A novel measurement method of, and factors associated with, the healthfulness of parent-child food purchasing interactions




Calloway, Eric Elyett

Journal Title

Journal ISSN

Volume Title



The aims of this research were to 1) demonstrate the validity of using a personally-worn micro-camcorder (PWMC) method to assess in-store parent-child food-purchasing interactions and environmental factors related to these behaviors; 2) examine the relationship between child at-home TV-exposure, home food availability/accessibility, parent dietary modeling, and child-feeding style with the healthfulness of child in-store food purchasing requests; and 3) examine the relationship between parent weight status, parent diet quality, food purchasing intentions, perceived relative cost of healthy food, and the use of nutrition facts labels with the healthfulness of parent responses to child in-store food purchasing requests. A total sample of 40 parent-child dyads completed the study. Parents were a mean age of 36.5 years (±6.3), and children were a mean age of 3.8 years (±1.1). Dyads were met at their usual grocery store and shopping time. Children wore a micro-camcorder or eButton on a hat to capture what they saw. Parents also completed a questionnaire about nutrition behaviors and the home food environment based on validated questions from the literature. Coded personally worn micro-camcorder (PWMC) data were highly correlated (rho = 0.345-0.911, p<0.01) with in-person observational data for assessing in-store behavioral and environmental factors, and the method demonstrated a high degree of reliability for assessing purchasing decisions compared to receipt data (Cohen's kappa = 0.787). Also, inter-rater reliability for assessing environmental/behavioral variables ranged from moderate to almost perfect (Cohen's kappa = 0.466-0.937). Children whose parents reported high levels of unhealthy dietary modeling had lower odds of a food request being healthy (OR=0.50, P=0.021), and having parents who report non-directive child-feeding had increased odds of a request being healthy (OR=1.66, P=0.028). Healthy weight parents were more likely to make healthy responses to child food purchasing requests than overweight/obese parents (OR=2.06, P=0.022). Behavioral interventions that seek to improve the healthfulness of food purchasing in families with young children should include components to promote non-directive feeding styles, discourage unhealthful dietary modeling, provide additional resources to target overweight/obese parents' responses to child requests, and use the PWMC method for efficient measurement of these behaviors.



LCSH Subject Headings