women (3) to determine whether there is an association between concordance and PA adoption among African American and Hispanic or Latina women. Community dwelling African American and Hispanic or Latina women participating in an ongoing HIP study self-reported their environmental perceptions at baseline (T1). In order to assess longitudinal PA levels and explore ethnic differences of neighborhood perceptions, we compared objectively measured neighborhood attributes with self-reported neighborhood attributes for African American and Hispanic or Latina women. Participants' (N=409) average BMI was classified as obese (M BMI=34.5 kg/m ², SD=7.9) and the mean body fat percentage was 42.8% (SD =7.1). BMI, body fat percentage, PA and ethnicity were not significantly associated with any built environment attribute, and no multinomial regression model significantly predicted indirectly measured built environment attributes. Repeated measures analyses suggested no significant relationships between any built environment attribute concordance value and PA adoption for total self-reported or objectively measured PA. Self-reported PA significantly increased over time (F(1,184)=7.82, p=.006), and this increase did not vary by ethnicity or any built environment attribute concordance value. Being less familiar with certain built environment attributes may not be associated with PA adoption. In an effort to promote PA, community leaders and investigators must consider the complex associations between built environment attribute concordance and PA adoption, particularly among the vulnerable population of minority women.