Defining perceptions of picky eating: a qualitative and quantitative investigation of early childhood mealtime behaviors

Picky Eating;Child Development;Childhood Nutrition;Parental Perception;Focus Group;Conjoint Analysis;Multiple Correspondence Analysis

Mayeaux, Mandy ; Lee ; Soo-Yeun

关键词: Picky Eating;    Child Development;    Childhood Nutrition;    Parental Perception;    Focus Group;    Conjoint Analysis;    Multiple Correspondence Analysis;   

Others  :  https://www.ideals.illinois.edu/bitstream/handle/2142/17068/1_Mayeaux_Mandy.pdf?sequence=2&isAllowed=y

美国|英语

来源: The Illinois Digital Environment for Access to Learning and Scholarship

【 摘 要 】

Picky eating is a mealtime struggle for many parents and, thus, a topic of interest for many researchers. Yet, the lack of an operational definition is a great limitation in measuring, quantifying, and truly understanding what behaviors define picky eating and how it affects early childhood development and food acceptance. This study had 3 main objectives to: 1) use qualitative and quantitative analyses to holistically assess: parental perceptions of picky eating and actual childhood mealtime behavior; 2) make recommendations on constructs that should define picky eating; and 3) propose alternative methods to classify picky eating status. To fulfill these objectives, this study was conducted in two phases – exploratory and explanatory phases. During the exploratory phase, 3 parental focus group sessions (n= 6, 7 and 6 parents / group) probed the perceptions and expectations of picky eating, as well as guided the development of four categories of picky eating behavior. These categories and their respective characteristics were used in a subsequent conjoint analysis of picky eating perceptions. In the conjoint analysis, a general population (N=359) evaluated concepts of mealtime behaviors and rated the frequency that these behaviors were used to describe a picky eater child. From the results of this study, four segments of picky eating perceptions were revealed, and were defined as—“The Sensory Dependent”, “The Behavioral Responders”, “The General Perfectionists”, “The Preferential Eaters”. Because the conjoint analysis was able to segment perceptions, it was hypothesized that this analysis could discriminate parent’s perceptions of their own child’s mealtime behavior and could objectively segment perceived Picky Eaters (PE) from Non-Picky Eaters (NPE). A parental conjoint analysis (N=498) was conducted as an explanatory study and a pre-screening tool for an Early Childhood Mealtime Study. Responses of the participants could be segmented into five groups which were similar to the previous conjoint. However, when the parent’s self-classified picky eater status of their child (based on control question: Is your child a picky eater? Responses – Never, Rarely, Sometimes, Often, and Always) was compared across the segments, there was no significant pattern within the segments. A sub-sample of this population was recruited to participate in the Early Childhood Mealtime Study which consisted of central location surveys (CLT) [Mealtime Assessment Survey and Food Inventory and History Survey] and a Home Use Test (HUT). Parent-child pairs were dichotomized into PE (n=83) and NPE (n=88) groups for comparison (based on same control question as conjoint analysis). For the CLT surveys which evaluated food consumption and mealtime behaviors of both parent and child, there were significant differences found between the PE and NPE children for the majority of measures, but respective parents had few significant differences. The 37 significant behaviors revealed in the Mealtime Assessment Survey (conducted on day 1 of the Early Childhood Mealtime Study) were analyzed on a respondent level by Multiple Correspondence Analysis (MCA). The MCA explained 71% of the variation, and the visual plots showed an underlying inter-relationship among picky eater status, the frequency at which behaviors were exhibited, and the respondents. This analysis, which is not traditionally applied to data in picky eating research, was able to differentiate the range and variation of childhood mealtime behaviors, and several behaviors were identified as potential criteria to assess picky eating. A second MCA was run on the 15 significant actual behaviors revealed during the HUT. The resulting MCA plot had a similar trend and shape as the previous plot, but there was less dispersion and differentiation of the participants and behaviors along the curve. In spite of this, the MCA was not only able to distinguish differences between perceptions and actual behaviors of NPE and PE, it was also able to demonstrate the degree of difference among NPE, moderate PE and extreme PE. Further research is needed to confirm the significant variables revealed in the MCA, but with future validation, surveys combined with MCA may provide a way to model and profile a child’s picky eater status objectively.