EFFECT OF A TECHNOLOGY-INTEGRATED , NUTRITION EDUCATION PROGRAM ON LOW-INCOME YOUTHS ’ DIETARY INTAKE

Childhood obesity is an important public health problem as it relates to several chronic diseases and continues to be high, particularly among low-socioeconomic (SES) and racial and ethnic minority populations. In 2011-2014, 25.0% of Hispanic 611-year-old school-aged children were considered to be obese or extremely obese, followed by 21.4% of non-Hispanic black children and 13.6% of non-Hispanic white children. When compared to higher-SES children of the same ethnicity and race, lowSES Hispanic, white, and black children were 2.7, 1.9 and 3.2 times more likely to be obese, respectively. Contributing to the obesity epidemic among children is the excess consumption of energy-dense snacks (EDS) and sugar-sweetened beverages (SSB) and not meeting the dietary recommendations for fruits and vegetables (FV). Given their wide reach, schools are an optimal location to educate on the importance of healthy foods and/or reduction of unhealthy foods that may influence dietary habits. The majority of school-based nutrition interventions have focused primarily on increasing fruit and vegetable (FV) consumption. However, this is problematic because while EDS and SSB provide very little in terms of nutrients, they more than likely replace healthy foods and also provide a lot of calories which can lead to weight gain. Furthermore, students respond favorably to technology, a tool that has been shown to increase nutrition outcomes, yet has been sparsely used in EDS and SSBtargeted nutrition education with low-SES school-aged students. Therefore, the first chapter focuses on the primary aim of this study which was to test the effect of a 13week school-based nutrition education program on EDS (sweet and salty) and SSB intake with low-SES 3 grade students utilizing the technology-integrated Body Quest: Food of the Warrior curriculum enhanced with additional nutrition education materials. The treatment 3 graders significantly decreased their EDS and SSB consumption from baseline (week 1) to post-assessment (week 13). When compared to the control group over time, the treatment 3 graders significantly decreased their EDS consumption. These results indicate that the school-based nutrition education program is effective in decreasing EDS consumption in low-SES 3 graders. While school-based nutrition education programs help improve what foods students consume, there is room for improvement. One way to improve these programs is by incorporating student feedback into nutrition education programs. Moreover, students’ perspectives may help provide a more complete picture on how a school-based nutrition education program can impact what they eat. They may also provide insight into the students’ perceptions of the program to help guide future programming. However, few studies have incorporated feedback from low-SES, racially and ethnically diverse school-aged students. Thus, the second chapter concentrates on the secondary aim which was to determine the acceptability and appeal of the school-based program, as well as barriers and/or facilitators to behavior changes by the 3 grade students, through semi-structured focus groups. Qualitative analysis found that the 3 grade treatment students enjoyed the program, yet had suggestions for improvement; perceived that the program influenced their attitudes towards making healthy choices and also affected what their family was consuming; and shared barriers such as appealing taste to unhealthy food that prevented them from eating healthier. The students’ insights help to inform future program content and understand what facilitates and prevents behavior change. Lastly, as parents/caregivers play a critical role in shaping the child’s environment and behaviors, they also need to be included in education efforts. However, parental involvement in nutrition education programs remains a challenge, and are often only provided indirect education through newsletters. Active involvement is successful in behavior change, yet is sparse, especially in the low-SES population. Therefore, the attention of the third chapter is of the third exploratory aim of this study. The third aim explored if students exposed to an additional group-based parental component would have greater improvement in EDS and SSB outcomes compared to those students who only receive the in school nutrition education program. As extensive recruitment and retention efforts were made for a 6-week “Family Night” program, this exploratory aim morphed into an opportunity to share “lessons learned” around recruitment, retention and family programmatic successes and challenges. Multiple modes of recruitment including flyers, stickers and text messages were used. Additionally, involving students in the program and reminder text messages encouraged repeated family attendance. From baseline (week 1) to postassessment (week 6), parents improved in nutrition-related parental practices, children increased their confidence with cooking skills, and both parents and children improved in nutrition-related habits. While recruitment and retention was a challenge, the “Family Night” program was successful in improving the involved families’ wellbeing. In conclusion, this multicomponent intervention targeted at low-SES 3 graders successfully decreased unhealthy dietary consumption, improved family nutrition-related habits, and provided a mode for students to express their thoughts, share insight, and contribute in a meaningful way to future programming.


INTRODUCTION
Given that childhood obesity is associated with many chronic diseases, prevention efforts are critical, especially among racial and ethnic minority populations. 1 In 2011-2014, 25.0% of Hispanic 6-11 year old school-aged children were considered to be obese or extremely obese, followed by 21.4% of non-Hispanic Black children and 13.6% of non-Hispanic White children. 2 Independent of ethnicity, lower socio-economic status (SES) is also associated with higher obesity prevalence.
When compared to higher SES children of the same ethnicity and race, low-SES Hispanic, White, and Black children were 2.7, 1.9 and 3.2 times more likely to be obese, respectively. 3 There is a need to address obesity-related behaviors among ethnic minority school-aged children as they have a greater propensity to live in poverty. 4 There are several potential factors to why low-SES and ethnic minority children are at higher risk of being overweight or obese including access to and consumption of low quality foods. In order to stretch the food dollar, low-cost meats, inexpensive grains, and nutrient-poor items that are low in cost are purchased. 5 As part of these low-cost food items, families purchase energy-dense snacks (EDS) and sugar-sweetened beverages (SSB), laden with fat, salt, and sugar, which are associated with excess weight. [6][7][8][9][10] Today's children are not meeting the dietary recommendations and are over consuming EDS and SSB. [11][12][13][14] According to 2007-2010 National Health and Nutrition Examination Survey data, on average EDS and SSB consumption makes up 37.8% of total calories of the 6-11 year old children. 12 This is especially true for lower SES, ethnic minority children as EDS and SSB consumption have been inversely associated with parental SES 15 and have increased in non-Hispanic Black children. 16 Given that low-SES ethnic minority children are more likely to consume an excess of EDS and SSB, there is a need for successful nutrition education interventions that target these behaviors.
The majority of school-based nutrition interventions have focused primarily on increasing fruit and vegetable (FV) consumption. 17,18 These interventions have focused on low-SES minority populations, and found success with improved FV consumption, knowledge, attitudes and/or beliefs. [19][20][21][22][23][24][25] Of school-based interventions involving low-SES minority children, few have focused on decreasing unhealthy habits such as EDS and SSB consumption. 22,[26][27][28] Not only do EDS and SSB provide very nutritional value at a high calorie cost, which can lead to weight gain, they replace healthy foods. 29, 30 Furthermore, children respond favorably to technology, a tool that has been shown to increase nutrition outcomes, 31, 32 yet has been sparsely used in EDS and SSB-targeted nutrition education with low-SES school-aged children. 22,27,28 Of the studies that have targeted unhealthy dietary behaviors through technology, Sharma et al pilot tested the Quest to Lava Mountain computer game with middle-to low-SES children (n=107), and found a significant decrease in sugar consumption in the treatment group when compared to the control group (β= -9.73; 95% CI= -18.00, -1.47, p=0.021). 22 However, there was no indication of which sugarcontaining foods or drinks decreased. The University of Alabama's school-based nutrition education curriculum Body Quest: Food of the Warrior (BQ) utilized technology with low-SES elementary-aged children to impact changes in FV consumption 24 and intention to change SSB consumption. 28 However, it has not assessed actual changes in EDS and SSB consumption. There is a need to improve EDS and SSB consumption among school-aged ethnic minority children through school based nutrition education approaches. This research article describes the results of a quasi-experimental, technology-integrated intervention on EDS (sweet and salty snacks) and SSB consumption with low-income 3 rd grade students. The school-based nutrition education curriculum used in the intervention is based on the Social Cognitive and Experiential Learning theories and utilized a modified version of the technology-integrated BQ curriculum. 24 The objectives of the research study were to determine the effect of the technology-integrated 13-week nutrition education program on low-SES 3 rd graders' consumption of EDS and SSB. It was hypothesized that the intervention students would decrease EDS and/or SSB consumption due to the program compared to the control students.

Study Design
This 2x2 quasi-experimental research study was one component of a clustered-

Participants and Recruitment
Providence, Rhode Island is one of the four core cities in the state, with an average 87.7% of public school students eligible for free or reduced-school meals. 33 The city population consists of 64% Hispanic and 17% Black/African-Americans. 34 Based on Principal and 3 rd grade teachers consent, the school district determined the initial treatment and control schools; in the next school year, the previous control school became the treatment school and stakeholder referrals were used to select the subsequent schools. Figure 1 provides details of number of classrooms and student participants. Parents and caregivers received a letter via their student's backpack describing the study and 3 rd grade students in participating classrooms engaged in the program as a part of their science curriculum as approved by the URI IRB #1213-106.

Instruments, Protocol and Data Collection
Cognitive interviews were held with six 3 rd grade students to ensure survey instrument clarity and understanding. Students did not understand the terms "Hispanic" or "non-Hispanic", but instead understood when interviewer asked if they spoke Spanish at home. Based on this information, ethnicity was defined as the student speaking Spanish at home. Additionally, "other" and "not sure" categories were added as options to the race question, as some students did not identify with any option provided or were unsure. No changes to the nutrition-related behavior questions on sweet snacks, salty snacks, SSB, fruits and vegetables were made. To ensure uniformity, a standard script to administer the survey was provided to data collectors.
Each student was assigned a unique identification number. Educators collected demographic information including age, gender, race and ethnicity as well as nutrition-related behaviors through Surveymonkey.com in both Spanish and English on iPads. The survey included instruments provided by USDA CYFAR and those adapted from the Beverage and Snack Questionnaire. 35 Through self-recall, nutritionrelated behavior questions assessed the following: "how many times did you eat a sweet snack yesterday between your meals?" (and same for salty snacks), "how many times did you drink a sugary drink yesterday? Do not include 100% fruit juice, chocolate milk or diet drinks.", as well as how many times in the previous day fruits included soda, sports drinks, energy drinks, sweetened iced tea, and fruit drinks. All questions were multiple choice, with range option of "0 times" to "5 or more times" consumed.
Data were collected at two time points (always a week day) for both groups: baseline (week 1) and post-assessment (13 weeks) with the control data collected within a 2-week period of treatment data. To complete all surveys, the students followed along as the educator read each question aloud to the class, allowing for visual and auditory understanding of the question. Set examples to clarify questions were provided with questions. The surveys took approximately 20 minutes to complete. If any student was absent, a rescheduled survey time was attempted to be made as close to the original date as possible. Process evaluation conducted throughout the intervention included weekly attendance of each student.

Intervention
The intervention school received a weekly one-hour in-class program for 13weeks while the control school received no programming. The curriculum involved interactive, hands-on activities as well as seven iPad applications created for the BQ curriculum to reinforce topics taught by the educators who were Registered Dietitians.
A modified version of the BQ curriculum was used. Modifications included extending all 13 lessons from 30 minutes to one hour in length. This allowed each topic to be more robust with additional hands-on activities. It also allowed for additional topics not covered in the original curriculum to be taught. Such additional topics included breakfast, "Go, Slow, Whoa", MyPlate, fast food, and sugar-sweetened beverages.
Lastly, the modified curriculum removed the FV tasting portion of the original curriculum and instead relied on the USDA Fresh Fruit and Vegetable Program which provided a fruit or vegetable in the classroom during the lesson. This modified curriculum was piloted with one 3 rd grade classroom. No major modifications were made after the pilot. Table 1 provides details of curriculum topics.

Statistical Analysis
G*Power version 3.0.10 was used to calculate sample size. Sample size calculations were performed based on expected changes in EDS and SSB from pilot year data. 36 In the pilot, the treatment group (n=70) had a significant decrease in EDS consumption between meals by 0.70±1.41 times per day and a significant decrease in SSB consumption by a mean of 0.94±1.85 times per day; the control group (n=59) had a significant decrease in EDS by 0.42±1.40 times per day and no change in SSB consumption (0.00±1.77 times per day) 36 . A required sample size of 768 and 118 3 rd graders were necessary to determine the effect of the intervention on EDS and SSB, respectively, with an alpha set at 0.025 and statistical power at the 0.80 level.
All statistical analysis for this project used IBM SPSS software (version 24.0, IBM SPSS Statistics, Armonk, NY, 2016). Numerical (skewness and kurtosis) and graphical (histogram) methods were used to determine normalcy. Baseline Pearson Correlation between variables was run for both treatment and control groups. One additional variable was created from survey questions: "EDSAVG" (sweet and snack variables combined and averaged, Cronbach alpha 0.72).
Independent t-tests and chi squared assessed any differences between the treatment and control group at baseline for continuous and categorical variables, respectively. Paired t-tests were used to assess within group differences and analysis of variance (ANOVA) for between group differences of EDSAVG, sweet snacks, salty snacks, and SSB. To account for the study design in which some, but not all schools involved, were both treatment and control groups, paired t-tests were ran for EDSAVG, sweet and salty snacks separately, and SSB for each treatment and control group involved in each year of data collection. Significance was set at p <0.05.

Descriptive characteristics
Two-hundred and seventeen (217) treatment and 242 control students completed baseline data. Overall, both groups had a high eligibility for free/reducedmeals, were on average approximately eight years old, and roughly equally split in gender distribution. There were no significant difference in baseline demographic characteristics between the two groups ( Table 2). Of note is the low attrition rate throughout the research study, with only a 10.1% and 11.2% loss for the treatment and control groups, respectively ( Figure 1). Most attrition was due to relocation of students to another school or absenteeism on data collection days, despite repeated efforts to survey all students. On average, treatment students attended 11.65 lessons, with 88.1% of students attending ≥ 75% of the lessons (10 or more lessons) and 42.7% having perfect attendance.

Within and Between Group Changes
Overall, consumption of EDS and SSB was high in both the treatment and control groups (Table 2). At baseline, 88.0% of the treatment students and 88.4% of the control students consumed at least one sweet or salty snack (using EDSAVG) in between their meals during the previous day. SSB were slightly less consumed at baseline, with 75.6% and 79.6% of treatment and control students, respectively, consuming at least one SSB in the previous day. As expected, there was a moderate positive correlation between EDSAVG and SSB consumption in both the treatment (r= 0.50, p<0.01) and control groups (r= 0.52, p<0.01). Sweet snack consumption had a smaller correlation (treatment r=0.41, p<0.01; control r=0. 41, p<0.01) to SSB consumption than salty snack consumption (treatment r=0.45, p<0.01; control r=0.512, p<0.01) to SSB consumption.
Paired t-tests revealed a significant decrease in EDSAVG consumed by the treatment group from baseline to post-assessment of M=0.55 times between meals in previous day, 95% CI [0.34, 0.76], t(194)=5.10, p<0.000, d=0.37 (Table 3). When analyzed separately, both sweet snacks and salty snacks significantly decreased in the treatment group from baseline to post-assessment (Table 3). There was also a significant decrease in SSB consumption in the treatment group from baseline to post assessment of M=0. 41  showed there was a statistical significant difference in EDSAVG between the students in the two groups, F(1, 399)=6.83, p=0.009; partial  2 =0.02. When each component of the EDSAVG variable was assessed with an ANOVA, there were between group differences for both sweet and salty snacks (Table 4). However, partial eta squared showed a small effect size and power was not met at 0.80.

Additional Analysis
Over the three-year data collection period, two schools served as both control and treatment groups (schools B and C), one school as only a control group (school D), and one school as only a treatment group (school A) ( Figure 2 Paired t-tests also revealed significant improvements in EDSAVG and salty snacks for all three treatment group schools, in sweet snacks for one of three treatment group schools, and in SSB for two of three treatment group schools. In the control group, the only significant improvement was with EDSAVG in one of three control group schools.

DISCUSSION
Eating behaviors of school-aged students are important for their current and future health. This is especially true for the low-SES and racially and ethnically diverse population who have a high prevalence of unhealthy behaviors as well as obesity. 13,15,16 Given that most technology-integrated, school-based nutrition education programs have focused on improving FV, 19,20,24,25,37,38 there was a need to explore the effect of such programs on decreasing school-aged student's EDS and SSB consumption. Results from this study indicate that after completing the program there was a decrease in low-SES 3 rd graders' EDS (both sweet and salty). To the author's knowledge, this is the first research study to evaluate low-SES school-aged student's EDS and SSB consumption from a technology-integrated nutrition education program.
Decreasing unhealthy dietary behaviors among school-aged students through such a program may be an effective way to decrease long-term health consequences associated with such behaviors like EDS consumption.
While U. of Alabama's Body Quest: Food of the Warrior curriculum showed positive effects on FV consumption 24 and intended change in SSB consumption 28 in low-SES students, it had not explored the effect on EDS and SSB consumption.
Modifications were made to the curriculum to expand on healthy and unhealthy choices in meals and snacks as well as integrate different interactive, hands-on activities on topics. Exploration of this effect and modifications to the curriculum were warranted as nationally 6-11 year olds overconsume unhealthy foods and drinks.
Based on 1999-2010 NHANES data, 73.9%, 59.4% and 76.7% of 6-11 year olds consumed sweet snacks, salty snacks, and SSB on a typical day, respectively. 39 The students involved in this research had very similar, but slightly higher than the national averages for sweet snack, salty snack and SSB consumption at baseline. Thus this cohort of students was in need of healthy behavior changes.
Often one unhealthy eating behavior is associated with another. 40 Based on baseline data of correlation coefficients and coefficient of determinations, EDS and SSB consumption were only moderately associated with each other in both the treatment and control group, with about one-quarter of EDS consumption related to SSB consumption. When divided into sweet and salty snacks, sweet snack consumption accounted for 17% of SSB consumption in both treatment and control groups and salty snack consumption accounted for 20% and 27% of SSB consumption in the treatment and control groups, respectively. These relationships are much lower than what is nationally reported. Through 24-hour recall NHANES 1999-2010 data, Bleich and Wolfson found that 74.6% of students ages 6-11 years old (n=6,266) who consumed SSB also consumed sweet snacks and 61.4% of students who consumed SSB also consumed salty snacks. 39 This discrepancy between the students in this study and national data may be due to the fact that this study specifically assessed sweet or salty snacks when consumed between meals, not with meals. However, sugary foods like pastries, and salty foods like chips and French fries, may be consumed with meals and thus were not captured in this study. Future data collection should consider inclusion of sweet and salty snack foods consumed at any time in the day.
Significant within group decreases in EDS (and both sweet and salty snacks when separated) and SSB consumption were found in the treatment group.
Additionally, there was a significant between group decrease in EDS (and both sweet and salty snacks when separated) from baseline to post-assessment. These results are similar to Rosário et al. who found a significant decrease in energy-dense foods, but not SSB, when a classroom-teacher taught model of a 6-month nutrition education program was implemented for 6-12 year old students in Portugal. 41 Additionally, Sharma et al. did find a decrease in sugar consumption, though it was non-specific as to if the sugar source was from food or drink. 22 Education focused on EDS and SSB can effect behavior change.
Nutrition education programs and multi-level, systems-based approaches 42 to decrease SSB consumption in children are much more common and have found that children exposed to these programs significantly decreased SSB consumption. 43,44 This research study found a significant decrease in the treatment groups' SSB consumption from baseline to post-assessment, but the effect size was small and there was no intervention effect found over time. This is similar to other studies that saw trends or modest improvements in SSB consumption in low-SES, racial and ethnic minority youth. 26,27 The lack of interaction effect over time may be due to in part to the timing of when the SSB lesson was taught (week 12). Given that it was at the end of the curriculum it is possible that the students had less time to implement a behavior change before the post-assessment (conducted in week 13). The length of time it takes to implement a behavior change is different for each individual, depending on where they are in the process of change. 45,46 Due to the study design and preference of the schools to ultimately receive the program, some but not all schools were used as both a control and treatment group. In addition, the study was conducted over several years. To overcome some of the limitations of this study design, paired t-tests were run for EDSAVG, sweet and salty snacks separately, and SSB for each treatment and control group involved in each year of data collection. These tests showed trends in the intervention's effect on the dependent variables when the same school acted as both a control and treatment group. For the two schools that acted as both control and treatment groups in different years, it appears that when they were treatment groups, they had significant improvements in salty snacks and thus EDSAVG, but not when acting as a control school. This indicates that the intervention itself was effective in behavior change, independent of the school environment.
For the treatment group, across the three-year period, paired t-tests showed that the intervention had an effect on change, particularly EDSAVG consumption, not the school. Likewise, the control groups across the three-year period had very little behavior change from baseline to post-assessment. This indicates that the intervention is replicable with different environments (schools) and different students.
As there have been very few studies to evaluate changes in EDS consumption in children from a nutrition education program, this research adds to the much-needed body of literature.
Decreasing EDS and SSB consumption in children is important. Not only are those who consume SSB are more likely to consume EDS, 39 there is also an inverse association between EDS consumption and healthy dietary habits such as consuming FV. 29 These habits are also associated with health outcomes such as overweight and obesity, 30, 47-49 as well as cognitive outcomes such as executive functioning, 50 academic grades, 51,52 and in-class behavior. 52

Strengths and Limitations
There were several strengths to this study. The first is this study involved low-SES, minority and ethnically diverse students, a population that has been shown to need more nutrition education than higher-SES populations to increase nutrition knowledge and combat the higher propensity to consume unhealthy foods and beverages. [53][54][55] The second strength was the low attrition rate throughout the study. 56 Third, researcher data entry error was very low since the data was electronically collected and downloaded into a data analysis sheet. Lastly, the data over the three years showed replicability with different students in different schools.
While there were several strengths to the study, there were also limitations.
First, the self-recall survey was modified and not validated for that age group and asked about previous day's consumption at one-week day time point at baseline and post-assessment, which may not represent a typical week day's food and beverage consumption or be enough to capture a usual consumption. 57 In addition, recall was always on a week day, not a weekend, which limits its ability to capture day to day variability in consumption. Weekend eating tends to be different from weekdays. 58 The post-assessment was given directly at week 13, possibly not allowing enough time for implementation of behavior change as each student may be in a different stage of change. 45,46 Lastly, the survey required self-recall by the 3 rd grader. While the survey was administered via an online tool and may be more engaging to the students, 59 selfrecall is shown to be difficult with children. 60 Electronic modes of collection such as digital imaging of lunch trays to assess consumption and variety of foods 61 and software included into cafeteria computers to assess student food choices, 62 should be considered for future data collection methods. Aside from the survey, another limitation included potential respondent bias by the students, especially at postassessment, as they may have wanted to please the researchers now known to them from the program. 63 Lastly, it was not feasible to have a randomized controlled trial, so some, but not all schools, served as both control and treatment groups, and they served as those groups in different years with different students. A design that uses the same students as both control and treatment group participants within the same school year is ideal to detect behavior change due to the intervention.

IMPLICATIONS FOR RESEARCH AND PRACTICE
As reported by students in this study and nationally, EDS are widely consumed on a daily basis. Results from this intervention show a technology-integrated, schoolbased nutrition education program is effective in improving EDS consumption in low-SES 3 rd graders. Future programming should consider continued use of technology to enhance learning. It should also consider inclusion of EDS consumed at any point in the day, delayed post-assessment to allow students time to implement behavior change, alternative modes to dietary recall with children, and modified study design to eliminate potential bias and confounding factors. Year 1

Chapter One Figures and Tables
Year 2 Year 3      While school-based nutrition education programs have helped improve what foods students consume, particularly FV, 6,7 there is room for improvement. A recent meta-analysis found FV school-based nutrition education programs moderately increased fruit intake, with minimal improvements in vegetable intake. 6 One way to improve these programs is by incorporating student feedback into nutrition education programs. Conducting qualitative research with students provides meaningful information to improve programming. 8 Previous studies have found students' perspectives may help provide a more complete picture on how a school-based nutrition education program can impact what they eat. 9, 10 They may also provide insight into the students' perceptions of the program to help guide future programming. [9][10][11][12] However, few studies have incorporated feedback from low-income, racially and ethnically diverse elementary-aged students. As this population is at higher risk for developing obesity, it is important to determine how to best intervene, from their perspective, to promote healthy eating habits.

Figure 2: Paired T-Test Results for Baseline to Post-Assessment to Depict Trends in Behavior Change Results when 1) School Serves as both Control and Treatment Groups and 2) Intervention is Carried Out in Different Schools
There have been limited opportunities for low-income, racially and ethnically diverse elementary-aged students to share their perceptions regarding nutrition education programs through focus groups. This research article aims to fill this gap.
This article describes the results of focus groups conducted with low-income, racially and ethnically diverse 3 rd graders who completed a 13-week school-based nutrition education program through the University of Rhode Island's Children Youth and

Families at Risk (CYFAR) project, Integrating Nutrition Education into Providence
Full Service Schools in Providence, RI. The objectives of the focus groups were to determine student's (i) perceptions on how the program impacted their food and beverage consumption, (ii) perceptions of the overall program and potential changes for improvement, and (iii) overall barriers, independent of the program, to eating behavior change. It was hypothesized that students would report positive eating behavior changes due to the program, find the program desirable, and would reveal barriers that prevent children like them from having healthy eating habits.

Study Design
This study assessed perceptions of students who participated in one component Food of the Warrior curriculum created by the University of Alabama. 15 Curriculum content included food group function and the concept of "Go, Slow and Whoa"; FV amounts, variety and importance; EDS, including both salty and sweet snacks; and SSB. The curriculum used interactive, hands-on activities as well as iPad applications to reinforce topics taught by the educators who were registered dietitians. This paper describes focus groups conducted with students who completed the 13-week program in two treatment schools.

Participants and Recruitment
Sixteen focus groups with four 3 rd grade students in each group were completed. 16 Sixty-four out of a possible 138 students who received the program were willing to participate and were selected by the classroom teacher. The classroom teacher was instructed to select students of both genders, all learning levels, and who had attended the nutrition program throughout the school year. Thematic saturation was reached after 16 focus groups in two schools were conducted. URI's ethics committee granted internal review board approval for this research study.

Procedures
Conducted during the school day approximately two months after completion of the 13-week program, all focus groups were held in quiet locations within the school the students attended. 17,18 The focus group guide was developed based on prior literature and organized in to five sections: influence of food selection, memorable topics from the curriculum, perceived behavior change from the program, barriers to behavior change, and potential changes to the program. The focus group guide was pilot tested with a small group of same-aged children (n=4) for comprehension and clarity of questions. Table 1 provides details of question asked.
Each focus group lasted approximately 20 minutes and was audio-recorded and included the lead researcher as moderator and the nutrition educator as note taker, both of whom the students knew through the program. When doing qualitative research with children it is important to establish rapport and felt it was important to have a familiar face to increase responses and comfort. Data saturation was reached when coding of data revealed no new themes.
As part of the 13-week program, demographic information was collected during an in classroom baseline assessment. Children responded to questions developed specifically for USDA CYFAR on individual iPads. School-wide information on Providence from the RI Department of Education was also collected.
To enhance clarity of questions, some USDA CYFAR items were modified.
Modifications included asking the students if they spoke Spanish at home; this replaced asking if they were Hispanic. Additionally, "other" and "not sure" categories were added as options to the race question, as some students did not identify with any option provided or were unsure.

Data Analysis
Audio-recorded focus groups (n=16) were transcribed verbatim by a professional transcription service, Verbal Ink, and were reviewed by the focus group moderator and note taker for accuracy. Each transcript was coded using a hybrid approach of inductive and deductive thematic analysis. 19 This approach acknowledged the sections in the focus group protocol and also included any additional themes that emerged from the data during the coding process. The lead researcher utilized thematic analysis to detect themes from the content of the transcripts. 20 A codebook of structural and content codes was created and updated based on transcription readings.
A second researcher (author #4) coded 25% of the data and inter-rater agreement was calculated. There was a 94% agreement of coding, determined by the number of agreements divided by the sum of agreements and disagreements. These codes led to patterns and themes within each section. Descriptive statistics summarized student demographic characteristics based on survey data and were analyzed in IBM SPSS software (version 24.0, IBM Statistics, Armonk, NY, 02016).

RESULTS
The focus groups were comprised of students who were an average of eight years old, 62% reported speaking Spanish at home, and 16% reported as non-Hispanic Black (Table 2) Tables 3 and 4.

Theme 1: Influence on Food Selection
As part of the icebreaker activity in the focus group, two food options were shown to the students; one of a typical fast food restaurant food such as a cheeseburger, and one of a typical healthier option such as a grilled chicken sandwich or turkey sandwich on whole wheat bread. The students were asked to point to which food they would eat if given the choice. Forty out of 64 (63%) students chose the healthier option. The common reasons for selecting the healthier option were that it was the healthier choice and that it had vegetables on it. When asked why they chose the healthier option, one student replied:

"Because it looks more healthier. This [cheeseburger] has meat and this [turkey sandwich] has tomatoes and lettuce."
The less healthy option was selected most commonly for its appealing taste. This theme also carried over into the discussion on barriers to eating healthy. During that discussion, students stated that they still consumed unhealthy foods and beverages because they taste good and also because they get sick of eating only healthy foods.
When asked why they eat unhealthy foods, one student replied: "I eat ice cream every day because it tastes good, and I just want to sneak up so I can have something … I can have something sweet and then eat something healthy." When asked how they felt when eating the unhealthy foods and beverages, most responded negatively, mentioning that the unhealthy foods make them feel "not that great" or "it feels, like, badder", but a few responded positively ("I'm happy because I eat chips"). Yet, students still consume these products, and as one student summed it up: "I feel like … they're not kind of good for me but they taste so good."

Theme 2: Perceived Behavior Change from the Program
While the curriculum did not emphasize empowerment specifically, students reported feeling more empowered to influence what they were eating at home as a result of participating in the program. This increased feeling of empowerment was reinforced by students who stated that they asked for healthier items in the home and often times reported that because of this, their parents would buy those items. One student stated: "Before I ate chips and everything, and now I eat a little bit of candy. I tell my mom to buy me baby carrots, grapes, watermelon. She buys me and I eat it.
I tell her to keep our family healthy." The students also reported that by sharing what they learned in the program with their families, family members also changed their eating habits. Students stated that they appreciated this aspect of the program.
"You can tell your whole family and then your whole family will live longer and healthy life." Aside from a perceived influence on their families and home environment, the students talked about how the program was helping them make healthier food choices.
Students reported that they began to limit unhealthy foods and drinks not only by decreasing how often they have them, but also by replacing them with healthier choices.

"I think I'm making a great decision because I'm getting salad instead of, like, hamburgers and chicken nuggets."
Students also stated that they became braver to try new foods, specifically fruits and vegetables. This bravery was a major theme of the Specific to the Body Quest curriculum, the most memorable topics were learning about the characters and using the iPad applications. The Body Quest curriculum is unique because of its six characters that represent healthy eating habits.
They were introduced to the students through posters hung on classroom walls, activities during the lesson, and the iPad applications.
"We get to talk about food, what makes you healthy and the Body Quest people. We got to go on the tablet and we get to learn about, uh, vegetables."

Theme 4: How to Make a Topic Memorable
Overall, students reported that the hands-on activities influenced their learning experience. Hands-on activities included Body Quest playing cards (used in seven lessons), four learning kits such as "Think your Drink" and rubber breakfast food models, nine interactive boards and games such as "Fruit and Veggie Bingo", the seven iPad applications, and use of paper and pencil (used in four lessons).

"I really liked using the [rubber] food models because they look like real food and we could just see if we could change the Coco Puffs into Raisin Bran."
Students also reported that if the activities were fun they were more likely to remember the content and apply what they learned to their own experiences with foods.
"I liked when Ms. [teacher's name] passed out these, um, cans that were unhealthy and healthy drinks…it was actually surprising to see how much I actually drank of that soda and I don't even pay attention to the labels on the back."

Theme 5: Potential Changes to the Program
Overall, the students reported enjoying and being satisfied with the 13-week program. However, students provided suggestions on how to improve the program, including duration and/or frequency and possible curriculum modifications. In all focus groups, students reported wanting longer and more frequent sessions throughout the school year. "You should do a game [on the iPad] like that like they ask you questions and then like they ask you a question about stuff to be a body warrior, and you could be a body warrior."

DISCUSSION
The goal of this study was to capture the perceptions of low-income, racially and ethnically diverse 3 rd grade students who participated in the URI CYFAR in school program. Findings from this study highlight the struggle that 3 rd graders have between what they know is healthy and what tastes most appealing to them. It also suggests that the program may have improved the student's knowledge, bravery and empowerment; students felt that because of this they and their families were choosing to eat healthier foods. Lastly, it provided detailed information for future modifications to the curriculum with regards to specific topics and how to best teach those topics.

Knowledge and Action Conflict
In conversations about food and drink choices, the struggle between what the students know to be healthy and what they consume based on what is most taste appealing came up repeatedly. These conflicts between knowledge and behavior are similar to what previous research has found in that children, adolescents, and even adults, find it difficult to refrain from unhealthy food because of its appealing aesthetics and taste. 21 23 This is also similar to the students in this research study as they reported that they craved food, even if they knew it was not healthy for them.
Future studies should explore ways to resolve this dissonance among elementary school aged children.
Methods previously found to align knowledge with behavior are to repeatedly expose students to healthy options and to decrease access to unhealthier ones.
Evidence suggests that repeated exposure to foods increases the likability and acceptability of the food. 24 Therefore, if schools and families repeatedly expose students to healthy options for meals and snacks, the students' perception of healthy food's appealing taste may change. This may lead to social norms changing over time, as was seen with the 2010 Healthy, Hunger-Free Kids Act, changes to school meals increased acceptability by students over time. 25 The environment in which the students spend time also determines food and drink choices. This includes their home, school, and surrounding neighborhoods. Shifts within environments have been show to alter eating habits. 26-28 Continued efforts to establish healthy eating norms in different settings where elementary aged children spend time is critical.

Students' Perceived Behavior Changes
Students may still consume unhealthy food and drink choices; however, they perceived that the 13-week program helped them to make positive behavior changes.
This finding is similar to what others have reported in that students decrease unhealthy foods and also replace unhealthy with healthy options after receiving school-based nutrition education programs. 15 One of the core principles of the Body Quest curriculum is the use of "bravery" which may have positively contributed to the student's perception of making changes to their behavior. The curriculum emphasized bravery by continually encouraging the students to try new, healthy foods, specifically fruits and vegetables. This emerged as a theme during the focus groups, as the students repeatedly reported that because of this encouragement, they tried various fruits, vegetables, and other healthy foods and drinks that they had either never tried before or did not like previously but tried again.
Part of the bravery concept included the use of repeated exposure of a food or drink item to help increase the chance of liking it. Thus, the students were repeatedly encouraged to try food and drink items. 24 Influential adults such as teachers and parents can practice positive feeding practices and encourage the students to continue to try new healthy foods and drinks. 38 Like bravery, empowerment was another concept that may have contributed to student's perceived eating behavior changes. Empowerment, "the process of becoming stronger and more confident, especially in controlling one's life and claiming one's rights", 39 when increased may lead to behavior changes. 40  study is similar to other research that focused on students as agents of change. 43,45,46 Heim et al. found both an increase in empowerment and fruit and vegetable availability in the home following their garden-based intervention. 43

Informing Future Programming
In order to help instill positive behavior changes in the students and their families, the curriculum needs to be relevant to the student and taught in a way that best engages the student. Overall, the students liked the curriculum and its activities.
This is similar to what previous qualitative research studies with school-aged students have found in that the students like the programs, especially the technology components. 9, 10 Five nutrition topics, along with the Body Quest warriors and use of the iPad applications, stood out the most by students: drinks (SSB and milk), how fruits and vegetables help your body, consequences to eating healthy and unhealthy foods, "Go, Slow and Whoa", and the balance of healthy and unhealthy foods. This information is helpful when planning sustainable school-based programs. In order for programs to be sustainable after funding ceases, the program needs to be feasible for use by the school community. This may mean having a flexible program that the schools can adapt as needed. 47 Therefore, when considering school use and sustainability of the nutrition education program, having an emphasis on "must have" curriculum topics that are most salient to students, with optional add-on topics as time permits, is a reasonable compromise to nutrition education conducted by the school staff. In addition, the curriculum's delivery method and its appeal by the students may influence sustainability.
How a topic is taught is equally as important as what is taught. The students' perception that hands-on, fun activities, particularly use of the iPads, are the best delivery modes to increase retention of concepts aligns with the two theories this research project is based on: the Social Cognitive Theory and the Experiential Learning Theory. The Social Cognitive Theory's constructs of behavior capability and self-efficacy, and the Experiential Learning Theory's concept of experience, all tie concrete experiences with change. 13,14 When the students have the opportunity to actively engage in a topic and make it applicable to themselves, the chance of behavior change increases.
Overall the students had positive feedback on the program, however, the students did have some critical feedback to consider. Critical feedback included increased time to use and discuss the iPad applications, having more peer-to-peer education, and an increased focus on nutritionally sound foods and drinks to choose as opposed to focusing on foods and drinks not to choose. This feedback will help plan future programming to focus on topics pertinent to students and the best ways to present such topics.

Strengths and Limitations
There are several strengths to this qualitative research study. First, this research focused on the perceptions of low-income, racially and ethnically diverse 3 rd graders, a population that has not been an emphasis in qualitative research. Second, the focus groups were conducted with a large number of students (n=64) to the point of saturation. This allowed for the maximum amount of information to be provided by the students to comfortably conclude that the opinions of these 64 students would resonate as opinions by the other 3 rd graders involved in the program. Third, the focus groups included a moderator, note taker and were recorded and professionally transcribed. This created verbatim data with no bias by the note taker. Fourth, the same moderator and note taker were utilized in all focus groups to ensure consistency.
While recommended protocols for conducting focus groups with children were followed, some aspects of the protocol also serve as limitations. First, use of staff known by the students during the focus groups may have affected how the students responded. Respondent bias can come into play if the students respond to answers based on what they think will please the moderator. 48 This was evident in one student response during unhealthy choices dialogue. Second, because the qualitative research involved small groups of students, there may have been social desirability bias, with students influencing the others students' responses. 48 Lastly, the outcomes of this qualitative research cannot be generalized to the whole population as the 3 rd grade students involved in the research were low-income and racially and ethnically diverse.

Conclusions and Implications for Research and Practice
The perceptions and attitudes of students, particularly low-income, racially and ethnically diverse elementary-aged students, are often not captured. This qualitative research study adds to the literature on why behavior change is so difficult, aligning with the Transtheoretical Model, which focuses on motivational readiness to change. 49 Knowledge of barriers to eating healthier foods may help better tailor future interventions and programs. Results from the focus groups also revealed how students can feel empowered to make changes not only to their own habits but to that of their families. Lastly, the students were able to express their enjoyment of the program, inform the researcher on topics and mode of delivery that impacted them most, and offer ideas of how to improve it for the future. This information is vital to expanding nutrition education programs out into the community in a sustainable way.
Future research should include evaluating and revising programs based on student suggestions. In addition, a pilot of the program taught by school staff should be conducted to assess the potential for sustainability.   It looks so much healthier than eating all that grease.

Chapter Two Figures and Tables
[It] has salad in it, and it has tomato, and it also has the-I know it has the protein. I told my mom that you two teachers told me not to eat, um, too much junk food. So now every time I go to, like, BJs or food markets, I start getting-my mom starts buying me vegetables and she doesn't give me any junk food or anything.
That every time when I eat dinner, my parents always give me soda and I say, 'no, don't give me soda, because save the soda for special occasion' and then they give me juice or water.

Perceived
Influence on Family Behavior I always ask my dad 'can we go and buy some Subway?' so he sometimes says yes and sometimes says no. So now in these times, actually, we don't go too much to those places. Now we just go to the supermarket and buy some vegetables, fruit, rice and chicken. I just have one thing to say is that when I was little, I didn't wanna eat yogurt. I didn't like yogurt. Now, I'm eating the whole cup of yogurt.
You told us to try new foods, like I never liked celery, but then I tried it and I liked it.

Table 4: Representative Quotations by Students (n=64) from Post-Intervention Focus Groups Regarding the 13-week School-Based Nutrition Education Program
Themes and Sub-Themes

Memorable Topics Nutrition Topics
We learned to not eat so much sugar and not to drink grown up drinks, and not to drink a lot of like, coffee stuff.
You have to eat vegetables like carrots so you can be very healthy, your body, because if you go to the hospital they might say 'have you been eating vegetables?' and you could say 'yeah, so my body could be healthy', and for your eyes.
I learned to eat a variety and not just eat one color [of fruits and vegetables] because that means that your body isn't getting that much of the colors that it needs on your body.
If you have whoa foods, those are only for parties and stuff like that.
Instead of picking a big French fry, picking a small French fry. Instead of picking a Big Mac, picking a small cheeseburger.

BQ-Specific Topics
And don't eat unhealthy food because you will be like [Trans] Fat Cat.
The body quest warriors. They had some powers about the food groups I remember that we used the [iPads] so that we can eat more healthy foods, and we can get healthy. We did the six of them. The thing that I wanted that is different, I wanted that it happens that we stay more longer in a class to learn more.

How to Make a Topic Memorable
What I would like to change about the program is that sometimes they should add more days like not just Wednesdays, they should do it Tuesdays, Fridays, Saturdays.

Curriculum Changes
We would be the teachers.
Instead of learning about sugars we could learn about other stuff that are in healthy food.
A game where we have a shield and a sword, and they're healthy. Then we can go to battle with unhealthy things…We can go fight with germs and everything.
If you could talk about how sports are also good for your body If we have extra time, we can have like free time on the iPads.

Recruitment
Initially, the "Family Night" program was just a program for parents.
Recruitment was through collaboration with a previously-established community group program in the school. However, very few parents of 3 rd graders became involved, thus new groups were created through cold calls to parents, advertisements and sign-up sheets during school events, and talking to parents during school drop-off and pick-up times. Unfortunately, these newly formed groups had poor attendance and through conversations with partners, it was deemed unsuccessful. It was noted however that for the parents that did attend, they often brought their children with them. As a result of this, the program format changed into a "Family Night" program for parents and children together.
Over the three years of programming, several recruitment methods were employed and refined in an effort to enroll a maximum number of families into the "Family Night" series. Figure 1 describes those methods.

Program and Retention
The "Family Night" program ran for 1.5 hours for six consecutive weeks. It Parents were asked to provide anecdotes on how they thought the program influenced their family as a result of participating in the "Family Night" program. Table 1 provides written quotes from parents.
Successes of the program include being able to continuously evolve and innovate the recruitment methods. This included changing the format of the program to encourage participation and increase retention. As a result, for the parents that participated they were able to improve their positive parenting practices as well as parent and child nutrition-related behaviors. However, there were several challenges.
Despite extensive efforts to recruit, the number of participants recruited into the program was small. Due to the small sample size, the study design was limited to nonexperimental pre-post which may have limited the ability to detect significant changes.
Lastly, it appeared that parents reported enjoying the program and making many behavior changes as a result. Unfortunately, this was not reflected in the survey data, indicating that the survey instrument may not have been sensitive enough to detect change or the sample size was too small.

CONCLUSION AND IMPLICATIONS
In conclusion, community programs that involve families are needed to foster a healthy home environment by improving parenting practices and also to improve children's nutrition-related behaviors. It appears that including the entire family in addition to communication with them via text messages are possible strategies to increase recruitment and retention. Future outreach and extension efforts should continue to explore novel ways to engage families, especially low-SES families that face increased barriers towards attending community outreach programs.

*method used in data not presented in this article
Flyers sent home with 3 rd grade students. Modifications included: • Advertisement of program incentives including a gift card for attending 5/6 lessons and raffle items • Student receive educational extender for returning parent-signed flyer indicating they can or cannot attend program • Classroom teacher competition-which classroom could return the most flyers • SMS text messaged classroom teachers to communicate about returned flyers and reminders to students • Stickers placed on 3 rd grader's t-shirt saying "Ask me about signing up for the family nutrition program" when flyers went home* • Classroom teachers added program advertisement to their family newsletter and sent it out via their email to parents/caregivers* • School principal posted program advertisement on school website* • One classroom teacher was the "lead" communicator with other classroom teachers and outreach educator (through SMS text message) and collected signed flyers from all classrooms* "Graduation Event" • held for completion of an in-school 3 rd grade student program to recognize students for success and sign parents/caregivers up for "Family Night" program.

SMS Text Messages
• Initial text message when they signed up for program; another message one week prior to start of program show the relationship of health to energy-dense snacks (EDS) and sugar-sweetened beverages (SSB). It will also describe the effect school-based nutrition education programs, specifically those involving technology or including family components, have on the student's EDS and SSB consumption. Lastly, it will provide background as to why students' perspectives and inputs are so valuable for a successful program.

Prevalence of obesity
Obesity among children in the United States continues to be an important public health problem. To assist in tracking the severity of obesity, the Centers for Disease Control (CDC) has provided more specific classifications of obesity based on Body Mass Index (BMI)-for-age charts. Class I obesity is defined as ≥95 th percentile (hereinafter referred to as "obese"), class II obesity as ≥120% of the 95 th percentile (hereinafter referred to as "extreme" obesity), and class III obesity as ≥140% of the

Why childhood obesity is a public health concern
Childhood obesity is a public health concern due to the consequences associated with an increase risk to several physiological and psychological diseases. and cost to the healthcare system. Childhood obesity is associated with increased physiological risks including cardiovascular disease, type II diabetes, hypertension, asthma and sleep apnea, joint problems, metabolic syndrome, and fatty liver disease. [2][3][4][5][6] Psychologically, obese children are at increased risk for anxiety and depression, low self-esteem, and social problems like bullying and stigma. [7][8][9] Compared to a normal weight 10-year old child who maintains a normal weight through adulthood, it is estimated that the incremental lifetime medical costs of an obese 10-year old child that remains obese as an adult is $19,000. 10 This is troubling as nearly one-fifth of schoolaged children are obese. 1 Given the potential consequences of childhood obesity, it is important to understand the contributing factors that can inform interventions.

Determinants of childhood obesity
There are several factors that can affect weight including race, ethnicity and socio-economic status as well as biological, social and environmental determinants. Various biological, social and environmental determinants also affect weight.
These determinants intertwine in children to lead to a greater propensity to be obese.
Biological sex, lack of safe places to play, food deserts where there is limited access to healthy, affordable food, 14 and home environments that allow for unhealthy food items and poor parent modeling all influence the risk of childhood obesity. 15 Of the social and environmental determinants, dietary behaviors such as fruits and vegetables (FV), EDS and SSB are important.

Fruits and vegetables
Contributing to the obesity epidemic among children is insufficient consumption of FV. 16 There are several nutrients in FV including vitamins, minerals and fiber that contribute to a healthy diet. 17 These nutrients not only help maintain a healthy weight, but may decrease the risk of chronic diseases associated with obesity. 18  populations all consume more fruit juice than the non-Hispanic White population (0.31 servings). 22 In regards to SES, there is an inverse association between SES and whole fruit consumption. 22 In fact, while 64.2% of high-SES 4-13 year-old children did not meet the recommendation of 1.5 servings of fruit in a day, 68.9% to 83.1% of lower-SES same-aged children did not meet the recommendation. 22 The same trend follows for vegetable consumption, with the non-Hispanic Black population (all ages) consuming 0.5 few daily servings of vegetables than the non-Hispanic White population and the lower-SES population consuming 0.22-0.35 fewer daily servings than the higher-SES population. 23 These disparities may be due to types of foods prioritized when shopping. Access to FV can be an issue in low-SES neighborhoods. Many low-SES neighborhoods are considered food deserts, lacking healthy affordable food. 14 Qualitative research has found low-SES parents of schoolaged children do not prioritize purchasing FV when on a limited budget because it is thought to be more expensive and less satiating than other food items. 24 These parents also found it frustrating that "junk food", such as EDS, was less expensive than FV. 24 Energy-dense snacks EDS contribute to the obesity epidemic not only because they may replace healthy FV options in a diet, but also because of their high energy density. 25 High energy-dense foods and drinks have a high amount of energy per gram of food. 26 Too much energy (in the form of kilocalories) can lead to weight gain. 27 Thus, snacks, including both salty and sweet foods, are a concern due to their energy density. 28 were surveyed. There were significantly higher consumption of SSB in non-Hispanic Black children compared to non-Hispanic White children, an inverse association between SSB consumption and parental education attainment, and a positive correlations between child SSB consumption and parental SSB consumption and child fast food consumption. 42 Thus, a combination of social and environmental factors may be predictive of SSB consumption.
Overconsumption of SSB, along with EDS, may be contributing to the obesity epidemic in children because of its high energy density and potential replacement of healthy foods like FV. This is particularly concerning with racial and ethnic minority or low-SES children whose diets are often less healthy than White or higher-SES

Theoretical frameworks
Schools are an ideal place to provide nutrition education to students as they are a captive audience. How much the student retains and puts into practice can depend on several curriculum factors. Behavior theories can help inform successful curriculums for health promotion programs. 44 The most common theories utilized in nutrition education programs include the Social Cognitive Theory (SCT) 45

and the Experiential
Learning Theory (ELT). 46 The premise behind Albert Bandura's SCT is reciprocal determinism, that is, learning and maintenance of a behavior happens in a social context where people learn through interaction, response to behaviors, and observation with others and their environment. 45 Aside from reciprocal determinism, other key constructs include observational learning, reinforcement, expectations, and selfefficacy. 44 The emphasis with David Kolb's ELT is the process of learning, where learning occurs and modifies with different experiences. 46 There are four stages: concrete experience, reflective observation, abstract conceptualization and active experimentation. 46 Thus, by providing students the opportunity to actively engage in a topic and make it applicable to themselves, their chances of learning, and subsequently, behavior change, increases.
FV consumption is often the focus of nutrition education efforts with schoolaged students, 47,48 with few concentrating on low-SES minority populations. [49][50][51][52][53][54][55] The development of obesity is complex with a number of environmental and individual contributors. Multi-level interventions have been successful in changing student behaviors. 56 This study specifically targets contribution of EDS, SSB and SES and may help inform effective obesity prevention focus areas. Thus the following is a thorough literature review of school-based nutrition education programs with low-SES students that focus on EDS and SSB.

Energy-dense snack focus in school-based nutrition education programs
There are very few school-based nutrition education programs that focus on energy-dense snacks with school-aged, low-SES students. In Lebanon, [9][10][11] year old students (n=188) from public (generally low-SES) and private (generally middle or high-SES) schools participated in a 12 lesson intervention. 57  Likert scale assessing how often sweet drinks such as soda, fruit-flavored drinks, and sports drinks were consumed (n=680). With a more specific SSB-related questionnaire that separated out fruit drinks and sweetened iced teas from other SSB, a different sub-sample of students (n=954) had a significant (p<0.001) decrease in how often fruitflavored drinks, sweetened iced teas, and soda/pop were consumed. 62 One limit to the study includes the data collection instrument as the reliability testing of practice-based instruments was restricted to internal reliability and not test-retest. Another limit was the study design, so while significant decreases on the frequency Likert scale were found, since there was no control group in the study, the assumption cannot be made that the intervention caused the behavior change.
Sichieri et al did use a control-group study design to assess the 7-month intervention on 1,140 4 th graders (9-12 years old) in 22 schools in Brazil. 63  Sichieri, a limit to this study is the lack of data collection on water consumption. In addition, while SSB drinks included soda and energy drinks, it also included fruit juice and flavored milk, sugar-containing drinks that do provide some nutrition. Since there were also several components to the intervention, the PSE or community aspects may have contributed to the decrease in SSB consumption. Lastly, the self-reported student data did not have any significant results. Self-reporting by school-aged students can be a challenge as they may over report, under report, or omit items. 65 To alleviate problems with student self-reporting SSB consumption, Feng et al asked parents to complete a survey of their student's typical weekday and weekend SSB consumption in their longitudinal, quasi-experimental intervention. 66 Five- the study found that daily TV time, fast food consumption, and types of SSB available at home were significantly (p<0.001) positively associated with predicting student's daily SSB consumption while family meals had an inverse (p<0.01) association. 66 One potential limit to this study is the inability to determine if the school-based education had an effect on SSB consumption or if the home visits (~40% of eligible families participated in this part of the project) contributed. Another limit is that while the parents may have more accurate recall of what their student consumed in the previous day, the parent is not with them throughout the day, so SSB consumption during time apart may be inaccurate.
The effectiveness of school-based nutrition education research on EDS and SSB consumption in low-SES school-aged students remains inconclusive, with SSB being more extensively studied. However, both unhealthy dietary habits warrant further investigation as results have been modest and often involve several intervention components that go beyond direct nutrition education with the student.
Additional strategies that more closely involve the students in school-based nutrition education may increase effectiveness of interventions.

Use of technology in nutrition education programs focused on EDS and SSB
Technology is integrated into students' lives at school and at home. While too much non-productive screen time is discouraged, there is value in productive screen time. 67 Students respond favorably to technology, a tool that has been shown to increase nutrition outcomes. 68,69 Two systematic reviews have focused on technology.
One systematic review focused on technology-based interventions that either targeted prevention or treatment of overweight and obesity in youth. 70 Of the 24 studies included in the systematic review, four involved school-aged students. Three focused on prevention of overweight and obesity by concentrating on fruit, vegetable, juice or physical activity. Two of the three studies found positive behavior changes due to the study; however, one had no control group to compare an interaction effect. 70 The second systematic review of technology involved media-based health interventions targeting behavior change in youth (not necessarily low-SES). 71 Like the other review, it found some significant improvements in the four studies that assessed dietary behaviors. 71 However, as noted in the review, interventions involving technology require more thorough and complete evaluation.
The research by Turnin et al 72 was one of the studies involved in the abovementioned systematic review. 71 In France, 1, 876 students (grades 3-5) participated in a research study to assess the effect of games on nutrition knowledge and behavior. 72 Over a 5-week period, the treatment group (n=1,003 students; 8 schools) received the 1-hour, twice a week nutrition education intervention through games while the control group (n=873 students; 7 schools) received the nutrition education through a teacher. These results were not sustained over the 2-year period. In regards to SSB, a limit to the reported data is only providing results of greater than or less than 400mL, not a smaller quantity. Smaller quantities may still be impactful in SSB consumption change.
Servings per day of SSB and FV, as well as screen time, were the focus of a 12-week mobile technology pilot RCT intervention with 9-14 year-old females (83.7% African American; 32.4% living in poverty) in Kansas. 74 The treatment group (n=26) received the technology that included setting goals for behavior change, selfmonitoring, and feedback and reinforcement on goal attainment. The control group (n=25) received manuals that contained screen shots from the electronic, treatment version of the intervention, and the control students had to initiate their own goal setting and self-monitoring, while receiving no feedback or reinforcement. One week day and one weekend day recall via 24-hour multiple pass method was used at baseline and week 8 for SSB with the female participants. While there was a decrease in SSB servings per day from baseline (M=1.20 serving/day; SD=0.92) to week 8 (M= 0.87 servings/day; SD= 0.93), it was not significant and the effect size was small/medium (d= -0.34). There was a significant association (r= 0.50, p=0.01) between the technology use and SSB consumption, with those girls who responded to more prompts had a greater reduction in SSB at week 8 compared to those who responded to fewer than six prompts. 74 As this was a pilot, the sample size was small.
In addition, it was with a specific population and thus cannot be generalized to males or other races/ethnicities. While school staff do their best to ensure accurate dosage of a program, research staff often implement the program themselves. This was the case with the U.

Quest to Lava
of Alabama's school-based nutrition education curriculum Body Quest: Food of the Warrior (BQ). 54 This study used iPad application technology with low-SES school-aged students to impact changes in FV consumption 54 and later in an additional study, intent to change SSB consumption. 75 The original quasi-experimental, clusterdesigned study by Struempler and colleagues assessed the effect of a 13-week technology-integrated nutrition education program on the FV consumption of 2,4777 3 rd grade students in 60 SNAP-eligible schools (i.e. ≥50% eligible for free/reduced meals). 54 Results showed a significant (p<0.001) increase in FV consumption during school lunch (the focus of assessment). The next school year, an additional clusterdesigned study was conducted with 3,568 3 rd graders from 80 SNAP-eligible schools 75 . In addition to the assessment of FV consumed during lunch, this study assessed knowledge, intention, and behavior of dietary and physical activity characteristics at weeks 1, 7, 12, and 17. In regards to beverages, the question "will you drink water instead of soda in the future?" was asked. At baseline there was no significant difference between groups in their response of "yes"; however, at postassessment, the treatment group (76.7%) responded "yes" significantly (p<0.001) more than the control group (64.0%). 75 However, a major limitation to the data collected was that it only assessed intent to change SSB consumption, not actual change in consumption.
Overall, technology-integrated school-based nutrition education programs involving low-SES school-aged students have shown positive results in EDS or SSBrelated changes. However, the studies have had limitations including no comparison groups, not meeting dose recommendations, not generalizable, and lack specific EDS and SSB consumption changes. Therefore, there is a need to assess the effect of a technology-integrated, school-based nutrition education program on EDS and SSB consumption of low-SES school-aged students through a two-group by two-time point study design.

Parent involvement in student's EDS and SSB consumption
In addition to technology, parent involvement may increase nutrition outcomes of children, as was shown by the above-mentioned Feng et al study. 66 Parenting practices are parent behaviors or actions towards their child that influence the child's attitudes, behaviors or beliefs. 76 These actions may be intentional or unintentional by the parent and include coercive control, structure and autonomy support. 76 Coercive control includes restriction of foods, pressure to eat, and threats or bribes. 76 Restriction of foods can negatively affect unhealthy food consumption. In a systematic review and meta-analysis, 5 out of 11 studies suggested that this parenting practice was associated with higher consumption of unhealthy foods by children ages 7-11 years old, 1 out of 11 studies found a decreased consumption, and 5 out of the 11 studies had non-significant findings. 77 When the parental practices were synthesized in a meta-analysis, pressure to eat (n= 9 studies; r= 0.04, 95% CI= 0.00, 0.08) and food as reward (n=4 studies; r= 0.14, 95% CI= 0.03, 0.25) were positively associated (p<0.05) with unhealthy food consumption. 77 Structure includes setting rules and limits with meal and snack schedules, modeling by parents of nutrition-related behaviors, food accessibility/availability in the home, and neglect or indulgence by the parent. 76 A cross-sectional study found such parenting practices to be associated with SSB consumption (specifically fruit drink/juice and soft drink) by children. 78 Children, ages 10-12 years, and their families across eight countries in Europe were involved (n=7,915 students; 6,512 parents).
Positive associations (p<0.05) were found between student SSB consumption and parent modeling, availability of SSB in the home, and consuming SSB with the parents. Permissiveness (allowing SSB), lack of monitoring, and low self-efficacy by the parent were associated (p<0.05) with increased soft drink consumption by children. 78 Autonomy support includes nutrition education by the parent, student involvement in meal preparation, encouragement, praise, reasoning and negotiation. 76 A cross-sectional study out of Canada found that with increased involvement by 5 th graders (n=3,398) in meal preparation, there was an increase in FV preference. 79 Rewarding with verbal praise has been associated (p<0,05) with a decrease in unhealthy food consumption, particularly in younger children (n=4 studies; r= -0.04, 95% CI= -0.07, -0.01). 77 80 Harris and Ramsey examined the association between African-American father's parenting practices and their child's SSB consumption. 81 The fathers (n=102) had children between the ages of 3-13 years old and completed usual consumption surveys for both themselves and their child. There were significant correlations between father's consumption of SSB and their child's SSB consumption (r=0.67; p<0.001), modeling (r= -0.21; p<0.05), and household availability (r= -0.36; p<0.001). 81 However, it is important to note this study did not compare outcomes of the African-American families to other races and ethnicities. Overall, parents can shape a child's eating behavior by their control over food, modeling of behaviors, child involvement with food, and availability and accessibility of foods in the home.
The Academy of Nutrition and Dietetics recommends a multicomponent intervention, including an interactive at-home parental component, to have the greatest impact on prevention. 82 Similarly, a systematic review on efforts to reduce childhood obesity found the most effective methods included both school-and parent-based aspects. 83 The combination of technology-integrated nutrition education and parentinclusive research is scarce. To date, these studies have mostly focused on FV outcomes, [49][50][51][53][54][55] with one study to the researcher's knowledge that has included EDS or SSB in their research. 75 Parent involvement in the interventions have varied from newsletters sent home with the students 49,54,55,75 to nutrition education classes for the parents themselves. 50,51,53 More parent-inclusive and parent-involved nutrition education programs that focus on an outcome of student's EDS and SSB consumption are needed.

Improvements to Nutrition Education Programs
While nutrition education programs have shown effectiveness in behavior change, there is always room for improvement. One way to improve these programs is by incorporating student feedback into nutrition education programs. Research has found that qualitative research conducted with students provides meaningful information to improve programming. 84 The Health-E-PALS pilot study in Lebanon conducted focus groups with students after a multicomponent, school-based intervention. 57 The goal of the focus groups was to determine the perception of the program, potential improvements to the program and also what the students perceived that they learned. The students found the interactive, fun activities most beneficial to their learning and behavior change, but wish the program was longer in duration. 57 Similarly, Grassi et al explored students' perspectives following a 10-week intervention in Italy. 85 The focus groups provided more insight to responses also obtained quantitatively, information on positive and negative reinforcements to behavior change, and explored the student's satisfaction with the program. 85 The information gained by focus groups can help refine curriculum content for more successful programs.
While the above focus groups aimed to improve programming, focus groups can also set out to provide formative information to create programming. Boddy et al aimed to learn what factors influence student's behaviors, both positively and negatively, as formative research to develop the CHANGE! school-based curriculum intervention. 86 Their qualitative analysis found the largest influence on student's nutritional habits were parents and their parenting practices, such as role modeling or rule setting, as well as siblings and grandparents. Barriers to healthy eating not only included parenting practices such as food as reward, but also preferred taste and smell of unhealthy foods, advertisement and convenience. 86 Knowing these influences and barriers to healthy eating can help mold an effective program.
The students' perspectives may help provide a more complete picture on how a school-based nutrition education program can impact what they eat. Students' perspectives may also provide insight to help guide future programming. However, few studies have incorporated feedback from low-SES, racially and ethnically diverse school-aged students. As this population is vulnerable to an increased risk for obesity, it is important to determine how to best intervene, from their perspective, to promote healthy eating habits.

Conclusion and gaps
As obesity continues to be an important public concern, the eating habits of school-aged students are imperative for their current and future health. For the secondary aim, qualitative data was collected via semi-structured focus groups with treatment group 3 rd graders. It was expected that the 3 rd grade students would have a positive experience with the 13-week school-based nutrition education program and make dietary behavior changes as a result of the program. In addition, it was predicted that students would report on several barriers to becoming and staying healthy such as lack or overabundance of foods and taste preferences. Focus groups were conducted before the final follow-up data collection with treatment groups in years 2 and 3.
The tertiary, exploratory aim, hypothesized that students who were exposed to an additional 6-week "Family Night" program would have a larger decrease in EDS and SSB consumption when compared to students who only receive the 13-week school-based nutrition education program. This "Family Night" program occurred in between the post-assessment (week 13) and follow-up (week 27) data collection.
Therefore, for this 2x2 design, post-assessment and follow-up data collected from 3 rd graders in the treatment schools were used in analysis.
The University of Rhode Island's ethics committee granted internal review board approval for this research study (IRB#HU1415-015).

Recruitment and Participants
Providence, RI is one of the four core cities in the state, with an average 87.7% year. Sixteen semi-structured focus groups with four 3 rd grade students in each group were completed. 89 Sixty-four out of a possible 138 students who received the program were willing to participate and were selected by the classroom teacher. Thematic saturation was reached after 16 focus groups in two schools were conducted.
Lastly, for the tertiary aim, the treatment school 3 rd graders who were involved in the "Family Night" programming were compared to treatment school 3 rd grader who only received the 13-week school-based program. Initially, the "Family Night" series program was just a parent/caregiver program (as the 3 rd grade students received the 13-week school-based program during the school day and an after school program if they participated in after school activities at the school). Recruitment occurred through collaboration with a previously-established community group program in the school. As there were very few parents/caregivers of 3 rd graders involved in the community group program in the school, new groups were formed through cold calls to parents/caregivers (Appendix B), advertisements and sign-up sheets during school events (Appendix C), and talking to parents/caregivers during school drop-off and pick-up times. These newly formed groups had poor attendance. As the parent/caregiver-only program was unsuccessful and parents/caregivers often brought their children with them, the program format changed. With children being interested in coming to the program, and realizing that students enjoyed the school-based and after school programs, the parent/caregiver program was converted into a "Family Night" series program. Thus, even if the parents/caregivers were not totally interested in the program, their children were, and the families would attend.
Over the three years of programming, several recruitment methods were employed in an effort to enroll a maximum number of families into the "Family Night" series. One recruitment method included flyers sent home with the students describing the program (Appendix D). These flyers were modified each year to increase recruitment. One modification included asking parent/caregivers to sign the flyer regardless as to if the family were planning to participate in the "Family Night" series (Appendix E); this ensured that the parent/caregiver read the flyer. Another was offering a nutrition education reinforcement item to the student for bringing back the signed flyer.
Another recruitment method was holding a "Graduation Event" for completion of the 13-week school-based program. This recognized the students for successful completion of the program and also provide the opportunity for parents/caregivers to discuss and sign up for the "Family Night" series. Lastly, SMS text messages were sent to parents/caregivers who signed up to remind them of the start date and time of the program. Over three years of data collection, 25 3 rd graders and their families from treatment schools participated in the "Family Night" programming, while the remaining 192 students from treatment schools did not participate.

Instruments
The primary and tertiary aims used a student survey that relied on self-recall of previous day's consumption. Cognitive interviews were held with six 3 rd grade students for understanding and clarity of the student survey instrument. Students were chosen by the classroom teacher at random. Students did not understand the terms "Hispanic" or "non-Hispanic", but instead understood when the interviewer asked if they spoke Spanish at home. Based on this information, ethnicity was determined by if the student spoke Spanish at home. Additionally, "other" and "not sure" categories were added as options to the race question, as some students did not identify with any option provided or were unsure of their race. No changes to the nutrition-related behavior questions on sweet snacks, salty snacks, SSB, fruits and vegetables were made and it was decided that a standard script would be provided when administering the survey for uniformity.
The student survey (Appendix F) included instruments provided by USDA CYFAR and those adapted from the Beverage and Snack Questionnaire. 90 Through self-recall, nutrition-related behavior questions assessed the following: "how many times did you eat a sweet snack yesterday between your meals?" (and same for salty snacks), "how many times did you drink a sugary drink yesterday? Do not include examples of SSB included soda, sports drinks, energy drinks, sweetened iced tea, and fruit drinks. All questions were multiple choice, with range option so of "0 times" to "5 or more times" consumed.
A semi-structured focus group guide was developed for the secondary, qualitative aim (Appendix G). This guide was pilot tested with a small group of sameaged children (n=4) for comprehension and clarity of questions. No questions were changed, but the order of asking questions was altered so a response to one question would not influence conversation later in the focus group. The semi-structured focus group guide was organized by five sections: influence of food selection, memorable topics from the curriculum, perceived behavior change from the program, barriers to behavior change, and potential changes to the program.

Protocol for Data Collection
Each 3 rd grade student was assigned a unique identification (ID) number. The For the secondary aim, semi-structured focus groups were conducted during the school day in a quiet location within the school the students attended. 91,92 The lead researcher was the moderator and the nutrition educator was the note taker, both of whom the students knew through the program. The moderator asked students if they could be taped via voice recorder, explained why the nutrition educator was taking notes, and provided ground rules and expectations of the conversation. Each focus group lasted approximately 20 minutes. Data saturation was reached when coding of data revealed no new themes.

Intervention
The primary aim used a modified version of the U. of Alabama's Body Quest: Food of the Warrior (BQ) curriculum (Appendix L). 54 The curriculum involved interactive, hands-on activities as well as seven iPad applications created for the BQ curriculum to reinforce topics taught by the educators who were registered dietitians.  Did anything you learn help you change the foods you eat and drinks you drink?
What are some things that you may be doing that you think may not be healthy? Can you tell me more about that? Does anyone have anything else you would like to say about the nutrition program last year? 1-hour lessons and was enhanced by URI CYFAR nutrition materials. It taught nutrition topics and positive parenting practices leading to nutrition and physical activity behavior change in the home. Nutrition topics included sugar-sweetened beverages, fruits and vegetables, fast food, serving sizes, family meals, screen time, and physical activity. Positive parenting practices included firm and responsive parenting, shaping the child's environment, leading by example, division of responsibility in regards to meal time, and the concept of "can do". Parents/caregivers also learned how to make small, attainable goals to slowly make behavior changes.
These 1-hour lessons were conversational and interactive between the educator and other parents/caregivers.
During the 1-hour of education, the children received a brief nutrition lesson, spent time on iPads to create content for a program recipe book, and prepared and cooked the recipe they were offered at dinner. Through interactive, hands-on activities, the brief nutrition lesson reviewed topics learned during the 13-week school-based program. Such topics included food groups and "Go, Slow, and Whoa" foods, MyPlate, variety and amount of fruits and vegetables, the importance of breakfast, and how to add vegetables into meals. The children also used iPad applications such as Doodle and iBook to create content and put together a recipe book as a final product of the "Family Night" series program. However, the real focus of the lesson was hands-on cooking. Using a modified version of the Cooking with Kids curriculum, 94 children had the opportunity to prepare, cook and sample the recipe served to them for dinner that evening (Appendix N). This showed the ease of making the recipe and motivated the children to try the recipe if they had not during dinner.

Data Analysis
Sample Size G*Power version 3.0.10 was used to calculate sample size for the primary hypothesis. Sample size calculations were performed based on expected changes in SSB and EDS from year-one data. 95 The treatment group (n=70) had a significant decrease in SSB intake by a mean of 0.943±1.849 times per day and a significant decrease in EDS by 0.700±1.408 times per day; the control group (n=59) had no change in SSB intake (0.000±1.771 times per day) and a significant decrease in EDS by 0.415±1.402 times per day. 95 A required sample size of 118 and 768 3 rd graders are necessary to determine the effect of the intervention on SSB and EDS, respectively, with an alpha set at 0.025 and statistical power at the 0.80 level.

Quantitative Analysis
For the primary and tertiary aims, the quantitative data collected via Surveymonkey.com was exported into Excel, saved, cleaned, and analyzed in IBM SPSS software (version 24.0, IBM SPSS Statistics, Armonk, NY, 2016). Numerical (skewness and kurtosis) and graphical (histogram) methods were used to determine normalcy. One additional variable was created from survey questions: "EDSAVG" (sweet and snack variables combined and averaged, Cronbach alpha .719).
For the primary aim, baseline Pearson Correlation between variables was run for both treatment and control groups. Independent t-tests and chi squared assessed any differences between the treatment and control group at baseline (week 1) for continuous and categorical variables, respectively. Paired t-tests were used to assess within group differences from baseline (week 1) to post-assessment (week 13) and repeated analysis of variance for between group differences of EDSAVG, sweet snacks, salty snacks, and SSB. To account for the study design in which two out of the four schools served as both treatment and control groups within the three years, paired t-tests were ran for EDSAVG, sweet and salty snacks separately, and SSB for each treatment and control group involved in each year of data collection. This analysis detected if changes in behavior were due to the intervention and not the school, as well as if it was replicable with different schools and students. Significance was set at p <0.05.
Similar statistics were run for the tertiary aim. Baseline Pearson Correlation between variables was run for both the treatment students who did the 13-week school-based program and their families attended the "Family Night" series program and for the treatment students who only participated in the 13-week school-based program. Independent t-tests and chi squared assessed any differences between the two groups at baseline for continuous and categorical variables, respectively. Paired ttests were used to assess within group differences from post-assessment (week 13) to follow-up (week 27) and repeated analysis of variance for between group differences of EDSAVG, sweet snacks, salty snacks, and SSB. Significance was set at p <0.05.

Qualitative Analysis
For the secondary aim, audio-recorded focus groups (n=16) were transcribed verbatim by a transcription service and were checked by the focus group moderator and note taker for accuracy. Each transcript was coded using a hybrid approach of inductive and deductive thematic analysis. 96 This approach acknowledged the sections in the focus group protocol and also included any additional themes that emerged from the data during the coding process. The lead researcher utilized thematic analysis to detect themes from the content of the transcripts. 97 A codebook of structural and content codes was created and updated based on transcription readings. A second researcher coded 25% of the data and inter-rater agreement was calculated. There was a 94% agreement of coding, determined by the number of agreements divided by the sum of agreements and disagreements. These codes led to patterns and themes within each section.
¡Esperamos que usted también se pueda involucrar! Usted es muy importante para los hábitos alimenticios de su hijo(a). Si le gustaría participar, se le pedirá que llene una encuesta de 15 minutos dos veces (6 semanas de diferencia) que le hará preguntas como de sus hábitos alimenticios, los hábitos alimenticios de su hijo(a) de 3 er grado y las prácticas de sus comidas familiares; qué tanto le gusta usar iPads; e información sobre su edad, nivel de educación y etnicidad. Como un agradecimiento por su ayuda, le regalaremos un utensilio de cocina tal como una tabla de cortar para apoyar una alimentación saludable. Todas las preguntas son opcionales. Los detalles se mandarán a casa con su hijo(a) de 3 er grado. Si tiene cualquier pregunta, por favor comuníquese con el Coordinador The purpose of the focus group is to determine the following: 1) Do the children feel they made any changes in their food/beverage behavior over the last year, were these changes related to the program and, if so, what was it about the nutrition program that affected their food/beverage behavior? 2) What are the barriers to becoming and staying healthy and did the program help reduce these barriers? If It did help reduce the barriers, what was it about the program that helped? 3) What would they like to see if we could change the program in the future? To help the students answer the questions honestly, make them feel welcome, explain that there is no right or wrong answer and that they are not being judged or graded on what they say. Explain that they are here to help us determine what works and what does not work with providing nutrition education to 3 rd graders.
Say: Thank you so much for coming! Today we are going to talk about eating habits. I am going to ask some questions, and after each question I will give you all some time to talk and answer if you want to. Remember, we want to hear what everyone has to say, so make sure to take turns and speak one at a time and let's make sure that we listen and respect each other. Also, you don't have to go in order, if you have something to say you can just say it. You do not have to answer a question if you do not want to. But, just so you know, there is no right or wrong answer and you will not be graded on anything you say. We just want truthful answers. Can anyone tell me what it means to tell the truth? Are there any questions before I get started?  Let's start with looking at some pictures (have a piece of paper with two choices, one healthy and one unhealthy). Which would you choose to eat? (use this as a baseline to start conversation)  Why did you choose the food that you did?  What do you remember learning last year in the class I taught? (make a list as a group on a large piece of paper) o Prompt: if they do not remember the curriculum, show them a picture of the BQ warriors  Did anything you learn help you change the foods you eat and drinks you drink? o Probe: Learning is one thing, but actually doing something because of it is another! For example, we can learn that milk is healthy to drink every day, but it does not mean we will do it, right? So, is there anything you learned that had an effect on what foods you eat?  What are some things that you may be doing that you think may not be healthy? Can you tell me more about that? o Probe: What do you like about these foods? How do you feel about these foods?  Does anyone have anything else you would like to say about the nutrition program last year?
o Probe -what would you like to see changed? What did you really like?  Thank you so much for taking the time to meet with me today. This information has been very helpful.
well these workshops change behaviors that may affect the health of you and your family and help us improve programs for other families.
Confidentiality: Your part in this study is confidential. None of the information will identify you by name. All records will be stored in a locked file cabinet at 80 Washington Street, Providence, RI.
Decision to quit at any time: The decision to take part in this study is up to you. You do not have to take part. If you decide to take part in the study, you may quit at any time. If you choose to quit, you can continue to participate in other adult programs and your children will not be affected in any way. FORMULARIO DE CONSENTIMIENTO PARA INVESTIGACIONES Se lo invita a participar en el proyecto de investigación de la Universidad de Rhode Island, descrito a continuación. El investigador le explicará el proyecto en detalle. No vacile en hacer las preguntas que desee. Si más tarde desea hacer otras preguntas, puede llamar al 401-874-2253 para comunicarse con Linda Sebelia y Kate Balestracci, quienes responderán sus consultas. Para participar en este proyecto de investigación, usted debe tener 18 años como mínimo. Descripción del proyecto: Este estudio se llevará a cabo con 150 padres de niños de tercer grado, de cuatro escuelas de Providence. El objetivo del proyecto es alentar a los padres para que elijan alimentos más saludables para sus familias.
Qué se hará: Si decide participar en este estudio, asistirá a un taller de nutrición dirigido a padres, que durará 6 semanas y donde se tratarán temas de alimentación de los hijos, mejora del rendimiento del dinero que se gasta en alimentos y cocina. El taller será dictado por nutricionistas de la Universidad de Rhode Island y se realizará en el edificio de la escuela comunitaria de servicio completo. Cada sesión durará una hora y media. Además, se le enseñará a usar un iPad. Al principio y al final del programa, se le pedirá que responda a algunas preguntas. Responder a esas preguntas le tomará unos 15 minutos.

Decision to quit at any time:
The decision to take part in this study is up to you. You do not have to take part. If you decide to take part in the study, you may quit at any time. If you choose to quit, you can continue to participate in other adult programs and your children will not be affected in any way. If you wish to quit, tell Linda Sebelia or Kate Balestracci 401-874-2253 of your choice. FORMULARIO DE CONSENTIMIENTO PARA INVESTIGACIONES Se lo invita a participar en el proyecto de investigación de la Universidad de Rhode Island, descrito a continuación. El investigador le explicará el proyecto en detalle. No vacile en hacer las preguntas que desee. Si más tarde desea hacer otras preguntas, puede llamar al 401-874-2253 para comunicarse con Linda Sebelia y Kate Balestracci, quienes responderán sus consultas. Para participar en este proyecto de investigación, usted debe tener 18 años como mínimo. Descripción del proyecto: Este estudio se llevará a cabo con 150 padres de niños de tercer grado, de cuatro escuelas de Providence. El objetivo del proyecto es alentar a los padres para que elijan alimentos más saludables para sus familias.