Influence of National School Lunch Program to Children's Food Preference and Children's Influence in Family Purchase

MISSING IMAGE

Material Information

Title:
Influence of National School Lunch Program to Children's Food Preference and Children's Influence in Family Purchase
Physical Description:
1 online resource (64 p.)
Language:
english
Creator:
Jiang, Yuan
Publisher:
University of Florida
Place of Publication:
Gainesville, Fla.
Publication Date:

Thesis/Dissertation Information

Degree:
Master's ( M.S.)
Degree Grantor:
University of Florida
Degree Disciplines:
Food and Resource Economics
Committee Chair:
HOUSE,LISA ANN OFFENBACH
Committee Co-Chair:
GAO,ZHIFENG
Committee Members:
BLIZNYUK,NIKOLAY A

Subjects

Subjects / Keywords:
children -- decision-making -- family -- food -- nslp -- preference
Food and Resource Economics -- Dissertations, Academic -- UF
Genre:
Food and Resource Economics thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract:
The National School Lunch Program is the second-largest government food-assistance program in the United States. Over 80% of all primary and secondary schools participate in the NSLP. Given this influence, it is an interesting question to investigate the effects of school meals on children's food preferences. In addition, considering that children have an impact on family's purchasing decisions as they directly and indirectly influence their parent's decision-making; a secondary goal of this study is to explore how children influence the family's grocery shopping. Data for this research was collected using a survey completed by school children and their parents. A convenience sample of one hundred and sixty-three matched pairs of 4th and 5th graders and their parents was collected. Results showed that although the school meal is consumed frequently, the quality is not as rated as high as home foods. Other responses indicate children play an important role in the family purchase decision, and the different children’s resource—children’s contribution towards housework, grade and their knowledge has a significant correlation with the degree of children’s influence.
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility:
by Yuan Jiang.
Thesis:
Thesis (M.S.)--University of Florida, 2013.
Local:
Adviser: HOUSE,LISA ANN OFFENBACH.
Local:
Co-adviser: GAO,ZHIFENG.

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Classification:
lcc - LD1780 2013
System ID:
UFE0046036:00001


This item is only available as the following downloads:


Full Text

PAGE 1

1 By YUAN JIANG A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILL MENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2013

PAGE 2

2 2013 Yuan Jiang

PAGE 3

3 To my Mom and Dad

PAGE 4

4 ACKNOWLEDGMENTS I would like to take this opportunity to give my gratitude to those who have helped me in the way of completing the project. I would thank my supervisor Dr. Lisa House whose profound knowledge in the field provides me valuable advices and sparks off my insightful thinking. And I am grateful for her tremendous help, support, patience and encouragement throughout this wonderful journey of challenge and fulfillment. All the achievements during the course of my study would not have been possible without her guidance and advice. I would also express my great appreciatio n to my co chair professor Dr. Zhifeng Gao He has offered me considerable help and guidance in various aspects, including software, model design, and cheerful encouragements. Besides, my great thankfulness also goes to my committee member, Dr. Nikolay Bli znyuk for his suggestions and help in the data anal ysis. And I would also thank Dr. Steven M. Slutsky, Dr . Demetris Athienitis and Dr Zhihua Su for their assistant with my graduate training in various aspects. I would like to thank the Department of Food and Resource Economics for providing me the chance to study an d obtain voluble professional training. Without its support, nothing would be porssible. In addition, I would express a heartfelt appreciation my dear friends and fellow graduate students for their support and encouragements. From the bottom of my heart, I want to express my gratitude to all my family for their love and supports, especially my parents.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ ............... 4 LIST OF TABLES ................................ ................................ ................................ ........................... 7 LIST OF FIGURES ................................ ................................ ................................ ......................... 9 ABSTRACT ................................ ................................ ................................ ................................ ... 10 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .................. 11 Study Background ................................ ................................ ................................ .................. 11 Purpose and Aim of the Study ................................ ................................ ................................ 12 2 LITERATURE REVIEW ................................ ................................ ................................ ....... 14 ................................ ................................ .... 14 Individual Factors ................................ ................................ ................................ ............ 14 Economic Factors ................................ ................................ ................................ ............ 16 Social Factors ................................ ................................ ................................ .................. 17 C ................................ ................................ 18 Product Types ................................ ................................ ................................ .................. 20 Decision Stage ................................ ................................ ................................ ................. 20 ................................ ................................ ............................... 20 Family Characteristics ................................ ................................ ................................ ..... 22 ................................ ................................ ................................ ... 23 3 DATA ................................ ................................ ................................ ................................ ..... 24 Methodology ................................ ................................ ................................ ........................... 24 Descriptive Statistics ................................ ................................ ................................ .............. 25 Demographic s ................................ ................................ ................................ .................. 25 Preference and Knowledge of School Meals ................................ ................................ ... 26 Family Charac teristics ................................ ................................ ................................ ..... 27 ................................ ................................ ............................... 27 4 INFLUENCE OF NATIONAL SCHOOL PROGRAM TO CHILDREN'S FOOD PREFERENCE ................................ ................................ ................................ ....................... 34 General Data Overview ................................ ................................ ................................ .......... 34 Hypothesis Testing ................................ ................................ ................................ ................. 35 5 ................................ ........ 40

PAGE 6

6 Model ................................ ................................ ................................ ................................ ...... 40 Ordered Probit Model ................................ ................................ ................................ ...... 40 Model Specification ................................ ................................ ................................ ......... 42 Results ................................ ................................ ................................ ................................ ..... 44 ................................ ........................ 44 ................................ ................................ ............................... 47 Family Shopping Habits ................................ ................................ ................................ .. 48 Summary ................................ ................................ ................................ ................................ 49 6 CONCLUSION ................................ ................................ ................................ ....................... 56 7 IMPLICATI ON ................................ ................................ ................................ ...................... 60 LIST OF REFERENCES ................................ ................................ ................................ ............... 61 BIOGRAPHICAL SKETCH ................................ ................................ ................................ ......... 64

PAGE 7

7 LIST OF TABLES Table page 3 1 Gender of parents ................................ ................................ ................................ ............... 28 3 2 Gender of children ................................ ................................ ................................ ............. 28 3 3 Marital status of parents ................................ ................................ ................................ ..... 28 3 4 Education level of p arents ................................ ................................ ................................ .. 29 3 5 Number of children in the family ................................ ................................ ....................... 29 3 6 Eligibility for school meals ................................ ................................ ................................ 29 3 7 Preference towards school meals and home meals ................................ ............................ 30 3 8 Frequency of children going grocery shopping with parents ................................ ............. 30 3 9 Frequency of parents agreeing to buy items at the grocery store if child asks for something ................................ ................................ ................................ ........................... 30 3 10 Self ................................ ..................... 30 3 11 Frequency of doing housework as reported by the children ................................ .............. 31 4 1 ..................... 37 4 2 ......... 38 4 3 meals and home meals: one sample statistics ....... 38 4 4 test ..... 38 4 5 test .................. 39 4 6 Fishers test for foods eaten at home and at school ................................ ............................. 39 5 1 ................. 50 5 2 Variables used in the ordered probit model ................................ ................................ ....... 51 5 3 Sample descriptive statistics ................................ ........................ 53 5 4 Sample descriptive statistics s characteristics ................................ .................... 54 5 5 Sample descriptive statistics shopping characteristics ................................ ...................... 54

PAGE 8

8 5 6 Ordered probit model result ................................ ................................ ............................... 55

PAGE 9

9 LIST OF FIGURES Figure page 3 1 Age distribution of parents ................................ ................................ ................................ 31 3 2 Distribution of annual household income ................................ ................................ .......... 32 3 3 Frequency of having school meals ................................ ................................ ..................... 32 3 4 Preference towards foods in s chool m eals ................................ ................................ ......... 33 3 5 Preference t owards f oods at h ome ................................ ................................ ..................... 33

PAGE 10

10 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science INFLUENCE OF NATIONA L SCHOOL LUNCH PROGRAM TO CHI L S FOOD PREFERENCE By Yuan Jiang December 2013 Chair: Lisa House Co chair: Zhifeng Gao Major: Food and Resource Economics The National School Lunch Program is the second largest government food assistance program in the United States. Over 80% of all primary and secondary schools participate in the NSLP. Given this influence, it is an interesting question to investigate the effects of school meals on children's food prefer ences. In addition, considering that children have an impact on family's purchasing decisions as they directly and indirectly influence their parent's decision making; a secondary goal of this study is to explore how children influence the family's grocery shopping. Data for this research was collected using a survey completed by school children and their parents A convenien ce sample of one hundred and sixty three matched pairs of 4th and 5th graders and their parents was collected. R esults showed that alt hough the school meal is consumed frequently the quality is not as rated as high as home foods Other responses indicate children play an important role in the family purchase decision, and resource housework, grade and their knowledge has a

PAGE 11

11 CHAPTER 1 INTRODUCTION Study B ackground Good nutrition during childhood plays a key role in ensuring adequate growth, preventing the long term ris k of obesity and other chronic diseases, and enhancing overall health and well being (USDA/HHS 2010). Since food habits are still developing during childhood, it is meaningful to help children adopt healthy eating habits in order to improve longer term hea lth outcomes. Although individual factors such as gender and age play an important role, and parental influence exists, there is an increasing understanding that children's eating habits are influenced by environmental factors, both in the home and at scho ol. As many children spend most of their time at school during the weekdays and on average obtain one third of their daily caloric intake from food consumed at school (Briefel et a l. 2009), schools are a natural place to provide nutrition education or imp lement policies to help children to improve healthy eating habits. Federal, state and local school nutrition programs can influence the foods provided to students in school. In this sense, the School Meal Program of U.S. Department of Agriculture (USDA), i ncluding both the National School Lunch Program (NSLP) and the School Breakfast Program (SBP) serve a very important role in developing children's food and eating habits. In 1946, t he National School Lunch Program (NSLP ) was established with the dual goal of reducing government commodity surpluses while providing nutritious meals to low income children (Ralston et al., 2008) to help reduce the malnutrition problem T he NSLP was designed to reduce malnutrition by p roviding meals whose nutritive contents met at least one third of the child's nutritional requirements for the day to help insure a healthy and balanced diet. At the same time, t hrough Federal, State and local support, the price of the meal was set at a low level

PAGE 12

12 Over time, the NSLP has become th e second largest government food assistance being ( USDA/HHS 2010 ) O ver 80% of all primary and secondary schools participate in the NSLP. T he current program provides free and reduced cost lunches for income eligible students, as well as subsidizing paid lunches for students whose families meet somewhat higher income eligibility requirements. Similar to NSLP, the SBP was designed to provide c hildren easy access to a healthy and well balanced breakfast on school days. Given the reach of these programs, it is an interesting question to investigate the effects of school meals on children's food preferences. Given the reach of these programs, mea suring the effects of school n the U.S. market has increased: over a twenty year period from the 1960s to the end of 1980s, spending of children as consumers in the U.S. market increased from $2b illion to $6b illio n and their influence extended to a staggering $132b illio n of household expenditure (McNeal, 1992). This trend continued with an increasingly prosperous teenage market in the 1980s despite declining of the population of teenagers by 15.5%. B y the end of the 1990s, it was estimated that children in the United States a ccounted f or $23b illio n in direct spending, and their indirect spending had increased to a further $188b illio n in family purchases (McNeal, 1999). Thus it is important to understand the power of e on family purchases. Purpose and A im of the S tudy The goal of this research is to explore whether school meals change children's food preference, and lead to choosing healthier options outside of the school envi ronment. In addition, considering that chi ldren have an impact on family's purchasing decisions as they directly and

PAGE 13

13 indirectly influence their parent's decision making, and that children's preference and shopping behaviors influence present and future consumption trends; a secondary goal of this study is to explore how children influence the family's grocery shopping, and whether or not participation in the NSLP influences parents food shopping behaviors.

PAGE 14

14 CHAPTER 2 LITERATURE REVIEW The purpose of this chapter is to review prior articles related to the topical area s of the towards family decision making. This chapter will be divided into three parts, starting with the s, followed by the previous studies on National School ve been investigate d frequently in Individual, Social and Economic Factors. Individual Factor s Individual factors refer to age gender, nutrition knowledge food taste. Aranceta et al (2003) analyze the food preference and eating habits of Spanish children and adolescents, and their relationship to sociodemographic factors, using a random sample of the Spanish population aged 2 24 years. Information was collected by trained dietit ians during personal interviews with each participant and their mother or other guardian responsible for feeding the children for those less than 8 years old. They assess food consumption by means of a 24 hour recall. Their results indicate that there is d ifference in preference for specific food items among different age groups. Nu, Macleod, Barthelemy (1996) conducted a research on French children and adolescents ages 10 20. They asked respondents to generate a maximum of ten liked and ten disliked food s and to recall at what age their food preferences had changed. The results show that most participants thought changes had occurred at the age of 10, and the ages of 12 and 15 seem also important as well. Thus it is concluded that there is a significant e ffect of age on the change

PAGE 15

15 On the whole sample, the age effect also exists: negative changes take place sooner (mean age = 10), whereas positive changes take place later (mean age = 11).Thus children are more likely to begin to dislike a food they liked previously before puberty ; and after puberty, they are more likely to begin to like a food they disliked before. However, as the data collection relied on memory interpretation of the age differences should be considered with caution. Skinne r, Carruth, a nd Bounds ( 2002 ) compared food preferences longitudinally to identify factors related to food preferences. They collected data by questionnaires completed by mothers for their children at 2 3 years of age, 4 years old and 8 yea rs old and also for mothers themselves, whose children are 2 3 years old, and 8 years old Children who were 8 years old were also asked to finish abbreviated food preference questionnaires. The result shows that few s food preferences during the 5 to 5. 7 year period although the number of foods disliked increase d The summary of these studies suggests that age Another individual factor that may affect food preferences is gender. Previous research concerning differences in food preference based on gender ha s not shown consistent findings. Lytle and Seifert (1999) used 24 behaviors, and compared them across grade s and other sociodemagraphic factors. Their findings suggest boys and girls had similar developmental trends in consumption and eating preferences. Aranceta et al (2003) found a similar result, with differences in food preferences in their sample of childr en and adolescents in Spain based on gender. However, in 2001, Wardle, Sanderson and Gilbson found that among 4 5 year old British children, girls preferred vegetables more than boys did. Additionally, Reynolds et al (1999) found that American boys are rep orted to consume

PAGE 16

16 less fruit and raw vegetables than girls. And Caine and Scheule (2008) had the similar result that there was a difference in food preferences between genders, and this difference varied among elementary school, middle school and high scho ol. Boys preferred the meat fish and poultry foods over girls, and girls preferred fruits and vegetables. These conflicting results suggest that Another individual factor that has been i is nutrition knowledge. Pirouznia (2001) argued that children were not mature enough to connect nutrition knowledge and food selection. However, Morgan and Warren (2009) found that school garden enhanced nu trition education did encourage more children to taste fruits and vegetables. selection. French et al (1998) accessed motivations for vending snack choices and found that ad olescents rated snack taste as the most important factor to consider, followed by hunger and price. And in the study constructed by Susan Barr (1994), it was concluded that taste enjoyment has a significant relationship with the calcium intake for adolesce nt children. Economic Factors food prices have become an important consideration in food choice. Drenowski and Darmen (2005) concluded that people usually make selectio n of foods, like energy dense grains, fats, and sweets, which are higher in sugar and fat, because such foods are usually among the lowest cost dietary options. Guil laume and Lapidus (1998) argued that lower educational status of parents has been associated with lower dietary quality, including higher fat and lower micronutrient intake in children. Employment also was correlated with diet quality. Maternal employment was found to

PAGE 17

17 be negatively associated with the frequency of family meals, which are, in turn, positively associated with diet quality. Wolfe and Campbell (1993) concluded that children with mothers employed outside had less diverse diets than those with m others at home. Social Factor s amily factors peer influence, school environment and market and media environment. Family factors, a ccording to the research of Nicklas et al (2001) inclu de food exposure and availability, parental modeling, parenting style, and food socialization practices. The more fruit and vegetables available at home, the larger consumption of fresh fruit and vegetables were reported. And the result is similar in the research constructed by Hearn et al (1998), which claimed that the availability and accessibility of fruits and vegetables enable consumption when controlling for other socioeconomic status and psychosocial characteristics. Young and Fors (2001) also foun preference. They found that the percentage of children who usually ate healthy foods decreased as the parental situation changed from a two parent, to a single parent, to other family members, to a f oster family. Parental modeling was also examined, and Fisher et al (2002) indicated in their research children, leading to higher micronutrient intakes and l attitudes and knowledge about nutrition have also been correlated with children's healthy eating ; Contento family purchas ed, and therefore the availability of healthy foods, as well as the size of portions of healthy foods served to the children

PAGE 18

18 Peers have an important and lasting influence on the food preferences of children. Hendy and Raudenbush (2000) investigated the ef peer modeling has been found to be the strongest predictor of younger children to try some new foods. French et a l (1998) examined 13 different motivations regarding snack selections among 419 adolescents, and they found that the influence of friends was significant. preference and f ood purchases. Marquis and Dagenais (2002) investigated the relationship between the frequency of consumption of specific foods and TV watching. Food advertising was found to significantly promote increased consumption of some specific food items, and this research further discussed that food or beverage commercial advertisements often contained misleading information, which might lead to confusion among children. food thy foods, while increasing the intake of total fat and sugar. It was also reported by Ishdorj, Crepinsek and Jensen (2012) that the School Lunch Program and School Breakfast Program lead to increased consumption of both fruit and vegetables in schools. Family decision making has an important role in consumption decision making in the market. Thus, there have been numerous research studies in the area of the process of family decision making. However, in the early years, most researchers tended to focus on the influence

PAGE 19

19 of husband and wife in the family decision making, and overlooked the role of children in the decision process. The first research trying to analyze the influences of children on family decis ion making more the mother is child centered, the less the mother would agree to buy the foods children liked, and instead they would to buy foods that were healthy for their children. S ince then, children have been researched and recognized as having an influential role on the process of family decision making. In the 1990s, research indicated the increasingly important role of children during family decision making. According to the report of McNeal (1992), children spent more than $132 billion on 62 product categories, and approximately 17% of family purchases were i nfluenced by the children in the United States. And in the late 1990s, McNeal (1998) indicated that the influence has increased to around $188 billion directly, and $300 billion indirectly. And the research by Laczniak and Palan (2004) on children aged fro m 3 11 years old and found that when shopping together with parents, children would asked for a purchase every two minutes on average. Thus, the power of children in the process of family purchase decision making has increasingly aroused the interests of researchers, marketers and manufacturers. And based on the economic factors

PAGE 20

20 Product Types One important factor is product type. According to the previous research, generally, it is indicated that children have more influence on the product for their own use than product for the whole family. (Foxman and Tansuhaj, 1988) M ore recently, Shoham and Dalaka (2005) claimed records/CDs. And as for the leisure time activity, children are also influential in decision making processes, becau se such activities are often child focused (Labrecque, Ricard, 1999). In contract, children had less influence on the products that was used by the entire family; arch indicated that children had little involvement in the purchase of cars, furniture, televisions and household goods which involve higher financial risk s (Be lch et al., 1985). Decision Stage Another significant factor influencing the power of children in the family consumption is decision stage. According to various research studies, including the study conducted by Beatty and Talpade, children have larger in fluence in the early stage of the decision making process (Beatty and Talpade, 1994). Yet, there is also existing contrary evidence, from the research conducted by Lee and Beatty in New Zealand (Lee and Beatty, 2002), which concluded that children were hig hly involved in both the earlier stages and final stages of the decision process. This may be caused by the increasing influence of children on the family purchase in recent years. Recently, more and more researchers like Foxman (1989a), Beatty and Talpade (1994),

PAGE 21

21 and the power of children in the family decision making process. From the resource theory, developed by Blood and Wolf i n 1960, which mainly explain the influence husband and wife in the family purchase, it is shown that in the decision making process, the family member who contributes more resources has larger influence on the decision making. And, grounded on this theory, influence in the process of family decision making. power in the family decision ma employment, grade and the perception of parental love and confidence as independent variables, and chose the product per sonal resources have a positive relationship to their influence on the family decision making process. Also among all the independent variables, grade had the largest influence towards the decision process. Secondly, considering the income contribution, M oschis and Mithchell (1986) indicated study using a wider range the aim of generating more robust results; their results suggested thatFinally, product knowledge power in h (1989) product knowledge has products used by children themselves. Furthermore, resear ch conducted by Beatyy and Talpade

PAGE 22

22 stage, initiating the idea to purchase the item. And for those products used by the whole family, children did not affect t he decision making process. Also, Thomson (2007), found that influence over the purchase decision. Family Characteristics In addition to the various aforementioned f actors, changes in family structure also can researchers have claimed that the increasing divorce rate has dramatically altered the traditional family structure, and tha t single parent families or blended families, may change the dynamic of single parent families were more involved in the family purchase decision than children from ordinary amilies. In the study conducted by Taylor et al.,(1985) found that children from single parent families tend to have greater influence on food purchase, and had more participation on the choice of foods. One reason explaining this may be that single parent may not have enough time to make food purchase decisions, thus children had to take more responsibility for the family decision making on food purchases. Secondly, in the previous research, many researchers claimed that children from higher income families would have more influence on the family purchase. Jenkins (1979) found that luence would be larger in family with higher socio economic status. It may be because higher income families may provide more opportunities to their children on the choice of products, and then be more o showed by the study of Breatty and Talpade

PAGE 23

23 (1994) that children in the dual income family would have greater influence on family purchase decisions. stics with are child centered would be less inclined to buy the cereal that children liked, because mothers who care more about their children would pay more attent ion to the nutritional content of foods, thus it may lead to fewer purchases of the foods that children liked. Another study conducted by Roberts et al (1981) found that the more conservative and traditional the mother, the less children influenced family purchase decisions.

PAGE 24

24 CHAPTER 3 DATA This chapter first introduces the process of gathering information about the influence of the National School Lunch Program on s well as the degree of s influence on family purchas e s Then, it will provide the descriptive analysis of the variables used in this study. Methodology A mail survey was delivered to parents of 4th and 5th graders in one county through the school system Because the research includes two aspects (the NSLP' s effects on children's food preference, and on parents' grocery shopping ) the survey was designed with two parts: one for parents to complete and the other for children. Children from 4th or 5th grade were chosen because they were more likely than younge r children to be actively engaged in family grocery shopping, they may be better able to express their opinions regarding foods, and they had a higher ability to complete the questionnaire. Elementary school children were selected as opposed to middle or h igh school students due to the rates of participation in school lunch, increased food options (non NSLP foods) at schools for older children, and because their preferences may not be as set as older children. Surveys were sent home with approximately 1,0 00 4th and 5th graders across 7 schools with varying levels of participation in the school lunch program. A pre paid envelope was included for participants to return the survey. In total, 1 72 respondents returned completed surveys, for a completion rate of 15%. However, 15 returned surveys had incomplete responses and were not included in this analysis The remaining 163 responses are both completed by parents in the first part, and children in the second part.

PAGE 25

25 There are 32 questions, which are divided int o two part s : the first part is developed for parents, and the second part is for children. I n each part, it can be further divided into four groups to investigate: 1) demographics; 2) knowledge and preferences towards school meals; 3) shopping habits; and s characteristics. Descriptive Statistic s The questionnaire was sent to seven elementary schools in Alachua County, Florida. The seven elementary schools can be divided into three groups: 1) high percentage of participants in the National Scho ol Lunch Program (NSLP); 2) medium percentage of participants in the NSLP; and 3) low percentage of participants in the NSLP. In total, 1,000 surveys were sent to the 4 th and 5 th graders in the seven elementary schools, and 172 were returned. Deleting 15 o f the returned surveys with incomplete responses, 163 completed surveys were used in this analysis. Demographics Demographic information was collected with the hope to determine if these factors could urchase. For parents, female s accounted for 90% of the respondents (Table 3 1). Gender of children was more evenly spread, with 59% girls, and 41% boys (Table 3 2). Most (93.7%) parents in the sample were more than thirty years old, with the most common ag e range of 36 40 years old (26.38%) (Figure 3 1). Given the 4 th and 5 th grade target, it is not surprising that 82.21% of children were ages 10 and 11. More than 70% of the children come from a dual parent family, while approximately 22.09% have a family w ith a single parent, either separated, divorced, widowed or never married (Table 3 3) From the education aspect, approximately 50% of the participants have a four year college degree or higher. Very few people (10%) had a high school diploma or lower (Tab le 3 4). Most of the participants had an annual household income in the range from $ 50,000 to $ 75,000 Participants with an annual household income less than $35,000 accounted for about 26 % of respondents and

PAGE 26

26 participants with a household income more than $149,000 made up about 6% of the total participants (Figure 3 2). Most (85%) of the participants had more than one child in their house, with 30% of families with three or more children. Only 15% of the participants had one child (Table 3 5). Preference a nd Knowledge of School Meals asked to indicate the frequency of their children having school lunch and home lunch; school breakfast, and home breakfast. In the sample approximately 36.4% of all the parents indicated their children had school lunch every day, while there was 24% that never had school lunch. Compared to school lunch, school breakfast is less popular. Only 12.3% of the responde nts indicated that their ch ild had s chool b reakfast every day, and 63.6% reported never having s chool b reakfast (Figure 3 3). Parents were asked to report if their child was eligible for the National School Lunch Program (they were reminded the prices of a full price, reduced price, and free lunch to help with the definitions) According to the respondents, 38% qualify for free or reduced fee lunches (Table 3 6). C hildren were asked to indicate how much they liked school lunch and home lunch on a 5 po int scale (indicated with smiley faces) Children indicated they preferred home lunch to school lunch, with a mean score of 1.44 for home lunch (1 being the best), and 2.97 for school lunch (3 being neutral). Only 9.82% of children indicated that they liked the s chool l unch very much an d another 13.5% report never having s chool l unch. Over half of the children ( 55.83% ) gave lunch brought from home highest score, and 19.02% never hav e l unch packed from home. As for s chool b reakfast, 10.56 % of children gave the s chool b reakfast the best r ating but 60.25% never hav e s chool b reakfast On the other hand, 70.37% like the home breakfast best, with only 3.09% never having home breakfast ( Table 3 7 ).

PAGE 27

27 Next, children were asked to identify how much they liked 12 different foods commonly included in school lunch and likely eaten at home. In all cases, on average, foods consumed at home were preferred to foods consumed at school, with the smallest difference in means of 0.24 for French Toast Sticks and the largest of 1.23 for steamed broccoli (Figur es 3 4 and 3 5 ). Family Characteristics Nowadays, characteristics of the family were examined, including family shopping habits, and the level of nutrition knowledge. To better understand family shopping habits, parents were asked to report how often their children came to the grocery store with them. Approximately 70% reported their child ren came grocery shopping more than frequently (Table 3 8). Nearly all (96%) of the parents thought that they would agree to buy an item that their children wanted during shopping more than occasionally, and 7% would agree to buy the foods all of the time (Table 3 9). When asking about nutrition, most of the parents (98%) would consider their knowledge of nutrition would be better than a letter grade of C (A is the best) (Table 3 10). Finally the children were asked about their s hopping habits, their contribution to housework, and their efforts related to studying, which may be important factors affecting the power of children in family shopping. Approximately 74% of children would ask their parents to buy something they like dur ing shopping. thought that they would agree to buy the foods that their children wanted during grocery shopping more than occasionally. It demonstrates that children are playing a v ery important role in family grocery shopping.

PAGE 28

28 The frequency of doing housework (including house cleaning, making bed, cleaning More than half (64%) of childre n though that they did house cleaning more than occasionally, and 22.07% of children indicated they made their bed every day. As for helping with the dishes, 43% of children reported that they helped at least 1 2 times each week. Nearly half (42%) indicate d they did laundry more than 1 2 times a week. Almost all (85.89%) indicated that they did homework every day, and 11.04% thought they did homework about 4 5 times a week (Table 3 11). Table 3 1 Gender of parents Gender Frequency Percent Female 147 90. 18 Male 16 9.82 Table 3 2 Gender of children Gender Frequency Percent Girl 91 55.83 Boy 72 44.17 Table 3 3 Marital status of parents Marital Frequency Percent Never married 11 6.75 Married 116 71.17 Living with partner 7 4.29 Separated 3 1.84 Widowed 3 1.84 Divorced 19 11.66 Other 2 1.23 Refused to answer 2 1.23

PAGE 29

29 Ta ble 3 4 Education level of parents Education Frequency Percent Less than high school diploma 6 3.68 High school diploma or equivalent (G.E.D.) 10 6.13 Some college (AA degree or courses without a final degree) 67 41.10 Bachelor's Degree 31 19.02 Post graduate or Professional Degree 49 30.06 Table 3 5 Number of children in the family Number of children in family Frequency Percent 1 25 15.34 2 87 53.37 3 28 17 .18 4 16 9.82 5 4 2.45 6 3 1.84 Table 3 6 Eligibility for school meals Meals Frequency Percent Free lunch 49 30.06 Reduced price meals 11 6.75 Regular school lunch ($2.15) 98 60.12 Do not know 5 3.07

PAGE 30

30 Table 3 7 Preference towards school meals and home meals I never have this School Lunch 16 (9.82%) 40 (24.54%) 44 (26.99%) 17 (10.43%) 24 (14.72%) 22 (13.50%) Lunch you bring from home 91 (55.83%) 31 (19.02%) 8 (4.91%) 1 (0.06%) 1 (0.06%) 31 (19.02%) School breakfast 17 (10.56%) 17 (10.56%) 8 (4.91%) 9 (5.59%) 13 (8.07%) 97 (60.25%) Breakfast I have at home 120 (70.37%) 33 (20.37%) 8 (4.91%) 1 (0.06%) 1 (0.06%) 5 (3.09%) Table 3 8 Frequency of children going grocery shopping with parents Frequency of children going grocery shopping with parents Frequency Percent Never 0 0 Rarely 7 4 Occasionally 42 26 Frequently 73 45 All of the time 41 25 Table 3 9 Frequency of parents agreeing to buy items at the grocery store if child asks for something Frequency of parents agreeing to buy Frequency Percent Never 0 0 Rarely 5 3 Occasionally 71 44 Frequently 76 47 All of the time 10 6 Table 3 10 Self reported g rade of trition Grade of knowledge of nutrition Frequency Percent A 57 35 B 75 46 C 25 17 D 3 2 E 0 0 F 0 0

PAGE 31

31 Table 3 11 Frequency of doing housework as reported by the children Type of work Never Not much About 1 2 times a week About 4 5 times a week Everyd ay Clean the house 10 (6.13%) 48 (29.45%) 69 (42.33%) 17 (10.43%) 19 (11.66%) Make the bed 20 (12.27%) 44 (26.99%) 35 (21.47%) 27 (16.56%) 37 (22.70%) Wash dishes 37 (22.84%) 55 (33.95%) 35 (21.60%) 20 (12.35%) 15 (9.26%) Do laundry 44 (26.99%) 49 (30. 06%) 53 (32.2%) 8 (4.91%) 9 (5.52%) Do homework 0 (0) 2 (1.23%) 3 (1.84%) 18 (11.04%) 140 (85.89%) Fi gure 3 1 Age distribution of parents

PAGE 32

32 Figure 3 2 D istribution of a nnual h ousehold i ncome Figure 3 3 Frequency of having school meals

PAGE 33

33 Figure 3 4 Preference towards foods in s chool m eals Figure 3 5 Preference towards f oods at h ome

PAGE 34

34 CHAPTER 4 IN F LU ENCE O F N AT I O NAL SCH OO L PROGRAM TO C HIL DR EN' S F O O D PRE F ERENCE The aim of this chapter is to analyze data from the survey res ponses The initial portion will provide an overview of the data using frequency analysis The next portion will show the analy sis of the influence of the National School Lunch Program by testing three hypothes e s about the NSLP. General Data Overview In total, 170 responses were collected ; h owever 15 were rejected due to incomplete responses. The valid 163 responses are both completed by parents in the first part, and children in the second part. As expected, even though only 38% of respondents qualify f or free or reduced fee lunches more children ate school lunch (defined as the School Lunch enceforth, home lunch will refer to lunches consumed at school but packed at home). In total, 36.2% of the respondents had School Lunch every day, 17.8% of ate Scho ol Lunch more than half of the time and 22.3% ate School Lunch some of the time. Only 23.7% of respondent reported never eating School Lunch. School Breakfast is less popular, with only 12.3% of responde nt s reporting eating it every day, and 63.8% indicat ing that they never have School Breakfast. Children were asked to indicate how much they liked school lunch and home lunch on a 5 point scale. Children indicated they preferred home lunch to school lunch, with a mean score of 1.43 for home lunch (1 being the best) and 3.12 for school lunch (3 being neutral). Children were also asked to identify how much they liked 12 different foods commonly included in school lunch and likely eaten at home. In all cases, on average, foods consumed at home were preferred t o foods consumed at school, with the smallest difference in means of 0.24 for French Toast Sticks and the largest of 1.23 for steamed broccoli. This indicates that although there are many

PAGE 35

35 students participating in the NSLP, most children are not as satisfi ed with the quality of school meals compared to home. When children were asked if they ever talked about the foods from school lunch at home, answers ranged from 76.3% for pizza to 24.3% for pineapple tidbits. Parents were also asked if their children ever talked about school lunch at home (in general, not for specific foods like the children) and 92.4% indicated their children talked about school foods at least sometimes, though 42.1% indicated it was rare. What is more, when considering the reason why the children hope to buy certain foods, the most important reason children indicate is because they tried it with a friend (51% ), saw a commercial (46%), or because the packaging attracted them (42%). Though lower in priority, 32% indicated they asked for so mething because they tried it at school. In addition to asking children why they want to purchase certain foods, p arents were asked if they knew why their children asked for foods at the store Parents most often attribute the requests to packaging (54%) a nd commercials (54%), and only mention school lunch 17% of the time (though 13% of parents indicating they do not know the reason the child asks). Though not a major reason, it does appear and family g rocery consumption. Hypothesis Testing In order to investigate the National School Lunch Program, perceived quality differences between School Meals and home meals is tested. To achieve this goal, matched pairs T test will be used. H1: Children Prefer Home Meals More than School Meals. To determine if children prefer home meals to school meals, mean ratings for both types of lunch were compared for respondents who reported having both school and home lunches ( 53 responde nts who never tried School Lunch or l unch brought from home were removed from this test).

PAGE 36

36 Among the participating children, 24.54% never had school lunch, and 20.37% of them never had lunch packed from home. Among those who had both, 6.36% prefer school lunch more than home lunch, 8.18% ran k the lunches same and 85.46% thought home lunch was better than school lunch (Table 4 2). For breakfast, 60.25% of responding students never had school breakfast, while 3.09% report never having breakfast at home. Among those students who both had school and home breakfast, 8.06% would prefer school breakfast over home breakfast, 67.74% had ranked the home breakfast higher, and 24.19% thought the two were the same quality. The paired t test indicates the mean values ( 3.12 for school lunch and 1.43 for hom e lunch) is different at the 95% confidence level (Table 4 4), with students preferring lunch from home to the School Lunch Program food Similarly, deleting 101 responders who report never having School Breakfast or home breakfast, the mean score of Schoo l Breakfast is 2.84, while the score for home breakfast is 1.30 (Table 4 5) This difference is also significant at the 95% confidence level, with children prefer ring home breakfast to school breakfast. Results from these two tests indicate the quality of school meals may be an issue as students significantly prefer home meals to school meals. H2: For each category of foods, children prefer home foods to school foods. Children were also asked to identify how much they liked twelve different foods commonly included in school lunch and likely eaten at home. P aired t test s were used to test each food Output from SAS is shown in table 4 5 In all twelve cases, children reported significantly prefer ring the foods eaten at home. The largest difference was found for pizza followed by h amburgers and the smallest difference is f rench t oast s ticks.

PAGE 37

37 Meals and Home Meals In addition to investigating whether or not children prefer home foods to school lunch foods, it is of interest if the ratings are related. Given the small sample size, a chi square test is not appropriate, so was used for the test Results are shown in Tables 4 6. Results indicate there is signific ant correlation ten of the twelve foods provided at school and eaten at home, indicating there may exist significant relationships between the preferences of school and home foods The two foods that did not show corr elation were pizza and hamburgers, which were also the two foods with the largest difference in means. This indicates the difference in quality for these foods may be very significant. Table 4 1 school lunch. Frequency Percent Cumulative Frequency Cumulative Percent Rank home lunch lower than school lunch 7 6.36 7 6.36 Rank home lunch higher and school lunch the same 9 8.18 16 14.55 Rank home lunch higher th an school lunch 94 85.45 110 100.00

PAGE 38

38 Table 4 2 breakfast and school breakfast Frequenc y Percent Cumulative Frequency Cumulative Percent Rank home breakfast lower t han school breakfast 5 8.06 5 8.06 Rank home breakfast lower and school breakfast the same 15 24.19 20 32.26 Rank home breakfast higher than school breakfast 42 67.74 62 100.00 Table 4 3 preference towards s chool m eals and h ome m eals: o ne s ample s tatistics Meals N Means Std Dev Std Err School Lunch 110 3.1182 1.1710 0.1116 Lunch Packed from Home 110 1.4364 0.7361 0.0702 School Breakfast 62 2.8387 1.5278 0.1940 Home Breakfast 62 1.3065 0.6675 0.0848 Table 4 4 p reference towards school meals and home meals: Matched Paired T test t df Sig. Mean Difference 95% CL Means lower upper Difference between school lunch than home lunch 12.65 109 <0.001 1.6818 1.418 4 1.9453 Difference between school breakfast than home breakfast 6.96 61 <0.001 1.5323 1.0919 1.7341

PAGE 39

39 Table 4 5 Child ood: m atched p aired T test Difference between home and school: t df Sig. Mean Difference Orange juice 2.55 83 0.0128 0.1786 Greens 5.73 72 <.0001 0.5479 Fruit cup 4.90 93 <.0001 0.3838 Fruit juice slushy 3.73 67 0.0004 0.2836 Fresh orange wedges 4.39 107 <.0001 0.2710 French toast sticks 1.47 69 0.1456 0.1000 Hamburger 6 .46 124 <.0001 0.5120 Pizza 8.22 148 <.0001 0.5503 Pineapple tidbits 2.41 74 0.0185 0.1733 Steamed broccoli 7.07 103 <.0001 0.6538 Steamed carrots 6.08 90 <.0001 0.5714 Sweet potato wedges 2.95 62 0.0045 0.3333 Table 4 6 Fishers test fo r foods eaten at home and at school Food O range juice 0.0117 G reens 5.758E 08 F rench cup 0.0072 J uice slushy 0.0231 O range wedges 6.786E 09 F rench toast sticks 2.140E 04 H amburgers 0.2487 P izza 0.1477 P ineapple tidbits 6.1 71E 06 S teamed broccoli 1.079E 04 S teamed carrots 2.453E 07 S weet potato wedges 4.261E 07

PAGE 40

40 CHAPTER 5 products, and their contribution to the housework are important factors that determine the Model Data were col lected by mailed questionnaires, as described in the previous chapter. Ordered Probit Model An o r dered probit model is used to estimate models whose dependent variables are ordinal, but not continuous in the sense that the metric used to code the variables is substantively meaningfu l (Simon Jackman, 2000). The ordered probit model is based on the central idea that there is a latent underlyingindex that is not observable by the analyst and is a continuous descriptor of the real responses. Thresholds partition the real line into a ser ies of regions corresponding to various ordinal categories. In addition the random error associated with this continuous descriptor is assumed to be normal distributed. The basic probit model is shown in Equation 5 1 = ~ N (0, 1) (5 1) where is a continuous variable which is a linear function of a set of dependent variables is the disturbance term that has a normal distribution. is the vector of the regression coefficient to be estimated.

PAGE 41

41 The relationship between and (for example following equations (Equation 5 2, 5 3, and 5 4): = 0, if 0 (5 2) = 1, if 0 < < (5 3) = m, if < < (5 4) Where is the observed ordinal variable taking on values 0 through m, and s are unknown threshold parameters which are estimated with To analyze decision, such as = 0 if parents totally disagree with t he statement that child ren have influence on their purchase decision. =1 if parents dis agree with statement that children have influence on their purchase decision. =2 if parents neither disagree nor agree with the statement that children have influence on their purchase decision. =3 parents agree with the statement that children have influence on their purchase decision. = 4 parents totally agree with the statement that children have influence on their p urchase decision. Consider the probabilities of each ordinal outcome, the following equations can be written: P [ = 0 ] = P [ < < ] = ( ) (5 5) P [ = 1 ] = P [ < < ] = ( ) ( ) (5 6)

PAGE 42

42 P [ = 2 ] = P [ < < ] = ( ) ( ) (5 7) P [ = 3 ] = P [ < < ] = ( ) ( ) (5 8) P [ = 4 ] = P [ < < ] = ( ) ( ) = 1 ( ) (5 9) The marginal effects of the independent variables on the probabilities can be derived and t hey vary by individuals. The marginal effects depends on the values of all ind ependent variable s and are calculated as the follows (Equation 5 10): = [ ( ) ( )] (5 10) Typical marginal effects are calculated at the mean of the variable. However, in the case of dummy variables the mean is not relevant, and the difference of two resulting probabilities when the dummy variable equals 1 and 0 is use d. Model Specification agreement towa 1). Independent characteristics (Table 5 2). SAS was used to estimate the model. The specification of the Ordered Probit Model is: Where Pa_study is statement listed in Table 5 1, which examine s whether or not parents would agree to buy a learning tool, like an electronic dev ice for reading or listening to books to help children with school The independent variables in the model can be divided into several groups. The first gender ( pargender ),

PAGE 43

43 ag e (parage), education level (paredu), marital status (marstatus), number of children in the family (nochild), income, and self rated nutrition knowledge (nutrigrade). The second group age (chiage), frequency of doing homework (fqhome), and frequency of doing housework (fqhouse). The frequency of doing housework is the sum of the reported frequency of doing laundry, washing dishes, making their bed, and helping to clean the house. The third group consists of three variables focusing on the interaction between children and parents regarding shopping behavior such as frequency of shopping with children (Freqshopchi), frequency of children asking for a certain item (Freqask), and frequency of parents agreeing to buy in general (Freqbuy). The Ordered Probit Model is repeated for each of the following dependent variables: agr eement/ disagreement with statements two to five, respectively, listed in Tabl e 5 1 The six statements listed in Table 5 1 can be considered as measurement of the relationship between These statements measure agree to buy an electronic device for re ading or listening o books to help children with school ; their progress in grade at school (grade), measured by if parents would be more likely to agree to buy a certain item when children get better grade in school; the information they provided during s hopping (information), whether parents would buy a certain brand when child ren told parents that brand is preferred because it works better for school or they know how to use it from school ; their preference towards a certain item (food), measured by whet her parents would be more

PAGE 44

44 likely to buy a certain food that children liked or asked for at grocery store; the nutrition knowledge of a certain item that children provided (nutrition), measured by whether parents would be more likely to buy a certain food w hen children have learned about the nutrition towards housework (housework), measured by whether parents would be more likely to purchase things for children wh en they helped a lot at home. Results Data from 163 surveys was used to estimate the ordered probit models. S ummary statistics for the dependent variables are shown in Table s 5 3, 5 4 & 5 5 Since the six statements making the dependent variables were answ ered by the same person, and their answers will inevitably affected by their personality the answers would likely be correlated. T hus the qualitative and l imited d ependent m odel (QLIM) procedure in SAS, which has the ability to analyze the models involvin g simultaneous association was used Correlation tests were run, and based on the results ((Rho < 0.0011 in all cases), the variables are correlated, supporting use of the QLIM procedure. Results of the ordered probit models are shown in Tables 5 6. Fami ly C haracteristics and P D emographics The coefficient for gender is significant in five of the six cases, with the exception being mothers (female) ar e more likely to be influenced by children during the shopping progress than fathers (male). In the case of whether or not a parent will purchase a food from the grocery store because the child likes it, gender was not significant. In fact, in this equati on, only two factors is positively related to likelihood to purchase a food because the child likes it, and their level of

PAGE 45

45 nutrition knowledge, which is nega tively related. This result is consistent with previous research constructed by Berry and Pollay (1968) that the more mothers cared about nutrition, the less influence the children has over food purchases ucation) is significant in two cases: when (pa_nutri). In both of the two s tatements, the sign of the coefficients is negative, indicating that ildren reasonable. It may be because that parents with higher education would gain more knowledge, thus they will be more stick to their own opinion, and more diffi cult to be affected by their children. However, in the other 4 cases, education is not an influential factor, making no difference on the power of children in family purchase. rated grade of nutrition knowledge (nutrigrade) is only significa nt in one preference towards certain food (pa_food). In this case, the grade of nutrition knowledge is y to explain the reverse relationship preference towards a certain food. With higher nutritious knowledge, when deciding whether or not to buy a certain food that children like, parents would consider more about its nutritious

PAGE 46

46 The coefficient of fami ly income is only significant in one case, which is examining the they provided during shopping (pa_info). The positive coefficient sign in this case shows that because that when deciding a certain brand, with high income, parents would pay more attention towards the functions and its help for children thus would care less a bout other factors for example the price. In this way, children will play a more important role in the process of purchase. This result is consistent with numerous of researches, like the one conducted by Jenkins (1979). Income is not statistically signifi cant in other cases. housework (pa_housework), the sign of the coeff older of the parents, the more difficult for them being influenced by children. This may be because that older parents would be more traditional and conservative, thus they would be less influenced by their ch ildren, which is consistent with previous research constructed by Roberts et al (1981) demographics, there do still exist two more variables: marital status and number of chi ldren in the family, which are statistically insignificant in all the six cases. It was expected that children in single family may have larger influence in family shopping according to the previous researches constructed by Taylor et. al,(1985) However, it may be because that we only checked

PAGE 47

47 influence in family purchase. C har acteristics (chiage), frequency of doing homework (fqhome), and frequency of doing housework (fqhouse) are also related variables which would impact the amount of c shopping. significant in three out the six cases, concerned respectively towards necessity of certain item in progress in grade (pa_grade), and contribution towards housework doing housework is positive. It is easy to conclude that in these three case, children do more h ousework at home would have larger influence towards family purchase. This result is easy to understand. Generally, children who did more housework at home would have more influence in the family because they made more contributions to the family, accordin g to the resource theory developed by Blood and Wolf (1960) Thus, when deciding whether or not to buy a certain item are necessary in study, children who made more contribution would be more influential towards their parents. with ts would be more

PAGE 48

48 agreed to buy certain item that children asked either as a helpful leaning tool, or as a prize for their progress in school or contributions towards housework when children do more homework at home. and gender are only statistically significant in the case concerned about the nutritious information provided by children in the process of family children, the more influential of them towards parents when buying a certain food. And the positive sign of gender shows that boys are more influential than girls in this case. Family S hopping H abits The group of f amily shopping habits is consisted of three variables: frequency of shopping with children (freqshopchi), frequency of children asking for a certain item during shopping (freqask), and frequency of parents agreeing to buy in general (freqbuy). The coefficient of frequency of children asking for a certain ite m (freqask) is statistically significant in three cases, correspondingly regarding to the necessity of certain items in study, three cases are positive, which means that the more children bargaining for a certain item, the larger possibility that parents would agree to buy a certain item which children asked as a prize for their progress in study or a learning tool in their schoolwork. This result is consistent with the Frequency of parents agreeing to buy items children asked in general (freqbuy) is only

PAGE 49

49 children asked in general, the la making. The frequency of shopping with children is insignificant in all the six cases, which indicate s that shopping with parents does not improve the influence of children in family shopping Summary By applying the O rdered P robit M odel in this chapter, we found the factors that affecting variables, and 20 independent variables In the chapter of conclusion, the relationship between

PAGE 50

50 Tabl e 5 1. Statement t esting p a ttitudes towards c Label Statement 1 If my child told me that he/she needs an electron ic device for reading or listening to books to help his/her with school, I would think about buying one. 2 The more my child studies and the better he/she does in school, the more likely I am to think about buying a new electronic device (for reading/list ening to books) for him/her. 3 When deciding what electronic device (for reading/listening to books) to buy, if my child told me that there is a certain brand they need because it works better for school or they know how to use it because they learnt it a t school, I would buy that brand. 4 If my child told me that she/he wants to get a certain food from the grocery store, because she/he likes it, I will be more likely to buy it for them. 5 If my child told me that she/he wants to get a certain food from the grocery store, because she/he leant how healthy it is at school, I would be more likely to buy it for them. 6 The more my child helps around our home, the more likely I am to purchase things for him/her as a reward

PAGE 51

51 Table 5 2 Variables used in the ordered probit model Variable Definition of Variable Coding Pa_study statement concerning study = 0 if parents totally disagree = 1 if parents disagree = 2 if parents neither agree or disagree = 3 if parents agree = 4 if parents totally agree Pa_grade statement concerning grade = 0 if parents totally disagree = 1 if parents disagree = 2 if parents neither agree or disagree = 3 if parents agree = 4 if parents totally agree Pa_info statement concerning knowledge of certain item/brand = 0 if parents totally disagree = 1 if parents disagree = 2 if parents neither agree or disagree = 3 if parents agree = 4 if parents totally agree Pa_food greement towards fourth preference = 0 if parents totally = 1 if parents disagree = 2 if parents neither agree or disagree = 3 if parents agree = 4 if parents totally agree Pa_Nutri stat ement concerning nutritious information = 0 if parents totally disagree = 1 if parents disagree = 2 if parents neither agree or disagree = 3 if parents agree = 4 if parents totally agree Pa_housework statement concerni contribution to housework = 0 if parents totally disagree = 1 if parents disagree = 2 if parents neither agree or disagree = 3 if parents agree = 4 if parents totally agree Pargender Female=1; Male=0 Parage Continuous Pareduc Continuous Marstatus Marital status single family=1; ordinary family=0

PAGE 52

52 Table 5 2. Continued Variable Definition of Variable Coding Nochild Number of children in family Continuous Income Household income Cont inuous Nutrigrade Grade of nutrition knowledge Continuous Chigender Female=1; Male=0 Chiage Continuous fqhome More than frequently=1; less than frequently=0 fqhouse of doing housework Continuous Freqshopchi Frequency of shopping with children Continuous Freqask Frequency of children asking for a certain item Continuous Freqbuy Frequency of parents agreeing to buy Continuous

PAGE 53

53 Table 5 3 Sample descriptive statisti cs Variable Variable Description Sample%(N=163) Gender Female 9 0.00 Male 10 .00 Age <=30 6.75 31 45 70.55 >45 22.7 0 Education Less than Some College 9.81 Some College 41.10 Bachelor or More 49.08 Marital Status Ordin ary Family 22.09 Single Family 75.46 Number of Kids 1 15.34 2 53.37 31.29 Annual Household Income $14999 or less 7.41 $15,000 $24,999 11.73 $25,000 $34,999 6.79 $35,000 $49,999 14.2 0 $50,000 $74,999 19.14 $75,000 $99,999 16.05 $100,0 00 $149,999 19.14 $150,000 $199,999 4.32 $200,000 or higher 1.23 Grade of Nutrition Knowledge A 34.97 B 46.01 19.02

PAGE 54

54 Table 5 4 Sample descriptive statistics s characteristics Variable Variable Description Sample%(N=163) Gender Gir l 55.83 Boy 44.17 Age 9 and under 15.95 10 47.85 11 and older 36.2 0 Frequency of doing homework 1 or 2 times each week or less 3.37 4 5 times each week 11.04 Everyday 85.89 Frequency of doing housework Rarely or less 40.26 Sometimes 45.28 Frequently of more 14.46 Table 5 5 Sample d escriptive s tatistics s hopping c haracteristics Variable Variable Description Sample%(N=163) Frequency of Shopping with Children About Half of times of Less 22.29 Usually 37.97 Always 29.75 Frequency of Children Asking for Things during Shopping About Half of times of Less 25.16 Usually 44.79 Always 30.06 Frequency of Parents Purchasing Things Children Asked Rarely or Never 3.09 Occasionally 43.83 Frequently or More 53.08

PAGE 55

55 Table 5 6 Orde red probit model result Pa_study Pa_grade Pa_info Pa_food Pa_nutri Pa_work Paage 0.0531 (0.070) 0.08 (0.070) 0.069 (0.068) 0.072 (0.074) 0.008 (0.075) 0.142** (0.070) Pagender 0.531* (0.312) 0.623** (0.305) 0.387* (0.305) 0.230 (0.317) 0.688** (0.34 7) 0.736** (0.310) Paedu 0.057 (0.106) 0.076 (0.105) 0.159* (0.104) 0.029 (0.110) 0.105* (0.113) 0.114 (0.105) Nochild 0.004 (0.095) 0.080 (0.095) 0.059 (0.094) 0.089 (0.100) 0.014 (0.100) 0.110 (0.097) Marstatus 0.096 (0.244) 0.010 (0.242) 0.27 9 (0.240) 0.168 (0.252) 0.188 (0.266) 0.245 (0.243) Income 0.067 (0.059) 0.018 (0.058) 0.194*** (0.059) 0.046 (0.061) 0.013 (0.062) 0.052 (0.059) Nutrigrade 0.044 (0.125) 0.212 (0.124) 0.007 (0.123) 0.194* (0.130) 0.051 (0.134) 0.190 (0.124) Chi gender 0.128 (0.183) 0.065 (0.181) 0.095 (0.179) 0.133 (0.190) 0.438** (0.196) 0.104 (0.180) Chiage 0.125 (0.123) 0.160 (0.122) 0.083 (0.120) 0.048 (0.125) 0.314** (0.129) 0.132 (0.122) fqhome 0.164 (0.520) 1.153** (0.554) 0.438 (0.508) 0.288 ( 0.551) 0.219 (0.549) 1.469** (0.584) fqhouse 0.055* (0.030) 0.064** (0.029) 0.063*** (0.029) 0.017 (0.030) 0.017 (0.031) 0.070** (0.029) fqshopchi 0.052 (0.123) 0.011 (0.123) 0.060 (0.120) 0.013 (0.127) 0.132 (0.132) 0.111 (0.122) Freqask 0.250*** (0.097) 0.320*** (0.097) 0.010 (0.095) 0.115 (0.100) 0.014 (0.102) 0.161* (0.096) Freqbuy 0.073 (0.145) 0.008 (0.144) 0.165 (0.143) 0.969*** (0.166) 0.074 (0.159) 0.223 (0.146) %

PAGE 56

56 CH A PTER 6 CO NCL U SIO N This research consists two parts: the influence of National School Program on purchases. I n the fi r st pa r t of the stu d y results show that app roximately half of st u d e nts report eating school lunch regularly; how e v e r, the qu a l i t y o f cu r r e nt school foods is not rated as high as home lunch Wh e n c h e c ki n g the c o r r e l a t i on b e tw ee n c hi l d re p r e fer e n c es for tw e lve foods w h ich a re c om m on l y found in S c hool L un c h a nd likely ea ten a t ho m e, ratings for both locations a r e s i g nifi ca nt l y c o r re l a t e d. However, children consistently prefer home food more than school meals. In two cases, pi zz a a nd h a mburg e r, there was no correlation between the ratings of home an d school lunch, indicating the quality of these two foods is significantly differen t. I n the s ec ond p a rt, c hi l d re n s infl u e n c e on f a m i l y pu r c h a s es is examined F rom p re vious stud i e s, it has been found that c hi l d re n s r e sou r ce s f a m i ly c h a ra c te r is t ics, c h i l d r e n s c h a ra c te r is t ics, a nd ot h e r soci o ec onom i c f ac tors pl a y a si g nifi ca nt ro l e in the amount o f c hi l d re n s influen c e in t he pr o ce ss of p a r e l grades, contribution towards housework, knowledge of certain brands, and nutrition information are important factors Results indicate that mo t h e rs are more likely to buy what children liked if the children really need cer tain learning tools for school work. The mo r e c hi l d r e n a sked or bargained, the higher possibility that parents would buy the items children needed. Other variables positively related to the likelihood to purchase what the child wants is a t ho m e study

PAGE 57

57 effo rt on (frequency of doing homework), and contribution to housework (frequency of doing housework). The more children do housework/homework, the larger the possibility that parents would agree to buy items that are helpful for their school work. However, in this case, other f a c tors, l ike p a r e nts incom e and e du c a t i on lev e l, childr e a ge a nd g e nd e r and family shopping habits do not si g nifi ca nt l y a f f ec t likelihood for parents to purchase the items. A second model examined whether or not parents were more likel y to purchase items if the child is doing well on their grades at school. Again, mothers are more easily influenced than fathers. Also similar to the previous case, the more children do housework/homework, the more children asking for a certain item, the h igher possibility that parents would agree to buy items for the children. for specific items based on the information their child has, the significant factors change, significant, with the hi g h e r level of e d u ca t io n of the p a r e nts, the less influ e n c e the c hi l d re n in f o r mation has on t h e ir d e c is i on. Another significant fac hi g h e r levels of household income, are more likely to be influ e n c ed b y c hi l d r e n to de c ide whi c h fo o d/br a nd to b u y contribution towards housework still weighs significantly in this case. The more children do the housework, the more important they weigh in the family, thus the larger influence of them habits are not s ignificantly related to influence. The fourth model examines whether or not parents would be more likely to buy a specific food when children have learned about the nutrition information of that food at

PAGE 58

58 school (and it is considered healthy). The significa nt factors in this model are different from ence. The younger the child, the larger possibility that gender, with boys being more influential than girls. Common with the other models, the frequency of child ren doing housework is still a positive factor affecting the amount of The next model examined whether or not parents were more likely to agree to buy a are significant: the likelihood to purchase the foods that the chil d liked. The frequency of agreeing with children in general is a positive factor, indicating that the more parents would listen to children liked. Finally, the last model examines whether or not parents would be more likely to purchase items if children made contributions towards housework. Mothers are more easily the parents, the less possibility that parents would decide to purchase items children want. In shopping and the frequency of doing housework are both significant fac tors, working

PAGE 59

59 the child does participate more frequently in helping with housework, the parents are more likely to purchase items for the children based on this f actor. different situations, the affecting factors will be different.

PAGE 60

60 CH A PTER 7 IMPLICATION Developin value to the schools, as well as the administrators of the Na t i on a l S c hool L un c h P r o g ra m. If a goal of the NSLP is to teach children healthy eating habits, it will be important to provide foods they like. Our research shows that current food served in the NSLP does not meet the preferences of students as well as food from home. Although the school lunch foods were not rated as highly as at home foods, 1/3 of students still iden tified the school lunch program as a source of information on new foods they ask for at the grocery store. Therefore, it does seem like there is potential to teach children about foods at school. To further understand this process, it is important to also study what makes children more influential in the family decision making process. Parents tend to be more likely child is doing well in school (grades). Results requests if the ir children have information about the nutritional value of the food. This implies that the NSLP could pair nutrition education with the school lunch to try to influence at home behavior. This c ould also potentially be carried over to food manufacturers and marketers if they provide positive nutrition information to children, it may influence the parents purchasing habits.

PAGE 61

61 LIST OF REFERENCES Aranceta J., C. Perez Rodrigo, L. Ribas & L Serra Majem (2003). Sociodemographic and lifestyle determinants of food patterns in Spanish children and adolescents: the enKid study European Journal of Clinical Nutrition 57(1) S40 S44 Baranowski, T., Domel, S., Gould, R., Baranowski, J., Leonard, S., Trei ber, F., & Mullis, R. (1993). Increasing fruit and vegetable consumption among 4th and 5th grade students: results from focus groups using reciprocal determinism. Journal of Nutrition Education 25(3), 114 120. Barr, S. I. (1994). Associations of social a nd demographic variables with calcium intakes of high school students Journal of the American Dietetic Association 94(3), 260 269 Beatty, S. E., & Talpade, S. (1994). Adolescent influence in family decision making: a replication with extension. Journal of Consumer Research 332 341. Belch, M. A., Krentler, K. A., & Willis Flurry, L. A. (2005). Teen internet mavens: influence in family decision making. Journal of Business Research, 58(5), 569 575. Briefel, R. R., Crepinsek, M. K., Cabili, C., Wilson, A. & Gleason, P. M. (2009). School food environments and practices affect dietary behaviors of US public school children. Journal of the American Dietetic Association 109(2), S91 S107. Blood, R. O., & Wolfe, D. M. (1960). Husbands & wives. Free Press of G lencoe Caine Bish, N. L., & Scheule, B. (2009). Gender Differences in Food Preferences of School Aged Children and Adolescents. Journal of School Health 79(11), 532 540. Contento, I. R., Basch, C., Shea, S., Gutin, B., Zybert, P., Michela, J. L., & Rip s, J. (1993). Relationship of mothers' food choice criteria to food intake of preschool children: identification of family subgroups. Health Education & Behavior 20(2), 243 259 Drewnowski, A. (2003). Fat and sugar: an economic analysis. The Journal of Nu trition 133(3), 838S 840S. Fisher, J.O., D.C. Mitchell, H. S. Wright & L. L. Birch (2002). Parental influences on young Journal of the American Dietetic Association 102(1), 58 64 Foxman, E. influence in family purchase decisions: patterns of agreement and disagreement. Advances in consumer research 15(1), 449 453.

PAGE 62

62 French, S. A., M. Story, P. Hannan, K. Breitlow, R. W. Jeffery, J. S. Baxter & M. P. Snyder (1999). Cognitive and demographic correlates of low fat vending snack choices among adolescents and adults Journal of the American Dietetic Association 99(4), 471 474 Guillaume, M & L Lapidus (1998). Obesity a nd nutrition in children. The Belgian Luxemebourg Child Study IV. European Journal of Clinical Nutrition 52, 323 328 Hanks, A., Just, D., & Wansink, B. (2012). Trigger Foods: The Influence of'Irrelevant'Alternatives in School Lunchrooms. Agricultural and Resource Economics Review Hearn, M. D., T. Baranowski, J. Baranowski, C.Doyle, M.Smith, L.S.Lin & K. Resnicow (1998). Environmental Influences on Dietary Behavior among Children: Availability and Accessibility of Fruits and Vegetables Enable Consumption Journal of Health Education 29 (1), 26 32 Hendy, H. M., & Raudenbush, B. (2000). Effectiveness of teacher modeling to encourage food acceptance in preschool children Appetite 34(1), 61 76 Ishdorj, A., Crepinsek, M. K., & Jensen, H. H. (2012, March). Fruits and Vegetables: Do School Environment and Policies Affect Choice in School Meals? In 2012 AAEA/EAAE Food Environment Symposium, May 30 31, Boston, MA (No. 123534). Agricultural and Applied Economics Association. Jenkins, R. L. (1979). The influence of children in family decision making: parents' perceptions. Advances in consumer research 6(1), 413 418. Labrecque, J., & Ricard, L. (2001). Children's influence on family decision making: a restaurant study. Journal of Busin ess Research 54(2), 173 176 Laczniak, R. N., & Palan, K. M. (2004). Under the influence. Marketing Research 16(1), 34 39. Lee, C. K., & Beatty, S. E. (2002). Family structure and influence in family decision making. Journal of consumer marketing 19(1) 24 41. Lytle, L. A., Seifert, S., Greenstein, J., & McGovern, P. (2000). How do children's eating patterns and food choices change over time? Results from a cohort study. American Journal of Health Promotion 14(4), 222 228. Marquis, M., Dagenais, F., & Filion, Y. P. (2002). The habit of eating while watching television, the frequency of consumption of specific foods and food preferences, as reported by Quebec children. Can J Diet Pract Res 63, S104. Mc Nea l, J. U (1992). T h e little shopp e rs Am e rican D e m ographi c s 14(2) 48 52

PAGE 63

63 Morgan, P. J., Warren, J. M., Lubans, D. R., Saunders, K. L., Quick, G. I., & Collins, C. E. (2010). The impact of nutrition education with and without a school garden on knowledge, vegetable intake and preferences and quality o f school life among primary school students. Public health nutrition 13(11), 1931. Moschis, G. P., & Mitchell, L. G. (1986). Television advertising and interpersonal articipation in family consumer decisions. Advances in consume r research 13(1), 181 186. Ni c klas, T. A. T. B a r a nowski, K. W. Cu l len, G. B e r e nson ( 2001), E a t i ng p a t t e rns, diet a r y qu a l i t y a nd o b e si t y J ou r nal of the A m e rican Co l lege of N utr i t i on, 22, 599 608 Nu, C. T., MacLeod, P., & Barthelemy, J. (1996). Effects of age and gender on adolescents' food habits and preferences. Food Quality and Preference ,7(3), 251 262 Prez Rodrigo, C., Ribas, L., Serra Majem, L. L., & Aranceta, J. (2003). Food preferences of Spanish children and young people: the enKid study.European Journal of clinical nutrition 57, S45 S48. Pirouznia, M. (2001). The association between nutrition knowledge and eating behavior in male and female adolescents in the US. International journal of food sciences and nutrition 52(2), 127 132. Shoham, A., assessment of children's influence on family consumption decisions. Journal of Consumer Marketing 22(3), 152 160. Skinner, J. D., Carruth, B. R., Bounds, W., & Ziegler, P. J. (20 02). Children's food preferences: a longitudinal analysis. Journal of the American Dietetic Association 102(11), 1638 1647. Wardle, J., Guthrie, C. A., Sanderson, S., & Rapoport, L. (2001). Development of the children's eating behaviour questionnaire. Jo urnal of Child Psychology and Psychiatry 42(07), 963 970. Wolfe, W. S & C. C. Campbell (1993). Food pattern, diet quality, and related characteristics of schoolchildren in New York State Journal of the American Dietetic Association 93(11), 1280 1284 Yo ung, E. M., & Fors, S. W. (2001). Factors related to the eating habits of students in grades 9 12. Journal of School Health 71(10), 483 488.

PAGE 64

64 BIOGRAPHICAL SKETCH Yu a n J iang is a student o f the Mast e r o f Sci e n c e p rog ra m i n F ood a nd Res o u rc e E c onom ic s D e p a rtme n t a t Unive r si t y of F lorid a Y u a n w a s born in S h a ndo ng Ch i n a S he a t te nd e d S h a ndo n g Uni ve rsi t y in C hina to ga in her b ac h e lo r egre e in e c o n om ic s f r om 2007 to 2011. A f ter su cce ssful gr a du a t i on, she w a s admitted to the U nive r si t y of F lorida a nd majo r e d in f ood a nd r e sour c e e c onom ic s. S he sch e duled to gra du a te i n December 2013. A f ter gra du a t i on, s he will c ont i nue h er edu ca t i on a nd s ee k f o r PhD in a g r icultur a l economics at University of Florida.