6.1:.”
.1 .aw .
évwgﬁ

”8.5! . . max... .
.mwmﬁw. 71v» 7
r: _ ‘ .
.; L was; ‘
. pa Hui...
xv... l. .
7 . a?“
_ a 3%

s. E

v: ‘
litwwymm 1T.

i.» . any ta ..
. . $.3Nwﬁﬁwv

ft: .35 a.
1.: an .33...
in? _. g.
. a .75.; .:.
a. 2.»: n3.
. 7.3%

3.1 i .

h.

h V

x. 1.3%.."

«9 .c. :. :7
cytnurh

vow ._ :sai. a...

J P31

.. ﬁﬁf a» .. A
” Emcmém f:
. ﬁriwﬁri 3 .
, .— .O
3.» (mars

» .
. ‘ . #1
gain . ,
, . . Gm?

.5“ . 1

Q. a

ﬁll:

3.1.? ass“! a.
.111 ‘

51....

.2!»

A.

_ 1....
. L).
. .v.

., a

".0an s n. p 1"
his”. a.

1.... w. a;

“mum“?

.z

. F
$35.

 

 

THESIS
/)

Zacl

This is to certify that the

thesis entitled

BEVERAGE PREFERENCES OF LOW-INCOME PRESCHOOL

CHILDREN: RELATIONSHIP TO GROWTH, BEVERAGE

AVAILABILITY AND BEVERAGE PREFERENCES OF
THE PARENT

presented by
Jennifer R. Almy

has been accepted towards fulfillment
of the requirements for

M-S. degree in HnmaLNntrition

    

Major professor

Date December 6. 2000

0-7639 MSU is an Afﬁrmative Action/Equal Opportunity Institution

 

 

LIBRARY
Michlgan State

Universlty

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

11/00 worm-m.“

 

___ ___—- _ _ _. _

 

BEVERAGE PREFERENCES OF LOW-INCOME PRESCHOOL CHILDREN:
RELATIONSHIP To GROWTH, BEVERAGE AVAILABILITY AND BEVERAGE
PREFERENCES OF THE PARENT
By

Jennifer Rebecca Almy

A THESIS
Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of
MASTER OF SCIENCE
Department of Food Science and Human Nutrition

2000

ABSTRACT
BEVERAGE PREFERENCES OF LOW-INCOME PRESCHOOL CHILDREN:
RELATIONSHIP To GROWTH, BEVERAGE AVAILABILITY AND BEVERAGE
PREFERENCES OF THE PARENT
By

Jennifer Rebecca Almy

FOod and beverage intakes of young children can help shape their future eating
habits. Relationships exist between the intake of children and their growth status, as well
as family food choices and the availability of foods in the home. This study examines
children’s and their parents’ beverage preferences relative to the children’s growth data,
availability of beverages in the home and the parents’ nutritional beliefs about beverages
tested. Fifty children and their parents or primary caregivers enrolled in an urban Head
Start program were examined for their beverage preference and intake habits. Signiﬁcant
relationships were found between selected beverages, weight for age and height for age
data, as well as the availability of beverages for the children at home and the same
growth values. Signiﬁcant results were also seen between selected child beverage
preferences and those beverages that are offered to the children at home. No signiﬁcant
results were found relative to beverages in the home and the parents’ beliefs about the
nutritive value of the beverages tested. Relationships between parent preferences and
certain beverages and the presence of beverages in the home were shown to exist.
Relationships are evident when looking at beverage preferences and growth, as well as
beverage preferences and availability, however, additional factors can affect the beverage

intakes of parents and children. Further studies are needed to clarify these relationships.

To my husband Erick, my mother, my sister Christine
and in memory of my father.

iii

ACKNOWLEDGEMENTS

I would like to express gratitude to my major professor Dr. Jenny Bond for
guiding me through my coursework, teaching experiences and research project. Her
wisdom and remarkable patience have enabled me to successfully complete my graduate
program. I would also like to acknowledge the other members of my guidance
committee, Mrs. Stella Cash and Dr. Stephen Yelon. Mrs. Cash introduced me to the
world of dietetics with her enthusiasm and energy toward the profession. I also thank her
for her support during the internship selection process. Dr. Yelon provided me with a
greater appreciation for and interest in the ﬁeld of education.

I would also like to thank several other department faculty members, Dr. Sharon
Hoerr, Dr. Won Song, Mrs. Esther Park and Ms. Marilyn Mook. Each provided valuable
coursework and teaching experiences guiding me toward my dietetics career. In addition,
I want to express my appreciation to Theresa Hargterink, as well as the teachers, social
workers and families from Head Start for their participation in this study.

I thank my friends in the department, Marci Scott, Sue Davis, Zalilah Mohd
Shariff, Sara Pocernik, Diane Golzynski and Danielle Davidson. Graduate school would
not have been the same without your continued ﬁiendship. I thank all of you for shaping
my Michigan State experience.

I thank the Clinical Nutrition Staff at Carolinas Medical Center in Charlotte,
North Carolina, for their support during the writing of my thesis. I would also like to

thank my sister Christine and my best friend Heather for their motivation and laughter. 1

iv

want to express my gratitude to my father- and mother-in-law Mr. and Mrs. Alan Almy
for making Michigan my second home.

Thank you to my mother for her constant love and words of encouragement
despite the geographical distance between us. Finally, thank you to my husband Erick for

his time, love and amazing patience. Without him, I would not be where I am today.

TABLE OF CONTENTS

TABLE OF CONTENTS .................................................................... vi
LIST OF TABLES ............................................................................ ix
CHAPTER I INTRODUCTION ................................................. 1
CHAPTER II LITERATURE REVIEW ........................................ 3
Introduction ............................................................................. 3
Food Programs Targeting Preschool Children ..................................... 5
Enviromnental Impacts on Children’s Food Consumption and Eating
Habits ..................................................................................... 9
Family and Peer Inﬂuences on Children’s Nutrition Choices .................... 11
Working with Preschool Children ................................................... 14
Determining Food Intakes of Preschool Children ................................. 15
Preschoolers’ Food Exposure and Acceptance ..................................... 17
Contingency and Reward Use of Foods ............................................. 19
Preschoolers’ Energy Intake and Growth Status .................................... 21
Excessive Fruit Juice Consumption In Preschool Children ........................ 23
Water and Other Beverages Consumed By Preschoolers .......................... 27
Conclusions .............................................................................. 30
CHAPTER III METHODS ............................................................ 31
Subjects .................................................................................. 3 1
Materials .................................................................................. 3 1
Research Site ............................................................................. 32
Procedures ............................................................................... 32
Statistical Analysis ..................................................................... 38

vi

Null Hypothesis Statements ........................................................... 38

CHAPTER IV RESULTS ............................................................. 39
Demographic Data ..................................................................... 39
Anthropometric Data .................................................................. 42
Beverage Collection and Caregiver Questionnaire ................................. 44
Hypothesis 1 ............................................................................. 46
Hypothesis 2 ............................................................................ 52
Additional Results ..................................................................... 55
Summary of Null Hypotheses ........................................................ 60

CHAPTER V DISCUSSION ........................................................ 61
General Information ................................................................... 61
Hypothesis 1 ............................................................................ 64
Hypothesis 2 ............................................................................ 67
Additional Information ................................................................ 68

CHAPTER VI CONCLUSION ...................................................... 69
Study Strengths and Limitations ....................................................... 69
Implications for Practice and Education .............................................. 70
Implications for Future Research ..................................................... 70

APPENDIX A UCRIHS Letter ........................................................ 73

APPENDIX B Parent/Guardian Questionnaire .................................. 75

APPENDIX C Cartoon Faces for Beverage Preference Exercise ............ 80

APPENDIX D Consent Form ......................................................... 82

APPENDIX E Preschool Child Information Sheet .............................. 84

vii

APPENDIX F Head Start Beverage Information Collection Sheet ......... 86
APPENDIX G Beverage Preference Data Collection Sheet .................. 89

REFERENCES ................................................................................ 9]

viii

TABLE 3.1

TABLE 4.1

TABLE 4.2

TABLE 4.3

TABLE 4.4

TABLE 4.5

TABLE 4.6

TABLE 4.7

TABLE 4.8

TABLE 4.9

TABLE 4.10

TABLE 4.11

TABLE 4.12

TABLE 4.13

TABLE 4.14

TABLE 4.15

LIST OF TABLES

Variable coding used in statistical analysis ................................. 37

Demographic data collected from the caregiver survey and Head Start

records ............................................................................ 41
Anthropometric data at birth and at time of the study ..................... 43
Factors contributing to signiﬁcant differences between weight .......... 47
Signiﬁcant differences between mean values of weight for age .......... 48

Signiﬁcant differences between means of weight for age grouped ....... 49
Signiﬁcant differences between means of height for age grouped. . . . . ...49

Signiﬁcant differences between means of weight for height relative
to the availability of a beverage at home ..................................... 51

Signiﬁcant differences between means of weight for height
grouped relative to beverage preference and availability .................. 51

Signiﬁcant differences between mean ability of children to
identify beverages relative to beverage availability.. . .5...........3

Signiﬁcant differences between means of children’s ranking of
beverages relative to the availability of beverages at home ............... 54

Signiﬁcant differences between means of parent’s ranking of
beverages relative to the availability of beverages at home ................ 54

Signiﬁcant differences with ability to identify beverages with
respect to gender .................................................................. 57

Signiﬁcant mean anthropometric differences with respect to
ethnicity ........................................................................... 58

Signiﬁcant differences between means for ranking beverages with
respect to ethnicity ............................................................... 59

Signiﬁcant differences between ability to identify beverages with
respect to ethnicity ............................................................... 59

CHAPTER I

INTRODUCTION

All individuals begin life on a milk-based diet, but the intakes of older infants and
children rapidly change and develop into eating patterns that are also shaped by their
culture and social environment. In today’s society, many individuals take for granted the
wide array of foods from which they choose at each meal. Whether or not a country is
fortunate enough to have the necessary foods available to feed its children, all efforts
should be made to provide children with foods that will enhance their growth and
development. This is important during the early years of life, particularly infancy and
preschool years. Children are learning about foods, experiencing eating in their particular
environment and physically growing. Such a period is ideal for the introduction of foods
that are an important contributor of essential nutrients to a young child’s diet.

During these critical years, children are trying new beverages, including fruit
juices (deﬁned as 100% juice beverages), juice drinks (beverages that do not contain
100% juice, such as Hi-C and Kool-Aid) and soft drinks. In addition to water and milk,
which they have had since birth, children are becoming familiar with these newer
beverages, especially if the beverages are readily available — at home and other frequently
visited places. Fruit juice is not introduced until four to six months of life to assure that
the child will gain many nutrients and protective factors from the breast milk or formula
which they have received until this point. In addition, parents often push ﬂuid
consumption as beverages are generally easy to prepare, readily available and accepted
by infants. Fruit juice consumption has increased for all age groups in recent years,

primarily due to the “healthy” and “natural” images attributed to it. Often fruit juice and

other beverages are used to replace milk, water and even solid foods. A growing child
needs the beneﬁt of variety in the diet. Diets high in fruit juice may lack the variety and
nutritional value that diets with lower amounts of juice contain. Excess intake of fruit
juice and other non-protein based beverages may lead to concerns about the growth of
young children, including failure to thrive, short stature and overweight. Also increased
juice consumption in young children may contribute to chronic non-speciﬁc diarrhea
(CN SD) and gastrointestinal problems due to the malabsorption of sugars, speciﬁcally
fructose and sorbitol found in juices. Such growth issues so early in a child’s life can
lead to increased problems, as they grow older.

Analyzing the food preferences and intakes of preschool children and their
parents’ juice consumption is one way to begin to understand what beverages children are
consuming, as well as determine the effects that parents have on these choices. This
study will look speciﬁcally at the beverage preferences of low-income preschool children
and the relation to children’s weight and height. In addition, the parents’ nutrition beliefs
and availability of the tested beverages to the child will be analyzed. The availability of
the tested beverages at home will provide further information as to the effect parents have

on the beverage preferences of their children.

CHAPTER II

LITERATURE REVIEW

Introduction

In the past several decades, childhood health and nutrition have been gaining
importance among nutrition professionals, educators, school systems and families.
Programs have been developed to speciﬁcally target childhood nutrition practices and
status, including the National School Lunch Program in 1946, Women, Infants and
Children (WIC) program in 1972, as well as the more recent National School Breakfast
program in 1975 (Boyle and Morris, 1994). Even as these programs prove to have a
positive effect on childhood nutrition, some children remain at risk for the physical,
sociological and intellectual impacts of malnutrition. Mild malnutrition, at the very least,
can result in children who are ﬁ'ustrated, initable, lack interpersonal skills and suffer
from increased infections due to diminished resistance (American Dietetic Association,
1990). While chronic hunger is a reality in the United States for some children, most
children are fortunate to have several food choices in adequate amounts at each meal.

On the other end of the spectrum, childhood obesity is a growing problem in the
United States. Factors contributing to the increased prevalence of overweight in children ,
include the increased availability and popularity of fast food choices and lack of physical
activity. For infants and preschool children, in particular, there has been recent concern
regarding the excessive consumption of fruit juice and other beverages, such as
carbonated soﬁ drinks and juice drinks, instead of milk, water and even solid foods. By
twelve months of age, 90% of infants in the United States are regularly drinking juice.

Juices are the most consumed beverage by children under the age of ﬁve. (Smith and

Lifshitz, 1994) The issues surrounding the relationship between increased juice
consumption and growth will be discussed further later in this review.

In a longitudinal study by Skinner and colleagues in 1999, fruit drink and
carbonated beverages were among the foods most commonly consumed by children aged
24-60 months. In fact, four of the top six foods were beverages, including fruit drinks,
carbonated beverages, 2% milk and apple juice. Fruit drink consumption was the highest
at ages four and ﬁve, while carbonated beverage intakes increase as age increased. Milk
and apple juice were consumed more oﬁen at the younger ages.

From a very young age most children are taught (most often by example) about
the consumption of food, including what is appropriate to consume, as well as where,
when and how to consume certain culturally appropriate foods. With an increasing
number of “non-traditional situations”, such as multi-cultural families and the increased
number of hours per day children spend in day care settings, many children are being
removed from the “traditional” home environment and experience new foods in new
settings. The new meal settings have an array of fast food and pre-prepared food items,
frequently targeting working mothers and fathers who are searching for easy ways to feed
a family and save time. Many of these foods have a low nutrient density and are high in
fat, saturated fat and cholesterol. Increased salt and sugar contents are also
characteristics of many foods presented to our children. Marketing tools are also
encouraging children to increase their intake. “Super-sized” beverages and other extra
large food items may even promote increased consumption at one setting at little extra
cost. Toys and games are more popular than ever at fast food restaurants, easily

attracting families with children for a quick “fast-food” meal.

To decrease intakes of “fast food” or low nutrient dense foods and encourage
foods that are more nutrient dense, our children need to be exposed to healthful food
choices at an early age if we expect them to continue to eat nutritious foods.

Unfortunately, today’s society does not make this a simple task.

Food Programs Targeting Preschool Children

The Women, Infants and Children Program (WIC) and the Child and Adult Care
Food Program (CACFP) provide direct assistance with foods for preschool age children.
WIC has gained nationwide recognition for its assistance to low income families, with
pregnant women, breastfeeding mothers and children up to the age of ﬁve. WIC provides
coupons for foods such as infant formula, milk, cheese, cereal, juice, eggs and peanut
butter. These are all foods that when incorporated into a preschool child’s diet will
increase the nutritive value of the diet. The supplemental foods from WIC do not serve
as a quick cure for malnutrition problems, but as an initial step to assist children of low-
income families in obtaining the nutrients they need. CACF P provides non-residential
child and adult care centers, both public and private, with federal funding and commodity
donated food items. This program helps to ensure that preschools and child care centers
are able to adequately feed children nutritious snacks and meals while they are away from
their parents (Boyle and Morris, 1994).

In 1999, Bruening and colleagues, studied the effectiveness of CACFP by looking
at food and beverage intake and health outcomes in urban preschool care centers.
Studied were 40 three to ﬁve year old children in one of two centers, one that used

CACFP and another that did not participate in the program; instead food and snacks were

brought to the center from home. The center participating in CACFP provided
signiﬁcantly more servings of milk and vegetables, as well as lower amounts of fat and
sweet items (Bruening et a1., 1999).

In 1994, the American Dietetic Association (ADA) published their standards for
the nutrition component of childcare programs, including nutrient requirements for those
eating meals at childcare settings, as well as supplemental nutrition education for children
and care-givers. These standards state that meals served to the children should be
adequate and in accordance with the Dietary Guidelines, emphasizing age appropriate
meals and snacks, food safety measures, minimal amounts of fats, sugars and sodium and
reﬂecting appropriate portion sizes (American Dietetic Association, 1994). This
information should, if possible, correspond to foodservice aspects of the childcare
facility. In addition, the ADA position paper recommends that children should receive
adequate information on the relation of health and food habits to the body (American
Dietetic Association, 1994). The opportunity for children to practice good nutrition habits
at the preschool or child care center is an excellent way of reinforcing previously learned
lessons.

The nutritive value of the meals provided at childcare centers needs to meet at
least one-half to two-thirds of the RDA for that age group if it is a full-day childcare
center (American Dietetic Association, 1994). In a 1989 study, Briley and colleagues
analyzed menus of several childcare centers in Texas. The researchers’ focus was on the
variety of the items served to the children, as well as the nutritive value of the overall
daily diet. Results from this study showed that many care centers provided an adequate

variety of vegetables, meats, breads and cereals, and dairy products, but were deﬁcient in

providing a variety of fruits/fruit juices. While, an average of six to seven vegetables
were served to the children each week, the researchers found that on the average only one
type of fruit was included in the meals each week. Often the fruit was served more than
once during the week. An example would be if the care center served peaches on two
days with no other fruit served at all during that speciﬁc week (Briley et al., 1989). Also,
at least three different types of low nutrient dense foods, including sweets, were provided
for the children each week. In addition to low fruit variety, the menus overall were
deﬁcient in calories and iron, failing to meet one half to two thirds of the RDA. Finally,
portion sizes were signiﬁcantly smaller than the CACFP requirements.

As previously stated, a variety of menu items introduced to children at this age
can help to increase their overall acceptance of new foods. Ashraf and colleagues (1990)
studied the addition of tofu to school menus as a meat substitute to examine overall
acceptability of new menu items and to analyze their nutrient content with toﬁr as the
replacement. The low saturated fat, cholesterol and sodium levels of toﬁr make it an
attractive alternative to regular meat products served. Results showed that the children
accepted the new item and that there was a signiﬁcant increase in the nutrient quality of
the meals consumed using the tofu. Simple substitutions can introduce unfamiliar foods
to children and can also increase the amounts of nutrient dense foods and decrease the
amounts of saturated fat and cholesterol offered to the children.

Positive inﬂuences on nutrition and health are the multidisciplinary ways in which
nutrition can be introduced to childrenand the individuals that administer these methods.
Day care, preschool and elementary school environments are ideal for young children to

learn and practice good nutrition behaviors. Often they provide meals and snacks, as well

as structured activities to incorporate basic information about food and nutrition. In
addition to the National School Lunch and Breakfast programs, which provide ﬁ'ee or
reduced meals to school-age children of families meeting certain income criteria, there
are large-scale programs which provide ﬁnancial assistance to day care and preschool
settings around the country, such as Head Start. Head Start was developed in 1965 as
part of the US. Department of Health and Human Services. This program provides
children aged 3-5 from lower income families with comprehensive education, health and
nutrition services. As a result of this program’s success, many children nationwide are
more adequately prepared for kindergarten and are healthier than their non-participating
counterparts (Crockett and Sims, 1995). Head Start also includes a component to help
parents take responsibility for and become active in their child’s education and overall
development. The CACFP provides food and funding for Head Start programs
nationwide. Therefore snacks and meals must meet the US Dietary Guidelines and food
safety requirements.

A study testing a Head Start curriculum, Children, Get A Head Start on the Road
to Good Nutrition, analyzed the knowledge, attitudes and behavior of the children
participating. This curriculum contains a series of nine lessons that are interactive and
multi-culturally based. At the post-test time, the treatment group of children displayed an
increased knowledge of nutrition items, but showed little change in attitudes and
behaviors. Two signiﬁcant positive behavioral changes of children in the treatment
group were the increase in number of requests for low sugar snacks and the increase in

overall acceptance of foods (Byrd-Bredbenner, 1993).

Environmental Impacts on Children’s Food Consumption and Eating Habits

A constantly changing world inevitably creates changing environments for
children. Affecting the food behaviors and habits of children is the number of different
individuals involved in each child’s life. Both the family environment and surrounding
community in which children live greatly contribute to the behaviors of children,
including food consumption and habits. An immediate and inﬂuential factor on food
consumption and eating behaviors of children is the family environment, including the
parents’ employment status, education level of mother, age of the mother and cultural
and ethnic traditions, as well as their basic socioeconomic level (Crockett and Sims,
1995). Many families are classiﬁed as dual income where both parents are working to
support the family unit. In fact, half of all preschool children have working mothers
(Ashraf et al., 1990). In these situations, time is of great value to the working mother, as
she must plan meals for the family, maintain work hours and the household. Extended
work hours or commutes often leave children at preschool for many hours each day;
therefore, forcing the child to make eating decisions under the guidance of school
personnel without the direct encouragement or inﬂuence of their parents. The education
level of the mother is also related to overall household earnings and food expenditures.
Finally, due to decreased social support and income, the growing population of
adolescent mothers frequently suffer from malnutrition or have developed deleterious
eating habits of their own. These poor habits may create detrimental eating environments
for their children.

At-home nutrition activities parents and children can share together are often

developed by nutrition educators based on knowledge of the parents’ educational level.

Programs such as Hearty Heart and the Home Team combine to focus on a parental
component and working with the children during the school day. With this family
involVement, parents have been able to increase their knowledge and attitude of heart
smart foods (Lytle and Achterberg, 1995). Parents who increase communication with
their children about healthy foods and offer good foods at mealtime can have a positive
impact on overall nutritional status of their children. The consumption of foods meeting
basic nutritional criteria at home is based on the availability of these foods to the children
for which parents are solely responsible. Children can begin to taste new foods talked
about in school if parents are aware and knowledgeable about healthy food choices.

The successful distribution of nutrition information to parents can often pose
some difﬁculties. The Home Plate study looked speciﬁcally at the delivery of nutrition
education to children and parents. A pre-prepared kit, containing a video, paper materials
for parents and joint parent-child activities, was distributed to families. Results showed
that 90% of the families preferred receiving information at home, as opposed to seminars
and workshops after school and work hours (Dunn et al., 1998). Activities at home can
be easy and convenient, plus allow parents a great deal of freedom as to what material is
covered and how closely they adhere to the kit instructions. Parent and farmly
involvement is necessary to increase the overall impact of a nutrition program as
classroom lessons are reinforced at home.

Cultural, ethnic and socioeconomic characteristics also individualize family food
consumption patterns and differentiate them from those of surrounding communities. As
the United States becomes more diverse culturally, speciﬁc foods are becoming

mainstream items in traditional childcare and school settings. With an estimated 30% of

10

the child population in 2000 being either Latino or African-American, new programs to
educate children, families and school personnel on the food consumption patterns of
these populations has reached new heights. (Crockett and Sims, 1995)

Increased participation in supplemental food programs, such as WIC and Food
Stamps, allows us to realize that there are many children suffering ﬁom malnutrition due
to their family’s low socioeconomic status. While children were only about 26% of the
total US population in the early 1990’s (National Commission on Children, 1991), they
comprised 39.6% of the poverty stricken (US Bureau of the Census, 1993). Working
mothers, especially if head of the household, often spend a larger percentage of their total
budget on food in order to feed their families quickly and easily.

The surrounding community environment also has a great impact on the
nutritional status of children, as it establishes the supply, safety and cost of food.
Location also plays a role in food availability. Urban children have the lowest kilocalorie
intake even with many food assistance programs centered in larger cities. The rural poor
and homeless suffer from poverty and decreased kilocalorie intakes leading to a
compromised health status. Less than 50% of a child’s Recommended Daily Allowance
(RDA) for zinc, iron, magnesium and folic acid was consumed by homeless children

according to Drake, 1992.

Family and Peer Inﬂuences on Children’s Nutrition Choices
While children are learning and making choices regarding food and nutrition at
preschool, the majority of their food experiences are with family members or peers.

Therefore, food preferences of children are affected by the choices that their family

11

members, in particular their parents and peers, are making. Children are easily affected
by the verbal and nonverbal messages of those around them (Anliker, et al., 1990). The
literature provides mixed results when looking at the extent of the effects of parents on
the food choices of children.

In 1990, Anliker and colleagues studied the effects of parental messages on
children’s awareness of nutrition. Children answered questions as to what foods they felt
made them “stronger”, “ grow more” and so forth. The children’s answers closely
corresponded to the amount of positive nutrition messages they received from their
parents. Again, this reiterates how easily children are inﬂuenced by those around them.

While the individuals in a child’s environment have an effect on their knowledge
of foods, studies show that there is little correlation between the intakes and preferences
of parents and those of their young children. The children may understand what is indeed
good for them, but are not sharing the preferences of their parents. Birch (1980b)
observed children’s eating patterns, obtained preference data for children and gathered
the parents’ preference information. While the children were eating foods they preferred
in the test setting, their preferences and intakes were different from the preferences of
their parents.

Skinner and associates (1998) tested toddler’s food preferences as a factor of the
preferences of various family members. The mothers reported the toddler’s intake. In
addition the intakes of the mother, father and a sibling, if applicable, were reported by
that individual. Results showed that no one family member had a signiﬁcant inﬂuence
over the child’s intake, in relation to other family members. The authors cited a meta-

analysis of ﬁve studies that showed weak correlation between the diet of parent/child

12

pairs (Borah—Giddens and Falciglia, 1993). The results of these analysis are supported in
the study above, showing a lack of signiﬁcant inﬂuence over the child’s diet. While there
is little effect of a single parent on the intake of a child, the overall effects of the family
have signiﬁcant implications as to the nutritional value of the child’s diet.

In 1996, Colavito and colleagues, found that the knowledge and attitudes of the
meal planners inﬂuenced the child’s intake of fat and ﬁber. The fat and ﬁber levels in the
preschoolers’ diets reﬂected those of the meal planners. Overall, when the meal planner
was concerned with taste, there was a decreased amount of ﬁber included in the diet of
the child.

Older peers also present an interesting inﬂuence, as they are still learning about
food and nutrition basics themselves. Birch (19803) studied the effects of peer modeling
on thefood choices of preschool children. Since eating is a social experience overall, the
author observed the children’s preferences for nine vegetables alone and then after seeing
what older children would choose. Surprisingly, the children exposed to older children
chose what the older child was eating, irrespective of whether it was a vegetable that they
preferred in the pretest session. The younger child would not only select the item, but
continued to consume the item as well. This poses an interesting twist as to what
inﬂuences older children may have at school on food choices of younger ones. So often
nutrition education focuses on the teacher and foodservice aspects, and ignores the older

children who may have a greater inﬂuence on choices of younger ones.

13

Working with Preschool Children

Preschool children are excellent targets for nutrition research and programs
because they are beginning to discover new foods, develop food habits and model the
habits and actions of family, friends and others around them. The most important aspect
for preschool children to understand is the concept that eating a variety of foods is
necessary. Nutrition programs for this age group can use several vehicles to reach
children, including preschools, daycare settings, their homes, television and even
supermarkets. In the late 1970’s, over 2 million children regularly attended preschool
(Jennings, 1979). Today, the number of children attending day care and preschool is
growing rapidly since the workforce is now comprised of two working parents in many
cases (Davis, et al., 1983). While this number is growing, sufﬁcient nutrition programs
are needed to handle the increases in attendance.

Children are at different levels of cognitive understanding; therefore research
methods must be at an appropriate developmental level. Younger children, two and three
year olds, are able to perform basic language skills such as naming and identifying
familiar items, as well as understanding their general surroundings. Children ages three
to ﬁve are beginning to be able to classify and identify items based on size, shape and
color. Children, as they begin to prepare for kindergarten, are able to understand
meaningful food groupings as taught to them. (Hertzler, 1994)

Piaget’s Cognitive Development Theory provides further insight as to the
developmental level of young children. The pre-Operational stage, children age two
through six, focuses on the use of primary language skills and the immediate

surroundings as more advanced cognitive skills are not developed. As the child reaches

14

the concrete operational stage, ages seven to eleven, they are able to understand abstract
thoughts and the reasoning process in a basic form. The researcher must understand the
skills of the age of the child that a research study will target for complete and accurate

performance by the child. All too often questions are posed at an advanced level so that

the child does not fully comprehend what is being asked.

Determining Food Intakes of Preschool Children

Past work has shown the limited ability of parents, most importantly mothers, to
report or recall their preschooler’s food intake for a speciﬁc period. In 1980, Phillips and
Kolasa found that mothers were not able to accurately report their child’s intake when the
children spent much of their day outside of the home, primarily in preschool or child care
settings. Baranowski et a1. (1991) looked speciﬁcally at the 24-hour recall abilities of
mothers of 3-and 4-year-old children. There was a 64% agreement between the mother’s
report and that of the trained observers of what the child ate during that period.
Underreporting foods was the common error made by those mothers.

Fortrmately preschool age children are able to provide researchers with some
information as to their food preferences and intakes. Simple methods and patience will
allow a researcher to carefully obtain diet information from a young preschool child. In
1979, Birch began to examine the food preferences of preschool children. With 17
preschoolers, ranging in age from 3 years and 7 months to 4 years and 7 months, the
researchers set out to assess the relationship between child preferences and their actual
intake of foods. To determine the preferences, Birch had the child select one of the food

items being tested (in this case eight small sandwiches with varied spreads), taste it,

15

identify the spread and then continue that process with the rest of the sandwiches. After
the tasting and identifying was complete, she had the children rank the foods in order of
preference. As a result, Birch found that when the children were allowed to sample any
sandwich spreads at another time, they would choose the items that they preferred during
the ranking session. She concluded in this early study that preference data is a good way
to begin to determine the general consumption of children at this age (Birch, 1979b).

In another 1979 study, Birch looked primarily at children’s preferences for eight
different fruits. The preschoolers were the same age of the previous study and performed
the same tasting and ranking exercise to establish a preference ranking of fruits. In this
study, however, the authors examined the stability of the preference data obtained. The
children were assessed again after seventeen days to see how preferences differed; then
again at a later time to assess the veriﬁcation of the preference method. Finally, they
were assessed one last time to see the initial effects of increased exposure of a food item.
The correlation for the ﬁrst and second assessments was 0.86 when looking at the data
from four-year-old children. The three year old children had a slightly lower correlation,
but still signiﬁcant at 0.71. Throughout this study, the focus was on the dimensions of
familiarity with the food and sweetness of the food. These dimensions, discussed later in
relationship to children’s food neophobia and exposure issues, are determinants of food
choices made by children. The familiarity, sweetness and other dimensions of the ﬁ'uits
were determined and then plotted against the children’s preference ranking. Over the
four assessments, children preferred the fruits that were familiar to them or that had a

distinct sweetness. The results of this study showed that three and four year old children

16

are capable of providing stable preference information over a period of time (Birch,
1979a).

To further simplify the ranking task of children, Birch (1981) utilized a process
for ranking preferences of preschool children in her study on the social contexts affecting
food preferences. This process began as that previously described with self-selection of
food items, tasting the item and then identiﬁcation of the item. After each item was
tasted and identiﬁed, the children then pointed to one of three cartoon faces representing
feelings about the food item: like, dislike or neutral. The food and face were recorded
along with comments from the children. This process was implemented for each of the
food items, one at a time. After completion, the children were asked which food in the
“like” category was the very best and that item was placed at the top of the preference
list. Then they were asked which item they liked second best and so forth in all
categories until all items were ranked. This procedure helped ease the process of straight
ranking by dividing up the items for the children initially after they tasted them. The
process took approximately 5 minutes per child and was accurate. Such information
from preschool age children is reliable and stable (Birch, 1979a). Other studies
including Birch et a1. (1982, 1984), Birch and Sullivan (1991) and Newman and Taylor
(1992) used this process to obtain preschoolers’ basic preference data for further

comparison with other variables in their respective studies.
Preschoolers’ Food Exposure and Acceptance

While Birch and colleagues demonstrated that valuable information can be

obtained from preschool children as to their food preferences, the next step was to

17

determine what information leads children to make the choices they do. Historical
research of the late 1930’s by both Davis (1939) and Duncker (1 93 8) provides a basis for
research today. Davis set out to examine the self-selection patterns of infants and young
children. Duncker, on the other hand, chose to study the effects of social inﬂuence on
food preferences. With their results and those of recent studies, it is known that there are
many environmental and social cues that help to direct food preferences and eating
behaviors. Taste and preference for foods are learned experiences, which are often
inﬂuenced by the body’s reactions to a food. If foods are positively experienced by the
child, there is an increased chance that the food will be accepted. It is often easy to tell
what initial reaction a food will cause by the facial expressions of an infant or child.
Since young children are not able to verbally express their feelings, often parents and
others rely on facial expressions as a sign of acceptance or rejection (Birch, 1990).
Expressions and initial reactions should not be the only factors that parents use to
determine a child’s preference for foods; however, such a means of communication can
assist the parents or caregivers.

Past exposure to foods can also increase the food’s acceptance by a child. A
study published in 1982 by Birch and Marlin, addresses the effects of exposure on
acceptance and preference in early preschool age children. With three- and four-year-
olds, studies suggest that familiarity and sweetness account for 50-60% of the variability
in preference data. While sweetness is a direct characteristic of the food, familiarity is a
factor related to the exposure the child has had with a food. Three-year-old children
relied on familiarity, while four-year-old children preferred foods that are sweeter (Birch

et al., 1982a). In this study, the researchers varied the number of exposures of cheeses

l8

and fruits for two-year-old children. They found that increased exposure of a food did
little to elevate the food’s preference of two-year—old children. Obviously, these children
are too young to familiarize themselves with a food item and recall that information at a
later time. Thus, the age of a child has an effect on the child’s food preferences (Alles-
White and Welch, 1985).

Perhaps one of the great mysteries to parents of young children is how to get them
to eat those foods that the parents claim the children do not like. When children are
introduced to new foods, there is an initial neophobic reaction that occurs. The children
are unfamiliar with this food and are not willing to accept it into their diet immediately.
About ten exposures to a food item are necessary for the children to feel comfortable
enough to ingest the food (Sullivan and Birch, 1990). Quite often parents of young
children give up long before ten exposures and determine that the child “will not eat” a
certain food. As they grow and experience more foods, children are able to increase the

variety in their diet more easily when the acceptance process becomes familiar.

Contingency and Reward Use of Foods

In addition to a child’s exposure to foods and the characteristics of the foods, the
use of food items as rewards or means to obtaining a reward have a distinct effect on the
child’s preference for that food. Not only does the quantity of food experiences determine
the child’s preference, but the quality and type of interaction that is experienced with the
food contributes. (Birch et al., 1982b). Two studies done by Birch and colleagues help to
relate preference of foods to their use in a means-end situation. In 1982, Birch studied

how children’s preferences changed for ﬁ'uit juices when the beverages were used as a

19

means reward in a fun child activity. The ranking for the juices followed the same
procedure used by Birch et al. (1980) and as described above using the cartoon faces to
help break down the activity. The control group of children was exposed to juices and
play activities as they normally would be in their day to day activities. There was a
decrease in preference for the juices when they were used as a means to reach an end
activity. For those children normally exposed to the juices, an increase in preference
resulted ﬁ'om the lack of the stigma of the juice as being a way to reach a reward.

In 1984 Birch and colleagues studied the effects of several ﬂavored milk
beverages as a means to reach a reward for three- to ﬁve-year-old preschoolers. They
found that negative shifts in preference resulted when the food was used as the means to
an end result. The cartoon ranking exercise was implemented to obtain the preference
information for each child. The children were presented with two types of praise, verbal
praise and a tangible reward (movie ticket) for drinking the milk beverages. There were
no differences in the results with respect to the age or sex of the children or the ﬂavor of
the milk.

Newman and Taylor (1992) examined the effects of the use of snack foods in
several situations, means-end contingency, temporal order and normal exposure. Their
results showed that when a snack was used as the means in a situation, the preference for
the snack diminished.

Opposite results occur when using food as the reward in situations. Birch and
colleagues (1980) examined the results of the use of food in four social affective
contexts. The preschool children were presented with a snack food item in one of four

social contexts: as a reward from their teacher, as part of non-contingency adult attention,

20

in a nonsocial situation and ﬁnally in a normal snack situation. The preferences for the
snacks were determined with the cartoon face exercise and a ranking of the snack foods
was completed. The researchers reported that there was increased preference for the
snacks after being presented with adult attention and as a reward from the teacher. The
nonsocial exposure and the normal snack exposure did not produce marked changes in
the preferences of the snack food items.

While rewarding children with food items does increase preferences for those
items, all too often such a plan backﬁres for parents. The foods used as rewards are often
sweets or savory snacks; however, the foods used as the contingency to reach a reward
are frequently vegetables or foods that the children do not want to eat in the ﬁrst place.
Due to the already innate preference for sweets and the use of sweets as a reward food, it
is obvious that children will prefer sweeter items and will reject the other foods that their

parents are trying to encourage them to eat.

Preschoolers’ Energy Intake and Growth Status

While determining the choices preschoolers are making is important, it is also
necessary to examine how those choices are affecting their energy and nutrient intake and
growth status. Children amazingly are able to regulate their overall energy intake. Birch
and colleagues (1991) examined the regulation of energy intake of preschool age
children. While the meal-to-meal intake showed 33% variability, the total intake for day-
to-day variability was 10%. Children easily compensated for a low kilocalorie dense
meal by increasing their intake of kilocalories, most often at the following meal. A

sample breakfast for the children ranged from 100 kcal to 350 kcal, but the children

21

would balance out increases or decreases soon thereafter. The children were offered a
maximum of 2700 kilocalories daily to eliminate any energy intake limitations (the RDA
for young children aged two to ﬁve is 1400 kilocalories/day) (Birch et al., 1991b). The
practical application for this research is aimed at parents. Without the understanding that
children are able to regulate their kilocalorie intake, parents often force children to
consume more than they need. As a result of this parental action, children are learning to
ignore their feelings of satiety and continue to eat excess amounts. Birch (1993) believes
that this provides a simple explanation for some cases of childhood obesity.
Unfortunately, this research does not take into account the increases in kilocalories and
percent fat of many of the foods that children are consuming.

Energy and nutrientintake and food preferences have a direct effect on a child’s
growth and development. This period of life is critical for individuals, as children are
gaining weight and height at a rapid rate. For low-income preschoolers, Such as the Head
Start population, there is an extra challenge in meeting nutritional needs for growth.
Often children born to low income families, regardless of participation in federal
assistance programs, are at an increased risk of being born at a low birth weight.
Whether they are born appropriate for gestational age (AGA) or small for gestational age
(SGA), these children experience challenges in catching-up in physical size and
development. Doong and colleagues (1998) examined the growth of a Head Start
population in four counties in a mid-westem state. They found that of the children that
were born LBW, 60% of the children were, at the time of the study, already below the
50‘11 percentile for height for age and 65% of the children were over the 50th percentile for

weight for height. While these children would not yet be classiﬁed as short stature or

22

overweight, respectively, they are falling into a pattern of abnormal growth. This pattern
is often exacerbated by poor nutrition and eating habits. Programs such as WIC attempt

to intervene with foods and nutrition information, but do not always reach the mothers in
time to stop deleterious feeding practices. One example of such practices would be

giving excess fruit juice.

Excessive Fruit Juice Consumption In Preschool Children

Fruit juices are often one of the ﬁrst foods that infants receive besides breast milk
or formula. Fruit juice has a positive connotation associated with it for most individuals.
As health professionals, we encourage Americans to eat ﬁve servings of ﬁ'uits and
vegetables each day (including fresh, frozen, dried, canned and juices). F ruit juice is a
good source of fruit and is recognized as a healthy beverage for adults (Dennison, 1997).
Controversy arises when adults take the messages that are tailored for them and translate
them to infants and very young children. Programs such as WIC often fail to
communicate information as to the amount of juice that parents should be feeding their
child. From personal experience in a WIC clinic, the researchers observed that many
mothers understand that WIC does encourage the drinking of juice by children once they
reach four months of age. Unfortunately, the mothers do not realize that the juice
consumption is to merely supplement the breast milk or formula the child is already
receiving. When coupons are presented to mothers for fruit juices for months, they may
see that as an approval for feeding their infants fruit juices frequently.

Fruit juice consumption has caused great controversy among pediatricians and

dietitians in the recent past. The types of juices consumed, the timing of the introduction

23

of juices and the amounts that should be consumed, as well as the adverse effects of
excess juice consumption have gained much publicity. Some of the adverse effects, both
direct and indirect, of excess fruit juice consumption include failure to thrive, obesity,
short stature, chronic non-speciﬁc diarrhea (CNSD) and gastrointestinal discomfort due
to the malabsorption of sugars, speciﬁcally fructose and sorbitol. Not all of the effects
are experienced by children consuming excess amounts of juice. Many of these effects
are speciﬁc to a particular age group or are the result of a particular juice product.

For infants, the need for formula or breast milk Outweighs many of the other food
items and beverages parents attempt to feed their young child. In 1992, approximately
164 million dollars were spent on bottled juice for infants alone (Klish, 1997). Smith
and Lifshitz (1994) studied the effects of excess ﬁ'uit juice consumption in children with
non-organic failure to thrive. These children, one- to two-years-old, were examined
through dietary and biochemical evaluations to determine the amount of juice being fed
to the child by the parents. Excesses of juice for the young children ranged ﬁom 12 to 30
ounces each day. The researchers found that all of the children were below the 5’11
percentile for weight for age and over half were below the 5th percentile for weight for
height. Such statistics led the researchers to intervene to change the diets of the children.
With a drastic decrease in juice consumption or removal of juice from the diet
completely, these young children showed growth improvement, including increases in the
total kilocalorie intake for each day. Smith and Lifshitz (1994) found kilocalorie per
kilogram intakes ranging from 80 — 94 kilocalories/ kg body weight for the children on
the excessive juice diets, with approximately 59-74% of the kilocalories from

carbohydrates. These excessive carbohydrate percentages were due to the excessive juice

24

intakes. The intake range for the juice restricted diets for the same set of children ranged
from 98 - 118 kilocalories per kilogram body weight, with a range of 42 -— 52% ﬁ'om
carbohydrates. In addition, there were marked gains in the weight of the children on the
juice restricted diets when examined in the study for the ﬁrst few months of the
intervention period. The mean weight gain after the intervention was 0.36 kilograms per
month, which is an increase over the 0.08 kilograms per month weight gain rate for the
children when consuming the excess fruit juice. The formula, milk or breast milk, as well
as infant foods in this population are essential for the children to reach the necessary
kilocalorie and nutrient intake they need to maintain the rapid growth and development
occurring at this age.

Dennison and colleagues found very different results in a study published in 1997.
The population in this study was healthy two- and ﬁve- year-olds. The researchers set out
to examine the fruit juice consumption and growth parameters of the children. They
looked at the consumption of milk, fruit juice, Kool-Aid juice drink and soda through the
use of a seven-day food record completed by the children’s parents. In addition, the
child’s activity level was assessed to gain perspective as to the energy each child might
expend each day. The weight and height of the children was aSsessed and used to
calculate Body Mass Index, BMI, and Ponderal Index. The researchers deﬁned an
excess fruit juice as over 12 ounces of ﬁ'uit juice per day and a lack of milk as less than
16 ounces per day. Pop, milk and other drinks were more commonly consumed by the
ﬁve-year-old children, while the two-year-olds were drinking excesses of fruit juice.
The younger children were also consuming signiﬁcantly more total beverages per day

than the older children, 43.2 vs. 32.6 ml/kg body weight/day, respectively.

25

The researchers concluded that 42 % of the children drinking excesses of fruit
juice were below the 20th percentile for height for age, compared to the normal ﬁ'uit juice
drinking group in which 14% were below the 20th percentile. Low milk intake was not
linked with short stature in this age group at the time of the study.

The prevalence of overweight was elevated in children drinking excessive fruit
juice, particularly in the two-year-olds. With 53% of the excess fruit juice drinkers
having a BMI above the 75th percentile (32% for normal juice drinkers) and with 32% of
those being above the 90th percentile (9% for normal juice drinkers), there is great
concern for the effects of excess juice consumption on the weight of young children. In
conclusion, Dennison and colleagues (1997a) determined that the excesses in juice
consumption were related to the overweight status of the two-year-old children, while a
lack of activity contributed to the overweigh of the ﬁve-year-old children. Unfortunately,
several concerns regarding this research exist. These include the limited sample size of
children consuming excess amounts of fruit juice (n=19), the lack of information as to
the child’s genetic makeup, the additional foods consumed by the child during the course
of the study and the cutoff points determined for the excess juice intakes (Dhurandhar,
1997; Kennedy, et al., 1997 and Stinson, 1997).

Concerns surrounding fruit juice consumption are not limited to a child’s growth.
Some juices cause great gastrointestinal discomfort in children and carbohydrate
malabsorption leading to increases in diarrhea. Different juices contain varied levels of
carbohydrates , including glucose, fructose, sucrose and sorbitol. Fructose is the sugar
that is most often malabsorbed in children. Increased levels of sorbitol in juices will

inhibit the absorption of fructose, as will decreased levels of glucose. Fructose can be

26

readily absorbed with the assistance of a glucose transport system, but adequate amounts
of glucose need to be present to handle the fructose. Smith et a1. (1995) assessed the
absorption of white grape juice and apple juice in a preschool population through
hydrogen breath analysis and reports of gastrointestinal symptoms from the children.
Apple juice contains sorbitol and twice the amount of fructose and glucose, therefore
causing malabsorption problems and gastrointestinal discomfort in many children. White
grape juice contains no sorbitol and equal amounts of fructose and glucose; therefore,
facilitating absorption easier. These researchers found that about 54% of the children
experienced some form of fructose malabsorption when consuming the apple juice,
whereas only 19% showed signs of malabsorption when consuming the white grape juice.
In 1991, the American Academy of Pediatrics recommended that apple and pear juices
should be used in moderation in children due to the increased prevalence of diarrhea and

discomfort reported in children consuming these (Klish, 1995).

Water and Other Beverages Consumed By Preschoolers

While much of the research related to beverage consumption in preschool
children focuses on fruit juices, there are several studies reviewing basic trends in
beverage consumption among young children. A British study by Petter and associates
(1995) surveyed the beverage consumption of young preschool and elementary age
children. Through recording of beverages for a two day period, one week day and one
weekend day, and an interview, the child’s parents provided the researchers with typical
beverage consumption. The most popular beverage categories included squashes (diluted

fruit drinks), milk, diet drinks and water, primarily with older children. For the preschool

27

children, 71.8% never drank plain water while only 50.0% of the older children
consumed water in the forty-eight hour period. Over the two day period 15.4% and
16.4% of the preschool and elementary children consumed water only once.

Petter and associates (1995) also determined that in both groups the consumption
of squashes was by far the highest of any beverage. Squashes would be the equivalent of
Hi-C, Kool-Aid and other fruit ﬂavored drinks in the United States. Interestingly, while
the children were consuming increased amounts of fruit drinks, there was little
consumption of carbonated beverages in relation to the other beverages consumed. The
sweet taste of the squashes and the apparent nutritious appeal of these fruit drinks to
parents make them a popular choice when feeding young children. Finally, Petter and
colleagues (1995) concluded that approximately 15% of the children received their total
recommended intake of carbohydrate from low-nutrient dense fruit drinks. As a result
they Concluded that future problems related to fruit drink intake may include excessive
energy intake, obesity and a lack of necessary nutrient intake through meals. In addition,
dental caries and cavities were concerns with the increased carbohydrate consumption.

Another study examining beverage intakes of children in Athens, Greece focused
on the relationship between bone fractures in children and their drink consumption
(Petridou et al.,1997). A 51-item semi-quantitative food frequency questionnaire was
completed for each child with information obtained from both the child and his/her
parents. Overall, increased amounts of non-carbonated beverages, carbonated beverages
and carbonated non-cola beverages were consumed, as compared to the other beverages.

Milk was also consumed in large quantities by many of the children. This study did not

28

ﬁnd that the beverage consumption type or amount was related to bone fracture
frequency.

Harnack and colleagues reviewed data from the 1994 Continuing Survey of Food
Intakes by Individuals (CSFII). Their focus was soft drink consumption by preschool
aged children, school age children and adolescents. Almost half of the preschool age
children in the sample (ages three to ﬁve) did not consume soda at all, while almost 40%
consumed up to nine ounces per day. The consumption percentages are markedly
increased for school age children and adolescents with intakes for the latter up to 26
ounces per day or more. In addition, Harnack et a1. (1999) determined that elevated
levels of carbonated beverage intake were related to decreased milk and fruit juice
consumption. Such beverage patterns may lead to poor eating habits in adults including
dependence on caffeine and increased kilocalorie intake.

A ﬁnal study focused on the intake of sugars, excluding milk and plant sugars, in
beverages by children ages 1.5 to 4.5 years old. (Gibson, 1997) These sugars, described
as “non-milk extrinsic sugars” in the study, include table sugar, syrups, honey and sugars
found in fruit drinks. The mean intake of energy ﬁ'om these sugars was 18.9% and
18.6% of total caloric intake for boys and girls, respectively. In addition, as higher levels
of non-milk extrinsic sugars were consumed, the author found that there were signiﬁcant
decreases in the intake of calcium, zinc, thiamin, riboﬂavin and niacin. Vitamin C
intakes were adequate or exceeded the recommended level for most children. The
increased percentage of energy from juices and decreases in micronutrients cause one to

wonder why it is that people are continually feeding children so much juice.

29

Conclusions

While there has been a great deal of research involving the eating behaviors and
food preferences of children, there are still many unanswered questions. All factors
affecting each child’s food habits and preferences are immeasurable. Targeting the most
inﬂuential factors in a child’s environment is a basic and logical ﬁrst step. Longitudinal
studies are necessary as we further examine relationships between food intake and growth
status of children. Exploratory research is also necessary to begin to discover other
possible relationships between children, their environment and their eating habits.
Methods for obtaining food intake and preference information from younger children and
their parents are beginning to take shape, but need further attention. This study will
examine relationships between the beverage preferences and growth parameters of Head

Start children, as well as the beverage preferences and nutrition beliefs of their parents.

30

CHAPTER III

METHODS

Subjects

The subjects for this study were 50 preschool children and their primary caregiver
who participated in the Capital Area Community Service (CACS) Head Start program in
a mid-western urban area. The Head Start program beneﬁts children approximately ages
3-5 from low socioeconomic status. Head Start focuses on early child development,
nutrition, health and general social services. Another unique feature of the program is the
required participation of the parents or caregivers in the child’s development and growth
(Boyle and Morris, 1994).

Human subject use approval was obtained from the University Committee on
Research Involving Human Subjects (UCRIHS) at Michigan State University prior to

beginning the data collection. This approval letter can be found as Appendix A.

Materials

Materials included the National Center for Health Statistics (N CHS) growth
charts for males and females aged birth to thirty-six months and aged two to eighteen.
Updated charts had not yet been published at the time of the data analysis. All growth
information plotted by the researcher on the NCHS growth charts was obtained ﬁ'om the
children’s Head Start records. A second researcher veriﬁed all data entered. A survey
was developed and distributed to the children’s primary caregiver (parent/ guardian) for

information as to the beverage preferences of the parents and the beverage consumption

31

habits at home (Appendix B). A pilot test of this questionnaire is described under
procedures.

The materials necessary for the beverage preference exercise included: clear
plastic tasting cups, plastic gloves for preparation and seven commercially prepared
beverages (2% white milk, 1% chocolate milk, bottled water, caffeine-free Coca-cola,
100% apple juice, Hi-C Boppin’ Berry Punch and Kool-Aid Strawberry-Kiwi drink). In
addition, laminated pictures of clearly drawn cartoon faces expressing the feelings of like,

dislike and neutral were used to help the children classify their beverage preferences

(Appendix C).

Research Site

The study took place at several of the Head Start classroom sites during the fall of
1998. The classroom sites varied, depending primarily on the availability and
participation of the children and caregivers at each site. All classrooms required the same
criteria for participation in the Head Start program and this study. All classrooms were
part of one overall Head Start program. Primary caregiver data were collected during

group parent/caregiver meetings at the various Head Start classroom sites.

Procedures
Consent was obtained ﬁ'om the primary caregivers prior to collecting the
children’s beverage preference data (Appendix D). In addition, each caregiver completed

a form containing child food and beverage allergy information (Appendix B). At the time

32

of the study, no child had allergies to any of the beverages tested as reported by the
caregiver.

The growth status of each child was examined through the use of the most recent
weight, height, age, birth weight and gestational age data as provided by the Head Start
site. These data were plotted on the age and sex speciﬁc NCHS growth charts. By
plotting the weight for height and the height for age, the researchers were able to assess
the children’s body weight and stature relative to nationally used standards. These values
are commonly used to assess the growth of young children. The weight for height data
takes into account the height of the child in determining the percentile for the child’s
weight, providing the examiner with the child’s weight percentile independent of age. A
useful assessment of a child’s weight cannot be determined by their weight for age value
(Falkner, 1985). The height for age value, examines the child’s stature and is a “stable
and valuable” indicator of short stature (Falkner, 1985). The data collected from this
portion of the study was used to identify correlations between beverage preferences and
growth. Weight for height percentile values above the 95m percentile were considered
overweight and under the 5th percentile were underweight. As for height, short stature
was deﬁned as values below the 5th percentile for height for age (Trowbridge, 1983).

To obtain information regarding parent and child beverage consumption and
beverage exposure in the home, the parent/ guardian was asked to complete a short
questionnaire. A pilot test of twenty-six primary caregivers participating in the Women,
Infant and Child (WIC) Supplemental Food and Nutrition program in the same mid-

westem urban area was conducted to examine the readability and overall format of the

33

questionnaire. Two versions were tested on the WIC participants and revised until the
ﬁnal version was developed.

The study questionnaires were completed at the monthly parent meetings held by
the Head Start staff. The researcher was present at all meetings to answer any questions
and assist anyone needing help completing the questionnaire. The primary caregivers
provided information regarding their assessment of the nutritional value of the beverages
the children were consuming. Speciﬁcally, they were asked to rank the beverages that
the children tested. Finally, the parent/ guardian answered questions pertaining to the
availability of beverages in the home for the child, focusing on the past three months.

The beverages offered to the children at Head Start as part of their regular feeding
were recorded for 5 consecutive days during a typical week. The type of drink, time of
day when the drink was offered and most recent meal offered to the children was
recorded on a data collection sheet included in the appendix. These data provided
information regarding the children’s exposure to beverages while at Head Start.
(Appendix F)

The beverage preference data was collected using a method developed, validated
and used extensively by LL. Birch (Birch et al., 1980). This method begins with the
self-selection of one of several food items by the children, continues with the tasting and
identiﬁcation of the item by the child and concludes with a ranking exercise. The
preference data collection was completed in a private area outside of the classroom with
the researchers. Children used three cartoon faces to assist in communicating their

beverage preferences. Prior to beginning the experiment, the researchers reviewed the

34

meanings of the cartoon faces (like, dislike and neutral) with each child individually,
using phrases the child could comprehend.

In this study, the children were asked to identify and taste the seven beverages in
a designated sequence: 100% apple juice, Kool-Aid Strawberry-Kiwi drink, Hi- C
Boppin’ Berry Punch, Caffeine-free Coca-cola, bottled water, 2% white milk and 1%
chocolate milk. Previous studies using this method vary the number of food items tested
ﬁom ﬁve to eight. These beverages were selected since they are often included in
children’s diets and because their long term use or lack of use can affect the physical
growth of the child and nutrient intake. Each child was presented with a tray containing
the seven beverages in small, clear plastic tasting cups. The child was asked to identify
the drink and then taste it. Since dealing with a young population, several words or
phrases were accepted as the identiﬁcation. For example, Kool-Aid and Hi-C could be
called fruit juice, juice drink or simply juice. Apple juice could also be identiﬁed by the
child as juice and it was accepted. For soda the following words and phrases were
accepted: Coke, soda, pop and cola. Water, milk and chocolate milk were the only terms
accepted to identify those products.

For children who were unable to identify a beverage, the researcher assisted the
child by prompting with descriptive words and phrases. For example, “What does this
drink look like?” or “What color is this drink?” If the child was still unable to identify a
beverage, the researcher told the child the name of the beverage and continued with the
tasting portion of the data collection. If a child was not willing to taste a particular
beverage, the researchers noted this and proceeded with the next beverage. The child’s

ability to identify a beverage as well as any comments (i.e. color, preference, texture,

35

taste) made by the children during the tasting session were recorded on the data collection
sheet (Appendix G).

After identiﬁcation and tasting of the beverage, the children were then asked to
point to the cartoon face that showed how they felt when they drank this beverage (like,
dislike or neutral). The experimenter used age appropriate words and phrases, in order to
help decrease any misunderstandings by the preschool children. For example, “How does
this drink make you feel?” and “What is this drink called?” The procedure of
identiﬁcation, tasting and preliminary ranking continued until all seven beverages had
been tested and grouped.

At this point, the child was asked to rank the drinks they placed in the “like”
category. The researcher then asked “Which of these is your favorite?” or “Which do
you like the very best?” This was recorded and the drink was removed from the child’s
sight. Then they were asked which of the beverages they like second best and continued
on throughout the remainder of the beverages and neutral and dislike groups, until a
preference ranking was established. The use of two steps in the ranking procedure
simpliﬁed the task for the children. In addition, the use of three cartoon faces, as
opposed to a more traditional ﬁve point Lickert scale, helped to simplify the task for the
children, while still allowing for the collection of more speciﬁc data. This method of
collecting children’s preference data has been successful and not invasive, showing no
risk or harm to the children, in previous studies with varied types of food items (Birch et

al., 1982; Birch et al., 1984; Birch and Sullivan, 1991 and Newmann and Taylor, 1992).

36

 

Table 3.1. Variable coding used in statistical analysis.

 

 

 

 

 

 

 

Category Code

Age Numerical value in months

Sex Male = 1
Female = 2

Ethnicity Caucasian = 1 Asian = 4
Hispanic = 2 Biracial = 5
Black = 3

Ethnicity Grouping Caucasian = 1 Not Caucasian = 2
Hispanic = 1 Not Hispanic = 2
Black = 1 Not Black = 2
Asian = 1 Not Asian = 2
Biracial = 1 Not Biracial = 2

Birth weight, Weight Numerical value in kg

Height Numerical value in cm

 

Weight for Age, Height for Age,
Weight for Height, Birth weight

> or = 95th percentile = 1

= 90 or 90-95th percentile = 2

= 75th or 75-90th percentile = 3
=50th or 50-75th percentile = 4

=25th or 25-50th percentile = 5

= 10th or 10-25th percentile = 6
=5th or 5-10th percentile = 7

<5th percentile = 8

 

Groupings for Wt/Age, Ht/Age, Wt/Ht

> or = 90th percentile = l
=10th or10-90th percentile = 2
< 10th percentile = 3

 

 

 

 

 

 

 

 

Beverage Availability in Home during past Yes = 1 N0 = 2

Three Months

Child’s Ability to Identify the Beverage Yes = 1 N0 = 2

Child’s Preference Ranking Most favorite = 1 Least favorite = 2
Grouping of Child’s Preference Ranking Top 3 of rank = 1 Bottom 4 of rank = 2
Parent’s Preference Ranking Most favorite = 1 Least favorite = 2
Grouping of Parent’s Preference Ranking Top 3 of rank = 1 Bottom 3 of rank = 2
Parent’s Idea of Nutritive Value of Beverage Most healthy = 1 Least healthy = 2
Grouping of Parent’s Idea of Nutritive Value Top 3 of rank = 1 Bottom 3 of rank = 2

of Beverage

 

 

37

 

Statistical Analysis

All data were analyzed using the Statistical Package for the Social Sciences
(SPSS) version 6.1.4, copyright 1995. Table 3.1 is a summary of how the variables were
coded for the statistical analysis. Basic student’s t (t-test) and one way analysis of
variance (AN OVA) tests, as well as the Tukey’s h post hoc test, were performed to

correlate variables. Signiﬁcance was determined by a p value less than 0.05.

Null Hypothesis Statements

The ﬁrst hypothesis (H 1) tested in this study states that there is no relationship
between the ﬂuid beverage preferences of a Head Start preschool sample and their most
recent weight for height and height for age data. The second null hypothesis (H2) tested
states that the ﬂuid beverage preferences of a preschool Head Start sample are not
indicative of their parents’ beverage preferences and nutritional beliefs. The objectives
for the hypotheses are stated below:

For the ﬁrst hypothesis (H 1):

H1 .. To determine any correlation existing between weight for height and
height for age data and the beverage preferences of the preschool
children.

H lb. To determine any correlation existing between weight for height and

height for age data and the beverages most often offered to the child as per

parental reporting.

For the second hypothesis (H2):

38

H23. To determine the relationship between the children’s ﬂuid beverage
preferences and the beverages offered to them.

H21, To examine relationships between the beverages offered to the children at
home and the belief of the parents as to their relative nutritive value.

H2c- To examine the relationship between beverages offered at home and

the beverage preference of the parents.

39

CHAPTER IV

RESULTS

Demographic Data

Fifty preschool age children enrolled in a Head Start program in a mid-westem
urban area were examined for their beverage preferences and exposure to beverages at
home. Weight and height data were collected for each child in the study. Of the children
in the sample, 54% were male and the average age of the children was 46 months. Table
4.1 lists the demographic characteristics of the sample. The children were from twelve
different classrooms participating in the same Head Start program. The children and
caregivers were enrolled in the study based upon the willingness of the caregiver to Sign
the consent form for the child to participate and complete the survey. In addition, the
children could not have any allergies to the beverages consumed in the experiment. Only
two children had food allergies as reported by the primary caregiver. These allergies
were to corn and oats; therefore, they had no effect on the ability of the children to taste
the beverages. No child was reported to have any choking or swallowing problems, nor
were there any reasons indicated by caregivers why their children could not taste a
beverage.

The ethnic proﬁle of the sample includes Caucasian (34%), Hispanic (14%),
Black (42%), Asian (4%) and children who are biracial (6%). In addition to different
backgrounds, ten children were born prematurely with gestation being on average 4.7
weeks shorter that the children born at term. This information was taken ﬁom Head Start

records. Prematurity was deﬁned as birth prior to 37 weeks of gestation.

40

 

Table 4.1 Demographic data collected from the caregiver survey and Head Start records.

 

 

 

 

Category Variables
Sex (11 = 50) Male 54% (n=27) Female 46% (n=23)
Age in months (n = 50) Mean = 46.0 (min. = 33.0; max. = 57.0)
Ethnicity (n = 50) Caucasian 34% (n=1 7)
Hispanic 14% (n=7)
Black 42% (n=21)
Asian 4% (n=2)
Biracial 6% (n=3)
Premature at birth (n = 50) Yes 20% (n=10) No 80% (n=40)

 

Number of weeks premature (n = 10)

Mean = 4.7 (min. = 1.0; max. = 12.0)

 

 

 

Have Food Allergies (n = 50) Yes 4% (n=2) No 96% (n=48)
Number of children in Mean = 2.46 (min. = 1.0; max. = 5.0)
household (n = 50)

Primary caregiver (n = 50) Male head of household 0% (n=0)

Female head of household 46% (n=23)
Both 36% (n=18) Other 18% (n=9)

 

 

Food shopper (n = 50) Male head of household 4% (n=2)
Female head of household 78% (n=39)
Both 0% (n=0) Other 18% (n=9)

Meal planner (n = 50) Male head of household 4% (n=2)

Female head of household 76% (n=38)
Both 20% (n=10) Other 0% (n=0)

 

Participate in food assistance
programs (11 = 50)

 

Yes 70% (n=35) No 30% (n=15)

 

41

 

The average number of children in a participating household was 2.46.
Additional household information obtained included the roles of the meal planner and
food shopper - 76% of the meal planner and 78% of the food shopper roles were ﬁlled by
the female. head of household. Only two households indicated the male head of
household as the primary meal planner and food shopper, while others listed both the
male and female sharing the roles, or another individual ﬁlling the responsibility, i.e. a
grandparent.

Many of the families participating in the Head Start program are often eligible for
additional supplemental food programs, such as Women, Infants and Children (WIC) and
Food Stamps. Seventy percent (35 households) were participating in an additional food

assistance program at the time of the study.

Anthropometric Data

All weight and height data were reported to the researcher by the Head Start staff
from past records. The weight and height data were collected by the Head Start staff
throughout the year. Birth weight was also provided by Head Start. Table 4.2 displays
the mean birth weight, weight and height data for the children, as well as the distribution
of the children according to the percentiles from the 1979 version of the NCHS growth
charts. An updated version of the NCHS growth charts which includes the body mass
index (BMI) calculation in assessing a child’s growth status, became available in late
2000 but was not available at the time the data was analyzed.

The mean birth weight was 3.23 kg and almost one quarter (24%) of the children

were at the 500' or between the 50th and 75th percentiles at birth. The average weight of

42

 

Table 4.2 Anthropometric data at birth and at time of the study.

 

 

Category Variable
Birth weight in kg (n = 50) Mean = 3.23 kg (min. = 1.0; max. = 4.20)
Birth weight (n = 50) > or = 95”1 percentile 6% (n=3)

= 90 or 90-95’h percentile 18% (n=9)

 

= 75th or 75-90th percentile 14% (n=7)
= 50th or 50-75th percentile 24% (n=12)
=25th or 25-50th percentile 8% (n=4)
= 10th or 10-25th percentile 20% (n=10)
=5th or 5-10th percentile 6% (n=3)
<5th percentile 4% (n=2)
Weight in kg (n = 50) Mean = 16.716 (min. = 11.0; max. = 24.50)

 

Weight/Age (n = 50) > or = 95th percentile 12% (n=6)
= 90 or 90-95th percentile 10% (n=5)
= 75th or 75-90th percentile 28% (n=14)
=50th or 50-75th percentile 14% (n=7)
=25th or 25-50th percentile 16% (n=8)
= 10th or 10-25th percentile 8% (n=4)
=5th or 5-10th percentile 2% (n=1)
<5th percentile 10% (n=5)

 

Height in cm (11 = 50)

Mean = 99.01

(min. = 73.8; max. = 110.0)

 

Height/Age (n = 50)

> or = 95th percentile 4% (n=2)
= 90 or 90.95th percentile 2% (n=1)
= 75th or 75-90th percentile 12% (n=6)

=50th or 50-75th percentile
=25th or 25-50th percentile

30% (n=15)
24% (n=12)

 

= 10th or 10-25th percentile 8% (n=4)
=5th or 5-10th percentile 6% (n=3)
<5th percentile 14% (n=7)
Weight/Height (n = 49)* > or = 95th percentile 12% (n=6)
= 90 or 90.95th percentile 16% (n=8)
= 75th or 75-90th percentile 28% (n=14)
=50th or 50—75th percentile 16% (n=8)
=25th or 25-50th percentile 12% (n=6)
= 10th or 10-25th percentile 12% (n=6)
=5th or 5-10th percentile 2% (n=1)

 

* One child's value was off growth chart.

 

43

 

the children was 16.716 kg with children spanning the growth chart. Twenty-eight
percent of children were at the 75th or between the 75th and 90th percentile. Ten percent
of the children (n=5) were below the 5th percentile for weight for age.

In looking at stature, height for age growth chart percentiles were used. The mean
height for the children was 99.01 cm. Twelve, thirty and twenty four percent of the
children were at the 75th or between the 75th and 90th percentile, at the 50th or between the
50th and 75th percentile, or at the 25th or between the 25th and 50th percentile, respectively.
Seven (14%) of the children were below the 5th percentile, indicating short stature or
growth stunting.

One can determine appropriate weight for age and can indicate incidence of over-
or underweight using the weight for age percentile. Only data from 49 children were
used, as one child’s height was below the lower value on the growth chart. Therefore,
this information could not be plotted on the readable area of the growth chart. Twenty
eight percent of the children were at the 75th or between the 75th and 90th percentile. The
remainder of the children were distributed throughout the weight for age growth chart
with twelve percent being overweight (greater than the 95th percentile) and no children

being below the 5“1 percentile.

Beverage Collection and Caregiver Questionnaire

F orty-four (88%) of the ﬁfty total children participating in the study tasted all of
the beverages offered. Two children were absent on the day of the testing and one child
refused to taste any of the beverages but did go through the ranking exercise. Three other

children did not taste all of the seven beverages. Two children tasted all but two and one

child tasted all but three. The tasting exercise took place at least one hour after the
children had a meal or snack. This was done to avoid confusion due to prior food intake
or consumption of a beverage. Overall, the children were cooperative and the researchers
kept the testing moving along, so as not to lose the attention of the children. Each
beverage exercise took approximately ﬁve to ten minutes. This included setting up the
beverages, explaining the exercise to the children and conducting the tasting exercise.

In addition, all tasting exercises were performed in an area outside of the classroom, so
that other children were not around to disrupt the tasting exercise. Due to the young age
of the children and the classroom setting, it was assumed that they would not share
information regarding the exercise so as to inﬂuence children who had not yet performed
the exercise. This would be a concern when dealing with older children. The children
made additional comments of note during the testing sessions. Many children were easily
able to identify soda as “pop”, as well as “Coke.” Chocolate milk was identiﬁed as
“coffee” by one Hispanic child. Most children easily identiﬁed the fruit drinks and apple
juice as such. It would be difﬁcult to distinguish between different brands of juices, such
as Kool-Aid and Hi-C, without showing the child a commercial or label to assist them in
distinguishing between the two.

All Head Start classroom sites served milk with either the breakfast and lunch
sessions or the lunch and snack sessions, depending on the time of day the classes met.
One site also served orange juice in addition to milk. All beverage preference test
sessions were conducted at least one hour after the child had a meal, either at Head Start

or home.

45

Caregiver questionnaire information was helpful in learning the beverages most
often found in the home, as well as the beverage preferences and nutritional beliefs of the
caregivers. However, the last page of the questionnaire, a simpliﬁed food frequency
questionnaire, was completed incorrectly by most caregivers even though it was
explained to them and the researcher was available to answer questions about the
completion of the questionnaire. As a result, the data from this section were not used in
this study. In addition, two questions regarding beverages actually offered to children at
home and beverages the child consumes while in the caregiver’s care were often not
completed. This poor response eliminates an attempt to gather information about the
child’s beverage consumption at home.

Thirty -two percent of the primary caregivers ranked soda as their least favorite
when asked to rank six beverages: soda, water, rrrilk, apple juice, chocolate milk and juice
drinks. Forty-eight percent stated that milk was either their ﬁrst or second favorite of the
same six beverages. When asked about the “healthiness” of the beverages, 80% of the
caregivers ranked soda or cola last, while 84% ranked water in the top three and 86%

ranked rrrilk ﬁrst or second.

Hypothesis 1 (H1):
H1: There is no relationship between the ﬂuid beverage preferences of a

Head Start preschool population and their most recent weight for height and

height for age data.

46

For this hypothesis the independent variables included the preference or ranking
of the beverages by the children, the rankings grouped into the top three favorites, the
bottom four preferences and the child’s ability to identify the beverage. In addition, the
availability of the beverages at home, taken from the parent survey, was an independent
variable. The dependent variables included all weight and height information available on
the children participating (weight, height, weight for age, height for age and weight for
height). T-tests and analysis of variance (ANOVA) were used due to the small sample
size. Total sample size (n) varied at times during the study, as some caregiver survey
questions were not completed correctly and two children never performed the beverage
preference exercise due to absences on the day of the exercise.

The mean body weight of those children who ranked water as a top three
preference was signiﬁcantly lower (p=0.027) than the weight of those who gave it a
lower overall rank (Table 4.3). Table 4.3 also shows that children who were able to
identify Hi-C beverage were lighter (mean weight of 16.5692 kilograms) than children
who were not able to identify it. No other signiﬁcant differences were found in body

weight relative to preference of beverage or ability to identify a beverage.

Table 4.3 Factors contributing to signiﬁcant differences

 

 

between weight.

N Mean * P Value

Rank of Water P=0.027
Top 3 ranking 14 16.0786
Bottom 4 ranking 32 17.0844

Able to ID Hi-C P=0.002
Yes 13 16.5692
No 34 16.9000

 

 

 

 

* Values represent weight in kilograms.

47

Differences were also present between preference ranking, ability to identify
beverages and the availability of beverages at home with respect to the weight for age
percentile on the NCHS growth charts. Table 4.4 and 4.5 show the differences that were
signiﬁcant. Children who gave Hi-C a higher preference rank, as well as those able to
identify it, had a lower mean weight for age (when numerical values are assigned to the
percentiles) than their counterparts with p values of 0.035 and 0.040, respectively.
Caregivers who indicated that there is bottled water at home had children with a lower
mean weight for age. These children were lighter than those whose caregivers did not
indicate bottled water was in the home. The households with low fat milk (1/2%,1% and
2%) milk had children with a higher overall weight for age percentile when grouped, as

seen in Table 4.5. This was signiﬁcant with a p value of 0.029.

Table 4.4 Signiﬁcant differences between mean values of

 

 

 

weight for age.
N Mean“ p Value
Rank of Hi-C P=0.035
Top 3 ranking 24 4.2917
Bottom 4 ranking 22 3.8182
Able to ID Hi-C =0.040
Yes 13 4.6154
No 34 3.7941
Bottled water at home P=0.022
Yes 18 1.9444
No 32 1.875

 

 

 

 

* Values for weight for age are deﬁned as follows:
1 = >95th %tile; 2 = 90.95“1 %tile; 3 = 75-90th %tile;
4 = 50.75th %tile; 5 = 25.50th %tile; 6 = 10-25th %tile;
7 = 50.40th %tile; 8 = <5‘h %tile

48

Table 4.5 Signiﬁcant differences between means of weight for

 

age grouped.
N Mean * P Value
Low fat milk at home P=0.029
Yes 28 1.7857
No 22 2.0455

 

 

 

 

* Values for weight for age are deﬁned as follows:
1 = >90th %tile; 2 = 10.90th %tile; 3 = <10th %tile

Table 4.6 Signiﬁcant differences between means of height for

 

 

 

age grouped.
N Mean" P Value
Rank of Chocolate P=0.0101
Milk

Ranked it #1 17 2.3529
Ranked it #3 6 1.6667

Able to ID Hi-C P=0.023
Yes 13 2.3077
No 34 2.0882

Low fat milk at home P=0.025
Yes 28 2.0000
No 22 2.3182

 

 

 

 

* Values for height for age are deﬁned as follows:
1 = >90th %tile; 2 = 10.90th %tile; 3 = <10th %tile

Signiﬁcant differences indicated in means of height for age were with the ranking

of chocolate milk, ability to identify Hi-C and the availability of low fat milk at home.

Children who ranked chocolate milk as the highest preference were signiﬁcantly shorter

than those children who ranked it third. The means for the numerical values assigned to

the percentiles were 2.3529 and 1.6667 for those ranking chocolate milk ﬁrst and third,

respectively (Table 4.6). There were no other signiﬁcant differences between height for

age percentiles relative to beverage preference rank or ability to identify a beverage.

49

Children able to identify Hi-C were also found to be shorter than their counterparts
unable to identify Hi-C, p = 0.023 (Table 4.6). Caregivers who indicated that low fat
milk is present in the home had children who were taller than those who did not have low
fat milk at home, p = 0.025 (Table 4.6).

In looking at a child’s weight for height value, one can see whether or not the
child may be over or underweight. Several differences in the mean weight for height
values were found in children with regards to beverage preference and availability of
certain beverages at home. Table 4.7 and 4.8 proﬁle this data. Mean values for weight
for height in looking at all percentiles individually (numerical values assigned to each
percentile to calculate the mean) for children with and without bottled water at home was
3.6667 and 3.3226, respectively (p=0.029). Therefore children with bottled water at
home were smaller overall. When grouping the percentiles, the same conclusion results
but with slightly more signiﬁcance, p=0.000.

Children ranking soda and water in the top three show different results with
regards to weight for height. Those ranking soda in the top 3 were heavier than their
counterparts who gave soda a lower ranking. Mean values were 1.7667 for a top three
rank of soda and 1.8667 for a lower rank of soda. The opposite result occurs for children
who ranked water in the top three, mean was 1.9286, versus children who did not rank
water high, mean 1.7419. Children who ranked water higher had a smaller weight for
height overall.

The presence of low fat milk at home versus no low fat milk at home has means
of 1.6786 and 1.8095, respectively; this indicates that the‘children with low fat milk at

home were larger than their counterparts. Reverse results were shown with children

50

having non-cola sodas (lemon-lime, orange ﬂavored, gingerale, etc) in the home. These
children were smaller than their counterparts without non-cola sodas in the home. Table

4.7 and 4.8 display these results.

Table 4.7 Signiﬁcant differences between means of weight for
height relative to the availability of a beverage at home.

N Mean * P Value

 

Bottled water at home P=0.029
Yes 1 8 3 .6667
No 3 1 3 .3226

 

 

 

 

"‘ Values for weight for height are deﬁned as follows:
1 = >95th %tile; 2 = 90.95th %tile; 3 = 75.90th %tile;
4 = 50-75th %tile; 5 = 25-50th %tile; 6 = 10-25‘h %tile;
7 = 5.10th %tile; 8 = <5th %tile

Table 4.8 Signiﬁcant differences between means of weight for
height grouped relative to beverage preference and availability.

N Mean * P Value

 

 

 

 

 

Rank of Soda P=0.048
Top 3 ranking 30 1.7667
Bottom 4 ranking 15 1.8667

Rank of Water P=0.001
Top 3 ranking 14 1.9286
Bottom 4 ranking 32 1.7419

Bottled Water at home P=0.000
Yes 18 1.8889 '
No 3 1 1.6452

Low fat milk at home P=0.015
Yes 28 1.6786
No 21 1.8095

Non-Cola Soda at home P=0.033
Yes 20 1.8000
No 29 1.6897

 

 

 

 

* Values for weight for height are deﬁned as follows:
1 = >90th %tile; 2 = 10.90th %tile; 3 = <10th %tile

51

Hypothesis 2 (H2):
H2: The ﬂuid beverage preferences of a preschool Head Start population are

not indicative of their parents’ beverage preferences and nutritional beliefs.

The second null hypothesis relies on information collected from the parent
surveys, the ability of the children to identify beverages during the preference testing and
their actual preference data. The children’s ability to identify two beverages correlated
with the fact that those beverages are more often present in the home. One beverage,
however, showed the opposite result (Table 4.9). Children were more able to identify
white milk if low fat milk was present in the home. (p=0.023). They were also more
likely to identify soda if the had non-cola sodas (such as orange soda, lemon—lime,
gingerale, etc) in the home as opposed to those who did not have these beverages in the
home with mean values of 1.0000 and 1.1111, respectively. The opposite result was seen
for water. Children with bottled water in the home had a higher mean value for the
ability to identify water and therefore were less able to identify it correctly.

Signiﬁcant differences existed between the mean beverage preference rank for
children who had chocolate milk at home and those that did not have the beverage at
home. (Table 4.10). Children with chocolate milk at home gave chocolate milk a mean
rank of 2.4000, while those who typically do not have chocolate milk in the home had a
mean rank of 3.6452; therefore, an opposite result was seen for children who had non
cola soda in the home. These children ranked regular soda lower overall (mean 3.2632),

when their counterparts without “other sodas” in the home gave it a mean rank of 2.5 1 85.

52

There were no signiﬁcant differences with regard to having a beverage at home
and the parent believing it is a healthy beverage. Finally, Table 4.11 shows results when
looking at the parent ranking of several beverages and whether or not they are in the
home. The ﬁrst result stated in Table 4.11, displays the parent’s mean rank of chocolate
milk, 1.4762 and 1.2400, for those who have whole milk or do not have whole milk in the
home, respectively (p=0.010). For both white milk and soda, the mean rank was higher
for those that have milk or diet caffeine free soda in the home. While there are
differences, neither of the mean rankings indicated that white milk or soda were in the top

three beverages most of the time.

Table 4.9 Signiﬁcant differences between mean ability of
children to identify beverages relative to beverage availability.

N Mean“ p Value

 

 

 

ID White milk
Low fat milk at home p=0.023

Yes 26 1.0000
No 21 1.0476

ID Soda

Non-Cola Sodas at home p=0.001
Yes 20 1.0000
No 27 1.1 11 1

ID Water

Bottled water at home p=0.009
Yes 18 1.0566
No 29 1.0000

 

 

 

 

*1= Able to ID beverage; 2= Unable to ID beverage

53

Table 4.10 Signiﬁcant differences between means of children’s
ranking of beverages relative to the availability of
beverages at home.

N Mean" p Value

 

 

Rank of Chocolate milk

Chocolate milk at home p=0.003
Yes 15 2.4000
No 31 3.6452

Rank of Soda

Non-cola sodas at home p=0.001
Yes 19 3.2632
No 27 2.5185

 

 

 

 

* Rank of beverages: 1= most favorite; 7= least favorite

Table 4.11 Signiﬁcant differences between means of parent’s
ranking of beverages relative to the availability of
beverages at home.

N value Mean“ p Value

 

Parent rank of Chocolate
milk
Whole milk at home p=0.010
Yes 21 1.4762
No 25 1.2400

 

Parent rank of Milk

Skim milk at home p=0.026
Yes 29 1.9310
No 16 2.0000

 

Parent rank of Soda

Diet caffeine free soda at p=0.011
home
Yes 18 1.9444

NO 28 2.0000

 

 

 

 

 

*Rank of beverages: 1 = Top 3 favorites; 2 = Least favorites

54

Additional Results

In addition to looking at results directly related to the null hypotheses, much of
the data were compared with respect to sex and ethnicity to indicate differences in the
means of this sample. The results are listed in Tables 4.12 - 4.15. Most notable is the
male mean ability to identify beverages. This was lower; therefore, more males
identiﬁed the beverages correctly than their female counterparts. This was true for Hi-C,
Kool—Aid, white milk, soda and water. All males correctly identiﬁed white milk, soda
and water.

Several anthropometric differences also existed with respect to race, particularly
whether or not a child was Black or Caucasian. Black children had a higher mean value
for height for age and were therefore shorter when coding was interpreted than children
who were not Black. A higher mean value was true for weight for age among children
that were Black as opposed to all others, indicating that Black children weighed less for
their age as opposed to their non-Black counterparts.

Caucasian children had a higher birth weight overall than their non Caucasian
counterparts, mean birth weight values of 3.4765 kilograms versus 3.1000 kilograms. In
addition, their value for birth weight for age was smaller indicating (when coding was
translated) that they were larger for their age at birth overall (p=0.030).

Tables 4.14 and 4.15 display mean differences in ranking of beverages and the
ability to identify beverages correctly among ethnic groupings. Both apple juice and soda
had an overall higher mean rank among Black children when compared to their non-
Black counterparts. The opposite result was seen for Kool-Aid. Apple juice was given

higher preference among children who were not Caucasian, as opposed to those who

55

were Caucasian. The mean ability to identify water and chocolate milk was higher (less
able to identify it correctly) among Black children, while the ability to identify soda was
lower (more able to identify it) among Caucasian children.

No data were signiﬁcant with respect to whether or not the children were Asian,
biracial or Hispanic. The sample sizes from each of those ethnic groups was too small to

determine an accurate value.

56

Table 4.12 Signiﬁcant differences with ability to identify beverages
with respect to sex.

 

 

 

 

 

N Mean * p Value
Ability to ID Hi-C
Sex p=0.018
Male 26 1.6538
Female 21 1 .8095
Ability to ID Kool-Aid
Sex p=0.000
Male 26 1.6538
Female 21 1.9048
Ability to ID white milk
Sex p=0.023
Male 26 1.0000
Female 21 1.0476
Ability to ID soda
Sex p=0.000
Male 26 1.0000
Female 21 1.1429
Ability to ID water
Sex p=0.023
Male 26 1.0000
Female 21 1.0476

 

 

 

 

* 1 = Able to ID beverage; 2 = Unable to ID beverage.

57

Table 4.13 Signiﬁcant mean anthropometric differences with

 

 

 

 

respect to ethnicity.
N Mean * p Value
Height for Age
Ethnicity p=0.006
Black 21 2.2857
Not Black 29 2.0345
Weight for Age
Ethnicity p=0.003
Black 21 1.9048
Not Black 29 1.6071
N Mean" p Value
Birth weight
Ethnicity p=0.045
Caucasian 17 3.4765
Not Caucasian 33 3.1000
N Mean *** p Value
Birth weight for age »
Ethnicity p=0.030
Caucasian 17 3 .41 18
Not Caucasian 33 4.5152

 

 

 

* Values for weight for age and height for age are deﬁned as
follows: 1 = >90th %tile; 2 = 10.90th %tile; 3 = <10th %tile
*"' Values in kilograms.
*** Values for weight for height are deﬁned as follows:
1 = >95th %tile; 2 = 90-95th %tile; 3 = 75.90th %tile;
4 = 50.75th %tile; 5 = 25-50‘h %tile; 6 = 10-25th %tile;
7 = 50.10th %tile; 8 = <5th %tile

58

 

Table 4.14 Signiﬁcant differences between means for ranking
beverages with respect to ethnicity.

 

 

 

 

N Mean * p Value
Rank of Apple juice
Ethnicity p=0.000
Black 21 1.5238
Not Black 25 1.8400
Rank of Kool-Aid
Ethnicity p=0.001
Black 21 1.8571
Not Black 25 1.6400
Rank of Soda
Ethnicity p=0.002
Black 21 1.4762
Not Black 25 1.2000
Rank of Apple juice
Ethnicity p=0.020
Caucasian 1 5 1 .8000
Not Caucasian 31 1.6452

 

 

 

 

* Rank of beverages: 1 = Top 3 favorites; 2 = Least favorites

Table 4.15 Signiﬁcant differences between ability to identify
beverages with respect to ethnicity.

 

 

 

N Mean * p Value
Able to ID chocolate
milk
Ethnicity p=0.022
Black 21 1.1905
Not Black 26 1.0769
Able to ID water
Ethnicity p=0.023
Black 21 1.0476
Not Black 26 1.0000
Able to ID soda -
Ethnicity p=0.009
Caucasian 15 1 .0000
Not Caucasian 32 1.0938

 

 

 

 

" l= Able to ID beverage; 2 = Unable to II) beverage

59

Summary of Null Hypotheses

Both null hypotheses are rejected in this study. The ﬁrst null hypothesis stated
that there is no relationship between beverage preferences and a child’s growth status.
The two objectives for this null hypothesis were to determine correlation between weight
for height and height for age data and the beverage preferences of the children; also to
determine correlation between the previously stated growth data and beverages most
often offered to the children participating. Signiﬁcant results for each objective exist and
therefore the ﬁrst null hypothesis can be rejected. Further studies are needed to provide
more detailed information regarding such relationships.

The second null hypothesis states that beverage preferences of the children are not
related to their parents’ beverage preferences and parents’ nutritional beliefs. The three
objectives for this hypothesis were to determine relationships between children’s
preferences and the beverages offered to them, to determine correlation between
beverages offered to the children at home and their parents’ nutrition beliefs about the
beverages and ﬁnally to examine the relationship of the beverages offered at home and
the beverage preferences of the parents. Two of the three objectives were completed
successfully. Signiﬁcant values indicate that there is a relationship between the
beverages offered at home and the preferences of both the parents and children. There
was however, no signiﬁcant data found to support a relationship with beverages offered
at home and the parents’ nutritional views regarding beverages. Therefore only part of
the second null hypothesis can be rejected. Again further studies are needed to validate

this study as well as provide more in depth examination of such relationships.

60

CHAPTER V

DISCUSSION

General Information

This study examined relationships between beverage preferences in preschool
children and their speciﬁc anthropometric data, as well as parent preferences and
nutritional beliefs about commonly consumed beverages. Previous studies have shown
relationships between food preferences, consumption, growth status and familial
inﬂuences on young children (Birch, 1979a; Birch 1980b; Birch and Billrnan, 1986a;
Colavito et al., 1996; Skinner et al., 1998 and Smith and Lifshitz, 1994). Little has been
done in the area of beverage preferences and consumption. This study was designed to
provide general information regarding beverage preferences and their relationship to
anthropometric data and availability of beverages at home.

Fifty children and their primary caregivers participating in an area Head Start
program served as the sample. The participants were selected randomly on the basis of
caregiver attendance at the Head Start parent meetings. The distribution of male and
female child subjects was almost even. No children participating had any allergies or
intolerances to the beverages served. Also no children had any swallowing or choking
problems, as indicated by the primary caregivers. Therefore, all children were eligible to
taste every beverage, although this was not the case due to absences and children refusing
to taste beverages. The children participated in the entire activity, with few exceptions.

The researcher performing the preference exercise had assistance to ensure that the

61

exercise moved along and that the child’s attention was kept throughout. Approximately
ten to ﬁfteen rrrinutes was spent with each child to conduct the exercise.

Over one-third of the children were Caucasian and over one-half were Hispanic or
Black. This exceeds the estimate that one third of the child population in 2000 would be
Latino or Black. (Crockett and Sims, 1995) A larger sample or a sample encompassing
several different urban and rural areas may indicate otherwise. In addition, there were a
few children that were Asian or biracial. Future studies would need to collect data from a
larger sample size in order to examine speciﬁc relationships between beverage
preferences and the cultural background of the participants.

In the majority of the households (over 75%), the female ﬁlls the traditional
“role” of meal planning and food shopping with occasional assistance from the male
counterpart or other household members. The primary caregiver is often female or
caregiver duties are shared by the male and female in the household. The socioeconomic
status of the families is consistent with the qualiﬁcations for the Head Start program. In
addition, a majority of the families participate in Food Stamps, WIC or another food
assistance program. The children and families participating in these federal programs
beneﬁt from the services provided. The focus on this subset of the population does assist
with our understanding of factors contributing to the growth and development of low
socioeconomic status children.

Eighty-eight percent of children tasted all of the beverages provided. Several
children did not taste one or two of the beverages and some children were unable to
perform the tasting exercise due to absence. All of the head start sites served milk

exclusively with snacks and meals provided to the children. One site also offered juice to

62

accompany a snack. Exposure to milk at Head Start daily is essential for children to
become familiar with the beverage and develop a taste for it. Whether milk is served at
home is the discretion of the caregivers and often the effects of their own beverage
preferences and nutritional beliefs. One Hispanic child identiﬁed chocolate milk as
“coffee.” Coffee is often consumed in Hispanic countries and when consumed it has
added milk. His identiﬁcation of chocolate milk as “coffee” indicates his awareness of
beverages at home.

Results from this study showed that most caregivers feel that water and milk are
the healthiest beverages to drink for themselves and their children. Soda or cola was
indicated as nutritionally inferior by most caregivers. It is unknown whether or not the
parent/caregiver provided information that they thought the researcher would want to
hear compared to accurate information. Such a variable is difﬁcult to identify and
control.

Weight and height data examined showed that most children were within
acceptable percentiles for their age. Almost one quarter (24%) of the children were
between 50-75th percentile at birth for weight for age and over one quarter (28%) of the
children were between the 75- 90th percentile at birth. Therefore over half of the children
were above the 50th percentile for weight for age. Only four percent of the children,
including two of the premature children, were below the 5m percentile for birth weight for
age. While this contradicts a common assumption that children from lower income
families are born at a lower birth weight than higher income counterparts, the sample size
in this study is small. A much larger sample size would be needed to further analyze that

relationship.

63

Stature is measured by height and can be analyzed looking at height alone or
height for age. Stature can also be predicted by looking at the heights of the child’s
parents. Those data were not collected in this study, but would be helpful to look at in
further studies with a speciﬁc focus on stature. Fourteen percent of the children were
considered “short-stature,” below the 5th percentile for height for age.

In looking at the weight for height data, one can determine if a child is deﬁned as
over or underweight. There were no children classiﬁed as underweight according to the
data provided by Head Start. Twelve percent of the children were plotted above the 95th
percentile for weight for age. Factors contributing to increased weight in children can
include excess calorie intake, especially from non-nutrient dense foods, as well as lack of
physical activity and heredity. This list is not all-inclusive, as additional environmental

and physical factors can lead to obesity in both adults and children.

Hypothesis 1 (H1):
H1: There is no relationship between the ﬂuid beverage preferences of a
Head Start preschool population and their most recent weight for height and

height for age data.

This null hypothesis examines the relationship between beverage preferences and
anthropometric data of the preschool child sample. In several instances there was a
signiﬁcant relationship evident between the weight and height data and the child’s
beverage ranking and or ability to identify a beverage. Not all beverages showed

signiﬁcant results.

Children ranking water in the top three of the seven beverages weighed less than
their counterparts who did not provide water with a high preference ranking. These
children were lighter overall, but not underweight. The lack of calories in water and the
“healthy” connotation of water may indicate that these children have healthier eating
habits at home and that their parents drink water. It is unknown as to the type of bottled
water purchased by these households. Reasons for purchasing bottled water were also
not examined and would be helpful in future studies. The children’s eating habits were
also not examined in this study. In addition, children with bottled water at home had a
mean smaller weight for age. With regard to weight for height values, children who
ranked water higher and those who have bottled water at home were smaller than those
children who did not rank water high or have bottled water at home. Water is most often
not the drink of choice for children. Exposure to this beverage may increase its
popularity and consumption among younger children. No estimation of tap water intake
or preference was obtained.

Hi-C and other popular fruit drinks ﬁll the stores shelves enticing children and
parents to purchase and try them. These beverages provide most of their calories as sugar
and lack many of the vitamins and minerals that other juices contain. In this study
signiﬁcant relationships were found with children who were able to identify Hi-C as a
ﬁ'uit drink or juice beverage. Children who were able to identify it correctly had a lower
weight overall and a lower weight for age. Also the children ranking I-Ii-C in the top
three of the seven beverages were smaller when looking at weight for age. This brings
forth an interesting point in that usually juice and fruit drink consumption often leads to

excessive calorie intake and overweight children, but this study shows the opposite result.

65

This study does not identify the amount of the Hi-C or other ﬂuid drink beverages
consumed by the children, just their rank of it and ability to identify it. Additional
information would be needed to make an assumption related to excess consumption and
weight for height data. Children able to identify Hi-C were also shorter when looking at
height for age data. This may be related to a lack of milk consumption among these
children.

When examining milk speciﬁcally, children with low fat milk at home were taller
(height for age) and larger (weight for age and weight for height) than children from
families who did not have low fat milk at home. Additional calories in the low fat milk
versus the skim milk may contribute to such differences, but overall information
regarding food intakes are essential to make such an assumption. No signiﬁcant
relationship existed with whole milk and weight and height data. Children ranking
chocolate milk high were overall shorter than their counterparts giving chocolate milk a
lower preference ranking. There are no logical explanations for this relationship.

Soda consumption is increasing among older, school-age children. However,
soda is consmned by preschool age children (Harnack, 1999). Whether cola, lemon-
lime, ginger-ale or other type of soda, children are drinking it. Heavier children in this
study showed a higher preference for soda that those children weighing less. Children
without non—cola sodas, such as lemon lime, ginger-ale, grape and orange, at home were
larger than their counterparts with these beverages at home. More data is needed to

further examine this relationship.

66

Hypothesis 2 (Hz):
H2: The ﬂuid beverage preferences of a preschool Head Start population are

not indicative of their parents’ beverage preferences and nutritional beliefs.

This hypothesis is also proven incorrect due to several signiﬁcant differences in
the information provided by the children and caregivers. Children with low fat milk at
home were more often able to identify white milk. Children with chocolate milk at home
gave it a higher mean ranking overall in the preference exercise. Increased exposure to
these beverages at home increased the probability that the child is able to identify them
correctly. Parents with whole milk at home ranked chocolate milk lower than parents
who do not typically have whole milk at home. This suggests that white milk may be
consumed more often at home as opposed to chocolate milk. Finally, parents with skim
milk at home ranked milk higher overall. The “healthy” label often associated with milk,
especially skim milk, was evident. It is encouraging to know that parents are making wise
beverage choices and hopefully set a good example for their children.

Children in households with non-cola sodas at home more often identiﬁed regular
soda correctly, yet had a lower rank for soda. Possible use of lemon lime and orange
sodas, etc could lead to a decreased use of regular cola sodas and therefore a lower
ranking among children. More information about consumption at home would be needed
to identify further relationships. On the other hand, parents with diet caffeine free soda at
home, gave soda a higher ranking. This could be due to the fact that diet caffeine free is
lacking two elements that regular soda contains — caffeine and calories. A great deal of

advertising may go into making this beverage alternative quite appealing to adults.

67

Additional Information

Additional data was examined to determine any further relationships among
beverage preferences, weight and height data and the ethnicity or sex of the children
participating. All males were able to correctly identify Kool-Aid, white milk, soda, Hi-C
and water more often than their female counterparts. There were no signiﬁcant
differences for any of the weight and height data collected among the sexes. However,
there were signiﬁcant differences involving Black and Caucasian children. Black
children were shorter and weighed less for their age than their non-Black counterparts.
Caucasian children weighed more at birth than their non—Caucaisan counterparts. Mixed
results were shown with respect to different ethnicities and their beverage preferences
and abilities to identify various beverages. Larger sample sizes are needed to validate
these results.

While relationships are evident in this study, there are other various factors linked
with growth status in children. Likewise, there are other factors inﬂuencing parent
choices for purchasing beverages for their household. Previous research relates child
food intakes and eating habits and environmental inﬂuences, as indicated in the literature
review. This study does not provide concrete evidence of relationships, but only

identiﬁes possible ones.

68

CHAPTER VI

CONCLUSION

Study Strengths and Limitations

Strengths of this study include the fact that one researcher recorded all beverage
preference data and comments made by the children. In addition, little has been done in
the area of beverage preferences with a young population. The controlled environment of
Head Start was an ideal setting to conduct the caregiver surveys, obtain recent weight and
height data, as well as collect the children’s preference data.

Limitations of the study include the small sample size and the single preference
exercise for each child. Repeated preference exercises with the same children and
beverages would have provided possibly more accurate data and may have shown
changes in preferences over a period of time. Children who were tested earlier in the day
may have been tired, since they could not sleep in as late. This was not necessarily
observed or recorded in this study.

Beverage availability at home was not examined by food records or visits to the
home sites; instead caregiver reporting on the questionnaire was used to collect this
information. In this population, home visits may not be well received and other methods
of collection of exact beverage information may need to be used. Beverage intake away
from home, beside the Head Start program, was not collected.

Also, the last page of the caregiver questionnaire was meant to collect valuable
beverage frequency data, but due to subject errors, much of the data was incomplete and

could not be used. Finally, an in depth examination of the foods consumed, as well as the

69

amounts of beverages consumed by the children in a given period of time would have
provided additional dietary data relating to growth status. Overall, several relationships
between children’s and parents’ beverage preferences, growth status, as well as

availability of beverages in the home were suggested.

Implications for Practice and Education

This study suggests several unexpected signiﬁcant relationships between child
beverage preferences, their grth data and the availability of beverages at home. In
addition, correlation was found with regards to parents’ preferences and beverages found
in the home. While further research is needed to validate this information and provide
additional details, these preliminary results provide areas for health and nutrition
professionals to focus upon with regards to child beverage consumption. Knowing that
such relationships exist can provide avenues for teaching both children and their
parents/caregivers good nutrition habits, not only with beverages, but with foods as well.

Health professionals can use the results from this study to further educate parents
and caregivers, as well as children, on appropriate beverage intake and the effects on the
body. Speciﬁcally, they can focus on excess juice consumption, white milk as opposed

to chocolate milk consumption and soda as a source of added calories in children’s diets.

Implications for Future Research

While it may seem as if it is a logical relationship, there was no data from this
study to support that the beverages in the home are relative to the parents’ nutritional
beliefs of the tested beverages. Other reasons for having beverages in the home were not

fully examined, including ﬁnancial limitations, preferences or needs of other family

70

members beside the child tested and the parents/caregivers or even the ability to obtain
certain beverage items. Additional research can focus on these speciﬁc areas to more
closely examine the relationship.

This study provides ideas for further studies focusing on beverage intake and
preferences of young children. Some of the results show that beverages children drink at
home or at Head Start are related to growth status and preferences. It is important to
begin teaching children good eating and drinking habits at an early age. Young children
are perceptive at home and school and will quickly deve10p habits that can mimic those
of the adolescents and adults around them (Birch, 1980a; Crockett and Sims, 1995).
Soda and non-nutritive fruit drinks, such as Hi-C and Kool Aid, in excess amounts can
mask a child’s preference for water, milk, juice and other healthier beverages. The
beverages most often seen in the results in this study included water, milk, chocolate milk
soda and Hi-C. Few signiﬁcant ﬁndings included Kool-Aid or apple juice.

The use of z scOres, a measure of standard deviation, in analyzing the height and
weight data could be used in future studies. This would allow each child in the study to
be compared to a “healthy” population, as well as comparing the mean anthropometric
data of the study sample to the “healthy” population mean. This information would
provide the researcher with an idea of the overall growth status of the study sample.

Future, more focused hypotheses can be derived from the results of this study.
Speciﬁc areas can include, but are not limited to, an examination of children’s beverage
preferences as related to their actual beverage intake, child beverage preferences in
relation to overall calorie and food intake and relation to growth parameters. An

exarrrination of cultural backgrOunds and beverage availability and child preferences

71

would also provide additional data necessary for health professionals when counseling

multicultural populations.

72

APPENDIX A

UCRIHS Letter

73

 

OFFICE OF
RESEARCH
AND
GRADUATE
STUDIES

niversity Committee on
Research involving
Human Subjects
(UCRIHS)

Michigan State University
46 Administration Building
East Lansing. Michigan
48824-1046

517/355-2180
FAX: 517/353-2976

rne Michigan State University
'DEA is institutional Diversity;
Excellence in Action

MSU is an affirmative-action.
equal opportunity institution

MICHIGAN STATE

U N l V E R S l T Y
December 9, 1998

 

TO: Dr. Jenny Bond
2100 South Anthony Hall

APPROVAL DATE: December 9, 1998

RE: IRB # 98765 CATEGORY: l-C,E,G
TITLE: BEVERAGE PREFERENCES OF LOW-INCOME PRESCHOOL CHILDREN

The University Committee on Research Involving Human Subjects' (UCRIHS) review of this
project is complete and I am pleased to advise that the rights and welfare of the human subjects
appear to be adequately protected and methods to obtain informed consent are appropriate.
Therefore, the UCRIHS approved this project.

RENEWALS: UCRIHS approval is valid for one calendar year, beginning with the approval date
shown above. Projects continuing beyond one year must be renewed with the green renewal form.

A maximum of four such expedited renewals possible. Investigators wishing to continue a project
beyond that time need to submit it again for a complete review. - _

REVISIONS: UCRIHS must review any changes in procedures involving human subjects, prior to
initiation of the change. If this is dOne at the time of renewal, please use the green renewal form.

To revise an approved protocol at any other time during the year, send yom’ written request to the
UCRIHS Chair, requesting revised approval and referencing the project's IRB# and title. Include

in your request a description of the change and any revised instruments, consent forms or
advertisements that are applicable.

PROBLEMS/CHANGES: Should either of the following arise during the course of the work, notify
UCRIHS promptly: 1) problems (unexpected side effects, complaints, etc.) involving human
subjects or 2) changes in the research environment or new information indicating greater risk to

the human subjects than existed when the protocol was previously reviewed and approved.

If we can be of further assistance, please contact us at 517 355-2180 or via email:
UCRIHS@pilot.msu.edu. Please note that all UCRIHS forms and instruction are located
via the web: http://www.msu.edu/unit/vprgs/UCRIHS/

Sincerely, f
_~:‘\.‘ ‘ "If '1 ’- i ‘1 . V/ l n.“ .
’,W ”W "Rf Cv’i ’11-" ’1‘“; . '

/ _.

David E. Wright, PhD. .
UCRIHS Chair ' -- "

DEW: db
Jennifer Krenos

74

APPENDIX B

Parent/Guardian Questionnaire

75

Preschool Children’s Beverage Preferences
Parent! Guardian Questionnaire

Subject # Date

Please respond to the following questions as they pertain to you and the members of your household. Your time and
effort is greatly appreciated and will assist us in manning the beverage prefaenece of preschool children.
GENERAL HOUSEHOLD INFORMATION

1. Whotalteacnreofthechildreninyourhouseholdandabouthowoldisenchpaaon?

 

 

 

Howmanyyearaofechooilnveeachoftherncompleted?

 

2. HowmanyotherMandadnb(agel3andm)liveinthehouae?
WIntaretheiragee?

 

3. How many lab-ta and Mee- live in the home?

 

Whataretheirape?
4. lntheMwhohasdecidedwhatMofthemeabwﬂlbe?
(check all that lpply) mothe/fenraleheadofhouaehold

fathclnralehadofhouaehold
__othcindivithraLpleaaeliat

5. lntheggtthne-onthwhohaadonemouofdrefoodalwppin'inthehunehold?
(check all that apply) rnothulferulehad of homehold
fatherlnralehudofhouechold

othcilldivithlaLpleaaeliet

6. Inhwhwdnmbaaoﬂhemehoﬂpanicipandhfoodmm?
(circle one)
Yes No
If yes. which 0116? (check all that W!) W1C (Wm Infam and Chikhen)
Food Slam
Other. plane list

 

76

10.

What type of water do the household members drink most often in the home?

(check all that apply)

Why?

well water

city water

bottled water

no water is consumed

 

What type of milk is most often consumed by the membas ofthe household?

(check .11 that apply)

Why?

whole. (Vitamin‘D)

2% milk

1% or V196 milk
skim milk

chocolate milk
lactose-free milk

no milk is cornumed

 

Winttypeofeodalmhmostoﬂenoomumedbythemanbcaofthehoueehold?

(check all that apply)

“caffeine-free

othcsodasmoumain 09",qu Sprite. 7-UP, lemon
limeac)

nosodaisoonsumed

 

L H ’SIN A N

Hasahealthprovidertoldyouthatmhasmyfoodallcgia? (circleone)

Ya No If yes, wilt are they?

Docs mg gmehooler I've any food alludes (ex. Milk allay, lactose intoIa-ancc) rigll now? (cicle one)

Yes No If ya. what re they?

77

12. Please indicate how von think vour preschooler would rank the following beverages according to
preference? (1- favorite, 2= second favorite, etc.) Place an X if your child has never tried the beverage.

whole milk

apple juice

water

regular cola

juice drinks (such as Kool-Aid, Hi-C. etc)
chocolate milk

13. What beverages do you offer your preschooler when they are in your care? ( please list)

 

 

14. What beverages does your preschooler drink most often when they are in your care? (please list)

 

 

15. Does your preschooler eat before going to Head Start uch day? (circle one) Yes No
lfycs, what foods do they gym have?

 

PARENT/GUARDIAN INFORMATION

16. Please rank the following bevaages according to m mm for them (1- favorite, 2- second favorite,
etc.) PhceanXifyourarelydlinkthcbcveage.

 

whole milk
_. Imbiuice
MW
nevi-r cola
juice thinks (such as Kool-Aid, Hi-C, etc)
chocolate milk

17. Pleaserankthcfollowing bevcages accordingtohow healthymtblnktheyarefor child'en. (I-healthlut.6
-Ieasthealthy) Thcearenorightorwronganswers.

whole milk

apple juice

Villa

result! 001!

juice chinks (such as KooI-Aid, Hi‘C, ac)
chocolate milk

78

l8. Please circle how often mhave had these beverages within the a three months:

 

 

 

 

 

 

 

 

 

 

Lam “___Mfﬁm d NFL-smea—
Water I 2 3 4 Sormore l Sormore
Whitemilk l 2 3 4 Sorrnore l Sorrnore
Chocolaternilk l 2 3 4 Sormore l Sorniore
Regularsoth/pop l 2 3 4 Sormore l Scrmore
Dietsoth/pop l 2 3 4 Sormore l Sorrnore
100%juice l 2 3 4 Sormore l Sormore
(«material-Squads)

JuiceDrinks l 2 3 4 Sormore l Scrmore
mmmwﬁnu) '

Coffee I 2 3 4 Sormore l Sarnore
Tea 1 2 3 4 Sormore l Scrum

 

l9. Whatothanon-alcoholichevaagadomconsmneﬂpleaselist)

 

 

79

APPENDIX C

Cartoon Faces for Beverage Preference Exercise

80

 

APPENDIX D

Consent Form

82

 

Consent for Research

Your child who is in Head Start,

 

(First name of child) (Last name of child)
is allowed to participate in a tasting exercise as a part ofa study through the Michigan State
University, Department of Food Science and Human Nutrition.

By signing below, you indicate that you have read the consent letter explaining the
research, understand the information listed below and allow your child to participate in

the study.

0 Yom'childwilltastcthefollowingdrinks: 2%whitctnilk, 1%dlocolatemiIKHi-CBtwin’Bufy
drink; Kool Aid Strawbary-Kiwi drink; apple juice, bodied water and caffeine-free Coca-cola.

0 Yourchildwillhaveastnalltasteofeadibevaagemdmenbeakedtomkmebevunysasto
whichtheylikethemost. ‘Ihetotaltimethechildwillspaidwiththeresearcherwillbe
approximatelyten(10)minmes.

- Yourchildwillnotbeforcedbymereseuchasmtastemybeveragcedmhdshedoesnotmto
taste.

- Yourchild’sMmtheiglmtmdweiglrtMuﬂbMdﬂewﬂlbedenmbyﬂn
HeadStartstaﬂ‘.

- All datacollected will remainanonymous. The information collected frorntheparent/guardianand
thetastingexercisewillbeusedbytheresearchersonly. Codenumbaswillbemedsothatnames
ofthehidiﬁdudswuldrmmdpummmnnmbenccdcdwhmdninfamaﬁmiswﬂeaed.

- Childrenwithallergiestothebeveragesinthestudy,mcdicalpoblemsprohibitingtl’teirintakeofthc
bevaagesa'dioldng/swallowing problems will notpanicipsteinthetastingportion of the study.

 

(paruﬂgwdian signature) (“0)

 

(print petal/guardian tame)

83

 

APPENDIX E

Preschool Child Information Sheet

84

Preschool Child Information Sheet

Child’s Na-e

 

(In) (ﬁrst) (initial)
Sea ofchlld (circle one): Male lie-ale
Cherno- Name:
Cl. tine (circle one): Mornlq Afternoon Al Day

1. Please listany PAST food allagiathatyour childhashad:
A_I_eln muse-cumming“

 

 

2. PleaselistanyCURRl-ZN'I'foodallcgiesyourchildhns. Howlonghavetheyhadlhisallagy?
_Alem Perm—'21?

 

 

 

3. Is it ok to give your child a small taste ofthe following drinks? Circle yes or no. lfno, please explain why (alla'gic,
diabaic, etc).

 

 

 

 

 

 

 

2% whhe it Yes No If no, why?
”6 chocolate .lllt Yes No If no, why?
Iii-C Bopﬂn’Berry Punch Yes No If no, why?
Kool-Ald Strawberry-Kiwi drink Ya No lfno, why?
Bottled Wm Yes No If no, why?
Apple Jdu Yes No If no, why?
Caﬂelne-bee Coca-cola Yes No lf no, why?
4. Does your child live any choking or swallowing prohlans? (circle one) Yes No
If yes, plane explain the prohlan.

 

 

85

APPENDIX F

Head Start Beverage Information Collection Sheet

86

‘ f5“??? - j.

Prescth Children’s Beverage Preferences
Head Start Beverage Information '

Record of Beverages offered to Head Start Participants

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Day of the Week Dale

___W TIL—Hf 0' mm.
l.

2.

3.

4.

5.

Day of the Week Date

Bevem Tlrne Dn Addiﬂl Foods m

l.

2.

3.

4.

5.

Day of the Wedt Date

23m: TM AW

 

 

 

 

 

 

 

87

.V-'

Day of the Week Date

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Beverag Time of Dav Additional Foods mend
l.

2.

3.

4.

5.

Day ofthe Week Date

__L-M 1222;: mm
1.

2.

3.

4.

 

 

88

 

APPENDIX C

Beverage Preference Data Collection Sheet

89

Beverage Preferences in Head Start Preschool Children

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Data Collection Sheet
Subject 4 Age (in yr. ltd mo.) W
am: From record a pan recall
momenta) Weightahs.) l-leight(al¢cs) Weight/Age
Weight no new hm taker sun:
“MSW Hair!" “ism/Act Mic W 50‘
Laun‘rnedrildudrhmk mm Feedlbevuage What?
mm” Exerebe Rails 2L
Tasting Orda:

Cornet Olihd Girl’s
W Lumen——
2. Keel-Aid Yes I No Yes [NO
3. iii-C Ya I No Ya / No
4. Set Yo I No Yes I No
5. Wu Yes I No Ya I No
6. With Imlk Yea / No Ya I No
7 (liooolde nnlk Ya I No Ya / No
Preference Order (must preferred-l, least #7):

W

 

 

 

 

 

 

 

 

 

9O

REFERENCES

Alles-White, ML. and Welch, P. Factors affecting the formation of food preferences in
preschool children. Early Child Dev and Care 1985;21:265-276.

American Dietetic Association. Position of the American Dietetic Association:
Domestic hunger and inadequate access to food. J Am Diet Assoc 1990;90:1437-1441.

American Dietetic Association. Position of the American Dietetic Association: Nutrition
standards for child care programs. J Am Diet Assoc 1994;94:324-328.

Anliker, J .A., Laus, M.J., Samonds, K.W.and Beal, V.A. Parental messages and the
nutrition awareness of preschool children. J Nutr Ed 1990;22:24-29.

Ashraf, H.R., Schoeppel, C and Nelson, J .A. Use of tofu in preschool meals. J Am Diet
Assoc 1990;90:1114-1116.

Baranowski, T., Sprague, D., Baranowski, J .H. and Harrison, J. Accuracy of maternal
recall for preschool children. J Am Diet Assoc 1991;91:669-674.

Birch, L.L. Children, parents and food. Br Food J 1993;95:11-15.
Birch, L.L. Development of Food Acceptance Patterns. Dev Psych 1990;26:515-519.

Birch, L.L. Dimensions of preschool children’s preferences. J Nutr Ed 1979a;] 1:77-
80.

Birch, L.L. Preschool children’s food preferences and consumption patterns. J Nutr Ed
1979b;11:189-192.

Birch, L.L. Effects of peer models’ food choices and eating behaviors on preschoolers’
food preferences. Child Dev 1980a;51:489-496.

Birch, L.L. The relationship between children’s food preferences and those of their
parents. JNutr Ed 1980b;]2:14-18.

Birch, L.L. Generalization of a modiﬁed food preference. Child Dev 1981;52:755-
758.

Birch, LL. and Billman, J. Preschool children’s food sharing with friends and
acquaintances. Child Dev l986a;57:387-395.

Birch, LL. and Deysher, M. Caloric compensation and sensory speciﬁc satiety:
Evidence for self regulation of food intake by young children. Appetite 1986b;7:323-331.

91

Birch, LL. and Marlin, D.W. I don’t like it; I never tried it: Effects of exposure on two-
year-old children’s food preferences. Appetite: J Intake Res 1982;3:353-360.

Birch, LL. and Sullivan, S.A. Measuring children’s food preferences. J School Health
l991a;61:212-214.

Birch, L.L., Birch, D, Marlin, D.W. and Kramer, L. Effects of instrumental consumption
on children’s food preference. Appetite: J Intake Res 1982b;3:125-134.

Birch, L.L., Johnson, S.L. and Fisher, J .A. Children’s eating: The development of food-
acceptance patterns. Young Children 1995;50:71-78.

Birch, L.L., Johnson, S.L, Andresen, G, Peters, J .C. and Schulte, MC. The variability of
young children’s energy intake. N Eng J Med 1991b;324:232-235.

Birch, L.L., Marlin, D.W. and Rotter, J. Eating as the “means” activity in a contingency:
Effects on young children’s food preference. Child Dev 1984;55:431-439.

Birch, L.L., Zimmerman, 8.1. and Hind, H. The inﬂuence of social-affective context on
the formation of children’s food preferences. Child Dev 1980;1z856-861.

Borah-Giddens, J. and F alciglia, GA. A meta-analysis of the relationship in food
preferences between parents and children. J Nutr Ed 1993;25:102-7.

Boyle, M.A. and Morris, D.H. Communig Nutrition in Action. West Publishing
Company, New York, New York, 1994.

Briley, M.E., Buller, A.C., Roberts-Gray, CR. and Sparkman, A. What is on the menu at
the child care center? J Am Diet Assoc 1989;89:771-774.

Bruening, K.S., Gilbride, J .A., Passannate, MR. and McClowry, S. Dietary intake and
health outcomes among young children attending two urban day-care centers. J Am Diet
Assoc 1999;99: 1529-1535.

Byrd-Bredbenner, C. Development of a nutrition education curriculum for Head Start
children. J Nutr Ed 1993;25:134-138.

Colavito, E.A., Guthrie, J .F., Hertzler, A. and Webb, R.E. Relationship of diet-health
attitudes and nutrition knowledge of household meal planners to the fat and ﬁber intakes
of meal planners and preschoolers. J Nutr Ed 1996;28:321-328.

Crockett, SJ. and Sims, L.J. Environmental inﬂuences on children’s eating. J Nutr Ed
1995;27:235-249.

92

 

Davis, C. Results of the self-selection of diets by young children. The Can Med Assoc J
1939;41 :25 7-261 .

Davis, S.S., Bassler, E.M., Anderson, J .V. and Fryer, H.C. A nutrition education
program for preschool children. J Nutr Ed 1983;15:4-5.

Dennison, B.A. Fruit juice consumption by infants and children: A review. J Am Coll
Nutr 1997;15:4S-11S.

Dennison, B.A., Rockwell, H.L. and Baker, S.L. Excess fruit juice consumption by
preschool-aged children is associated with short stature and obesity. Pediatrics
1997a;99215-22.

Dennison, B.A., Rockwell, H.L. and Baker, S.L. Fruit juice consumption (letter).
Pediatrics 1997b; 1 00:729-730.

Dhurandhar, N.V. Fruit juice consumption (letter). Pediatrics 1997;100:728-729.

Doong, J .Y., Shariff, Z.M., Hoerr, S., Bond, J .T. and Hartgerink, T. Growth status of
low-income preschool children and the relation of birth weight to short stature and
overweight. J Fam Cons Sci 1998:12-19.

Drake, M.A. The nutritional status and dietary accuracy of single homeless women and
their children in shelters. Pub Health Reports 1992;197:312-319.

Duncker, K. Experimental modiﬁcation of children’s food preferences through social
suggestion. J of Abnorm Soc Pscyh 1938;33:489-507.

Dunn, P.C., Lackey, C.J., Kolasa, K. and Mustain, D. At-home nutrition education for
parents and 5- to 8-year-old children: The Home Plate Study. J Am Diet Assoc
1998;98:807-809.

Falkner, F. Monitoring Growth. In Gracey, M. and Falkner, F. eds., Nutritional Nee__d§
and Assessment of Normal Growth. Vevey Raven Press, New York, New York, 1985.

Gibson, S.G. Non-milk extrinsic sugars in the diets of pre-school children: association
with intakes of micronutrients, energy, fat and NSP. Br J Nutr 1997;78:367-378.

Harnack, L., Stang, J. and Story, M. Soft drink consumption amount US children and
adolscents: Nutritional consequences. J Am Diet Assoc 1999;99:436-441.
Hertzler, A.A. Preschoolers’ developmentally appropriate food and nutrition skills. J

Nutr Ed 1994;26: 166B.

Jennings, C. Children in the United States. Fam Econ Review Winter 1979:17-21.

93

Kennedy, E., Riddick, H., Kramer-LeBlanc, C. and Davis, C. Fruit juice consumption
(letter). Pediatrics 1997;100:728.

Klish, W.J. Use of fruit juice in the diets of young children. Pediatrics 1995;95:433.

Lytle, L and Achterberg, C. Changing the diets of America’s children: What works and
why? J Nutr Ed 1995;27:250-260.

National Commission on Children. Beyond rhetoric. A new American agenda for
children and families. Washington DC, US. Government Printing Ofﬁce, 1991.

Newman, J. and Taylor, A. Effect of a means-end contingency on young children’s Food
Preferences. J Exp Child Psych 1992;64:200-216.

Petridou, E. Karpathios, T., Dessypris, N, Simou, E. and Trichopoulos, D. The role of
dairy products and non alcoholic beverages in bond fractures among schoolage children.
Scand J Soc Med 1997;25:119-125.

Petter, L.M.P., Hourihane, J .O. and Rolles, C]. Is water out of vogue? A survey of the
drinking habits of 2-7 year olds. Arch Dis Child 1995;72:137-140.

Phillips, B.K. and Kolasa, K.K. Vegetable preferences of preschoolers in day care. J
NutrEd 1980;12:192-195.

Skinner, J ., Carruth, B.R., Houck,K.S., Bounds, W., Morris,M., Cox, D.R., Moran,J. and
Coletta, F. Longitudinal study of nutrient and food intakes of white preschool children
aged 24 — 60 months. J Am Diet Assoc 1999;99:114-1521.

Skinner, J., Carruth, B.R., Moran, J ., Houck, K., Schmidhammer, J ., Reed, A., Coletta,
F ., Cotter, R., and Ott, D. Toddlers’ food preferences: Concordance with family
members’ preferences. J Nutr Ed 1998;30:17-22.

Smith, M.M. and Lifshitz, F. Excess fruit juice consumption as a contributing factor in
nonorganic failure to thrive. Pediatrics 1994;93438-443.

Smith, M.M., Davis, M., Chasalow, RI. and Lifshitz, F. Carbohydrate absorption from
fruit juice in young children. Pediatrics 1995;95:340-344.

Stinson, W. Fruit juice consumption (letter) Pediatrics 1997;100:728.

Sullivan, SA. and Birch, L.L. Pass the sugar: Pass the salt: Experience dictates
preference. Dev Psych 1990;26:546-551

Trowbridge, F .L. Prevalence of growth stunting and obesity: Pediatric Nutrition
Surveillance System. MMWR 1983;32:23-26.

94

 

 

 

United States Bureau of the Census. Poverty in the United States: 1992. Current
Population Reports (185). Washington, DC, US. Government Printing Ofﬁce, 1993.

95