McNair Scholars Research Journal
Abstract
Emotional eating is an important precursor of weight gain and obesity among adolescent girls (Halberstadt et al., 2016). Researchers have defined emotional eating as individuals’ eating behaviors in response to the positive or negative emotions they endure (Bongers & Jansen, 2016). In past studies, stress has been found to be an important indicator of emotional eating in adolescent girls (Corsica, Hood, Katterman, Kleinman, & Ivan, 2014). However, not all girls who experience stress will engage in emotional eating. The stress-diathesis model suggests that certain traits of vulnerability may predispose some individuals towards mental health problems (e.g., eating disorders) in response to stress (MacNeil, Esposito-Smythers, Mehlenbeck, & Weismore, 2012). One of the important traits that may moderate the effect of stress on mental health problems is attachment styles (Chow & Ruhl, 2014). Attachment styles are defined as the internalized mental representations of individuals’ key attachment figures (e.g., mothers; Cooper, Shaver, & Collins, 1998). Attachment styles are measured by the dimensions of anxiety (e.g., fear of being abandoned by others) and avoidance (e.g., fear of being too close to others). Combining the stress-diathesis model and attachment theory, the current study aims to investigate whether adolescent girls’ attachment security within close relationships moderates the link between experiences of stress and emotional eating. Specifically, it is hypothesized that girls who are low in attachment security engage in more emotional eating when under stress. In contrast, it is hypothesized that girls who are high in attachment security engage in less emotional eating, regardless of their stress levels. To test the hypothesis, data will be drawn from archival data including 100 adolescent girls between 11 and 18 years old. Participants answered questionnaires on stress attachment security to their parents and whether they engaged in emotional eating. Moderation hypothesis will be examined with multiple regression implemented in R.