A Collaborative Extension GardenBased School Nutrition Program Measuring

A Collaborative Extension Garden-Based School Nutrition Program: Measuring the Interests, Behaviors, and Self. Efficacy of Third Grade Youth Participants and Their Families Matthew Kararo Dr. Kathryn Orvis Dr. Neil Knobloch Dr. Linda Prokopy

Introduction � 32% of youth, 68% of adults in U. S. are overweight (Flegal, et al. , 2010; Ogden, et al. , 2010) �Only 8. 8% of adolescents meet the 5 -a-day F&V recommendations (CDC, 2009) �Third graders are forming their habits for the rest of their lives (Crain, 2005) �May be first generation to not have higher life expectancy (Olshansky, et al. , 2005)

Eat Your Way to Better Health �Extension offered school-based garden nutrition program �Offered since 2007 (Light, 2007) �Funded by Indiana State Department of Health �Cooperation between CES & local schools �Uses JMG® curriculum plus additional materials and a school garden � 6 -8 weeks of lessons, once a week, at least 2 hours/day in third grade classrooms

Purpose The purpose of this study was to describe the relationships between youth participation in the Eat Your Way to Better Health (EYWTBH) program and variables pertaining to the eating habits of the youth and their families.

Conceptual Framework Personal Factors Five research questions looked at these personal and environmental factors of F&V consumption Environmental Factors

Theoretical Framework �Social Cognitive Theory Personal Factors Environmental Factors Behavior (Bandura, 1977; Bandura, 1986; Bandura, 1989) �Expectancy-Value Theory Expectancy Behavior Value (Ryan & Deci, 2000; Wigfield & Eccles, 2000)

Review of Literature Instrument Development Baranowski (2003), Domel, et al. (1996), and Watson, et al. (2006) have developed reliable instruments with this age group and were used as a basis for the EYWTBH evaluation. Theoretical Framework Social Cognitive Theory (Bandura, 1989) has been used in many youth nutrition education studies (Blanchette & Brug, 2005; Geller, et al. , 2009; Morris & Zidenberg. Cherr, 2002) Nutrition Education Program Results Studies that integrate the most ecological levels (family, school, community) show the strongest impact on healthy behavior, such as F&V consumption (Gentile, et al. , 2009)

Methodology �Pre-experimental design �Pre- and Post-EYWTBH both Youth & Parent One week before beginning of program One week after end of program �Administered by Extension Educators in Indiana third grade classrooms �Completely quantitative

Instrument �Original instrument was used in pilot test, both quantitative and qualitative (Light, 2007) �Two versions used in this study Version 2 (Fall 2008, Spring 2009, Fall 2009, Spring 2010) Version 3 (Fall 2010)

Instrument VERSION 2 VERSION 3

Instrument YOUTH � � � Youth fruit consumption (20 items) Youth veg consumption (21 items) Youth self-efficacy (11 items) Youth fruit preferences (20 items) Youth veg preferences (13 items) Youth social intentions (positive or negative) PARENT/GUARDIAN � Family fruit consumption (20 items) � Family veg consumption (21 items) � Family fruit availability (20 items) � Family veg availability (13 items)

Population Summarized Racial Demographic Response Data for Fall 2010 (Version 3) Semester Fall 2010 Youth (N=222) Parent (N=117) African. American Indian Asian Pacific Islander Hispanic White, not Hispanic 10 16 1 3 1 172 0 0 0 114 Summarized Household Annual Income Response Data for Fall 2010 (Version 3) $20, 001$40, 001 Semester $0 -$20, 000 $40, 000 $60, 000 Parents/ Fall 2010 12 22 21 Guardians (N=108) Same as Indiana median More than one 19 3 $60, 001$80, 000 $80, 001$100, 000 $100, 001 and up 27 13 13

Population Summarized Parent/Guardian Participant Education Level Response Data for Fall 2010 (Version 3) High 2 -year college 4 -year college Semester Grade/middle school School degree Parents/ Fall 2010 5 47 26 20 Guardians (N=119) Graduate degree Summarized Location of Residence Demographic Response Data of youth and parents for Fall 2010 (Version 3) Medium City Small Town Suburb of a city Large City with (between Semester On a Farm (<10, 000 with >50, 000 10, 000 and >50, 000 people) people 50, 000 people) Youth 78 103 16 2 15 (N=218) Fall 2010 Parent 33 53 23 3 7 (N=119) 21 Very Large City with >100, 000 people 4 0

Instrument Reliability Index reliability (Cronbach’s alpha) for Version 2 and Version 3 survey instruments Index Youth fruit consumption behavior Youth vegetable consumption behavior Family fruit consumption behavior Family vegetable consumption behavior Youth self-efficacy for food choices and habits Youth fruit snack interest Youth vegetable snack interest Youth junk food snack interest Youth healthy eating social intentions Household fruit and vegetable availability Household fruit and vegetable consumption Version 2 Pre Post α =. 84 α =. 85 α =. 82 α =. 85 α =. 62 α =. 66 α =. 62 α =. 63 α =. 86 α =. 88 α =. 82 α =. 85 α =. 80 α =. 79 α =. 92 α =. 91 α =. 83 α =. 90 α =. 84 α =. 86 α =. 73 α =. 77 Version 3 Pre Post α =. 82 α =. 84 α =. 80 α =. 67 α =. 66 α =. 74 α =. 67 α =. 82 α =. 86 α =. 89 α =. 91 α =. 82 α =. 86 α =. 92 α =. 87 α =. 91 α =. 85 α =. 79 α =. 81 α =. 79

Key Research Questions �Was there a difference in youth F&V consumption after EYWTBH? �Was there a difference in youth food choice self-efficacy after EYWTBH? �Was there a difference in youth F&V preferences/interests after EYWTBH? �Was there a difference in youth healthy eating social intentions after EYWTBH? �Can a model be constructed?

Conclusion #1 �Three relationships explain a portion of the variation in youth F&V consumption after participation in EYWTBH (36. 3%) Model Standardized Beta Coefficient t Significance Zero-order correlation Collinearity Statistics Tolerance VIF Constant -1. 20. 231 Youth fruit/ vegetable . 480*** 9. 18. 000. 55. 92 1. 09 consumption preprogram Youth self-efficacy . 188*** 3. 66. 000. 29. 96 1. 05 after program Family fruit/ vegetable . 163** 3. 19. 002. 26. 96 1. 04 consumption postprogram Note: R 2 =. 304 for Step 1, ΔR 2 =. 033 for Step 2 (p <. 001), ΔR 2 =. 026 for Step 3 (* p <. 05, ** p <. 01, *** p <. 001).

Conclusion #2 �There was a relationship between youth self- efficacy and EYWTBH participation, because post-program youth self-efficacy was higher than pre-program youth self-efficacy Youth self-efficacy index pretest versus posttest for Version 2 Index N Pretest Mean Posttest Mean Youth self-efficacy for food choices and habits 3. 03 3. 11 674 SD = 0. 69 SD = 0. 72 Mean Difference 0. 08 SD = 0. 62 P-value Effect Size 0. 001 A 0. 11 Youth self-efficacy index pretest versus posttest for Version 3 Index N Pretest Mean Posttest Mean Youth self-efficacy for food choices and habits 2. 93 3. 12 168 SD = 0. 73 SD = 0. 76 Mean Difference 0. 19 SD = 0. 76 P-value Effect Size 0. 001 A 0. 27 A Significant at the alpha = 0. 05. SD = Standard deviation. Scale: 1 = disagree a lot; 2 = disagree a little; 3 = agree a little; 4 = agree a lot.

Conclusion #3 �There was no relationship between household availability of F&V and EYWTBH Index Household fruit and vegetable availability (Version 2 data) Household fruit and vegetable availability (Version 3 data) N Pretest Mean Posttest Mean Difference 357 8. 73 SD = 5. 00 9. 09 SD = 5. 11 10. 93 SD = 5. 42 11. 01 SD = 4. 72 87 P-value Effect Size 0. 36 SD = 4. 56 0. 14 NS 0. 07 0. 08 SD = 5. 67 0. 90 NS 0. 02 A Significant at the alpha = 0. 05. SD = Standard deviation.

Implications �Unique and reliable instrument �When published, instrument will be used by other garden-based nutrition programs �EYWTBH is an Extension offered program, no longer grant funded, but still producing data Retest results in future with Scantron data �Data can be used in impact statements

Recommendations �Improve precision of instrument Measure not only diversity, but also frequency �Increase parent/guardian involvement Weekend garden work days/family night More take-home materials �Increase education about junk food Steady level of junk food interest was observed �Increase the length of EYWTBH

A Collaborative Extension Garden-Based School Nutrition Program: Measuring the Interests, Behaviors, and Self. Efficacy of Third Grade Youth Participants and Their Families Matthew Kararo Dr. Kathryn Orvis Dr. Neil Knobloch Dr. Linda Prokopy