The Obesity Parenting Intervention Scale: Factorial Validity and Invariance Among Head Start Parents

Obesity affects 15.7% of U.S. preschoolers, with higher rates among low-income and Spanish-speaking populations. Food, physical activity, and sleep parenting practices, referred to collectively as obesity-related parenting practices, are linked with children's risk of obesity and are a common target in family-based obesity interventions. Yet, there is no brief, validated measure of obesity-related parenting practices that is appropriate for use in intervention studies and for diverse audiences. This study tests the factorial validity of a brief measure of obesity-related parenting and measurement invariance of the English and Spanish versions of the scale, as well as among mothers and fathers.

Parents of children enrolled in Head Start (n=578; 500 mothers and 78 fathers) completed a brief survey of food (7 items), physical activity (5 items), and sleep parenting (3 items) in fall of 2017 and 2018. Scale items were drawn from existing measures and the evidence base, initially drafted in English, and then translated to Spanish. One parent per family completed the scale independently in English (n=448) or Spanish (n=130). A confirmatory factor analysis framework was adopted to test a 3-factor model for the total sample. Multi-group structural equation modeling was used to assess measurement invariance of the scale by the language of administration (English or Spanish) and among mothers and fathers separately.

Results supported a 3-factor model of obesity parenting with a single factor each for food, physical activity, and sleep parenting. There was statistically significant measurement invariance across all groups (p<0.05). Internal consistency was adequate across factors (α=0.65-0.80).

This brief obesity-parenting scale demonstrates adequate factorial validity in English and Spanish and among mothers and fathers. This measure has been integrated into an intervention, and future work will test sensitivity to change.