Food parenting measurement issues: working group consensus report

Abstract Childhood obesity is a growing problem. As more researchers become involved in the study of parenting influences on childhood obesity, there appears to be a lack of agreement regarding the most important parenting constructs of interest, definitions of those constructs, and measurement of those constructs in a consistent manner across studies. This article aims to summarize findings from a working group that convened specifically to discuss measurement issues related to parental influences on childhood obesity. Six subgroups were formed to address key measurement issues. The conceptualization subgroup proposed to define and distinguish constructs of general parenting styles, feeding styles, and food parenting practices with the goal of understanding interrelating levels of parental influence on child eating behaviors. The observational subgroup identified the need to map constructs for use in coding direct observations and create observational measures that can capture the bidirectional effects of parent-child interactions. The self-regulation subgroup proposed an operational definition of child self-regulation of energy intake and suggested future measures of self-regulation across different stages of development. The translational/community involvement subgroup proposed the involvement of community in the development of surveys so that measures adequately reflect cultural understanding and practices of the community. The qualitative methods subgroup proposed qualitative methods as a way to better understand the breadth of food parenting practices and motivations for the use of such practices. The longitudinal subgroup stressed the importance of food parenting measures sensitive to change for use in longitudinal studies. In the creation of new measures, it is important to consider cultural sensitivity and context-specific food parenting domains. Moderating variables such as child temperament and child food preferences should be considered in models.