Date on Master's Thesis/Doctoral Dissertation
Committee Co-Chair (if applicable)
Sleep; Physical Activity; Food Choices; Stress
College students are at increased risk for weight-gain and obesity compared to their non-student peers. Although many studies have confirmed these findings, few address the underlying causes of weight-related behaviors in college students. Associations between sleep behaviors (sleep duration and sleep quality), stress and weight-related behaviors (physical activity and food choices) have been identified, but studies have not adequately explored these associations among college students. Further, the influence of bioecological determinants of health cannot be overlooked when addressing health-related behaviors in diverse populations. The purpose of this dissertation was to explore factors impacting sleep behaviors and weight-related behaviors in college students. Three manuscripts comprised this dissertation and included: a state of the science review of sleep behaviors in college students; a systematic review of instruments used to assess physical activity in college students; and a cross-sectional study identifying predictors and moderators of weight-related behaviors in college students. A critical review of sleep behaviors in college students revealed that developmental changes occurring in older adolescents and young adults results in delayed sleep times. Decreased sleep duration in this population is associated with poor health outcomes (weight-gain), higher risks for accident related death/disability, and academic underperformance. Findings from this review enforced the need for healthcare providers working with college students to assess for and educate students regarding the risks associated with inadequate sleep. The second manuscript evaluated the psychometric properties of commonly used self-report physical activity measures used with college student populations. Three instruments including: the International Physical Activity Questionnaire; the Leisure Time Exercise Questionnaire; and the National College Health Assessment II, were identified as the most commonly used instruments used with college student populations. Results indicated that researchers must consider the physical activity domains of interest, the completion process, and the reliability and validity of instruments when selecting self-report instruments of physical activity for use in studies examining college students. The third manuscript summarized a cross-sectional, quantitative research study designed to identify predictors and moderators of weight-related behaviors among college students to provide a better understanding of the association between sleep behaviors (sleep duration and sleep quality) and perceived stress and what role they play in predicting weight-related behaviors. The aims of the study were to (1) characterize sleep behaviors, perceived stress and weight-related behaviors in college students stratified by bioecological determinants of health and (2) identify predictors and moderators of weight-related behaviors in college students. Approximately 394 undergraduate nursing students were recruited to complete self-report instruments related to sleep behaviors, perceived stress, physical activity, fruit intake, vegetable intake, sugar sweetened beverage consumption, fast-food consumption, and bioecological determinants of health (n=268). Data were analyzed to determine group differences in sleep behaviors, perceived stress and weight-related behaviors based on bioecological determinants of health. Next, path analysis was conducted to establish correlations and model fit. Finally, multiple group analysis was performed to identify the moderating effect of perceived stress. The findings indicated that having children/step-children and eating the majority of meals at home had significant indirect effects on physical activity and sugar-sweetened beverage consumption when mediated by sleep duration. Perceived stress was also found to moderate the predictive effect of bioecological variables and sleep behaviors on weight-related behaviors. Limitations of the study included limited interpretability of cross-sectional data, exclusive use of self-report data, a homogenous sample that limited generalizability, and a lengthy survey.
Owens, Heather, "Identification of predictors and moderators of weight-related behaviors in college students." (2016). Electronic Theses and Dissertations. Paper 2436.