University of Massachusetts Amherst University of Massachusetts Amherst

ScholarWorks@UMass Amherst ScholarWorks@UMass Amherst

Doctoral Dissertations 1896 - February 2014

1-1-1978

The effect of sex-role stereotyped picture books on the play The effect of sex-role stereotyped picture books on the play

behavior of three- and four-year-old children. behavior of three- and four-year-old children.

Eleanor Ashton University of Massachusetts Amherst

Follow this and additional works at: https://scholarworks.umass.edu/dissertations_1

Recommended Citation Recommended Citation Ashton, Eleanor, "The effect of sex-role stereotyped picture books on the play behavior of three- and four-year-old children." (1978). Doctoral Dissertations 1896 - February 2014. 3364. https://scholarworks.umass.edu/dissertations_1/3364

This Open Access Dissertation is brought to you for free and open access by ScholarWorks@UMass Amherst. It has been accepted for inclusion in Doctoral Dissertations 1896 - February 2014 by an authorized administrator of ScholarWorks@UMass Amherst. For more information, please contact scholarworks@library.umass.edu.

THE EFFECT OF SEX-ROLE STEREOTYPED PICTURE

BOOKS ON THE PLAY BEHAVIOR OF THREE-

AND FOUR-YEAR-OLD CHILDREN

A Dissertation Presented

By

ELEANOR ASHTON

Submitted to the Graduate School of theUniversity of Massachusetts in partial fulfillment

of the requirements for the degree of

DOCTOR OF EDUCATION

March 1978

Education

© ELEANOR ASHTON 1978

All Rights Reserved

11

THE EFFECT OF SEX-ROLE STEREOTYPED PICTURE

BOOKS ON THE PLAY BEHAVIOR OF THREE-

AND FOUR-YEAR-OLD CHILDREN

A Dissertation Presented

By

ELEANOR ASHTON

Approved as to Style and Content by:

Grace J. Craig, Chairperson

Linda Blane, Member

Daniel R. Anderson, Member

Mario Fantini, DeanSchool of Education

iii

To

Waver ley

Amber

And Tom

IV

ACKNOWLEDGMENTS

I wish to thank the members of my committee for

their continual help and encouragement. In addition, I am

grateful to Lisa Weld of the University of Massachusetts

Computing Center for her valuable assistance. And, of

course, a special word of thanks to the teachers and chil-

dren of The Human Development Laboratory School; without

their willingness to participate in the study, none of it

would have been possible.

ABSTRACT

The Effect of Sex-Role Stereotyped PictureBooks on the Play Behavior of Three-

and Four-Year-Old Children

(March 1978)

Eleanor Ashton

B.A., New York UniversityM.Ed., University of MassachusettsEd.D., University of Massachusetts

Directed by: Dr. Grace J. Craig

The purpose of this study was to ascertain the effect

of sex-role stereotyped picture books (as compared with non-

stereotypic picture books) on the behavior of three-to-f ive

year old children. It was hypothesized that children's

literature models sex-role behaviors and these behaviors in

turn affect the sex role behaviors of children.

Thirty-two children, sixteen boys and sixteen girls,

were individually given the opportunity to play with a select

group of male, female, and neutral toys in an experimental

room (located down a corridor from their nursery classroom)

,

where their play behavior was recorded on videotape through

a one-way mirror. Following this play session, each subject

was read a picture book that presented a same-sex main char-

acter playing with either a sex-role stereotypic or non-

stereotypic toy. After this exposure, the subjects were

VI

given the opportunity to play with the same toys and their

behavior was again recorded.

It was found that boys and girls exposed to a stereo-

typic picture book significantly more often select a stereo-

typic toy with which to play (manipulate and visually regard)

than children exposed to a non-stereotypic book. Conversely,

boys and girls exposed to a non-stereotypic book significant-

ly more often select a non-stereotypic toy with which to play

than children exposed to a stereotypic book. The hypothesis

was substantiated by these results.

In addition, female subjects appeared to be more

strongly affected by the picture books than their male coun-

terparts. Both male and female groups, however, were more

affected by the non-stereotypic picture book than the male

and female groups exposed to the stereotypic picture book.

It was concluded that picture books, through their

characters, may offer children models that depict sex-role

behaviors. When children are exposed to these characters,

they may in turn model these same behaviors. Therefore, to

encourage the development of each child's potential for per-

sonal growth, storybook characters should portray a wide

variety and range of sex-role behaviors that are not limited

by cultural stereotypic definitions.

Vll

TABLE OF CONTENTS

PageACKNOWLEDGMENTS •••••••••••• V

ABSTRACT

LIST OF TABLES x

LIST OF FIGURES xi

Chapter- I. INTRODUCTION 1

II. REVIEW OF THE LITERATURE 5

Definitions 5

Harmful Aspects of Rigid Sex-RoleStereotyping 6

The Development of Sex-Roles 11The Onset of Sex-Role Development 11Biological Theory of Sex-Role Development . . 12Theories of Sex-Role Development as Learned . 16

Picture Books in the Child's Environment ... 22Sex-Role Stereotyping in Children's Books ... 23

Sex of Characters 23Activity of Characters 25Language 26

Relationship Between Picture Books andTelevision 30

III. METHOD 34

Hypotheses 34

Design Rationale 35

Subjects 38

Setting 39

Equipment 39

Toys 39

Picture Books 41

Procedure 45

Data Collection 45

Data Recording 47

Comparability of Groups 49

Data Analysis

viii

Chapter^ IV. RESULTS AND DISCUSSION

Results 51Discussion 59

Female Group Toy Manipulation 59Male Group Toy Manipulation 62Modeling 65Toy Choice 65Charm Choice 67Story Recall 69Friendship 71Summary and Suggestions for Future Research . 72

V. EDUCATIONAL IMPLICATIONS 75

Specific Implications 75General Implications 76

REFERENCES 84

AppendicesA. PICTURE BOOKS 9 3

B. TABLES OF RESULTS 142

IX

LIST OF TABLES

Table Page1. Male and Female Group Mean Post-Test Scores for

Manipulation and Visual Regard of Female Toys . 522. Male and Female Group Mean Post-Test Time Scores

for Manipulation and Visual Regard of MaleToys 55

3. Pre- and Post-Test Mean Time Scores for ToyManipulation and Visual Regard 56

4. Mean Time Difference Scores for Female and MaleToy Manipulation and Visual Regard 58

5. Mean Time Difference Scores for Toy Manipulationand Visual Regard 60

6. Mean Time Scores for Female Groups forManipulation and Visual Regard of IndividualToys 61

7. Mean Time Scores for Male Groups for Manipulationand Visual Regard of Individual Toys 64

8. Frequency and Range of Subject's Modeling .... 669. Frequency of Subject's Verbal Toy Selection ... 68

10. Frequency of Subject's Charm Selection 6811. Subject's Responses to Questions 7012. T-Test Analysis of Toy Manipulation and Visual

Regard on Pre-Test Scores Indicating GroupEquivalence 14 3

13. Analysis of Variance of Mean Post Test Scores forToy Manipulation 144

14. Analysis of Variance of Mean Post Test Scores for

Toy Visual Regard 145

15. Analysis of Variance of Mean Post Test DifferenceScores for Toy Manipulation 146

16. Analysis of Variance of Mean Post Test DifferenceScores for Toy Visual Regard 147

X

LIST OF FIGURES

Figure Page1. Research Design 34

XI

CHAPTER I

INTRODUCTION

The intent of this study is to ascertain the effects

of sex-role stereotyped picture books (as compared with non-

stereotyped picture books) on the behavior of children aged

three-to-f ive years. The study grew out of the hypothesis

that adherence to rigid stereotypical sex-roles can limit the

opportunity for both males and females to reach their full

potentials for personal growth. As Margaret Mead (1935)

aptly stated:

No skill, no special aptitude, no vividness of imagi-nation or precision of thinking [should] go unrecog-nized because the child who possessed it was of onesex rather than the other. No child [should] be re-lentlessly shaped to one pattern of behavior, but in-stead there should be many patterns in a world thathas learned to allow to each individual the patternwhich was most congenial to his gifts.

Yet, this concern for individual differences and

potential is not demonstrated by picture book authors and il-

lustrators who present stereotypic information. Picture

books, the literature of young children, play an important

role in the socialization of the young child. Through books,

children can learn about the world outside of their immedi-

ate environment. They learn about what other boys and girls

do, say, and feel. They learn about what is right and wrong.

1

2

and what is expected of children their age. Picture books

are all the more influential because they are directed at an

audience that is in the process of developing their sex-roles

and are particularly impressionable. In addition, picture

books provide children with role models— images of what they

can and should be like when they grow up. Masha Rudman , a

professor at the University of Massachusetts and author of

Children’s Literature; An Issues Approach , illustrates the

powerful lingering effect of picture books in her class in

children's literature. She asked college students to de-

scribe their impressions of a "princess." Their replies re-

vealed the stereotypic "princess" character found in many

picture books. This stereotype may bear no resemblance to

reality. The power of picture books to create this long-

term stereotyped impression can also be applied to stereo-

types dealing with sex-roles. The sex-roles that picture

books represent follow the stereotypes that the culture pre-

scribes, whether or not the prescription is based on reality

The following text from the picture book I'm Glad I'm a Boy ,

I'm Glad I'm a Girl by Whitney Darrow and published by Simon

and Schuster in 1970 is one example of the repressive stereo

types the culture, through picture books, perpetuates.

Boys have trucks

.

Girls have dolls.

Boys are strong.Girls are graceful.

Boys are doctors.Girls are nurses.

3

Boys are policemen.Girls are metermaids.

Boys are football-players.Girls are cheerleaders.

Boys are pilots.Girls are stewardesses.

Boys fix things.Girls need things fixed.

Boys invent things.Girls use what boys invent.

Boys build houses.Girls keep houses.

I'm glad you're a boy.I'm glad you're a girl.

We need each other.

Although it is generally assumed that books like I'm

Glad I'm a Boy^ I'm Glad I'm a Girl limit the opportunity of

both males and females to reach their full potential for per-

sonal growth, we are currently operating more under assump-

tions than proven fact. Researchers have not thoroughly

examined to what extent books affect children. Therefore,

the present study was designed to determine whether picture

books presenting stereotypic or non-stereotypic behaviors

differentially affect specific behaviors in preschool age

( three-to-f ive) children.

It was hypothesized that following exposure to a

sex—role non—stereotypic picture book, both girls and boys

will more often select a non-stereotypic toy with which to

4

play. Conversely, subjects exposed to a sex-role stereotypic

book will more often select a stereotypic toy with which to

play.

Thirty-two children, sixteen boys and sixteen girls,

participated in this study. Upon entering the experimental

room, the subjects were given an opportunity to play with

a select group of male, female, and neutral toys while their

play behavior was recorded. Following this play session each

subject was read a picture book that presented a main charac-

ter playing with either a sex-role stereotypic or non-stereo-

typic toy. After this exposure, the subjects again played

with the same toys and their behavior was recorded.

CHAPTER I I

REVIEW OF THE LITERATURE

Definitions

For the purposes of this study, sex-role stereotype

is taken to mean all the personal qualities, behavioral

characteristics, interests, abilities and skills which one

is expected to have because one is a certain sex. These

stereotypes are pervasive in our culture. Through an exten-

sive questionnaire survey, Broverman (1972) found evidence

of these sex-role stereotypes which were consistent across

groups varying in age, sex, marital status, and education.

Commonly shared male stereotypes included such descriptions

as independent, objective, active, competitive, logical,

skilled in business, worldly, adventurous, able to make de-

cisions easily, self-confident, and demonstrates leadership

abilities. The stereotypic perception of women was defined

by a relative absence of these traits. Women were perceived

to be dependent, subjective, passive, non-competitive, il-

logical. The positively valued feminine traits formed a

cluster which reflected warmth and expressiveness. Attri-

butes such as gentle, sensitive to the feelings of others,

tactful, religious, neat, and quiet were included in the

5

6

cluster. In addition, male and female sex-roles appear to be

bipolar in nature, in that the adjectives describing mascu-

and femininity are considered to be opposing quali-

ties. To be highly masculine is to possess mostly male

traits to the practical exclusion of female traits and vice

versa (English and English, 1958) . Masculinity-femininity

measures (Terman and Miles, 1936; Strong, 1936; Campbell,

1966) for the most part have been constructed so that on each

item the response scored "masculine" is the one that is en-

dorsed by most males, whereas the one scored "feminine" is

the one favored by the majority of females. Any individual's

masculinity-femininity score, therefore, serves as an index

of how closely his/her self-reports, self-descriptions, and

preferences correspond with those of the majority in his/her

sex. It is evident that these "majority attributes" involve

a stereotype about modal or typical characteristics rather

than a precise behavior description. Although this research-

er finds this concept of masculinity and femininity somewhat

contrived, the present study was constructed based on this

widely held definition.

Harmful Aspects of RigidSex-Role Stereotyping

Although this study grew out of the belief that ad-

hering to rigid stereotypic sex-roles limits the growth of

individuals , that belief needs to be documented by research

7

investigation. Stereotypically the female in this culture

is expected to be unaggressive, passive, nurturant, attrac-

tive, and oriented toward others (Kagan, 1964) . This orien-

tation may limit the intellectual growth of the female in

that passivity and dependency do not foster intellectual

growth in girls, while assertive and active behaviors sup-

port intellectual growth (Maccoby, 1966) . To be active and

assertive, however, goes against the female sex-role stereo-

type prescribed by the culture. Problem solving, another in-

dicator of intellectual ability, was studied by Witkin (1962)

who showed that children who are dependent in their inter-

personal relations, suggestible, conforming and likely to

rely on others for guidance and support show difficulty break-

ing set in problem solving. Yet females in our culture are

encouraged to be dependent and conforming (Frazier and Sad-

ker, 1973, p. 95)

.

In addition to the limiting of intellec-

tual capacities high femininity (as determined by a high

score on a M-F scale such as the California Psychological

Inventory) in females has consistently been correlated with

high anxiety, low self-esteem, and low social acceptance

(Costentina and Heilbron, 1964; Gall, 1969; Gray, 1957;

Sears, 1970; Webb, 1963). Indeed girls in our culture often

express a wish to become a boy or daddy yet relatively few

boys state a desire to be a girl or a mommy (Brown, 1957)

.

This may indicate that little girls perceive the greater

prestige and numerous privileges connected with the masculine

8

role

.

Adherence to the male sex-role stereotype requires

aggressiveness and independence, while suppressing emotional

expression and nurturant behavior (Kagan, 1964). Hartley

(1959) indicates that many boys give evidence of anxiety cen-

tered in the whole area of sex-connected role behaviors, an

anxiety which frequently expresses itself in over-straining

to be masculine, in virtual panic at being caught doing any-

thing traditionally defined as feminine. For male children

"the outward semblance of non-femininity is achieved at a

tremendous cost of anxiety and self alienation" (Hartley,

1959, p. 462). In addition to the emotional stress experi-

enced in childhood, high masculinity has been correlated

during adulthood with high anxiety, high neuroticism, and

low self-acceptance (Hartford et al . , 1967; Mussen, 1962).

Although the research clearly indicates the rigid

female stereotypes that are found in picture books, little

is said about the equally rigid male sex-role stereotypes

found in the literature. The pressures that demand inactiv-

ity from females also demands activity from boys. Boys per-

ceive that they "need to be strong, they have to be ready to

make decisions; they must be able to protect women and chil-

dren in emergencies; they have to have more manual strength

than women and are the ones to do the hard labor, the rough

work, they must be able to fix things; they must get money

to support their families and they need 'a good business

9

head'" (Hartley, 1959, p. 461).

Creativity may also be affected by a strong identifi-

cation with the masculine role. Creativity was defined as

originality related to independence of judgment, to personal

complexity, preference for complexity in phenomena, to self-

assertion and dominance, and rejection and suppression as a

mechanism for the control of impulse (Barron, 1955) . MacKin-

non (1961) reports that men who are outstanding in original-

ity and creativity (as measured by the Barron-Welsh Art Scale

of the Welsh Figure Preference Test) score more toward the

feminine end of a M-F scale than do their less creative coun-

terparts. This is seen as a reflection of the greater

breadth of interests among creative men. Yet, based on re-

search findings, boys consistently make appropriate sex-typed

choices (Rabban, 1950) and display a lack of flexibility

within the whole ar3a of sex-typed related activities. Mac-

coby's (1966) review of the studies indicates that analytic

thinking, creativity, and high general intelligence are as-

sociated with cross-sex typing, in that men and boys who

score high are more feminine, and the women and girls are

more masculine than their low-scoring, same sex counterparts.

In addition to the creative and intellectual re-

sources that are lost to society by adherence to strict sex-

roles, individuals may suffer emotional strain as well. Bern

(1976), using 2,000 undergraduates in a study focusing on the

concept of psychological androgyny, found that highly mascu-

10

line men and highly feminine women (as measured by the Bern

Sex-Role Inventory) experienced the most discomfort and felt

the worst about themselves after performing cross-sex activi-

ties. These activities were categorized as male stereotypic

(nail two boards together, attach artificial bait to a fish-

ing hook) or female stereotypic (iron cloth napkins, wind a

package of yarn into a ball) . It was concluded that cross-

sex activities are problematic for strongly sex-typed indivi-

duals, and that traditional sex-roles do produce an unneces-

sary and perhaps dysfunctional pattern of avoidance for

many people. The ironic character of these findings is high-

lighted in Slater's work (1964). He suggests that the abil-

ity to alternate instrumental (male) and expressive (female)

role performance rapidly—what he calls interpersonal flexi-

bility— is coming to be more highly valued in our society.

This cultural contradiction may be an indication that soci-

ety is in the process of changing. The strong emphasis

placed on rigid sex-role definitions may be altering to ac-

commodate the need for more flexibility in a rapidly changing

world.

This review strongly indicates that adherence to

rigid sex-roles may limit the healthy development of indivi-

duals. It is important to now consider how sex-roles develop

and to understand the part that picture books may play in

fostering that development.

11

The Development of Sex-Roles

The onset of sex~irole developmen t. There is evidence to sug-

gest that the development of sex-roles begins at a very early

age. Researchers using toy and activity preference and area

usage tests (Sears, Rau, and Alpert, 1965) as an indicator

of sex-role development have found that three-year-old chil-

dren are able to make sex-role distinctions, e.g., make sex-

appropriate toy choices, are aware of self as male or female,

are aware of clothing and hair style as ‘features of maleness

and femaleness, and make appropriate projections regarding

future parental role (Rabban, 1950; Vance and McCall, 1934).

However, Rabban found that three-year-olds were less able

than older children to make these accurate sex-role distinc-

tions. In addition, Hampson's (1965) research with hermaph-

roditic children indicates that children who, through sur-

gical correction, are assigned to a sex after two years of

age show poorer adjustment than younger children.

Sears (1965) ,working with four-year-olds, found

that boys spent more time in the portion of a large nursery

school playroom where blocks, wheel toys, and carpentry

tools were to be found, whereas girls spent more time in the

area having the dress-up clothes, the cooking equipment, and

the doll house. Five- and six-year-olds demonstrate a

knowledge of the appropriateness of toys, games and activi-

ties (Anastasiow, 1965; De Lucia, 1963; Brown, 1956) and

12

perceive that parents prefer activities appropriate to the

child's sex (Fauls and Smith, 1956) . In summary, these

studies indicate that children as early as three years of

age can demonstrate an ability to make culturally established

sex-role distinctions, and this ability develops with age.

The question to turn to next, is how do children develop

sex-appropriate patterns of behavior?

Biological theory of sex-role development . A number of ex-

planations have been offered regarding the origin and devel-

opment of sex-role patterns. These theories vary along a

continuum from largely biological explanations (Hutt, 1972;

Terman and Miles, 1936; Stoller, 1968; Tiger, 1969) to

theories which consider socialization as the primary deter-

minant of sex-role behavior. The biological point of view

suggests that sex differences in cognitive abilities and per-

sonality traits seen in adults result from physiological

and anatomical differences at birth. One commonly presented

concept suggests that male size, strength and motor ability

tend to set standard patterns of dominance, aggression and

0nergetic activity for all males and contrasting patterns

for all females (Terman and Miles, 1936). Hutt (1972, p. 37)

states that "hormones have many effects other than those on

sex organs and functions, and it is often as a result of

these non-sexual effects that we may observe a characteris-

tically 'feminine' or 'masculine' response in matters other

13

than those concerned with reproductive functions.

"

These researchers, in taking the "naturists" position

in presenting their arguments for the biologically determined

sex-role behaviors, fail to consider the influence of social-

ization on the development of sex-roles. While some evidence

suggests that genital and hormonal sex differences influence

behavior, they may not be the major determinant of sex-role

behaviors. Research suggests a biological predisposition

toward some sex-linked behaviors. For example, males appear

to be more aggressive than girls, and excel in visual-spatial

ability while girls appear to have greater verbal ability

than boys (Maccoby and Jacklin, 1974) . Although these re-

searchers give evidence for the biological predisposition for

these behaviors, they do not believe that these behaviors are

unlearned. They state that both sexes have the capacity to

perform these behaviors but one sex may do so more readily

while the other sex requires more of a push to do so. Rosen-

berg and Sutton-Smith (1972, p. 77), however, clearly under-

emphasize the role bf heredity:

Perhaps we may illustrate the problem by saying that

if an animal's ratio of heredity to environment is

1:1, in human beings the ratio is probably close to

1:10. Given this degree of flexibility in man, it is

hard not to believe that the form of his sex role

development will depend on whether contemporary cul-

ture wishes to underemphasize the differences between

genders or to overplay them.

Hamburg and Lunde (1966, p. 16) who reviewed studies

of pseudohermaphroditic children go even further than Rosen-

14

berg and Sutton Smith in their support of environmental in-

fluences on the development of sex-roles. They report that:

. . . gender role is entirely the result of a learn-ing process which is guite independent of chromosomal,gonadal, or hormonal sex. In other words, genderrole is, for all practical purposes, determined en-tirely by environmental conditions during the firstfew years of life.

Their statement reflects the extreme environmentalist posi-

tion in the ongoing nature-nurture debate. In addition to

their strong emphasis on the role the environment plays in

the development of sex-roles, they place a critical period

for establishing gender role between the age of roughly

eighteen months and three years. Hampson's (1965) research

supports the "critical period" theory. He found that her-

maphroditic individuals who had had little sex-role experi-

ence prior to the time of the surgical sexual correction

(i.e., were two years of age or less) adjusted well to the

new sex-assignment and showed normal psycho-sexual adjust-

ment. But, sex assignments made after age two resulted in

progressively poorer adjustment.

The importance of socialization experiences is in-

dicated by anthropological research. In a study of three

New Guinea tribes Mead (1935) found a tribe in which both men

and women showed cooperative, non-aggressive, maternal be-

havior (considered feminine in our society) . Both sexes

of a second tribe were ruthless, aggressive and violent

(behavior considered masculine in our culture) . A third

15

trib© showsd a rsvarsal of our S6x—rol© pattsrn. Men were

dependent and emotional while women were impersonal, domi-

nant, and managerial. More recently in a study of children

three-to-eleven years-old in six cultures, Whiting and Ed-

wards (1973) found that sex differences in behavior could be

almost entirely accounted for by the type of tasks children

were assigned. Children assigned tasks which kept them in

the home such as cooking and child care showed more behavior

appropriate to the female stereotype (e.g., helping, respon-

sibility and social skills) . Children assigned tasks that

took them outside the home (e.g., gardening, animal husband-

ry) showed more male appropriate behavior such as aggressive-

ness and rough and tumble play. These studies suggest that

sex-role patterns are highly malleable and vary consider-

ably according to cultural expectations.

An exhaustive review of the literature on sex dif-

ferences has recently been published by Maccoby and Jacklin

(1974) . These authors conclude that there is evidence for

biological differences favoring greater aggressiveness in

males and superiority in some cognitive skills (e.g., visual-

spatial abilities) and female superiority in verbal skills.

However, Maccoby and Jacklin point out that these differences

do not suggest exclusion of one sex or the other from certain

vocations or roles in life but rather that more effort be di-

rected toward providing additional learning in areas of weak-

ness .

16

Theories of sex-role development as learned . Since biologi-

cal factors do not seem to be a sole determinant of most sex-

role behaviors, researchers have looked to the child's en-

vironment to explain the basic mechanisms which underlie the

learning of sex-role behaviors. Three theories of sex-role

development have been proposed: reinforcement-learning

theory; imitation-learning theory; and cognitive-developmen-

tal theory.

Reinforcement-learning theory of sex-role develop-

ment . The reinforcement-learning theory of sex-role develop-

ment emphasizes the role played by reward and punishment in

the development and strengthening of sex-typed behavior.

When a child behaves in a sex-appropriate fashion, he/she

will be rewarded and will repeat and value that behavior.

Behavior which is not reinforced extinguishes. In support

of this view, researchers have found that parents have a dif-

ferent set of expectations for male and female children.

Pitcher (1974) found that both fathers and mothers clearly

regard it as feminine to be interested in pretty clothes,

domestic habits, families, babies, and to be more social and

guileful. In contrast parents regarded it as masculine to

be interested in objects and ideas (not persons) and in mak-

ing things work (process). Rubin, Provenzano, and Luria

(1974) using 30 pairs of new parents, found that parents had

different sets of adjectives to describe infants which are

based on gender alone. They found that newborn infant girls

17

are seen as "softer," "littler," and boys are seen as

"firmer," "more alert," and "stronger." Yet, hospital scales

showed that male and female newborns did not significantly

differ on any physical or health measures. Kagan (1964)

has also noted the differential treatment accorded male and

female infants. Lewis (1972) found that mothers are more

responsive to three-month-old girls exhibiting affective

behavior (i.e., crying and smiling) than to boys showing

this behavior. Boys, however, receive more attention for ac-

tivity and movement than do girls. In addition, laboratory

research has demonstrated that children learn many behaviors

through observation, but tend to perform and practice only

those behaviors which receive reinforcement (Bandur and Wal-

ters , 1963)

.

In summary, the research suggests that parents have

different expectations concerning the behavior of boys and

girls and explicitly convey these expectations and shape

the child's behavior through a system of rewards and punish-

ments .

Picture books can be seen as an instrument that aids

the development of sex—roles according to the reinforcement—

learning theorists. A child's behavior can be reinforced

by viewing the reinforcement of similar behaviors of story-

book characters. In a similar way a child could question or

diminish a behavior that is punished in the children's book.

Imitation-learning theory of sex-role development.

18

The imitation-learning theory views sex-role development as

a product of identification. Children identify with role

models (usually their parents and teachers) , and try to

duplicate or emulate the model's behaviors, feelings and

values. The evidence for the importance of modeling in the

sex-typing process consists largely of observations of simi-

larities in the behavior of parents and their children. Ban-

dura and Walters (1963) have shown that such similarities oc-

cur in many behavior qualities, including aggression, de-

pendency, withdrawal, etc. In addition, Bandura (1969) at-

tributes sex-role behavior to " identificatory processes" and

suggests that multiple models including teachers, adults in

the neighborhood and the media are sources for the modeled

behaviors. Kagan (1964, p. 215) has suggested that children

identify with models in order to "reduce anxiety over antici-

pated aggression from a model and to experience or obtain

positive goal states which he perceives that the model com-

mands. "

Freud distinguished various types of identification,

but based the identificatory process on an individual's in-

tense dependency on a parent. As the child grows older he/

she is motivated to imitate a parent in an attempt to either

recapture the close relationship with that parent or as a

vehicle for relieving feelings of anxiety (Bronfenbrenner

,

1960) . The motivation for identifying with a model may be

influenced by many factors, but researchers agree that model-

19

ing does take place and that it is a "powerful source of in-

fluence upon the nature of sex-typed behaviors that children

adopt" (Maccoby and Jacklin, 1974, p. 275).

The relationship between picture books and the imi-

tation-learning theory of sex-role development is clear when

picture books are perceived as a source of symbolic models

for children. Walter Mischel's (1966, p. 61) statement per-

taining to symbolic models underlines the importance of pic-

ture books as a mechanism of sex-role learning: "With humans,

iriformation transmitted through verbal or other symbols,

can dramatically reduce the number of trials needed for er-

rorless performance." Thus, if the book supplies symbolic

models, and the models transmit exaggerated sex-role stereo-

types, then children observing these models may more easily

perform these distinctly sex-typed behaviors. Maccoby and

Jacklin (1974, p. 301) suggest that "the modeling process is

crucial in the acquisition of a wide repertoire of potential

behaviors." For example, Duncker (1938) told preschool

children an exciting, vivid story in which the hero. Eagle-

feather, violently disliked a pleasant-tasting food while he

supposedly relished a more noxious food. After listening

to the exploits and emotional reactions of the hero, the

children tended to adopt his food preferences. In addition

to the models found in picture books, the person who reads a

picture book to children is usually a significant adult,

i.e.

,

parent or teacher. These adults may in fact be models

20

for the children involved. Bandura and Huston (1961) were

able to demonstrate that nurturance of the child by the model

increased the probability that the child will imitate the

model's behavior. If the teacher as nurturing model reads

a picture book, one possible result may be that the chil-

dren accept the message of the book as being part of the

modeled behavior.

Cognitive-learning theory of sex-role development.

The cognitive-developmental theory which has been explicitly

enunciated by Kohlberg (1966, 1969) conceives the development

of sex-roles as an aspect of cognitive growth in which ob-

servational learning plays an important part. In a general

sense, "cognitive representation is a coding of the informa-

tion in the environment . . . one that indicates relation-

ships among items as well as items themselves" (Boldwin,

1969, p. 333). According to Kohlberg (1966, p. 156), the

"egocentricity of children's thought causes them to view

themselves in a positive light." Thus when the child labels

himself/herself as male or female the child chooses to do

those things that are sex-appropriate. The development of

the concepts of sex—role that a child considers appropriate

remains under the control of internal, cognitive factors.

Sex-typed behaviors and sex-role concepts parallel the

child's conceptual growth.

Of course, the cognitive-developmental theory calls

A child's concept of what isupon the other two theories.

21

appropriate behavior for male or female depends both upon

what he/she sees males and females doing and upon the ap-

proval or disapproval that these actions elicit. Both of

these kinds of events constitute information the child can

draw upon in building his/her concept of sex-appropriate

behaviors. Kohlberg (1969, p. 415) acknowledges that "there

can be little question that imitation is an extremely basic

mechanism in the formation of social knowledge." However,

neither modeling nor reinforcement is thought to operate

in any automatic way to produce sex-typed behavior.

The picture book can be seen as an important vehicle

in the development of sex-role behavior for the cognitive-

developmentalists . They emphasize the role of observational

learning and, for this researcher, picture books can be an

important source of information for children. Here the pic-

ture book can be seen as the means by which children can widen

their concept of what is appropriate sex-role behavior. Chil-

dren can be exposed to the feelings, attitudes and behaviors

of children and adults that are not in their immediate en-

vironment, thus, extending their information and broadening

their generalizations. Unfortunately, most picture books

offer a very limited and stereotypic view of sex-role be-

havior for children to internalize.

In general, each theory of sex—role development pre-

sents an important analysis of sex-role learning, but tends

to underemphasize the importance and influence of the other

22

theories. Based on the presented review of the literature,

this researcher hypothesizes that all three learning

theories contribute to the understanding of sex-role dif-

ferentiation. That is, sex—role differentiation takes place

primarily through learning processes involving: modeling-

imitation, reinforcement, and cognition based on observation.

Picture books have the potential to act as tools that imple-

ment that development. These three theories of sex-role

development are rooted in well established mechanisms of

learning. Learning (including sex-role learning) is a com-

plex process and is not solely dependent on any single proc-

ess, but rather occurs through an integrative system that

utilizes a variety of mechanisms.

Picture Books in the Child's Environment

Little information is available that directly ad-

dresses whether picture books have any effect on children

and represent a significant element in their environment.

There is, however, some indication that picture books are

readily available to many children in our culture. Rheingold

and Cook (1975) examined the rooms of 96 middle-class chil-

dren under six years of age. They found that the number of

children's books increased markedly from a mean of 1.9 at

years one to a mean of 60.2 at years six. Young children

can take books out of libraries and often receive them as

Publishers of books that receive the Caldecott Awardgifts

.

23

can expect sales of 60,000 copies, and are ordered by prac-

tically all children's libraries in the United States. The

^^ttle Golden Books, that are available for purchase in

grocery stores often sell 3 million copies (Weitzman et al.,

1972, p. 1127). In addition, picture books are considered

an important part of the curriculum in schools and child-care

centers for young children. It seems evident that picture

books can be considered a part of a child's environment. It

is important now to consider how picture books, as tools in

the development of sex-roles, present sex-roles.

Sex-Role Stereotyping in Children's Books

Sex of characters . There have been a few articles written on

the subject of sex-role stereotyping in children's picture

books. These articles vary in scope and quality, but all

agree that most picture books present a clear sex-role

stereotypic representation of the roles, attitudes and be-

haviors of males and females. Nilsen (1971) reviewed the

80 books that received the Caldecott Award^ and runners-up

of the last twenty years and 58 books used by teachers of

children’s literature at Eastern Michigan University. The

^Although the specific criteria for selecting an

award-winning book is not given, the committee that presents

the award is made up of professionals and they are invited

to serve on the committee by the American Library Associa-

tion Children's Services Division. The award is given an-

nually to the artist of the most distinguished picture book

for children published in the United States during the pre-

ceding year.

24

rationale for using the Caldecott winners in her sample high-

lighted the fact that these books are considered the most

distinguished picture books for children published in the

United States and are selected by the Children's Service

Division of the American Library Association. in addition,

another author comments that the Caldecott winners are

clearly less stereotyped than the average book, and do not

the most blatant examples of sexism (Weitzman et

al., 1972). Nilsen (1971) found that 386 females were pic-

tured compared to 579 males. Of the 80 books, there was not

a single one that did not have a male (human or animal), but

there were six books in which females were completely absent.

Trumpeter and Crowe (1971) who also analyzed Caldecott win-

ners, found that males in 23 of the 33 books reviewed oc-

cupied starring roles as opposed to females who starred in

three books. In addition, Weitzman (1972) reviewing the

titles of books receiving the Caldecott Award found eight

titles with male names, three with female names, one with

both a male and a female name together, and twenty-two titles

without names of either sex. This results in an 8:3 male/

female ratio.

The statistics for the titles of Newburry winners

show twenty titles with males names, six with female names,

none with both, and twenty-three titles without the names

of either sex. This results in a 10:3 male/ female ratio.

Weitzman (1972, p. 1129) suggests that due to the under-

25

representation of females in the titles, central roles, pic-

tures and stories, girls may "receive the impression that

they are not very important because no one has bothered to

write books about them.

"

Qf characters . A more subtle indication of sex—role

stereotyping is gained from an analysis of the activities of

male and female picture book characters. Czaplinski (1972)

analyzed the 32 books that received the Caldecott Award from

1941 to 1972 and the 32 picture books that received the Car-

2roll Award since 1958. The study was constructed around

scales measuring physical, intellectual and emotional activ-

ity. Czaplinski found that males are given a wider range

of activity levels, their number of actions is far greater

than the number of actions performed by females. Females

performed more actions when they were portrayed sleeping,

fantasizing, watching, making small hand movements, articu-

lating concern, and crying. Males are most often shown

exerting power over others, taking physical risks, taking

trips, building, inventing, controlling others, solving prob-

lems, and aggressively expressing anger. Other researchers

tend to corroborate these findings.

^The Carroll Award is sponsored by the University of

Wisconsin and is given annually (since 1958) . Publishers

submit books to a reviewing committee made up of publishers,

librarians, teachers, parents and editors who review the

books on the basis of imagination, originality, genuine emo-

tion, consistent characters, plausible events, and a plot

that unfolds gently and logically.

26

Weitzman (1972) indicates that most girls in picture

books are passive and immobile, they are most often found in-

doors engaged in cooking, cleaning and serving activities or

peering out of windows observing the activity of others.

Males are pictured as adventurous and clever, they rescue

others and are generally very independent and portray leader-

ship behaviors. In addition, Heyn (1969) indicates that

there is an extreme scarcity of women projected in leader-

ship, political, or creative positions. It is interesting

to note that "not one woman in the Caldecott sample (winners

and runners-up for the past five years) had a job or a pro-

fession" (Weitzman, 1972, p. 1141)

.

Language . Language is another indicator of sex-role stereo-

typing found in picture books. Kay (1971, p. 175) focuses

on early signs of linguistic development gleaned from chil-

dren's picture books. She found in the books that she re-

viewed :

. . . a pathetic lack of conversation with bright,adventurous females of any age. Rarely is there a

give-and-take dialogue in which a female is shown

to be capable of making a decision or where the in-

put of the female is intelligent and useful informa-

tion. The things which girls and women say in these

books too often reflect the stereotypes of society:

'Women are emotional.'. . . If one compares this

image which crystallizes in the formative years of

child development, with the potential of women in

adulthood, it becomes apparent that both male and

female have difficulty in participating in equal

sharing dialogues at the professional level. Males

who have grown up learning dialogues as are in

children's books today are not able to listen to

a female in adult life. . . . There are no linguis-

27

tic models in this early literature for females totake active parts in the dialogue nor for males torespond with dignified acceptance and a willingnessto listen.

Nilsen (1971, p. 922) also discusses the use of lan-

guage. She points out that the English language contributes

to sex-stereotyping since there is no neuter which can be

used for inanimate objects or animals of unknown sex. Con-

sequently, "he" is usually the choice of writers who may not

be completely aware that they are sex-stereotyping the object

or animal. Nilsen suggests that this is a critical issue be-

cause she assumes that children interpret language quite

literally. For example, if children hear such expressions

as chairman, brotherly love, ten-man team, and fellow-man,

they think of men, not the whole human race. With further

research, this very believable assumption could be clarified.

Although this is an area where research needs to be done, it

is simply helpful that Nilsen raises these important ques-

tions.

From the above discussion, it is apparent that picture

books reflect the rigid sex-role stereotypes found in our

culture. The question that seems to follow is whether there

is evidence to indicate that picture books affect children's

behavior. One researcher, Jennings (1975) , using 32 pre-

school boys and girls, found that children significantly re-

membered longer and in more detail ,stories with reversed

sex-roles as compared to stories where the main character be-

28

haves in a manner appropriate to his/her sex. Results of a

study by Mussen and Distler (1960) showed that subjects

Identified themselves more with the same-sex leading charac-

ter and tended to recall somewhat better the actions and

verbalizations of the model of their sex. In addition, when

the leading character in a story is of the same sex as the

reader and is also presented as a hero figure, the behavior

of children can be strongly affected. For example. Masters

and Driscall (1971) using a story in which "Tarzan" was the

main character, found that boys modeled the behavior of the

protagonist more often than did girls. In a similar way,

Ducker (1938) found that children's taste preference could

be altered upon hearing a story about "Eaglefeather .

"

Eaglefeather and Tarzan act as symbolic models

(models which are presented through oral or written instruc-

tion, pictorially, or through a combination of verbal and

pictorial devices (Bandura and Walters, 1963)), and the

limitation of a model has been demonstrated to be a power-

ful mechanism in the acquisition of a wide variety of behav-

iors in children. For example. Wolf (1973) found that boys

would more likely play with a toy stove or doll if they had

previously seen a boy playing with these and the same was

true for girls playing with sex-inappropriate toys.

Maccoby and Jacklin (1974) found that imitation is

more likely to occur when learners have little information

or when their own past experience provides little guidance

29

as to what behavior is appropriate. In addition, Mischel

(1966) states that the manner in which the model's behavior

is presented, with respect to frequency, rate and clarity of

presentation, critically affects the extent to which the

modeled behaviors are acquired. From this research it would

seem that a picture book could be a highly effective medium

to transmit sex-role behaviors.

The picture book offers a visual and verbal presenta-

tion, thus adding to the clarity of its message. The fre-

quency is high in that the message is often stated and re-

stated, and the picture book is directed at an audience that

is in the process of developing their sex-roles and by the

nature of their age have a minimum of information concerning

sex-role behaviors. Picture books often demonstrate some of

the criteria established by researchers in that picture

book characters are similar to the children hearing the

stories, but these characters often possess some qualities

that the audience is meant to admire. Mischel (1970, p. 45)

acknowledges the importance of picture books when he states;

"Undoubtedly, books and stories, and other symbolic media

play an important part in the transmission of information

about sex-typed behavior and the diverse consequences to

which they may lead when displayed by males and females."

Relationship Between Picture Books andTelevision

30

Although it seems reasonable to say that picture

books can have an effect on the behavior of young children.

Particularly their sex-role behaviors, research addressing

this problem is limited. More research of this kind has

been conducted using another media, television. Picture

books and television are similar in that both mediums offer

two dimensional visual representation and narrative com-

ponents, both are seen as a means of instruction and offer

children a vehicle for contact with the world outside their

immediate environment, both can be seen as a means for the

presentation of social values, and both present symbolic

models. Because of these similarities, some of the conclu-

sions drawn from the research of television may extend to

picture books.

Researchers using television and television-like

formats have found that children learn, and then often use,

novel acts they have observed (Liebert, Neale and Davidson,

1963) . The Educational Testing Service (Miles, 1975, p. 12)

"found that television can be an effective tool for teaching

children many of the things that the program intended them

to learn. " These tests were designed to measure specific

learnings, i.e., body parts, letters, numbers, forms, match-

ing, relationships, sorting and classification. The children

who watched this program (Sesame Street) improved more than

31

children who did not. in addition to these specific learn-

ings, social behaviors can also be affected. Friedrick and

Stein (1973) conducted a study of 93 preschool children in a

nursery school setting showing them one of three types of

television programs in a four week period in order to deter-

mine if television could promote prosocial or positive be-

havior. Children's free play was observed prior to, during

and after the showing of the film. The researchers found

that children followed their viewing of aggressive material

with an increase in interpersonal aggression and that pro-

social programs increased sel f-regulatory behaviors.

Fryrear and Thelen (1969) using 60 three- to four-

year-olds found that after the children viewed films of

affectionate behavior, they later modeled those behaviors

in a free play situation, especially when like-sex models

were used. Sex-role behaviors were also found to be affected

by television viewing. Frueh and McGhee (1975) found that

elementary school children who are considered "high" televi-

sion watchers (25 hours per week or more) received a high

stereotypic sex-role rating (as measured by the IT Scale)

as compared to low television watchers (10 hours or less per

week) . Thus high amounts of television viewing are clearly

associated with stronger stereotypic sex-role development.

Frueh and McGhee suggest that this is due to the highly tra-

ditional stereotypic portrayal of sex—roles on commercially

produced children's television programs and commercials.

32

Turning to specific sex-role behaviors, in a study by

Wolf (1975) , six-year-old boys and girls were exposed to a

same or opposite-sex televised peer model who played with

and made unusual responses with a toy that was sex-inappro-

priate for the subject viewing the model. After the subjects

were exposed to this televised condition, they were given an

opportunity to play with the sex-inappropriate toy. In ad-

dition, the subjects were asked to exhibit as many unusual

responses that the model performed with the toy. Wolf found

that girls touched the sex-inappropriate toy more quickly

and longer than did boys. In addition, the subjects were

more influenced by a same-sex model than by an opposite-sex

model. No difference was found between boys' and girls'

exhibition of the modeled unusual responses in the play ses-

sion. However, boys displayed significantly more unusual

responses than did girls when the experimenter asked them to

demonstrate all the unusual behaviors they could remember.

The ability of a televised model to affect specific

sex-role behavior was similarly demonstrated by Kobasigawa

(1968, p. 94). He used live, filmed, and orally presented

models in his studies and he indicates that his findings sup-

port "the importance and prevalence of modeling processes

in the development and modification of children's behavior

in general .

"

These studies seem to indicate that children s be-

haviors are affected by television, and sex-role behaviors

33

are a part of this learning. How does television present

sex-roles? Several studies employing a content analysis ap-

proach have found that the portrayal of sex-roles is stereo-

typic and much of it perpetuates a rather biased reflection

of the world around us (Women on Words and Images, 1975; Lie-

bert et al . , 1973) . Although the present study is not pri-

marily concerned with the influence that television per se

has on the development of an individual's sex-role, it is

concerned with the effect that another medium, picture books,

has on that development. Picture books and television both

present symbolic models, and the research indicates that sym-

bolic models can affect sex-role behaviors. The present

study was designed to determine whether having a picture

book read to a child, which presents a symbolic model, will

affect sex-role behaviors.

CHAPTER IIIMETHOD

To test the following hypotheses, thirty-two chil-

dren were randomly assigned to four groups (Figure 1) and

the opportunity to play with a select group of mascu-

line, feminine, and neutral toys. Following this pretest,

each subject was read either a stereotypic or non-stereotypic

picture book (experimental treatment) . A post-test followed

in which the subjects were again allowed to play with the

same toys and their selections recorded.

FIGURE 1

RESEARCH DESIGN

GIRL

BOY

The Study hypothesizes that characters in children's

literature model sex—role behaviors and these behaviors in

turn affect the sex-role behaviors of children. Specifi-

SEX-ROLE STEREOTYPIC SEX-ROLE NON-STEREOTYPIC

n=8 n=8

n=8 cx>IIC

Hypotheses

34

35

cally

:

1. Three to five-year-old girls exposed to a female

stereotypic picture book will select a female stereotypic toy

with which to play significantly more often than three to

five-year-old girls exposed to a non-s tereotypic picture

book

.

2. Three to five-year-old girls exposed to a non-

stereotypic picture book will select a non-stereotypic toy

with which to play significantly more often than three to

five-year-old girls exposed to a female stereotypic picture

book

.

3. Three to five-year-old boys exposed to a male

stereotypic picture book will select a male stereotypic

toy with which to play significantly more often than three

to five-year-old boys exposed to a non-stereotypic picture

book

.

4. Three to five-year-old boys exposed to a non-

stereotypic picture book will select a non-stereotypic toy

with which to play significantly more often than three to

five-year-old boys exposed to a male stereotypic picture

book

.

Design Rationale

In an attempt to construct an experimental design

that would minimize contamination of the data, a review of

the various studies using mascul ine/feminine tests was made.

36

For the most part, these research designs were constructed

so that the measures of the subjects' masculine or feminine

tendencies (Terman and Miles, 1936) were derived from self

reports of attributes, attitudes, values, and preferences.

Problems that result from this method are the possibilities

that: (1) the subject is merely saying what he/she feels

the experimenter wants --conforming to social expectations,

rather than actually revealing their own preferences; (2) the

subject doesn't wish to disclose the information to the re-

searcher; and (3) the subject is not aware or can't clearly

articulate his/her reasons.

Other measures of M-F have avoided asking the sub-

ject to characterize his/her own traits and instead utilize

a projective technique (such as the "it" Scale for children

(Brown, 1957)) . A frequent criticism leveled against this

technique is that the "IT" actually looks more like a boy

than a girl, and many children think they are supposed to

make choices for a male figure rather than projecting their

own choice onto "it" (Brown, 1962)

.

Other experimental techniques require the subject to

indicate the kind of toys, games r and activities (all shown

on pictures) he/she would prefer. The Toy Preference Test,

and Picture Test, are examples of such designs (Sears, Rau,

and Alpert, 1965). Rabban's (1950) design is basically a

variation of the toy preference test, but uses real toys as

opposed to pictured toys. In addition, Rabban's choice of

37

masculine/feminine toys reflects the inherent assumption that

masculinity and femininity are polar opposites. Rabban's toy

selection, however, appears to have been rather limited.

Of the eight "feminine" toys, five include dolls and of the

eight "masculine" toys, five represent some kind of vehicle.

The method of observing children's spontaneous play

to determine sex-role preference has also been used with

varying success (Bridges, 1927, 1929; Farwell, 1930). Re-

ports of such investigations cite limitations in this method.

Bridges stated that the children were hindered or sometimes

forced to use materials they did not prefer because the

material they actually preferred was being used by other

children. Farwell considered the addition of new materials

an influencing factor in her study—the children preferred

the new to the old. Adult suggestion and the impact of the

peer group have also been shown to strongly influence behav-

ior (Fagot and Patterson, 1969)

.

Although the shortcomings of these measures are ap-

parent, they have provided a valuable foundation for the de-

velopment of this study. To demonstrate the maximum effect

of the experimental condition, a classic pre-post-test de-

sign was implemented. The toys used in these tests were se-

lected based on a review of the toys used by other research-

ers and were most often found to be stereotypic. Toys as op-

posed to pictures were selected because it was felt that the

influence of the experimenter on the subject's verbal re-

38

sponse could not be effectively eliminated, and the use of

the actual item was preferred to a verbal choice of a sym-

bolic (pictured) toy. In addition, children in the age

range of the experimental population are considered to be at

the concrete operational stage of development (Piaget, 1964)

,

where learning is tied to direct perception and action.

Therefore, the use of real toys perhaps stimulated the child's

involvement in the experiment. In addition, the toys pro-

vided a natural extension to the materials and activities in

the preschool classroom, thus creating a somewhat naturalis-

tic setting. To reduce the possibility of misrecording the

children's play behavior, both the pre- and post-tests were

videotaped, allowing for a check on reliability.

Subjects

The sample consists of sixteen boys and sixteen

girls or a total or 32 children ranging in age from 2.7 to

5.4 years, who are attending two nursery school classes at

the Human Development Laboratory School at the University

of Massachusetts. The school typically enrols children of

the professional staff and students of the University, but

enrolment is not limited to the academic community. The

school is located on the first floor of a building located

on the campus of the University. The two nursery classes

occupy the same classroom, one however meets in the morning,

the other in the afternoon. The random placement of subjects

39

into the four groups resulted in an average group age ranging

from 4.2 to 4.0 years. Children of this age range were se-

lected because evidence suggests that by the age of three

children are able to make basic distinctions between the

sexes and are able to distinguish themselves as being either

boy or girl (Kagan, 1969) . In addition, by the time the

child is five he/she is aware of the different behavior pat-

terns that are expected of men and women, and boys and girls

in this culture (Brown, 1956; Fauls and Smith, 1956; Rabban,

1950) . This awareness increases with age, with preschool

children showing the most flexibility. It was hoped that

this flexibility would allow for the experimental treatment

(exposure to a picture book) to have an effect on their sex-

role behaviors.

Setting

The experimental room is located down a corridor

from the nursery classroom. The room contained a large of-

fice desk and adult chair placed against one wall, a table

and two chairs designed for the use of preschool children.

The child's table was located beneath a one-way mirror and

the two chairs were arranged in one corner of the room.

Equipment

Toys. Researchers have conducted studies that analyze a

child's overt behavior or his/her verbalized preferences of

40

play materials in order to determine the development of a

child's sex-role. These studies assume that the choice of

play material is a reflection of a child's sex-role develop-

ment because the materials have been categorized as mascu-

line or feminine in this culture (Biller, 1971) . The re-

searchers used a variety of techniques to determine the mas-

culinity, femininity, or neutrality of the toys used in

their studies. From a review of these studies, six toys

were selected that represent highly masculine, feminine, or

neutral toys for use in this study. Two of the toys, the

gun and the dump truck, represent male stereotypic toys

(Rabban, 1950; Rosenberg and Sutton-Smith , 1959; Brown,

1956; DeLucia, 1963). The doll and dishes represent female

stereotypic toys (Rabban, 1950; Brown, 1956; Benjamine,

1932; Hartup, Moore and Sager, 1963; Fogot and Patterson,

1969) . A ball and peg board represent neutral toys (Vance

and McCall, 1934; Hartup, Moore and Sager, 1963). The final

selection of masculine, feminine and neutral toys was made

on the basis of similarity in color, size, cost, and manipu-

lative interest (Benjamine, 1932), and were all purchased

in the toy department of local stores.

The dump truck was made of red and white plastic

and measured approximately twelve inches. The back of the

truck could be moved up and down and the large wheels turned

freely. The gun measured seven inches in length and was a

metallic color. The trigger could be depressed. The doll

41

was made of plastic and measured twelve inches. The doll was

dressed in red slacks and a blue turtleneck sweater and had

short brown hair. The doll did not appear to be either mas-

culine or feminine. The head, eyes, arms and legs of the

doll were movable. The dishes consisted of two cups and

saucers, two spoons and a pitcher. They were made of plas-

tic and metal and were colored red and chrome. The rubber

ball was red, white and blue and measured five inches in

diameter. The peg board was made of wood and measured four

by six inches. A clear plastic cup contained five red and

blue pegs.

Picture books . Four picture books were developed for this

project. Each book describes a male or female child select-

ing a toy (either a doll or dump truck) from a grouping of

toys present in the child's room. The story and illustra-

tions go on to describe five ways the main character manipu-

lates the toy selected. The books were constructed so that

they would appear to be similar to other books the child may

encounter. A special reproducing process was used to insure

that the text, illustrations and general appearance of the

books were consistent. The sex of the main character was in-

dicated by the altering of hair, facial features, clothing

and name. The toys represented in the book were illustrated

so that they matched the toys available in the experimental

setting. (See Appendix A for copies of the experimental

42

books -

)

Non-stereotypic picture books . Book one depicts a

male child (Andrew) choosing a female stereotypic toy (doll)

with which to play. Book two depicts a female child (Kathy)

choosing a male stereotypic toy (dump truck) with which to

play.

Stereotypic picture books . Book three depicts a male

child (Andrew) choosing a male stereotypic toy (dump truck)

with which to play. Book four depicts a female child (Kathy)

choosing a female stereotypic toy (doll) with which to play.

The picture books that have been developed offer the

child symbolic models. Masters and Driscoll (1971) define

symbolic models as models which are presented through oral

or written instruction, pictorially or through a combination

of verbal and pictorial devices. In addition, Mischel (1966)

describes the symbolic modeling process as an extremely ef-

ficient and effective method of transmitting information.

Research has shown that a variety of conditions aid this

modeling process. The perceived similarity between the model

and observer has been shown to increase the amount of ob-

servational learning in laboratory studies (Kagan, 1958;

Rosenkrans, 1967) . Children have been shown to attend more

closely to same-sex models (Maccoby and Wilson, 1957; Mac-

coby , Wilson and Burton, 1958) and observational learning

was promoted by exposure to same-sex models (Hetherington

and Frankie, 1967). Wolf (1973) found that when a child al-

43

ready has built up inhibitions against playing with a given

toy, a same—sex peer can symbolically "give permission,"

or make the situation seem safe, to play with the normally

forbidden toy. Therefore, every attempt was made to repre-

sent the main character in the book as being similar to the

subjects in the experiment. The storybook characters were

similar in age, dress, sex, and in morning routine to be

followed. The sex of the subject was always the same as that

of the storybook character. In addition, the toys represent-

ed in the story are commonly found in the rooms of many chil-

dren.

In his review of the research on imitative behavior,

Flanders (1968) indicates that the degree of realism of the

model's performance is another condition that affects the

modeling process. Bandura, Ross and Ross (1963) found that

the more remote the model was from reality, the weaker would

be the tendency for subjects to imitate the behavior of the

model. Taking this data into consideration, the books used

in the present study were constructed so that the behaviors

of the main characters were typical of middle class children.

The subjects participating in the study were also considered

to be from middle class families.

Kagan (1958) maintains that it is not only important

for the subject to perceive similarity between the subject

and model, but he or she must experience some of the desired,

affective goal states of the model. Thus, if the model were

44

successful or happy and the subject believed that the model

was experiencing positive affoct, the subject would also feel

positive affect, and this experience would aid in the model-

ing process. Although specific situations that would stimu-

late positive affective states in the subjects were not di-

rectly known, it was assumed that demonstration of some basic

positive human feelings would be desirable to the subjects

in this study. Therefore, the characters in the picture

books both verbally and behaviorally demonstrated their en-

joyment and sense of accomplishment. The characters hold the

toys closely demonstrating warmth and affection for the toys,

they describe how they "loved" playing with the toy, and

through their rapt attention, demonstrate their enjoyment

in mastering the play material that they manipulated.

In addition to the physical and emotional similari-

ties between the model and subject, Mischel (1966) indicates

that the frequency, rate and clarity of the presentation

critically affects the extent to which the modeled behaviors

are demonstrated. Therefore, a clear, short, direct text

was written for the picture books. The illustrations were

accomplished through simple black and white line drawings

making sure that the background elements would not distract

from the behavior of the main character. The drawings al-

ways illustrated the text of one page thus offering a clear,

direct, well articulated message.

45

Procedure

Data collection . The head teacher and facilitator of the two

nursery classes were contacted and informed about the nature

of the study. They were told that the study was concerned

with the effect that picture books may have on a child's toy

choice and play behavior. After obtaining their approval,

a visit was made to each classroom to explain the procedure

but not the intent of the study to other teachers working

with the children and to solicit their cooperation and to ob-

tain class rosters.

Before beginning any data collection, the experimen-

ter visited each class for a minimum of three two-hour ses-

sions in which the experimenter participated in the classroom

activities and became known to the children. When the ex-

perimenter perceived that the children were comfortable with

her presence, she individually asked each child if they would

like to play with some toys that were in a room down the

hall. In time, each child accepted the invitation to play

the toys. When each child was brought into the experi-

mental room, he/she was led to a table on which the toys

were randomly arranged directly beneath the observation mir-

ror. The subjects were positioned so that they faced the

one-way mirror with their backs to the experimenter. The ex-

perimenter told the subjects that they could play with the

toys while the experimenter was to fill out a form at the

46

desk. The subject's play behavior in this pretest session

was recorded on videotape through the one-way mirror for 120

seconds. Pilot data indicated that children became anxious

to return to their classrooms after two minutes and the

period of time was sufficient for the subjects to use all

the experimental toys if they chose to. At the end of the

pretest the experimenter interrupted the subject, placed the

toys on the table as before, and directed the subject to

the two children's chairs; where the experimenter read the

appropriate picture book. Girls always were read a story

about a girl (Kathy) and boys always were read a book about

a boy (Andrew) . Following this, each subject returned to

the table with the experimental toys and was given the op-

portunity to play with these toys as before. Similarly,

the subject's play behavior was recorded on videotape for

a period of 120 seconds.

To gain some knowledge of individual response and

recall concerning the story and the toys, the subjects were

asked the following questions at the end of tne second play

session

.

1. Can you tell me what ( name of main character ) did

in the story?

2. Would you like to be 's friend?

3. Why?

4. Would you change anything in the story?

47

5. If you could take any one of these toys home

with you now, which one would you choose?

Before the subject was returned to the classroom, the

experimenter said: "The other children need to have a turn to

play with these toys (experimental toys), but you can have

one of these." At that time, the subject was presented with

the two charms and the subject chose one before returning to

the classroom. The subject's response to the questions and

the final charm selection were recorded by the experimenter.

Data Recording

The video recordings of the subject's pre- and post-

play sessions were analyzed by the researcher to obtain the

total amount of time that the subject visually regarded

(looked at) each toy, and the total amount of time they

manipulated (touched or picked up) each toy during each ses-

sion. If a subject simultaneously manipulated two or more

toys, the amount of time that each toy was manipulated was

recorded. This process was referred to as "double timing."

For example, if a subject fed the doll using the toy china,

the amount of time that the doll as well as the china were

manipulated was recorded. To obtain a measure of reliabil-

ity, a graduate student reviewed the tapes of 20% of the ex-

perimental population (that were randomly selected) . The

student was not aware of the purpose or intent of the study.

She was asked to time toy manipulation and visual regard.

48

When the results of her review were compared with the re-

searcher's, there was 90% agreement within three seconds and

within a five second span, 93% agreement. This computation

was arrived at by dividing the total number of timed epi-

sodes by the total number of episodes that agreed with the

researcher's review tihin a three second span and the proce-

dure was repeated using a five second span.

If the subject modeled any of the five behaviors

demonstrated in the book during the post-play session, the

specific behavior was recorded on a check sheet. The be-

haviors were identified by the following criteria;

Truck modeling . (Male stereotypic and female non-

stereotypic books)

:

1. Subject rolls the truck.

2. Subject lifts the back of the truck indicating

a dumping action.

3. Subject turns the truck upside-down.

4. Subject rolls the truck under a table.

5. Subject spins the wheels on the truck.

Doll modeling . (Male non-s tereotypic and female

stereotypic books)

:

1. Subject moves arms and/or legs of the doll.

2. Subject feels the doll's hair.

3. Subject holds doll closely in a rocking position.

4. Subject looks at and feels the eyes of the doll.

Subject turns the doll upside-down.5 .

At the end of the experiment the subjects chose

either a doll or car charm as a reward for participating

in the experiment. Their selection was indicated on a check

sheet and a frequency count was made for each of the four

groups

.

Comparability of Groups

To determine the comparability of the four experimen-

tal groups, the total amount of time that a group visually

regarded each toy was divided by the total amount of time

the group visually regarded/manipulated all toys in the pre-

test session. T-tests were performed and no significant dif-

ferences were indicated between groups (see Appendix B)

.

In addition, to gain an understanding of the sub-

ject's toy preference as determined by time scores in the

pre-test, mean scores were calculated for each variable

(manipulation and visual regard) for each toy group (mascu-

line, feminine, and neutral) . Although the mean scores

should not be interpreted as being statistically significant,r

the scores indicate that in terms of total playing time boys

and girls show no difference. They spend an equal amount of

time manipulating and visually regarding all toys. In addi-

tion, they similarly spend an almost equal amount of time

manipulating and visually regarding neutral toys (ball and

peg board) . However, boys play longer with male toys (gun

and dump truck) and spend a comparatively short time playing

50

with female toys (doll and china) . Girls, however, reveal a

different pattern. They spend almost the same amount of time

playing with female and male toys. This pattern has been

found by other researchers who have shown that girls demon-

strate a preference for and adopt aspects of the opposite sex

to a greater extent than do boys (Lynn, 1959; Kagan, 1964;

Brown, 1957; Hartup and Zook, 1960; Mussen, 1960) . The repe-

tition of this pattern of toy preference in the pre-test ses-

sion, perhaps indicates that the subjects can be seen as more

representative of a wider middle-class population.

Data Analysis

A split plot analysis of variance of post-test time

scores for toy manipulation and again for visual regard us-

ing the computerized Statistical Package for the Social Sci -

ences format was performed on the following variables : sex

of subject (male or female) , book type (stereotypic or non-

stereotypic) , and toy type (neutral, feminine, or masculine).

In addition, the same analysis was conducted on difference

scores (difference between pre- and post-test time scores)

.

Difference scores were analyzed so that the size and direc-

tion of shifts in behavior could be examined. Toy type was

treated as an independent variable in that it was possible

for the subject to simultaneously play with all the available

toys. Many subjects played with a number of the toys at one

time and "double timing" was used to record this play behavior.

CHAPTER I V

RESULTS AND DISCUSSION

Results

The study hypothesized that children's literature

models sex-role behaviors and these behaviors in turn affect

the sex-role behaviors of children. The results of this

study support this hypothesis. An analysis of variance of

post-test time scores indicates that no significant differ-

ences were found for the main effects of sex or book type

on either the manipulation or visual regard measures. There

was a significant interaction of Sex of Subject X Book Type

for female toy manipulation (F = 5.64, df = 1/28, p < .02)

(see Appendix B)

.

Females exposed to the stereotypic picture book

manipulated the female toys longer than the females who were

exposed to the non-stereotypic book. The mean manipulation

score for the stereotypic group was 56.62 seconds, while the

mean score for the non-stereotypic group was 35.87 seconds

(see Table 1) . Males exposed to the non-stereotypic picture

book manipulated female toys almost three times as long as

the males exposed to the stereotypic book. The mean manipu-

lation score for males in the non-stereotypic group was 44.62

51

52

TABLE 1

MALE AND FEMALE GROUP MEAN POST-TESTTIME SCORES FOR MANIPULATION AND VISUAL

REGARD OF FEMALE TOYS*

GROUPMANIPULATION VARIABLESTEREO. NON-STEREO.

Male 11.62 44.62

Female 56.62 35.87

kSignificant sex by book type

interaction P = .02.

GROUPVISUAL REGARD VARIABLE

STE REO. NON-STEREO.

Male 10.50 42.87

Female 43.75 29.25

*Significant sex by book type

interaction P = .02.

53

seconds , while the score for the male stereotypic group was

11.62 seconds (see Table 1).

Mean scores for visual regard indicate that females

in the stereotypic group visually regarded female toys for

43.75 seconds, while the non-stereotypic group had a mean

score of 29.25 seconds. The males exposed to the non-stereo-

typic book had a mean visual regard score of 42.87 seconds,

while the score for the stereotypic group was 10.50 seconds

(see Table 1)

.

These results indicate that picture books presenting

a same-sex character manipulating a female toy is a powerful

experimental treatment that significantly affects the play

behavior of young children.

It is interesting to note that when comparing the

female stereotypic group with the male non-stereotypic group

(both groups were read a similar story)

,

both groups spent

a comparable amount of time manipulating the female toys.

Perhaps this indicates that both girls and boys will react

similarly to a picture book that presents a same-sex charac-

ter playing with a doll. In addition, males in the non-

stereotypic group have a higher mean time female toy manipu-

lation and visual regard score than females who are exposed

to the non-stereotypic book.

Although the analysis of variance did not indicate

a significant result for the interactive effect of male

toys, the mean scores for manipulation and visual regard sup-

54

port the trend indicated by the female toy analysis (see

Table 2) . Males in the stereotypic group and females in the

non-stereotypic group (who were exposed to similar books)

manipulated male toys more than the other two groups. in

addition, the female non-stereotypic group had a higher mean

score than the male non-stereotypic group for male toy manip-

'^^^tion. The same pattern was indicated for the visual re-

gard mean scores.

A more complete presentation of the mean manipulation

and visual regard time scores for each group is presented in

Table 3, Pre— and Post—Test Mean Time Scores for Toy Manipu-

lation and Visual Regard. This table indicates the mean

scores by group for both the pre- and post-test, revealing

the differences in size and direction of the shift in play

behavior after exposure to the picture books.

An analysis of variance of difference scores re-

vealed similar results as those indicated in the analysis

of post-test scores. No significant main effects were indi-

cated. However, there was a significant interactive effect

for Sex of Subject X Book Type for female toy manipulation

(F = 10.45, df = 1/28, p < .003) and for male toy manipula-

tion (F = 6.80, df = 1/28, p < .01) (see Appendix B) . Fe-

males exposed to the non-stereotypic book decreased their

mean manipulation score of female toys, while increasing

their mean score for male toys. Females exposed to the

stereotypic book, however, indicated a reversal of this pat-

TABLE 2

MALE AND FEMALE GROUP MEAN POST-TESTTIME SCORES FOR MANIPULATION AND VISUAL

REGARD OF MALE TOYS

GROUP MANIPULATION VARIABLE

STEREO. NON-STEREO.

Male 66.37 58.12

Female 43.62 65.75

GROUPVISUAL REGARD VARIABLE

STEREO. NON-STEREO.

Male 51.75 50.00

Female 34.87 56 . 25

POST-TEST

MEAN

TIME

SCORES

FOR

TOY

MANIPULATION

AND

VISUAL

REGARD

56

t

(

57

tern.

Male mean scores demonstrate the same pattern. Boys

exposed to the non-stereotypic book decreased their manipula-

tion of male toys, while increasing their score for female

toy manipulation. The reversal of this was true for males

exposed to the stereotypic book (see Table 4)

.

The same pattern of significance was found for the

visual regard measure; Female toys (F = 8.16, df = 1/28,

p < .007); Male toys (F = 4.97, df = 1/28, p < .03) (see

Appendix B)

.

Females exposed to a non-stereotypic book dramati-

cally increase their mean visual regard score for male toys

while decreasing their visual regard of female toys. Fe-

males in the stereotypic group, however, increased their mean

score for female toy visual regard while decreasing their

male toy mean score.

Males reveal a similar pattern with the stereotypic

group increasing male toy and decreasing female toy visual

regard mean scores. The boys exposed to the non-stereotypic

book dramatically increase their female toy mean score and

show a slight decrease in their male toy mean score (see

Table 4 )

.

In addition, when looking at difference scores, the

effect of the experimental treatment appears stronger for

females than males. Females show a larger difference be-

tween their pre- and post-test scores than their male coun-

TIME

DIFFERENCE

SCORES

FOR

FEMALE

AND

MALE

TOY

MANIPULATION

AND

VISUAL

REGARD

58

• . •

0 cn 00 0

* K LT) in 0 • •K s inM W CN • rH * PI CN n2 Eh • . 0 Q EhO cn m 1—

1

II• CO 1 CDH 1 r-H < 1 n

Eh 1 + Oi II CJ 2 +<c 0 0PI 2 U) 04 S 2D >1CI4 0 0) PIH 4-» <2 0 D

•(U -P CO .

0 CN m f—

1

M 0 0 00 fd 0) > m in

>H W • . g rH s .

0 W 0 ro 0) (d >H w CD CTi

Eh Eh CN 1—

1

14-1 g 0 Eh 1—

1

1

CO + 1 Eh CO +W P P0 0

s MH iw a0) C c: 0)

0^ I—

1

0 0 O4 rHD QJ eg •H -H D Q) Id

0 6 4-> -P 0 rH gPi td 0) U 0 Pi (d 00 S pH fd fd 0 2 pL,

M U(D Q)

-P 4-J

C C•H 'H

0) (U

cu a+j -P

• .

0 0 0 00 0 0 PI

*. s 0 Si XI * 05 0 02 PI 0 ro 9 PI 0 00 EH . • >1 >1 Eh • •

H CO CN iH Si Si < CO 0 00

EH 1 ro 1—

1 CJ 1 ro 1

C 2 + 1 X X 0 2 +PI 0 0 0 2 0D 2 w w 2O4 PIM p -p <z c c D2 0 0 CO

• u u H *

0 0 r' •H •H > 0 0 0>H 0 ro 4-^ 14H 0 in 00 2 • . •rH •H >H K • •

Eh PI CD m C c 0 pa 0 CD

Eh rH CN tT> Cr> Eh Eh rH rH

W CO 1 + •iH •H CO 1 +PI CO CO pa

•K K PI* 2

PI 2Pi Q) PI 0

04 iH Pi Oi rH

D Q) fd D 0 00 1—

1

g Q rH ga fd 0 Pi 0 00 2 P4 a 2 Pi

Significant

sex

by

book

type

interaction

for

female

toys^P

-

.007.

Significant

sex

by

book

type

interaction

for

males

toys

P

=

.03.

59

terparts. For example, females in the non-stereotypic group

increased their mean score for toy manipulation by 41.75

seconds, while males in the non-stereotypic group increased

their score by 32.00 seconds (see Table 5)

.

When comparing the four groups, the non-stereotypic

book had a stronger effect than the stereotypic book on

both male and female groups. The non-stereotypic groups

showed a dramatic increase in their mean post-test scores

as compared to males and females who were exposed to the

stereotypic book.

Discussion

Female group toy manipulation . The significant three-way

interaction results found in the analysis of variance clearly

indicate that the play behavior of boys and girls is affected

by the experimental treatment. Reviewing the mean pre- and

post-test scores helps to clarify the significant interac-

tive results (see Table 6)

.

Females exposed to a non-stereotypic picture book

increase their mean manipulation score of male toys, while

decreasing their score for female toys. The highest mean

toy manipulation score for this group in the pre-test was re-

corded for the doll at 30.62 seconds while the highest score

in the post-test shifts to the gun at 34.25 seconds, with the

truck following closely at 31.50 seconds.

Females exposed to a stereotypic book reveal a rever-

60

LO

W

sEh

•

O«pa o o (N CN O OEh o o lO CO in inCO • • • • • •

rH CO o 00 o COu 1 1—

1 CM 1 t

—

i rH

§ 2 1 + 1 +

< sowPi •

oC SD uCO Eh in o in CN O mM> CO CM o CN CO O CN

2 (N cn cn o 1

Q 1 ro rH cn

§pa + 1 1 +

ZoHEh •

< oi-q

D Sa< pa in m r' o oH CN m m o m2 CO • • • • • •

3 m m ro 00 CO cr>

2 pa m (N rH CN (H 1

1 + 1 1 +>H

o spaPm

Di

oPm •

OCO

aS paOu CO m r' m in o r-

CO 1 (N m r' CN o m2 • • • • • •

w m iH rH n 00 CO

u pa 1 i~< 1 1 m2 m:i 1 + +PJ

pa patM t.PmMQWS •

H a •

Eh •H a. •

c •H •

2 (d c a . Pi

2 fd •H > •

U S c > Pi

2 >1 (d >1 •

0 >1 s O >Eh o Eh O

Eh >1 Eh >1

0 i-H 0ttJ (U Eh fd 0) Eh

>H rH H *H4J (d Q) P fd QJ

{3 e rH 0 B rH

0) QJ td QJ 0) fd

2 Pm S 2 Ph S

61

99

DQ

W(Xi

§O Cn

Q2

W

Cm Ci3

DCO CO

ouCO

VO o CTi 00 VO CN in CN H* H* no rH CN 00 00 in O VO CTi 00Q 00 00 in o cn 00 rH 'cr in VO >o 00 00 in CN o 00 iH CN VO CJV

COCN 00 o o 00 (T> 00 OO VO o in VO o 00 rH OsJ OO O (T> o(N 00 00 TT OO CN OM CN Ovj rH 00 CN rH •H <-{ 00

Q Ehz CO

pa CN m o in in o o in o in O o in CN o m in CN in

2Eh1

< VO•

in CN (N in in o 00 in m VO m CN 00 rH 00 00 n-

O Eh (T> 00 VO VO rH cr« CN r- 00 rH CN CN 00 o CN 00 OM 00 00 r'HEh

COO

rH r-| OO CN 00 CN CN rH rH rH OO CN iH rH

< a•

D CO s •

a >H o o (N CN CN 00 VO CN rH Osl o CN CJV o CN OM O' iH CNM oCO

pa

00 nH rH m 00 O' H* rH ov CN CN VO VO H* 00 o cjv VO 00Eh rH tH CN CN CN CN rH rH rH • CN rH rH OO

QCO

Eh

COpa 2Eh sW §Pia

sDCO

E-t

H>MEhU<

CO

00 CN (N o VO 00 CJN 00 O' 00CN O VO in VO CN OO O' rH

7. 4,

CO o o 9.P' 00 00

CN CN rH rH rH OO OM rH rH

o o O r' m CN CN CN OM OM CNin o 00 in CO CN rH VO VO VO VO VO

4. 3.rH H* 1

—1

9,

00 O CN VOCN OO fH rH OO OM rH rH

VO H* C3> CN ov o 00 CTV in rH 00 O'OO CN (T> rH (Ti r' VO 00 OO rH

11 rH rH CN 1—

1

rH rH

a•H •

c cc:

(d • •

• S > a •

a • •H • a•H • TJ TJ a c Pi •H •

c Pi U U •H • (d • c a(d • (d (d c Pi s > (d •

S > 0 0 (d • s >(H a s > M

rH rH u O rH iH

rH rH CP1 CP c c 3 3 rH rH

fd (d 0) 0) 3 B M U 0 0CO CQ a a C5 Eh Eh Q Q

a•Hc(TJ

s

(25

UC•Hjg x:u u

(Nr-ic3>rOrH'^0'^ror^oooc^llr)CN^Hr^oooM'X>n^o^^

ooovovoot^r^cT>r^roo(T>rHrHrOrO'^CNrNCNrH.HCNrH

or'C^minocNr^or^mr^orororM(NOr-i<^irif^r^f^

CNr'(NrH>-ICX3'sDvOr-l(J\Cr»CX)nrO-^TCN^rHiH iHiH

v£>CT^o^oO'^•o^tHCNlna^r'cr»ir>mmn(N(Nro<yir^in^

<NrMrncN.HrH rHr^

a•H •

C ct5

n •

S >

(C3 it3

a•H •

C Ct5

icJ•

S >•

T3 'C3 Cij

M -H(T5 fO

0 O(U CQ

Cnj

S

Pi

Cu•Hc(tj

s

pt::

CP CP0) Q)

aHc(T3

S> ^5 >5

U U -H

a• c

05 (TJ

05

C 33 U

CQCDaaC5CDEHEH

3 rH rH

s >rd (d

C CHJG

•HJG

U U

62

sal of this pattern. They spend more time manipulating fe-

male toys and less time with male toys. In addition, the

mean manipulation score decreases for every toy in the

post-test except china which increases by 28 seconds to be-

come the highest score recorded for this group.

The pattern revealed by the above review is supported

by comparing the two female groups (stereotypic and non-

stereotypic) . Females in the stereotypic group have a higher

female toy manipulation score than the females in the non-

stereotypic group. On the other hand, the non-stereotypic

group received a higher male toy manipulation score.

In summary, these results support the original hypo-

thesis. Girls exposed to a female stereotypic picture book

will more often select a female stereotypic toy with which to

play than girls exposed to a non-stereotypic picture book.

Girls exposed to a non-stereotypic picture book will more

often select a non-stereotypic toy to play with than girls

exposed to a female stereotypic picture book.

Male group toy manipulation . The results of the analysis of

variance of difference scores indicates a significant inter-

active effect for male toy manipulation and visual regard

measures. This significant interaction is delineated when

analyzing the difference between pre- and post-test mean

scores. Males, exposed to a non-stereotypic picture book

(male character plays with doll) increase their mean manipu—

63

lation score for female toys while decreasing their score for

toy manipulation. In addition, this increase is pri-

due to an increase in doll manipulation rather than

china manipulation (see Table 7)

.

Males exposed to a stereotypic book (male character

plays with truck) reveal a reversal of this pattern. Their

score for male toy manipulation increases while their score

for female toy manipulation decreases. The increase in male

toy manipulation is directly related to an increase in truck

manipulation (rather than gun manipulation) , which was the

only score to dramatically increase in the post-test.

This trend continues when comparing the two male

groups. The non-stereotypic group has a dramatically higher

female toy manipulation score than the stereotypic group,

while the stereotypic group has only a slightly higher male

toy manipulation score than the non-stereotypic group. Per-

haps these scores indicate that males rigidly adhere to the

male stereotypic pattern as demonstrated by a strong prefer-

ence for playing with male-stereotypic toys. However, that

rigidity was somewhat relaxed by the non-stereotypic group

as reflected in the lower mean male toy manipulation score

and the dramatic increase in the female toy manipulation

score

.

These results support the hypothesis that boys ex-

posed to a male stereotypic picture book will more often se-

lect a male-stereotypic toy to play with than boys exposed

64

w

9Eh

m <Ti rH H* o o 00 CJC CDQ n r' ch m 00 m CTi o rH oc (N H*CD

•

KO O CN CN cn O H* in in (Nm rH rH rH m m rr ro ro ro CN CN

Q Eh

CDCD CNJ o o o in CN CN o O

J2Eh

1

< CD•

m o in m CD r- rH rH ro in in

O Eh c» CD 00 CD O o cr\ in CT» 00MEh

CDo

nH cn m CN rH ro ro

aOi

D0^ CD s

D <T> (N CN CT> CD ro rH in CD 00M >HCO 'S' in CD in <T» H’ CD in 00 p' CDo

Eho rH rcj CN rH rH CN CN

CD

« < Pio DPm Q CD r-H «r ro CD H* m o o in in rH CD

CDH> Q

EhCT\

•

tH CD CD rH in o in CN CN CD CD

a^ M 00 (N rH O CD CTC o CN CD CD CD 00D Q CD fH rH <N (N H* CNo 2« H 1

o EhCm CD 2 r' r- o in O p' in r' P' O

pj O CD 3 00 n o CN in 00 00 ro CN ro ro inEh S O •

<! Q 1 1

—

) 00 a\ 00 H* m CD CD H* 00 00S Pi y 2 rH t-H rH in ro rH rH< «Pi O 0^o yPm 2 CD •

LD r' (N CD CD rH m t—

1

m P' 00CD y rn hD o^ CD in ro ro ro ro ro CD CD

9 rH rH (N rHK D <O CDCJ H SCD >Ws .

M Q.Eh fH •

c PCJ

*z (d • • •

• S > a • acu • •H • cu •H •

s •H • T3 TJ C Pi •H • G Pi>H c Pi >H U •H • fd • C Pi fd •

EH (TJ • fd (d c Pi S > fd • s >H 2 > 0 0 fd • s >> OQ CQ S > fd fd

M rH rH u u rH rH C aEh r-H rH CP Cn c C p P rH rH •H •HU fd (d 0) (1) 0 P M u 0 0 x: X3< CQ m 0^ {J CD Eh Eh Q Q u u

OWPi

wEh

W

W

cs

ro in rH H* CD O o (T> 00 00 P'00 rH 00 OC H* CD CN (N <T> rH in CO

rH CD (TC ro in in in rH CN ro1—

1 ro ro ro rH rH rH

P' o m CN o in in p' m in in00 00 O CN CD in (N ro p' CN p^

00 ro CO O CN rH ro o rH rH o 00ro ro CN rH

'

rH

rH rH 00 CN rH CN O CN rH CN op- ro CD H' 00 <T> in CN rH 00 p'

CN rH rH ro ro

CN rH ro 00 CN CN cn CTi 00 cr» CNin p' CO in CN CO 00 oc rH CN in rH

o 7,rH p- CN CO o 6. 4.

p' CTi1—

1

CN ro CN rH rH rH CN f-H

p- P- o p' o CN m CN CN o o oro 00 in 00 in CO CN CO CO in o in

6, 3.cr> 00 00 m 7. 9.

3.1—

1

CT\

ro ro ro CN CN rH

rH f—

1

CD rH 00 m 00 o> fN CN•

CDin ro rH rH o o in p' CN rH (Ti in

ro ro ro CN rH rH

a•H •

C Pi•

• s > a aa • •H . a •H •

•H • TJ '0 a G a: -H • G Pic u u •H • fd • G CtJ fd •

fd • Id fd G Pi 2 > fd • 2 >S > 0 0 •3 J 2 >

CQ CQ s > fd fd

rH 1—

1

U U rH rH G GrH rH CP CP G G P P .H -H •H •Hfd fd 0) 0) p p M >-( 0 0 x: siCQ CQ CM CM CD CD E-i E-i Q Q u u

65

to a non-stereotypic picture book. Boys exposed to a non-

stereotypic picture book will more often select a non-stereo-

typic toy with which to play than boys exposed to a male-

stereotypic picture book.

Modeling . To gain an understanding of the subject's modeling

of the specific behaviors presented in the books, a frequency

count was made of those specific behaviors performed by the

subjects in the post-test (see Table 8) . Although these

scores should not be interpreted as statistically signifi-

cant, they do suggest that picture books offer children

models, in that the subjects often demonstrated the specific

behaviors modeled by the characters in the picture books.

Seventy-five percent of the male and fifty percent of the

female subjects modeled the "doll" or "truck" behaviors de-

scribed earlier (see Data Recording) . Males modeled these

behaviors more often than females and also performed a wider

range of these specific behaviors.

These data suggest that exposure to picture books

may cause children to model the specific behaviors indicated

in the book. However, more research needs to be done to

demonstrate this conclusively.

Toy choice . To determine whether the picture books have an

effect on the subject's selection of a toy, each subject was

given the opportunity to choose one toy from the experimental

5, Procedure). A test of frequencytoy group (see Question

66

TABLE 8

FREQUENCY AND RANGE OF SUBJECT'S MODELING

GROUP NUMBER OF SUBJECTSMODELING BEHAVIORS

Female Stereotypic 4

Female Non-Stereotypic 4

Male Stereotypic 6

Male Non-Stereotypic 6

TYPE OF BEHAVIORNUMBER OF TIMES SUBJECTS MODELED BEHAVIORS

F STER. F N-STER. M STER. M N-STER.

1. Rolls truck 3 5

2. Dumps truck 4 4

3. Turns truckupside-down

3

4 . Rolls truckunder table

1

5. Spins wheels 2

6. Moves doll'sarms/legs

2 1

7. Feels doll'shair

5

8. Rocks dollin arms

1

9. Looks/feelsdoll's eyes

3 6

10 . Turns dollups ide-down

1 4

TOTAL 7 7 15 16

67

(see Table 9) indicated that overwhelmingly the subjects

chose male-stereotypic toys. Of the 32 subjects, 21 chose

either a truck or gun. Males adhered to a very stereotypic

pattern of choice in that they selected a male stereotypic

toy 14 times and only selected the ball and china once.

Girls on the other hand show a wider range in their selec-

tion, choosing male toys 7 times, female toys 5 times, and

neutral toys 4 times. This behavior is similar to the pat-

tern found in the pre-test, where girls spent almost the

same amount of time playing with male, female and neutral

toys. This data, however, does not clearly indicate that

the picture book had an influence on the subject's toy

choice

.

Charm choice . Although the picture book did not seem to have

an effect on the subject's selection of a toy, there is some

indication that the book may have had more of an influence

on the subject's selection of a charm (car or doll). The

subjects selected one of these charms as a reward for par-

ticipating in the study, and were permitted to return with

it to the classroom (see Table 10) . Although this discus-

sion is not based on statistically significant evidence, the

pattern revealed by the subject's charm selection may be di-

rectly related to the picture books.

Males in the stereotypic group chose the car charm

7 times and the doll charm only once, again demonstrating

68

TABLE 9

FREQUENCY OF SUBJECT''S VERBAL TOY SELECTION

GROUP BALL TRUCK GUN DOLL CHINA

Female Stereo. 2 4 2

Female N-Stereo. 2 2 1 2 1

Male Stereo. 1 3 3 1

Male N-Stereo. 1 7

TABLE 10

FREQUENCY OF SUBJECT'SCHARM SELECTION

GROUP CAR DOLL

Female Stereo. 4 4

Female N-Stereo. 3 5

Male Stereo. 7 1

Male N-Stereo. 5 3

69

the highly stereotypic pattern that the book perhaps rein-

forced. Boys in the non-stereotypic group selected the car

charm 5 times and the doll charm 3 times. Although this is

not an overwhelming split, it does indicate that this group

is more willing to choose a non-stereotypic charm than the

stereotypic group of males. This willingness to choose a

non-stereotypic charm may have been influenced by the will-

ingness of the male character in the picture book to play

with a non-stereotypic toy.

For girls, however, the book had little influence on

their charm selection. Females in the stereotypic group

chose the car and doll charms four times respectively and

girls in the non-stereotypic group chose the car charm three

times and the doll charm five times.

Story recall. In order to gain some understanding of the

subject's ability to recall the story, at the end of the

post-test the experimenter asked each subject to tell her

what had happened in the story (see Question 1, Procedure)

.

If the subject was able to accomplish this, the experimenter

indicated this appropriately on a check sheet. Although this

evidence is not statistically significant, the data revealed

that both girls and boys were able to retell the story, with

girls showing a slightly higher score based on a frequency

count (see Table 11) . This data indicates that the subjects

were interested in and able to follow the text of the story.

SUBJECT'S

RESPONSES

TO

QUESTIONS

70

71

No apparent differences were found in recall ability between

groups. In addition to story recall, the experimenter asked

each subject if he/she would change anything in the story

(see Question 4, Procedure) . Their suggestions were recorded

by the experimenter, and a frequency test was performed for

each group of subjects. Boys suggested changes more often

than girls. These suggestions were primarily the addition

of a favorite toy or allowing the main character the freedom

to play with other toys mentioned in the book. In addition,

the subjects in the non-stereotypic groups suggested adding

a more sex-appropriate toy to the story.

Friendship . The recording and analysis procedures used for

the questions of story recall/change were repeated for the

question dealing with friendship (see Question 2, Procedure).

Here again, the results should not be considered statisti-

cally significant (see Table 11) . The response to this ques-

tion indicates that girls in the stereotypic group appear to

be the most willing (as compared to other groups) to be

friends with the picture book character. Boys in the non-

stereotypic group show the reversal of this behavior. They

did not want to be the friend of the storybook character more

than subjects in the three other groups. In addition, sub-

jects in the stereotypic group often suggest that their

friendship choice was made on the basis of toy interest or

other elements of the story. The responses in the non-

72

stereotypic groups were predominantly based on affective

statements (I like, I love)

.

In summary, with respect to the question of friend-

ship, the picture book appeared to have the strongest in-

fl^snce in the male non—stereotypic and female stereotypic

groups. It is interesting to note that the female stereo-

typic group also had the highest mean female toy manipula-

tion score recorded for any group in the pre- or post-test.

The picture book appeared to reinforce the stereotypic play

behavior of girls as well as encourage a desire to be friends

with others who perform in a stereotypic fashion.

Males in the non-stereotypic group showed a dramatic

change in their pre-post-test female toy manipulation score

(increased by 32.00 seconds). Although exposure to a non-

stereotypic book appears to increase their willingness to

play with a sex-inappropriate toy, it has the opposite ef-

fect on their willingness to be friends with a character

that demonstrates sex-inappropriate behavior. Perhaps this

question reveals how strongly boys react to a sex-inappro-

priate situation when it is placed in a social context.

Summary and suggestions for future research . In summary

,

the significant results of this study support the hypothesis

that boys and girls exposed to a stereotypic picture book

will significantly more often select a stereotypic toy with

which to play (manipulate and visually regard) than children

73

exposed to a non-stereotypic book. Conversely, boys and

girls exposed to a non-stereotypic book will significantly

more often select a non-stereotypic toy with which to play

than children exposed to a stereotypic book. In addition,

the results indicate that females were more affected by the

picture books than males. Their mean difference scores re-

fl®*^ted larger pre—post—test differences than their male

counterparts. However, the non-stereotypic picture books

appeared to be a more powerful treatment than the stereo-

typic books. Both female and male non-stereotypic groups

showed larger differences in pre-post-test scores than both

male and female stereotypic groups.

Although the significant results of this study pre-

sent important evidence that supports the hypothesis that

picture books influence specific sex-role behaviors of

young children, the study was limited in a few areas. More

specifically, the experimental population consisted of a

relatively small number of children coming from a highly

educated, middle-class group. More research needs to be con-

ducted using larger and more varying groups of children. In

addition, the present study used the pre-test as a method of

control. By extending the experimental population a separate

control group could be included.

Turning to the experimental toys, the use of the gun

as a male stereotypic toy should be closely examined. Some

families and schools forbid the use of this toy, placing it

74

in a taboo category. This treatment may influence the

child's interaction with the gun. Although this problem was

not overtly apparent in this study, this researcher suggests

that the gun be removed from the experimental toy selection

and substituted with another male stereotypic toy.

Aside from these specific considerations and sugges-

tions, a more general problem that applies to research of

this nature needs to be considered. When children (in a

somewhat isolated setting) are exposed to an experimental

treatment, that gives specific directives, the children may

be "set-up" to perform as directed. Perhaps the cause leads

to the effect only in an experimental environment. There-

fore, there is a need for more research that utilizes a

variety of techniques (especially using natural environ-

ments) in order to determine more definitely the effect of

picture books on the behavior of young children. In addi-

tion, future research should consider if exposure to stereo-

typic and non-stereotypic picture books have a long term

effect on children and if that effect changes with age.

Older children, who are able to read their own books, may

demonstrate a different pattern of behavior.

Although it seems clear that there is a need for more

research in this area, the present study indicates that pic-

ture books can influence specific sex-role behaviors of chil-

dren. Therefore, the educational implications of these re-

sults are now considered.

CHAPTER V

EDUCATIONAL IMPLICATIONS

Specific Implications

This study represents an initial effort to demon-

strate the effect of one curriculum material, picture books,

on the sex-role behaviors of young children. The need for

the replication of this study and for future research in this

area is great and educators should be cautioned against im-

plementing changes in their classrooms (and homes) based on

limited research.

However, the significant results of this study clear-

ly indicate that picture books affect sex-role behaviors of

young children and these results have important educational

implications. More specifically, teachers and parents should

provide children with picture books that present children and

adults demonstrating a wide range of activities, interests,

and emotional expressions. In addition, other curriculum

materials that symbolically present sex-role behaviors (pos-

ters, games, movies) may similarly affect sex-role behaviors.

Therefore, the selection of these materials should follow

the same criteria developed for the selection of picture

books

.

75

76

Although the study indicated that the behavior of

both males and females were affected by the picture books,

females reacted more strongly. Therefore, educators should

be aware of the need to present females with picture books

that present females in a wide variety of activities. In

addition, males may require frequent exposure to picture

books (as well as other curriculum materials) presenting

males in a wide variety of activities (both stereotypic and

non-s tereotypic) in order to reduce the reticence demon-

strated by males.

Although these specific educational suggestions are

few the more general educational implications of this study

are more numerous and encompass a broader educational scope.

Just as one small function of an organism can affect the en-

tire function of that organism, the significant results found

in this study, can affect the educational process as a whole.

Therefore, it is important now to consider the more general

educational implications of this study.

General Implications

Florence Howe and Paul Lauter wrote in the Conspiracy

of the Young , "Before one can begin to change a condition,

one must believe in the possibility of change." Taking this

statement one step further, before a condition can be changed

it first must be understood. It was hoped that the present

study would contribute to an understanding of the role that

77

education plays in the development and continuance of stereo-

typic sex-roles by demonstrating the effect of one curriculum

material, picture books, on the behavior of young children.

One of the most important purposes of education is to

put all children in complete possession of their abilities

and talents. There is an overwhelming agreement that educa-

)tion is concerned with the sacredness of each individual and

that one of its prime mandates is to encourage each child to

grow and to develop to his or her fullest potential. Yet,

beginning in nursery school and continuing through an indivi-

dual's entire educational career, schools channel children

according to their sex into a culturally defined norm. This

channeling into a sex-role stereotype begins in the home and

is continued in the nursery school which is usually the

child's first institutional contact outside of the home and

can be seen as performing critical educational and socializa-

tion functions. This process takes place through a child's

interactions with teachers, peers, and curriculum materials.

Carol Joffee (1974) reports her observations of one nursery

school that was strongly committed to avoiding imposition

of sex-role stereotypes. She indicates that girls were more

frequently admired than boys, and received more compliments

on days they wore dresses rather than pants; reinforcing once

again the societal message that it is how a girl (and later

a woman) looks which is of primary importance. Boys received

positive reinforcement when tney ably defended themselves

78

against physical attack. Sex-role stereotypes were prevalent

in the storybooks read to the children, as well as the tradi-

tional children's games and songs taught at music. For exam-

ple, one song describes the tasks each character performs:

"And the daddy went spank-spank. And the mommy went shh-shh."

Joffe's case study is not atypical; most preschool

programs do not openly segregate activities according to sex.

However, girls may not be encouraged to use the workbench

and boys the housekeeping area. An example of this inadver-

tent form of stereotyping found in nursery schools is illus-

trated by this interaction (Frazier and Sadker, 1973, p.

186) :

In the nursery school there were two rooms. Theblocks, wheel toys, and manipulation toys were keptin One room, and all the dolls, role play and cre-ative arts materials in another. The mother askedher daughter one day, "Do you have a girls' room anda boys' room?" And Eva said, "Oh no, the girls areallowed to go into the boys' room too."

Curriculum materials contribute to the child's devel-

oping concept of the appropriate sex-role. Materials such

as the community worker block accessories often found in

schools for young children provide a stereotypic image of

family groupings (mother has a painted-on apron and father

has a briefcase in hand) and occupational roles. Lotto

games, puzzles, flannel board figures, charts, posters and

dolls all perpetuate the cultural norm.

Teachers' expectations and interactions may also con-

tribute to a child's sex-role patterning. Rosenthal (1968)

79

has shown that teacher expectation is one of the most impor-

tsnt determinants of student achievement. He gave teachers

false evidence of student's I.Q.'s, and found that the stu-

dents whose teachers were led to believe they were the most

intellectually capable improved more than a group of students

whose teachers were not given this false information. This

evidence is somewhat controversial; many studies have not

replicated their findings while others have (Thorndike,

1968 ) . Even with rats, those experimenters who were told

that their rats had been especially bred for brightness ob-

tained better learning from them than did those experimenters

believing their rats had been bred for dullness (Rosenthal

and Fode, 1960 ; Rosenthal and Lawson, 1961 ). Therefore, if

teachers expect girls to be submissive, dependent, docile,

diligent, good at details and to like domestic activities,

and if teachers expect boys to play tough, ask questions,

be logical, like math and science, these expectations may

operate as a self-fulfilling prophecy.

Turning to differential treatment in terms of teach-

er-interactions, research indicates that males receive high-

er rates of teacher—child interactions, including competence

encouragement and more of an effort is made to keep them in-

volved in the activity at hand. However, boys also receive

significantly more disapproval or blame than the girls did

(Meyer and Thompson, 1963 ;Appleton, 1975 ) . Research done

by Spaulding ( 1963 ) ,supports these findings, but also indi-

80

cates that teachers interacted more with boys than with girls

on every one of the four major categories of teaching be-

havior; approval, instruction, listening to the child, and

disapproval. In addition, a disproportionate number of

teachers' negative remarks to girls concerned incorrect an-

swers. One consequence of these findings might be a cumula-

tive increase in independent, autonomous behavior by boys as

they are approved, praised, listened to and taught more ac-

tively by teachers. Another might be a lowering of self-

esteem generally for girls as they receive less attention

and are criticized more for lack of knowledge and skill. In

fact, the Sears study ( 1963 ) found bright fifth- and sixth-

grade girls to be significantly lower than boys of the same

intelligence in their own self-concepts of mental ability.

The stereotyped educational materials found in nur-

sery schools continues into elementary school. School

readers are a glaring example of such a material, represent-

ing an extension of the picture books found in the nursery.

These readers have a special influence on the socializing

process of children. They are used in state public schools

which are attended under state compulsion until a certain

age is reached. In many states they have to be approved by

state officials since they are purchased with state funds.

Therefore, the readers arrive in the classroom with the sup-

port and authority of a bureaucracy. Finally, every child

must read them. Through the readers, society says, "This is

81

what we would like you to be." The studies of elementary

school readers usually follow a content analysis format. The

clearly indicate the presence of strong stereotypic

sex—roles matching the results found when picture books

were analyzed (Miles, 1971; U'Ren, 1971; Women on Words and

Images, 1972; Michigan Women's Commission, 1973). Even in

elementary math texts, sex stereotyping is blatant. Mathe-

matical problems in one series of math texts presented math

concepts in social contexts that strongly reinforced stereo-

typic sex-roles (West, 1971)

.

What is the result of all this educational stereo-

typing? 'For girls, the school's expectations and the tradi-

tional sex-role expectations are congruent and provide a

strong double-barreled message reinforcing girl's obedience,

docility, and dependence. Nevertheless, these qualities for

which the girls receive good grades and teacher's praise

have little to do with active intellectual curiosity, analy-

tical problem solving, and the ability to cope with chal-

lenging material. The passive approach to learning, this

avoidance of the challenging situation, the fear of failure

are obviously unfortunate detrimental attitudes and can be

translated into a decline of confidence as documented by

Sears' research (1963). Girls are generally found to under-

estimate their ability, while boys overestimate theirs

(Brandt, 1958; Crandall, 1969). By effectively reinforcing

a passive approach to learning, the school runs the risk of

82

decreasing the female student's self-concept of ability.

Ironically, while attempting to increase potential, the

school, in reality, may be likely to limit it.

For boys the educational package is not very encour-

aging. The schools' expectations, often conflict with tradi-

tional . masculine sex-role expectations, resulting in a con-

fusing double message; Be aggressive, active, achieving and

independent (be masculine) , but also be passive, quiet, and

conforming (be a good pupil) . It is no wonder that boys in

vastly disproportionate numbers are the truants, the delin-

quents, the underachievers, the inattentive, and the rebel-

lious. In New York City sixty-three percent of all dropouts

are boys and twice as many boys as girls under fifteen are

currently first admitted to public mental hospitals (Sexton,

1969). Schools must accept some of the responsibility of

these statistics. Phil Jackson (1968, p. 5) in Life in

Classrooms , underscores how great an influence, in terms of

time, the elementary school actually has: "Aside from sleep-

ing, and perhaps playing, there is no other activity that

occupies as much of the child's time as that involved in at-

tending school." (However, research indicates that televi-

sion viewing may occupy more time cumulatively (Lesser,

1974) .)

The impact the schools have on children is obvious

and the responsibility great. However, schools cannot create

a nonsexist world. They can join in the struggle or continue

83

to thwart it. They can either continue to perpetuate the old

roles and relationships, or they can begin to free boys and

girls from the rigid and stunting identities that have been

imposed by our culture. It is hoped that the results of this

study will add to the arguments that will ultimately affect

change in our educational institutions. After all, "breaking

the system of sex stereotyping in early childhood could ac-

complish wonders in eradicating the discriminatory stereotyp-

ing and tracking of children in their later education"

(Frazier and Sadker, 1973, p. 27).

REFERENCES

Anastasiow, N.J. Success in school and boys' sex-role pat-terns. Child Development , 1965, 1053-1066.

Appleton, C.P. Distribution of teacher attention in thepreschool classroom: The effects of child sex andchild behavior. Unpublished doctoral dissertation.University of Massachusetts, 1975.

Bandura, A. Social-learning theory of identif icatory process.• Handbook of Socialization Theory and Research.

Edited by D.A. Goslin. Chicago : Rand McNally and Co . ,

1969.

Bandura, A., and A. Huston. Identification as a process ofincidental learning. Journal of Personality andSocial Psychology , 1961, 311-318.

Bandura, A.; D. Ross; and S.A. Ross. Imitation of film-mediated aggressive models. Journal of AbnormalSocial Psychology , 1963, 3-11.

Bandura, A., and R. Walters. Social Learning and PersonalityDevelopment. New Yorkl Holt , Rinehart and Winston,1963.

Barron, F. Originality related to independence of judgment.Journal of Abnormal Social Psychology, 1955, 51,478-485.

Bern, S.L. Probing the promise of androgyny. In: Beyond Sex -

Role Stereotypes: Readings Toward a Psychology ofAndrogyny . Edited by A.G. Kaplan and J.P. Bean.Boston: Little, Brown and Co., 1976.

Benjamine, H. Age and sex differences in the toy preferencesof young children. Journal of Genetic Psychology ,

1932, 417-429 .

Biller, H.B. Father, Child, and Sex-Roles . Lexington, Mass.:D.C. Heath and Company, 1971.

84

85

Boldwin, A. A cognitive theory of socialization. In: Hand-book of Socialization Theory and Research . Editedby D.A. Goslin. Chicago: Rand McNally and Co., 1969.

Bowers, K.S. Situationalism in psychology: An analysis anda critique. Psychological Review , 1973, 80, 307-336.

Brandt, R.M. The accuracy of self-estimate: A measure off“^o^cept . Genetic Psychology Monographs, 1958,

58 , 55 .

Bridges, K.M.B. Occupational interests of three-year-olds.Ped. Sem. , 1927, 34, 415-423.

Bridges, K.M.B. Occupational interests and attention offour-year-old children. Journal of Genetic Psychology,1929,. 36, 551-570.

Bronfenbrenner , U. Freudian theories of identification andtheir derivatives. Child Development, 1960, 31, 15-40.

—Broverman, D. Sex-role stereotypes: A current appraisal.

As quoted in Undoing Sex Stereotyping Research andResources for Educators . Edited by M. Guttentag andH. Bray. New York: McGraw-Hill Book Co., 1976.

Brown, D.G. Sex-role preference in young children. Psycho-logical Monographs , 1956, 7£, 1-19.

Brown, D.G. Sex-role preference in children: Methodologicalproblems. Psychological Reports , 1962, 477-478.

Brown, D.G. Masculinity-femininity development in children.Journal of Consulting Psychology , 1957, _3, 197-202.

Campbell, D.P. Revised Manual for Strong Vocational InterestBlanks. Stanford: Stanford University Press, 1966.

Cosentino, F. , and A.B. Heilbrun. Anxiety correlates ofsex-role identity in college students. PsychologicalReports , 1964, 1±, 729-730.

Crandall, V. Sex differences in expectancy of intellectualand academic reinforcement. In: Achievement RelatedMotives in Children . Edited by C.P. Smith. New York:Russell Sage Foundation, 1969.

Czaplinski, S.M. Sexism in Award Winning Picture Books .

Pittsburg, Pa.: Know, Inc., 1972.

86

DeLucia, L.A. The toy preference test: A measure of sex-roleidentification. Child Development ^ 1963, 34, 107-117.

Ducker, K. Experimental modification of children's foodpreference through social suggestion. Journal ofAbnormal Psychology , 1938 , 211-216^

English, H.O., and A.B. English. Comprehensive Dictionary ofPsychological and Psychoanalytical Terms . New York:Longman, Green, 1^58

.

Fagot, B.I., and G.R. Patterson. An in vivo analysis of re-inforcing contingencies for sex-role behaviors in thepreschool child. Developmental Psychology, 1969, 1,563-568. ^

Farwell, L. Reactions of kindergarten, first- and second-grade children to constructive play materials.Genetic Psychology Monograph , 1930, £, 431-560.

Fauls, L.B., and W.D. Smith. Sex-role learning of five-year-olds. Journal of Genetic Psychology, 1956, 89, 105-117.

Feminists on Children's Literature. A feminist look at chil-dren's books. School Library Journal, 1971, 17, 19-24.

Flanders, J.P. A review of research on imitative behavior.Psychological Bulletin , 1968, 316-337.

Frazier, N., and M. Sadker. Sexism in School and Society .

New York: Harper and Row, 1973.

Friedrich, L.K. , and A.H. Stein. Aggressive and prosocialtelevision programs and the natural behavior of pre-school children. Monograph of Society for Researchin Child Development , 1973, 38 .

Frueh, T., and P.E. McGhee. Traditional sex role developmentand amount of time spent watching television. Devel -

opmental Psychology , 1975, 109.

Fryrear, J.L., and M.H. Thelen. Effect of sex of model andsex of observer on the imitation of affective behav-

ior. Developmental Psychology , 1969,

Gall, M.D. The relationship between masculinity-femininityand manifest anxiety. Journal of Clinical Psychology ,

1969, 25, 294-295.

87

Gray, S.W. Masculinity-femininity in relation to anxiety andsocial acceptance. Child Development

^ 1957, 28, 203-

Haraburg, D.A., and D.T. Lunde . Sex hormones in the develop-ment of sex differences in human behavior. In: TheDevelopment of Sex Differences . Edited by E.E. Mac-coby. Stanford; Stanford University Press, 1966.

Hampson, J.L. Determinants of psychosexual orientation.In: Sex and Behavior . Edited by F.A. Beach. New York:JohnWiley and Sons, 1965.

Hartford, T.C.; C.H. Willis; and H.L. Beabler. Personalitycorrelates of masculinity-femininity. PsychologicalReports , 1967, 881-884 .

Hartley, R.E .' Sex-role pressures and the socialization of themale child. Psychological Reports , 1959, 456-468.

Hartup, W.; S. Moore; and G. Sager. Avoidance of inappropriatesex-typing by young children. Journal of ConsultingPsychology , 1963, 467-473.

Hartup, W., and E.A. Zook. Sex-role preferences in three-and four-year-old children. Journal of ConsultingPsychology , 1960, 420-426.

Hetherington, E., and G. Frankie. Effects of parental domi-nance, warmth, and conflict on imitation in chil-dren. Journal of Personality and Social Psychology,1967, 6, 119-125.

Heyn , L. Children's books. Women: A Journal of Liberation,1969, 1.

Howe, F., and P. Lauter. The Conspiracy of the Young . NewYork: World, 1970, p. 306.

Hutt ,C. Males and Females . Harmondsworth ,

Middlesex,England: Penguin Books Ltd., 1972.

Jackson, P. Life in the Classrooms . New York: Holt, Rinehartand Winston , 17568.

Jennings, S. Effects of sex typing in children's stories on

preference and recall. Child Development , 1975, 46 ,

220-223.

88

Joffe, C. As the twig is bent. In: And Jill Came TumblingAfter: Sexism in American Education. Edited by

^

J. Stacey, S. Bereaud, and J. Daniels. New York:Dell Publishing Company, 1974.

Kagan, J. The concept of identification. Psychological Re-.view , 1958, 296-305.

Kagan, J. Acquisition and significance of sex-typing andsex-role identity. In: Review of Child DevelopmentResearch , Vol. 1, pp. 137-167. Edited by M.L. Hoffmanand L .W . Hoffman. New York: Russell Sage Foundation,1964.

Kagan, J. On the meaning of behavior: Illustrations from theinfant. Child Development

, 1969, 4^, 1121-1134.

Key, M.R. The role of the female in children’s books--dis-pelling all doubts. Wilson Library Bulletin, 1971,

Kobasigawa, A. Inhibitory and disinhibitory effects of modelson sex-inappropriate behavior in children. Psycho-

logia , 1968, 86-96.

Kohlberg, L. Cognitive developmental analysis of children'ssex-role concepts and attitudes. In: The Developmentof Sex Differences . Edited by E.E. Maccoby. Stanford:Stanford University Press, 1966.

Kohlberg, L. Stage and sequence: The cognitive developmentalapproach to socialization. In: Handbook of Socializa -

tion Theory and Research. Edited by D.A. Goslin.Chicago : Rand McNally and Co., 1969.

Lesser, G.S. Children and Television. New York: Random House,

1974 .

Lewis, M. Parents and children: Sex-role development. SchoolReview , 1972, 229-240.

Liebert, R.M.; J.M. Neal; and E.S. Davidson. The Early Win-

dow: Effects of Television on Children and Youth .

New York: Pergamon Press, Inc., 1973.

Lynn, D.B. A note on sex differences in the development of

masculine and feminine identification. Psychologica_l

Review , 1959, 6^, 126-135.

Maccoby, E.E., ed. The Development of Sex Differences. Stan-

ford: Stanford University Press ,1966.

89

Maccoby, E.E., and C.N. Jacklin. The Psychology of Sex Dif-f££ence^. Stanford: Stanford University Press, 1974.

Maccoby, E.E. , and W.C. Wilson. Identification and observa-tional learning from films. Journal of AbnormalSocial Psychology , 1957, ^,”76-87.

Maccoby, E.E.; W.C. Wilson; and R.V. Burton. Differentialmovie-viewing behavior of male and female viewers.Journal of Personality

, 1958, 26, 259-267.

MacKinnon, D.W. Fostering creativity in students of engineer-ing. Journal of Engineering Education, 1961. 52. 129-142.

~ — —Masters, J.C., and S.A. Driscoll. Children's imitation as a

function of presence or absence of a model and thedescription of his instrumental behavior. ChildDevelopment , 1971, £2, 161-170.

Mead, M. Sex and Temperament in Three Primitive Societies .

New York : Morrow , 1935

.

Meyer, W.J., and G.G. Thompson. Teacher interaction with boysas contrasted with girls. In: Psychological Studiesof Human Development . Edited by R.G. Kuhlens andG.G. Thompson. New York: Appleton-Century-Crofts

,

1963 .

Michigan Women's Commission. Sex Discrimination in an Ele -

mentary Reading Program , 1973

.

Miles, B. Harmful lessons little girls learn in school.Redbook , 1971, 168-169.

Miles, B. Channeling Children: Sex Stereotyping in Prime -

Time T.V. Princeton, N.J.: Women on Words andImages, 1975.

Mischel , W. A social-learning view of sex differences inbehavior. In: The Development of Sex Differences .

Edited by E.E. Maccoby. Stanford : Stanford UniversityPress, 1966.

Mischel, W. Sex-typing and socialization. In: Carmichael '

s

Manual of Child Psychology , Vol. II. Edited by P.H.

Mussen. New York: John Wiley and Sons, Inc., 1970.

Mussen, P.H. Long-term consequents of masculinity of inter-

ests in adolescence. Journal of Consulting Psychol -

ogy, 1962, 435-440.

90

Mussen, P.H., and L. Distler. Child rearing antecedents ofmasculine identification in kindergarten boys.Child Development , 1960, 11 , 89-100.

NLlsen, A.P. Women in children's literature. College English1971, 8.

^

^

Piaget, J. Judgment and Reasoning in the Child. Patterson.N . JTI Littlefield, Adams , 1964

.

Pitcher, E.G. Male and female. In: And Jill Came TumblingAfter: Sexism in American Education . Edited by J.Stacey, S. Bereaud, and J. Daniels . New York:Dell Publishing Co., 1974.

Rabban, M. Sex-role identification in young children in twodiverse social groups. Genetic Psychology Monographs,1950, 81-158.

Rheingold, H.L., and K.V. Cook. The contents of boys' andgirls' rooms as an index of parents' behavior.Child Development , 1975, 456-463.

Rosekrans, M.A. Imitation in children as a function of per-ceived similarity to a social model and vicariousreinforcement. Journal of Personality and SocialPsychology , 1967, 1_, 307-315.

Rosenberg, B.G., and B. Sutton-Smith . The measurement ofmasculinity and femininity in children. ChildDevelopment , 1959, 22.' 373-380 .

Rosenberg, B.G., and B. Sutton-Smith. Sex and Identity .

New York: Holt, Rinehart and Winston, Inc., 1972.

Rosenthal, R. Pigmalion in the Classroom: Teacher Expecta -

tion and Pupils* Intellectual Development . New York

:

Holt, Rinehart and Winston, 1968.

Rosenthal and Fode, 1960, and Rosenthal and Lawson, 1961.Information from N. Weissten. Kinder, kuche, kircheas scientific law: Psychology constructs the female.The New Woman; An Anthology of Women's Liberation.New York: Fawcett World Library, 1971

.

Rubin, J.Z.; F.J. Provenzano; and Z. Luria. The eye of the

beholder: Parents' view of sex of newborns. AmericanJournal of Orthopsychiatry , 1974 , 44^^ 512-519"^

Rudman, M. Children's Literature; An Issues Approach . Lexing-

ton. Mass.: D.C. Heath and Company, 1976.

91

P ‘S . The Effect of Classroom Conditions on theStrength of Achievement Motive and Work Output ofElementary School Children^ Cooperative ResearchProject No. OE-873. Washington, D.C.: U.S. Dept,of Health, Education, and Welfare, 1963.

R.R. Relation of early socialization experiences toself-concepts and gender role in middle childhood.Child Development , 1970, 4]^, 267-289 .

R.R.; L. Rau; and R.A. Alpert. Identification andChild Rearing. Stanford: Stan fo rd University Press,1965.

Sexton, P.C. The Feminized Male Classrooms, White Collarsand the Decline of Manliness . New York; VintageBooks , 1969

.

Spaulding, R.L. Achievement, Creativity, and Self-ConceptCorrelates of Teacher-Pupil Transactions in Elemen-tary Schools . Cooperative Research Project No. 1352.Washington, D.C.: U.S. Dept, of Health, Education,and Welfare, 1963.

Stoller, R.J. Sex and Gender . New York: Science House, 1968.

Strong, E.K. Interests of men and women. Journal of SocialPsychology , 1936 , 1_, 49-67.

Terman, L.M., and C.C. Miles. Sex and Personality . New York:McGraw-Hill, 1936.

Thorndike, R.I. Review of pygmalion in the classroom.American Educational Research Journal, 1968, 5, 708-

711.

Tiger, L. Men in Groups . New York: Random House, 1969.

Trumpter, M. , and L.D. Crowe. Sexism in picture books.Illinois Libraries , 1971, 5.

U ' Ren , M. The image of women in textbooks. In; Women in

Sexist Society. Edited by V. Cornick and B. Moran.

New York : Basic Books, 1971.

Vance, T.F., and L.T. McCall. Children's preferences among

play materials as determined by the method of paired

comparisons of pictures. Child Development , 1934,

267-277.

Sears

,

Sears

,

Sears

,

92

Webb, A.P. Sex-role preference and adjustment in earlyadolescents. Child Development , 1963, 34, 609-618.

Weitzman, L.J.; D. Eiffer; E. Hokada; and C. Ross. Sex-rolesocialization in picture books for preschool chil-dren. American Journal of Sociology , 1972, 77.

A.G. , ed.^®Port on Sex Bias in the Public Schools.

New YorkT NOW, 1971.^

“

White, L., Jr. Educating our Daughters. New York: Haroer.19 50.

Whiting, B., and C.P. Edwards. A cross-cultural analysis ofsex differences in the behavior of children agedthree through 11. Journal of Social Psychology, 1973,9]^, 171-188.

Wolf, T.M. Response consequences to televised modeled sex-inappropriate play behavior. Journal of GeneticPsychology

,

1975, 127, 35-44.

Wolf, T.M. Effect of live modeled sex-inappropriate play be-havior in a naturalistic setting. DevelopmentalPsychology , 1973, 9_, 120-123.

on Words and Images. Dick and Jane as Victims: SexStereotyping in Children's Readers. Princeton, N.J.,1972.

Women

APPENDIX A

PICTURE BOOKS

93

94

95

i

98

99

OIMz|>->o

e- E-9DO W

t/3

EhOQ CuOtnM >HX X

"=£Q MU] 2UZ >H< <o a«

z og3 Eh

tii &Hz

ei

ANDREWS

EYES

RUSHED

TO

SEE

HIS

DUMP

TRUCK

THAT

WAS

SITTING

ON

A

SPECIAL

PLACE

ON

HIS

SHELF.

ANDREW

WANTED

TO

PLAY

WITH

HIS

DUMP

TRUCK.

102

103

OR

SOMETIMES,

ANDREW

TURNED

THE

DUMP

TRUCK

UPSIDE-DOWN

AND

TURNED

THE

WHEELS

SO

THEY

WOULD

SPIN.

HE

WANTED

TO

SEE

HOW

PAST

HE

COULD

MAKE

THEM

GO.

IP

HE

LISTENED

CLOSELY,

THE

WHEELS

MADE

A

SOPT

SOUND

WHEN

THEY

WERE

SPINNING

SLOWLY

AND

A

LOUD

NOISE

WHEN

THEY

WERE

GOING

PAST.

105

106

107

110

113

ANDREW

WOULD

TURN

THE

HANDS

AND

LEGS

OF

THE

DOLL

SO

THAT

THEY

POINTED

IN

DIFFERENT

DIRECTIONS,

ONE

ARM

WOULD

BE

UP

AND

ONE

POINTED

OUT

STRAIGHT.

HE

EVEN

TRIED

TO

MAKE

THE

DOLL'S

LEG

BEND

SIDEWAYS.

114

Sometimes

Andrew

just

held

the

doll

in

his

arms,

that

reminded

Andrew

of

the

way

HIS

dad

held

him.

115

116

117

118

119

122

125

KATY

WOULD

TURN

THE

HANDS

AND

LEGS

OF

THE

DOLL

SO

THAT

THEY

POINTED

IN

DIFFERENT

DIRECTIONS.

ONE

ARM

WOULD

BE

UP

AND

ONE

POINTED

OUT

STRAIGHT.

SHE

EVEN

TRIED

TO

MAKE

THE

DOLL’S

LEG

BEND

SIDEWAYS.

127

MOTHER

HELD

HER

128

129

SUDDENLY.

KATY

HEARD

DIFFERENT

NOISES.

THEY

WERE

BATHROOM

NOISES

AND

KITCHEN

'NOISES.

KATY

KNEW

THAT

HER

SECRET

TIME

WAS

OVER

AND

THAT

IT

WAS

TIKE

TO

START

SETTING

DRESSED.

130

1

131

KATY

SAW

HER

BALL.

NO

SHE

DIDN'T

WANT

TO

PLAY

WITH

HER

BALL

134

asM 6-*

OftT

EhEh:d(XX Cus

§"=

egas£-•>-«

"=2QOiU4oo

2 e-*02es

w-

aatn a

IkATY'S

eyes

rushed

to

see

her

dump

truck

that

was

sitting

on

a

special

place

on

her

shelf,

katy

wanted

to

play

with

her

dump

truck.

I

I

I

138

139

140

141

APPENDIX B

TABLES OF RESULTS

142

-TEST

ANALYSIS

OF

TOY

MANIPULATION

AND

VISUAL

REGARD

ON

PRE-TEST

SCORES

INDICATING

GROUP

EQUIVALENCE

143

Eh

144

TABLE 13

ANALYSIS OF VARIANCE OF MEAN POST TESTSCORES FOR TOY MANIPULATION

SOURCE df SUM SO. MEAN SQ. F P

N E U T R A L TOY S

Sex 1 2346.125 2346.125 1.72608 .19887Book 1 21.125 21.125 .01535 . 90230Sex By Book 1 2211.125 2211.125 1.60623 .21546

Error 28 40776.750 1359.225

FEMALE TOYS

Sex 1 2628.125 2628.125 2.26884 .14246

Book 1 300.125 300.125 .29309 .59253

Sex By Book 1 5778.125 5778.125 5.64260 .02461

Error 28 34750.750 1158.358

MALE TOYS

Sex 1 457.531 457.531 .24944 .62111

Book 1 385.031 385.031 .20420 .65483

Sex By Book 1 1845.281 1845.281’ .97863 .33101

Error 28 55026.437 1834.214

145

TABLE 14

ANALYSIS OF VARIANCE OF MEAN POST TESTSCORES FOR TOY VISUAL REGARD

SOURCE df SUM SQ. MEAN SQ. F P

N E U T R A L TOY S

Sex 1 385.031 385.031 .51945 .47705Book 1 140.281 140 . 281 .18926 .66687Sex By Book 1 2295.031 2295.031 3.09626 . 08940Error 28 20754 . 375 741.227

FEMALE TOYS

Sex 1 770.281 770.281 1.00186 . 32543

Book 1 639.031 639.031 .83115 . 36972

Sex By Book 1 4394.531 4394.531 5.71570 .02378

Error 28 21527.875 768.852

MALE TOYS

Sex 1 225.781 225.781 .15113 .70040

Book 1 770.281 770.281 .51558 .47868

Sex By Book 1 1069.531 1069.531 .71589 .40467

Error 28 41831.875 1493.995

146

TABLE 15

ANALYSIS OF VARIANCE OF MEAN POST TESTDIFFERENCE SCORES FOR TOY MANIPULATION

SOURCE df SUM SQ. MEAN SQ. F P

N E U T R A L TOY S

Sex 1 1953.125 1953.125 1.38168 .24971

Book 1 1431.125 1431.125 1.01241 . 32294

Sex By Book 1 2211.125 2211.125 1.56490 .22140

Error 28 39580.000 1413.580

FEMALE TOYS•%

Sex 1 8 . 000 8.000 .00583 .93968

Book 1 253.125 253.125 .18448 .67084

Sex By Book 1 14365.125 14365.125 10.46919 .00311

Error 28 38419.750 1372.133

MALE TOYS

Sex 1 850.781 850.781 . 36214 .55216

Book 1 935.281 935.281 . 39811 .53318

Sex By Book ] 15978.781 15975.781 6.80023 .01445

Error 28 65780.375 2349.299

147

TABLE 16

ANALYSIS OF VARIANCE OF MEAN POST TESTDIFFERENCE SCORES FOR TOY VISUAL REGARD

SOURCE df SUM SQ. MEAN SQ. F P

N E U T R A ]L TOY!

Sex 1 116.281 116.281 .11349 .73872

Book 1 957.031 957.031 .93403 . 34209

Sex By Book 1 2032.031 2032.031 1.98319 .17006

Error 28 28689.625 1024.629

FEMALE TOYS

Sex 1 264.500 264.500 .25957 .61441

Book 1 544.500 544.500 .53435 .47086

Sex By Book 1 8320.500 8320.500 8.16536 .00797

Error 28 28532.000 1019.000

MALE TOYS

Sex 1 225.781 225.781 .14329 .70789

Book 1 1696.531 1696.531 1.07669 . 30832

Sex By Book 1 7843.781 7843.781 4.97799 .03386

Error 28 44119.75 1575.691