THE RELATIONSHIP BETWEEN PERSEVERANCE IN SCHOOL
AND CHILDHOOD LOSS OF PARENTS IN ONE
MIDWESTERN SCHOOL DISTRICT '

Dissertation for the Degree of Ph. D.
MICHIGAN STATE UNIVERSITY
DONALD EUGENE SPOHN
1978

LIBRAR Y
18335:“;

U

ABSTRACT

THE RELATIONSHIP BETWEEN PERSEVERANCE IN SCHOOL
AND CHILDHOOD LOSS OF PARENTS IN ONE
MIDWESTERN SCHOOL DISTRICT

By
Donald Eugene Spohn

The Problem

 

1

Each year increasing numbers of children suffer the loss of
parents. The primary purpose of this study was to determine the
relationship, if any, between the childhood loss of parents and
perseverance in school. The secondary purpose of this study was to
determine if relationships exist between perseverance in school and
the following variables: age at loss; duration of loss; number of
losses; sex of child; sex of parent; socioeconomic background of
family; ability as perceived by mother; cause of loss; substitute

parents; and substitute homes.

Methods and Procedures

 

Data were obtained in personal interviews with mothers of
T42 subjects, l6 through 2l years of age, approximately 6 percent of
the total number of possible subjects. The sample consisted of 40
dropouts and l02 perseverers. The sampling procedure included
random selection of streets in the school district with all homes on
the selected streets visited and revisited until every subject was

identified. No subjects were lost.

Donald Eugene Spohn

Analysis was conducted through the use of chi-square test of
independence and the t-test for proportion. Significance was set at

the .05 level of confidence.

Findings

Primary Hypothesis

The relative intactness of the family and the subjects' per-
severance in school were found statistically dependent. Dropouts were
more likely to have lost parents than to have come from intact homes,
and subjects suffering losses of parents were even more likely to drop

out than to persevere.

Secondary Hypotheses

 

l. The age of subjects at the time of loss was found to be
statistically independent of perseverance in school.

2. The duration of loss of parents was found to be statistically
dependent of perseverance in school. Losses of parents
spanning one or more years are more closely related to per-
severance in school than are losses recorded as less than one
year in duration.

3. The number of losses of parents experienced by subjects and
perseverance in school were found to be statistically depen-
dent. Subjects suffering more than one loss were less likely
to persevere than were subjects suffering only one loss and

subjects suffering no loss were most likely to persevere.

l0.

Donald Eugene Spohn

The proportion of subjects suffering no loss of parents and
persevering in school exceeded the proportion of subjects
suffering losses of parents and persevering in school.

The sex of subjects suffering losses of parents was found to be
statistically independent of perseverance in school.

The sex of lost parents was found to be statistically indepen-
dent of perseverance in school.

The socioeconomic background of subjects' families, as measured
by family income, and perseverance in school were found to be
statistically dependent. Subjects from lower income families
were less likely to persevere in school than were subjects

from families with average or above average incomes.

Mother's perception of subject's ability in school was found to
be statistically dependent of perseverance in school. The

more ability mothers perceived in subjects the more likely the
subject was to persevere in school.

The cause of subjects' loss of parents and perseverance in
school were found to be statistically dependent. While per-
severing subjects were seldom observed to have lost parents,
those who did were more likely to experience that loss because
of working mothers than because of separations.

The proportion of perseverance in school for subjects suffering
no loss of parents exceeds the proportion of perseverance for
subjects suffering a loss caused by divorce or desertion.

While most subjects from intact families persevered in school,
only a small number of those experiencing divorce or separation

did so.

Donald Eugene Spohn

ll. The proportion of perseverance in school for subjects suffering
no loss of parents exceeds the proportion of perseverance for

subjects experiencing a working mother.

THE RELATIONSHIP BETWEEN PERSEVERANCE IN SCHOOL
AND CHILDHOOD LOSS OF PARENTS IN ONE
MIDNESTERN SCHOOL DISTRICT

By

Donald Eugene Spohn

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Administration and
Higher Education

l978

Capyright by
DONALD EUGENE SPOHN
l978

This dissertation is dedicated to my family who in tolerance
of my many demands supported me in love, patience, and strength,
assuring my success. I am deeply indebted to my family:

My wife, Rita;
My daughter, Heidi; and

My son, Daniel.

ii

ACKNOWLEDGMENTS

My sincere thanks are due to the members of my Committee,
Professors Louis G. Romano, Norman T. Bell, and Daniel H. Kruger,
for their encouragement and help during the preparation of this
dissertation.

Invaluable assistance was provided and helpful suggestions
were made by Professor Archibald B. Shaw, Chairman of my Committee.
Illness prevented him from continuing for a period of time and Dr.
Louis Romano efficiently guided my progress until Dr. Shaw resumed his
duties as Committee Chairman.

Finally, deep appreciation is extended to the many parents
who participated in this study, who welcomed this investigator into

their homes, giving of their time and sharing their experiences.

TABLE OF CONTENTS

Page

LIST OF TABLES . . . . . . . . . . . . . . . . . vii
Chapter

I. IDENTIFICATION OF THE PROBLEM. I

The Need for This Study . 4

Definitions . . 5

Assumptions . . 7

Limitations of the Study. 7

Research Questions. 8

Primary Research Question. 8

Secondary Research Questions. 8

Procedure for Data Analysis. 9

Overview . 9

II. REVIEW OF THE RELATED LITERATURE. . . . . . . . . ll
PART I . . . . . . . . . . . . . . . . l2

Loss of Primary/Secondary Object, General . . . . l2
Primary/Secondary Attachment Object Loss by Death . . l9
Primary Attachment Object Loss: Working Mother . . . 22
Loss of Secondary Attachment Object. . . . . . 24
Separation Caused by Parental Discord . . . . . . 26
Part I Summary . . . . . . . . . . . . . . 30

PART II . . . . . . . . . . . . . . . . 3l

A Complex Behavior Pattern: School Withdrawal. . . . 3l
I. Q. and Grade Point Average. . . . . . . . . 32
Ability and Achievement . . . . 33
Irregular Attendance and Excessive Transfers . . . 34
Discipline, Violence, Crime, and Drugs . . . . . 34
Social Economic Status of Family . . . . . . . 35
Broken Homes and Low Self-Esteem . . . . . . . 38

Part II Summary. . . . . . . . . . . . . . 40

iv

Chapter Page

III. DESIGN OF THE STUDY. . . . . . . . . . . . . 41
Population . . . . . . . . . . . . . . . 41
Sample Selection . . . . . . . . . . . . . 43
Procedure . . . . . . . . . . . . . . . 48
Question Sequence. . . . . . . . . . . . . 48
Question Analysis. . . . . . . . . . . . . 50
Interview Procedure . . . . . . . . . . . . 52
Testable Hypotheses . . . . . . . . . . . . 54

Primary Hypothesis. . . . . . . . . . . . 54
Secondary Hypotheses . . . . . . . . . . . 54
Analysis. . . . . . . . . . . . . . . . 55
Summary . . . . . . . . . . . . . . . . 56
IV. ANALYSIS OF DATA. . . . . . . . . . . . . . 57
Introduction . . . . . . . . . . . . . . 57
Procedure . . . . . . . . . . . . . . . 57
Test of Primary Hypothesis. . . . . . . . . . 59
Test of Secondary Hypotheses . . . . . . . . . 60
Secondary Hypothesis 1 60
Secondary Hypothesis 2 51
Secondary Hypothesis 3 52
Secondary Hypothesis 4 63
Secondary Hypothesis 5 55
Secondary Hypothesis 6 55
Secondary Hypothesis 7 55
Secondary Hypothesis 8 57
Secondary Hypothesis 9 58
Secondary Hypothesis lO 59
Secondary Hypothesis ll 7I
V. SUMMARY AND CONCLUSIONS . . . . . . . . . . . 75
Smmmy. . . . . . . . . . . . . . . . 75
Primary Hypothesis. . . . . . . . . . . . 76
Secondary Hypotheses . . . . . . . . . . . 75
Discussion . . . . . . . . . . . . . . . 78
Primary Hypothesis. . . . . . . . . . . . 78
Secondary Hypotheses . . . . . . . . . . . 30
Summary of Discussion . . . . . . . . . . . 84

Chapter

Implications.
Recommendations.

REFERENCES

vi

Page

86
87

89

Table

000090
#WN

LIST OF TABLES

Population Distribution for Secondary Students.
Racial Composition of Secondary Student Population
Secondary Students--Population (February 23, l978)

Secondary Students--Persevering Population
(February l978) .

Secondary Students--Dropout Population (February
l978 . . . . . . . . . . . ,. . .

A Comparison of Population and Sample Distribution
Based on Location. .

A Comparison of P0pulation and Sample Distribution
Based on Race . . .

A Comparison of Population and Sample Distribution
Based on Sex ' . .

The Relationship Between the Hypotheses and the
Individual Questions Designed to Collect Data to
Test the Hypotheses . . . . . .

Test of Primary Hypothesis
A Comparison of Age at the Time of Loss (Under Six

Years of Age/Six Years of Age and Older) with
Perseverance in School (Drap/Persevere)

A Comparison of Duration of Loss (One Year or Less/
Longer than One Year) with Perseverance in School

(Drop/Persevere) .

A Comparison of Loss of Parent (No Loss/One Loss/ Two
or More Losses) with Perseverance in School (Drop/

Persevere) .

vii

Page
42
43
44

44

44

47

47

47

53
59

GT

62

63

Table
4.5

The Proportion of Perseverance in School (Drop/Persevere)
by Subjects Who Have Lost Parents as Compared to the
Proportion of Perseverance in School by Subjects from
Intact Families. . . . . . .

A Comparison of Sex of Subjects with Perseverance in
School (Drop/Persevere) .

A Comparison of Sex of Lost Parent with Perseverance
in School (Drop/Persevere) .

A Comparison of Socioeconomic Background (Below
Average/Average, and Above Average) of Subjects
with Perseverance in School (Drop/Persevere) .

A Comparison of Ability (Below Average/Average/Above
Average) with Perseverance (Persevere/Drop)

A Comparison of Cause of Loss (Separation/Working
Mother/Other) with Perseverance in School (Drop/
Persevere) . . . . . . .

The Proportion of Perseverance in School (Drop/
Persevere) by Subjects Who Have Lost Parents Caused
by Separation as Compared to the Proportion of
Perseverance in School by Subjects of Intact
Families . . . . .

The Proportion of Perseverance in School (Drop/
Persevere) by Subjects Who Have Lost Parents Caused
by Working Mothers as Compared to the Proportion
of Perseverance in School by Subjects of Intact
Families . . . . . . . . . . .

A Table of the Research Findings.

A Summary of the Research Findings .

viii

Page

64

65

66

67

68

69

7O

72
74
79

CHAPTER I

IDENTIFICATION OF THE PROBLEM

The purpose of this study is to measure the relationship
between childhood loss of primary/secondary attachment objects and
later failure to persevere in school. The family is one of the oldest
human institutions. In its simplest form it is composed of mother,
father, and child. Margaret Mead (l97l), who has long studied
families in various cultural and geographic settings, says of the.
mother-infant relationship, "During the long months of infancy, the
child's tie to the mother is a bodily one, the child learns through
its own body what the world is like and what is like to be a person."
In terms of the father-child relationship, Mead (l97l) states,

". . without a recognized father, whether he is living or dead,
and whether he is the child's biological father or another man who
stands in his place, the child has no social existence."

Man has yet to plumb the depth and measure the breadth of the
parent-child relationship. It is not known under what conditions and
to what extent children can endure the loss of parents. Few pretend
to know the effect of parental loss on children at any age. Research
has yet to measure the effect of parental loss on social adjustment,

school achievement, or success in later life and the consequence of

separation, divorce, parental death, working mothers, absent fathers,

and birth out of wedlock on children. Of one thing we are certain, the
number of children affected is growing year by year. Each year, our
children are spending more of their time away from their parents. Of
the primary causes of loss of parents listed above, divorce is one of
the most common.

In 1900, only 8 percent of American marriages ended in divorce
(Peterson, l977). Twenty (20) years later, it was l3 percent and by
l950, it had grown to one out of every four, 26 percent. In 1970, it
increased to one out of every three (U.S. Department of Health,
Education, and Welfare, Vital Statistics of the U.S., l972). In l976,
nearly one out of every two marriages ended in divorce, or 49 percent
(U.S. Department of Health, Education, and Welfare, Vital Statistics
Report, l976). In the fifteen year period, l96l-75, inclusive, a
total of ll,460,000 school children were affected by divorce.

Another large group of children are affected by working
mothers. Not only do divorced mothers work away from home, but a
greater proportion of them with children under the age of eighteen do
so (85%), than do women without children (U.S. Department of Labor,
Bureau of Labor Statistics, l977). In 1940, only one out of ten
mothers worked. Three out of ten worked in 1960 and four out of ten
in 197T (U.S. Department of Labor Handbook, 1975). By January l976,
roughly 60 percent of all women between the ages of seventeen and
twenty-four were working (U.S. Department of Labor, 1977).

Then there are the fatherless homes. As might be expected,
these, too, have increased from roughly lO percent of the households
in T950 to nearly 24 percent in 1975 (U.S. Department of Commerce,

Bureau of the Census, l976). Approximately half a million illegitimate

births were recorded in T975, an increase of 7 percent over T974. Over
l4 percent of the live births in T975 were illegitimate (U.S. Depart-
ment of Health, Education, and Welfare, Vital Statistics, T975).

Based on these two figures, divorces and live births, we know that
close to 39 percent of our children lived in fatherless homes in T977.

Death, too, takes its toll on the child-parent relationship.
Brown (l96l) estimated T2 percent of our children suffer the loss of a
parent by the time they reach the age of sixteen. Munro (l965) found
that TO percent of children in the general population have lost one or
both parents by age sixteen.

The family is often considered society's giant shock absorber.
Toffler (l97l) believes the family is receiving and will continue to
receive shocks which will tear it to pieces. Peterson (T977) feels
the nuclear family is under pressure and it may be unable to survive.
Mead (l97l) tells us, "The closeknit relationship between parent and
child has been altered. . . ." Mead (T956) poses the question, "How
soon will all humans break down under the strain of modern life?"
Skinner (T948) says educators themselves cannot save the situation as
the causes are too deep and involve the whole structure of society.
The family is sometimes likened to the soil which stabilizes our
emotional roots. Before he died, Arnold Toynbee (T964) complained,

"A human being has roots, like a tree and they tether him. . . ." He
warned, too, we are "plowing the sands" and unless we retreat to some
of our grandparents' practices in raising our children, our institu-

tions of learning will become little more than social clubs for sexual

mating.

Over the past twenty years, most recognized thinkers claim to
have seen the writing on the wall, expressing concern for the nurturing
practices of society. Sociologically, these undesirable characteris-
tics have been associated with poverty. Costin (T972) expressed the
feeling, children are not safe and happy if parents cannot protect them
from poverty. The growing belief that poverty is the root of childhood
problems has been taught by sociologists and acted on by our govern-
ment. Unfortunately, however, over the past several years broken
homes have increased about as fast as the increasing billions spent on
welfare. We observe, too, that the educationally handicapped, dropouts,
lower national test scores, drug abuse, childhood crimes, venereal
disease, and childhood suicide have all increased over the years even
as we have increased our billions in Federal aid to education.

What is the cumulative effect of the phenomena reflected in
these statistics on our children? Does the increasing loss of parents
affect children's performance in school? Has the effect been measured?
Can it be measured? If one believes in prophecy, he might believe the
problem and its solution were predicted thousands of years ago. "And
he will turn the hearts of the fathers to their children and the
hearts of the children to their fathers, lest I come and smite them
with a curse" (Malachi 4:5). It is the purpose of this study to add
some knowledge to that known about the relationship which may exist

between perseverance in school and childhood loss of parents.

The Need for This Study

 

Data were collected in this study to provide insight into the

nature of childhood loss of primary/secondary attachment objects,

hereafter referred to as P/SAOL as it is related to perseverance in
school. The design provided for noting coincidence of certain condi-
tions and certain apparent consequences, providing certain valuable
clues as to the cause-effect relationship.

This study is important for the following reasons. First,
childhood P/SAOL is a growing phenomenon and may be associated with
another phenomenon, failure to persevere in school. Both phenomena
are of great concern to parents and educators. A second is, although
studies have related childhood P/SAOL to later adult abnormalities,
few studies pretend to study its effect on perseverance in school.
Many investigators have examined variables related to dropouts. Few
studies claim to relate dropping out of school to P/SAOL. A third
reason is parents have little knowledge concerning the effect of their
absence from their child on their child's ability to persevere in
school. It is important that parents and educators better understand
this parent-child relationship to improve their ability to make
decisions in this critical area. Finally, it is hoped that this study
will provide a foundation upon which further studies in this area may

be conducted.

Definitions
Definitions for key terms used in this study will provide a
common basis for understanding.
Dropout-~This term refers to the student who withdraws from
school, failing to persevere through graduation. The terms withdrawing

and dropping out will be used interchangeably.

Intact Family--In this study, the term refers to families in

 

which subjects have not experienced the loss of a parent.

Loss of Primary or Secondary_Attachment Object (LP/SAQ)7-This
term refers to the loss of the primary or the secondary attachment
object, for any cause, of more than fourteen days for children 0-5
years of age or for more than thirty days for children 6-l5 years of
age. This term also includes the working mother.

Primary Attachment Object (PAO)--This term refers to the adult
to whom the child has formed a primary attachment during the first six
months of its life. In this study, the term always refers to the
mother.

Parent;-This term refers to those adults, normally the father
and the mother, to whom the child forms primary and secondary attach-
ments during the first six months of its life.

Perseverance--This term refers to the act of prevailing or
remaining in school regardless of ability, accomplishments, etc.

Primary or Secondary Attachment Object Loss (P/SAOL)--This
term refers to the loss of the mother to whom the child has formed a
primary attachment or to the father to whom the child has formed a
secondary attachment.

Real Parent--This term refers to the biological or natural

 

parent as differentiated from a temporary, substitute, or foster
parent. In this study parents who adopt children prior to attachment
formation are defined as real parents.

Separation--This term refers to any loss of a parent due to

 

marital discord and includes divorce, desertion, temporary separation,

etc.

Scholastic Ability--This term refers to the mother's perception

 

of the subject's scholastic ability.

Substitute Parent--This term refers to those adults who take

 

the place of'the lost primary and secondary attachment objects.

Success in School-~This term refers to the act of persevering

 

in school.

Withdrawal--This term refers to the failure to persevere in

 

school, the act of dropping out of school.

Working Mother--In this research, the term will be defined as

 

working outside of the home full time, continuously for thirty school

days or more.

Assumptions

 

It is assumed in this study that a relationship exists between
the loss of parents by children and later ability to persevere in
school. It is further assumed that the following variables may relate
to perseverance: the age of the child at the time of the loss: the
cause of the loss; the sex of the child as well as the sex of the
parent: the duration of the loss: the number of losses; the socio-
economic background of the family; the child's ability in school; and

the nature of the substitute parent and substitute surroundings.

Limitations of the Study

 

This study was limited by one particular type of community,
one school district. The size of the sample, T42, limited the number
of observations. The nature of categorical data limited the useful-
ness of quantitative predictions. The study was also limited by the

ability of subject mothers to recall accurately. Due to the

limitations of time and resources, the study was further limited to one

variable, childhood loss of parents and one hypothesized related com-

plex behavior pattern, perseverance in school.

Research Questions

The research questions addressed in this study are as follows:

Primary Research Question

 

I.

Are relative intactness of the family and subjects' persever-

ance in school statistically independent?

Secondary Research Questions

 

l.

Is the age of subjects at the time of loss statistically
independent of perseverance in school?

Is the duration of loss of parents statistically independent
of perseverance in school?

Is the number of losses of parents experienced by subjects
statistically independent of perseverance in school?

Is the proportion of subjects suffering no loss of parents and
persevering in school equal to the proportion of subjects
suffering loss of parents and persevering in school?

Is the sex of subjects statistically independent of persever-
ance in school?

Is the sex of lost parents statistically independent of
perseverance in school?

Is the socioeconomic background of subjects' families

statistically independent of perseverance in school?

9

8. Is the subject's ability in school statistically independent
of perseverance in school?

9. Is the cause of subjects' loss statistically independent of
perseverance in school?

TO. Is the proportion of perseverance in school for subjects
experiencing loss of parents caused by separation equal to the
proportion of perseverance in school for subjects experiencing
no loss of parents?

Tl. Is the proportion of perseverance in school for subjects
experiencing loss of parents caused by working mothers equal
to the proportion of perseverance in school for subjects
experiencing no loss of parents?

12. Are substitute parents and substitute surroundings statis-

tically independent of perseverance?

Procedure for Data Analysis
The Michigan State University standard statistical program for
the construction of descriptive statistics was used in this study. As
the data were almost exclusively frequency counts, contingency tables
of analysis were used to see if expected relationships existed among
the variables. The t-test for proportions was used with research

questions involving proportions.

Overview
A review of the literature is presented in Chapter II.
Emphasis is placed in two areas, the area of the variable and the area

of the complex behavior problem.

10

Chapter III describes the population, the sample, the instru-
ment, the research hypotheses, and an explanation of the design and
methods of analysis.

In Chapter IV, the data are presented with a restatement of
the hypotheses and a probability statement of acceptance or rejection,
followed by a statement of acceptance or rejection based on the sig-
nificance levels. The data are interpreted under the title, Discussion,
in Chapter V.

Chapter V includes a summary of the facts, conclusions about
the findings as they relate to both the theory this research was based
on and on other findings. Finally, implications for the future are

discussed.

CHAPTER II

REVIEW OF THE RELATED LITERATURE

This review of the literature is divided into two main sectors.
Part 1 deals with five aspects of the variable, primary/secondary
attachment object loss. The review follows patterns set by research
in the field and focuses on five questions as they relate to the
variable and the complex behavior pattern. These questions are:

(I) Is P/SAOL associated with negative effects on children and if so,
what are some of the effects noted by investigators? (2) Has research
shown the death of one or both parents to affect children's adjustment?
(3) What are the effects of working mothers on child devel0pment?

(4) Has the loss of father been associated with childhood adjustment
problems? (5) Parental discord often separates parents from each
other and from their children. Has research shown the effect of such
separation on children?

Part II will be limited to one aspect of the complex behavior
pattern, perseverance in school--failure to persevere in school.
Although this research is limited to one variable, P/SAOL, as it may
be related to failure to persevere in school, it becomes necessary to
look at other variables before proceeding with the research. These
five categories commonly associated with failure to persevere are:

(T) intelligence quotient, grade point average, ability, and

ll

12

achievement; (2) irregular attendance and excessive tardiness;

(3) discipline, violence, crime, and drug use; (4) socioeconomic
status of the family; (5) broken homes and self esteem. By gaining
some understanding of these variables and their possible relationship
to failure to persevere in school, this research can be designed to

more clearly focus on the variable, P/SAOL.

PART I

Loss of Primary/Secondary Object, General

 

The family is often viewed as society's womb. It provides our
children with love, affection, and security, as well as introductory
contacts with society and the world at large during their term of
maturation. Margaret Mead (l97l) explains,

During the long months of infancy the child's tie to the mother
is a bodily one, the child learns through its own body what the
world is like and what it is like to be a person. As the
mother feeds the child when it is hungry the child learns that
the world can be trusted to supply its needs. . . . There are
no people anywhere, whatever their race and however simple or
primitive their way of life, who do not, in the simplest sense,
differentiate between a child with a recognized father and a
child without one . . . without a recognized father, whether
he is living or dead and whether he is the child's biological
father or another man who stands in his place, the child has
no social existence.
Wiene (T973) conducted a study of the family and peer relationships
as they affect identity and interdependence of five year olds. He,
similarly, concluded, "It appears that the child's parents and family
are primary and fundamental in development of beneficial relationships
with the outside world."
In times past, our children enjoyed the benefits of an

extended family. Several pillars supported a roof of love, affection,

l3

and security over the heads of our children. Today, in the nuclear
family, the pillars are reduced to two, father and mother. If one or
both of the remaining pillars are damaged or broken, how are those
children affected? Bowlby (T972) reviewed more than five dozen studies
conducted over a period of more than sixty years. He developed a
thesis concerning the effects of object loss during childhood on later
adult behavior. Bowlby explains that during an infant's early months,
it learns to discriminate a particular figure, usually its mother, and
develops a strong attachment to her company. It is particularly
attached to her during the latter half of its first year and during
the whole of its second and third years. The child is content when
she is present and distressed when she is absent. It often protests
even momentary separations and will always protest longer ones. The
strength of the attachment commonly diminishes after the third birth-
day but remains strong nevertheless throughout the period of
maturation.

A minority of children experience a loss of their primary
attachment objects. Mothers die, desert, are institutionalized, etc.
Bowlby believes a sequence of behaviors can be predicted when a child
is separated from his mother if he has experienced a reasonably secure
relationship to his mother during his first fifteen to thirty months.
He describes three phases the child will pass through: protest,
despair, and detachment.

At first, during the protest phase, he will demand his mother
back with tears and anger and seems hopeful of succeeding in getting
her back. As he enters the phase of despair, he becomes quieter but

is still preoccupied, as much as ever, with the absence of his mother.

14

He still yearns for her return, but his hopes have faded in the phase
of despair. He may alternate between these two stages as hope fades
into despair and renewed hope grows from despair. Eventually, a
greater change occurs. The child appears to forget his mother and if
she returns he remains uninterested in her and may not even appear to
recognize her. This phase is called detachment. During this stage
the child is prone to temper tantrums and may exhibit violent destruc-
tive behaviors.

When a child is returned to his mother after reaching the
early state of detachment, he may be unresponsive for an hour or even
a day or more. When his intense ambivalent feeling for his mother
finally breaks, it may do so in a storm of feeling for her. He may
exhibit intense clinging, acute anxiety, and rage when his mother
leaves him even for a moment. On the other hand, if he is separated
from his mother for more than six months or separations have been
repeated, there is danger he may reach the advanced stage of detach-
ment and may remain detached, never recovering his affection for his
parents.

Many studies associate the antisocial destructive behavior
described by Bowlby in the detached phase with parental deprivation.
Brown and Epps (T966) studied 546 women and T68 men prisoners. They
found a significant relationship between adult crime and unstable
parent figures in childhood for both males and females. Glueck and
Glueck (l95l) found that parents of delinquent children were nearly
twice as likely to be divorced or separated (2l.l%) as were parents
of nondelinquents (ll%). Parents of delinquents were found to be

three times more likely to be unmarried to each other (6.3%) as were

15

parents of nondelinquents (2.2%), and delinquents were found to have a
(4.l%) higher rate of parental death than nondelinquents. Costin
(T972) lists problems related to parents not married to each other as
one of five major problems facing American children.

Deutsch (T967) found overcrowding in the home related to poor
school performance. Interestingly, the intact homes in his study were
more crowded than broken ones, but children from intact homes did
better in scholastic achievement. He concluded that crowding in the
home is less likely to have a negative effect on scholastic achievement
than is the fact of coming from a broken family background.

Holzner and Ding (T973) related an unstable family life with
school withdrawal and drug addiction. Greer (T966) compared the
frequency of parental loss among 466 patients with neurotic and psycho-
pathic disorders with parental loss found in the control group.
Parental loss was recorded where there was a definite evidence of loss
or continuous absence of parents for at least twelve months before the
age of fifteen. He found four times the number of losses of both
parents among the experimental group (l6%) as he did from the control
group (4%). He categorized the parental losses as: (1) death,

(2) separation, (3) divorce, (4) illegitimacy, and (5) others. He
found no difference between the experimental group and the control
group in the incident of loss due to death, specifically.

Despert (T953) concluded that illness of children is frequent
when the parents are emotionally rather than legally divorced. She
found earlier (Despert, T946) that unfavorable birth conditions are

important contributing factors in genesis of anxiety in young children.

16

Over the past eighty years researchers have sought to relate
abnormal behaviors with loss of a loved object, pathological mourning.
and childhood experiences. Most investigators have taken as their
starting point their abnormal patients. Freud (1897) was one of the
first to do 50. Mason (172), on the other hand, studied successful
women. He studied 485 women with Ph.D. degrees who held the rank of
Assistant Professor or above. The women were discovered to have five
common characteristics, one of which was that they all came from stable
homes in which both parents were present while they were growing up.

The biological parents affect their children in another way,
genetically. Erlenmeyer and Kimling (T963) reviewed fifty-two studies
and concluded that his survey of the literature over the past fifty
years revealed a remarkable consistency in accumulated data relating
mental functioning to genetic potentials. For unrelated people in a
large random-mating population, the theoretical correlation for general
intelligence is considered to be zero. He found a (+.23) correlation
for foster siblings whereas for siblings reared apart, the correlation
was (+.42) and siblings raised together scored (+.49). They found the
foster parent-child correlation to be (+.20), and the true parent-
child two and one-half times as great, (+;50). Clearly, parents affect
their offsprings' intelligence. Honzik (T967) associated intelligence
and environmental setting. She found sons with a close mother-son
relation had higher I.Q.'s. She similarly found daughters' test
performance positively related to fathers' friendliness and to parental
compatibility. Deutsch and Brown (1964) found that the "father-in-
home“ is a variable positively associated with children's I.Q.'s.

Clark-Stewart and Alison (1973) suggest that responsive maternal

l7

behavior influenced the child's intellectual development. The child
can and does reflect his parents' attitudes as early as 6-9 months.
On the other hand, Antonovsky (T959) and Ausubel (1958), found that
although infants react to parental attitudes, the reaction is not as
intense or permanent as has been supposed. Boshier and Thom (1974)
found conservatism and liberalism in parents associated with similar
attitudes in their children and greater attitudinal similarities were
found between father and son and mother and daughter.

LaRoch (1933) found a significant difference between the step-
father and the natural father in the involvement of child rearing.
The areas she found significant were guidance, discipline, and decision
making. Considerable concern has been expressed as Indian children's
educational level is less than one-half that of the national norms
whereas their dropout rate is more than double the national norm.
Some studies have shown that the more English Indian children speak
before they start school, the more successful they are in school.
With these and other facts in mind, the Church of Jesus Christ, Latter
Day Saints (Mormons) initiated a foster home placement program. They
hoped to improve these Indian children's scholastic achievement and
intelligence. A study of the program was reported by Cundick et a1.
(1974), Brigham Young University. Eighty-four children, ages 8-14
were enrolled for five years at the time of the study. Iowa Test of
Basic Skills and the California Test of Mental Maturity test scores
were compared with children in other Indian educational programs and
with national norms. The children's test scores declined with

successive years of participation in the foster home program. 1.0.

18

scores showed no significant change. Canning (1974) found that foster
children have serious problems of adjustment to the school environment.
Freud laid the foundation in the last century. Building on
his legacy, researchers continue to this day to associate the child-
parent relationship to success or failure to relate with the outside
world. This review has shown the following general variables to be
correlated with P/SAOL: general social adjustment (Wiene, T973; Mead,
1971); general success in life (Mason, 1972); emotional security
(Bowlby, 1972); delinquency (Glueck and Glueck, lSl); drug addiction
(Holzen and Ding, T973): and neurotic and psychotic disorders (Greer,
1966; Despert, T953; and Freud, 1897). It is not difficult to see a
possible relationship between these variables and failure to persevere
in school. Indeed a direct association will be shown with most of
these variables in a later portion of this review. Another group of
studies more clearly relates P/SAOL with perseverance in school.
Erlenmeyer and Kimling (1973), Honzik (T967, and Clark-Stewart and
Alison (T973) associate intelligence with "parents in home." Canning
(1974) found foster children have serious problems adjusting to the
school environment. Cundick et al. (1974) found children separated
from their parents do poorly on standardized tests. Mason (1972)
found perseverance in school is directly associated with a stable
family life with both parents present in the home while the children
were growing up. Holzen and Ding (1973) found early school withdrawal

related to an unstable family life.

19

Primary/Secondary Attachment Object Loss by Death

 

Death is the most clear cut and irreversible of all causes for
parental loss. It is the most universal, too. Death is unaffected by
the economic status, cultural status, geographical setting, and his-
torical setting of the family. Unless we first experience death
ourselves, we all experience the loss of our parents through death.
Death is different from divorce and separation in another way. It is
not the result of personal or interpersonal discord in the family.

The percent of children who experience the death of a parent
by the time they reach sixteen years of age has been found to be as
high as 19.5 percent (Munro, T965) and as low as 12 percent (Brown,
1961).

Most bereavement studies explore the incident of P/SAOL by
death among adult psychiatric populations and compare such statistics
to the nonpsychiatric population. Inferences are then made regarding
the significance of childhood P/SAOL by death for later psychiatric
disturbances. Such studies cannot therefore tell us about the typical
child who lost a parent due to death. The readers can, however, by
reading these studies, gain some insight regarding the mental perils
of childhood P/SAOL by death.

Brown (T961) found 41 percent of a group of depressed patients
lost a parent before age fifteen compared with a 12 percent loss
established by the death census. He concluded that bereavement in
childhood, P/SAOL, is one of the most significant factors in the
development of depressive illness. He theorizes that depression recurs
in later life, triggered by minor rejection situations, in children

apparently recovered from P/SAOL by death. Barry (1949) found

20

maternal deaths occurred in the histories of psychotic patients with
increased frequency before the patients were eight years of age. Barry
and Lindemann (1960) found that death of the mother before the child is
five years of age is more frequent in psychoneurotics than in the total
population. They found the most critical age for maternal death in
little girls who developed neuroses is from birth to age two.

Birchnell (T970b) suggests that for psychiatric patients as a whole the
experience of P/SAOL by death is a significant predisposing factor to
mental illness.

Greer (1966) found that P/SAOL was four times as common among
patients who attempted suicide as among the control group. On the
other hand, when P/SAOL was divided into categories--death, separation,
divorce, illegitimacy, etc.--there was no difference between their
attempted suicide and controls in the incident of P/SAOL by death.

Hilgard et al. (1960) examined normal adults from the general
p0pu1ation who suffered P/SAOL by death in childhood. They found
21 percent of the subjects experienced P/SAOL by age nineteen, whereas
27 percent of state mental hospital patients sustained a similar loss
by age nineteen. Of the general population, twenty-nine (29) subjects
were women who lost fathers. Fourteen of the twenty-nine women were
classified as well adjusted. This definition was based on an intact
home, a satisfactory marriage, and adequate relations with children.
Similar information was not available for other subjects or for these
twenty-nine subjects relating to maternal loss.

The mother-child relationship appears to outweigh the father-
child relationship in most studies. Barry and Lindemann (1960) found

the relationship between death of mother and subsequent deveTOpment

21

of psychoneurosis to be consistent with what other investigators found
for other psychiatric conditions including psychosis and delinquency.
Barry (1949) and Barry and Lindemann (1960) all found a relationship
between loss of mother and later psychoses. They found the death of
the father, on the other hand, followed a pattern which is more nearly
that which is found in the general population. Brown (1961) found the
death of the father was more significant in age groups 5-9 and 10-14,
whereas the loss of the mother was equally significant in each of the
five year periods. Barry (1949) found maternal bereavement due to
death before age eight may be a sensitizing factor in the development
of subsequent psychoneurosis.

Death of a parent or parents is an experience common in the
general population. Twelve to 19 percent of our children experience
the death of a parent by age sixteen. Most studies explore the inci-
dent of P/SAOL by death among adult psychiatric populations and
compare such statistics to the nonpsychiatric population. Almost all
such studies find the incident of P/SAOL by death to be significantly
higher among psychiatric patients than among nonpsychiatric patients.
Investigators theorize that the depression occurs in later life,
triggered by minor rejection situations. ‘The death of the mother
appears to be more significant to later depression than is the death
of the father. One study examined the incident of P/SAOL by death
experienced in the general population and compared such loss to the
incident sustained by state mental hospital patients. State hospital
patients experienced 9 percent more P/SAOL by death than did the
general population. Delinquency, too, has been related to death of

parents.

22

There is some general agreement in the literature. Childhood
P/SAOL due to death is more common among neurotic, psychotic, and
depressed patients than are similar deaths in the general population.
The death of both parents is more significant than the death of only
one parent. The death of the mother is more significant than the death
of the father. Investigators are not in agreement concerning the

significance of the age category at the time of the bereavement.

Primary Attachment Object Loss: Working Mother

The most common cause of primary attachment object loss is of
a temporary repetitive nature caused by mother working out of the home.
The U.S. Department of Labor (1976) reports that women made up 37 per-
cent of the work force in 1952. This figure increased to 39 percent
in 1959 and to 45 percent in 1975. Bowerman and Elder (1964) found
women more frequently play a primary role in the family. The U.S.
Department of Labor, Women's Bureau (1972) reports the increasing
percentage of mothers entering the work force. In 1940 only one out
of ten mothers with school aged children worked. In 1960, three out of
every ten such mothers worked, and by 1971, four out of ten mothers
were working. One out of every three working mothers had children
under six years of age in 1962, and 16 percent of these children were
fatherless. By 1974 more than half of our mothers with school aged
children (54%) were working out of the home. Female heads of house-
holds increased from 10 percent in 1950 to 23.6 percent by 1975
(U.S. Department of Commerce, 1976).

The effect of the working mother on the child's education is

not as clear as we might desire. Banducci (1964) studied more than

23

3,000 seniors of working mothers. He measured three socioeconomic
groups which he labeled laborers, skilled workers, and professionals.
Among the first two groups he found that mothers' working had little
if any detrimental effect on the children in regard to educational
aspirations and expectations. He found a slight trend for educational
aspirations and expectations to be higher than those measured from
nonworking mothers. Among the children of the professional mothers,
however, he measured their educational aspirations and expectations
lower than those found in the nonworking mothers' children. Glueck
and Glueck (1951) and (1957) found the working mother had a special
impact on delinquency. In a more recent study, Hoffman and Nye (1974)
found a similar relationship between working mothers and delinquency
in the middle class. Hoffman (1963) found, however, that maternal
employment and delinquency are strongly associated with broken homes
and when the latter variable was controlled the relationship dis-
appeared. Hoffman found, too, that children of working mothers had
Tower 1.0. scores and lower performance records in school. In a
study of dropouts, Van Dyke and Hoyt (1958) found mothers' working
was not associated with dropping out of school.

Honzik (1967) found sons' I.Q.'s to be higher when they
experience a close mother-son relationship. Clark and Alison (1973)
found a relationship between maternal behavior and her child's
intellectual development. Brown (1974) found lower scores on
California Achievement Tests for middle class eighth and ninth grade
sons of working mothers. Roy (1974) found, on the other hand,
children of working mothers among rural high school students were more

likely to go to college.

24

To summarize, a review of the literature indicates a relation-
ship between working mothers and variables, delinquency, I.Q., and
school performance. The findings are not consistent and investigators

agree on the importance of further studies.

Loss of Secondary Attachment Object

 

The loss of secondary attachment objects (fathers) is even
more common than the loss of primary attachment objects. Six out of
ten mothers of school children worked in 1975 (U.S. Bureau of Labor
Statistics, 1977). Sixteen percent of the children involved, 4.1
million, were fatherless. Although father's working out of the home
is regarded as normal, Landy et al. (1969) found it feasible to treat
long term nightshift work by fathers on a father-present-absence
continuum. They found the effect of father's absence due to night-
shift work similar to father's absence from the home. They found such
absences negatively affect school performance as measured by American
College Entrance Examinations. They found, too, father's absence due
to nightshift work has a negative effect on the cognitive development
of daughters under age nine, an effect similar to that caused by
father's complete absence from the home.

Honzik (1967) related the father's presence in the home and
his friendliness to his daughter's test performance. Mead (1971)
relates the father's presence to the child's social existence and to
his son's sense of triumph all of his life. Bunch et al. (1971) found
more than the expected number of suicides kill themselves in a time
period close to the anniversary of their father's death, but not to

the mother's death. They found more females than males suicided near

25

their father's death anniversary. Munro (1965) views the loss of the
father as a more emotional traumatic experience than the loss of the
mother.

In a study on relationships between selected factors and
school achievement among Mexican-American children, Cain (1970) found
the father's presence and love positively related to arithmetic
achievement. He found father's casualness negatively related with
reading and general performance. Deutsch and Brown (1964) and Deutsch
again (1967) investigated experiences which influence the development
of intellectual functions in children. They found "father in home"

a variable positively associated with children's 1.0. They explain
the lowering of Negro children's I.Q., in part, by the prevalence of
broken homes in the Negro pOpulation. I

Kutcher (1972) studied absenteeism in low income schools. He
found the father's presence in the home and his involvement with
discipline moderately correlated with attendance and positive group
adjustment. Bowerman and Elder (1964) found high school scholastic
performance among high school boys most frequent when the father was
seen as the most powerful family decision maker. Burns (1972) found
the father's occupational status to be positively correlated with
student success in school, as measured by the Michigan Assessment
Program. Sutton-Smith et al. (1969) found the father's absence has a
depressive effect throughout the child's cognitive deveTOpment with
‘the greatest effect during early and middle years. Boys without

t>rothers were found to be more affected than those with brothers.

L.aRoch (1933) found a difference at the .001 level between the natural

26

father and the foster father in the involvement of child rearing,
particularly areas of guidance, discipline, and decision making.

The number of fatherless homes is increasing in the United
States. A review of the literature indicates a positive relationship
between father's presence and cognitive development. Nightshift work
may affect cognitive development, at least for many children, in the
same way a complete absence does. Both boys and girls are affected
by their fathers' presence or absence, but the effect may be different
for each sex. There is no clear indication of how the absence affects
children differently at different age levels. The effect of fathers'
absence is often measured in terms of I.Q., test performance, atten-

dance, and social adjustment.

Separation Caused by Parental Discord

P/SAOL caused by parental discord is the largest of all
categories and includes a number of subcategories. Subcategories,
loss of mother and loss of father, for instance were examined under
separate headings in this review. Unlike death, separation can be
influenced by society. Society can affect the incident of separation
caused by parental discord.

Costin (1972) lists "parents not married to each other" as one
of five major problems facing American children. Despert (1953) found
illness of children more frequent when parents are emotionally rather
'than legally separated. The visible discord itself may be more
damaging to the child than the absence of a parent along with the
iabsence of discord. Studies indicate a definite relationship between

sseparations caused by parental discord and later illness. Greer

27

(1966) found loss caused by divorce, separation, and illegitimacy four
times more common among patients who attempted suicide as among his
control group.

Derdeyn (1976) concluded that we need: (1) more empirical
studies to help predict the effect of divorce on children, and (2) more
involvement of child psychiatrists in consultations, education, and
research in this area. Bane (1976) similarly concluded that research
has raised questions rather than answered questions on how best to
meet special needs of children affected by divorce.

Cervantes (1965) associated divorce with dropping out of
school. Glueck and Glueck (1951) found that parents of delinquent
children were nearly twice as likely to be divorced or separated
(2.1%) as were parents of nondelinquents (11%). They found, too,
parents of delinquents were three times more likely to be unmarried
to each other (6.3%) as were parents of nondelinquents (2.2%). Holzner
and Ding (1973) related an unstable family life with school withdrawal
and drug addiction. Deutsch (1967) found children from intact homes
did better in scholastic achievement than did children with broken
family backgrounds. He found that coming from an intact home to be a
more important variable than coming from an uncrowded home. Mason
(1972) studied variables associated with success. One variable common
to all of his successful subjects was both parents in the home while
the child was growing up.

Wallerstein (1976) found children's initial reaction to
separation, caused by divorce, to be reactions of anger and denial as
predicted by Bowlby (1969); Wallerstein goes beyond Bowlby to relate

the child's reactions to school performance, as well as later

28

depressive behavior patterns. His study is of particular interest as
he relates the lowering of the child's self-esteem to separation by
divorce. Self-esteem has in turn been associated with school with-
drawal in many studies. Such associations will be viewed in Part II
of this chapter.

Separation of parents may be one of the major problems facing
American children, according to Costin (l972). Illness (Despert, 1953)
and suicide (Greer, 1965) have been associated with separation.
Cervantes (T965) and Holzner and Ding (1973) found school withdrawal
associated with separation. Glueck and Glueck (1965) found delinquency
related to separation and drug abuse. Deutsch (1976), Mason (1972),
and Holzner and Ding (1973) all found children from intact homes do
better in school and are more successful in life than children from
broken homes. Derdeyn (1976) and Bane (1976) both found research
raises more questions than it answers and recommends further research
to help meet special needs of children affected by divorce.

Research is often criticized on the grounds that other vari-
ables may be responsible for the observations recorded. Bowlby's
(1972) work, already quoted, is criticized on such grounds. Critics
explain the child's deviant behavior may not be due to the loss of the
mother but to his new strange environment, the expectation of a rival,
the quality of the substitute mother, previous unsatisfactory rela-
tions with his mother, etc.

Some studies of childhood behavior in the mother's absence
claim to have accounted for these other variables. Spitz and Wolf
(1946) observed infants who remained in the same institution during

the absence of their mother. It was these infants' behavior which led

29

Spitz to delineate the syndrome, "anaclitic depression." They found
only one change was necessary, provided it occurred within three
months, to restore the children to their previous condition-~the
return of the mother. Two other studies reported the main variable is
always the absence of the mother, Deutsch (1919) and Spiro (1969),
both case studies.

One group experimented by removing eighty-four children, ages
8-14 from their parents. By placing the children in foster homes con-
ducive to learning, they planned to raise the children's scholastic
achievement and 1.0. Cundick et al. (1974) reported the results. Over
a five year period, achievement declined each succeeding year and I.Q.
scores remained constant. Although Hilgard et al. (1960) studied well
adjusted adults who suffered death of a parent as a child, and Mason
(1972) studied successful subjects, such studies are rare.

A few case studies have been popularized and generally support
Bowlby's position that children separated from their mothers too long
or too often may never fully recover. One such case study, Axline's
(T964) Dibs, describes a boy who sat mute for hours and exhibited
violent temper tantrums while his peers were in school learning to
read. Dibs, who was rejected by his parents at early infancy, was
suspected of suffering mental retardation or deep-seated illness but
was later proven brilliant. Wexler's (155) Sandy is a similar case
study. Sandy was rejected by his mother during the first months of
his life and was later given to his grandparents by his father. He
was diagnosed by doctors as hopelessly retarded, an imbecile. Yet,
he~seemingly fully recovered by age eighteen. In both cases the boys

exhibited autistic symptoms.

3O

Bettelheim (1967) described a boy who developed normally until
the age of two when his mother entered a nursing home. The boy
regressed, losing those words already learned, and became autistic.
Bettelheim reports too that recovering autistic children regress
temporarily with the slightest form of rejection. Ekstein (1966),
Eisenberg (T957), Mahler (1963), and Bettelheim all associate autism
with parents, usually the mother. Bowlby (1973) lists inappropriate
mothering as one of three possible causal factors. Other investi-
gators feel another set of variables is as important or more important

than is the parents' variable.

Part I Summary

 

In a review of the literature, not a single study failed to
associate at least one undesirable variable with childhood loss of
primary or secondary attachment objects. Few studies, however, pre-
tended that all separations of children from their parents have
resulted in observed negative effects on children. Most studies
indicate only a minority are affected. The most common research is
one which starts with abnormal subjects and attempts to associate
the symptoms to an earlier experience, P/SAOL.

The literature seems to indicate that most serious mental
illnesses are associated with loss of parents, particularly with the
loss of mother. While working mothers seem to have some negative
effects on children, the effect has not been noted on all children.

In the case of father's absence, not all children are noted to
be affected, but the effects reported were mostly negative. The death

of one or both parents has been found to have a negative effect on a

31

majority of the children. There seems little agreement on how the
length of the loss affects children, except that longer periods may
have a more damaging effect than shorter periods. Similarly, research
does not demonstrate a definite age or age group at which children are
most susceptible or the sex of the child most likely to be affected.
Most research indicates the child is more sensitive to the loss of
mother, and more recent research demonstrates the importance of
father's presence in the home while the child is growing up.

A review of the literature highlights two points. (1) Parents
risk the normal potential development of their child absenting them-
selves from their child. This observation is particularly true if the
absence is often, for long periods of time, or at critical deve10p-
mental periods of the child's life. The danger is even greater for
children whose potential for maturation is rooted in psychologically
weak ground. (2) Childhood loss of primary and secondary attachment
objects is a fertile, potentially productive, and important area in

which research may greatly benefit society.

PART II

A Complex Behavior Pattern: School Withdrawal
Dropping out of school is a phenomenon generally regarded as a
social illness. School withdrawal is often perceived as a waste of
human resources. During periods of full employment, dropping out of
school is deplored as a manpower loss. School withdrawal is suspected
as a liability to society, and the national incident of dr0pout is

closely monitored by the U.S. Government. The U.S. Bureau of Census

32

(1976) reported 23 million teenage dropouts in 1975. Of the amount,
8.4 million were female and 14.6 million were male.

The phenomenon, dropping out of school, is one of the most
common subjects of educational research. Investigators most frequently
seek to identify symptomatic variables related to school withdrawal.
Such variables are sought as indicators of potential dropouts. Inves-
tigators grope for tests with predictive value (Keene, 1976). Finally,
special programs are devised for potential dropouts to strengthen the
holding power of the schools.

This review will now examine those variables most often
related to school withdrawal. They are: intelligence quotient, grade
point average, ability, and achievement; irregular attendance and
excessive tardiness; discipline, violence, crime, and drug use;

socioeconomic status of the family; broken homes; and self-esteem.

I.Q. and Grade Point Average

 

Investigators most commonly associated grade point average,
academic ability, and intelligence quotient with school withdrawal.
O'Neill (1963) believes it is pretty well established that low I.Q.
is one of several factors which lead to withdrawing from school.

Yudin (1973) compared fifty dropouts with fifty graduates admitted for
college. He found I.Q. to be a significant factor as early as the
first grade. Matthews and Bowman (1960) found low I.Q. associated
with dropping out of school and that children from broken homes have
lower I.Q.'s. In a recent study in Michigan, Fuller and Friedrich
(1972) found the average I.Q. for thirty-nine dropouts to be 94.92,

and they measured the average nondropout's 1.0. at 108.63. Cervantes

33

(1965), on the other hand, states that although a low I.Q. is the
intellectual milieu of the dropout, it is by no means a determinant in
withdrawal. Closely associated with I.Q. is grade point average.
Harding (1967) found G.P.A. a variable related to dropouts at the .001
level. Kilpatric (1973) found dropouts do not differ significantly

from nondropouts in G.P.A.

.Ability and Achievement

 

Brookover (1965) participated in a longitudinal study. He
found a significant relationship between G.P.A. and academic ability.
Harding (1967) found academic ability an important variable signifi-
cantly related to dropping out. Tseng (1972), Kowalski (1974), and
Yudin (1973) all found substantially the same thing: school withdrawal
and academic ability are significantly related. Cundick (1974) found
children who are removed from their parents suffered decreasing school
achievement, and Cain (1970) found fathers' love and presence
associated with school success. Berry (1974) administered an Adjective
Checklist to eighty sixteen-seventeen-year-old reentry dr0pouts,
matched for race, sex, and age. Dropouts received scores signifi-
cantly lower than the control group on achievement. The U.S. Depart-
ment of Health, Education, and Welfare (1964) found underachieving
significantly related to school withdrawal. Kilpatric (1973) found
no significant relationship between academic ability and G.P.A., and
Lee and Crump (1963) found ability is not always a deciding factor.
They point out that low achievers often graduate whereas students

with greater ability often drop out.

34

Greaney (1973) found school withdrawal related to verbal
reasoning ability. Fuller and Friedrich (1972) found verbal reasoning
ability, numerical reasoning, abstract reasoning, and mechanical
reasoning all associated with school withdrawal.

Irregular Attendance and
ExcessiveTTransfers

 

Irregular attendance is so strongly associated with school
withdrawal in the minds of investigators that it is often left
unmeasured in their research. It is noted as a significant variable,
however, by O'Neill (1963), U.S. Department of Health, Education, and
Welfare (1964), Kutcher (T972), Yudin (T973), and again by Greaney
(T973). O'Neill (1963) and Yudin (1973) noted excessive transfers

from school to school as significant variables.

Discipline, Violence, Crime,
and'Drugs

Kutcher (1972) found fathers' involvement in discipline

 

positively related to attendance. Discipline problems was one of the
variables related to school withdrawal by U.S. Department of Health,
Education, and Welfare (1964). Yudin (1973) associated negative
behaviors with dropping out of school. Blake (l974)’similar1y
associated delinquent behavior with dropping out and O'Neill found

a relationship between delinquency and school withdrawal. Carter
(1972) found school behavior problems and crime associated with
dropping out. Berry (1974) found aggression related to school with-
drawal. Brown and Epps (1966) found a relationship between crime and

unstable parent figures in childhood for both males and females.

35

Ahmad (1966) found delinquents to have lower I.Q.'s as compared to
normal adolescents.

Holzner and Ding (1973) related an unstable family life with
school withdrawal and drug addiction. Wiener and Egan (1973) studied
one-hundred heroin addicts and related addiction to dropping out of
school. They found, too, as compared to the normal population, these
students are more likely to have only one parent in the family. Loken
(1973) found dropouts twice as likely to use drugs as are nondropouts.
Winburn and Hays (1974) found marijuana, tobacco, and stimulants
significantly more prevalent among dropouts than among nondropouts.
Annis (1975) conducted a longitudinal study of 915 ninth grade
students. For almost every drug used by students, the use was found
significantly greater among dropouts than among nondropouts. They
found evidence of equally strong use by dropouts before withdrawing
from school.

Discipline, violence, aggression, crime, and drugs are identi-
fied as variables closely associated with dropping out of school.
These variables have been related to other variables: broken homes,
irregular attendance, achievement, ability, and the socioeconomic

status of family which are in turn related to school withdrawal.

Social Economic Status of Family

Sexton (1961) studied 285,000 students and 10,000 teachers in
three-hundred big city schools. She found family income is one index
to social class, no matter how the term is defined. She found many
other indices: occupation of the father, type of housing, educational

levels, and others. She found family income highly correlated with

36

all other indices. She found a "lower-class" family will tend to have
a lower income, lower status, lower paying occupations, poor housing,
and lower educational levels. She found a family possessing one of
these characteristics will tend to possess all of them. Carter (1972)
found broken homes, family size, and unemployed parents all associated
with juveniles referred to the court three or more times.

Brookover (1965) found ability in school associated with socio-
economic status of parents. The U.S. Department of Health, Education,
and Welfare (1964) found a relationship between school withdrawal and
family income. Greaney (1973) compared school entrance with terminal
leaves and found socioeconomic status of family an important indica-
tion. Van Dyke (T958) participated in a cooperative research project.
They found children of unskilled workers are nine times more likely to
drop out than are children whose fathers are classified as profes-
sionals. The absent father was found to affect school performance,
too (Landy, 1964). In a study of Caucasian males, Harding (1967)
found a significant relationship between dropping out and socioeconomic
status of family. Kowalski found a relationship between adverse
economic conditions of the family and school withdrawal. O'Neill
(1963) considers it a well established fact that low socioeconomic
status of the family is a factor which leads to dropping out.

The literature is not in agreement, however. Kilpatric (1973)
reported in his study which associated school withdrawal with socio-
economic status that his study did not support other studies. In a
longitudinal study starting in December 1964 and continuing through
November 1971, Blake (1974) found the occupation of the father to have

little influence on school dropouts when student status in school was

37

controlled. He found low aspirations and delinquent behavior associ6
ated with school withdrawal. Blake used the father's occupation as a
measure of socioeconomic status. In a study of dropouts, its causes
and cures, Cervantes (1965) found dropping out is rarely precipitated
by financial conditions. Matthews and Bowman (1960) found only 5 per-
cent of the dropouts in their study claimed to leave for financial
reasons. Matthews found dropouts more often come from broken homes.

Education of parents is, according to Sexton and others
already quoted, one aspect of the socioeconomic status. Van Dyke
(T958), Bledsoe (1959), Williams (1963), and Duncan (1965) all found
parents of dropouts have less education than do parents of nondropouts.
Harding (1967) found the child's concept of the parents' expectations
related to school withdrawal. Kilpatric (1973) found the child's
perceived parental expectations similarly correlated with drOpping out.
O'Neill (1963) and U.S. Department of Health, Education, and Welfare
(1964) found parental attitude toward school affected dropping out.
Kowalski (1974) found lack of reinforcement from the home associated
with school withdrawal.

School withdrawal is associated with socioeconomic status of
the family in most studies. Investigators do not agree, however, that
the family's socioeconomic status is a causal factor in dropping out
of school.

Deutsch (1967) explains a poor cultural environment increases
children's need for stimulation in school. This same poor cultural
background, however, does little to prepare the child to accept school

experiences. Deutsch, therefore, recommends that society accept

38

responsibility--through the schools--to prepare the child for school,

as well as providing school for the child.

Broken Homes and Low Self-Esteem

Deutsch (1967) studied children coming from overcrowded,
broken, and intact homes. In his sample he found intact homes more
crowded than broken ones but found children from intact homes do better
in scholastic achievement. He found coming from an intact home signi-
ficantly correlated with achievement, and achievement scores for his
total experimental group were influenced by the large proportion of
children from broken homes. O'Neill (1963), Matthews and Bowman
(1960), and Carter (1972) found broken homes correlated with school
withdrawal. Holzner and Ding (1973) found school withdrawals associ-
ated with unstable family backgrounds. Cundick (1974) found children
removed from parents caused scholastic achievement to decrease, and
Canning (1974) found children of foster families have more adjustment
problems in school. U.S. Department of Health, Education, and Welfare
(1964) reports associated school withdrawal with weak or absent
fathers. Wiener (1973) found dropouts and drug addicts are more
likely than the control group to have only one parent in the family.
In a study of successful women, Mason (1972) found one constant
factor among all subjects, and that was the presence of both parents
while the girls were growing up. Wallerstein (1976) found a signifi-
cant relationship between problems in school, broken homes, and low
self-esteem. It is interesting to note several investigators found

relationships between school withdrawal and low self-concept.

39

. Fifield (1963) and Beaird (1964) both found school withdrawal
related to poor self-concepts. Kilpatric (1973) found school with-
drawal related to poor self-concept, low parental expectations, and
the absence of educational goals. Blake, too, found dropouts related
to low aspirations, and Berry (1974) related dropping out to poor
self-esteem. U.S. Department of Health, Education, and Welfare (1964)
found poor attitude toward school and education, lack of an educational
perspective, and disinterest in the curriculum related to school with-
drawal. Davis (1962) found poor "self-esteem" and "not wanted"
variables associated with dropping out while O'Neill (1963) found a
correlation between failure to persevere in school and a lack of
goals, unmotivated by the curriculum, inability to adjust, emotional
instability, and personality defects. Cervantes (1965) and Harding
(1967) both found school withdrawal related to self-esteem. Harding
found parental expectations and goals related to school withdrawals.

Tseng (1972), however, found no significant difference between
dropouts and nondropouts in self-esteem. His findings were supported
in a study the next year by Kowalski (1974) who found a relationship
between perceiving fathers' unacceptance and withdrawal, but no
significant difference between dropouts and nondropouts in self-
esteem. Mason (1972) studied the effects of special counseling pro-
grams designed to raise reentry students' self-esteem and achievement.
Analysis of the test data suggested counseling did improve self-

esteem but did not improve achievement.

40

Part II Summary

 

Ability, achievement, I.Q., and grade point average are clearly
associated with school withdrawal as are irregular attendance and,
excessive transfers. So, too, have discipline, violence, crime, and
drugs been closely related to dropping out of school. A review of
the literature indicates that the socioeconomic status of the family,
self-esteem, and broken homes are variables directly correlated with
school withdrawal. These variables are not only related to school
withdrawal but are clearly associated with each other.

Although these variables are clear indices of school with-
drawal, research has yet to prove which, if any, of these indices are
a cause of the phenomenon. It is not clear either that special
programs for potential dropouts affect the incidence of school with-
drawal. Is it possible that terminability precedes symptom identifica-
tion or that the cause and cure for the malady are more closely
associated with the home than the school? Wilson (1976) suggests
factors affecting dropouts are established prior to high school.
Matthews and Bowman (1960) found indicators of school withdrawal as
early as the first grade, and Barrea (1976) found practices contribu-
ting to school dropout begin at an unusually early age. All of the
variables identified in this review are interrelated and associated

with the home as well as with school withdrawal.

CHAPTER III

DESIGN OF THE STUDY

This chapter will present the research procedures used to
determine the dependence of the complex behavior problem, withdrawing
from school and the variable, primary or secondary attachment object
loss. This chapter will include the following:

1. A description of the population.
A description of the sample selection.

A description of the instrument.

wa

A description of the data analysis procedure.
A list of the testable hypotheses.

6. A summary of the chapter.

Population

 

Subjects were drawn from a single school district, Bayside
Public Schools. Bayside, with a population of 18,000, is the largest
of a dozen small cities and villages in a rural agricultural county of
63,000 people (1970 U.S. Census). More than 86 percent of Bayside's
5,807 students, K-12, come from the city itself and its close in
suburbs. The remainder of the students live in the country, mostly on
farms. (See Table 3.1 below.) Bayside was chosen on the basis of

accessability.

41

42

Table 3.l.--Population Distribution for Secondary Students.

 

 

Location Number Percent
City 1,342 67
Suburb 400 20
Country 272 13
Total 2,014 100

 

Bayside, a two hour drive from Detroit, is surrounded with
fields of grain and dairy farms. Within the school district are a
number of small private schools, all church connected: three elementary
schools; a junior high school; a high school; and a college. Residents
of Bayside are within a 45-minute drive of a major university.

Approximately 50 small manufacturing firms are located in the
school district. The Michigan Security Agency reported a total county
labor force of 27,000, as of April 1978. Of this amount 7.6 percent or
2,100 were described as unemployed.

Public records indicate that 51 percent of the current
secondary enrollment are male students while the remaining 49 percent
are females. The students are almost all caucasian. The racial
composition of the secondary school population is pictured in
Table 3.2.

The population studied consisted of those youths from the
school district who possessed all of the following characteristics:

1. Youths who had birthdays prior to February 23, 1962, and were

not more than 21 years of age by March 8, 1978.

43

Table 3.2.--Racial Composition of Secondary Student Population.

 

 

Racial Division Percent of Population
Caucasian 98.8
Latino 1.2
Other 0
Total 100.0

 

2. Youths who attended the public schools for one or more days
between September 5, 1975, and March 8, 1978.
3. Youths who were not found in categories other than "in school"
or "live dropouts" during the period of the study.
4. Youths who had at least one parent living in the area of the
study during the period of the survey.
The population is further defined in Tables 3.3-3.5.

Sample Selection

 

To aid in determining whether a dependent relationship exists
between the loss of parents and later perseverance in school, a random
sample was chosen in the following manner. The names of all streets
and roads in Bayside school district, 371 in all, were cut from city
and county maps provided by the Chamber of Commerce. Each name was
cut into equal sizes of paper to assure equal chance of selection.

The names were placed in a hat and thoroughly mixed. One name at a
time was drawn by the blindfolded investigator and placed on the list
in order of its drawing. This process continued until a predetermined

number (9%) of the streets were drawn.

44

Table 3.3.--Secondary Students--Population (February 23, 1978).

 

 

Year Number
1977-78 In-school population, 16 years of age and older 1,100
l977-78 Drops, excluding deaths, transfers, etc. 74
1977 Graduates 397
1976-77 Drops, excluding deaths, transfers, etc. 201
1976 Graduates 388
1975-76 Drops, excluding deaths, transfers, etc. 205

Total 2,365

 

Table 3.4.--Secondary Students--Persevering Population (February 1978).

_‘

 

Year Number
l977-78 In-school population, 16 years of age and older 1,100
1977 Graduates 397
1976 Graduates 388

Total 1,885

 

Table 3.5.--Secondary Students--Dropout Population (February 1978).

 

 

Year Number
1977-78 Drops, excluding deaths, transfers, etc. 74
1976-77 Drops, excluding deaths, transfers, etc. 201
1975-76 Drops, excluding deaths, transfers, etc. 205

Total 480

45

As names were selected, they were marked on a map and every
home on that road or street was visited. On first contact, an occupant
was found in 48 percent of the homes. A number of occupants were home
for unusual reasons, with illness being the most common reason. Other
occupants were home because their automobiles would not start or were
stuck in the snow or ice.

If no one answered the door, people in the first or second
house to the right or left of the homes were asked if the home con-
tained subjects. If these people were not home either, or they did not
know, the people in the three homes most directly across the street
were asked. If this method failed, names were taken from the house,
the mailbox, or from neighbors, and the people were contacted by phone
in the evening and residents were identified as subjects or nonsubjects.
If few people were home in a neighborhood, the area was revisited later
in the day or in the evening.

During the first one-third of the survey, the interviewer
obtained a second opinion from another neighbor if he was informed no
subjects lived in a home. Ten percent of the double-checked homes
were cross-checked again by phone. No information pertinent to the
study obtained from adult neighbors was found to have been in error.
During the second third of the survey, second opinions were not
obtained from a second neighbor, but one out of twelve homes was
double-checked by phone. No errors were found. In the final third
of the survey, empty homes were double-checked with any two neighbors
who claimed to know the occupants. Spot checking by phone was dis-

continued. Subjects identified by neighbors were revisited in their

46

home or interviewed by phone. Parents of three drop subjects and two
persevering subjects were interviewed by phone.

Over one hundred and fifty homes were surveyed in a nine-hour
period between 9:00 A.M. and 6:00 P.M. Return visits were made later
in the same day, evening, and weekends. Homes were recontacted until
the resident family was identified as having subjects or nonsubjects.
Seven contacts were required in one instance before the subject was
identified and an interview was completed.

Between February 23, 1978, and March 8, 1978, more than 2,600
homes were contacted. Parents of 142 subjects, 40 drops, and 102
perseverers were interviewed. Occupants of streets were similar and
streets were similar in the amount of subjects produced. Nearly
9 percent of the streets in the district were sampled. Only one
occupant refused to cooperate. Subject identification was not deter-
mined for this occupant. Five fathers were interviewed alone.
Fathers and mothers were interviewed together on eleven occasions.

On one occasion, paternal grandparents who lived across the street
from the subject family were interviewed in place of the parents who
were awayii Two adoptive mothers were interviewed, one the mother of ‘
twin boys adopted at the age of six weeks, and the other the mother
of a boy adopted at the age of three days. All other interviewees
identified themselves as the biological mothers of the subjects.

The subjects identified in this survey included 76 males and
66 females. Three of the males were of Latino origin. The remaining
males and all females were Caucasians.

Listed below are Tables 3.6-3.8 which compare sample findings

with: population distribution; racial distibution of the population;

47

Table 3.6.--A Comparison of Population and Sample Distribution Based
on Location.

 

 

Location Percent of Population Percent of Sample
City 67 69
Suburbs 20 22
Country 13 9
Total 100 100

 

Table 3.7.--A Comparison of Population and Sample Distribution Based

 

 

on Race.
Racial .
Division Percent of Population Percent of Sample
Caucasian 98.8 97.9
Latino 1.2 2.1
Other 0 0
Total 100.0 100.0

 

Table 3.8.--A Comparison of Population and Sample Distribution Based

 

 

on Sex.
Sex Percent of Population ‘ Percent of Sample
Male 51 53
Female 49 47

Total 100 100

 

48

and sex distribution of the population. The findings pictured in the
tables suggest that the sample finally achieved in this survey was

quite representative.

Procedure

Bayside has a city ordinance requiring door-to-door solicitors
to register with the local police. This investigator registered with
both the local police and the Chamber of Commerce.

As street names were randomly selected, every house on the
street was visited. At each door the investigator knocked or rang and
waited twenty seconds, knocked or rang and waited fifteen seconds,
knocked or rang a third time and waited another fifteen seconds. If
no movement was detected by the end of the third wait, the initial
contact was terminated.

Once a resident answered the door the investigator presented
his Michigan State University identification card and stated, "I am

from Michigan State University and I am conducting a survey."

Question Sequence

 

The questions listed below were devised to collect data which
were used to answer the research questions identified in Chapter I.

1. Do you have members of your family or are there people living
in this home, age 16 through 21?

2. Has any member of your family, age 16-21, moved away from
your family or home?

3. How many people in your family or home fall into this
category?

4. Subject's sex is male or female?

49

5. Subject's age is?

6. You are the subject's mother?

7. Is subject in school? Graduated? Dropped? From subject
school?

8. At what age did subject discontinue school?

9. As subject's mother, how would you describe subject's ability
in school, not necessarily subject's mark or what subject
accomplished in school, but subject's real ability: above
average, average, or below average?

10. Would it be accurate to say your family income is above or
below $15,000? Below $50,000 or over?

11. You said you are subject's mother. Are you subject's real
mother? The mother to whom subject was born?

12. Has subject always lived with you? And with subject's real
father? The father to whom subject was born?

13. Subject's real father is deceased then? Divorced?

14. What was subject's age at the time of the death, divorce, etc.?

15. Did subject ever have another father? At what age? Does
subject still live with second father?

Subsequent fathers, mothers, ages, and durations were recorded.

16. From the time subject was born until subject started school,
was there ever a time when you were away from subject for more
than two weeks, due to illness, vacation, or any reason? Was
the father away for more than two weeks due to business,
illness, vacation, etc.?

17. At what age? For how long?

18. Did you ever work out of the home?

50

19. What age was subject when you started working? How long did
you work?
As the interview ended, the investigator thanked the parent for the
information.
The chart below shows the relationship between the variables,
the hypotheses, and the instrument. The questions listed below were

devised to elicit data about the variables to test the hypotheses.

Question Anelysis

 

A rationale for the questions in this survey follow in the
numerical order of the hypotheses to which they relate.

V1 Persevere--All questions relate to persevere and persevere

 

is measured against all other variables. A relationship between
school perseverance and parental loss was suggested by Deutsch (1967),
O'Neill (1963), Matthews and Brown (1960), Carter (1972), Holzner
(1973), Cundick (1974), and Canning (1974).

V23 Number of Losses-~Bow1by (1972) concluded that repeated

 

loss of mother is dangerous to a child's emotional development.
V2_Age--In a review of the literature Bowlby (1972) found
children react differently to parental loss at different ages. Barry
(1949), Brown (1961), and others found children exhibit somewhat
similar reactions if they fall into similar age groupings, 0-5, 6-11,
and 12-18 years of age. Due to insufficient data collected in the
last two categories and due to the fact that mothers often remembered
events as happening before or after their child started school,

categories preschool and school age were used in this research.

51

V17 Duration--Bowlby (1972) concluded that if preschool

 

children are separated from their mothers for more than six months,
there is danger that they may never recover their emotional attachment
to their mothers. Glueck (1951), Deutsch (1967), and Costin (1972)
found prolonged separation related to later emotional problems.

V2 Sex of Child--Research (see Chapter II) shows young boys

 

sometimes react differently to a parent's absence than do young girls.

V3 Sex of Parent--Most recent research, Smith (1969), Burns

 

(1972), and others show that the sex of the lost parent determines to
some extent the type of behavior the affected child will exhibit.

V14 Socioeconomic Background of the Family--In this study

 

family income was used as an index of social class. Sexton (1961)
studied 285,000 subjects. She found family income an accurate index
of both social class and perseverance in education. In this study,
$15,000 was used to separate below average from average income. This
sum was used at the suggestion of welfare personnel in the county
where the study was conducted.

V13 Ability--Ability is a common variable found related to

 

perseverance. In this study the variable, mother's perception of her
child's ability, was used. The perception variable has been used by
Fifield (1963), Beaird (1972), and by others (see Chapter II).

V15 Cause--A review of the literature indicates the following
types of childhood parental losses most affect children: death;
parental discord; working mothers; and others. Due to insufficient

data, death and others were combined in this study.

52

Table 3.9 listed below further shows the relationship between
the research hypotheses and the individual questions designed to

gather data to test the hypotheses.

Interview Procedure

 

The first portion of the interview, questions 1-8, was struc-
tured to elicit specific responses: yes, no, age, sex, etc. After
providing the information, mothers volunteered explanations for their
answers. Although the additional information was not useful, mothers
were not discouraged from talking about their children whom they were
often proud of or disappointed in. Frustration was often vented upon
the school. Question 9, the mother's description of her child's
ability in school appears as a structured question calling for one of
three specific answers. In fact, all mothers gave more than a specific
answer and in every case justified the answer they gave. If rapport
was not established prior to this question, it was established at this
point and probably assured a response for the income question which
followed.

With question 11, the survey was unstructured and by this
point mothers were almost always persuaded that the interviewer was
interested in the parent-child relationship and not the husband-wife
relationship. Most mothers answered with pride that there were no
separations or divorce in their family. Those mothers who experienced
a divorce usually volunteered a cause.

Although specific questions were elicited from subjects, the
“interview procedure was informal and unstructured. All questions were

(open-ended in that subjects felt free to add to, quality, or explain

53

Table 3.9.--The Relationship Between the Hypotheses and the Individual
Questions Designed to Collect Data to Test the Hypotheses.

 

 

Hypotheses Variables Questions
Primary V1 8 V23 Number--all loss 11-13, 15, 16
1 V1 8 V9 Age--preschool, school age 14
2 V1 8 V17 Duration--l yr. or less,
1 yr. plus 5. 14. 7
3 V1 8 V23 Number--one, one plus,
no loss 11-13, 15, 16
4 V1 8 V23 Number--a11 loss, no loss 7, 11-13, 15,
16
5 V1 8 V2 Sex--child--male, female 4
6 V1 8 V3 Sex--parent--father, mother, 6, ll-13, 15,
both 16
7 V1 8 V14 Socioeconomic--below average,
average, above average 10
8 V1 8 V13 Ability--below average,
average, above average 9
9 V1 8 V15 Cause--separation, working
mother, other 11-13, 15
10 V1 8 V15 Cause--separation 11-13, 15
11 V1 8 V15 Cause--working mother 18, 19
All hypotheses V1 Persevere--drop, persevere l-l9

Subject identification .

. l, 2, 3, 5, 8

 

54

their answers. The financial question was the only question in which
the investigator discouraged more specific information. Listening was
used by this investigator as a means of establishing rapport. Subject
parents were surprisingly willing to answer all questions with apparent
openness and honesty. Nearly 80 percent of the parents responded with
both a smile and a "you are welcome" at the end of the interview.

More than 10 percent of the parents thanked the interviewer for coming.

TestableyHypotheses

 

Listed below are the hypotheses which were tested with the

data collected with the questionnaire.

Primary Hypothesis

 

The relative intactness of the family and subjects' persever-

ance in school are statistically independent.

Secondaryyflypotheses

 

1. The age of subjects at the time of loss is statistically
independent of perseverance in school.

2. The duration of loss of parents is statistically independent
of perseverance in school.

3. The number of losses of parents experienced by subjects is
statistically independent of perseverance in school.

4. The proportion of subjects suffering no loss of parents and
persevering in school is equal to the proportion of subjects
suffering loss of parents and persevering in school.

5. The sex of subjects is statistically independent of persever-

ance in school.

55

6. The sex of lost parents is statistically independent of
perseverance in school.

7. The socioeconomic background of the subjects' families is
statistically independent of perseverance in school.

8. The subject's ability in school is statistically independent
of perseverance in school.

9. The cause of the subjects' loss is statistically independent
of perseverance in school.

10. The proportion of perseverance in school for subjects experi-
encing loss of parents caused by separation is equal to the
proportion of perseverance in school for subjects experiencing
no loss of parents.

11. The proportion of perseverance in school for subjects experi-
encing loss of parents caused by working mothers is equal to
the proportion of perseverance in school for subjects experi-
encing no loss of parents.

12. Substitute parents and substitute surroundings are statis-

tically independent of perseverance.

Analysis

The Michigan State University Standard Statistical Program was
used to analyze the data. Contingency tables of analysis were used
as the data were frequency counts. A chi square test was used for the
primary hypothesis and for hypotheses 1-3 and 5-9. Hypotheses 4, 10,
and 11 require examination of proportions and the t-test for propor-

tions was therefore used. Alpha was set at .05 level of confidence

56

for both tests and the data collected from the selected streets were

generalized to the population.

Summary
One hundred and forty two (142) subjects between the ages of

16 and 21 were randomly selected from a single school district. Forty
(40) of the subjects were defined as dropouts and 102 were defined as
perseverers. Mothers of the subjects were interviewed individually in
their homes and no subjects were lost.

The data were organized into categories to test the hypotheses,
to measure the independence of variables, loss of parents, and per-
severance in school. The primary hypothesis, hypotheses 1-3 and 5-9
were tested with the chi square test of independence while hypotheses
4, 10, and 11 were tested with the t-test for proportion. Dependence
was determined if either the t-test or the chi square was significant

at .05 level of confidence.

CHAPTER IV

ANALYSIS OF DATA

Introduction

 

The data collected for this research are presented in this
chapter. Each hypothesis of this study is stated and followed by the
statistics related to the hypothesis.

The original data presented in this chapter were collected from
personal interviews with the mothers of 142 subjects between the ages
of 16 and 21 years of age. The interviews were conducted according to

the methods described in Chapter II.

Procedure

A chi square test of significance or a t-test for proportion
is submitted for each comparison of variables. The major variable,
perseverance in school or dropouts, is tested in each hypothesis
investigation against one of three other major variables: student
ability as perceived by mother; socioeconomic status; and relative
intactness of family. This last variable is further broken down into
five categories: age of child at loss, duration of loss, sex of the
lost parent, and the cause of loss.

Originally proposed was a further categorization of the intact-
Iiess variable according to the quality of the substitute homes and

TJarents. Secondary hypothesis 12 was to have been tested against

57

58

these quality variables. In the data gathering, it became evident that
the subjectivity of the responses and the difficulty of distinction
among the various quality factors made this both imprudent and unpro-
ductive to pursue. This hypothesis then remains untested.

In both statistical tests that were used here, the chi square
and the t-test for proportion, it was predetermined that alpha must
exceed the .05 level of confidence before the hypothesis would be
accepted.

The data presented in this chapter were collected from a
single Michigan school district. The 142 subjects were randomly
chosen, and all participated in the study.

The analysis of the data gathered in the interviews with
respect to the childhood loss of parents and later perseverance in
school is presented as follows:

1. The primary hypothesis and subsequent secondary hypotheses are
presented in the same order in which they were presented in
Chapter I.

2. The source table for the chi square or for the t-test is
presented with each subproblem.

3. The .05 level of confidence for the degrees of freedom stated
will be used as the criterion of accepted or rejected for
each hypothesis.

4. A report of the findings on the research hypotheses will
follow the presentation of each research table.

This research was based on 1,363 separate observations,
collected from 142 subjects, which were used to test 12 hypotheses.

No subjects were lost, but 4 observations were missing on a

59

persevering subject and 1 observation was missing from a dropout
subject. Each of the 5 missing observations is noted at the base of
the appropriate research table. Observations missing in 4 out of the

12 tests were taken into account in the statistical procedures.

Test of Primary Hypothesis

The primary hypothesis is that the relative intactness of the
family and subjects' perseverance in school are statistically indepen-
dent.

Data gathered from personal interviews with mothers of the
subjects are set out with their statistical significance in Table 4.1.
The comparison is between youths who have or have not experienced a
loss of a parent with respect to whether they persevered in school or
dropped out. It presents, too, the figure for the corrected chi
square with 1 degree of freedom.

Table 4.l.--A Comparison of Intactness of Family (Loss/No Loss) with
Perseverance in School (Drop/Persevere).

 

No. of Cases with Chi Square with

 

 

Perseverance No. Of History Of: 1 Degree
Cases of Freedom
Intact Family Loss of Parent
Dropout 4O 8 32
Persevere 102 52 50
Total 142 60 82 10.07*

 

*Significant at the .05 level.

60

The research demonstrates that relative intactness of the
family and the subjects' perseverance in school are not independent.
As the chi square figure of 10.07 is significant at the .05 level of

confidence, the null hypothesis is rejected.

Test of Secondary_Hypotheses

 

Twelve secondary hypotheses were tested. For nine of them the
statistical test, the chi square test of significance, was applied.
For three of them the t-test for proportion was applied. In each, the
critical level of confidence was established at .05.

Data for all were gathered in personal interviews with mothers
of the subjects and were tabulated according to the several variables

and subvariables.

Secondary Hypothesis 1

 

The age of subjects at the time of loss is statistically
independent of perseverance in sChool.

Table 4.2 sets out the data on perseverance in school
(Dropout/Persevere) for all subjects who have experienced a loss of
parent. It compares those whose loss occurred before their sixteenth
birthday with those whose loss was on or after that birthday. Finally,
it presents the figure for the corrected chi square with 1 degree of

freedom.

61

Table 4.2.--A Comparison of Age at the Time of Loss (Under Six Years
of Age/Six Years of Age and Older) with Perseverance in
School (Drop/Persevere).

 

No. of Cases with

 

 

No. Parental Loss at: Chi Square with
Perseverance of 1 Degree
Cases 0-5 Yrs. 6 Plus Yrs. of Freedom
of Age of Age
Dropout 32 8 24
Persevere 49* 13 ‘ 36
Total 81 21 60 .Ol**

 

*1 missing observation, persevere.

**Not significant at the .05 level.

The results demonstrate that the age of subjects at the time
of loss of parents and perseverance in school are independent. As the
chi square figure of .01 is not significant at the .05 level of confi-

dence, the null hypothesis is not rejected.

Secondary Hypothesis 2

 

The duration of loss of parents is statistically independent
of perseverance in school.

Table 4.3 sets out the data on perseverance in school (Drop/
Persevere) for all subjects who have experienced a loss. It compares
those whose losses were one year or less in duration with those whose
losses exceeded one year in duration. Finally, it presents the figure

for the corrected chi square with 1 degree of freedom.

62

Table 4.3.--A Comparison of Duration of Loss (One Year or Less/Longer
than One Year) with Perseverance in School (Drop/Persevere).

 

No. of Cases with

 

 

No. Parental Loss for: Chi Square with
Perseverance of 1 Degree
Cases 1 Yr. or Longer than of Freedom
Less 1 Yr.
Drop 31* 12 19
Persevere 49* 8 41
Total 80 20 60 3.10**

 

*1 missing observation in each category, drop and persevere.

**Significant at the .05 level.

The results demonstrated that the duration of the loss of
parents is not independent of perseverance in school. As the chi
square figure of 3.10 is significant at the .05 level of confidence,

the null hypothesis is rejected.

Secondary Hypothesis 3

The number of losses of parents experienced by subjects is
statistically independent of perseverance in school.

Table 4.4 sets out data on perseverance in school (Drop/
Persevere) for all subjects. It compares those experiencing no loss,
those who experienced a single loss of parent, and those who experi-
enced two or more losses of parents. Finally, it presents the figure

for the raw chi square score with 2 degrees of freedom.

63

Table 4.4.--A Comparison of Loss of Parent (No Loss/One Loss/Two or
More Losses) with Perseverance in School (Drop/Persevere).

 

No. of Cases with:

 

 

No. of Chi Square with
Perseverance Cases No One Two Plus 0% 2:22:66
Loss Loss Losses
Drop 40 8 23 9
Persevere 102 52 42 8
Total 142 60 65 17 13.35*

 

*Significant at the .05 level.

The results demonstrate that the number of losses of parents
is not independent of perseverance in school. As the chi square
figure of 13.35 is significant at the .05 level of confidence, the null

hypothesis is rejected.

Secondary Hypothesis 4

 

The proportion of subjects suffering no loss of parents and
persevering in school is equal to the proportion of subjects
suffering loss of parents and persevering in school.

Table 4.5 sets out the data on perseverance in school (Drop/
Persevere) for subjects who have experienced loss of parents as com-
pared to subjects who have not. The table shows the t-value and the
degrees of freedom. Significance is established at the .05 level of

confidence if the observed level of t is equal to or less than the

critical value of t.

64

 

 

 

no.m- ow No— Nep Pouch
mam. Room. w mm om mmob oz
pmc. mmom. mm om mm mmoJ
Ecummgu we mmmcmmo oep cowamw>mo cop Logo; uzoaogo mcm>wmgma momma pcmgmm
saw: ma—m>uu um>gmmno cgcucmum .u a mo .oz mo mmOJ

"as; momma to .oz

 

.mmePEmm eunucH Ease mpummnsm xn Poogom cw mucmem>mmcma we :owpcogoga as» on umgmasou mm
mucosa; “mo; m>o= on: maumwnzm >5 Amcm>mmcma\aogov .oozum cw wucmcm>mmgmm we comucoaoga mghnu.m.e anmk

65

The observed level of t, -3.67, was less than the critical
value of t, -l.96, therefore, the null hypothesis is rejected. The
results demonstrate that the proportion of youths suffering no loss
of parents and persevering in school is significantly greater than
the pr0portion of youths suffering losses of parents and persevering

in school.

Secondary Hypothesis 5

 

The sex of subjects is statistically independent of per-
severance in school.

Table 4.6 sets out data on perseverance in school (Drop/
Persevere) for all subjects and compares them on the basis of sex.
The table shows, too, the corrected chi square with 1 degree of
freedom.

Table 4.6.--A Comparison of Sex of Subjects with Perseverance in
School (Drop/Persevere).

 

No. of Cases

 

 

, Chi Square with
Perseverance No. Of Found to Be. 1 Degree
Cases of Freedom
Male Female
Drop 4O 19 21
Persevere 102 57 45
Total 142 76 66 .51

 

The results demonstrate that the sex of the youths and perse-
verance in school are independent. As the chi square figure of .51 is
not significant at the .05 level of confidence, the null hypothesis is

not rejected.

66

Secondary Hypothesis 6

 

The sex of lost parents is statistically independent of
perseverance in school.

Table 4.7 sets out data on perseverance in school (drop/
persevere). It compares subjects who lost fathers with those who lost
mothers. The table shows, too, the corrected chi square with 1 degree
of freedom.

Table 4.7.--A Comparison of Sex of Lost Parent with Perseverance in
School (Drop/Persevere).

 

No. of Cases Experi- Chi Square with

 

 

Perseverance Eggegf encing LOSS Of: 1 Degree
Father Mother Of Freedom
Drop 32 14 18
Persevere 49* 15 34
Total 81 29 52 .33

 

*1 missing observation, persevere.

The results demonstrate that the sex of the lost parent is
independent of perseverance in school. As the chi square figure of
.33 is not significant at the .05 level of confidence, the null

hypothesis is not rejected.

Secondary Hygothesis 7

The socioeconomic background of the subjects' families is
statistically independent of perseverance in school.

Table 4.8 sets out data on perseverance in school (Drop/

Persevere) for all subjects and compares them on the basis of

67

socioeconomic background (Below Average/Average, and Above Average).
The table shows, too, the corrected chi square of 1 degree of freedom.
Table 4.8.--A Comparison of Socioeconomic Background (Below Average/

Average, and Above Average) of Subjects with Perseverance
in School (Drop/Persevere).

 

No. of Cases with Socio-

 

 

No of economic Background Chi Square with
Perseverance Cases 1 Degree
Below Average and of Freedom
Average Above Average
Dr0p 4O 23 17
Persevere 102 15 87
Total 142 38 104 24.71*

 

*Significant at the .05 level.

The results demonstrate that the socioeconomic background of
the subjects' families and perseverance in school are not independent.
As the chi square figure of 24.71 is significant at the .05 level of

confidence, the null hypothesis is rejected.

Secondary Hypothesis 8

 

The subject's ability in school is statistically independent
of perseverance in school.

Table 4.9 sets out data on perseverance in school (Drop/
Persevere) for all subjects and compares them on the basis of ability
(Below Average/Average/Above Average). The table shows, too, the raw

chi square with 2 degrees of freedom.

68

Table 4.9.--A Comparison of Ability (Below Average/Average/Above
Average) with Perseverance (Persevere/Drop).

 

No. of Cases with Ability:

 

 

Perseverance No. Of Chi1S3:;::ewith
Cases Aeglgge Average Aeggzge of Freedom
Drop 40 5 27 8
Persevere 102 5 50 47
Total 142 10 77 55 9.21*

 

*Significant at the .05 level.

The results demonstrate that the subject's ability and per-
severance in school are not independent. As the chi square figure of
9.21 is significant at the .05 level of confidence, the null hypothesis

is rejected.

Secondary Hypothesis 9

 

The cause of subjects' loss is statistically independent
of perseverance in school.

Table 4.10 sets out data on perseverance in school (Drop/
Persevere) for all subjects experiencing a loss of parent. Losses are
compared on the basis of cause of lost parents (Separation/Working
Mothers/Others). Finally, the table shows the raw chi square with

2 degrees of freedom.

69

Table 4.10.--A Comparison of Cause of Loss (Separation/Working Mother/
Other) with Perseverance in School (Drop/Persevere).

 

No. of Cases with Causes:

 

 

Perseverance No. Of . ChiZSg:;::e:ith
Cases Separation HEEHLCQ Other of Freedom
Drop 32 13 16 3
Persevere 49* 6 33 10
Total 81 19 49 13 9.08**

 

*1 missing observation, persevere.

**Significant at the .05 level.

The results demonstrate that the cause of the lost parent and
perseverance in school are not independent. As the chi square figure
of 9.08 is significant at the .05 level of confidence, the null

hypothesis is rejected.

Secondary Hypothesis lO

 

The proportion of perseverance in school for subjects

experiencing loss of parents caused by separation is equal

to the proportion of perseverance in school for subjects

experiencing no loss of parents.

Table 4.11 sets out data on perseverance in school (Drop/
Persevere) for subjects who have experienced a loss of parent caused
by separation as compared to subjects who have not lost a parent.

The table shows the t-value and the degrees of freedom. Significance
is established at the .05 level of confidence if the observed level

of t is equal to or greater than the critical value of t.

7O

 

mv.¢1 Fm mm ow Peach

l' U U

 

 

men. meow. m mm om wee; ez

awe. mmwm. mp 0 mp cewueweeem

seeeecw we meecmee oew :ewuew>eo cop Lego; pseeego eee>emwee memeu ewamwgeueegecu
new: espe>1u ee>eemeo eceeceum .g a we .ez xwwsew

”on: memeu we .ez
.mewwwsem peech
we mueenezm we weesem cw eecege>emwea we :ewugeeege may on emceeeeu me :ewuegeeem he eemzeu
mpcegea “we; e>ez on: mueenesm we Aegm>emgem\eeeov weecem cw eucece>emgee we :ewpgeeega ecwni.—w.e eweew

71

The observed level of t, -4.45, was greater than the critical
value of t, 1.96, therefore the null hypothesis is rejected. The
results demonstrates that the proportion of perseverance in school for
youths experiencing no loss of parents is significantly greater than
the proportion of perseverance in school for youths experiencing loss

of parents by separation.

Secondary Hypothesis ll

 

The proportion of perseverance in school for subjects
experiencing loss of parents caused by working mothers is
equal to the proportion of perseverance in school for
subjects experiencing no loss of parents.

Table 4.12 sets out data on perseverance in school (Drop/
Persevere) for subjects who have experienced loss of parents caused
by working mothers as compared to subjects who have not lost parents.
The table shows the t-value and the degrees of freedom. Significance
is established at the .05 level of confidence if the t is equal to or

greater than the critical value of t.

71

The observed level of t, -4.45, was greater than the critical
value of t, 1.96, therefore the null hypothesis is rejected. The
results demonstrates that the proportion of perseverance in school for
youths experiencing no loss of parents is significantly greater than
the proportion of perseverance in school for youths experiencing loss

of parents by separation.

Secondary Hypothesis ll

 

The pr0portion of perseverance in school for subjects
experiencing loss of parents caused by working mothers is
equal to the proportion of perseverance in school for
subjects experiencing no loss of parents.

Table 4.12 sets out data on perseverance in school (Drop/
Persevere) for subjects who have experienced loss of parents caused
by working mothers as compared to subjects who have not lost parents.
The table shows the t-value and the degrees of freedom. Significance
is established at the .05 level of confidence if the t is equal to or

greater than the critical value of t.

72

 

 

L¢.N 4N mm mo. Pope»
mem. “new. m mm om mmos oz
4N4. mmke. ew mm me nausea: mcwxeoz

Eeeeeeu we meecmeo oe_ :ewpew>en :erLeeege uzeeego mge>emgee memeu uwumwweueegesu
sow: a=_a>-o ea>eamao ceaucaom . "on: mange we .oz to .oz »_wsaa

 

.mewwwsem peeucw we muueneam
an —ee;em cw eecece>emcea we :ewuceeega egg on emceeseu we mgegaez mcwxgez ze eemzeu
mucecem ewe; e>ez e53 mueeweam Am Aege>emwea\eegov weenem cw eezege>emwea we :ewugeeewe egwiu.mp.¢ eweew

73

The observed level of the t, 2.47, was greater than the
critical level of the t, 1.96, therefore, the null hypothesis is
rejected. The results demonstrate that the proportion of perseverance
in school for youths experiencing no loss of parents is significantly
greater than the proportion of perseverance in school for youths
experiencing loss of parents due to working mothers.

In the application of the tests of the data for these twelve
hypotheses, nine were not accepted and three were accepted, in their
null form. There is a statistical relationship between the variable
persevere (drop/persevere) and each of the following variables: loss
of parents, duration of the loss of parents; the number of losses of
parents; proportion of subjects who experienced loss with the propor-
tion of subjects who had experienced no loss; socioeconomic background
of the subject's family; school ability of the subjects as perceived
by the subjects' mothers; causes of the loss of parents; separation
as the cause of the loss of parents; working mothers as the cause of
the loss of parents.

No statistical relationship was found, however, between the
variable persevere (drop/persevere) and variables: age of the subject
at the time of the loss; sex of the subjects; sex of the parents that
the subjects lost.

The findings of the research are summarized in Table 4.13.

74

 

eeeeeeee eew we.N eeee-w ewes we eeeee ww

 

:ewueweeem

eeeeeeee eew ee.e- emee-w wees we eeeee ew
pcewea

eeeeeeee _ me.e eeeeem wee ewes we eeeee e
maeeneam

eeeeeeee N _N.e eeeeem wee we wewwwee e
eczewmxeem zwwsew

eepeenem _ ww.em eweaem wee ewseceeeewuem w
ucewea

eeeeeeee eez _ mm. eeeeem wee ewes we xem e

eeeeeeee eez _ we. eeeeem wee eeeeeem we xem e

eeeeeeee eew we.m- emee-w eeweweeewe\eeee e

eeueewem N mm.m_ ewezem wgu memmee we weesaz m
mmee

eeeeeeee w e_.m eeeeem wee we eeweeeee N
mucewee we

eeeeeeee eez _ . we. eeeeem wee eeee we ee< w
mucewea we

eeweenem _ wo.o_ ewezem wee mmee e2\mme4 zgeswwe
peenem wez eeeeeww ezwe>uw we “mew eweewwe> memegpeexz

we peewem we meewmeo ewezem wzu weewumwuepm

.meeweeww eeweeeem wee we eweew <--.m_.e eweew

CHAPTER V

SUMMARY AND CONCLUSIONS

Summary
Man has yet to plumb the depth or measure the breadth of the

parent-child relationship. In the general field of parent-child
relationships, this research aimed to explore the relationship, if any,
between the loss of parents and the child's later perseverance in
school. Causes of childhood loss of parents are grouped here as death,
divorce, working mothers, and other.

This study is designed to test the assumption that a relation-
ship exists between the loss of parents by children and their
perseverance in school. Other common assumptions tested included the
relationships between school perseverance and the following family-
based variables: the age of the child at the time of the loss; the
cause of the loss; the sex of the child as well as the sex of the lost
parent; the duration of the loss; the number of losses; the socio-
economic background of the family; the school ability of the child;
and the nature of the substitute parents and substitute surroundings.
These assumptions emerged from a review of the literature.

The sex of child, sex of parent, age of child at loss,
duration of loss, number of losses, cause of loss, mother's perception

of the learning ability of the child, socioeconomic background of

75

76

family, and substitute parents and surroundings were all selected as

variables to be tested against school perseverance. The following

research hypotheses were developed.

Primary Hypothesis

The relative intactness of the family and subjects' persever-

ance in school are statistically independent.

Secondary Hypotheses

 

l.

The age of subjects at the time of loss is statistically
independent of perseverance in school.

The duration of loss of parents in statistically independent
of perseverance in school.

The number of losses of parents experienced by subjects is
statistically independent of perseverance in school.

The proportion of subjects suffering no loss of parents and
persevering in school is equal to the proportion of subjects
suffering loss of parents and persevering in school.

The sex of subjects is statistically independent of persever-
ance in school.

The sex of lost parents is statistically independent of
perseverance in school.

The socioeconomic background of subjects' families is
statistically independent of perseverance in school.

The subject's ability in school is statistically independent
of perseverance in school.

The cause of subjects' loss is statistically independent of

perseverance in school.

77

10. The proportion of perseverance in school for subjects experi-
encing loss of parents caused by separation is equal to the
proportion of perseverance in school for subjects experiencing
no loss of parents.

11. The pr0portion of perseverance in school for subjects experi-
encing loss of parents caused by working mothers is equal to
the proportion of perseverance in school for subjects experi-

encing no loss of parents.

An instrument was devised to collect the data to be tested.
An informal open-ended questionnaire was administered in the home of
each subject. Data were collected from mothers of subjects in the
personal interviews.

A single school district was chosen and the p0pulation was
defined as youths between the ages of 16 and 21 who were attending or
had once attended the schools in the district and who had one or both
parents still living in the district. The sample consisted of 142
boys and girls. Streets and roads within the district were randomly
selected from a hat and every home on the selected street was visited.
Homes were revisited until all subjects were identified on each
street. No subjects were lost.

As the data collected were categorical in nature and because
some hypotheses required proportional data, the chi square test of
independence and the t-test for proportion were selected as testing
instruments. Significance was set at the .05 level of confidence.

Each of the research hypotheses was tested and the results

are summarized below.

78

Discussion

 

Following are a list of the research hypotheses with a

discussion of the findings.

Primary Hypothesis

 

The relative intactness of the family and the subjects'
perseverance in school are statistically independent.

The literature seems to indicate that most serious mental
illnesses are associated with loss of parents, particularly with the
loss of mother. While working mothers seem to have some negative
effects on children, the effect has not been noted on all children.

In the case of father's absence, not all children are noted
to be affected, but the effects were reported negative. The death of
one or both parents has been found to have a negative effect on a
majority of children. Children suffer the loss of parents for varying
lengths of time. There seems little agreement on how the length of the
loss affects children, except that longer periods may have a more
damaging effect than shorter periods. Similarly, research does not
demonstrate a definite age or age group at which children are most
susceptible or the sex of the child most likely to be affected. Most
research indicates the child is more sensitive to the loss of mother,
and more recent research demonstrates the importance of father's
presence.

This research sought to discover whether there is a relation-
ship between childhood loss of parents and later perseverance in
school. The results of this study indicate that for this population
there is such a relationship. Loss of parents and perseverance in

school are found to be statistically significant at the .05 level of

79

 

 

empeewem ovw we.~ weep-» wee; we emzeu ww
eewpeweeem

eeueenem oew me.e- umewuw mmee we emaeu ow
pcewee

eeeeeeem w Ne.e eweeem wee wees we emeee e
mpeenezm

eeeeeeem N wN.e eeeeem wee we eewwwee e
eczewaxeem xwwsew

eepeenem w ww.e~ ewezem wee uwseceeeeweem w
ucewee

eeeeeeee eez w mm. eweeem wee ewes we xem e

eeueenem uez w wm. ewezem wsu newnesm we xem m

eeeeeeee ee_ we.m- eeee-w eeweweeewe\eeee e

eeweenem N mm.mw ewezem wcu memmee we weeszz m
wee;

eeeeeeee _ ew.m eweeem wee we eeweewee N
mwcewee we

empeemem uez _ we. ewezem wzu wee; we em< _
mucewee we

eeueewem — wo.ow ewezcm wgo mmee ez\mme4 xweewwe
weenem pez Eeeeeww eawe>1w we peep 11
we peewem we meewmeo ewezem wgu weuwpmwueum ewaewwe> memesuee»:

 

.mmcwecww seweemem use we wweesam <1-._.m eweew

80

confidence. Only 20 percent of the sampled dropouts came from intact
families while more than 50 percent of the sampled perseverers came

from intact families.

Secondary Hygptheses

 

Hypothesis 1: The age of subjects at the time of loss is
statistically independent of perseverance in
school.

 

Research does not demonstrate a definite age or age group at
which children are most susceptible to loss of parents. This study
found the variable, age at the time of loss, statistically independent
of variable perseverance at the .05 level of confidence. Twenty-five
percent of the sampled dropouts and 26.5 percent of the perseverers
who suffered losses of parents, suffered the losses before starting
school. Similarly, the remaining 75 percent of the dropouts and the
73.5 percent of the perseverers who suffered losses, did so after
starting school. Age at the time of loss and perseverance are
statistically independent.

Hypothesis 2: The duration of loss of parents is statis-
tically independent of perseverance in school.

 

There is little agreement in the literature on the effects
of different durations of parental loss on children, except that
longer periods of time may have a more damaging effect than shorter
periods of time. Duration of loss of parents and perseverance in
school have a statistically significant relationship. Both categories
of suffering losses suffered more losses for longer periods of time

than they did for shorter periods of time.

81

Hygpthesis 3: The number of losses of parents experienced by
subjects is statistically independent of per-
severance in school.

Most research indicates that numerous losses of parents are
more likely to have a negative effect on children than a single loss,
providing the duration of the single loss is not longer or the loss
is not at an earlier age. This research found the number of losses and
later perseverance to be dependent variables. Although only 20 percent
of the dropouts suffered no loss, 51 percent of the perseverers sus-
tained no loss. Where 41 percent of the perseverers suffered a single
loss, the dropouts sustained 57 percent in that category. Similarly,
less than 8 percent of the perseverers suffered more than one loss,
while 22 percent of the dropouts did so.

Hypothesis 4: The proportion of subjects suffering no loss of
parents and persevering in school is equal to
the proportion of subjects suffering loss of
parents and persevering in school.

This study found the proportion of youths suffering no loss

of parents and persevering in school exceeds the proportion of youths
experiencing loss of parents and persevering in school at the .05
level of confidence. While nearly 87 percent of the youths who
suffered no loss of parents were found to persevere in school, only
61 percent of the youths who suffered a loss did so.

Hypothesis 5: The sex of subjects is statistically independent
of perseverance in school.

Many studies find the sex of the child to be significantly

related to developmental problems, while others do not. This study

did not find the variables, sex of the child and perseverance,

 

82

dependent. Among the dropouts, 48 percent were males while 52 percent
were females. The perseverers were similarly divided with 56 percent
males and 44 percent females.

Hypothesis 6: The sex of lost parents is statistically
independent of perseverance in school.

 

Most studies find the child is more sensitive to the loss of
mother. Recent research demonstrates the importance of father's
presence. This research found the sex of the parent to be independent
of perseverance in school. Although 44 percent of the dropouts' loss
was due to absence of fathers, 56 percent was caused by loss of
mothers. Perseverers, who suffer a loss, were even more likely to
have lost a mother (69%) than to have lost a father (31%).

Hypothesis 7: The socioeconomic background of the subjects'

families is statistically independent of
perseverance in school.

 

Most research finds a significant relationship between the
socioeconomic background of the family and perseverance in school.
Other studies have found the relationship disappeared when student
status in the school was controlled.

This research found family income and perseverance in school
to be dependent. While 57 percent of the dropouts were found to come
from families with lower than average incomes, only 15 percent of the
perseverers fell into this category. Where 42.5 percent of the
dropouts came from families with average or above average incomes,

more than twice that percent, 85 percent, of the perseverers did.

83

Hypothesis 8: The subject's ability in school is statistically

independent of perseverance in school.

This research found ability in school, as perceived by the
subject's mother, and perseverance in school to be dependent at the
.05 level of significance. Perseverers formed 50 percent of the below
average group, 65 percent of the average group, and nearly 86 percent
of the above average group.

Hypothesis 9: The cause of subjects' loss is statistically

independent of perseverance in school.

Other research seems to indicate that most serious mental
illnesses are associated with loss of parents. The causes considered
by most investigators are: death, divorce, working mothers, and other.

Students who suffer losses caused by separation are nearly
three and one-half times as likely to be a dropout (40%) as they are
to be a perseverer (12.8%). In considering all subjects, perseverers
make up 87 percent of the no loss category, 31 percent of the
separation group, and 67 percent of the working mothers category.

Hypothesis TO: The proportion of perseverance in school for

subjects experiencing loss of parents caused
by separation is equal to the proportion of

perseverance in school for subjects experi-
encing no loss of parents.

 

Most research indicates that the child is sensitive to parental
discord. Some investigators have found parental discord to be more
damaging to the child than an actual separation. This research found
the proportion of perseverance in school for subjects experiencing
no loss of parents exceeded the proportion of perseverance in school
for subjects experiencing losses of parents caused by separation at

the .05 level of confidence. While nearly 87 percent of the subjects

 

84

who suffered no loss of parents were found to persevere in school,
only 31 percent of the youths suffering a divorce or desertion

similarly persevered.

Hypothesis 11: The proportion of perseverance in school for
subjects experiencing loss of parents caused
by working mothers is equal to the proportion
of perseverance in school for subjects experi-
encing no loss of parents.

 

Investigators are divided on the effect of mothers working
outside the home. Some research indicates there is no effect while
other investigators have found positive or negative effects. This
research found the proportion of perseverance in school for subjects
suffering no loss of parents exceeded the proportion of perseverance
in school for subjects experiencing loss of parents caused by mothers
working at the .05 level of confidence. While 87 percent of the
children who suffered no loss of parents persevered in school, a

lesser number, 67 percent, of those with working mothers did so.

Summary of Discussion

 

Of the twelve research hypotheses, nine were rejected and
three were not rejected in their null form. The following variables
were found to be significantly related to the variable perseverance
(drop/persevere): relative intactness of the subjects' families
(loss/no loss); duration of the loss of parents; number of losses of
parents; proportion of loss of parents; socioeconomic background of
the subjects' families; ability of the subjects in school as perceived
by the subjects' mothers; cause of the loss of parents; separation as
the cause of the loss of parents; and working mothers as the cause of

loss of parents.

 

85

Statistically significant relationships were found between the
variables described above regardless of the age of the subject at the
time of the loss, the sex of the subject, and the sex of the parent
which the subject lost.

Among the subjects studied, dropouts were four times as likely
to have suffered the loss of parents as they were to have come from
intact families. The proportion of intact families for persevering
youths was significantly greater than the proportion of youths
suffering losses of parents and persevering in school.

The number of losses of parents experienced by subjects was
found to be related to perseverance in school. The more losses of
parents subjects suffered, the more likely they were to be found in
the dropout category. The fewer losses of parents subjects experienced,
the more likely they were to be identified as perseverers.

A statistically significant relationship was found between
cause of the subjects' loss of parents and perseverance in school.

Of the three causes measured, separation, working mother and other,
separation was found to be the most significant. Although a
statistically significant relationship was found between the duration
of the subjects' losses of parents and perseverance in school, the
relationship is the reverse of what might be expected. Perseverers
experienced longer periods of loss of parents than did dropouts.

Like other studies, this research found the socioeconomic
background of the subjects' families and ability of the subjects in
school both related to perseverance in school. Unlike other studies,
this study measured the mothers' perception of their children's

ability in school. Dropouts were less likely than perseverers to come

86

from families with average and above average incomes, and their ability
in school was more likely to be perceived by their mothers as below

average.

Implications

 

This research is the result of the study of a single school
district. In this single community, a web of statistical relationships
developed. Although the purpose of this study was to determine the
relationship between losses of parents by school age children and
perseverance in school, it was not the intention of the study to
assign a causal relationship to the findings. The observed relation-
ship, however, leads to some speculation.

If we observe a relationship between instability of the family
and dropping out of school, efforts may be justified to find ways of
reducing the number of losses of parents sustained by children.
Efforts may be justified to develop methods of strengthening the
family.

As childhood loss of parents is an increasing phenomenon, it
becomes increasingly desirable to conduct research measuring its
effect on children. Parents can be made more aware of this effect on
their children. From this study in one district, it would appear that
those families which need the most help in understanding how loss of
parents can affect their children may be families with average and
below average income, families in which mothers perceive the ability
of their children in school as below average, and families in which
losses are repeated or extended over longer periods of time. These

families may need help in understanding the effect of such losses

87

regardless of the sex of the child, the age of the child at the time of

the loss, and the sex of the parent which the child loses.

Recommendations

 

The experience of interviewing mothers of 142 subjects spread
throughout the district impressed upon this investigator that parents
seldom knowingly set out to damage the emotional growth of their
children. Most of the parents interviewed seemed willing to make
sacrifices for the good of their children. Many mothers appear to be
at a loss to understand their child's failure to continue through
school.

Based on these assumptions, this investigator makes the
following recommendations. Further research is needed in the effects
of separation caused by marital discord on children involved.
Research must be designed to hold constant such variables as socio-
economic background of the family and ability or perceived ability.

Research is recommended in which the proportion of separations
suffered by children in the following categories are measured against
the proportions of separations sustained in the general population:
children who are screened into special education programs; who
exhibit antisocial behavior; who become addicted to drugs; who obtain
criminal records; who are diagnosed as autistic; who are committed to
mental institutions; who commit suicide; and who conceive their own
children out of wedlock.

It is recommended, too, that research be conducted to isolate
the most important and most common causes of marital discord and the

most effective techniques in minimizing and managing causes of marital

88

discord. It is not enough to conduct the research and publish the
findings. Methods must be devised to make this knowledge available
and useful to parents and prospective parents.

Efforts may be justified in developing parenthood training,
training aimed at teaching children how to recognize those elements
in their own personalities and in the personalities of others, which
will increase their ability to recognize and select compatible mates.
Parenthood training may aim at recognizing and minimizing causes of
marital discord and its effects on children. Finally, counseling
services with similar goals and based on similar research may be

developed as a continuing extension of parenthood training.

REFERENCES

REFERENCES

Ahmad, Farrukh Z. "A Study of the Relationship Between Test Intelli-
gence and Delinquency.“ Psychologia: An International Journal
of Psychology in the Orient 9 (l966):24-26.

 

Annis, Helen M., and Watson, Carol. "Drug Use and School Dropout: A
Longitudinal Study." Canadian Counselor 9 (1975):155-62.
In Psychological Abstracts (July-December l976):9798.

 

 

Antonovsky, Helen F. "A Contribution to Research in the Area of the
Mother-Child Relationship." Child Development 30 (1959):37-51.

 

Ausubel, David P. Theory and Problems of Child Development. New York:
Cruen 8 Stratton, 1958, p. 259.

 

Axline, Virginia M. Dibs In Search of Self. New York: Ballentine
Books, 1964, p. 220.

 

Banducci, Raymond. "The Effect of Mothers' Employment on the Achieve-
ment, Aspirations and Expectations of Their Child." Personnel
and Guidance Journal 64 (l964):263.

Bane, May J. "Marital Disruption and the Lives of Children." Journal
of Social Issues 32 (1976):TO3-17.

 

Barrera, Eduardo. "School Dropouts in Hidalgo County, Texas: History
of Remedial Programs and Assessment of Current Dropout
Situation." Dissertation Abstract International (1976):
2700-A.

Barry, H., Jr. "Significance of Mental Bereavement Before Age Eight
in Psychiatric Patients." Archives of Neurology and
Peychiatry 62 (l949):630-37.

 

Barry, H., Jr., and Lindemann, E. "Critical Ages for Maternal
Bereavement in Psychoneuroses." Eeychosomatic Medicine 22
(l960):l66-81.

v’ Beaird, R. E. S. "Self-Concept as Related to Adolescent School Drop-
outs." Doctor's Thesis, University of Nebraska Teacher's
College, Lincoln, 1964. In_Journal of Educational Research 65
Tseng, M. S. (August-Septemberl972):462.

89

90

y, Berry, Gordon L. “Self-Concept and Need Factors of Inner-City School
Adolescents and Dropouts.“ Child Study Journal 4 (l974):21-31.

 

Bettelheim, Bruno. The Emotnyortress: Infantile Autism and the Birth
of the Self. New YoFk: The Free Press, 1976, pp. 425 8 427.

Birchnell, J. "Early Parental Death and Mental Illness." British
Journal of Psychology 116 (l970b):28l-88.

\/.Blake, Gerald F. "School Dropouts: A Study of the Antecedents and
Consequences of Dropping Out of School." Dissertation
Abstracts International 34 (1974):788l-82.

 

Bledsoe, J. C. "An Investigation of Six Correlates of Student With-
drawal from High School." The Journal of Educational
Research 53 (September 1959):3-6.

Blood, Robert, and Wolfe, Donald. Husbands and Wives. New York:
Free Press of Glencoe, 1960. In American Sociological
Review, Bowerman and Elder, 1964, p. 571.

Boshier, Roger, and Thom, Elizabeth. "Do Conservative Parental Nature
Cause Conservative Children?" Auckland, New Zealand:
University of Auckland, 1974. In Eeychological Abstracts 51
(l964):108-TO.

 

Bowerman, Charles E., and Elder, Glenn H., Jr. "Variations in
Adolescent Perception of Family Structure." American
Sociological Review (l964):567.

 

Bowlby, John. Attachment. New York: Basic Books Inc., 1969,
Preface xi-xviii.

 

. "Childhood Mourning and Its Implications for Psychiatry."
In Childhood Beychopathology: An Anthology of Basic Reading,
pp. 265-89. Edited by 5.71:7Harrison and John F. McDermott.
New York: University Press, Inc., 1972.

. Separation: Anxiety and Anger. New York: Basic Books Inc.,
1973, p. 456.

V/Brookover, Wilbur B. "Self-Concept of Ability and School Achievement,
II." U.S. Office of Education, Cooperative Research Project,
No. 1636. East Lansing, Michigan: Michigan State UniverSity,
Bureau of Educational Research Services, 1965, p. 51.

 

Brown, F. "Bereavement and Lack of Parent in Childhood." In
Foundation of Child Peychiatry. Edited by E. Miller. New
York: Pergamon Press, 1968.

Brown, F., and Epps, Phyllis. "A Childhood Bereavement and Subsequent
Crime." British Journal of Psychiatry 112 (l966):lO43-48.

91

Brown, F. "Depression and Childhood Bereavement." Journal of Mental
Science 107 (1961):754-77.

 

Brown, S. W. A Comparative Study of Mental Employment. Mississippi
State University: University Microfilms, 70-8610. In Working
Mothers, Hoffman and Nye, Washington: Jossey-Bass Publishers,

I974, p. 162.

Buckley, William F. "Points of View." Detroit Free Press, 25 March
1977, p. ll-A.

Bunch, J.; Barraclough, B; Nelson; and Sainsbury. "Early Parental
Bereavement and Suicide." Social ngchiatry 6 (l97l):200-202.

Burns, Donald William. "An Investigation of Selected Factors that
May Influence Pupil Achievement in Michigan Schools." Ph.D.
dissertation, Michigan State University, 1972.

,rCain, Alexander M. "A Study of Relationships Between Selected Factors
and the School Achievement of Mexican-American Children."
Ph.D. dissertation, Michigan State University, 1970.

Canning, Rebecca. "School Experiences of Foster Children." Child
Welfare 53 (l974):582-87.

Carrier, Robert H. "Peer Group Violence as a Factor in the High
School Dropout Problem." In Dissertation Abstracts Inter-
national 34 (October 1973):l471-72.

Carter, Kenneth G. "A Selected Study of Variables Between Juveniles
Referred to the Juvenile Court Only Once and Juveniles
Referred Three Times or More." In Dissertation Abstracts
International 33 (December l972):2760.

\/Cervantes, Lucius F. The Drgpout, Causes and Cures. Ann Arbor: The
University of Michigan Press, 1965. pp. 96-101 8 199.

Clark-Stewart, K., and Alison. “Interactions Between Mothers and
Their Young Children: Characteristics and Consequences."
Monogreph of the Sociegy and Research in Child Development
38 (1973):l-108. In Peyphdlogical Abstracts Tl975):53.

v/Costin, Lela B. Child Welfare: Policies and Practices. New York:
McGraw Hill Book Company, 1972, pp. 18-20.

\/ Cundick, Bert; Gottfredson, Douglas, K.; and Wilson, Linda. "Changes
in Scholastic Achievement and Intelligence of Indian Children
Enrolled in Foster Placement Programs." Developmental
Psychology 10 (l974):815-20.

 

\// Davis, Donald A. "An Experimental Study of Potential Dropouts." IHe_
Personnel and Guidance Journal 40 (May l962):799-802.

92

Derdeyn, Andre P. "A Consideration of Legal Issues in Child Custody
Contents." Archives of General Psychiatry 33 (February 1976):
165-71.

 

Despert, Louise P. _CHildren of Divorce. Garden City, New York:
Doubleday and 66mpany,il953, p. 282.

 

The Relationship of Emotional Adjustment to Intellectual
Functionieg, Provincetown, Massachusetts: The OburnET'Press,
1946, p. 282.

 

Deutsch, H. “A Two Year Old Boy's First Love Comes to Grief."
‘ Qynamic Psychopatholggy in Childhood. Edited by L. Jessner
and E. Pavenstedt, l9 8. *InTAttaCHment, Bowlby, John,
New York: Basic Books, Inc., 1969, p. 30.

I'Deutsch, Martin. The Disadvantaged Child. New York: Basic Books,
Inc., 1967, pp. 103-6.

 

Deutsch, Martin, and Brown, Bert. "Social Influence in Negro-White
Intelligence." Journal of Social Issues 20 (l964):24-35.

 

Duncan, B. "Family Factors and School Dropouts: 1920-1960." U.S.
Department of Health, Education, and Welfare, Office of
Education, Cooperative Project No. 2258. Ann Arbor, Michigan:
University of Michigan, 1965, p. 229.

Eisenberg, L. "The Fathers of Autistic Children." American Journal
of Orthopsychiatry 27 (1957):715-24.

 

 

Ekstein, R. Children of Time and Space, of Action and Impulse. New
York: Appleton-Century-Croft. In Infantile Autism; Pro-
ceedings of the Indiana University COTquuium. Edited by
Churchill et al. Springfield: Charles C. Thomas Publishing
Co., 1966, pp. 163-64.

 

 

 

Erlenmeyer, and Kimling. "Genetics and Intelligence: A Review."
Science 142 (l963):l477-79.

’ Fifield, M. G. "The Self-Concept as an Identifying Factor of School
Dropouts." Ph.D. dissertation, Washington State University.
In Journal of Educational Research 65 (September-August 1963):
462.

 

Freud, S. The Origins of Psychoanalysis, Letters to Wilhelm Fliess,
Drafts, and'Notes. TLetter No. 69. New York: Basic Books,
5934’ pp- 2 55-58.

 

Fuller, Gerald B., and Douglas, Friedrich. "Rural High School Drop-
outs: A Descriptive Analysis." Perceptual and Motor Skills
35 (August 1972):l95-201.

93

Glueck, Sheldon, and Glueck, Eleanor. "Working Mothers and Delin-
quency." Mental Hygiene 41 (1957):349.

 

Glueck, Sheldon, and Glueck, Eleanor. Unraveling Juvenile Delin-

uenc . Cambridge, Massachusetts: Harvard University Press,
1951. PP. 90-91.

V/Greaney, Vincent. "A Comparison of Secondary School Entrants, Vocation
School Entrants and Terminal Leaves." Irish Journal of Educa-
tion 7 (l973):79-101. In Eyschological AbstractsITJuly-Decem-
ber l976):6639.

 

Greer, S. "Parental Loss and Attempted Suicide." British Journal of
Psychiatry 112 (l966):465-70.

 

l/Harding, Kenneth L. "A Comparative Study of Caucasian Male High
School Students Who Stay in School and Those Who Drop Out."
Dissertation Abstracts 27 (l967):2884.

Harrison, S. I., and McDermott. John F. Childhood ngchopathology:
An Antholggy_of Basic Reading. New York: University Press,
1972, PP. 273-87.

Hilgard, J. R.; Newman, M. F.; and Fisk, F. "Strength of Adult Ego
Following Childhood Bereavement." American Journal of Ortho-
psychiatry 3O (1960):788-98.

 

Hoffman, Lois, and Nye, Ivan. Working_Mothers. Washington: Jossey-
Bass Publishers, 1974, pp. 164-65.

Hoffman, Lois W. Emgloyed Mothers in America. Chicago: Rand McNally
and Co., 1963, p. 95.

\//Holzner, Ann S., and Ding, L. K. "White Dragon Pearls in Hong Kong."
v’ International Journal of the Addictions 8 (l973):253-63.

Honzik, Marjorie P. "Environmental Correlates of Mental Growth:
Predictions from the Family Setting of 21 Months." Child
Development 38 (l967):337-64. .

Janssen, Calvin W. "Comparative Creative Scores of Lower Social-
Economic Dropouts and Non-Dropouts." Dissertation Abstracts
27 (l965):966.

Keene, Richard. "Validity of School Motivation Analysis Test and
Prediction of Dropout Status of Secondary Students."
Dissertation Abstracts International (1976):2802-A.

xj/ Kilpatric, Ronald N. "A Comparison of Black Male High School Students
Who Graduated with Those Who Dropped Out." Dissertation
Abstracts International 34 (November l973):2218-19.

94

b/Kowalski, Casimir, Jr., and Cangemi, Joseph P. "High School Dropouts:
Last Resource." College Student Journal 8 (November-December
l974):71-74.

 

Kutcher, Louis William. "Part Time Dropout: Family and School
Achievement of High School Absentee in Low-Income Elementary
Schools." Ph.D. Dissertation, University of Minnesota, 1972.

Landers, Ann. ”Unsigned Survey Answers Differ." Detroit Free Press,
Monday, 28 March 1977, p. 2-B.

 

Landy, R.; Rosenberg, B. G.; and Sutton-Smith, B. "The Effect of
Limited Fathers' Absence of Cognitive Deve10pment.“ Child
Development 40 (l969):94l-44.

 

LaRoch, Shirley S. "The Role of the Stepfather in the Family." Ph.D.
Dissertation, University of New Mexico, 1933.

Lee, Beatrice, and Crump. “School Dropouts." NEA Research Bulletin
(1963). In "The Relationship Between Dropouts and the Graded
School." Phi Delta Kappan 46 (November 1964):129-33.

 

 

Loken, Joel. "Patterns of Drug Misuse and School Dropout." Canadian
Counselor 7 (October l973):279-82. In Educational Psyehology

(July-December l974):ll-l9.

Mahler, M. "Thoughts About Development and Individualization."
PsychoanaLytic Study of the Child 18 (l963):307-24.

 

 

Mason, Cathryn T. '"The Effects of Counseling on Self-Concept and
Academic Achievement of Disadvantaged Dropouts." Dissertation

 

Abstracts International 33 (December l972):2718.

 

Mason, Rela Geffen. "The Relationship, The Nuclear Family Structure
and Female Achievement." Ph.D. Dissertation, University of
Florida, 1972.

\/ Mathews, C. V., and Bowman, Peter. "The Motivations of Youth in
Leaving School." In The Dropout Causes and_§ures, pp. 99-102.
Edited by Lucious F. Cervantes. Ann Arbor: The University of
Michigan Press, 1960.

Mead, Margaret. Family. New York: McMillan Co., 1971, pp. 18-49.

New Lives for Old. New York: William Morrow and Co., 1956,
p. 3.

Munro, A. "Childhood Parent--Loss in a Psychiatrically Normal Popula-
tion." British Journal of Preventive Social Medicine 19
(1965):69-70.

O'Neill, John H. "High School Dropouts." Education 84 (November
1963):156-59.

\J

95

Peterson, Bill. The Last Deys of Man. New York: Warner Books, 1977,
pp. 81-90.

 

Roy, P. “Adolescent Roles: Rural-Urban Differences." In Working
Mothers, p. 158. Edited by Nye and Hoffman. Washington:
Jossey-Bass Publishers, 1974.

Sexton, Patricia Cay. Education and Income: Inequalities of Oppor-
tunity in Our Public Schodls. New York: Viking Press, 1961,
p. 260.

 

 

Skinner, B. F. Walden Two. New York: The MacMillan Company, 1948,
p. 260.

 

Spiro, M. E. "Children of the Kibbutz." Cambridge, Massachusetts:
Harvard University Press, 1969. In Attachment, p. 30.
Edited by John Bowlby. New York: Basic Books, Inc., 1969.

 

Spitz, R. A., and Wolfe, K. M. "The Smiling Response: A Contribution
to the Ontogenesis of Social Relations." Genetic Psycho-
logjcal Monograms 34 (l946):57-125. In Attachment, pp. 29-31.
Ed1ted by John Bowlby. New York: Basic Books, Inc.

 

 

 

Sutton-Smith, 8.; Rosenberg, B. 6.; and Landy, Frank. "Fathers'
Absence on Cognitive Development." Child Development
Abstracts and Bibliography 43 (1969):407.

 

 

Toffler, Alvin. Future Shock. New York: Bantam Books, 1971,
pp. 241-56.

 

Toynbee, Arnold. "Why I Dislike Western Civilization." The New York
Times Magazine, May 10, 1964, pp. 15-23.

 

Tseng, M. S. "Comparisons of Selected Familial, Personality, and
Vocational Variables of High School Students and Dropouts."
Journal of Educational Research 65 (July l972):462-66.
IfiiEducationaTiAbstracts (January-June 1973):3337.

 

 

Tseng, M. S., and Thompson, P. L. "Differences Between Adolescents
Who Seek Counseling and Those Who Do Not." Personnel and
Guidance Journal 47 (December 1968):333-36.

 

U.S. Bureau gf Labor StatistiCs. Employment and Earnings 24 (February
1977 :28.

 

U.S. Bureau of the Census. Statistical Abstracts of United States.
Washington, D.C., 97th edition, 1976.

U.S. Department of Commerce, Bureau of Census. A Statistical Portrait
of Women in the U.S.A., 1976, p. 29.

Statistical Abstracts of United States, 1976, pp. 24-42.

 

96

V/IU.S. Department of Health, Education and Welfare. "A New Look at
School Dropouts." Health, Education and Welfare Indicators,
April 1974.

 

Monthly Vital Statistics, 1975, p. 4.

Monthly Vital Statistics Final Report: Live Births, 1975,
p. l.

. Monthly Vital Statistics Report: Births, Marriages, Divorces
and Deaths, 1976, p.71.

 

Monthly Vital Statistics Report: Summary Report, Final
Divorce and Statistics, 1972, p. l.

 

Vital Statistics of U.S., 1972, p. 170.

 

. Vital Statistics of U.S.: Marriage and Divorce, 1972,
III: I-5 and II-5.

U.S. Department of Labor. Bureau of Labor Statistics: Monthly Labor
Review, January 1977, p. 49.

U.S. Department of Labor Statistics. Employment and Earnings,
February 1972, p. 28.

 

1976 Handbook on Women Workers, p. 3.
Women's Bureau, Who Are the Workipngomen, 1972, pp. 3-6.

»//Van Dyke, L. H., and Hoyt, K. B. "The Dropout Problem in Iowa
Schools." U.S. Department of Health, Education and Welfare,
and Welfare Office of Education, Cooperative Research Prgject
No. 160. Iowa City: University of Iowa and Iowa State
Department of Public Instruction, 1958, p. 88.

 

V/ WallerStein, Judith S., and Kelly, Joan B. "The Effects ofParents'
Divorce: Experiences of the Child in Later Latency.“
American Journal of Orthopeychiatry 46 (April l976):256-59.

Wexler, Suzan Stanhope. The Story of Sandy, New York: Bobbs-Merrill
Co., Inc., 1955, p. 174.

Wiene, Jacob G. "A Study of Family and Peer Relationships as They
Affect Identity and Interdependence of Five Year Olds,
Children in Group Day Care Centers." Ph.D. Dissertation,
New York University, 1973.

v/ Wiener, Jerry M., and Egan, James H. "Heroin Addiction in an Adoles-
cent Population." Journal of American Academy of Child
Psyehiatry 12 (January 1973).

 

97

Williams, P. V. "School Dropouts." NEA Journal 52 (February 1963):
1012.

Wilson, Maurice Valentino. "A Comparative Study of the Dropout Rate
Between Students in the North Eastern High School Accelerated
Learning Achievement Program and Those Eligible for the
Program.“ Dissertation Abstracts International (l976):2618-A.

Winburn, G. Michael, and Hays, J. Ray. "Dropouts: A Study of Drug
Use." Journal of Drug Education 4 (Summer l974):249-54.

 

Yudin, Lee W.; Ring, Stephen 1; Nowakiwska, Myra; and Heinemann,
Shirley H. "School Dropouts and College Bound: Study in
Contrast." Journal of Educational Research 67 (October 1973):

87-93.

MICHIGAN STATE UNIV. LIBRARIES
IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII
31293102308933