INFORMATION TO USERS

This material was produced from a microfilm copy of the original document. While the most advanced technological means to photograph and reproduce this document have been used, the quality is heavily dependent upon the quality of the original submitted.

The following explanation of techniques is provided to help you understand markings or patterns which may appear on this reproduction.

1.The sign or "target" for pages apparently lacking from the document photographed is "Missing Page(s)". If it was possible to obtain the missing page(s) or section, they are spliced into the film along with adjacent pages. This may have necessitated cutting thru an image and duplicating adjacent pages to insure you complete continuity.

2. When an image on the film is obliterated with a large round black mark, it is an indication that the photographer suspected that the copy may have moved during exposure and thus cause a blurred image. You will find a good image of the page in the adjacent frame.

3. When a map, drawing or chart, etc., was part of the material being photographed the photographer followed a definite method in "sectioning" the material. It is customary to begin photoing at the upper left hand corner of a large sheet and to continue photoing from left to right in equal sections with a small overlap. If necessary, sectioning is continued again — beginning below the first row and continuing on until complete.

4. The majority of users indicate that the textual content is of greatest value, however, a somewhat higher quality reproduction could be made from "photographs" if essential to the understanding of die dissertation. Silver prints of "photographs" may be ordered at additional charge by writing the Order Department, giving the catalog number, title, author and specific pages you wish reproduced.

5. PLEASE NOTE: Some pages may have indistinct print. Filmed as received.

University Microfilms International 300 North Zeeb Road Ann Arbor. Michigan 40106 USA St John s Road. Tyler s Green High Wycombe. Bucks. England HP10 8HR 7902081

BORNS, TINTTIHY HATTHEN THE IMPACT OF TELEVISION FOOD COMMERCIALS YOUNG CHILDREN.

THE OHIO STATE UNIVERSITY, P H .D ., 1976

* 1 8 3 * . raUIMdOnil MO N. » » W M O ; A W M tK W . Ml W I N

© 1978

TIMOTHY MATTHEW BURNS

ALL RIGHTS RESERVED THE IMPACT OF TELEVISION FOOD COMMERCIALS

ON YOUNG CHILDREN

DISSERTATION

Presented in Partial Fulfillment of the Requirements for

the Degree Doctor of Philosophy in the Graduate

School of The Ohio State University

By

Timothy M. Burns, B.A., M.A.

* * * +

The Ohio State University

1978 Reading Committee: Dr. Alexander Frazier *u " p " °v».r T - V' Dr. Frederick Staub Advi ser Department of Early and Professor Isabel Miller Middle Childhood Education ACKNOWLEDGMENTS

It is with sincere appreciation that I acknowledge the help provided by my major adviser, Professor Alexander

Frazier. I also wish to acknowledge the professional and personal guidance of the other members of my committee,

Professors Isabel Miller and Frederick Staub, who have been supportive of me both as a person and as a graduate student.

My sincere thanks to the students, teachers, and administrators in the Pasadena, California Unified School

System who participated in the research which resulted in this dissertation, and particular appreciation goes to my wife Kathy and our close personal friends, Mary Bologna and

Jim McManus without whose encouragement and enthusiastic support I would never have completed this dissertation at such great distance from Ohio State University. VITA

February 8 , 1944 ...... Born - Defiance, Ohio 1966 ...... B - A. - Liberal Arts John Carroll University, Cleveland, Ohio 1970 ...... M.A. - English Literature Georgetown University Washington, D.C. 1967-1968 ...... Seventh Grade English, and Social Studies Teacher, Fairfax County Schools, Fairfax, Virginia 1968-1970 ...... Senior High School English Teacher (Grades 10, 11, 12), Fairfax County Schools, Fairfax, Virginia 1970...... M.A., Georgetown University, Washington, D.C. 1970-1972 ...... Sixth and Seventh Grade English Teacher, The Columbus Academy for Boys, Columbus, Ohio 1972-1975 ...... Head of the Lower School (K-6), Columbus School for Girls, Columbus, Ohio 1975-1976 ...... Teaching Associate, Department of Early and Middle Childhood Education, The Ohio State Uni­ versity, Columbus, Ohio 1976-1978 ...... Headmaster, Mayfield Senior School, Pasadena, California

FIELDS OF STUDY Major Field: Early and Middle Childhood Education Studies in Early and Middle Childhood Education Professor Alexander Frazier Studies in Early Childhood Education Professor Isabel Miller Studies in Educational Administration Professor Frederick Staub iii TABLE OF CONTENTS

Page ACKNOWLEDGMENT ...... ii

VITA ...... iii LIST OF TA BLES...... V

LIST OF FI G U R E S ...... viii Chapter

I. INTRODUCTION ...... 1

II. REVIEW OF THE LITERATURE RELATED TO THE EFFECTS OF TELEVISION AND TELEVISION ADVERTISING ON YOUNG CHILDREN ...... 17

III. METHODS AND PROCEDURES...... 46

IV. PRESENTATION AND ANALYSIS OF D A T A ...... 77

V. SUMMARY AND CONCLUSIONS...... 133

APPENDICES A Definitions of Attributes ...... 166

B Instrument for Gathering Frequency Data for Commercials in the Content Analysis. . 169

C Instrument for Gathering Product Data and Sales Persuasion Technique Information for the Content Analysis...... 172

D Random Number Table ...... 175

E Instrument for Gathering Data Regarding Familiarity of Cereal Product, Cereal Package, and J i n g l e ...... 177

F Parents' Questionnaire ...... 183

G Instrument for Gathering Data for Familiarity with Cereals...... 185

BIBLIOGRAPHY ...... 189 iv LIST OF TABLES

Table Page

1. Study One: Reliance on Attributes of Product Data and Sales Persuasion Techniques for all Commercials...... 80

2. Study One: Use of Product Data Attributes in All Commercials...... 81

3. Study One: Use of Sales Persuasion Technique Attributes in All Commercials...... 84

4. Study Two: Four-Year-Old Subjects' Famili­ arity with Product Appearance ...... 87

5. Study Two: Six-Year-Old Subjects' Famili­ arity with Product Appearance ...... 89

6. Study Two: Four-Year-Old Subjects' Famili­ arity with Packaging...... 90

7. Study Two: Six-Year-Old Subjects' Famili­ arity with Packaging...... 92

8. Study Two: Four-Year-Old Subjects' Famili­ arity with Slogan/Jingle...... 93

9. Study Two: Six-Year-Old Subjects’ Famili­ arity with Slogan/Jingle...... 94

10. Study Two: Four-Year-Old Subjects' Famili­ arity with Cheerios Cereal in Three Attri­ butes of Product Appearance, Packaging, and Slogan/Jingle ...... 96

11. Study Two: Four-Year-Old Subjects' Famili­ arity with Frosted Flakes Cereal in Three Attributes of Product Appearance, Packaging and Slogan/Jingle...... 97

v Page

Study Two: Four-Year-Old Subjects' Famili­ arity with Lucky Charms Cereal in Three Attributes of Product Appearance, Packaging and Slogan/Jingle ...... 98

Study Two: Four-Year-Old Subjects' Famili­ arity with Raisin Bran Cereal in Three Attributes of Product Appearance, Packaging and Slogan/Jingle ...... 99

Study Two: Four-Year-Old Subjects' Famili­ arity with Rice Krispies Cereal in Three Attributes of Product Appearance, Packaging and Slogan/Jingle ...... 100

Study Two: Six-Year-Old Subjects' Famili­ arity with Cheerios Cereal in Three Attri­ butes of Product Appearance, Packaging and Slogan/Jingle ...... 102 Study Two: Six-Year-Old Subjects' Famili­ arity with Frosted Flakes Cereal in Three Attributes of Product Appearance, Packaging and Slogan/Jingle ...... 103

Study Two: Six-Year-Old Subjects' Famili­ arity with Lucky Charms Cereal in Three Attributes of Product Appearance, Packaging and Slogan/Jingle ...... 104

Study Two: Six-Year-Old Subjects' Famili­ arity with Raisin Bran Cereal in Three Attributes of Product Appearance, Packaging and Slogan/Jingle ...... * 105

Study Two: Six-Year-Old Subjects' Famili­ arity with Rice Krispies Cereal in Three Attributes of Product Appearance, Packaging and Slogan/Jingle ...... 106

Study Two: Comparison of Familiarity with Product Appearance Between Four- and Six- Year-Old Subjects ...... 109 Study Two: Comparison of Familiarity with Packaging Between Four- and Six-Year-Old Subjects ...... 110

vi Table Page

22. Study Two: Comparison of Familiarity with Slogan/Jingle Between Four- and Six-Year- Old Subjects...... Ill

23. Study Three: Four-Year-Old Children's Requests for Parents to Purchase Cereals . . 116

24. Study Three: Six-Year-Old Children's Requests for Parents to Purchase Cereals...... 118

25. Study Three: Parents of Four-Year-Old Chil­ dren Purchasing Of Cereals as Result of Children's Requests ...... 119

26. Study Three: Parents of Six-Year-Old Chil­ dren Purchasing of Cereals as Result of Children's Requests ...... 120

27. Study Three: Cereal Purchases Made by Parents of Four-Year-Olds and the Relation­ ship to Attributes of Product Data ..... 127

28. Study Three: Cereal Purchases Made by Parents of Six-Year-Olds and the Relation­ ship to Attributes of Product D a t a ..... 129

29. Study Three: Cereal Purchases Made by Parents of Four-Year-Olds and the Relation­ ship to Attributes of Sales Persuasion T e c h n i q u e ...... 130

30. Study Three: Cereal Purchases Made by Parents of Six-Year-Olds and the Relation­ ship to Attributes of Sales Persuasion T e c h n i q u e ...... 132

vii LIST OF FIGURES is Pa-ge i Framework for the Content Analysis of Children's Food Commercials 6

2 Scatter Diagram of Familiarity vs. Attributes of Product Data and Sales Persuasion Tech­ niques 113 3 Scatter Diagram of Parents' Responses vs. Product Data 122 4 Scatter Diagram of Parents' Responses vs. Sales Persuasion Techniques 123 5 Scatter Diagram of Parents' Purchasing vs. Attributes of Product Data 124

6 Scatter Diagram of Parents' Purchasing vs. Attributes of Sales Persuasion 12 5 7 Histogram Indicating Total Percentages of Attributes of Product Data and Sales Per­ suasion Techniques 142 8 Histogram of Product Data Attributes vs. Sales Persuasion Techniques 143 9 Histogram of Percentages of Individual Attributes of Product Data & Sales Persua­ sion Techniques for Cereals in Study Two 145

10 Histogram of Four-Year-Old Subjects' Famili­ arity with Three Attributes Under Study 147

11 Histogram of Six-Year-Old Subjects' Famili­ arity with Three Attributes Under Study 148

viii CHAPTER I

INTRODUCTION

The Nature of the Study

Examination of children's television programming and advertising in recent years has produced a variety of theories about the influence of mass media on youngsters. While a number of studies have tested the proposition that preschool children are affected by television advertising, few have attempted to explore the impact of particular types of com­ mercials on the young child's perception. The purpose of the present research is to analyze television advertising of food products for young children with particular emphasis on cereal advertising for children ages four and six. Such in­ formation should help to test the legitimacy of assumptions now being made about children's responses to television ad­ vertising.

Televison commercials have been studied from several different points of view in recent years. Several investi­ gations have examined the level of recall of commercials shown to housewives and college students (Shulman, 19 72;

Hollander, 1973; Haller, 1972); and extensive examination has also been made of the actual words adult consumers use to give their opinions of television commercial messages

1 2

(Wells, 1972). Advertising aimed at young children has also been an area which has received selected analysis and attention. It is to this aspect of television advertising that the present study is directed.

Recent research has been directed at the potential ability of advertising to manipulate children's thinking

(Disson, 1973; Breen & Powell, 1973) and the question of children's abilities to be selective and discriminating in their responses to visual stimuli (Pearce, 1974; Feldman and

Wolf, 1974). Earlier studies gave little serious considera­ tion to the impact of advertising on youngsters; in fact, the primary role of television as a commercial advertising vehicle went unquestioned for years. Television was seen as a superior instrument for advertising; and since networks exercised total control over early television operations, advertising became practically synonymous with television.

But the emphasis in studies of its advertising was clearly on economics, not on the quality or effects of the commercial messages. This was particulary apparent in the commercial messages aimed at children (Melody, 1974).

It was not until early 1970 that Action for Children's

Television (ACT), a private group of citizens who had orga­ nized in 1968, operated as a catalyst in this area by pre­ senting a petition to the FCC "...to substitute a new system of financial support of children's programing by commercial underwriting and public service funding in the belief that this system would look to the benefit of children rather

than the profits of advertisers" (FCC Regulation 1971).

From that time on, the consciousness of intellectuals,

politicians, educators, and laymen has been raised about the

issue of the effects of programming for children. Research

has consisted largely of investigations into the effects of

violence and aggression in children's television, but few studies have considered the effects of commercials. Many

persons have alleged that television commercials directed

at children have negative effects upon children's values,

beliefs, and behavior, but such charges have been raised without directly applicable data about the effects of

television advertising directed at children.

In view of the somewhat emotional responses to the whole subject of children's television commercials by groups

such as Action for Children's Television, the National

Association for Better Broadcasting, and the Council on

Children, Media, and Merchandising, it seems that more

research is needed in the area if the claims of groups such

as these are to be taken seriously.

Williams (19 74) has reported on the consumer needs of young children and has proposed a constructive utilization

of commercial television to meet those needs. Robertson

and Rossiter (1974) have provided insight into children and 4 commercial persuasion, and other studies have reported on children’s reactions through group interviews with children ranging in age from five to twelve years (McNeal, 1974;

Blatt, 1971; Ward, 1972). Both Blatt and Ward suggested that children's responses to commercials may be related to stages in cognitive development and based their findings on studies by Kohlberg, Loevenger, and Piaget.

The intent of the present series of studies of the effects of commercial advertising on the young child is

(1) to provide a content analysis of commercial advertising directed to youngsters; (2) to investigate young children’s familiarity with specific products; and (3) to describe parent-child interation that results from the impact on children of commercial advertising.

Framework of the Three Studies

The concepts of children's interests and advertising techniques have been combined in order to implement investi­ gation into children's behavior related to commercial adver­ tising. First, a frequency study was made of food commer­ cials directed to children and shown during a three-hour period of Saturday morning programming. An effort was made to determine if breakfast cereals are advertised more than other items such as candy, gum, soft drinks, pastries, or restaurant foods such as hamburgers. These same commercials were then studied in a Content Analysis (Study One) to 5

determine the presence of and reliance on specific attri­ butes that may be categorized as: {1) Product Data and (2)

Sales Persuasion Techniques (Figure 1). Utilizing this basic framework, an assessment of children's familiarity with specific television cereal commercials was then under­

taken (Study Two), as well as an exploration of the extent to which parents pay attention to children's requests as a re­

sult of television advertising of food products (Study Three).

Description of the Studies Descriptions of these three studies that tested the effects of food-product television advertising on young children follow.

Study One: A Content Analysis of Food Commercials Directed to Young Children The purpose of this study was to develop data on television food commercials directed to young children. The basic frame of reference was extracted from a study entitled,

Children's Television Commercials: A Content Analysis by

Winick, Williamson, Chuzmir, and Winick (1973). Modifica­ tions were made to coincide with questions to be answered in this research.

The frequency study first examined food commercials televised during a three-hour period (8:00 - 11:00 A.M.) of

Saturday morning children's programming (Appendix B).

Frequencies of these food commercials on the three major television networks (ABC, NBC, CBS) were counted and

recorded in an attempt to determine that cereal commercials

appear more frequently than any other kind of commercials for food during children's Saturday morning programming.

Product Attributes Categories

Product Data A. sweetness B. physical/oral character­ istics C. nutritional value D. product appearance or shape E. packaging of product

Sales Persuasion A. fun food/fun notion Techniques B. premium included with product C. association with physical prowess D. cartoon/fantasy character as spokesman E. use of slogan or musical j ingle

Figure 1. Framework for the Content Analysis of Children's Food Commercials

These same food commercials were then analyzed, and tallies were made for presence of or reliance on items of content as expressed in an instrument dealing with product attributes. Commercials had been viewed in preliminary studies for three-hour periods on three consecutive Saturday mornings in April and May, 197 6 in Columbus, Ohio, to 7

determine what might constitute an appropriate series of

content attributes or dimensions (as suggested by the earlier

Winick study] to examine in this investigation. The two categories of product attributes (Product Data and Sales

Persuasion Techniques) became apparent through this prelimi­ nary work. The actual commercials examined in this Content

Analysis were those televised between 8:00 and 11:00 A.M.

on the three major television networks (ABC, NBC, CBS) on

July 3, 1976, in Columbus, Ohio.

Within the Content Analysis, two distinct divisions were utilized. As found by Winickrs more comprehensive

study, commercials have a number of attributes which may be present in all commercials. The commercials in this study were analyzed to determine which of the ten selected attri­ butes were present. Further, the attributes themselves were classified into two main categories, namely: (1) Pro­ duct Data and (2) Sales Persuasion Techniques. Relationships between the two divisions in the Content Analysis provided insight into the significance of these two main categories of attributes in children's television commercials. Defini­

tions of items in these two categories of attributes

(Product Data and Sales Persuasion Techniques) are presented in Appendix A. 8

The Content Analysis of children's food commercials provides insight into the following questions:

IA. To what extent do the commercials under study differ in their reliance on items of Product Data and Sales Persuasion Techniques?

IB. To what extent are there differences in use of or reliance upon, the following Product Data items in commercials?

1. sweetness 2. physical/oral characteristics 3. nutritional value 4. product appearance or shape 5. packaging of product

IC. To what extent are there differences in use of or reliance upon, the following Sales Persuasion Technique items in commercials?

1. fun food/fun notion 2. premium included with product 3. association with physical prowess 4. cartoon/fantasy character as spokesman 5. use of slogan or musical jingle

Data derived from the Content Analysis have been utilized in Study Two which examines children's familiarity with advertised cereal products and Study Three which inves­ tigates parent-child interaction as a result of children's exposure to commercial advertising of certain cereal products.

Study Two: Children's Familiarity with Specific Cereal Advertisements

The purpose of this study has been to test young chil­ dren’s recognition of specific cereal advertisements. Based on the assumption that nonreading children can learn things by association, this study examined children's ability to recognize cereal products and the appropriate packaging and advertising jingle or slogan, using an instrument developed to record children's familiarity with products, packaging, and audio clues. Nonreading children (ages four and six) were given samples of cereal products in plastic bowls to determine their recognition of them. Children were asked specifically,

"What is the name of this cereal?" Then the children were shown cereal packages and again asked to identify the brand names. These two tests measured children's familiarity with both product and packaging. A third test of familiarity measured children's abilities to match packaging with audio clues. Jingles and/or slogans were played from a tape recorder and children were asked to match the jingle or slogan with one of five cereal packages placed before them.

The children were interviewed individually for this investi­ gation.

Data obtained from this study were analyzed in light of the two categories of product attributes in Study One, namely, aspects of Product Data and aspects of Sales Persua­ sion Techniques. Looking at the question of familiarity in these ways has provided information about the significance of these two kinds of attributes in cereal commercials.

These explorations into children's familiarity with cereal advertisements provided insight into the following questions: 10

I1A» To what extent do the measures of familiarity of five cereal products vary:

1. According to product appearance/shape a. Four-year-olds b. Six-year-olds

2. According to packaging a. Four-year-olds b. Six-year-olds

3. According to jingle/slogan a. Four-year-olds b. Six-year-olds

4. For each cereal a. Four-year-olds b. Six-year-olds

XIB. To what extent do four- and six-year-old chil­ dren's responses to the three measures of familiarity differ for the following:

1. product 2. packaging 3. jingle/slogan

IIC. To what extent do the three measures of famil­ iarity in this study (product/appearance, packaging, jingle) relate to the attributes of Product Data and Sales Persuasion Techniques from Study One:

1. Attributes of Product Data: a. sweetness b. physical/oral characteristics c. nutritional value d. product appearance or shape e. packaging of product 2. Attributes of Sales Persuasion Techniques a. fun food/fun notion b. premium included with product c. association with physical prowess d. cartoon/fantasy character as spokesman e. use of slogan or musical jingle 11

Study Three: Parent-Child Interaction as a Result of Television Advertising of Food Products for Children The purpose of this study was to investigate rela­ tionships that develop between parent and child as a result of food items the child sees advertised on television. A questionnaire# directed to parents of youngsters involved in

Study Two# listed the five cereal brand names of products utilized in Study Two. Parents were asked to respond to each of the items listed by checking one of the following categories: (1) "My child has requested me to buy this cereal as a result of television advertising" and (2) "I have purchased this cereal as a result of my child’s request for it." The results of this questionnaire were analyzed in an attempt to see the relationship between the requests chil­ dren make of parents and what action parents choose to take as a result of such requests. The analysis of these results has provided data on the following questions:

IIIA. To what extent do children's requests for parents to purchase products as a result of television advertising vary among the following indicated categories:

1. Four-year-old children 2. Six-year-old children

IIIB. To what extent do parents' purchase of products as a result of children's requests vary among the following indicated categories:

1. Among parents of four-year-olds 2. Among parents of six-year-olds 12

IIIC. To what extent are the attributes of Product Data and attributes of Sales Persuasion Techniques used in Study One related to parents1 responses to their children's requests to purchase certain cereals according to children's ages.

1. Parents of four-year-olds 2. Parents of six-year-olds

IIID. To what extent are there relationships between the frequency with which parents have purchased certain products their children have requested and particular attributes of Product Data and Sales Persuasion Techniques as described in the Content Analysis?

General Assumptions and Limitations

This investigation of the impact of television adver­ tising on young children is based on the assumption that the young child is influenced by commercials. It is further assumed that companies directing advertisements to young children rely on two basic elements in preparing the adver­

tisements: (1) Product Data which gives information about the item and (2) Sales Persuasion Techniques which employ tactics to gain the child's acceptance of the product. The first two studies in this research concentrate heavily on these basic assumptions. The third study adds the dimension that television advertising directed to children in turn affects parent purchasing habits as well, since young chil­ dren may not make many purchases themselves. This research has been an attempt to account for some of the parent-child interaction that results from advertising aimed at youngsters. 13

A recognizable limitation of these studies is the fact that the commercials used in the Content Analysis will

represent a three-hour time period for only one Saturday

morning in July, 1976. Commercials aired at other times

during the year may be programmed differently and may include

different kinds of products. Generalizations of this re­

search are limited to the commercials viewed for this study

alone.

Another acknowledged limitation of Study Two of this

research is the fact that it deals with young children's

familiarity with products. It is entirely possible that

young children's familiarity with products will be due to

having been exposed to these products through parents pur­ chasing them by chance and not necessarily because of tele­

vision advertising. This fact is simply acknowledged but

hopefully will not color the results of the test. This re­

search is simply investigating children's familiarity with

certain products, packaging, and jingles or slogans.

The subjects involved in this research, both children

and parents, are a select group. The children in Study Two were not randomly selected from the possible universe of all

four- and six-year-old children, and the parents, likewise, were not randomly selected either. Consequently, generali­

zations regarding their responses in this study will be

limited. 14

This investigation has also been limited to specific types of cereal commercials aired during children's Saturday morning programming. No attempt was made to deal with food products other than those described in Study One or with non­ food items. Such research may be needed in the field, but has not been included in the scope of this study.

Statistical Analysis and Instrumentation

Because of the nature of this research and the data that were collected, the most frequently used statistic was the chi square distribution. Emphasis was placed on counting the frequency of occurrence of items surveyed and comparisons were made between these frequencies through the use of histo­ grams. Other than the one borrowed instrument modified for

Study One, original instrumentation was created to collect data for these studies. Examples of instruments used appear in the Appendices of this investigation.

Definition of Terms

Commercial Attributes. The ten different aspects of food commercials which are classified within the two cate­ gories of (1) Product Data and (2) Sales Persuasion Tech­ niques which are included within the commercials themselves. Familiarity. Children's ability to properly identify a sample of the product, the packaging, and the slogan or jingle will constitute familiarity with the product. 15

Product Data. References in the copy of the commer­ cial to particular items of information about the product

(Appendix A). Sales Persuasion Techniques. References in the copy of the commercials which engage the child's interest or attention to aspects other than Product Data of the item being advertised (Appendix A).

Parent-Child Interaction. The verbal exchange that occurs between a child and parent as a result of the child desiring the parent to purchase a product.

Young Children. The study will rely on responses from four- and six-year-old children who are nonreaders.

Summary This introduction has presented the framework for a Content Analysis of specific food commercials and two studies on the impact of televised food commercials on young children.

It has listed the questions asked by each of the three studies and has acknowledged the general assumptions and limitations of the research. Definitions of key terms are also presented.

The first study analyzes the frequency of types of commercials and the attributes of Product Data and Sales

Persuasion Techniques that are included in them. Study Two isolates certain cereal commercials from the commercials analyzed in the Content A.ialysis and tests young children's 16 familiarity with the product, the packaging, and slogans or jingles. Information obtained in this study are compared with findings obtained from the Content Analysis. The involvement of parents in children's reactions to the tele­ vision commercials analyzed in the Content Analysis is the focus of Study Three. The questions asked by this research attempt to note significant characteristics of the commercials themselves as well as distinguish between the responses of four- and six- year-old children and their parents. Through these explora­ tions, hopefully a better understanding of food commercials aimed at young children has evolved. CHAPTER II

REVIEW OF THE LITERATURE RELATED TO THE EFFECTS OF TELEVISION AND TELEVISION ADVERTISING ON YOUNG CHILDREN

Although there is widespread interest in the effects of television on children, this interest has been focused primarily on the effects of television programming rather than upon television advertising. Consequently, while there is a large body of literature concerned with the former, there is relatively little material concerned mainly with the latter. The literature related to the effects of tele­ vision and television advertising on young children is dis­ cussed here under these headings: the nature of the study of the impact of television advertising on children; the physiological and psychological effects of television view­ ing on children; consumer information processing— the adult model and the child model; and finally, enhancing the child's ability to process televised consumer information. Under the section titled Nature of the Study, child consumerism will be explored, while the child's receptiveness to television commercials and his resistance and immunization to television advertising will be discussed in the section titled Enhancing the Child's Ability to Process Televised Consumer Information.

17 18 Nature of the Study of the Impact of Television Advertising on Children

In recent years, the growing concern over the impact that television advertising may have on children has been shared by various groups, including governmental agencies, broadcasters, advertisers, and consumer educators. Studies have shown that the average child between the ages of two and eleven watches between two and three hours of television each day (Liebert, Neale, & Davidson, 197 3)- It has also been estimated that 20 million children will spend approximately

15 thousand hours observing television programs prior to reaching their eighteenth birthday. This figure represents more time than they will spend in school. Furthermore, one out of every five of these hours is spent watching commercial messages. In 19 73, children between the ages of eight and twelve spent approximately $2.5 billion on items purchased as the result of television ads (Williams, 19 74). Seventy- five million dollars were used to sponsor programming directed at children. Eight companies, whose main products were break­ fast cereals and toys, accounted for half of these sponsor­ ships (Ward, 1972b). Studies in 1971 by Choate and in 1972 by Ward, Levinson, and Wackman supportedly claimed that 50% of all commercials that appeared during programs for younger children were for food products and toys. Much of the analy­ sis developed in this particular area over a period of years has for the most part been quantitative. Aside from that, 19 other sample studies have centered more generally on research

directed toward the influence of television programs on chil­

dren as opposed to the influence of television commercials

(Schramm, Lyle, & Parker, 1965).

Possibly one factor that has discouraged the genera­

tion of such studies is the difficulty involved in determin­

ing where the responsibility for advertising lies. Original­

ly, television began as a means to sell products, not as a

cultural force to educate children. Industrial firms set up

sponsorships with the idea that programming costs would

eventually be met through consumer spending (Henson, 19 70).

And yet, although networks were not set up with the purpose

of educating children, they did and still do exert an influ­

ence on them. Margaret Heckler in a letter to the Federal

Communications Commission (FCC) in 1970 was concerned with

this influence and urged the federal government to accept

more responsibility in regulating not only the quality of

the programs viewed by children, but also the amount and quality of commercial messages:

The child imitates what he sees. Therefore, we must be sure that impressionable young minds are not subjected to debilitating and destructive influences.... It is the govern­ ment's obligation to protect the interests of the citizen, especially the young citizen, who may suffer even more harshly from our inattention to his needs (Heckler, 1970).

Another comment regarding this concern came out in an

editorial in Advertising Age which remarked: "...if some­

thing is wrong with advertising directed at children, the 20 blame should rest with the parents, since they have the ulti­ mate control and responsibility for their children's viewing habits" (Christopher, 1971). It is obvious that the ques­ tion of responsibility cannot easily be resolved. However, commercial monitoring has been ongoing since 1952 (Schramm et al., 1965) . The main organization involved with the regulation of commercials is the Code Authority of the National Association of Broadcasters. In 1952, this agency issued standards re­ garding commercial messages, which have been revised several times since then. Recommendations on television advertising for young people have also been issued by the Association of

National Advertisers. However, no formal machinery has been instrumented to enforce these recommendations (Schramm et al.,

1965). In 1974, the Canadian Radio Television Commission acted to remove all advertising to childrens' programs sup­ plied over the Canadian Broadcasting System. Action for

Children's Television (ACT) has not only ventured to elimi­ nate commercials from programs viewed by children, but also has urged that the quality of such programs be improved

(Cooney, 1970) . Agencies such as these are concerned with developing programs that will enhance the growth of the child and will fortify the child's belief in his work, his background, and his potential (Heckler, 1970) . Child Consumerism

Considering that various agencies have alerted atten­ tion to the monitoring of television advertisements as a public service, it would seem likely that the general public itself would be supportive in this concern. And yet, the

Roper Research Association found in 1972 that as many as 93% of the American public supported the principle of children's advertising, especially when reminded of how children's shows are financed. Acceptance of the role and nature of adverti­ sing is general. Many of the same techniques used by Madison

Avenue firms to sell products were incorporated by Sesame

Street to teach children basic skills (Cooney, 1970). On the one hand, child consumerism may aid in the process of maturation since it involves decision making (Ward, 1972a).

On the other hand, evidence has been found suggesting that children may develop somewhat cynical attitudes toward com­ mercials at a fairly early age (Ward, 1972b) and that, fur­ thermore, the older the child, the less attention he pays to commercial messages in the course of a program (Ward,

Levinson, & Wackman, 1972). Patterns of mistrust may develop in a child when the image conveyed by a commercial is not confirmed in reality. These patterns could lead to further cynicism toward other sources of information and authority

(Winick, Williamson, Chuzmir, & Winick, 1973). Yet, if com­ mercials were banned from children's programs, as prescribed by ACT, local stations would be faced with a financial dilemma (Melody, 1973). 22 Since the total elimination of commercials appears impractical, perhaps what is needed is an improvement in their content and context (Heckler, 1970). Many commercials take advantage of a child’s inability to distinguish between fantasy and reality. One of the basic concerns about tele­ vision advertising directed to children is the criterion that children do not have fully developed powers of reasoning and are unable to examine closely and evaluate sales messages directed at them (Winick et al., 1961). There is little indication in our society that the education of very small children is believed to be just as important, if not more so, as the education of anyone else (Culkin, 1970). This lack of concern may account for some of the problems involved in programming for small children. The outcome is the super­ imposition of the adult who wants to do things that he thinks are good for the child (Heineman, 1970).

A child lives in a visual culture. Yet, institutions are not set up to help him to interpret and to become selec­ tive and discriminating about visual stimuli. It is becom­ ing more apparent that if children are to be trained for consumerism, then objective methods of communication should be employed that will enable the child to discuss material conveyed through television media (Culkin, 1970). In essence, the child must be given the facts along with the social im­ plications so that he may be able to arrive at his own con­ clusions and make value judgements that he intends to live 23 by in his milieu (Heineman, 1970). In order to define these methods needed to insure that children may be properly pre­ pared for consumerism, it is essential to understand more readily what occurs both physiologically and psychologically in children after viewing television commercials.

Physiological and Psychological Effects of Television Viewing

The reaction time between stimulation and response decreases as a child approaches adolescence (Wickens, 1974) .

Younger children tend to have a shorter attention span than older children {Ward & Wackman, 1973). However, visually complex ads are observed for a longer period of time and are better remembered than visually simple ads. Furthermore, products that are easy to identify are better recalled re­ gardless of viewing time (Morrison & Dainoff, 1972) . The relationship between commercial stimulation and commercial recall is in some ways contingent upon the attractiveness of the message being transmitted. In a study conducted by

Breen and Powell (1973), children between the ages of two through five rated video-taped television commercials, se­ lected at random on a scale of (1) true/false, (2) real/not real, (3) happy/not happy, (4) stupid/smart, (5) nice/not nice, (6) good/bad with "I don't know" as an alternative in each category. It was found that children, in general, "tend to believe a commercial message more if they find it attrac­ tive," that girls like and believe television commercials 24 more than boys, and that children tend to say that commercials are stupid regardless of like or dislike (Breen & Powell,

1973). This internalization of information by children allows television to become not simply an active force being imposed on a passive child, but a stimulus that children mediate

(Schramm et al., 1965). The concept of mediational representation is one of several theories developed to interpret perceptual processes in children. Data processing by children is a result of ex­ ternal information input and subsequent overt responses to that data. This idea criticizes Jean Piaget's theory which is more structural than dynamic (Calder, Robertson, &

Rossiter, 1975) . Piaget's theory of perceptual boundedness "refers to the tendency for children to focus immediately on perceived stimulus features rather than on more delayed or abstract features" (Calder et al., 1975). Cognitive develop­ mental stages impose limits on the amount and type of infor­ mation that children are capable of processing at various points of their sequence of mental maturation. Thus, per­ ceptual boundedness is closely related to age (Flavell, 1973) .

However, with this notion, there is no reliance on direct representation of stimulus information. The focus with Piaget's view is rather nonrepresentational which

(schematically) is more biological in structure. Biologi­ cally-based cognitive structures do not really provide ade­ quate explanations of children's information processing 25 (Calder et al. , 1975). The "social learning theory" appears

to be more reasonable since it describes the limited set of

dimensions that a child uses in processing information as

those "with the most salient reinforcing value in terms of

the child's cumulative experience with all the stimulus dimen­

sions presented" (Calder et al. , 1975).

Consumer Information Processing: The Adult Model and the Child Model

Comparison of the adult model for consumer informa­

tion processing (CIP) with that of children resulted in the

following findings:

- Direct motor imitation (of scenes from commer­ cials, from the home, and media in general) does play a relevant role in children's consumer learning.

- Children cannot encode as much as adults. Names and information used in ads must be repeated several times.

- Children's CIP models will surely have to allow for nonlinguistic information representation. This is a less elaborate processing step in which comprehension does not really occur. Familiarity may be more important than compre­ hension .

- Acceptance and product choice for children may be contingent upon attention to isolated bits of information regarding products and attributes. There may be no attempt by children to integrate stored information.

The conclusion here may well be that because of chil­

dren's psycholinguistic comprehension problems and greater de­

pendence on non-linguistic information, frequency and latency measures might be better for children's research than rating 26 scales and multi-category verbal responses scoring (Calder

et al., 1975). Fishbein’s model (1972) of adult consumer

information processing focuses on comprehension and accep­

tance. Here, the attributes of products are treated as a series of beliefs which may have some internal psychological

reality. It presumes linguistic representation (Fishbein,

1972). Furthermore, information processing both in adults

and in children has been analyzed as a multi-staged process

involving attention, comprehension, acceptance, retention, and behavior (McGuire, 1974).

It has already been pointed out that a child's in­

ability to detect information, i.e., his "reaction time," diminishes rapidly with increase in age. However, in trying

to determine what caused "selective" attention, researchers

found that children are quite sensitive to auditory informa­ tion cues in commercials, whicn might compensate for their

inattention to the visual message. Although children have clear visual imagery of such things as cereal brand cartoons, their visual encoded preferences may be quite at variance with verbally indicated preferences (McGuire, 1974).

The preoperational child will exhibit less differ­ entiation in his responses to visual and auditory stimuli

than the concrete operational child since there is decreasing perceptual boundedness (Ward & Wackman,, 1973). Thus, tele­

vision programs and commercials are on the whole quite

similar perceptually in their manner of visual and auditory 27 presentation, though stimuli differ in content. And, the younger child who is more sensitive to the perceptual aspects of stimuli would be more stable in his attention across com­ mercials and programs than the older child (concrete opera­ tional) who is likely to be sensitive to the content aspects of messages as well as the perceptual aspects (Ward & Wack- man, 1973).

In the sensory-motor stage (0-2 years), the infant's behavior is not mediated by thought at all, but rather by set behavior patterns called schema. For example, the infant may see an object, then reach for and grasp it, but he does not

"think" about what he is doing. In the preoperational stage

(2-7 years), the child is developing symbolic abilities, such as language and mental imagery, but his behavior is still very closely linked to perception. Piaget characterizes the mental processes of this stage as a "mental experiment" in which the child's mental life is no more than a replication in mental imagery of various stimuli which often bear no logical relation to each other.

In the concrete operational stage (7-12 years), the child has developed conceptual skills which enable him to effectively mediate perceptual activity, but only when deal­ ing with concrete objects. For example, the child can sort objects into subordinate and superordinate classes and answer questions about the inclusion of one class in another, but only if the objects are visually present. 28 Finally, in the formal operational stage (adolescence through adulthood), the individual develops the ability to mediate perceptual activity with abstract thought. Thus, in this stage, the individual is released from the need for con­ stant perceptual information to direct his behavior and now may be directed by theories, values, and ideas which have no referent in perceptible reality (Wartella & Ettema, 1974).

From this, a developmental continuum emerges: the more perceptually bound the child, the greater the influence of perceptual attributes of stimuli on his behavior. Further­ more, for children younger than six years of age, the greater the complexity of a stimulus, the greater the attention to it

(Wartell & Ettema, 1974).

The concept of auditory imagery is also important in children's consumer behavior. Extensive research has indi­ cated that, despite previous beliefs concerning pupillary dilation, pupils dilate when an individual is paying atten-' tion to what he is viewing, not just when he likes what he sees (Rice, 1974). In fact, preschool children do not even recognize a commercial nor understand that its purpose is to sell a product (Disson, 1973). Children often repeat brand names, slogans, or jingles from advertisements (Rossiter,

1975). These may become imitative stored information which when retrieved from memory may strongly influence children's product preferences even though the information may be mean­ ingless in a comprehensive sense (Wells, 1965). 29 In general, young children are not typically capable

of distinguishing commercials and the economic motive behind

them from ordinary program content (Ward, Wackman, Faber, &

Lesser, 1974). Thus, it is essential to educate children so

that they may be able to separate reality from fantasy and other similar conceptions that attempt to put fantasy on the same plane as reality (Howitt & Cumberbatch, 1974) .

In 1970, the ACT Symposium reported that in Washing­

ton, D.C., 98% of all children’s programs were entertainment

and that 84% consisted of cartoons (Jennings, 1970)* Further­ more, children's programs contained at least twice as many commercials as seen on adult programs during prime time. For example, in November, 1969, on a single Saturday morning, CBS broadcasted 72 minutes of commercials in which 130 individual sales pitches were directed to children. Sixty-three of

these, amounting to 35 half-minutes, consisted of toy com­ mercials. Under pressure, the television industry in 1970 ran advertisements in national magazines boasting about tele­ vision programs that were educational, including showings such as "Save on Your Taxes" and "Sixty Minutes" as "exempli­

fying special attention to children" (Jennings, 1970).

The appropriateness of classifying shows such as these as relevant for children is questionable. Both the public and the broadcasting system must increase their efforts toward improving television to the benefit of children. The

job cannot be left solely to the industry. The public must 30 become part of the process of developing children's television

and must pressure the broadcasting industry to stop imple­

menting television as a means to maximize profits by exploi­

ting the young (Jennings, 1970). This exploitation, for the most part, relies heavily on the child's inability to dis­

tinguish between the "fantasy" of commercials and the "re­

ality" of regular programming. Certain key devices are used by advertisers that may

confuse the child as to what is real and what is fabricated.

Some of these divices may be summarized as follows:

- Advertisers use "figurative distortions" such as large boxes and signs.

- Advertisers rely on the conversational habits of children: such slogans as "whiter than white" or "buy the one with the white tornado."

- Advertisers use misleading impressions, such as, "Wonder Bread helps build bodies in twelve ways." The Federal Trade Commission (FTC) claimed this was misleading since other competitive breads have the same composition as Wonder Bread.

- Advertising is based on adult conversation and perceptual habits, yet presented to young children (Bever, Smith, Bengen, & Johnson, 1975).

It seems that television advertising confronts the

child with a subtle conflict between two ways of processing

information. This conflict deals with Piaget"s distinction between the "operational" and "figurative" aspects of simula­

ting data. The figurative aspects are based on predisposi­

tions to perceive and act in certain ways while the opera­

tional aspects are based on reality testing and logical pro­ cesses that a child can apply to the world in consistent 31 relationship to each other (Sever et al., 1975). The results of this conflict vary with age. Most three-to four-year-olds can distinguish between fantasy and reality but have diffi­ culty articulating it. Five-to six-year-olds have less con­ flict. Since they are too immature, they view all television content as entertainment. "Because they discriminate between figurative impressions so little and make operational judge­ ments about advertising so seldom, television commercials appear to have only peripheral impact on five-and six-year- olds - However, those children ranging between seven and ten years of age show tension and anger over their inability to cope effectively with what they believe to be misleading advertising" (Bever et al., 1975).

Enhancing the Child's Ability to Process Televrsed Consumer Information

One way in which children can begin to learn to dis­ tinguish between fantasy and reality is by being made aware of the techniques employed in advertising and especially in advertising directed toward them. One of the methods employed by advertising is the use of clear, sheer repetition. Repeti­ tion has always played a key role in learning, especially in younger children’s learning (Wartell & Ettema, 1974). Grass,

Winters, and Wallace (1969) examined the short and long-term effects of repeated exposures on attention for a television advertising campaign. Fifteen males and fifteen females were shown a campaign consisting of three commercials in a 32 laboratory setting for each of three monthly sessions.

Similar commercials were shown four times within a session.

The results indicated that attention increases/ reaches a peak, and then decreases. Thus, mass campaign presentation would become a progressively inferior strategy as the cam­ paign grows older.

Other research has shown that attitudes change under conditions of distraction, and that the persuasiveness of television advertising cannot be attributed solely to learn­ ing measured by recall (Bither, 1972). Potential contri­ butors to recall of a television commercial have been cate­ gorized in four communication elements; pictorial value, empathy, interest, and clarity (Haller, 1972). Quantitative judgements, pooled in order to determine how much each ele­ ment would contribute to the ability of the commercial to be remembered twenty-four hours after viewing, confirmed the accuracy of this prevention method.

Other research indicates that the time of day in which interviewing occurs affects recall. Higher recall is obtained with earlier rather than later interviewing follow­ ing previous day commercial exposure. Thus, time lapse following exposure is an important factor to recall (Shulman,

1972). In children's commercials, probably the primary means of insuring recall is through the use of jingles and slogans.

The majority of mothers in the Los Angeles area reported that many of their children could sing television jingles by two 33 years of age and that over 90% could sing them by three years of age (Lyle & Hoffman, 1972).

Another technique used by advertising firms is the amount of commercial time used and the day of the week that most children would be viewing programs. Investigations were conducted by Blackwell and Yawkey (1975) to determine dif­ ferences between the total amount of television programming time devoted to advertising on the three major U.S. networks.

The total amount of commercial time was arbitrarily classi­ fied into categories of: (1) food, (2) sweet-treats, (3) toys, (4) commercial spots, (5) others. Saturday morning children's programming and advertising were the focus of the study. Using chi square experiments, statistically signifi­ cant differences were not observed among the television net­ works on the total amount of Saturday morning programming devoted to advertising for young children. Within the ad­ vertising categories across television networks, however, statistically significant differences were evidenced in food and toy classifications. Thus, the study showed that Satur­ day morning was the best time for children's commercials

(Blackwell, Yawkey, 1975). The development of "KIDVID

GHETTO" on Saturday mornings, for example, was clearly a response to the call for programs to cultivate child audiences that will be receptive to direct, pinpoint advertising

(Melody & Ehrlich, 1974). 34 By routing the audience flow throughout the morning with programs designed specifically to capture and hold the attention of different age groups of children, advertisers have an efficient vehicle for their advertising messages and broadcasters have been able to maximize profits from the sales of advertising during this time period. Saturday morn­ ing represents the best time for advertisers to exploit a more homogeneous submarket of children without paying for a lot of the heterogeneous "waste” circulation encountered during prime time television. The consensus is that weekend children-oriented television is a financial boon to the net­ works which can command top dollar and assure the advertiser that he can reach children at a CPM (cost-per-thousand) which is tremendously efficient (Melody & Ehrlich, 1974). This

Saturday morning monopoly led concerned groups to take some action against commercials viewed by children.

In 1970, the FCC accepted for consideration a peti­ tion for rulemaking filed by the citizens' organization,

ACT, which requested that financial means of funding chil­ dren's programs be investigated so that a new system of sup­ port would be incorporated that would benefit the child rather than the profits of the advertiser. Despite certain changes, problems still remained. In 1974, certain groups, including ACT, came up with a new set of proposals for the advertising code and presented this to the FTC. The FTC, which has the statuatory authority to regulate advertising, 35 advised the industry to regulate itself or face the conse­ quences {Melody & Ehrlich, 1974). The Commission's mandate includes the authority to prevent advertising abuses as well as to seek solutions by litigating cases after the fact. The

FTC is also the recipient of complaints made regarding cer­ tain practices in television advertising to children which are deceptive and unfair (Howard & Hulbert, 197 3).

The Child’s Receptiveness to Television Commercials

The question then arises as to a child's receptive­ ness to television commercials. Rossiter and Robertson

(1974) conducted a study designed to investigate the "per­ suasion processes in children's susceptibility to television ads." This particular piece of research differs from prior studies in that it focused on behavioral impact and not just on cognitive persuadability to behavioral outcome persuasion.

Using 289 primary school boys from five schools within the

Philadelphia-area Catholic school system as subjects, the study found that there was a persuasive cultural influence of advertising on children. To this result, the questions posed by several other studies may be related:

- Do children automatically "tune in" to commercials? (Ward, 1972)

- Do children develop a receptiveness to advertising techniques through exposure to television that could be used to teach subject matter? (Winick et al., 1973) 36 Rossiter and Robertson also reviewed children's levels

of understanding of television commercials and their effects

upon attitudes and purchase request tendencies. The central

idea here, known as the "attribution theory," is that an in­

dividual is motivated "to attain a cognitive mastery of the

casual structure of his environment" through casual relation­

ships. When a child attributes "persuasive intent" to com­ mercials, he believes them less, likes them less, and is less

likely to want to purchase the products advertised (Robertson

& Rossiter, 1974) . The study, involving third grade children, yielded the following results:

- Children who observe commercials designed to induce purchases are older and have parents of higher educational levels and sophistication.

- Children who can recognize commercials as persuasive meet certain criteria:

1. They can distinguish commercials from the program 2. They can recognize a sponsor 3. They perceive the idea for an intended audience 4. They are aware of the symbolic nature of commercials 5. They can cite instances of negative discrepancies where the product does not meet the expectations once it has been purchased.

- Children holding "assistive intent" attributions (advertisements which tell them about something) tend to trust commercials more and therefore like them more.

The findings also indicate that first-grade children

appear to be quite susceptible to these forms of persuasion:

two-thirds indicate they trust all commercials and a half 37 want the things they see advertised (Robertson & Rossiter,

197 5). Advertisers who are well aware that children can influence many types of purchases, control the spending of about $20 billion a year and know it is important to consider children when constructing commercials (Wainwright, 1970).

All this implies that children need to be protected from commercials until such time as they are able to perceive persuasive techniques. Setting limits on the amount of ad­ vertising and the form of advertising would be a step in helping children learn how to differentiate (Robertson &

Rossiter, 1974) , especially since children are enthusiastic and loyal viewers. It has been estimated that 87% of the two-to three- year-olds in the Los Angeles area asked for food seen on television (Lyle & Hoffman, 1972), and that San Francisco children 9-13 years of age spend about $10 a month mainly on toys and sugar-related snacks (Williams, 1974). The latter

finding was the result of a study which also showed that

(1) children's purchases were indeed influenced by television, and that (2) children were skeptical about the credibility of television commercials. This study, conducted by the

Committee on Children's Television (CCT) and five commercial television stations in the San Francisco area, designed and broadcasted television messages for children to help them to develop healthy eating habits and sound consumer skills.

The consumer message team devised by CCT consisted of experts 38 on medicine, nutrition, and child development as well as parents and television producers. The televised series of educational "spots" or "ads" had a sufficiently significant impact on children’s behavior (Williams, 1974). Since children tend to believe what they see, it is essential not only that we have credibility in television advertising but complete truth (Wainwright, 197 0).

A child's purchasing desires are not only influenced by what he sees, but by what he hears as well, especially if what he hears is repeated aloud and often. Behaviorally, if

the auditorily-represented information is repeated by the child, it may suggest a preference to the child's parents

(Wells, 1965). Then, in many instances when children re­ quested that their parents purchase certain products, the

products are bought. However, correlations of such nature

are difficult to establish and the validity of mothers'

subsequent responses regarding the influence of television on their children is debatable (Bechtel, Achelpohl, & Akers,

1972). Other areas related to such studies also have a

present degree of ambiguity.

One such area is the notion of "brand loyalty,"

i.e., are products continually bought because people really

like them or as a result of a commercial? Jacoby and Kyner

(1973) noted that a single unidemensional measure is probably insufficient for evaluating a complex concept such as brand

loyalty. Their six criteria of brand loyalty were, (1) that 39 it is biased, (2) that it results in a purchase, (3) that it is expressed over time, (4) that it is held by a decision­ making unit, (5) that it is held with respect to one or more alternative brands, and (6) that it is a function of psycho­ logical processes. The pilot study used these criteria on

80 six-to nine-year-old children with the product involved being candy bars. The project was subjected to three phases: evaluation, reviewing of a cartoon show, and hypothesis test­ ing. The results provided some understanding of the dynamics of brand loyalty. However, further research was recommended before any definite conclusions were drawn (Jacoby & Kyner,

197 3), especially since commercial communication proceeds on several tracks simultaneously and not along a unidemensional cognitive-affective contimum (Hornik, 1973).

Thus far, the aspects of television ads that have attracted researchers include the content of commercials, children's attention to commercials, their information pro­ cessing of commercials, their attitudes toward television

advertising, and their attempts to influence their parents to purchase the products advertised (Sheikh, Prasad, Kanti

& Rao, 197 4) . While much attention has been focused on the emotional responses of groups such as ACT for more account­

ability in advertising to children, certain studies have been more statistical in nature. One such report submitted by ACT on Boston commercial television studied 1,12 5 min­

utes of Saturday morning children's programs and commercials 40 in May and June, 1971 (Barcus, 1971)* The following statis­

tics were compiled:

- 406 commercials were observed occurring on the average of one every 2.8 minutes.

- 73% had white actors only.

- 24% had a racial mix of actors-

- 3% had only minority actors.

- 49% were enacted by males.

- 16% were enacted by females.

- 35% were enacted by both males and females.

- ads for girls centered on beauty and popularity.

- ads for boys focused on size, power, noise, and speed■

- 24% featured cereals.

- 22% featured toys.

- 22% featured other foods and snacks.

- 20% featured candy and sweets.

- Special visual techniques such as camera angles, close-ups and animation were very apparent.

In still other investigations, negative effects of

television advertising were examined by Feldman and Wolf

(1974). Eight charges made by critics concerning the nega­

tive effects of television advertising directed toward

children were used for one such study. Criteria established

for the study involved stimulus materials, the type of audi­

ence, and the materality of evidence. These were then con­

ceptualized so as to be tested concerning appropriate infor­ mation for the child's confrontations and arguments with 41 parents. The study revealed that it is dangerous to extra­ polate from research in one area (e.g., the effects of vio­ lence in programming) to what might be found in another area

(e.g., the effects of commercials).

Results such as these have urged researchers to exa­ mine all analysis carefully through the use of testing methods based on such statistical measures as reliability, sensitivity, independence of measures, comprehensiveness, relationship to other tests, and acceptability (Plummer, 1972). Compared with other aspects of television, the research com­ pleted on the impact that television commercials have on children is very simplistic and lags behind other areas of investigation. At this point, the findings in this category cannot be used for anything more than generating hypotheses for future study (Sheikh et al. , 1974).

Resistance and Immunization to Television Advertising

Other studies have been conducted to determine if resistance to commercials affects purchasing tendencies.

Findings revealed that resisting attitudes and immunized behavior toward television advertising can be considered obstacles to successful advertising campaigns. This analysis investigated children's resistance to and immunization against behavior on the assumption that psychological characteristics of individuals are often barriers to good commercial communi­ cation (Hornik, 1973). "Children's skill in acquiring 42 impressions of reality from advertising exceeds their skill in understanding them logically" (Bever et al., 1975).

Young children are not uncomfortable with the dis­ parity between these skills and are not upset if advertising is misleading. By middle childhood, however, the conflict becomes so frustrating that the child becomes an absolutist and begins to view all advertising as "a sham" (Bever et al.,

1975). This long-term effect of commercials on children can result in poor consumerism on their part in the future. Ad­ vertising deals with short-range goals (i.e., selling products to a child such as toys that are made to look like steel but in reality are cardboard). When that child becomes an adult, however, he will be bitter, and the long-range effects of commercials may boomerang. Since television is here to stay, we must attempt to deal with it in the best way possible

(Cooney, 1970). Three forces shape the thinking of children: parents, school, and television. Parents and schools view everything they do toward the child with some degree of responsibility; television does not, and it probably will not until there are some economic reasons to do so (Henson, 1970). One economic benefit of children's television is that it is reusable. Therefore, a budget for a long term can pay for good children's television programs. Other means of ensuring good programming for children could be through public or federal funding of the shows. "Sesame Street" is 43 one such program that is sponsored by federal and educational institutions (Henson, 1970). Public television, still in its infant stages, may well be the solution to quality education programming. However, it too needs funding in order to main­ tain this status. Without this funding, it may well be diffi­ cult to reduce the impact that advertising has on children and thus reduce the child's credibility as a future consumer.

Since commercials appeal to non-rational concerns and con­ siderations, the material involved, if non-informational, will raise questions about fundamental fairness and confor­ mity with traditional concepts of advertising which have offered the consumer free choice. Furthermore, exaggerated claims made by commercials may influence the child's learn­ ing of responsibility to himself. He may feel inadequate if he cannot do what the commercial tells him to do or he may want something he does not really need. Thus, commercials create conflicts that may confuse a child’s materialistic values (Winick et al-, 1973) .

Summary

Within this review of the literature, the research pertaining to young children's awareness of television food commercials has been explored. While the research reviewed has dealt with multiple concerns within television adver­ tising, all studies examined have been oriented to adver­ tising directed toward young children. They have been dis­ cussed under the following headings: the nature of the 44 study of the impact of television advertising on children; the physiological and psychological effects of television viewing on children; consumer information processing; and finally, enhancing the child's ability to process televised consumer information. It has become apparent from the body of literature reviewed, however, that very little research exists in the area of a content analysis of commercials aimed at young children. Studies have measured the race and sex of the actors in commercials, the frequency of commercials, and the types of products advertised, but few have investigated exactly what is contained within the commercials. In addi­ tion, there is a small body of knowledge which deals with children's recognition of jingles and cartoon figures, but there has been only limited investigation of the concept of familiarity with specific products, their packaging, and their jingles or slogans. Finally, the question of children's in­ ternalization of commercial messages needs to be expanded so that the interaction between parent and child which results from these commercials can be studied. After reviewing the current literature, it became apparent that this study should focus on the following: a

Content Analysis of food commercials aired during Saturday morning television; the familiarity of young children (ages four and six) with specific products, their packaging, and their jingles; and finally the interaction between parents 45 and their children that results when children are exposed to advertisements for specific cereals. CHAPTER III

METHODS AND PROCEDURES

This chapter will explain the procedures used in the

development of the research instruments, selection of the

sample, collection of the data, and selection of the statis­

tical methods used to treat the data in accordance with the questions posed in the introductory chapter.

Development of Instruments Three instruments were developed, one for each

exploratory study, along with three separate instruments used in an initial pilot study. Since information was taken

from sources such as television commercials themselves and then from young children ages four and six, two of the instruments were designed to be filled out by testers re­ cording the data. The third was a questionnaire which was

given to the parents of the same four- and six-year-olds who participated in the study. A fourth form, the Instru­ ment for Gathering Frequency Data, was devised to facilitate gathering information on how many times commercials were repeated on all three networks. This form was used in

46 47 conjunction with Study One. Experimentation with the three pilot study instruments led to the form of the final instru­ ments used.

Instrument for Gathering Frequency Data

This instrument was devised to provide a form on which data regarding the frequency of the commercials would be recorded. All food commercials that aired on the three major networks (NBC, CBS, ABC) on Saturday, July 3, 1976, were videotaped. This instrument presents a summarized listing of food commercials on all three networks as one index and correlates it to the three networks which form the other index (Appendix B). In this way, it is possible to see whether commercials were repeated on all three networks. The form yielded data which were used for basically two purposes. First the data indicated that cereals were the most advertised food product during the three-hour period of time. This led to the decision to examine only cereals in

Study Two and Study Three. Secondly, the data were later compared with data from Study Two to show relationships between frequency and familiarity. Only the five most familiar cereals-as found in Study Two - were analyzed with chi square analysis. 48

Instrument Analyzing Attributes of Product Data and Sales Persuasion Techniques

The instrument for this study was modified from a study entitled Children's Television Commercials: A Content

Analysis by Winick, Williamson, Chuzmir and Winick (1973).

Modifications were made to answer only those questions to be explored in this study. The Content Analysis form lists all fifty-three commercials aired on three major networks during the three-hour period on July 3, 1976, as one index. The other index is a combination of the five Product Attributes and the five Sales Persuasion Techniques chosen from the

Winick, Williamson, Chuzmir, and Winick (1973) study. The form has spaces where the examiner can record the presence of each of the five Product Attributes and the five Sales

Persuasion Techniques for each product advertised. The five

Product Attributes are (1) sweetness, (2) physical/oral characteristics, (3) nutritional value, (4) product appear­ ance or shape, and (5) the packaging of the product. The

Sales Persuasion Techniques are (1) fun food notion, (2) premium included with the product, (3) indication of having greater physical prowess for having used a certain product,

(4) presence of a cartoon or fantasy character spokesman,

(5) presence of a musical slogan or jingle. The form was filled in with a simple check mark whenever the examiner found the presence of any of these ten items within the food commercials. Totals were found for each of the ten items 49

being investigated in the study, for the five Product

Attributes and the five Sales Persuasion Techniques sepa­

rately, for the number of the ten items present in each

separate food commercial, and finally for the combined total

of all ten items present for each product (Appendix C).

Variability between these totals was measured by use of the

chi square statistic.

Instrument for Study of Children’s Familiarity with Specific Products (Appendix E)

The scope of this study was confined to five cereals.

The five cereals selected represented those which were most

familiar to students in a pilot study conducted on August 18,

1977. This instrument was designed to be used by the exam­ iners to record responses. Subjects, because of their young

age, did not mark anything. The testers administering the

three simple tests of familiarity simply recorded an "X" on

the instrument whenever a correct response was made. The instrument had been devised in such a way as to

have randomized numbers already on the forms before the

children came to each testing station. This allowed the

testers to move the children through each testing station

quickly and efficiently.

Instrument for Gathering Information Regarding Parent/Child Interaction Regarding Five Cereal Products (Appendix F)

The instrument utilized the same five cereal products

that were used in Study Two. Parents were asked to respond 50 to two basic statements: (1) "My child has requested me to buy this cereal as a result of television advertising" and

C2) "I have purchased this cereal as a result of my child's request for it." Parents were asked to simply check the appropriate box next to each of the five cereals, indicating that either one or both of the above statements could apply.

Verification of Instruments

These instruments were analyzed by two statisticians who were acting as advisers regarding the various ways to approach data treatment for the investigation. The instru­ ment for Study Two underwent several changes through a pilot study of twenty-five four- and six-year-olds. Modifications were made in consideration of the following facts: (1)

Since the subjects for Study Two were very young children, the instrument could not be completed by the children them­ selves. All information needed to be recorded by the testers.

C2 > Instrumentation needed to be prepared beforehand to accommodate the randomization technique used. Randomization and construction of five separate testing stations were recognized as solutions to the problem of having a subject see all five cereal products, packaging, and slogans/jingles at one time. It was found in a pilot study that when all five examples of product, packaging, and slogan/jingle were given to children at one time, the children could usually give correct responses for the fourth and fifth items 51 primarily because these could become items that were left over after three responses had been made. Also, it seemed that the children were somewhat overwhelmed by being shown five of anything at once. Randomization seemed to be the solution to the problem. (3) In the pilot study, young children became bored quickly and they fidgeted easily when their short attention spans were challenged. Much greater success was achieved in the final data gathering by having the subjects pass by each of five testing stations where only one product sample, one package, and one slogan were pres­ ented. Children moved very quickly from station to station and seemed to enjoy playing what they were told was a game.

C4) Recognizing that young children have difficulty focusing attention on one visual stimulus at a time, it was decided that when the subject came up to the tester's table, a large cardboard (3ft. wide x 1 ft. high) prevented the subject from seeing what was behind it. Then as each product sample and each package was shown to the subject, it was brought out from behind the cardboard and placed so the child could focus all attention on one singular item. For the test of familiarity with the slogan/jingle, the cardboard was removed, displaying all five packages used for the study. The child pointed to the matching package when the slogan/jingle was played on a small cassette recorder. Because each slogan/ jingle played sixty or ninety seconds, the subjects had 52 plenty of time to look at each package while the slogan/ jingle was played. The instrumentation was set up clearly beforehand with the randomized numbers already on the sheets to make the testing and recording processes as efficient as possible.

The other change in instrumentation that occurred took place on the instrument for Study Three# the survey of parent responses to children's requested cereals. In the first version of this instrument, all fifty-three products advertised throughout the three-hour period were placed on the survey with the same two questions being asked. A pilot study conducted in a shopping center near a supermarket involving thirty mothers who had just completed their mar­ keting revealed that most mothers (86%) could not recall all the names of the fifty-three items; and 8 3% of the mothers interviewed became impatient after about the tenth item on the survey. Because of this experience, the number of cereals was altered to five and it seemed ideal to have the five be the same five that the children were using in the test of familiarity with product, packaging, and slogan/ j ingle.

Study One; Frequency and Content Analysis

This study was designed to produce descriptive statistics. During a three-hour period (8:00-11:00 A.M.) on July 3, 1976, food commercials were video-taped for all 53 three networks (ABC, NBC, CBS) and the commercials were then analyzed. Frequency data were collected first (Appendix B) and then the commercials were analyzed to determine reliance on items of Product Data or Sales Persuasion Techniques

(Appendix C). The Content Analysis was suggested by a larger work, Children's Television Commercials: A Content

Analysis by Winick, Williamson, Chuzmir and Winick (197 3).

Modifications were made to answer the specific questions posed in this study.

From the Winick study, certain items were selected by the investigation in this study. Under the two broad categories of attributes of Product Data and attributes of

Sales Persuasion Techniques, five indices of each were selected. These attributes are described in Appendix A.

Attributes of Product Data Attributes of Sales Persuasion Techniques

Sweetness Fun food/fun notion Physical/oral character­ Premium included with product istics Association with physical Nutritional value prowess Product appearance or shape Cartoon/fantasy character as Packaging of product spokesman Presence of slogan or jingle

Each commercial was analyzed to note the presence of these ten attributes. It was decided that the attributes that appeared most frequently would be the ones used in the test 54

of familiarity with the subjects in Study Two. The profile

shown below indicates the frequencies of each attribute.

It was decided to consider those attributes which earned a

score of either 4 or 5.

Product Data Sales Persuasion

Cereal

Name Slogan Fantasy Premium Cartoon Package or Jingle or Physical Prowess... Fun Food Fun Sweetness Physical Food Shape Food Nutritional Anoearance j j Cheerios ✓ ✓ ✓ ✓ Frosted ✓ ✓ V" ✓ Flakes Lucky ✓ v' •>/ ✓ Charms V ✓ / Raisin s/ v ' Bran Rice ✓ ✓ / Krispies ✓ s

Totals 1 5 4 5 5 2 4 2 t - • I3 3 Profile of Attributes of Product Data and Sales Persuasion Techniques

The objective was to use attributes which were abundantly apparent in the commercials - as determined by the examiner in the Content Analysis - and compare these with subject's familiarity in Study 2 . Once the data were collected in the 55

Content Analysis, a decision had to be reached to determine which of the high-scoring attributes would be used in Study

Two. In the category of Product Data attributes, three top attributes emerged: nutritional value (5), product appear­ ance or shape (4), and packaging (5). Certainly it would be easy to test children's familiarity with the product itself and with the packaging because a child could be shown both of these items. Nutritional value was discarded, however, because it would involve a chemical analysis of the cereals and this would be inappropriate for this study. Therefore, it was decided that in a test of children's familiarity with cereals, from the category of Product Data, the attributes of product shape or appearance and packaging would be measured. In the category of Sales Persuasion Techniques, a similar situation became apparent. The only two items achieving a score of 4 or 5 were fun food/fun notion (5), and presence of slogan or jingle (4). The attribute of fun/ food/fun notion was discarded because it is too abstract an index. No technique could be conceived that would accurately measure children's assessment of "fun" notion. The entire subject of fantasy and creative imagination in young children could be a study in itself. Therefore, the only attribute selected from the category of Sales Persuasion Techniques was presence of a slogan or jingle. 56

These three attributes selected from the Content

Analysis formed the basis for the tests of familiarity in

Study Two. In chapter four where the data will be treated comparisons will be made which show the unique relationships between these three items. The chi square will be used to show these relationships.

Pilot Study for Study Two Study Two as originally presented in the prospectus for this study indicated that fifteen cereals would be used to test children's familiarity. The thought behind this was that exactly fifteen cereal brand names had been presented during the three-hour period of video-taped Saturday morn­ ing commercials. This did not, however, take into account that these were multiple commercials for several of the cereals. One particular cereal like Kellogg's Frosted

Flakes, for example, advertised the same product but with three distinctly different commercials. One focused on inter­ play between a cheerleader and an animated character named

Tony the Tiger, another focused on a dialogue between Tony the Tiger character and his mother, and finally, one focused on Tony the Tiger giving batting instructions to an eight- year-old, aspiring major league baseball batter. The pilot study for this investigation into Children's televised food commercials utilized all three versions of this particular 57 commercial and ultimately, the one selected for use as one of the five cereals in the final study was simply chosen at random.

The fifteen cereals presented to the children in the pilot study were analyzed to determine which were the five most often recognized in all three areas - product sample, packaging, and slogan/jingle (Appendix G). This was done because of the extreme uneasiness that children exhibited in being tested for all fifteen cereals. Even though the fif­ teen were divided so as to test five cereals - in all three dimensions - at each testing station, it seemed too long a time for the children to concentrate. After tabulation of the findings of the pilot study, the five most familiar cereals were chosen to be used in the final study. In this way, it was hoped that the final study would be organized around cereals that would be a fair assessment of children's familiarity.

Randomi z ation

In the pilot study, children were shown individual clear plastic bowls of product sample, and they were then shown individual samples of the cereal boxes. For the testing of familiarity with the jingles, however, the process was slightly different. All five cereal boxes at each of the five testing stations were lined up side by side in a row in front of the subject. Then a series of the five jingles that 58 matched the boxes was played from a tape recorder. As each jingle was played, the child pointed to the matching box.

It was, in most cases, after the third jingle was played that subjects began making comments such as, "It must be one of these two that are left." After the fourth jingle was played, again, many children pointed to the one remaining box as the response to match the fifth jingle even before it was played.

As adjustments were being made in preparation for the final data gathering, it was determined that randomization of the cereals - in all three dimensions - would prevent children from using a process of elimination in their re­ sponses. First, each of the five cereals selected through analyzing the pilot study results was given a number from one through five. The cereal names were first alphabetized and then the numerals were attached.

Cheerios #1 Frosted Flakes #2 Lucky Charms #3 Raisin Bran #4 Rice Krispies #5

Then a table of random numbers was generated (Appendix D) for the numerals one through five to correspond to the idea of having five separate testing stations with just one prod­ uct sample, one example of packaging and one slogan/jingle being presented at each station. The subjects then had the randomized numerals one through five arranged for each of the 59 three areas of familiarity - product, packaging, slogan/ jingle - so that at each testing station the subject had a presentation of only three items - one in each area of famil­ iarity. When all five stations were completed, the subjects had been exposed to all five cereals in all three dimensions

(Appendix E). The same instruments and set of randomized numbers were used for both four-year-olds and six-year-olds tested.

Selection of Subjects

The subjects for studies Two and Three of the final data gathering consisted of thirty-five four-year-old chil­ dren and their parents. Both groups attended Jefferson

Elementary School and Children's Center in Pasadena,

California. Jefferson is located in a low-income neighbor­ hood of Pasadena.

Pasadena is a city of 112,000 people located in the northeastern quadrant of Los Angeles County, California. Primarily a low to middle-class suburb of Los Angeles,

Pasadena is composed mainly of residential housing and small businesses. Pasadena has an unusually large number of pri­ vate elementary and secondary schools which indicates the great interest in education and research. The public school system consists of twenty-eight elementary schools, five junior high schools, and five senior high schools. Jefferson is an urban elementary and day-care complex that accommodates

910 children per day. Located in the central part of 60 Pasadena, it serves primarily a low-income neighborhood near a freeway redevelopment area. Eighty-eight percent of

Jefferson's parent population have annual incomes of less than $8,000 while only 1% of the families earn over $16,000 annually. The geographic area which Jefferson serves con­ sists largely of multi-family dwellings many of which are part of a Federal housing project. It is estimated that approximatley 10% of the student body changes each month because of families moving or entering the area. The racial/ cultural balance of the student body is unique because it consists of 45% Spanish, 27% White, 24% Black, and 4% Oriental and other non-White. Of the 910 children enrolled at the complex, 220 are non-English speaking or limited English speaking. The selection of Jefferson as a setting for gath­ ering data insured the availability of a cross-cultural population of subjects.

The Children's Center of Jefferson Center also pro­ vided an excellent population of four-year-old subjects. The

Children's Center is a service provided for low-income work­ ing parents or for children from homes where only one adult is present. Both the elementary school and the Children's

Center participate in a combination of district, state, and federally-funded programs. All children in the study have access to a television in their homes. 61

Data Collection Procedures

The Pilot Study

The studies of familiarity in Study Two and the determination of parent involvement in children's cereal preferences in Study Three began with a pilot study con­ ducted on August 18, 1977. The pilot study was conducted at

Maryvale, which is an elementary school maintained by the

Catholic Church and the Daughters of Charity. The children who attend Maryvale are in most cases from low to lower middle class families. The children range in age from four through six; the program is primarily day-care and first grade, organized to assist working parents. The children represent a mixture of Spanish- and English-speaking but all understand and speak English fluently. After the cooperation of the director of Maryvale was assured, attention was directed to working with the six testers who would administer the tests of familiarity to the children. Instead of relying on the school's teachers, this investigation utilized adult teachers and high-school age assistants who were known to the designer of this investiga­ tion. The plan for gathering data needed to be rehearsed several times in order to have consistency in the approach used with the children. The practice for this data gathering took place on two separate occasions for an hour each time.

The three adults became the persons to work with the children in soliciting responses while the high-school age students 62 concentrated on recording the data. The children could not see the assistants keeping a record of their responses.

In this pilot study, information about fifteen cere­ als was being solicited. The fifteen cereals were divided into three groups of five and set up in three testing sta­ tions. The cereals chosen for each station were selected on a basis of which ones looked similar - for example, Rice

Krispies looked very much like Frosted Rice — and care was taken to keep such similar looking cereals in different groups. The groupings of cereals included:

Station #1 Group Station #2 Group Station #3 Group

Captain Crunch Peanut Cocoa Puffs Alphabits Butter Cereal Corny Snaps Cheerios Raisin Bran Frosted Rice Frosted Flakes Rice Krispies Honeycomb Grins & Smiles Sugar Smacks Sugar Crisp & Giggles & Laughs Lucky Charms

The director of the school selected four- and six- year-old children for the study who were capable of conducting themselves comfortably in the presence of the six testers all of whom were unfamiliar to the children. The director took care to insure that the children represented both low income families and middle income families. Each child was asked "Do you have a television at your house?" and "Do you watch television on Saturday mornings?" In order to partici­ pate in this pilot study, each child had to answer "yes" to both questions. Once the children were selected in this way, 63

They were brought to a classroom where we had set up the three testing stations utilizing child-size tables and chairs that were familiar to all youngsters in the school. Since the subjects for this investigation were all young children, certain considerations were made for the children.

(1) Children would be asked their names and ages.

C2) Testers would refer to the process as "game," saying, "Would you please play a game with me today?"

(3) The subjects responses needed to be recorded by a tester; children this age could not be relied on for marking anything themselves.

04) Only five items would be presented at each of three testing stations so that subjects would not have to deal with all fifteen items at once.

The actual testing process in the pilot study con­ sisted of five steps at each of the three testing stations.

(1) First, the subject was shown five individual plastic containers of the product sample. Each container was shown one at a time. In each case, the child was asked, "What is the name of this cereal?" Only the correct responses of the exact commercial brand name were recorded. This was considered a pretest to see if the child knew the product before seeing the cereal boxes or hearing the jingle.

(2) Secondly, subjects were shown cereal boxes in exactly the same manner - one at a time - and asked the same, "What is the name of this cereal?"

(3) Thirdly, all five jingles were played consecutively from a tape recorder and subjects were told, "Point to the cereal box that goes with the music you hear." Subjects began to utilize a process of elimination to match the last two jingles with boxes. This was recognized as a problem in collecting accurate data.

(4) Fourthly, subjects were then given the same plastic containers used in Step #1 above - again, one at a time - and asked the same questions again, namely, 64

"What is the name of this cereal?" The idea was that after seeing the packaging and hearing the jingles, subjects might identify the product samples more correctly. What was found, however, was that there was only a 12% improvement in responses overall for both four- and six-year-old subjects with slightly better recognition for the six-year-olds. Four- year-olds' improvement measured 7% and six-year-olds' improvement measured 5%. The indication was that this brief exploration with the children was not very successful as a method of familiarizing the subjects with the cereal products themselves. This was considered a post-test for determining the impact of showing subjects the boxes and then playing the jingles. At no time in the testing were children told if their responses to the packaging or jingles were correct; instead the tester would just say "thank you" as subjects made replies. In situations where subjects remained silent and offered no response at all, testers waited only ten seconds for a response. After ten seconds passed, the tester would say "okay" and move on to the next item.

(5) The fifth and final step in the testing was to put all five plastic containers in a group in front of the five cereal boxes lined up across the table and children were instructed, "Put the cup of cereal in front of the box it came in." The children seemed to enjoy this matching process but again, they would utilize a process of elimination in matching the last two of the five items. When this task was completed, subjects were told they could have one piece of each of the five cereals. Each subject was thanked for playing the "games" and sent to the next testing station.

Results of the Pilot Study

The pilot study described above yielded interesting results. There se med to be very little improvement in the subjects' ability to recognize product sample, even though the product sample was shown twice in steps one and four.

In addition, the problem of working with the process of elimination became apparent in steps three and five. 65

Finally, and perhaps most important, it was determined that these young children ages four and six did not maintain interest long enough to work with five cereals at three testing stations going through five steps at each station.

By the time subjects reached the third testing station, they were fatigued and they fidgeted a great deal. The process was obviously too long and included too many steps. Modifi­ cations were made in the studies and data gathering proce­ dures before final data were collected. Steps in the pilot study were eliminated such as the pretest and post-test measuring familiarity with product sample. Also, matching product samples (Step #5) with cereal boxes was eliminated because it did not seem to be a valid measure of familiarity.

The most important change, however, was an elimination of ten of the cereals from the study for the simple reason that fifteen was too large a number to work with for the children.

Young children experienced difficulty working with this many items. It was decided that the final data gathering would work with only the five most "familiar" cereals (considering three indices, namely, product sample, packaging, and slogan/ jingle). The pilot study showed too, the need for randomi­ zation to prevent children from relying on using a process of elimination. The final data gathering procedure was completely reorganized to reflect information gained from procedures used in the pilot study. 66

Final Data Gathering With reorganized instruments and knowledge gained from working with the pilot study, the final data gathering procedure was established. Data for Study Two were gathered on one day at the Jefferson School complex. A classroom in a section of a new building provided a pleasant setting in which to work. The children at the school had never used the building and they were excited to come to the room. The principal of the school was cooperative and had instructed the teachers that this investigation would be conducted on this particular day.

We arrived at the school at 8:30 a.m. and immediately began a training session for the eleven persons who would administer the tests of familiarity to the subjects. The testers were all high school age students who were simul­ taneously taking part in a two-week course at their high school which explored the subject of child development. It was thought that such an opportunity to work in this data gathering experience would provide information to these high school age students about young children. This heightened sense of awareness on the part of the testers helped them to be conscientious about the data gathering experience. The testers responded positively to the two- hour training session which would insure the uniformity of the investigation. 67

Testers were encouraged to recognize the need for

consistency in conducting the data gathering. Each person

memorized exactly what was to be said to each child, and

this was practiced several times. In addition, since the

testers worked in groups of two, one actually worked with

the child while the other tester made all behind-the-scene

arrangements such as getting the correct sample of the *

cereal product ready to be shown to the subject along with

the correct cereal box and the correct tape-recorded slogan/

jingle. The tester who made these arrangements was given

an instrument (Appendix E) upon which was indicated the

number of the product sample, the packaging, and the slogan/ jingle that was to be played. These random numbers had been

designated earlier by the person conducting the entire

investigation. The testers at each station simply had to

follow the order given on the instrument. During the two-

hour practice session, both testers practiced the process

several times to insure the uniformity of procedure at each

of the five testing stations. Each testing station consisted of a small child's

table plus two small chairs to be used by the testers. One

tester sat on the front side of the table, and the other sat

on an adjacent side at a right angle to the first tester.

This allowed the tester who was working with the child to

simply be given items she needed to conduct the investigation without drawing the subject's attention to the second tester. 68

In the pilot study, a chair had been provided for the subject so that the child would be on a level physically with the tester. It was learned in the pilot study that having the subjects sit on a chair next to the tester seemed to inter­ fere with subjects interacting with the tester who was a stranger. Simply having subjects standing at the table in the final testing seemed to eliminate this problem as illus­ trated below.

Each station was equipped with a three foot wide by one foot high cardboard that separated the subject from the equipment and items being used in the testing. This allowed the subject to concentrate on only one item before him at one time. Behind the cardboard screen, however, were all five boxes of cereal, five samples of the cereal in small plastic bowls, and five cassettes of commercials which corresponded to the five cereals. Each station had one battery operated tape recorder on which to play the five cassettes but only one cassette was played for each subject. 69

In keeping with the randomization described earlier, each cereal product, each box, and each taped slogan/jingle was numbered in this order:

#1 Cheerios #2 Frosted Flakes #3 Lucky Charms #4 Raisin Bran #5 Rice Krispies

This method allowed the second tester, who was preparing the equipment, to arrange items according to the pre-arranged numbers of the data-gathering instrument.

As each subject entered the testing room he was immediately given a very large nametag upon which was writ­ ten both first and last names. Children were then kept in order so that each subject would pass by all five testing stations in the same order. The large nametags allowed two things. First, it helped the tester who worked directly with the subject to call the subject by name as he approached the testing station. This helped overcome some of the natural hesitancy on the part of the subjects. Secondly, the name­ tags helped the tester who arranged the equipment to be certain that she was recording the data by the correct sub­ ject's set of random numbers. The recording tester's first item to write down was the child's name. At the first of the five testing centers, the tester who worked directly with the subject began by first greeting the subject by name and then saying, "I'd like you to play a game with me today. This is a game about cereals, but there are no right or wrong answers. 70

All you must do to play the game is answer a couple of questions. Then when you finish at my table you will go over to another table and the girl there will ask you some questions too. When you finish the questions at all the tables, the game will be over. Now let's begin." At the last of the five tables, the subjects were thanked for playing the game and each was asked to go back to his class­ room. At each of the stations between, the subjects were greeted by name and told, "I have been waiting to play my part of the cereal game with you, (child's name). I'm so glad you're here."

One of the reasons for emphasizing the game idea and the use of the child's first name was based on the assump­ tion that young children can be frightened by a new situa­ tion. The attempt was to make them as secure and as confi­ dent as possible. In addition the hope was to solicit a response from each child so as to minimize the number of non-respondents. The testers had been instructed that if a child hesitated for longer than fifteen seconds, the tester should go on with the next question and ignore the fact that the subject had not responded. Testers were instructed to avoid making the subjects feel they were not responding adequately. At all times the setting was to be reassuring and positive. In this way it was anticipated that young children would respond most comfortably. 71

At each of the five stations, subjects were asked questions in exactly the same manner. When the randomly- numbered plastic container of cereal was placed in front of the child, the child was asked, "What is the name of this cereal?" Subjects were not then told if the response was correct or incorrect, but simply told "okay." Then, when the appropriate cereal box was placed in front of the child, the child was again asked, "What is the name of this cereal?"

For the familiarity of slogan/jingle, the process was slightly different. The carboard barrier was removed from in front of the child, exposing the five cereal boxes used in the investigation. Then the subject was told, "When you hear music from this tape recorder, point to the box that goes with the music." As soon as the child completed these three tests of familiarity, the tester said, "Thank you very much for playing this part of the cereal game with me and now (the next tester's name) would like to play another part of the game with you." And the tester would point the way to the next testing station. No more than five subjects were ever in the testing room at one time which eliminated the problem of having subjects learn how to react from observing others. Also, the initial contact with the children at the first station seemed critical in engaging their positive response to the process. The tester selected for station #1 was chosen because of her unusual ability to make unfamiliar children feel as comfortable as possible in a very short 72 time. All testers were also instructed to sit on children's chairs so that a young child standing next to the tester would be on eye level with the tester. This, it was hoped, would prevent the testers from appearing to be overpowering to the subjects.

The subjects were divided into groups according to age. Subjects ranged in age from four years to four years eleven months; six-year-old subjects ranged from six years to six years eight months. Classroom teachers were advised to recommend only non-reading children to participate in the investigation. Four-year-old subjects were tested first because the process would be somewhat slower due to the subjects' unfamiliar feelings about the testers and the pro­ cess. Also, their exposure to the television commercials and products in question was more limited than the six-year-olds..

After the fourr*year-olds were finished and the six-year- olds began the tests, the pace of the testing increased dramatically. Six-year-old subjects responded more rapidly and, as they encountered successive testing stations after the first one, they had identified the steps of the process and were helpfully responding with occasional phrases like r'I get how to do it."

When all seventy students had passed through the five- station process and had returned to their classrooms, the completed instruments were collected and scanned quickly to see that all information was complete. Also, to test the 73 process to see that a child had been exposed to all five products, packaging, and slogan/jingles through randomized assignment, three randomly-selected subjects' names were called and testers at each station commented on the numbered products, packaging, and slogan/jingles that had been shown at each of the five stations. The subjects had definitely seen— through randomly assigned numbers— a different product, a different package, and a different slogan/jingle at each station; yet in passing through all stations, each subject had seen all five of each of the three items of familiarity.

Thus, the data for Study Two were collected quickly and efficiently in a total of about four hours.

The parent questionniare (Appendix F) was distributed to the tested subjects with instructions from the classroom teachers to encourage the subjects to take the forms home and then return them as soon as possible. Of the thirty-five questionnaires sent home to parents of four-year-olds, only eighteen responded (51%), while only thirteen questionnaires were returned by six-year-old' subjects (30%) . Despite urging by classroom teachers and the promise of a reward of a box of raisins for each child who returned a form, only

4 4% of the subjects responded. Factors which might have affected the responses were the families' socio-economic position, level of sophistication, and education. School personnel attested to the fact that school forms sent home 74 requesting personal information of any kind were often ignored by parents. Since subjects' names were written on the questionnaires before they were taken home, however, it is possible to compare a child's record of familiarity with product sample, packaging and slogan/jingle with the same child's response to the questionnaire. Results of these questions will be explained in the treatment of the data in

Chaper IV.

When three weeks had passed and it was certain that as many parent questionnaires had been returned as were ever going to be, a final visit was made to the school. The ten teenage testers had been thanked before and with this last school visit, the principal, the director of the child care center and the classroom teachers were all thanked for their cooperation.

Treatment of the Data

The format of the data collection for the three studies involving children's food commercials enabled totals to be calculated based on the thirty-nine questions asked in

Chapter One for each of the exploratory studies. These questions can be summarized as follows:

A. To what extent do all commercials under study in the Content Analysis vary with respect to:

1. Product Data 2. Sales Persuasion Techniques 3. Product Data and Sales Persuasion Techniques 75

B. To what extent do the three measures of famil­ iarity (product, packaging, and slogan/jingle) of five specific cereal commercials vary:

1. For four-year-olds 2. For six-year-olds 3. Between four-year-olds and six-year-olds 4. For each cereal (a) Four-year-olds (b) Six-year-olds

5. According to presence of ten attributes of Product Data and Sales Persuasion Techniques

C. To what extent do children’s requests for parents to purchase the five cereals under study vary:

1. For four-year-olds 2. For six-year-olds 3. According to Product Data 4. According to Sales Persuasion Techniques

D. To what extent do parents’ purchases of the five cereals as a result of children’s requests vary:

1. For four-year-olds 2. For six-year-olds 3. According to their Product Data attributes 4. According to their Sales Persuasion Techniques 5. According to their particular attributes for each age group Since the data collected for these questions was in the form of frequencies, the chi square was the primary test of significance used to compare the various combinations. The .05 level of significance was employed throughout this research. Responses from four- and six-year-old subjects were not combined in the data analysis because of the pos­ sible violations of the independence assumption which is necessary for the valid use of the chi square statistic.

The various cross variable comparisions for certain studies were made using either 2x2, 2x3, or 2x5 chi square 76 matrices for uncorrelated proportions; and the final analy­ sis of the three studies involving children's food commer­ cials was an overall comparison of the patterns displayed in the category totals for each study. This is presented in Chapter IV of this investigation. In certain areas of the data analysis, differences were not sufficiently appar­ ent to warrant the use of the chi square statistic. In certain such instances, scatter diagrams were employed in order to present the information essential to this study.

Summary This chapter has explained the development of the four instruments used in this research along with the data collection procedures employed in the tests of familiarity of four- and six-year-old subjects. The latter portion of this chapter introduced the statistical analysis technique employed in treating the data with regard to the questions posed by the three studies set forth in Chapter One. CHAPTER IV

PRESENTATION AND ANALYSIS OF THE DATA

The previous chapter has identified the methods and

procedures used in collecting the data for the three studies

on children's food commercials. This chapter will present

and analyze the data. The present chapter consists of three major sections. Part I deals with the Content Analysis of

a series of commercials aimed at children; Part II presents

the results of children's responses with respect to the

three measures of familiarity with advertised cereals, namely,

product, packaging, and slogan/jingle; and Part III presents

the findings of parent-child interaction resulting from the

influence of televised food commercials' on children.

Since the major statistical analysis of these data concerns proportional comparisons, the chi square statistic was employed. The basic equation used for these comparisons was (Blalock, 1960} :

where; fQ = observed value fe = expected value

Each study required a table with expected values that

corresponded to the observed values. These expected values 78

could be computed by multiplying the row total of the ob­

served values by the column total of the observed values and

then dividing the total by the grand total of all the ob­

served values (Blalock, I960). In mathematical terms, this

would yield the following equation:

^ ~ (row total) (column total) grand total

where: i = row j - column

For the particular tests involved in the research, i.e.,

independence in a two-way contingency table, the degrees of

freedom were calculated using this basic formula (Blalock,

1960) :

df = (rows-1) X (columns-1)

Study One: Content Analysis of Commercials

All food commercials aired on the three major net­ works (ABC, NBC, and CBS) during a three-hour period on

Saturday morning, July 3, 1976, in Columbus, Ohio, were

videotaped and analyzed. Based on the Winick study mention­

ed earlier, two distinct aspects of the commercials were

examined: five items of Product Data and five items of

Sales Persuasion Techniques. The questions in Sections IA,

IB, and IC of this study yield interesting findings regard­

ing these two divisions of the Content Analysis. 79

Product Data/Sales Persuasion Techniques - All Commercials

Question IA of Study One asks: "To what extent do the commercials under study differ in their reliance on items of Product Data and Sales Persuasion Techniques?"

Table 1 presents the comparisons for all the commer­ cials under study based on a chi square value of 3.84

(1 df, p / .05). The column on the far right represents the computed chi square value for the 2 x 2 comparison. The value of 1.12 indicates that the overall 2 x 2 comparison is not significant and that meaningful differences do not exist within the given matrix. It can be therefore concluded that in the Product Data/Sales Persuasion Techniques category, no significant variations exist in these items for the total number of commercials under study.

Product Data Only

Question IB asks: "To what extent are there differ­ ences in the use of, or reliance upon, the following Pro­ duct Data items:

A. sweetness B. physical/oral characteristics C. nutritional value D. product appearance or shape E. packaging of product

Table 2 presents the comparisons for all the com­ mercials under study based on a chi square value of 9.49

(df=4, p .05). The value of 59.75*indicates that the

2 x 5 comparison is significant at p ^ .001, far greater Table 1

Study One: Reliance on Attributes of Product Data and Sales Persuasion Techniques for All Commercials

Attributes of Attributes of Product Data Sales Persuasion Techniques Total N % N % N %

Yes 149 48.06 161 51. 94 310 100

No 116 52.73 104 47.27 220 100

Totals 265 50.0 265 50.0 530 100 Table 2

Study One: Use of Product Data Attributes in All Coiranercials

______Chi square=59.75* Physical/oral Nutritional Product Appearance Sweetness Characteristics Value or Shape Packaging Totals N% N%N%N% N%N%

Yes 15 9.74 35 22.73 28 18.18 24 15.58 52 33. 77 154 100

No 38 34 .23 18 16.22 25 22.52 29 26.13 1 1.0 111 100

Totals 53 20.0 53 20.0 53 20.0 53 20 .0 53 2.0 265 100

*p < .05 df = 4

03 82 than that required for significance in this research, and that, therefore, meaningful differences exist somewhere with­ in the matrix. For example, from the percentages included in the table, 9.74% of the total "yes" responses were in the area of sweetness compared with 33.77% "yes" responses in the area of packaging. From this analysis, it can be seen that in the Product Data category, significant variations occur in the use of these five attributes in the commercials under study.

As has been pointed out earlier in this investigation, it was because of the data presented in Table 2 that the two attributes of product appearance or shape and packaging were selected for use in the investigation of familiarity in'

Study Two. Of the five attributes, it was clear that the attribute showing highest significance was packaging, with

52 of 53 commercials utilizing the exact picture of the packaged product in the test. The two attributes having the next highest significance, physical/oral characteristics at

35 of 53 and nutritional value at 28 out of 53, had to be rejected because neither attribute could be measured in a test of familiarity with children. The subject matter would be too subjective to be measured, and young children ages

four and six would have difficulty relating to questions about either attribute. The next highest attribute, pro­ duct appearance or shape at 24 of 53 possible commercials, was selected as the next significantly high attribute to measure in the tests of familiarity in Study Two. 83

Sales Persuasion Techniques Only

Question IC asks: "To what extent are there dif­ ferences in the use of, or reliance upon, the following

Sales Persuasion Technique:

A. fun food/fun notion B. a premium included with the product C. association with physical prowess D. cartoon fantasy character as spokesman E. use of a slogan or musical jingle" Table 3, given as a 2 x 5 matrix, presents this com­ parison for all commercials under study based on a chi square value of 9.49 (df = 4 , p ^ .05). The analysis yielded a rather large value of 140.96^; significant at p ( .001, again far greater than that required in this research. It is in­ teresting to note that in the category of Sales Persuasion

Techniques all the commercials under study reported the fun food notion as a technique used in television commercials

(32.72%). Not far behind were the reliance upon a slogan or jingle (28.4%) and the use of a cartoon fantasy character

(26.54%). It is evident from the percentages listed in the table that the total "yes" responses of both the premiums included (3.08%) and the association with physical prowess

(9.26%) lagged behind. Therefore, it can be concluded that in the Sales Persuasion Technique category, significant variations exist in the use of these five items in commer­ cials directed at children. Table 3

Study One: Use of Sales Persuasion Technique Attributes in All Commercials

Chi square = 140.96*

Fun Food/ Premium Physical Fantasy Slogan/ Fun Notion Included Prowess Character Jingle Total N % N % N % N % N % N %

Yes 53 32.72 5 3.09 15 9.26 43 26.54 46 29.63 162 100

No 0 0 48 46.60 38 36.89 10 9.71 7 6.80 103 100

Total 53 20.0 53 20.0 53 20.0 53 20.0 53 20.0 265 100

* p < .05 df = 4 85

Further, the analysis of attributes of Sales Persua­ sion Techniques was useful in selecting the attributes to be used for the Study Two test of familiarity. The attribute having the highest rating, namely, fun food/fun notion at

53 out of 53, was rejected because it would not be measured by young children's responses. The concept of "fun" is simply too subjective for young children's experience base.

The attribute having the next highest rating, then, is slogan or jingle with a score of 46 out of a possible 53.

Summary - Study One

The analysis of the data for Study One shows that while there is little variability between reliance on attri­ butes of Product Data and Sales Persuasion Techniques— in fact, they are equal— nevertheless, there is significance of variability within both categories. It is primarily this variation within each category that was employed in the selection of the three attributes to be used in the famili­ arity investigation of Study Two.

Study Two: Children’s Familiarity with Commercials

Two attributes of Product Data, product appearance and, packaging, and one attribute of Sales Persuasion Tech­ nique, of slogan/jingle, were selected from the Content

Analysis of Study One. A process was designed whereby young children ages four and six would be asked to identify 36 a sample of five cereal products in terms of product appear­ ance or shape, the packaging of the five cereals, and finally, the slogan/jingle related to the five items. The analysis of the data yielded interesting information about young chil­ dren's familiarity with these three aspects of television cereal commercials.

Section 1: Familiarity of Product, Packaging and Slogan/ Jingle: Four-and-Six-Year Old Subjects

Product Appearance Familiarity

Question IIAl asks: "To what extent do the measures of familiarity of the five cereal products vary according to product appearance?" Since groups of four-year-olds and six-year-olds were used in this study, comparisons of each group as a separate entity were essential. Combining both groups would violate the independence assumption necessary for the valid use of the chi square statistic.

The five cereals which proved to be of interest in

Study Two were determined in a pilot study conducted at an earlier date which made use of all fifteen cereals. Table 4 presents a 2 x 5 matrix of responses made by four-year-old children to product appearance familiarity. The computed chi square value of 9.70*attained the necessary 9.49 value needed for significance at the .05 level. It can be con­ cluded, therefore, that the five cereals did vary in measure of familiarity of product appearance for four-year-old children. Table 4

Study Two: Four-Year-Old Subjects' Familiarity with Product Appearance

Chi square = 9.700*

Frosted Lucky Raisin Rice Cheerios Flakes Charms Bran Krispies Total N % N % N % N % N % N %

Yes 14 35.90 5 12.82 6 15 .38 9 23.08 * 12.82 39 100 No 21 15.44 30 22.06 29 21. 32 26 19.12 30 22.06 136 100

Totals 35 20.0 35 20.0 35 20.0 35 20.0 35 20.0 175 100

* p < .05 df = 4

00 -J 88

Even more significant was the resulting chi square value for six-year-old subjects when given the same test

(Table 5). The value of 27.53* far above the .05 level,

registers significance at p ^ .001 and thus establishes that variations do occur in the measure of familiarity of product appearance for six-year-old children. Both groups responded

favorably to Cheerios as a product shape that was more easily recognizable (and therefore more familiar) versus

Frosted Flakes and Rice Krispies, both of which shared the position of being least familiar to the subjects. Raisin

Bran rated second place and Lucky Charms rated third place

for both age groups.

Packaging Familiarity Question IIA2 asks: "To what extent do the measures of familiarity of five cereal products vary according to package?" Table 6 presents a 2 x 5 matrix of responses made by four-year-old subjects to packaging familiarity.

The chi square value of 17.11*was significant in light of the fact that it resulted in a larger value than that of

9.49 needed for the .05 level of significance with four degrees of freedom. This value of 17.11* proved to be signi­ ficant at p ^ .01. This indicates that the five cereal products did vary in measure of packaging familiarity for four-year-old subjects. Table 5

Study Two: Six-Year-Old Subjects' Familiarity with Product Appearance

Chi square=27.53*

Frosted Lucky Raisin Rice Cheerios Flakes Charms Bran Krispies Total N % N % N % N % N % N %

Yes 19 36.54 3 5.77 10 19.23 16 30.77 4 7.69 52 100

No 16 13.01 32 26.02 25 20.32 19 15.45 31 25.20 123 100

Totals 35 20.0 35 20.0 35 20.0 35 20.0 35 20.0 175 100

* p < .05 df = 4

1003 Table 6

Study Two: Four-Year-Old Subjects' Familiarity with Packaging

______Chi sguare-»17 .11*

Frosted Lucky Raisin Rice Cheerios Flakes Charms Bran Krispies Totals N %N % N% N %N % N%

Yes 11 37.93 1 3.45 5 17.24 10 34.48 2 6.90 29 100

No 24 16.44 34 23.29 30 20.55 25 17.12 33 22.60 146 100

Totals 35 20.0 35 20.0 35 20.0 35 20.0 35 20.0 175 100

* p < .05 df = 4

VO o 91

Similiarily, Table 7 presents the same analysis for six-year-olds. The computed chi square value of 25.8 3*re­ presents significance at p ^ .001 and yields the conclusion that package recognition as a measure of familiarity for six- year-old-children does vary with respect to cereals. It is interesting to note that once again, Cheerios and Raisin

Bran were the two highest recognized cereals both for four- year-old and six-year-old subjects; and Frosted Flakes was the least-recognized cereal. From a merely speculative point of view, it appears that in both of these tests, that is, product appearance and package recognition, familiarity does not seem to be dependent upon age.

Slogan/Jingle Familiarity

Question IIA3 asks: "To what extent do the measures of familiarity of five cereal products vary according to slogan/j ingle?"

Table 8 presents a 2 x 5 matrix of responses made by four-year-old subjects to slogan/jingle familiarity. The value of 15.32*indicates that significance was attained at p ^ .01, which is a higher level than that needed in the study. Unlike the value obtained for four-year-old children, the chi square statistic for the same test for six-year-old children did not yield conclusive results (Table 9). In this analysis, the value of 2.16 did not meet the require­ ments of 9.49 for a .05 level of significance. Thus, Table 7

Study Two: Six-Year-Old Subjects' Familiarity with Packaging

Chi square = 25.83*

Frosted Lucky Raisin Rice Cheerios Flakes Charms Bran Krispies Totals N% N %N %N% N % N%

Yes 26 32.10 8 9.88 13 16.05 22 27.16 12 14.81 81 100

No 9 9.57 27 28.72 22 23.41 13 13.83 23 24.47 94 100

Totals 35 20.0 35 20.0 35 20.0 35 20.0 35 20.0 175 100

* p < .05 df = 4

VO to Table 8

Study Two: Four-Year-Old Subjects' Familiarity with Slogan/Jingle

Chi square = 15.32*

Frosted Lucky Raisin Rice Cheerios Flakes Charms Bran Krispies Totals N% N % N% N%N% N %

Yes 21 17.21 26 21.32 28 22.95 30 24.59 17 13.93 122 100

No 14 26.42 9 7.38 7 13.21 5 9.43 18 33.96 53 100

Totals 35 20 .0 35 20.0 35 20.0 35 20.0 35 20.0 175 100

P < .05 df = 4

VO LJ Table 9

Study Two: Six-Year-Old Subjects' Familiarity with Slogan/Jingle

Frosted Lucky Raisin Rice Cheerios Flakes Charms Bran Krispies Totals N % N % N % N % N % N %

Yes 31 19.13 33 20.37 32 19,75 34 21.00 32 19.75 162 100

No 4 30.77 2 15.38 3 23.08 1 7.69 3 23.08 13 100

Totals 35 20.0 35 20.0 35 20.0 35 20.0 35 20.0 175 100

\o although measures of familiarity of the five cereal products varied according to slogan or jingle for four-year-olds, no similar conclusion could be made for six-year-olds.

The consistently high familiarity factor as tabulated in Table 9, by itself, supports the null hypothesis that the five cereal products do not vary in measure of familiarity of jingle for six-year-old children.

Section II: Measures of Familiarity Within Each Cereal

For Four-Year-Old Subjects Question IIA4a asks: "To what extent do the measures of familiarity of the five cereal products vary for each cereal for four-year-old subjects?" Tables 10 - 14 present the comparisons for each cereal product based on a chi square value of 5.99 (df = 2,

The computed values for the 2 x 3 matricies of the five cereals are as follows:

Cereal Name Chi Square Probability

Cheerios 6 .11* p < .05 Frosted Flakes 48.63* p < .001 Lucky Charms 40 .23* p< .001 Raisin Bran 32.22* p< .001 Rice Krispies 20.42* p< .001 These values indicate that each of the comparisons is significant and that meaningful differences exist within each cereal with respect to the three measures of familiarity

{at the .05 level of significance) for four-year-old sub­ jects. It is interesting to note that Frosted Flakes had Table 10

Study Two: Four-Year-Old Subjects' Familiarity with Cheerios Cereal in Three Attributes of Product Appearance, Packaging, and Slogan/Jingle

Chi square = 6.113*

Product Appearance Packaging Slogan/Jingle Totals N % N % N % N %

Yes 14 30.43 11 23.91 21 45.66 46 100

No 21 35 .59 24 40,68 14 23.7 59 100

Totals 35 33.33 35 33. 33 35 33.34 104 100

*p < .05 df = 2

VO CTi PLEASE NOTE: F i 1 med as received, without page 97- Not available for microfilming.

UNI VERS ITY MlCROFt LMS Table 12

Study Two: Four-Year-Old Subjects' Familiarity with Lucky Charms Cereal in Three Attributes of Product Appearance, Packaging, and Slogan/Jingle

Chi square = 40.23*

Product Appearance Packaging Slogan/Jingle Totals N%N%N % N %

Yes 6 15.38 5 12.82 28 71.79 39 100

No 29 43.94 30 45.45 7 10.60 66 100

Totals 35 33.33 35 33.33 35 33.34 105 100

*p < .05 df = 2

ID 00 Table 13

Study Two: Four-Year-Old Subjects' Familiarity with Raisin Bran Cereal in Three Attributes of Product Appearance, Packaging and Slogan/Jingle

Chi square = 32.22*

Product Appearance Packaging Slogan/Jingle Totals N % N % N % N %

Yes 9 18.37 10 20.41 30 61.22 49 100

No 26 46.43 25 44.64 5 8.93 56 100

Totals 35 33.33 35 33. 33 35 33.34 105 100

*p < .05 df = 2 Table 14

Study Two: Four-Year-Old Subjects' Familiarity with Rice Krispies Cereal in Three Attributes of Product Appearance, Packaging and Slogan/Jingle

Chi square = 20.42*

Product Appearance Packaging Slogan/Jingle Totals N % N % N % N %

Yes 5 20.83 2 8.33 17 70.83 24 100

No 30 37.04 33 40.74 18 22.22 81 100

Totals 35 33.33 35 33. 33 35 33.34 105 100

*p .05 df = 2 100 101 the highest chi square value indicating that a higher degree of difference for the three measures of familiarity existed with that product.

The data, even without chi square analysis, support the conclusion that meaningful differences exist within # each cereal with respect to the three measures of familiarity because the percentage for product appearance (15.6 2%) and package familiarity (3.13%) were considerably lower than that for slogan/jingle familiarity (81.25%).

Similarly, slogan/jingle familiarity in all cereals ranked higher in percent than the other two attributes:

Cereal Name Product Appearance Packaging Slogan or Jingle

Cheerios 30.43 23.91 45.66 Frosted Flakes 15.63 3.13 81.25 Lucky Charms 15.38 12.82 71.79 Raisin Bran 18.37 20.41 61.22 Rice Krispies 20.83 83.33 70.83

For Six-Year-Old Subjects

Question IIA4b asks: "To what extent do the measures of familiarity of the five cereal products vary among each cereal for six-year-olds?"

Tables 15 - 19 present the comparisons for each cereal product based on a chi square value of 5.99 (df = 2, p ^ .0 5). The computed values for the 2 x 3 matrix of the five cereals are as follows: Table 15

Study Two: Six-Year-Old Subjects' Familiarity with Cheerios Cereal in Three Attributes of Product Appearance, Packaging, and Slogan/Jingle

Chi square = 9.56*

Product Appearance Packaging Slogan/Jingle Totals N % N % N % N %

Yes 19 25.00 26 34.21 31 40.79 76 100

No 16 55.17 9 31.03 4 13.80 29 100

Totals 35 33.33 35 33.33 35 33.34 105 100

*p <, .05 df = 2

o ro Table 16

Study Two: Six-Year-Old Subjects1 Familiarity with Frosted Flakes Cereal in Three Attributes of Product Appearance, Packaging and Slogan/Jingle

Chi square = 60.63

Product Appearance Packaging Slogan/Jingle Totals N % N % N % N %

Yes 3 6.81 8 18.19 33 75.00 44 100

No 32 52.46 27 44.26 2 3.28 61 100

Totals 35 33.33 35 33.33 35 33.34 105 100

*p < .05 df = 2

o u> Table 17

Study Two: Six-Year-Old Subjects' Familiarity with Lucky Charms Cereal in Three Attributes of Product Appearance, Packaging, and Slogan/Jingle

Chi square = 32.61*

Product Appearance Packaging Slogan/Jingle Totals N % N % N % N %

Yes 10 18.18 13 23.64 32 58.18 55 100

No 25 50.00 22 44.00 3 6.00 50 100

Totals 35 33.33 35 33.33 35 33.34 105 100

*p < .05 df = 2

i Table 18

Study Two: Six-Year-Old Subjects' Familiarity with Raisin Bran Cereal in Three Attributes of Product Appearance, Packaging, and Slogan/Jingle

Chi square = 22.27*

Product Appearance Packaging Slogan/Jingle Totals N % N % N % N %

Yes 16 22.22 22 30.56 34 47.22 72 100

No 19 57.58 13 39.39 1 3 03 33 100

Totals 35 33. 33 35 33.33 35 33.34 105 100

*p < .05 df = 2

o Table 19

Study Two: Six-Year-Old Subjects' Familiarity with Rice Krispies Cereal in Three Attributes of Product Appearance, Packaging, and Slogan/Jingle

Chi square = 47.89*

Product Appearance Packaging Slogan/Jingle Totals N % N % N % N %

Yes 4 8.33 12 25.00 32 66.67 48 100

No 31 54.39 23 40.35 3 5.26 57 100

Totals 35 33.33 35 33.33 35 33.34 105 100

*p < .05 df - 2

o 107

Cereal Name Chi Square Probability

Cheerios 9.56* p < .01 Frosted Flakes 60.63* p < .001 Lucky Charms 32.61* p<.001 Raisin Bran 22.27* p < .001 Rice Krispies 47.89* p < .001

These values indicate that each of the comparisons is signi­

ficant and that meaningful differences exist within each cereal, and the measures of familiarity for six-year-old children will be beyond the .05 level of significance.

Once again, Frosted Flakes had a higher chi square value than the other cereals. And, as in the previous

analysis, jingle familiarity stands out (in terms of per­ centage) with respect to the other attributes:

Cereal Name Product Appearance Packaging Slogan or Jingle

Cheerios 25.00 34 .21 40.79 Frosted Flakes 6 .81 18 .19 75.0 Lucky Charms 18 .18 23 . 64 58.18 Raisin Bran 22 .22 30 .56 47.22 Rice Krispies 8.33 25 .00 66.67

It appears that the attribute of slogan/jingle in the

Sales Persuasion Technique of television commercials is wide­ ly recognized and familiar among four-and six-year-old chil­ dren .

Section III: Comparison: Four-and Six-Year-Old Responses to Attributes of Product Data and Sales Persuasion Techniques

Response to Product Appearance

Question IIB1 asks: To what extent do four- and six- year-old children's responses differ according to product appearance?" 108 Table 20 presents a 2 x 2 matrix comparing the re­ sponses between four-year-old and six-year-old children’s

familiarity with product appearance. The value of 2.51 in­ dicates that the overall comparison is not significant at

the .05 level and that meaningful differences do not exist within the given matrix. The responses of the two age groups do not differ in the measure of familiarity of pro­ duct appearance.

Responses to Packaging

Question IIB2 asks: "To what extent do four- and six-year-old children's responses to package familiarity differ?"

Table 21 presents data that yields a chi square value of 35.85? a value greater than that needed to achieve the

.05 level of significant at p ^.001. This indicates that children's responses to package familiarity differ between the two age groups.

Responses to Slogan/Jingle Question IIB3 asks: "To what extent do four- and six-year-old children’s responses differ in relation to slogan/jingle recognition?"

Table 22 presents the comparisons for slogan/jingle

familiarity between the two age groups based on a chi square

value of 3.84 (df = 1, p ^ .05). The value of 29.88* indicates that the 2 x 2 comparison is significant at Table 20

Study Two: Comparison of Familiarity with Product Appearance Between Four-and Six- Year-Old Subjects

Four-Year-Old Subjects Six-Year-Old Subjects Total N % N % N %

Yes 39 42.86 52 51.14 91 100

No 136 52.51 123 47.49 259 100

Totals 175 50.0 175 50.0 350 100

i-* o Table 21

Study Two: Comparison of Familiarity with Packaging Between Four- and Six-Year- Old Subjects

Chi square = 35.85*

Four-Year-Old Subjects Six-Year-Old Subjects Total N % N % N %

Yes 29 26 . 36 81 73.64 110 100

No 146 60.83 94 39.17 240 100

Totals 175 50 .0 175 50 .0 350 100

*p < .05 df = 1 110 Table 22

Study Two: Comparison of Familiarity with Slogan/Jingle Between Four- and Six- Year-Old Subjects

Chi square = 29.88*

Four-Year-Old Subjects Six-Year-Old Subjects Total N % N % N %

Yes 122 42.96 162 57.04 284 100

No 53 80.30 13 19.70 66 100

Totals 175 50 .0 175 50 .0 350 100

*p < .05 df = 1 111 112 p < .001, which is far greater than that required for signi­ ficance within this research, and that meaningful differ­ ences exist somewhere within the matrix. This concurs with the earlier test on the familiarity of slogan/jingle which indicates that slogan/jingle recognition for the five cereals did not vary among six-year-old children, but did vary among four-year-old children.

Section IV: Comparison: Measures of Familiarity with Attri­ butes of Product "Data and Sales' Persuasion ~ Techniques

Questions IIC1 and 2 ask: "To what extent do the three measures of familiarity in this study (product appear­ ance, packaging, and slogan/jingle) relate to the attributes of Product Data and Sales Persuasion Techniques from Study

One :

A. sweetness A. fun food/fun notion B. physical/oral B. premium included characteristics C. physical prowess C. nutritional value D. fantasy character D. product appearance E. slogan/jingle E. packaging

The chi square statistic was not employed for this comparison since the commercials involved were not different enough to be able to discriminate between familiarity

(Study Two) and attributes of Product Data and Sales Per­ suasion Techniques (Study One). Thus, the use of a scatter diagram seemed more feasible as a means of comparison

(Figure 2). Familiarity Score per Cereal 100 110 ScatterDiagram of Familiarityvs. AttributesofProduct Eachcereal ahad possible attribute score 10 of Eachcerealhad apossible recognition score 105 of Four-year-oldchildren's familiarity score percereal o Six-year-oldchildren's familiarity score cereal per * 10 20 60 30 40 50 70 80 90 ! 1 1 1 — r i (5ProductData 5 and Sales Persuasion Techniques) i i 1 3 sbet, ttlatiue sd 3 3 35105)x(35= subjects, 3 total attributes used: i i i l 1 1 j i | : t L i ! ! : ! ■ j i i t ; r 2 1 TotalNumberof Combined Attributes PerCereal DataandSales Persuasion Techniques ! 1 , , I I l ; | ; * ’ ; t 1 1 i i i i ! i 1 . i 1 1 3456789 i i iue 2 Figure _ i . i . | i i i t i - - - - < < 9 ( 1 ►

■ c 0 t i l 7 i r 1 i i j ( ) j i 10 113

114

From this diagram, it can be ascertained that two cereals had a total of six attributes, another two had seven attributes, and only one of the five cereals had eight attri­ butes . In all but one case, familiarity was higher for the six-year-old subjects (*) than for the four-year-old sub­ jects (o), and approximately the same for the remaining cereals. According to the scatter diagram, there does not appear to be any direct correlation between number of attri­ butes of Product Data and Sales Persuasion Techniques in a cereal, and a child's familiarity with that cereal. In other words, a higher number of attributes does not neces­ sarily yield a higher level of familiarity.

Summary - Study Two Data from Study Two have been utilized to analyze the concept of children's familiarity with televised cereal commercials. The data have indicated that there is vari­ ability among the three indices of familiarity (product appearance, packaging, and slogan/jingle), as well as be­ tween the responses of four- and six-year-old subjects.

Finally, the data do not indicate that there is correlation between the three measures of familiarity and attributes of

Product Data and Sales Persuasion Techniques.

Study Three: Parent-Child Interaction

Study Three is an analysis of data gained from in­ vestigating parent-child interaction resulting from televised 115 cereal commercials. In general. Study Three examines parents' responses to two basic statements, namely, "My child has requested me to buy this cereal" and "I have purchased this cereal as a result of my child's request for it." The par­ ents used for this study were the same parents of the four- and six-year-old subjects used in Study Two. Furthermore, parent responses will be compared with attributes of Product

Data and Sales Persuasion Techniques of Study One.

Section I: Children's Requests for Parents to Purchase Five Cereal Products

Four-Year-Old Children's Requests. Question IIIAl in Study Three asks: "To what extent do children's requests for parents to purchase products as a result of television advertising vary for four-year-old children?"

Table 2 3 presents the comparisons for the five cereals versus children's requests of parents to purchase products, based on a chi square value of 9.49 (df = 4, p ^ .05). The value of 8.06 indicates that significance was not attained at the .05 level. Thus, it can be concluded that the pro­ bability of four-year-old children's requests for parents to purchase the five cereals does not vary among cereal products.

Six-Year-Old Children's Requests. Question IIIA2 asks: "To what extent do children’s requests for parents to purchase products as a result of television advertising vary for six-year-old children?" Table 23

Study Three: Four-Year-Old Children's Requests for Parents to Purchase Cereals

Questionnaire: "My Frosted Lucky Raisin Rice child has requested Cheerios Flakes Charms Bran Krispies Total me to buy this cereal." N % N % N % N % N % N %

Yes 12 22 .22 8 14 .81 15 27.78 11 20 . 37 8 14.81 54 100

No 6 16.67 10 27.78 3 8.33 7 19.44 10 27 .78 36 100

Totals 18 20.00 18 20.00 18 20.00 18 20.00 18 20.00 90 100

CTi 117

Unlike the previous results, significance at the .05 level was reached in the similar study made for six-year-old children {Table 24). Here, the chi square value of 15.52*was large enough for rejection of the null hypothesis. Thus, the probability of parents of six-year-olds purchasing cereals as a result of children’s requests to do so, did vary among cereal products. Section II: Parents' Purchasing of Cereals as a Result of Children’s Requests

For Parents of Four-Year-Olds. Question IIIB1 asks:

"To what extent do parents' purchases of products as a re­ sult of children's requests vary among parents of four-year- olds?" Table 25 shows the comparisons for this study. The chi square value of 9.49 {df = 4, p ^ .05) was not attained.

The value of 6.57 indicates that the five cereal products do not vary with parents' purchases as a result of four-year- old children's requests.

For Parents of Six-Year-olds. Question IIIB2 asks :

"To what extent do parents' purchases of products as a re­ sult of children's requests vary among parents of six-year- olds?"

The 2 x 5 matrix in Table 26 presents this compari­ son. Once again the .05 level of significance was not reached since the chi square value was only 5.58. There­ fore, the five cereals do not vary with parents’ purchases as a result of six-year-old children’s requests. Table 24

Study Three: Six-Year-Old Children’s Requests for Parents to Purchase Cereals

Chi square - 15.52*

Questionnaire: "My Frosted Lucky Raisin Rice child has requested Cheerios Flakes Charms Bran Krispies Total me to buy this cereal." N % N % N % N % N % N %

Yes 11 25.58 5 11.63 10 23.26 12 27.90 5 11.63 43 100

No 90.90 8 36.36 3 13.64 4.54 8 36.36 22 100

Totals 13 20.00 13 20.00 13 20.00 13 20.00 13 20.00 65 100

*p < .05 df = 4 Table 25

Study Three: Parents of Four-Year-Old Children Purchasing of Cereals as Result of Children's Requests

Questionnaire: “I have purchased this cereal as Frosted Lucky Raisin Rice a result of my child’s Cheerios Flakes Charms Bran Krispies Totals request for it." N % N %N%N % N% N %

Yes 10 24.39 8 19 .51 12 29.26 6 14.63 5 12.20 41 100

No 2 15.38 0 0 3 23.07 5 38.47 3 23.07 13 100

Totals 12 22.22 8 14.81 15 27.78 11 20. 37 8 14.81 54 100 Table 26

Study Three: Parents of Six-Year-Old Children Purchasing of Cereals as Result of Children's Requests

Questionnaire: "I have purchased this cereal as Frosted Lucky Raisin Rice a result of my child's re- Cheerios Flakes Charms Bran Krispies Totals quest for it." N %N%N%N%N%N %

Yes 11 31.42 3 8.57 8 22.86 10 28. 57 3 8.57 35 100

No 0 0 2 .25 2 .25 2 .25 2 .25 8 100

Totals 11 25.58 5 11.62 10 23.26 12 27.90 5 11.62 43 100 121

Section III: Relationship Between Attributes of Product Data and Sales Persuasion Techniques and Cereal Purchases Made by Parents of Four- and Six- Year-Old Subjects

For Parents of Four-Year-Olds. Question IIIC1 of

Study Three asks: "To what extent are the attributes of

Product Data and Sales Persuasion Techniques related to

parents' responses to their four-year-old children's re­ quests to purchase certain cereals?"

This particular comparison could not make use of the

chi square statistic since the commercials involved were not

different enough to be able to discriminate between parents'

responses and attributes of Produce Data and Sales Persua­

sion Techniques. Scatter diagrams were employed as a means

of comparison (Figures 3 and 4). There does not appear

to be a linear correlation between the number of attributes

of Product Data or Sales Persuasion Technique present in

commercials, and parents1 responses to children's requests

to purchase certain cereal products. Figures 5 and 6 re­

present comparisons of parents' purchasing of products and

attributes of Product Data and Sales Persuasion Techniques.

In this comparison, there is a suggestion that the higher

percentage of parents' purchasing products as a result of

their four-year-old children’s requests, correlates to the

higher number of attributes of Product Data and Sales Per­

suasion Techniques. Yet, it is difficult to decide about

the magnitude of the influence of the attributes of Product to to

o o <£} o 00 o -J O It CTl O J I' 0 U1 o -- t \ i - - ib o to o ■ ■* ■ —* -- to o that Certain Cereals be Purchased: has Purchased: be child Cereals "My Certain that requested me to buy this cereal." this buy to me requested — Indication by Parent (%) that Child Has Requested Has Child that (%) Parent by Indication - — - 1 I— to u> lb Ln H 0 CL 0 a rt o &> rt > rt rt tr 0 cn rt a> H- * 1 1 0 ►of 3 l-tl Hi H fD 0 0 HH H a 3 0 H* H a h (D 3 1 0 * * 1 1 - 1 (D

Scatter Diagram of Parents' Responses vs. Product Data 123

01 90 Pi >1 E 01 5 80 t► i T3 t : - r—t a) * i -i“* 01 i-H 1 J3 (d * 4-1 01 0> -H 4 .— . J Q -C dO 4J — • 01 50 i-H >i 4J fO 3 c (1 u 4-> [ & c 01 > | - H S J3 rd 30 4-1 4t 4I C M ai O

o Parents of four-year-old children * Parents of six-year-old children

Figure 4 Scatter Diagram of Parents' Responses vs. Sales Persuasion Techniques Scatter Diagram of Parents' Purchasing vs. Attributes Purchasing Parents'Diagram of Scatter

Parents purchasing: "Yes, I have purchased these <4-1 TJ rH sz + to S-l O’ 3 0) co to 0J S-l d f to fd CO fd u 0) o u u <1> a) j 100 o * 60 70 80 90 10 20 30 40 50 | Parents of six-year-old children six-year-old ofParents children four-year-old of Parents 1 ! 1 I 1 i 2 4 5 4 3 2 1 t i i I 1 of Product DataofProduct Product DataProduct 1 l M i 1 ! i ! i ' iue 5 Figure i i- i t 1 i ; t 1 I l : ■ l : i i i 1 i i ; I i i : I : 1 i j ! i i 1 i : . 1 i < it. it. * I r V 1 w W i i ! a * 1 ! r i i 1 . ' * * 7 A a 1 ■ i | l i ! : \ \ \ I I i

124 ScatterDiagram ofParents' Purchasing vs. Attributes

rents purchasing: "Yes I have purchased these cereals as a result of child's request. 100 Parents four-year-old of children o Parents six-year-oldof children * 20 90 40 50 70 60 30 80 I 1 1 I 1 1 1 1 1 i 1 I i 1 . 2 4 5 4 3 2 1 ofSalesPersuasion i Sales Persuasion ! 1 ! 1 1 1 v ( j 1 Figure 6 j \ h w 4 t % i i Jr J * * 4 > j ) 1

126

Data and Sales Persuasion Techniques on the children's re­ quests or the parents' purchasing of the products.

For Parents of Six-Year-Old Children. Question IIIC2 asks: "To what extent are the attributes of Product Data and Sales Persuasion Techniques related to parents' responses to their six-year-old children's requests to purchase cer­ tain cereals?"

Once again, scatter diagrams were used for these comparisons {Figures 3 thru 6). Similar statements can be made for six-year-olds as were made in the previous analysis for four-year-olds. There appears to be no linear correla­ tion and the results are not conclusive beyond mere specu­ lation .

Section IV: Frequency of Parents1 Purchases and the Rela­ tionships to Attributes of Product Data and Sales Persuasion Techniques'

Question IIID asks: "To what extent are there rela­ tionships between the frequency with which parents have purchased certain products their children have requested and particular attributes of Product Data and Sales Per­ suasion Techniques as described in the Content Analysis?”

Table 27 presents comparisons for each of the Pro­ duct Data items with respect to parents of four-year-olds and the purchasing of particular cereal products. For each comparison, the chi square value of 3.84 was needed for significance at the .05 level. From the analysis presented, m o n tr U £ o E fl U

n 3 in rent Questionnaire: "I have pur- in u n 41 41

4) this cereal as a result of my chile' *+4 ■H - u o Total No e 19.51 B Yes N a 0 Yus 14. 01 Sweetness 0 * N 46 J100 1J JB0.49 JJ Mo 85.19 Study Three; Cereal Purchases Made by Parents of Four-Year-Olds and and Four-Year-Olds of Parents by Made Purchases Cereal Three; Study % Tot.il N 54 11 41 N 16 1 1 ) YUS 29.62 21.08 Hit? 11. ? Characteristics Hiysical/oral eainhp o trbts f rdc Data Product of Attributes to Relationship % N 10 18 868. 29 28 Nu 70. 17 69 100 0 0 100 1 3 1 3 76.92 al 27 Table % Total 64 41 14 1 N 4 0 0 0 100 54 1 100 41 Yus Yus Nutr Nutr i t inna1 Va Va 1 uo % N 0 0 Ho Ho Total 11 54 41 N N 46 1 3 1J Yus Appear Appear anee/Shape 05.19 100 80. 49 N % I’ rod net 0 0 8 No 48 54 14.81 .5 41 J 9. 5; 0 1 Total 1 3 N % N 54 11 41 Yes Yes lOO 103 0 0 100 I’ai k i i . j i n i j 0 N 0 No 0 0 t 0 To To t a 1 _ 54 _

II 41 1 1 127 128 no one attribute attained a level of significance; that is, the purchase of cereal products by parents of four-year-olds did not vary with respect to Product Data. Certain single attributes were present in every cereal, thus making the analysis somewhat inconclusive. Since, for example, all commercials communicated nutritional value or package appear­ ance, there is no opportunity to observe differing responses dependent on presence or absence of these attributes.

Table 28 presents a similar analysis for parents of six-year-old children. Once again with df = 1, a value of

3.84 is essential for significance at the .05 level. In this particular test, the only attribute that attained significance at the .05 level was physical/oral character­ istics. This particular attribute varied among cereals with respect to parents of six-year-olds purchasing parti­ cular cereal products. Table 29 presents comparisons for each of the Sales

Persuasion Technique items with respect to parents’ of four- year-olds purchases of products. The chi square value of

3.84 was needed for significance at the .05 level. From the analysis presented in the 2 x 2 matrices, no one Sales

Persuasion Technique item achieved significance at the .05 level. In particular, the fun food/fun notion item was present in all the commercials, thus making the analysis of that item meaningless. From the analysis presented, there was no evidence of variation among parents of four-year-olds rant OuastJ.onna.lre: *3 have pur- tms cara&l as a result oi ny child1 Total Ho Yes N 5 25.0 2 1 Yes Sweetness 10 IB 11.01 B. S7 % N 12 6 Ho 91.4 ! so. i 75.0 1 ; Total H 41 IS S.04 SO. 0 4 B td he: eel ucae hd b aet o i-erOd ad the and Sis-Year-Olds of Parents by hade Purchases Cereal Three: Study 1 N 0 21.26 10 17.14 6 Yea Character1st Character1st ios

Physical/Oral h sur 3.91* square Chi H 29 11 eainhp oAtiue fPout Data Product of toAttributes Relationship Ho U2 . 06 76.74 so.o p -05‘p f- 1 - df I

Total 141 41 IS H B

N IS e Yes Nutri tional Table10 100 100 100 Value t 1 N 0 0 0 0 0 NO 0

Total 41 IS N u 11 12 IB 6 Yes Appearance/Shape du. du. 7 S. 0 91.41 1 7S .-41 U.6- S J7 Product N 25.0 2 B.57 1 Ho

% Total N S15 IS a 41 4 e Yes 100 100 100 1 PdL kaijlny N 0 0 0 0 0 Ho 6 0 Total N 4 1 15 a Froir, Parent Questionnaire: "I have pur­ chased this cereal as a result o f my child's reauest for it."

i ^ | z

o o

jo o

j o f o C *

c to n 0 IX

J* ft l >

£ — * r w c 3 1 T **- o T ST

(*t V* c ft r - z r* Uj Cl |rfr ft S' H c r & r * *< ft tfi *x> i t Li o ** rv I© H-, rt 'sC

W t r & (fl

I t c i © [fl p*n -y T* ft o' c Ui 1 Li K ft #.• X

x 1 o

c c w Cr 3 X .£ C ft 1“ C ft

un !Z

OCX 131 purchasing products as a result of a certain Sales Persua­ sion Technique item.

Comparisons for each of the Sales Persuasion Techni­ que items with parents' of six-year-olds purchase of pro­ ducts as a result of children's requests can be found in

Table 30. Once again, the .05 level of significance was not attained, and no conclusive findings were made.

In general, it appears that the attributes of Product

Data or Sales Persuasion Technique, as separate entities, do not vary in relationship to actual purchase of cereals by parents of four- or six-year-olds.

Summary - Study Three

Study Three has presented data which has measured the parent-child interaction resulting from young children making requests of parents to purchase certain cereal pro­ ducts. An exploration has been made of two basic statements:

"My child has requested me to buy this cereal" and "I have purchased this cereal as a result of my child's request for it." Statistical comparisons using the chi square statistic have indicated that while certain items vary, on the whole, there is little variability among the five cereals under study in two aspects:

1. Four- and six-year-old children's requests of parents to purchase the five cereals, and,

2, Parents' purchases of the five cereals as a re­ sult of their children's requests. u. to ‘O O. 4J W tl « tr r t « O E m u 3 n] Wi >u u .fi ai C a o w « 41 41 « w

u u Questionnaire : "I have pur- UJ 41 41 41 41 cereai as a result of iry thild'

Yes I ota t ^‘L

4 3 5 100 35 Yes”“ ’ ”“ s e “Y 8_ Fun Fun 100 100 od i n tio o N Food “No­ 0 li

td Tre Cra Prhss ae y aet o SxYa-ls n the and Six-Year-olds of Parents by Made Purchases Cereal Three: Study Total N 4 3 35 t N 21 9 54.29 19 2 Yes 40.8 40.8 3 25.0

Premium Included Included Premium eainhp o trbts f ae Pruso Teetmiijue Persuasion Sales of Attributes to Relationship . 6 1 . N 2 51.161 22 75.0 6 No 45.71 ___ i ToEaT N 43 35 8 N 24 8 65.11 28 4 .„50 Ys o N “Yes hscl Prowess Physical 68. 57 % Tab Tab It- 3 0 N 11 5 48 43 34.88 15

5(K 0 4 3 1.4 t _ N 35 8

t N 0 b5 5.4 43 53. 24i J 4b.51 20 14 .40.0 7 02 5086 8 25.0 2 75 .0 6 e o Total No Yes ats Character Fantasy Cartoon Cartoon 1 % 11 i 60.0 ii N N 3 2 3 5 3 6 Ys N TotIT No j "Yes 88 , i7 11 ' 5 . 4 3 6 1 5 1 i n^un/d r n j 1 e 7 5.0 N I 2 25.0 ! 25.0 2 ; H.5i) i II 1 I 35 i * i

CHAPTER V

SUMMARY AND CONCLUSIONS

After a restatement of the purpose and a review of

the procedure, this final chapter focuses on and summarizes

the findings of the three studies, with due recognition of

the limitations of the research. Possibilities for further

study are considered in the concluding section.

Restatement of Purpose

The purpose of this research was to enlarge the scope

of information about the impact of television food commer­

cials on young children. Through a content analysis of a series of commercials aired on the three major networks (ABC,

NBC, and CBS) during a three-hour period of time on Saturday morning, July 3, 1976, an investigation was made of the

attributes of Product Data and Sales Persuasion Techniques

contained within the commercials. The commercials under study were analyzed from the

point of view of two broad categories of items: Product Data

and Sales Persuasion Techniques contained within the com­ mercials. Product Data items were those which refer in any

133 134 way to the existence or elaboration of product or services information; and Sales Persuasion Techniques refer to tech­ niques used to promote the sale of the product. These two broad categories were taken from a 197 3 study by Winick which investigated food commercials for children. For this investigation, five attributes were selected for each category. Product Data items include: {1) sweetness,

C2) physical/oral characteristics, (3) nutritional value,

(4) product appearance or shape, and (5) packaging of prod­ uct. Sales Persuasion Techniques include: (1) fun food/ fun notion, (2) premium included with the product, (3) asso­ ciation with physical prowess, (4) cartoon/fantasy character as spokesman, and (5) use of a slogan or musical jingle.

Results of this analysis led to decisions about which attributes would be utilized in a later study in­ volving children's familiarity with particular aspects of the commercials. The three attributes selected after conducting the analysis were product appearance, packaging, and slogan/jingle.

The study of children's familiarity with the three dimensions of television food commercials was designed to broaden the range of inquiry into what particular aspects of 135

television cereal commercials contribute to advertising for

young children. This test of familiarity was chosen as the

central focus of concern for this entire investigation.

The concept of familiarity with three aspects of

cereal commercials was chosen for analysis because of the

obvious impact that a commercial would have on a child. The

assumption was that a child who could recognize the cereal

by its appearance, or its packaging, or its slogan/jingle

must have been exposed to it through commercial messages of

some kind. Testing the children's familiarity was an attempt

to quantify the extent to which the children were familiar with these three aspects.

Finally, then, parents of the subjects used for the

familiarity study were given a questionniare which investi­

gated parent-child interaction resulting from children's

requests for parents to purchase certain cereal products.

Although certainly not conclusive about a direct relationship

between television commercials and parents' actions regarding

their children's requests for certain cereals, nevertheless,

the data present an interesting exploration into the effec­

tiveness of the commercial messages. This has been first of all an investigation of tele­

vision cereal commercials aimed at children (Study One) .

The investigation has been further narrowed to five of the most recognized cereals. In study Two, these five were used

to gain insight into three aspects which may contribute to 136 their high degree of familiarity. The contact with parents of the children in the familiarity study took the notion of the impact of cereal advertising one step further in Study

Three. The intent was to investigate the effect of commer­ cials on parents who purchase cereal products for young children. The first study used a framework from Winick's work

<]9 7 3) to explore ten attributes of Product Data and Sales

Persuasion Techniques. The Content Analysis consisted of first defining the ten attributes (Appendix A) and then analyzing each commercial to determine presence or absence of these attributes. Study Two then utilized three attributes investigated in Study One and went on to determine what level of familiarity with these three attributes existed in young children. In Study Two, major independent variables were employed: age (four- and six-year-olds), attributes of

Product Data, attributes of Sales Persuasion Techniques, and the five individual cereal products, namely: Cheerios,

Frosted Flakes, Lucky Charms, Raisin Bran, and Rice Krispies.

Study Three was primarily concerned with three of the same independent variables, namely: age, attributes of Product

Data and attributes of Sales Persuasion Techniques.

Questions were asked for each study. The basic pat­ tern of the thirty-nine questions posed for Studies Two and

Three follow: 137

IIA. To what extent do the measures of familiarity of five cereal products vary:

1. According to product appearance/shape a. Four-year-olds b. Six-year-olds

2. According to packaging a. Four-year-olds b. Six-year-olds

3. According to slogan/jingle a. Four-year-olds b. Six-year-olds

4. For each cereal a. Four-year-olds b. Six-year-olds

IIB. To what extent do four- and six-year-old chil­ dren's responses to the three measures of famil iarity differ for the following:

1. Product 2. Packaging 3. Slogan/jingle

IIC. To what extent do the three measures of famil­ iarity in this study (product/appearance, packaging, slogan/jingle) relate to the attri­ butes of Product Data and Sales Persuasion Techniques from Study One:

1. Attributes of Product Data a. Sweetness b. Physical/oral characteristics c. Nutritional value d . Product appearance or shape e. Packaging of product

2. Attributes of Sales Persuasion Techniques a. Fun food/fun notion b. Premium included with product c. Association with physical prowess d. Cartoon/fantasy character as spokesman e. Use of slogan or musical jingle

IIIA. To what extent do children's requests for par­ ents to purchase products as a result of 138

television advertising vary among the following indicated categories:

1. Four-year-olds 2. Six-year-olds

IIIB. To what extent do parents' purchase ofproducts as a result of children's requests vary among the following indicated categories:

1. Among parents of four-year-olds 2. Among parents of six-year-olds

IIIC. To what extent are the attributes of Product Data and attributes of Sales Persuasion Tech­ niques used in Study One related to parents' responses to their children's requests to pur­ chase certain cereals according to children's ages :

1. Parents of four-year-olds 2. Parents of six-year-olds

HID. To what extent are there relationships between the frequency with which parents have purchased certain products their children have requested and particular attributes of Product Data and Sales Persuasion Techniques as described in the Content Analysis?

Procedures

The Content Analysis was conducted on the basis of

two dimensions of Product Data and Sales Persuasion Techniques borrowed from an earlier study by Winick {]973). The Winick

framework was modified to fit the needs of the investigation

in Study One. The subjects selected for participation in Study Two were chosen from the student population at Jefferson Elemen­

tary School and Children's Center in a low-income area of

Pasadena, California. Since the data were gathered on one 139 day, the sample size did not vary. Exactly thirty-five four- year-old children and thirty-five six-year-old children participated. Study Three, which was a questionnaire directed to the parents of the same subjects who partici­ pated in Study Two, produced a varied sample size, however, with 51% of parents of four-year-olds and 30% of parents of six-year-olds responding.

The data gathering for Study Two was conducted on

April 6, 1978 during which time the instrument designed for this research {Appendix E) was administered. In the case of

Study Two, the administration of the tests of familiarity was supervised closely by the researcher. Standardization of procedures and general environment within the classroom was controlled as much as possible. Extensive training of the persons involved with the testing stations preceded the actual data gathering to insure uniformity and consistency of technique. The data from Study Two were tabulated and ana­ lyzed to reflect findings for four-year-olds and six-year- olds . The instrument used for Study Three {Appendix F), the

Parent Questionnaire, was sent home to parents with the four- and six-year-old subjects themselves. This insured that the parents selected for the questionnaire would be the parents of the same subjects who participated in Study Two. The questionnaire consisted of a very uncomplicated checklist and parents were only asked to check small boxes to indicate 140

"Yes" or "No" to statements that were posed. The statements were simple and clear: "My child has requested me to buy this cereal" and "I have purchased this cereal as a result of my child's request for it." The questionnaire was com­ pleted by eighteen parents of the thirty-five four-year-olds and thirteen parents of the thirty-five six-year-olds. The chi square was the statistical tool used in examining the data collected in Studies One, Two and Three.

Chi square matrices of 2 x 2, 2 x 3 , and 2 x 5 were used for category and group comparisons. Chi square values were accepted as significant at the p .05 level and more specific levels of significance were mentioned where appropriate throughout the analysis of the data in Chapter Four. Those findings possessing significance were designated by the pres­ ence of an asterisk.

General Findings

Of the three studies of children's food commercials, Study One attempted to provide insight into the content of the commercials; Study Two explored children's familiarity with five specific cereal advertisements; and Study Three examined parent-child interaction resulting from televised food commercials. While the three studies were different to some extent, all three were unified by attributes of Product

Data and Sales Persuasion Techniques. Therefore general comparisons of the three studies were possible. 141

For Both Product Data and Sales Persuasion Technique Attributes

In Study One, the Content Analysis, there is no vari­ ability between the two groups of attributes of Product Data and Sales Persuasion Techniques. Television commercials under study do not seem to rely more on one set of attributes than the other. Nevertheless, there are significant differ­ ences within each category of attributes of Product Data and

Sales Persuasion Techniques {Figure 7).

In both categories of attributes, certain attributes became apparent because of their high level of significance.

In the Product Data category, packaging, physical/oral characteristics, and nutritional value were among the highest.

In the Sales Persuasion Techniques category, of high signifi­ cance were fun food, slogan/jingle, and the appearance of a cartoon fantasy character {Figure 8).

In Study Two, the three measures of familiarity

(product appearance, packaging, and slogan/jingle)and their comparison with attributes of Product Data and Sales Persua­ sion Techniques gave no conclusive results.

In Study Three the two categories of Product Data and Sales Persuasion Techniques were examined in relation to children’s requests of the parents to purchase cereals, and in relation to parents purchasing these same cereals as a result of their children's requests. In addition, each of the individual attributes within each category was examined Percentages 100 20 00 90 10 40 50 60 10 7(1 itga Idctn oa Pretgs f trbts f rdc Dt and Data Product of Attributes of Percentages Total Indicating Histogram Product 56.211 Data Sales Persuasion Techniques Persuasion Sales ■ V > ' .’V iue 7 figure ,

Salts Persuasion Techniques Techniques Persuasion Salts

62.891 2 4 1 143

Slogan/ Jingle 03 & \ Fantasy Character

Premium

Fun Food/ Notion

x- Product Appearance

Nutritional Value

Physical/Oral Characteristics \a so 00 Sweetness 00 f'J Histogram Histogram of Product Data Attributes vs. Sales Persuasion Techniques

© o Ol o00 oo o in 144 with respect to children's requests and parents' purchasing.

No conclusions could be made for six-year-old children's

requests in both categories. However, some linear correla­

tion is apparent for four-year-olds' requests in Sales

Persuasion Techniques (Figure 3). No conclusion could be made concerning parents' of four-year-olds purchasing of cereal products as a result of children's requests.

Within each category of attributes, there were, in

general, significant findings between the frequency with which parents purchased certain products that their children

requested, and particular attributes of Product Data and Sales

Persuasion Techniques as described in the Content Analysis

of Study One. It is important to note that certain single

attributes were present in each of the five cereals (Figure

9), thus making the analysis somewhat meaningless.

For Both Four- and Six-Year-Olds

In Study Two, variability existed between compari­ sons of fcur- and six-year-old's familiarity with product

appearance/shape. However, a high correlation of signifi­

cance at the .001 level was apparent for familiarity with

respect to package and slogan/jingle. This is an indication

that there is a relationship between age and children's

familiarity with package and slogan/jingle. Percentages 100 'JO 110 60 40 1)0 Jo in 10 histogram of Percentages of Individual Attributes of Product Data Data Product of Attributes Individual of Percentages of histogram trbtso Pout Data Product of Attributes V / 20 % A 0 A A t IflU e n 401 X 'ft y *i u fiw u t*N o h x n ito n p » p i» p, u o H H X X