UVM Theses and Dissertations
Format:
Print
Author:
Buckwalter, Erin
Dept./Program:
Community Development and Applied Economics
Year:
2011
Degree:
MS
Abstract:
The prevalence of childhood overweight and obesity is increasing across the United States. There is evidence that children who eat more fruits and vegetables have a lower likelihood of becoming overweight and obese. Farm-to-school (FTS) programs are one method schools are using to increase children's fruit and vegetable consumption, a goal that is becoming central to the fight against childhood overweight and obesity. Social Cognitive Theory (SCT) provides a framework for understanding how to change children's dietary preferences and behaviors, and has been proposed as an appropriate theory for use in developing and evaluating FTS programs. This study reports on a 2010 evaluation of FTS programs in Vermont and builds upon current research efforts in two areas: first to understand the relationship between personal constructs of SCT and children's food-related behavior, and secondly to determine if a composite measure of these personal constructs can be used to predict meeting the Dietary Guidelines for Americans (DGA) recommendations for fruit and vegetable consumption in 9-13 year old children.
Based on SCT, a relationship between the measured variables representing personal constructs and student food-related behavior was expected. Two-stage cluster analysis was performed on the student data to determine if any relationships existed between the multiple index variables of personal constructs. Three clusters were identified: Knowledgeable with Peer Support (12%), SelfConjident & Needs Reinforcement (27%), and Needs Broad Based Interventions (61 %). Bi-variate analysis methods were used to determine if the clusters are significantly distinct and if cluster membership is associated with fruit and vegetable consumption. An important finding is that the clusters are significantly related to respondents' meeting the DGA combined recommendations for fruits and vegetables. This significant relationship suggests that SCT is an appropriate theory to apply to FTS programs.
A recursive model incorporating empirical knowledge and theoretical constructs of SCT was constructed of variables hypothesized to affect both cluster membership and meeting the DGA recommendations. Two-stage logistic regression analysis tested the model. Being in the Knowledgeable with Peer Support and SelfConjident & Needs Reinforcement clusters significantly increased the probability that a student would eat the DGA recommended amount of fruits and vegetables. Years of FTS programming at a school also significantly increased the probability of a student in that school meeting the DGA recommendations.
The findings of this study point to a distinct set of food-related attitudes of each cluster, warranting interventions that are targeted specifically to each cluster to effectively encourage health behavior change. In addition, the model determined that there is a predictive relationship between personal constructs of SCT and food-related behavior. While only 1 of the 4 specific FTS tested interventions predicted cluster membership, as noted above, an increase in the number of years of FTS programming led to a significant increase in the probability that a student would meet the DGA recommendations. Future research confirming the model, and testing other variables that potentially influence cluster membership and that can predict food-related behavior will allow FTS programs to create more targeted interventions, possibly increasing their effectiveness and success in supporting health behavior change.
Based on SCT, a relationship between the measured variables representing personal constructs and student food-related behavior was expected. Two-stage cluster analysis was performed on the student data to determine if any relationships existed between the multiple index variables of personal constructs. Three clusters were identified: Knowledgeable with Peer Support (12%), SelfConjident & Needs Reinforcement (27%), and Needs Broad Based Interventions (61 %). Bi-variate analysis methods were used to determine if the clusters are significantly distinct and if cluster membership is associated with fruit and vegetable consumption. An important finding is that the clusters are significantly related to respondents' meeting the DGA combined recommendations for fruits and vegetables. This significant relationship suggests that SCT is an appropriate theory to apply to FTS programs.
A recursive model incorporating empirical knowledge and theoretical constructs of SCT was constructed of variables hypothesized to affect both cluster membership and meeting the DGA recommendations. Two-stage logistic regression analysis tested the model. Being in the Knowledgeable with Peer Support and SelfConjident & Needs Reinforcement clusters significantly increased the probability that a student would eat the DGA recommended amount of fruits and vegetables. Years of FTS programming at a school also significantly increased the probability of a student in that school meeting the DGA recommendations.
The findings of this study point to a distinct set of food-related attitudes of each cluster, warranting interventions that are targeted specifically to each cluster to effectively encourage health behavior change. In addition, the model determined that there is a predictive relationship between personal constructs of SCT and food-related behavior. While only 1 of the 4 specific FTS tested interventions predicted cluster membership, as noted above, an increase in the number of years of FTS programming led to a significant increase in the probability that a student would meet the DGA recommendations. Future research confirming the model, and testing other variables that potentially influence cluster membership and that can predict food-related behavior will allow FTS programs to create more targeted interventions, possibly increasing their effectiveness and success in supporting health behavior change.