An exploratory factor analysis of the Parenting strategies for Eating and physical Activity Scale (PEAS) for use in Hispanic mothers of adolescent and preadolescent daughters with overweight

• PEAS factor structure was tested in 134 mothers of adolescent and preadolescent girls.
• The original five-factor model exhibited poor fit in this sample.
• Removal of thirteen items resulted in a three-factor model with good fit.
• Revised Limit Setting, Monitoring, and Discipline factors were reliable.

Existing measures of feeding and physical activity parenting strategies have not been validated for use among Hispanic mothers of adolescent and preadolescent daughters with overweight. The purpose of this study was to investigate the factor structure of the Parenting strategies for Eating and Activity Scale (PEAS) among 134 Hispanic mothers (Mage = 38.81, SD ± 6.34) of adolescent and preadolescent daughters with overweight. Results from this exploratory study revealed that a three-factor structure (Limit Setting, Monitoring, and Discipline) and a 13-item revised PEAS measure might be better suited for use with this population. The revised Limit Setting, Monitoring, Discipline, and total subscales had good reliability (α = 0.89, α = 0.88, α = 0.90, and α = 0.87, respectively) and the subscale inter-item correlations were strong. To assess the concurrent validity, the revised PEAS subscales were correlated with both subscales of the Behavioral Strategies to Reduce Fat and Increase Fiber (Parent Report) in a subset of the sample (n = 78). The Monitoring and Discipline subscales were found to be correlated with the Behavioral Strategies to Reduce Fat (r = .36 and r = .27, p < .05, respectively) and Increase Fiber (r = .40 and r = .35, p < .01, respectively) subscales. However, the revised PEAS Limit Setting subscale was not correlated with either Reduce Fat or Increase Fiber strategy. Taken together, these results indicated some degree of concurrent validity. Results from this study should be cross-validated using confirmatory factor analysis approaches.