THE EPIDEMIOLOGY OF COMMON FEARS AND DIFFUSE PHOBIAS:
A SOCIODEMOGRAPHIC ANALYSIS
Billy L. Williams
A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL
OF THE UNIVERSITY OF FLORIDA
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE
DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
This dissertation is dedicated to the graduate students, the
staff, and the faculty of the Department of Sociology, the
University of Florida, September, 1971 through June, 1975.
With whom does one begin to thank all those persons who helped
reach a goal he once thought unobtainable? Achieving a Doctor of
Philosophy degree is not a lone venture; millions of people are involved.
At the most broad level are those taxpayers who support the graduate
schools and the administrators who guide them. And, of course, there
is the faculty, charged with the responsibility of providing the best
education possible for its students, and, at the same time, attempting
to socialize its students into possessing those values, attitudes, and
work habits which will make him an asset to the profession. The anxieties
of attaining a doctorate are made bearable only through the formation
of close friendships with fellow graduate students; without them, all
candidates for postgraduate degrees would surely lose their sanity. And,
finally, there are the undergraduate students who often serve as guinea
pigs for frightened graduate students placed in charge of a classroom for
the first time. As important and essential as these people are, there
are always a few who play especially important roles, and I would like
to acknowledge at this time their contributions to my education and to
the successful completion of this dissertation.
First, my appreciation is extended to Charles E. Holzer for his
invaluable assistance in programming the computer and retrieving the data.
Without the services he provided, this dissertation would literally
have been impossible to write. The research was funded in part by
NIMH Grant 15900-05, and all statistical analyses were performed through
the facilities of the Northeast Florida Regional Data Center located
on the campus of the University of Florida, Gainesville, Florida.
My parents, Pauline Jimmerson Scott and Charles Travis Williams,
and my stepparents, Jack Lewis Scott and Norma Dobbyn Williams, are to
be thanked for their support, both emotional and financial, for these
past four years, and for raising me in such a manner that I could
achieve this degree.
I also wish to acknowledge the time, interest, and energy spent
on my initial graduate school education by my Master of Arts committee:
Richard F. Larson and Felix M. Berardo. Their insistence that I write
a thesis of which I could be proud made writing this dissertation all
the easier. In addition, Benjamin Gorman and Joseph S. Vandiver, whom
I consider to be very good friends, have also served, probably unknowingly
to them, as role models for me.
My doctoral committee, however, has had the most direct influence
on the quality of my education and on the quality of this dissertation,
which, I believe, is to their credit. Mary Anna Baden is an extremely
capable sociologist, and I thank her for the guidance she has provided
and for the constructive criticisms she has made of my work. In 1967, I
began my career at the University of Florida with a course in statistics
from a young professor, James McClave, and now, eight years later, I
finish that career with Professor McClave serving on my doctoral
committee; whatever expertise I possess in the theory underlying statistical
analysis.l owe to him. The debt of gratitude that I owe to Gerald
R..Leslie is truly enormous, for it was he who first recognized that
I might one day leave the University of Florida with a Ph.D. in
sociology; I only hope that one day I will be in a position to repay
this debt. Charles E. Frazier served as co-chairman of my committee
until I passed the qualifying examinations, and I thank him for
preparing me well enough in the sociology of deviant behavior so that
I had very little difficulty in either the written or the oral exams.
The chairman of my committee is George J. Warheit who always managed to
find time for me in his unbelievably busy schedule; if I needed help,
or felt that I needed help, he was there to offer it. I am not sure
that I could have passed my doctoral examinations without his continuing
emotional support and reassurance, and I know for a fact that this
dissertation is all the better for his suggestions and criticisms.
I would also like to express my appreciation to the departmental
secretaries, and my friends, Dianne Wright and Jean Sneeringer. Their
assistance in overcoming the rigidities of the bureaucracy helped to make
my career as a graduate student more bearable and a lot more fun. Thanks
also are due Robert Beckley, Jeff Johnson, LeRoy Martinson, Nancy
Sederberg, Ellen Maher, and Joan Ritter, who constitute the faculty and
staff of the Department of Sociology at Valparalso University, for giving
me my first fulltime employment as a sociologist. I am also indebted
to my student assistant, Mary Sunsdahl, for her help in finishing this
dissertation; she has spent many hours typing and retyping the
manuscript for too little renumeration.
And, finally, I wish to thank those people who made my stay in
Gainesville among the most rewarding years in my life: the graduate
students in the Department of Sociology. They not only played an
essential part in my training, but also played a significant part in
my life. The friendships that were formed will surely last a lifetime
and they all hold a special place in my heart. There are two, however,
who are especially close to me: Sharon Woodruff and Steve Wray.
My love for them is immeasurable and the memories irreplaceable. Two
finer people never existed; two finer friends never will be found.
TABLE OF CONTENTS
ACKNOWLEDGMENTS . . . . . . . . ... . . iii
LIST OF TABLES . . . . . . . ... . . x
ABSTRACT . . . . . . . . .. . . . . xvlil
CHAPTER 1 INTRODUCTION . . . . . . . ... 1
WHAT CAN SOCIOLOGY OFFER? . . . . . . . 1
OBJECTIVES OF THE DISSERTATION . . . . . . 2
STRUCTURE OF THE DISSERTATION . . . . . . 2
CHAPTER 2 REVIEW OF THE LITERATURE . . . . . . 6
HISTORICAL DEVELOPMENT . . . . . . . . 6
MAJOR THEORETICAL FORMULATIONS . . . . . . 7
REVIEW OF EPIDEMIOLOGICAL STUDIES . . . . ... 25
TREATMENT PROCEDURES . . . . . . . ... .27
IN CONCLUSION . . . . . . . ... .. . 30
CHAPTER 3 THEORY . . . . . . . . ... . .31
CULTURAL INFLUENCES AND THE IMPACT OF
DIFFERENTIAL LIFE CHANCES . . . . . . ... 32
IN CONCLUSION . . . . . . . . ... .. .39
CHAPTER 4 RESEARCH DESIGN . . . . . . . .. 40
DESCRIPTION OF SAMPLE SITE . . . . . . . 40
THE SAMPLE . . . . . . . . .
DEVELOPMENT OF PSYCHOMETRIC SCALES . . .
OPERATIONAL DEFINITIONS . . . . . .
STATISTICAL TECHNIQUES USED IN THE STUDY . .
CHAPTER 5 F tIDINGS . . . . . . . . .
COMPREHENSIVE PHOBIA SCALE . . . . .
PERSONAL PHOBIA SCALE . . . . . . .
SITUATIONAL PHOBIA SCALE . . . . . .
INDIVIDUAL PHOBIAS . . . . . . .
IN CONCLUSION . . . . . . . . .
CHAPTER 6 CONCLUSION . . . . . . . .
FEAR AND SEX . . . . . . . . .
FEAR AND RACE . . . . . . . . .
FEAR AND AGE . . . . . . . . .
FEAR AND EDUCATIONAL ATTAINMENT . . . .
FEAR AND FAMILY INCOME . . . . . .
FEAR AND SOCIOECONOMIC STATUS . . . . .
FEAR AND MARITAL STATUS . . . . . .
FEAR AND CHILDHOOD HAPPINESS . . . . .
FEAR AND SATISFACTION WITH RETIREMENT . . .
FEAR AND PLACE OF RESIDENCE . . . . .
FEAR AND ATTENDANCE AT CHURCH . . . . .
FEAR AND SOCIAL PSYCHIATRIC SYMPTOMATOLOGY
QUESTIONS . . . . . . . . . .
FEAR AND FATALISTIC ATTITUDES . . . . .. 142
IN RETROSPECT . . . . . . . .... 142
APPENDICES . . . . . . . ... . . 145
APPENDIX A QUESTIONNAIRE ITEMS . . . . .. .146
APPENDIX B DISTRIBUTIONS OF COEFFICIENTS FOR
THE RELATIONSHIPS BETWEEN SCORES ON THE PHOBIA
SCALES AND SUBJECTS' PERCEPTIONS OF THEIR
RELATIONSHIPS WITH THEIR RELATIVES AND FRIENDS .167
APPENDIX C ANALYSIS OF VARIANCE TABLES FOR
THE PHOBIA SCALES BY SEX, BY RACE, AND BY
AGE, MARITAL STATUS, AND SOCIOECONOMIC STATUS . 171
APPENDIX D A RETROSPECTIVE LOOK AT SOME
WEAKNESSES IN DESIGN AND CONCEPTUALIZATION . . 190
BIBLIOGRAPHY . . . . . . . . . . . 195
BIOGRAPHICAL SKETCH . . . . . . . .... 201
LIST OF TABLES
1. Frequency Distribution of Responses to
Comprehensive Phobia Scale . . . . . . .. 52
2. Frequency Distribution of Responses to
Comprehensive Phobia Scale by Sex . . . . .. 52
3. Frequency Distribution of Responses to
Comprehensive Phobia Scale by Race . . . . .. 53
4. Frequency Distribution of Responses to
Comprehensive Phobia Scale by Age . . . . . 54
5. Frequency Distribution of Responses to
Comprehensive Phobia Scale by Family Income . . .. 55
6. Frequency Distribution of Responses to
Comprehensive Phobia Scale by Socioeconomic Status . 56
7. Frequency Distribution of Responses to
Comprehensive Phobia Scale by Marital Status . . .. 57
8. Frequency Distribution of Responses to
Comprehensive Phobia Scale by Reported
Degrees of Childhood Happiness . . . . .... 58
9. Frequency Distribution of Responses to
Comprehensive Phobia Scale by Educational
Attainment . . . . . . . ... .... . 61
10. Gamma and Tauc Coefficients for the Relationship
Between Comprehensive Phobia Scale Scores and
Socioeconomic Status Controlling for Race, Sex . . 64
11. Gamma and Tauc Coefficients for the Relationship
Between Comprehensive Phobia Scale Scores and
Childhood Happiness Controlling for Race, Sex,
and Race-Sex . . . . . . . .... ..... 67
12. Gamma and Tauc Coefficients for the Relationship
Between Comprehensive Phobia Scale Scores and
Satisfaction with Retirement Controlling for
Race, Sex, and Race-Sex . . . . . .... .. . 68
13. Gamma and Tauc Coefficients for the Relationships
Between Comprehensive Phobia Scale Scores and the
Major Independent Variables . . . . . .... 70
14. Gamma and Tauc Coefficients for the Relationships
Between Comprehensive Phobia Scale Scores and Social
Psychiatric Symptomatology Questions . . . ... 71
15. Gamma and Tauc Coefficients for the Relationships
Between Comprehensive Phobia Scale Scores and
Questions Tapping Fatalistic Attitudes . . . ... 72
16. Prediction of Comprehensive Phobia Scale Scores
by Race, Sex, Age, and Socioeconomic Status ...... 73
17. Frequency Distribution of Responses to Personal
Phobia Subscale . . . . . . . ...... 74
18. Frequency Distribution of Responses to Personal
Phobia Subscale by Sex. . . . . . . . .75
19. Frequency Distribution of Responses to Personal
Phobia Subscale by Race . . . . . . . .. 75
20. Frequency Distribution of Responses to Personal
Phobia Subscale by Age . . . . . . . . 76
21. Frequency Distribution of Responses to Personal
Phobia Subscale by Family Income . . . . .... 77
22. Frequency Distribution of Responses to Personal
Phobia Subscale by Socioeconomic Status . . . . 78
23. Frequency Distribution of Responses to Personal
Phobia Subscale by Marital Status . . . . .... 79
24. Frequency Distribution of Responses to Personal
Phobia Subscale by Reported Degrees of Childhood
Happiness .. .... . . . ...... . . 7 79
25. Frequency Distribution of Responses to Personal
Phobia Subscale by Educational Attainment . . . .. 81
26. Gamma and Tauc Coefficients for the Relationship
Between Personal Phobia Subscale Scores and Family
Income Controlling for Race, Sex . . . . .... 82
27. Gamma and Tauc Coefficients for the Relationship
Between Personal Phobia Subscale Scores and
Socioeconomic Status Controlling for Race, Sex .... .83
28. Gamma and Tauc Coefficients for the Relationship
Between Personal Phobia Subscale Scores and Childhood
Happiness Controlling for Race, Sex, and Race-Sex . . 84
29. Gamma and Tauc Coefficients for the Relationship
Between Personal Phobia Subscale Scores and
Satisfaction with Retirement Controlling for Sex,
Race, and Race-Sex. .... . . . . . . .85
30. Gamma and Tauc Coefficients for the Relationships
Between Personal Phobia Subscale Scores and the
Major Independent Variables . . . . . .... 87
31. Gamma and Tauc Coefficients for the Relationships
Between Personal Phobia Subscale Scores and Social
Psychiatric Symptomatology Questions . . . ... 88
32. Gamma and Tauc Coefficients for the Relationships
Between Personal Phobia Subscale Scores and Questions
Tapping Fatalistic Attitudes . . . . . . 89
33. Prediction of Personal Phobia Subscale Scores by
Race, Sex, Age, and Socioeconomic Status . . ... 89
34. Frequency Distribution of Responses to Situational
Phobia Subscale . . . . . . . . .. ... . 90
35. Frequency Distribution of Responses to Situational
Phobia Subscale by Sex . . . . . . . .. 91
36. Frequency Distribution of Responses to Situational
Phobia Subscale by Race . . . . . . .... 92
37. Frequency Distribution of Responses to Situational
Phobia Subscale by Age. . . . . . . . .93
38. Frequency Distribution of Responses to Situational
Phobia Subscale by Educational Attainment . . ... 94
39. Frequency Distribution of Responses to Situational
Phobia Subscale by Family Income . . . . ... 95
40. Gamma and Tauc Coefficients for the Relationship
Between Situational Phobia Subscale Scores and
Family Income Controlling for Race, Sex . . . ... 95
41. Frequency Distribution of Responses to Situational
Phobia Subscale by Socioeconomic Status . . . .. 96
42. Gamma and Tauc Coefficients for the Relationship
Between Situational Phobia Subscale Scores and
Socioeconomic Status Controlling for Race, Sex .... .97
43. Frequency Distribution of Responses to Situational
Phobia Subscale by Marital Status . . . . .... 98
44. Frequency Distribution of Responses to Situational
Phobia Subscale by Reported Degrees of Childhood
Happiness . . . . . . . . . ... . 99
45. Gamma and Tauc Coefficients for the Relationship
Between Situational Phobia Subscale Scores and
Childhood Happiness Controlling for Race, Sex, and
Race-Sex . . . . . . . .. ... . 99
46. Gamma and Tauc Coefficients for the Relationship
Between Situational Phobia Subscale Scores and
Satisfaction with Retirement Controlling for Sex,
Race, and Race-Sex . . . . . . . .... 100
47. Gamma and Tauc Coefficients for the Relationships
Between Situational Phobia Subscale Scores and the
Major Independent Variables . . . . . .... 101
48. Gamma and Tauc Coefficients for the Relationship
Between Situational Phobia Subscale Scores and Social
Psychiatric Symptomatology Questions . . . . .. 102
49. Gamma and Tauc Coefficients for the Relationships
Between Situational Phobia Subscale Scores and
Questions Tapping Fatalistic Attitudes . . . ... 103
50. Prediction of Situational Phobia Subscale Scores
by Race, Sex, Age, and Socioeconomic Status . . .. 104
51. Frequency Distribution of Responses to Fear of
Animals or Insects . . . . . . . . . 105
52. Frequency Distribution of Responses to Fear of
Animals or Insects by Sex . . . . . . .... 105
53. Frequency Distribution of Responses to Fear of
Animals or Insects by Race . . . . . .... 105
54. Chi Square, Gamma, and Tauc Coefficients for the
Relationships Between Fear of Animals and Insects
and Major Independent Variables . . . . .... 106
55. Prediction of Fear of Animals and Insects by
Race, Sex, Age, and Socioeconomic Status . . ... 107
56. Frequency Distribution of Responses to Fear of
High Places . . . . . . . . . . . 107
57. Frequency Distribution of Responses to Fear of
High Places by Sex . . . . . . . . .108
58. Frequency Distribution of Responses to Fear of
High Places by Race . . . . . . . . . .108
59. Chi Square, Gamma, and Tauc Coefficients for the
Relationships Between Fear of High Places and
Major Independent Variables . . . . . .... 109
60. Prediction of Fear of High Places by Race, Sex,
Age, and Socioeconomic Status . . . . . .... 109
61. Frequency Distribution of Responses to Fear of
Closed In Places . . . . . . . .... .110
62. Frequency Distribution of Responses to Fear of
Closed In Places by Sex . . . . . . .... 110
63. Frequency Distribution of Responses to Fear of
Closed In Places by Race . . . . . . .... 111
64. Chi Square, Gamma, and Tauc Coefficients for the
Relationships Between Fear of Closed In Places and
Major Independent Variables . . . . . .... 111
65. Prediction of Fear of Closed In Places by Race,
Sex, Age, and Socioeconomic Status . . . .... 112
66. Frequency Distribution of Responses to Fear of
Thunder and Lightning . . . . . . .... 112
67. Frequency Distribution of Responses to Fear of
Thunder and Lightning by Sex . . . . . .. 113
68. Frequency Distribution of Responses to Fear of
Thunder and Lightning by Race . . . . . .... 113
69. Chi Square, Gamma, and Tauc Coefficients for the
Relationships Between Fear of Thunder and Lightning
and Major Independent Variables . . . . . .. 114
70. Prediction of Fear of Thunder and Lightning by
Race, Sex, Age, and Socioeconomic Status . . ... 114
71. Frequency Distribution of Responses to Fear of
the Dark . . . . . . . ... .. . 115
72. Frequency Distribution of Responses to Fear of
the Dark by Sex . . . . . . . . ... .. .115
73. Frequency Distribution of Responses to Fear of
the Dark by Race . . . . . . . .... .116
74. Chi Square, Gamma, and Tauc Coefficients for the
Relationships Between Fear of the Dark and Major
Independent Variables . . . . . . . .... 116
75. Prediction of Fear of the Dark by Race, Sex, Age,
and Socioeconomic Status . . . . . . .... 117
76. Frequency Distribution of Responses to Fear of
Driving or Riding in a Car . . . . . . . 117
77. Frequency Distribution of Responses to Fear of
Driving or Riding in a Car by Sex . . . . .... 118
78. Frequency Distribution of Responses to Fear of
Driving or Riding in a Car by Race . . . . .. 118
79. Chi Square, Gamma, and Tauc Coefficients for the
Relationships Between Fear of Driving or Riding in
a Car and Major Independent Variables . . . ... 119
80. Prediction of Fear of Driving or Riding in a Car
by Race, Sex, Age, and Socioeconomic Status ...... 119
81. Frequency Distribution of Responses to Fear of
Crowds . . . . . . . . . . . . 120
82. Frequency Distribution of Responses to Fear of
Crowds by Sex . . . . . . . .... ..... 120
83. Frequency Distribution of Responses to Fear of
Crowds by Race . . . . . . . . ... .. .121
84. Chi Square, Gamma, and Tauc Coefficients for the
Relationships Between Fear of Crowds and Major
Independent Variables . . . . . . . .... 121
85. Prediction of Fear of Crowds by Race, Sex,
Age, and Socioeconomic Status . . . . . .... 122
86. Frequency Distribution of Responses to Fear of
Strangers . . . . . . . . . . . 122
87. Frequency Distribution of Responses to Fear of
Strangers by Sex . . . . . . . .... . .123
88. Frequency Distribution of Responses to Fear of
Strangers by Race . . . . . . . .... . .123
89. Chi Square, Gamma, and Tauc Coefficients for the
Relationships Between Fear of Strangers and Major
Independent Variables . . . . . . . .... 124
90. Prediction of Fear of Strangers by Race, Sex,
Age, and Socioeconomic Status . . . . . .... 124
91. Frequency Distribution of Responses to Fear of
Bodily Harm . . . . . . . . ... .... . 125
92. Frequency Distribution of Responses to Fear of
Bodily Harm by Sex. . . . . . . . . ... 125
93. Frequency Distribution of Responses to Fear of
Bodily Harm by Race . . . . . . . .... 126
94. Chi Square, Gamma, and Tauc Coefficients for the
Relationships Between Fear of Bodily Harm and
Major Independent Variables . . . . . .... 126
95. Prediction of Fear of Bodily Harm by Race, Sex,
Age, and Socioeconomic Status . . . . . ... 127
96. Frequency Distribution of Responses to Fear of
Being Hexed, Witched, or "Evil Eye" . . . . .. .127
97. Frequency Distribution of Responses to Fear of
Being Hexed, Witched, or "Evil Eye" by Sex . . ... 128
98. Frequency Distribution of Responses to Fear of
Being Hexed, Witched, or "Evil Eye" by Race ...... 128
99. Chi Square, Gamma, and Tauc Coefficients for the
Relationships Between Fear of Being Hexed, Witched,
or "Evil Eye" and Major Independent Variables . . .. .129
100. Prediction of Fear of Being Hexed, Witched, or
"Evil Eye" by Race, Sex, Age, and Socioeconomic
Status . . . . . . . . .. . . . 129
101 Frequency Distribution of Responses to Fear of
Other Things . . . . . . . .... ... .. 130
102 Frequency Distribution of Kinds of Responses of
Fear of Other Things as Reported by Subjects ..... 130
Abstract of Dissertation Presented to the Graduate Council
of the University of Florida in Partial Fulfillment of the Requirements
for the Degree of Doctor of Philosophy
THE EPIDEMIOLOGY OF COMMON FEARS AND DIFFUSE PHOBIAS:
A SOCIODEMOGRAPHIC ANALYSIS
Billy L. Williams
Chairman: George J. Warheit
Major Department: Sociology
While some disciplines in the behavioral sciences, particularly
psychology'and psychiatry, have devoted a substantial part of their
literature towards understanding the dynamics of the etiology of phobia,
they have neglected to consider the influence that social structures,
and the differing histories of various peoples, might have in creating
the social climate in which phobias are acquired. Generally, behavioral
scientists have not failed to consider cultural influences when studying
"individualistic" behavior; the study of phobia, however, is an
exception as very little attention has been given to rates of phobic
neurosis and cultural factors involved in its etiology. This
dissertation seeks to make a substantive contribution in that direction.
The dissertation has as its first objective, to discover which
groups of people have higher rates of fears and phobias than others.
This is to be accomplished by searching for relationships between a
set of sociodemographic variables and the subjects' scores on the scales
designed to measure gross phobic symptomatology. The second objective
Is to propose a sociological model to explain and account for those
patterns uncovered in the first phase of the dissertation, both for
phobia rates and for specific phobias.
In 1969, researchers at the University of Florida began an
intensive five year study designed to evaluate the mental health
needs and services in the southeastern United States. As part of
the prevalence component of that study, a random sample of 2333 persons
residing in Alachua County, Florida, was chosen. The interview
schedule, consisting of 317 items and 15 scales, was administered
to the 1967 persons who could be located and who agreed to participate.
Included in the interview schedule were 10 items designed to tap
fear and phobic behavior and thought patterns among the general
population. Specifically, they dealt with the following fears:
driving or riding in a car, being in crowds, strangers, bodily harm,
being hexed or witched, animals and insects, high places, closed in
places, thunder and lightning, and the dark. An interval level scale score
was obtained for each subject by summing the number of positive responses.
Although specific hypothesis testing was not undertaken, the data were
subjected to statistical analysis and levels of significance were
reported for those who desire such an interpretation of the data.
The major inferential techniques utilized were Chi Square Test of
Independence, Kendall's Tauc, gamma, one-way analysis of variance,
and stepwise multiple regression. The procedure used when imposing
controls upon the data was to generate separate tables for the categories
of the appropriate variables and to compute the various statistics.
In considering rates of phobic neurosis by sex and race, the
data indicate that women and blacks consistently have higher scale
scores than men and whites, respectively, for each of the items
comprising the scale. Black females, without exception, scored
highest on the scale and white males scored lowest. Generally, white
females scored higher than black males on items that are personal-
specific and black males scored higher on items that are situation-
specific. Statistically significant relationships were also observed
for phobia scores with the following variable list: level of
educational attainment, family income, socioeconomic status, marital
status, self-reported degrees of childhood happiness, other questions
measuring social psychiatric symptomatology and impairment, and
fatalistic outlooks on life. No definite statements could be made
concerning the relationships between phobia scores and the subjects'
places of residence, religiosity, or satisfaction with retirement.
With the exceptions of psychology and psychiatry, the behavioral
sciences have ignored phobic behavior as social phenomena. It is the
central tenet of this dissertation that phobic neurosis, traditionally
viewed as symptomatic of a maladjusted personality, can be studied
with the unit of analysis being the group, rather than the individual.
Additional insights into the nature and origin of common fears and
phobias may be obtained by studying their distribution among the general
population and by identifying those groups which seem to be particularly
prone to acquire phobia.
What Can Sociology Offer?
While the disciplines of psychology and psychiatry have devoted
a substantial part of their literature towards understanding the dynamics
of the etiology of phobia, they have neglected to consider the influence
that social structures, and the differing histories of various peoples,
might have in creating the social climate in which phobias are acquired.
Ever since Emile Durkheim took what was generally considered to be
the ultimate individualistic act, suicide, and illustrated how rates of
suicide vary by different categories of persons and remain relatively
constant over time, social scientists have usually not neglected to
consider cultural influences when studying "individualistic" behavior.
Somehow, the study of phobia has not been pursued at this level; very
little attention has been given to rates of phobic neurosis in the
literature. This dissertation seeks to make a substantive contribution
in this direction.
Objectives of the Dissertation
The dissertation has as its first objective, to discover which
groups of people have higher rates of fears and phobias than others.
This is to be accomplished by searching for relationships between a set
of sociodemographic variables and the subjects' scores on the scales
designed to measure gross phobic symptomatology.
The second objective is to propose a sociological model to explain
and account for those patterns uncovered in the first phase of the
dissertation, both for phobia rates and for specific phobias. Anticipating
the analysis of the data, the model will likely rely quite heavily on
cultural differences between groups to explain differential phobia rates.
Structure of the Dissertation
The dissertation is divided into five substantive areas. The
first consists of a review of the literature (Chapter 2). It begins
with a brief introduction to the history of the study of phobic neurosis
and to the major concepts involved in that study. Following a cursory
look at how psychology and psychiatry have traditionally defined and
diagnosed phobic neurosis, the dissertation presents the basic principles
underlying the major theoretical orientations used in the study of phobia:
the psychoanalytic perspective, including both the "displacement of
anxiety" approach and the "defense against threatening impulses"
approach, and the behavioristic or learning perspective, which includes
discussions on the general principles underlying the process of
socialization and operant conditioning and avoidance learning. The
work of Isaac M. Marks is given consideration apart from these two
general perspectives because of the comprehensiveness of his model
which includes both the psychoanalytic and the behavioristic perspectives
plus the influence that phylogenetics have on phobia acquisition. A
review of past epidemiological studies then follows, and the chapter
concludes with a short discussion on current procedures used to treat
phobic patients as well as on some practical measures which might be
utilized by such persons to overcome their handicap without submitting
to chemotherapy, psychotherapy, or desensitization.
Chapter 3 presents a tentative theory to be used in the
dissertation while attempting to explain observed differences in phobia
rates. Beginning with a general statement on cultural influences and
the impact of differential life chances on "individualistic" behavior,
the chapter first considers how one's racial or ethnic membership might
conceivably affect the probability of acquiring phobia. This section
is followed by discussions on the relationships between phobic neurosis
with dependency needs and sex role identification, with age and
retirement status, with the type of locale in which an individual
resides, with religion and the degree of religiosity, and with the
level of educational attainment.
The methodology used in the study is presented in Chapter 4.
It includes sections describing the sample site, the sampling techniques
utilized, and the development of the psychometric scales. The
operational definitions are then presented for the investigation's
major dependent and independent variables: the Comprehensive Phobia
Scale, the Personal Phobia Scale, the Situational Phobia Scale, the
specific item phobias, and a standard set of sociodemographic variables.
The chapter concludes with a short discussion on the statistical
techniques used in the study, the method of presenting the data in
subsequent chapters, and the method chosen for implementing controls
upon the data.
Chapter 5 presents the major findings of the investigation, and
reports the various correlation results, including the appropriate
levels of significance. Also, a preliminary analysis of the data is
provided, linking the findings with the discussions presented in the
third chapter. This procedure is followed for all three phobia scales
and for each of the ten items comprising those scales.
The conclusions made after examining the data are presented in
Chapter 6. Whereas in the previous chapter the discussion is broken
down according to the dependent variables, the present chapter uses the
independent variables to regroup the major findings. That is, the findings
of Chapter 5 are now categorized according to age, sex, race, and so on.
Hopefully, this allows the reader to grasp more easily the effect each
independent variable has on phobia in general. Naturally, the theory
of Chapter 3 is re-examined when the findings are not consistent with
the principles outlined previously, and alternative explanations are
made. The chapter concludes with a retrospective look at the dissertation
as a whole, stating the major difficulties encountered during the
research and the analysis of the subject matter, and the weaknesses
of the research design.
The remaining pages in the dissertation are devoted to additional
tables not included in the main text, the reproduction of the items
on the interview schedule that were used in this investigation, and
the list of references consulted and quoted.
REVIEW OF THE LITERATURE
The word, phobia, was taken from the name given the ancient
Greek god thought to be responsible for the presence of fear and panic
in one's enemies (Marks, 1970a: 377). Its sole medical usage prior to
the 20th century was to describe the symptom, hydrophobia, a term
for rabies coined by Celsus (1814: 236).
It was Le Camus (1722-1772), however, in his book Des Aversions,
who first attempted a comprehensive medical study of phobias. He
described the predominate attributes of his patients' aversions, noted
their frequent geneses from the early developmental years, and
classified them according to the sense (touch, sight, smell, hearing,
and taste) which was most affected (Errera, 1962: 326-329).
A number of theories have offered explanations of the dynamics of
phobic behavior and the etiology of phobia. Some, for example, have
attributed the acquisition of phobic behavior to "poor up-bringing"
(Rush, 1818), "stomach ailments" (Benedikt, 1870), and a ". . morbid
predisposition of the visceral ganglionic nervous system" (Morel, 1866).
Current theories as to genesis are correspondingly varied. There are
explanations in terms of "primary insufficiency of the ecto-mesodermal
barrier" (Brun, 1951), the principles of learning, and, under the
heading of analytic theories, a multiplicity of formulations. It is
on the latter two general perspectives that this dissertation now
Major Theoretical Formulations
An Introduction to the Concepts
Errera (1962: 325) defines phobia as ". .. a persisting fear of
an object or of an idea which does not ordinarily justify fear." Rachman
(1968: 5) believes phobias can best be viewed as conditioned fear, or
anxiety, reactions. Thus, he considers phobias to be anxiety states in
which the focus of the anxiety-producing stimulus is "sharpened." James
C. Coleman and William E. Broen, Jr. (1972: 770) define it simply as an
irrational fear; dozens of similar definitions can be cited from the
From the definitions found in the literature, it appears that there
are two primary distinguishing characteristics separating phobias from
other psychoneurotic disorders: immobilizing and irrational fear. Fear,
of course, is an emotion which occurs under stress conditions when the
individual is faced with uncertainty and/or danger (Gove, 1965: 305).
All stress situations, naturally, are not fear producing.
The particular emotional states accompanying reactions to stress
vary greatly, from grief and depression at one extreme to emotional
mobilization for emergency action at the other. According to Coleman
and Broen (1972: 120-121), three emotional patterns are commonly
observed when persons face stress situations: anger, fear, and anxiety.
Frustration, for example, tends to elicit anger or hostility: the
individual's immediate reaction to frustration is typically one of
anger. And anger, in turn, typically leads to an aggressive action
(that is, attack) as the person tries to remove the frustrating obstacle.
When the sense of frustration is continuous, as when the individual
is faced with a succession of frustrating situations stemming either
from the same or different sources, anger gradually blends into
hostility. Concomitant with hostile feelings is a tendency to destroy,
damage, or inflict pain upon the person, or object, perceived as the
source of frustration (Buss, 1961). "Anger and hostility are extremely
difficult emotions to deal with," Coleman and Broen (1972: 120) state,
"inasmuch as civilized living permits few direct outlets. Consequently,
these emotions are often expressed in indirect and deviant but 'safe'
ways." One such "safe" way is through the development of phobias.
Second, specific dangers tend to elicit fears. The perception of
danger usually arouses fear; and fear, in turn, tends to elicit withdrawal
or flight. The actual behavior evoked by fear, however, depends greatly
upon the individual and the specific situation.
Third, a sense of threat tends to elicit anxiety. Anxiety is
a feeling of apprehension and fearfulness in the absence of specific
danger. The source of threat here is usually not apparent to either
one's self or to others (Coleman and Broen, 1972: 120-121).
The Diagnosis of Phobic Neurosis
To reiterate, a phobic neurosis is a persistent fear of some object
or event that presents no actual danger to the person or in which the
danger is magnified out of proportion to its actual seriousness. In
addition, phobic neurosis need not apply solely to a learned fear of
an object or event, but may also refer to a general pattern of fear
and avoidance behavior (Coleman and Broen, 1972: 227).
The clinical picture of severe phobic neurosis is one in which
the patient is plagued by a fear totally out of proportion to the
demands of the situation, beyond voluntary control, cannot be explained
or reasoned away, and leads to :incapacitating avoidance behavior. Each
encounter, or potential encounter, with the object of fear is preceded
by an intense anxiety reaction.
According to Rachman (1968), this reaction can be analyzed into
three basic components: subjective, autonomic, and motor. The
subjective component consists of a feeling of alarm and may include
fear, tension, or full panic. The autonomic reaction can produce any
of the following physiological changes: nausea, palpitations,
breathlessness, involuntary excretions, dryness of mouth, trembling,
sweating, and weakness. The motor response is usually one of flight
but some persons become, on occasion "frozen" and unable to move.
Furthermore, the patient who acknowledges the irrationality of
his behavior may create additional psychological problems by handicapping
himself with a sense of shame and the desire and drive to conceal his
-difficulties from-others (Julier;-1973: 52).
In attempting differential diagnosis, Julier suggests the clinician
seek answers to five basic questions: what are the feared objects and
situations? is depression present? is there evidence of other
psychopathology? are there features suggestive of epilepsy? and could
the symptoms be explicable in terms of organic pathology?
Concerning the first question, Julier suggests that if there
are a great many and a large variety of fears present since childhood
or adolescence, the patient may have a personality disorder for which
the phobic behavior is merely a secondary symptom of more intense,
underlying psychopathology. Similarly, if phobic symptoms have been
present only a short time, they may indicate nothing more than a
temporary anxiety state, a state which is often accompanied by
diffusely increased fearfulness.
If depression is present in the patient, and if it can be
established whether the phobic state arose in the context of a depressive
illness, then it is likely once again that the phobia symptoms are
secondary and temporary in nature.
Third, phobic symptoms are present in about ten percent of the
schizophrenics but do not usually call for special management themselves.
Obsessional fears, such as fear of contamination or of causing harm to
others, are more difficult to categorize. These obsessional fears are
believed to be based on repetitive, stereotyped ideas or Impulses which
then dictate the need to avoid a range of situations linked by a
common theme, and anxiety derives from the imagined consequences of
exposure. In phobic, as distinct from obsessional, disorders, the fear
seems to derive from the situation itself.
Fourth, depersonalization and intense anxiety are well recognized
in temporal-lobe seizures. The phobic patient will tend to experience
a greater range of type and intensity of experience and the provoking
stimuli are not of the sort most commonly responsible for triggering
Last, organic pathology should be ruled out as a possible cause
of manifested phobic symptoms. Among the organic determinants of
phobic symptomatology are the following: paroxysmal tachycardia,
hyperthyroidism, anemia, coronary artery disease, left ventricular
failure, and asthma (Julier, 1973: 53-54).
Once the diagnosis has been made, treatment can commence.
Techniques for treatment will be covered later in this chapter. The
focus of this dissertation now shifts to a review of the basic
principles underlying the major theoretical perspectives on phobia
and its acquisition and etiology.
Dynamic Patterns of the Etiology of Phobia
Traditionally, phobias have been considered to be attempts by the
individual to cope with internal or external dangers by his carefully
selecting circumstances which would decrease the likelihood of coming
into contact with the phobic stimulus. Consequently, phobias have
been viewed as simple defense reactions in which the individual feels
he must concede to his fear in order to protect himself.
Three major dynamic patterns have been delineated in phobia
acquisition from two separate perspectives: displacement of anxiety,
defense against threatening impulses, and avoidance learning (Coleman
and Broen, 1972: 228).
The psychoanalytic perspective
Displacement of anxiety. A phobia may represent a displacement
of anxiety from the stress situation that elicited it to some other
object or situation. This mechanism forms much of the basis for
psychoanalytic theory and is derived principally from Freud's case
history of Hans published in 1909. On the basis of his experiences with
Hans and others, Freud concluded that phobias represent displaced
anxiety associated with the Oedipus and Electra complexes. Regarding
the Oedipus complex, the child is thought to desire a sexual union
with his mother. As such, he is in competition with his father for the
mother's love. The result is jealous and hostile feelings toward the
father. The child, therefore, fears his father and, in particular, fears
being castrated by him. This fear may then be displaced on a more
socially acceptable, formerly innocuous, object. In the case of Hans,
Freud reasoned that when the horse pulling the carriage occupied by
Hans and his mother fell down and was injured, the child had become
quite frightened and displaced his fear of castration to fear of being
bitten by a horse. In conclusion, Freud stated that phobias in adults
develop only in people with disturbed sexual relationships (Coleman
and Broen, 1972: 228).
Later investigators have noted that many different stress situations
may lead to the displacement of anxiety and the creation of phobic
symptomatology. Coleman and Broen (1972: 228-229) cite an example:
A young man had started his first business venture by
renting a very small store near the entrance to a large building
and stocking it exclusively with neckties. He managed to make
ends meet over a trial period of several months, but it was
becoming obvious that his business venture was doomed to failure.
About this time he noticed that the shop "seemed stuffy," that
he didn't have any "elbow room." It seemed as if the walls
were closing in on him, and he would feel compelled to go
outside and get his "lungs full of fresh air." These feelings
increased in frequency and intensity until he was forced to
close the shop.
Coleman's and Broen's analysis is simple: there was no need to admit
failure or poor business acumen since the man was forced to go out
of business for other reasons.
Many phobias may indicate a more generalized anxiety state than
the examples given above. Arieti (1961) has observed that intense guilt
feelings may lead to phobias or situations that are symbolically
associated with anticipated punishment. Coleman and Broen cite the
example of a person characterized by a phobic reaction to the possibility
of speaking in public as being related to the individual's more
generalized fear that others will "detect his insufficiencies" and
"see him for what he really is." Likewise, a phobia of traveling may
hide a greater fear of making excursions into life (Coleman and Broen,
Defense against threatening impulses. Salzman (1968a; 1968b) and
Lief (1968) have noted that phobia may represent a defensive reaction
that protects the individual from situations in which his repressed
aggressiveness or sexual impulses might become dangerous. As in the
immediately preceding section, anxiety is displaced; one's conscious
fear is not the basic cause of one's anxiety. The only difference
between phobias of this type and the phobias discussed above is that
here phobic symptoms develop as a result of unacceptable impulses arising
from the individual rather than from a fear that someone will do
something to him. The distinction lies with the activity-passivity of
the subject: in phobias of the type being considered in this section,
the phobia prevents the subject from performing some act he considers
immoral. In the previous section, phobias arise to prevent the
person from dwelling on an event that might force its way upon him
(for example, the castration complex of Hans).
As an illustration of phobias as defense reactions against
threatening impulses, Coleman and Broen (1972: 229) cite the following:
. 24 year-old Herbert S. sought psychological
assistance because of a "morbid fear of syphilis which makes
it impossible for me to have sexual relations." In the
course of psychotherapy it became apparent that this youth's
syphilophobia represented a displacement of a fear of engaging
in homosexual relations. His fear of contracting syphilis
helped him to maintain what he considered sexually moral
behavior. That it also prevented him from engaging in
heterosexual relations -- which he viewed as moral -- apparently
stemmed from the fact that during intercourse his "mind was
completely dominated by homosexual fantasies." These fantasies
were highly anxiety-arousing and on a number of occasions
had terminated in impotency, which he had found acutely
embarrassing. Thus the syphilophobia kept him from situations
that might arouse his sexual conflicts.
In summation. The two sections above are generally considered
to comprise the psychoanalytic perspective of phobia acquisition rather
than as constituting two separate viewpoints. The reason appears
twofold: (1) both involve anxiety displacement as the primary mechanism
involved in the formation of phobias, and (2) both deal with internal
and internalized conflicts rather than with current conflicts with
the external world.
Hurry and Sandier (1971: 379), in discussing this point, note
that this is so partly because the initial public impact of psychoanalysis
occurred during the second phase of its development (1897-1923) at a
time when Freud had pretty much abandoned the traumatogenic theory
of neurosis and was mainly concerned with the way in which unconscious
instinctual drives and wishes expressed themselves Lith adult patients
in the clinical psychoanalytic situation. The main emphasis at
this time was on the way in which such drives were themselves in
conflict with one another, as well as on the role of the childhood
internalizations of external reality, and the past and present conflicts
to which these internalizations contributed. Freud saw the ego as
a central regulating agency, mediating between the demands of the id,
the superego and external reality, and defenses were seen as being
set in motion by an anxiety signal which could arise from any of
The behavioristic perspective
Conditioning and avoidance learning. In their now famous
experiment on eleven month old Albert, Watson and Rayner (1920) showed
that phobias could be experimentally induced. Albert, who was fond of
animals, was conditioned to exhibit fear symptomatology at the sight
of a white rat. By striking a steel bar with a hammer whenever
Albert reached for the rat, the experimenters were able to elicit
fear from the child and make him cry. After several repeated trials,
Albert came to be disturbed merely by seeing the rat even without the
loud noise. The fear reaction then generalized to include other furry
animals and objects.
Four years later, Mary Cover Jones (1924) succeeded in eliminating
a fear of furry animals by presenting a white rabbit at a distance when
a child, Peter, was reacting positively to food -- an anxiety inhibitor.
By gradually bringing the rabbit closer to Peter, the boy was able
to lose his fear of it. Generalization also occurred in this instance
as Peter eventually lost his fear of all fuzzy animals.
These two classic experiments began what today has become
known as the learning school or the behavioristic perspective. In
this school, phobias are often seen as the learned result of trauma
in a fear situation. And this fear may generalize to include a
greater variety of similar situations.
Since this perspective views phobias as learned behavior, all
requisites for "normal" learning are also necessary for the learning
of phobias. Psychological maturation is a key concept of relevance at
this point. The person must, regardless of the degree of physical
maturity or age in years, months, or days, have attained a given level
of psychological maturity before the learning of a specific fear can
occur. Hebb (1966:146) illustrates, for example, that a baby's fear of
strangers rarely develops before age six months regardless of the
frequency or quality of contact with strangers prior to that time.
Similarly, Hebb also notes that fears of monsters and other imaginary
creatures, and of the dark, rarely occur before age three years although
they are relatively common among children after that age.
Probably best exemplified in the works of Stanley Rachman and
H. J. Eysenck, behaviorism holds that phobias are best understood if
viewed as conditioned fear, or anxiety, reactions (Rachman, 1968: 5).
In essence, the theory states that phobias are:
. acquired by classical conditioning in which the
(future) stimulus . is paired together with or shortly
before a noxious stimulus. Conditioning occurs through this
temporal contiguity, and the phobia increases with the frequency
of the pairing, with the strength of the noxious stimulus, and
when the pairing occurs in conditions of confinement or when
nothing can be done to stop the noxious stimulation. (Marks,
The essentials of the theory may be summarized in ten statements:
1. Phobias are learned responses;
2. Stimuli develop phobic qualities when they are associated
temporally and spatially with a fear-producing state of
3. Neutral stimuli which are of relevance in the fear-producing
situation and/or make an impact on the person in the
situation, are more likely to develop phobic qualities than
weak or irrelevant stimuli;
4. Repetition of the association between the fear situation
and the new phobic stimuli will strengthen the phobia;
5. Associations between high intensity fear situations and
neutral stimuli are more likely to produce phobic reactions;
6. Generalization from the original phobic stimulus to stimuli
of a similar nature will occur;
7. Noxious experiences which occur under conditions of
excessive confinement are more likely to produce phobic
8. Neutral stimuli which are associated with a noxious
experience may develop (secondary) motivating properties.
This acquired drive is termed the fear-drive;
9. Responses (for example, avoidance) which reduce the fear-drive
are reinforced; and
10. Phobias can be acquired vicariously. (Eysenck and Rachman,
1965: 81-82; Rachman, 1968: 31) Each of the above statements is
based on experimental evidence with animals and, to a lesser extent,
humans. Furthermore, they do appear to be consistent with clinical
experience (cf. Eysenck and Rachman, 1965; Wolpe, 1958; Eysenck, 1960a).
In short, any neutral stimulus, simple or complex, which makes
an impact on an individual.about the time that a fear reaction is
evoked, acquires the capability to evoke fear in subsequent encounters.
If the fear in the original conditioning situation is of high intensity,
or if the conditioning is repeated a good many times, the conditioned
fear will show the persistence that is characteristic of neurotic
fear; and there may be generalization of fear reactions to stimuli
resembling the conditioned stimulus (Rachman, 1968: 5).
Principles of the socialization process. No matter what the
discipline, most discussions of the socialization process emphasize
the development of the self. Writing at the turn of the 20th century,
Sigmund Freud recognized three analytically separable dimensions of
the self: the id, the ego, and the superego. The id was viewed as the
unconscious reservoir of libidinal energy, operating solely on the
pleasure principle, that was primarily concerned with the biological
and Instinctual properties of the individual. The ego, an outgrowth
of the id, was seen as the conscious dimension of the self that serves
primarily as moderator between the id and the superego, and
". . represents what we call reason and sanity." The superego, an
outgrowth of the ego, is largely, but not totally, unconscious, and
its chief function is to serve as the ego's "moral watchdog."
Furthermore, the superego is that product of socialization that we
commonly refer to as the "conscience" (Healy et al., 1944: 34-48).
The phenomenological psychologists try to understand behavior
from the point of view of the individual rather than from the point of
view of the outside observer looking in. Within this framework an
individual's behavior is always considered reasonable and behavior which
appears illogical at times to others is always logical from the
perspective of the individual actor. In other words, it is important
to recognize that ". . people perceive what is appropriate for persons
with their self-concepts to perceive" (Blocker, 1966: 32).
A phenomenological psychologist such as Carl Rogers defines
the self-concept as
. an organized conceptual gestalt consisting of
the individual's perception of himself alone and of himself
in relation to other persons and objects in his environment
together with the values attached to these perceptions. The
self-concept is not always in awareness, but it is always
available to awareness. . The self-concept is considered
to be fluid and changing, a process rather than an entity,
but at any given moment it is a fixed entity. . The self-
concept is the picture an individual has of himself along
with his evaluation of this picture. (Grummon, 1965: 40)
Rogers' six basic propositions provide a good overview of the
phenomenological model. These are:
1. The individual exists in a continually changing world of
experience of which he is the center. His perceptual field
is not completely known to any other person. This world is
his phenomenological field, that is, his own private world
2. The individual reacts to his private world as he perceives it.
3. Individuals react to their perceptual fields as organized
wholes . as total human organisms.
4. Every human being has within himself one basic tendency or
striving. He is constantly struggling to enhance and
5. Behavior is the goal-directed attempt of the human being
to satisfy his needs as experienced in the field as perceived.
All behavior is rational and goal-directed from within
the perceptual field of behavior.
6. Emotion accompanies and in general facilitates goal-directed
behavior. The intensity of emotion is related to the
degree of perceived significance or ego-involvement of the
behavior. Emotions are not disruptive, but facilitate the
striving of an individual to develop. (Blocker, 1966: 32-33)
Writing in the mid-1930s, George Herbert Mead helped to found what
has subsequently become identified as symbolic interactionism. The
symbolic interactionist perspective of the socialization process
maintains that the individual, entering into a social relationship, views
both himself and his actions as though they were objects; that is, he
views them as he perceives others view them. Thus, in essence, the
individual becomes an ". .. object to himself . by taking the
attitudes of other individuals toward himself within a social environment"
(Mead, 1934: 138). Clarifying this statement, Mead (1934: 158)
mentions that there are two general stages of self-development. In
the beginning, the individual's self is a reflection of his perceptions
of how his significant others view him as a person. At the second
stage, the individual incorporates his perceptions of how his generalized
others and other social groups view him.
Charles Horton Cooley, also one of the founders of symbolic
interactionism, parallels many of Mead's ideas. Cooley, for example,
proposed his theory of the "looking-glass self" in which there are
three elements. First is the individual's imagination of his appearance
to others. Second is his imagination of others' judgments of that
appearance. These are then followed by ". . some sort of feeling,
such as pride or mortification" within the individual (Manis and
Meltzer, 1972: 231).
The above discussion has relevance to the understanding of the
acquisition of phobia by outlining the basic principles and concepts
involved In the socialization process. As noted earlier, behaviorists
do allow for the possibility that phobic behavior can be learned through
continued interaction with phobic individuals.
Socialization as a partial explanation of the etiology of phobia. The
brief excursion into the principles underlying the socialization process,
above, has been intended to accomplish two objectives. The first was to
illustrate the means by which behavior patterns are learned. The second
objective was to introduce the following statements, based on
Sutherland's (1970: 75-77) nine assertions, concerned with the dynamics
of the socialization process in the formation of phobic neuroses:
S 1. Phobic behavior can be learned;
2. Phobic behavior can be learned in interaction with others
during a communication process;
3. The principal part of the learning of phobic behavior
occurs within intimate personal groups;
4. The greater the degree of identification of the individual
with those he interacts with, the greater will be their
influence on the individual's behavior patterns;
5. When phobic behavior is learned, the learning includes
(a) the object of fear, (b) avoidance patterns, (c) coping
mechanisms, and may include (d) convenient rationalizations
for the avoidance behavior; and
6. The process of learning phobic behavior by association and
identification with phobic and nonphobic patterns involves
all of the mechanisms that are involved in any other learning.
As noted earlier, behaviorists do allow for the vicarious
learning of phobias (cf. Eysenck and Rachman, 1965; Rachman, 1968;
Marks, 1970b), and there is ample experimental evidence for the process
of fear transmission through modeling, both in animals and in man
(Marks, 1969: 65; Bandura, 1965; Bandura and Rosenthal, 1966). The
dynamics of the process may be illustrated in the following manner.
Suppose a parent has a phobia. During the socialization process, the
child incorporates, generally speaking, his parent's attitudes as his
own; one of the attitudes he acquires may be that snakes are to be
avoided at all costs. Consequently, the child develops a phobic reaction
to snakes. Further, the child's avoidance behavior reinforces that of
his parent. In contrast, should the child manage to avoid being
indoctrinated, his nonphobic behavior may have a mitigating influence
on his parent's fear reaction.
Psychoanalytic versus behavioristic: Isaac Marks
Isaac M. Marks, a British psychiatrist and perhaps the foremost
authority on phobias, represents the most comprehensive perspective
to be found. Essentially, Marks combines the psychoanalytic, behavioristic,
and genetic approaches into one. The discussion that follows is meant
to sum up the section of this dissertation concerned with the dynamics
of the etiology of phobia and to provide a concise reformulation of the
basic viewpoints' most elementary and important concepts and how they
are theoretically interrelated.
Phylogenetics. Concerning the impact of genetic constitutions
on the tendencies of persons to acquire phobias, Marks states that,
although phylogenetic influences shape the form of phobias in man as
a species, there is no firm evidence that genetic inheritance plays
a significant part in the development of phobic states in particular
patients (Marks, 1969; 79; 1970b: 656). It is .undeniable, however, that
certain stimuli have prepotency over others as triggers for phobias,
given a similar frequency of exposure to all stimuli (Marks, 1969: 100).
Furthermore, the age of an individual has an influence, due to the
effects of psychological maturation and frequency of exposure to given
stimuli, on the types of phobia acquired. Genetic endowment also influences
the timidity of a patient and this, in turn, affects a patient's liability
to develop phobia. Phobias, he says, appear to be more common among
persons who are shy, timid, immature, and introverted. The possession
of such personality traits, however, constitutes neither a necessary
nor a sufficient condition for the formation of phobias, but clinical
experience does appear to substantiate their covariation (Marks, 1969:
100; 1970b: 657).
Psychoanalytic perspective. Marks also allows for both the
"displacement of anxiety" and the "defense against threatening impulses"
as outlined earlier. Further reiteration is deemed unnecessary at
this point except to repeat that some phobias may have their roots in
subconscious conflicts resulting from unsatisfactory interpersonal
Learning perspective. Trauma and stress are believed by Marks
to have a direct influence on phobia acquisition. After an exposure to
a traumatic event, the individual may associate the event with some
neutral stimulus present at the time or with the active stimulus itself.
The same effect can also be observed among people exposed to psychological
trauma or a prolonged period of stress.
Vicarious learning, or modeling, also has a direct influence on
phobias and can be called, without sacrificing Marks' intent, phobias
produced during the socialization process or through prolonged interaction
with phobic persons. For example, a phobic parent can, quite
unconsciously, socialize his child into fearing any given object.
Furthermore, once the child acquires the phobic behavior, the effect
is one of reinforcing phobic responses in the parent (Marks, 1970b: 663).
Vicarious learning is especially important and significant in producing
minor and persistent fears, that is, fears which do not substantially
reduce the individual's ability to function adequately in normal social
Other family influences are also thought to contribute to
phobia formation, especially when one considers fears of death or
sickness. Marks (1970b: 663) attributes this to instances where
threats of illness or death are used by significant others as
punishment for misbehavior.
In summation of Marks. In discussing the relative strengths
and weaknesses of the behaviorist and psychoanalytic perspectives,
Marks (1969: 101) offers the following comments:
Learning theory is . useful in accounting for the
onset, spread and course of the simpler phobias. It also
helps the understanding of some features of more complex
phobias. Learning theory does not account for the association
of certain phobias with other psychiatric disorders, nor does
It pay sufficient attention to their symbolic associations.
The response of phobias to desensitisation is partly a process
of counter-conditioning, but no conditioning or learning
paradigm explains those few cases which remit after abreaction.
Phobias may generalise not only by simple sensory
associations but also by symbolic mediational cues of special
significance to the patient. This type of spread has been
stressed by psychoanalytic writers. These writers have further
suggested that phobias are the manifest symbolic facade which
has been displaced from other hidden sources of anxiety.
Criteria for the validity of symbolism are often too loose
to be convincing. Symbolic material in a phobic patient may
be a secondary effect of the phobia itself or relevant to
problems other than the phobia. When relief occurs after
abreaction of symbolic material, this does not necessarily prove
the importance of that material in the genesis of the phobia.
Psychoanalytic writings have usefully shown that emotional
factors help to maintain phobias after they have developed.
The focus of this dissertation now shifts to a review of the major
epidemiological findings presented in the literature.
Review of Epidemiological Studies
In 1969, Agras et al. estimated the total prevalence rate to be
76.9 per thousand while Langner and Michael (1963), in their study of
life stress and mental health, estimated the general rate to be 260
per thousand. Warheit et al. (1973: 19) have reported that 15.2 percent
of a random sample scored "high" on a ten-item phobia scale.
At the conclusion of their article, Agras et al. (1969: 154)
state that "most" of the population is affected by "common fears"
while mild phobias affect a significant but relatively small proportion
of the population, and severely disabling phobias occur even less
frequently. Lemkau et al. (1942) found for severe phobias a
prevalence rate of 0.5 per thousand, as did Hollingshead and Redlich
Of the cases undergoing psychiatric treatment, Hollingshead
and Redlich (1958) have estimated that phobias account for 20 percent.
Marks (1970b: 656) estimates that phobic states form between 0.5
percent and 2.8 percent of the psychiatric patients in both the United
States and Great Britain for which the phobia is the major complaint,
but that phobic symptomatology affects from 20 percent to as much as
44 percent of all psychiatric patients.
Age, Sex, and Race Factors
It has been established that different phobias have varying
prevalence rates when controls are implemented for the subject's sex.
Macfarlane, Allen, and Honzik (1954) have observed that before puberty,
both sexes appear equally liable to acquire most fears. After puberty,
however, women tend to be characterized by higher phobia rates than
do men (Marks and Gelder, 1966; Agras et al., 1969; Warheit et al.,
1973). The preponderance of women is less marked, however, for
"social" phobias, according to Marks and Gelder (1966), and for
isolated phobias such as fear of darkness, heights, thunder, and
accidents (Marks, 1969: 75).
As mentioned earlier, the age of an individual also influences
the kind of phobia he is likely to acquire. This is due primarily
to the effect of maturation and frequency of exposure to the stimuli.
Sudden noise or movement triggers fear in young infants, fear of
strangers is common in older children, fear of animals begins in
pre-school children, and fear of open places or social situations
usually start after puberty (Marks, 1969: 100). Warheit et al.
(1973: 18-20) confirm that females score significantly higher on a
ten-item phobia scale than do men, but mean scores do not vary
significantly with age; the data, however, exclude as subjects
children under the age of 18 years.
Phobia rates by race have not been reported in the literature
except by Warheit et al. (1973: 18-20). They found that, while 15.2
percent of the subjects scored "high," about three times as many
blacks as whites scored "high." Analyzing the data with multiple
regression, the investigators determined that the variables sex, race,
socioeconomic status, and age account for almost 18 percent of the
variance in phobia scores. In presenting the results, the
The B of 3.76 for femaleness, the most significant
variable, corresponds to an increase in the Phobia scale
score of nearly 4 points for being a female. The B of 3.88
for blackness corresponds to an increase of nearly 4
points in the total score for being black. The B of
-0.041 for SES, corresponds to a decrease of about 0.8
points in the Phobia scale for every quintile increase
in SES. Age was not statistically significant. (Warheit
et al., 1973: 20)
In the treatment of phobias, two main strategies have generally
been followed; considerably less reliance has been placed on two
other strategems. The latter shall be described first.
Practical Measures to Diminish Handicap
Julier (1973: 57) recommends that some agoraphobic patients
may be helped to function in spite of their phobia by their using
a stick as they walk or a dog to accompany then; they also appear
to be less anxious on cloudy days and at night, so sunglasses on
bright days may help. Julier also suggests they consider riding
bicycles instead of walking and to tell a number of friends about
their phobia to establish a series of "safe" places to seek refuge
when traveling essential routes. Agoraphobics in London are also
encouraged to explore the benefits in "The Open Door", an association
for agoraphobics. While these practical measures receive little
space in the literature, they do appear promising, especially for the
less severe phobic neuroses, and should be more systematically and
Minor tranquilizers have proved very useful in controlling
anxiety symptoms of phobic patients. Although rarely used as the
sole treatment procedure, chemotherapy is often used in conjunction
with both psychotherapy and behavior therapy (Julier, 1973: 57).
Based on the premise that phobias symbolically represent more
basic anxiety states, psychotherapy focuses on helping the patient
understand his fear and learn more effective techniques for coping.
However, this focus on development of understanding or insight does
not appear very effective. Salzman (1968a: 464-465), a psychoanalytically
oriented psychiatrist, has remarked:
It has been known for some time that understanding,
alone, is ineffectual in resolving the phobic state. It
is commonplace that, while the patient may have adequate
insight into the origin, symbolism, and function of his
phobia, he is still unable to risk the initial venture
into the heretofore out-of-bounds area of living.
Behavior Therapy: Systematic Desensitization
The second major strategem involves behavior-modification
techniques. The underlying mechanism believed to be operational in
all the various behavior-modification techniques is continued exposure
of the phobic patient to the phobic situation until he gets used to
it. The exposure to the aversive situation can be either to internal
stimuli, such as fantasy, or to the actual phobic stimulus.
The original explanation for the action of systematic desensitization
was that it acts by reciprocal inhibition (Wolpe, 1958). According to
this theory, the active ingredient of treatment was the neutralization
of anxiety (usually the imagining of phobic scenes) by an antagonistic
response such as muscular relaxation or assertive responses (Marks, 1973:
9), Recent articles, however, have cast some doubt on reciprocal
inhibition as the key mechanism involved in systematic desensitization
(cf. Benjamin et al., 1972; lIelch and Krapfl, 1970; Mathews, 1969).
In the procedure known as "flooding," the patient is immersed,
totally and suddenly, into contact with the phobic stimuli (Watson
and Marks, 1971; Watson, Gaind and Marks, 1971). Modeling therapy,
as indicated in the works of Bandura et al. (1969) and Ritter (1969),
has shown that when the patient watches a model coping with the phobic
situation he becomes more able himself to cope.
Flooding and modeling therapies are just two of the many
conditioning techniques used by clinicians in assisting patients to
overcome their fears. The only thing they all have in common is the
mechanism the procedures act through. It is by no means clear at
present which of the techniques is best adapted to clinical utilization,
and the debate is voluminous in the literature (larks, 1973: 10-11).
In short, behavior therapy operates under the following premise:
since learned behavior patterns tend to weaken and disappear over time
if they are not reinforced, often the simplest way to eliminate a
maladaptive pattern is to remove the reinforcement for it (Coleman and
Broen, 1972: 668). Behavior that is being negatively reinforced, such
as phobias, is harder to deal with, since the patient becomes anxious
and withdraws at the first sign of the aversive stimuli and never gets
a chance to discover whether the aversive consequences he fears are
still in operation. His avoidance is anxiety-reducing and hence is
itself reinforced. The method of desensitization is aimed at teaching
the patient to emit a response which is inconsistent with anxiety while
in the presence, either real or imagined, of the anxiety-producing
stimulus (Coleman and Broen, 1972: 669).
The literature in the field on phobias and phobia acquisition
theory and research is staggering if one approaches the topic from
a psychological or psychiatric perspective. In the preceding pages
a brief but adequate review of the major principles of those perspectives
has been presented. In the ensuing chapter, a sociological view,
which has been conspicuously lacking in the literature, will be
Current and past theories of phobia acquisition have, without
exception, been individualistic explanations. The psychoanalytic
and behavioristic perspectives appear to be theoretically sound in
their attempts to account for the formation of fear among specific
individuals. Both models recognize that phobias may be acquired in
two basic situations: as the result of exposure to a traumatic event
and as the result of exposure to a prolonged period of anxiety-provoking
stress. In the first instance, the phobia acquired is likely to be
situation-specific, such as fear of snakes, school, or lightning, and
is perhaps most easily accounted for by reference to either learning
theory (cf. Watson and Rayner, 1920; Jones, 1924; Eysenck and
Rachman, 1965; Eysenck, 1960a; Wolpe, 1958; Rachman, 1968) or the
analytic perspectives (cf. Coleman and Broen, 1972: 228-229; Arieti,
1961; Salzman, 1968; Lief, 1968). The second instance, such as
agorophobia, appears best understood when viewed from Marks' comprehensive
perspective (Marks, 1966; 1969; 1970a; 1970b; 1973), with its inclusion
of the importance of constitutional and personality bases in predisposing
the individual with an affinity for acquiring phobia.
In the subsequent pages of this dissertation, a sociological
perspective of fear formation will be presented. Specifically, it will
be shown how cultural influences may predispose some groups of persons
to be characterized by higher rates of phobias than other groups.
Before proceeding, however, it is necessary to indicate a
change in the operational definition of phobia. In the psychiatric
literature, phobia has been used to refer only to a persistent,
irrational fear of an object or event characterized by some loss of
function when exposed to the aversive stimuli. For our purposes,
however, the concepts, phobia, fear, and phobia neurosis, will be
used interchangeably to indicate only a persistent fear of an object
or event. Loss of function and irrationality are not considered
to be requisites for defining and identifying phobia.
It is necessary to make this distinction because, though the
different components of fear are-congruent, they are related imperfectly
with one another (Marks, 1969: 5). Lang (1966) has shown that the
subjective and behavioral aspects of fear are separable. In his
study, Lang found that, whereas 20 percent of his subjects reported
a fear of snakes on a questionnaire, only one to two percent actually
avoided the snakes when tested. Behavioral and subjective indices
of fear were correlated, however (r = 0.40; n = 23; p / .05).
Cultural Influences and the Impact of Differential Life Chances
The discussion which ensues has been sectioned into subject
areas which correspond to the major sociological variables as they are
generally reported in the literature. Before proceeding, however, it
would appear desirable to reiterate at this time the direction that the
discussion will pursue.
The assumption is made, on the basis of the studies presented
in the previous chapter, that phobias can be the product of socialization
and exposure to varying cultural climates. The main contention in
subsequent pages is that groups of people, because of different social
experiences, are differentially vulnerable to acquire phobia.
Consequently, the dissertation will focus on the structural contingencies
influencing the acquisition of phobia rather than on the processual
aspects of the acquisition.
Racial and Ethnic Influences
To the extent that members of one particular race or ethnic
group, however membership in such groups may be defined, constitute
an aggregate of persons exposed to similar life chances, such as
discrimination and prejudice, those life chances are expected to be
reflected in the cultural traits of the aggregate. Thus, one might
expect to observe high rates of phobias among Negroes in the United
States since they are generally considered to suffer from social
oppression and, consequently, may place greater reliance on supernatural
phenomena to explain their social situations rather than attribute
their positions to personal abilities. To the extent that a group is
characterized by feelings of social determinism, one may anticipate
that that group will be more consciously aware of supernatural forces
determining their social positions. As a result, these groups should
have higher rates of phobias regarding such phenomena, even after the
implementation of controls for the subjects' socioeconomic status
Similarly, one would expect to observe high phobia rates among
groups of persons which are more frequently exposed to specific dangers
than among groups which are relatively isolated from those dangers.
For example, in the United States, Negroes have historically been an
oppressed people, and it has been well documented in criminology
literature that they have consistently and significantly higher rates
of violent crimes committed against them and are themselves more often
convicted of perpetrating violent crimes than is characteristic of
Caucasians. Sutherland and Cressey (1970: 136-139) note that the general
crime rate of Negroes exceeds the rate among Caucasians, with the former
having arrest rates three to four times higher than the latter; the
arrest rate in 1965 for felonies was four times higher for Negroes than
Caucasians; and the rate of commitment of Negroes to state and Federal
prisons is about six times the rate for Caucasians. Johnson (1968: 78)
remarks that Negroes rank "perceptibly higher" than Caucasians in
relative size of rates for ". . gambling,aggravated assault, weapons,
and robbery. Whites rank perceptibly higher for automobile theft,
driving while intoxicated, embezzlement and fraud, forgery and
counterfeiting, and sex offenses other than rape." The ranking of other
offenses within each race are similar, however. Clinard (1974: 297-298)
observes that in 1971, when Negroes comprised 17 percent of the total
U.S. population, they accounted for 62.2 percent of all arrests for
criminal homicide. Similar statistics can be quoted by citing Wolfgang
(1958), Pokornoy (1965), Voss and Hepburn (1968), Hepburn and Voss (1970),
Bensing and Schroeder (1960), Reckless (1967), and Cohen and Short (1971).
Official statistics, however, are not necessarily accurate measures of
true crime rates (cf. Clinard, 1974; Shulman, 1966; Wolfgang, 1963;
Sutherland and Cressey, 1970).
Accepting that Negroes have been historically oppressed, and
accepting for the moment that crime statistics do, in fact, reflect
real differences in types of crimes committed, one might expect Negroes
to be characterized by higher rates of self-reported phobic neurosis
than are Caucasians. It stands to reason that, if the black subculture
is more prone to violence than the white subculture, one should expect
to observe a greater prevalence of fear of violence among Negroes than
In addition to the above, the Negro is characterized by a
shorter life expectancy than is the Caucasian, is more likely to become
seriously ill, has a higher infant mortality rate, and also has a higher
maternal mortality rate (U.S. Bureau of the Census, 1974: 58-64).
These facts, combined with those already mentioned, in effect condition
Negroes to be fearful of such things as bodily harm, strangers, and
illness. In short, Negroes constitute a population of high risk in
terms of physical injury, morbidity, and mortality. One would expect
to observe, therefore, a greater prevalence of self-reported fear of
such occurrences and situations among Negroes than among Caucasians
simply because of the different social realities involved for the two
races. These fears do not, of course, constitute a phobia according
to the traditional criteria because Negroes have, in fact, perceived
reality accurately. It is for this reason that the operational
definition of phobia for this dissertation does not include the concept
Dependency Needs, Sex Roles, and Phobias
A second factor which could conceivably have great impact on
understanding gross rates of phobic neurosis involves different
conceptions of behavior appropriate for each sex. Obviously, distinct
behavior patterns do exist in the United States by sex with masculinity
being equated with fearlessness and femininity with dependency. That
this belief has traditionally been so pervasive is undeniable. The
female sex role has been one of passivity and almost total dependency
on her spouse, fiance, or any nearby male. Examples are so abundant
that further discussion is deemed unnecessary.
Given that the traditional female sex role is one characterized
by helplessness and dependency, then it follows that women should be
characterized by higher rates of phobic neuroses than are men. And,
as reported in an earlier chapter, studies have shown that phobia
rates do vary by sex (cf. Marks, 1969, 1970a, 1970b; Julier, 1973;
Warheit et al., 1973). Further, Alexandra Symonds, in her article
"Phobias After Marriage: Women's Declaration of Dependence," addresses
this very point by suggesting that marriage may cause women to become
more dependent than when they were single which, in turn, may lead to
an increased affinity to acquire irrational fears. In addition, this
may be especially true of professional women who give up their careers
at marriage (Symonds, 1971: 145, 150). In commenting on her theory,
Symonds (1971: 151) states:
The traditional explanation for phobias is related
to the fear of loss of control. Freud felt that it was loss
of control of sexual or aggressive impulses. Others, such
as Leon Salzman . broadens it to be a fear of loss of
control of any impulses that would be considered a threat
to the integrity of the personality. . This would
include tender impulses, power drives, needs for
detachment or closeness, etc. The fear of humiliation
Is closely tied up to the anxiety involved.
I would like to add another dimension to the
understanding of the function which these phobias played
in this type of patient. These women were actually
afraid to be in control. They feared the consequences
of taking their life into their own hands, of setting
their own direction (as driving a car), of movement on
their own, of exploring, of enjoying, of discovering.
Age and Retirement Influences on
the Acquisition of Phobias
In a similar manner, the ages of the subjects may be expected
to influence a particular cohort's predisposition for acquiring phobias.
If the assumption is made that education will reduce an individual's
tendencies to fear the unknown, then one would expect to observe
among small children high phobia rates. As the children progress
through the various levels of schooling, the phobia rate should decrease.
Once the individual completes his formal education, however, a valley
should be reached and the individual characterized by his lowest level
of phobic neurosis since infancy. Furthermore, this should continue
at least until he reaches retirement. After retirement, the individual
may begin devoting more of his time to the ponderment of the existence
of God and what death will bring. Thus, in anticipation of death and
with the increasing recognition of his own mortality, the individual
may become more and more fearful of becoming ill, of death, and of
the supernatural and its symbols.
Type of Residence and Phobic Neurosis
The type of community in which an individual resides, for example,
rural-urban, might also be expected to contribute to one's affinity
for acquiring phobic neurosis. All other things being equal, one
would expect urbanites to be characterized by high phobia rates relative
to rural residents. This would seem plausible for two reasons:
first, the more complicated the culture base, the more objects and
social experiences that exist which could serve as foci for fear, and
second, the characteristics of city life (that is, masses of people,
anonymity, the relative lack of primary relationships) would appear
to provide fodder for the relatively easy acquisition of fear.
Religion and Religiosity as a Factor
Influencing the Etiology of Phobia
It is often maintained that the primary function of religion is
to serve as a supraempirical referent to explain those aspects of
nature and social behavior which the rest of culture cannot. In their
attempts to fulfill this function, various religions tend to
differentially emphasize man's relation to the supernatural and his
ability to control his own future. Applying the same logic utilized
in the above discussions on age and race, one would expect to observe
variations in phobia rates between persons belonging to separate faiths
that are characterized by radically different dogmas.
Furthermore, a most important intervening variable would be
the degree of one's commitment to a particular religious faith. For
example, it would appear logical to expect persons who do not believe
in the existence of God to have no fear of His wrath. From a slightly
different perspective, one can state that to be fearful of an object
or situation requires, by definition, that that object or situation
exist in the person's frame of reference; to deny the existence
of God precludes a fear of Him. There does exist, however, the
possibility that a more pervasive fear of the Unknown might supplant
a fear of God should one not believe in His existence.
Educational Attainment and Phobic Behavior
Reference has already been made during the discussion on age
of the expected variation in phobia rates by the level of educational
attainment. As implicitly stated then, the phobia rate should vary
inversely with the level of educational attainment. The reasoning
is simple: by definition, a phobia constitutes an irrational fear.
The educational system has as one of its manifest functions, the
elimination of superstition and blind faith and their replacement with
logical, rational thought and action. If the educational system
is achieving this goal, then as persons are exposed to it they should
become progressively less irrational or nonrational and, therefore,
less inclined to acquire irrational fears.
It has not been the purpose of this chapter to exhaustively
list all possible cultural traits and relate them theoretically to the
acquisition of phobic neuroses. To undertake such a task is clearly
unrealistic. Nevertheless, the importance of cultural influences on
phobia formation appears theoretically plausible and totally consistent
with the literature in the field. The only remaining task is to see how
the data fit the schema. It is to this that the remainder of the
dissertation is devoted.
In 1969, researchers at the University of Florida began an
intensive five year study designed to evaluate the mental health needs
and services in the southeastern United States. Alachua County,
Florida, served as their initial research site. Their study consisted
of three major components: a socioanthropologlcal analysis of the
county; a rates-under-treatment study; and a prevalence study of
social psychiatric impairment (Warheit et al., 1973: 1). This
dissertation uses data collected as part of the prevalence phase of
the study, particularly that related to the distribution of phobic
symptomatology among the population represented by the sample.
Description of Sample Site
Alachua County is located in north central Florida. The county
land area comprises 982 square miles; the 1970 population was 104,764.
Gainesville, the county seat, is the home of the state's largest
university, and has within the last quarter century become one of the
Southeast's leading centers in medicine, research, technology, and
education. The city had a 1970 population of 69,441 persons and is
surrounded by six small towns, a number of rural non-farm settlements,
and much farmland.
The prevalence component of the study began by enumerating the
37,000 households In Alachua County; then 2333 addresses were randomly
selected. To insure randomness of contacting residents within each
household, Kish's (1965) tables were utilized. Comparison of the
major sociodemographic variables of the sample with the 1970 Census
Reports confirmed that the sample was representative of the county's
population (Warheit et al., 1973: 4).
After revision of the interview schedule following an analysis
of pretest results with 322 respondents, an additional 1645 interviews
were gathered. A total of 1967 interviews were obtained, including
those selected for the pretest. This represented a completion rate
of 84.3 percent of the total sample. The refusal rate was 8.8 percent
and an additional 6.9 percent could not be located or interviewed
despite repeated visits by interviewers to the households. All subjects
were interviewed in their homes (Warheit et al., 1973: 4-5).
The interview schedule, consisting of 317 items and 15 scales,
encompassed the following informational items: (1) demographic data
and comprehensive social histories; (2) items on familial and other
interpersonal relations; (3) questions about life satisfaction and
life crisis events; (4) indices about religiosity, racial distance,
anomie, perceptions of social change and social aspirations; (5) a
medical systems review and detailed physical symptomatology data;
(6) a detailed inventory of mental health signs, including the Health
Opinion Survey (HOS) (Leighton et al., 1963); and (7) a series of
items tapping attitudes toward and utilization of health service
facilities (Warheit et al. 1973: 4-5).
The pretest respondents and a. patient group were used for
validation of the research instrument. These subjects were rated
for social psychiatric impairment according to traditional criteria
by three psychiatrists. Their concept of "social psychiatric
impairment" encompassed psychosocial distress along four dimensions:
(1) traditional definitions of psychiatric symptomatology; (2) levels
of functioning; (3) the quality and quantity of interpersonal
relationships; and (4) indices of aspiration and satisfaction (Warheit
et al., 1973; Schwab et al., 1970).
Development of Psychometric Scales
Part of the study was concerned with the development of a
number of scales designed to measure the degree of psychiatric
symptomatology present in the general population. Of the 317 items
on the interview schedule, 98 were selected to be factor analyzed
after first being divided into the subsets of somatic health,
psychosomatic health, phobias, gross psychopathology, worry, and
nervousness (Schwab et al., 1973: 102).
The scales were not constructed for use in the diagnosis of
individual pathology, and were not designed to measure psychiatric
disorder at the construct level, such as described in the Diagnostic
and Statistical Manual of Mental Disorders (Holzer, 1973: 17). Holzer
further states that evidence supporting the validity of the scales is
provided from four sources:
. (1) item content was drawn from the psychiatric
literature; (2) a board psychiatrist has examined their
content and judged it to be appropriate; (3) factor analytic
procedures empirically confirm the grouping of items into
scales and (4) the scales have an acceptable level of internal
consistency as measured by Cronbach's Alpha (Cronbach, 1951).
(Holzer, 1973: 17)
To handle variations in response coding, all scale items were
recorded so that the range for each was from 0 to 4. Three point items,
for example, were coded 0, 2, and 4, and two point items were coded 0
and 4. This was done for two reasons: to ". . permit each item
in a scale to have an equal impact on that scale, even though in
practice the contribution of an individual item is related to the
mean for that item" (Holzer, 1973: 17) and to provide consistent and
easy interpretations of the data across each of the 15 scales, that is,
all scales with the same number of items have identical upper and lower
The procedures used to develop the psychometric scales have been
reported by Warheit et al. (1973) and are as follows:
Step One: Factor analysis and isolation of subsets.
The researchers attempted to factor analyze all 98 items
but because of the heterogeneity of the item content, the
distinctions between underlying patterns were all but
eliminated. Because they were unable to extract groups of
items on an empirical basis, more manageable subsets were
identified on the basis of their content, as reported on
the previous page of this dissertation.
Step Two: Analysis of subsets.
The subsets identified in the previous step were then
analyzed using principal component analysis. Since the
only purpose on this occasion is to reduce the number of
items to be used for prediction, only the common variances,
the communalities, are needed. The specific variance and
error in the unities can then be dropped from the diagonal
cell values leaving a "reduced correlation matrix" and
resulting in factor loadings closer to zero.
This procedure measures those clusters of items which
comprise the subscale, insuring that the subscale measures
only one psychiatric variable and that the internal
consistency of the items in a given subscale be maximized.
As a result, each subscale is homogeneous and unidimensional
(Schwab et al., 1973: 10').
Step Three: Formation of subscales.
As reported earlier, all items were recorded between a range
of 0 and 4 because of variations in response coding, and to
allow each item to have equal impact on the subscale under
consideration. Further, Cronbach's Alpha was calculated
to measure internal consistency for the preliminary subscales
using the method developed by Bohrnstedt (1969). In addition,
the item-subscale or part-whole correlations were calculated
for each item in each subscale.
Items with insufficient part-whole correlations were then
eliminated from the scales. Subscales which had low alpha
coefficients, despite the removal of the individual items
with low item-subscale correlations, were deleted. Some
items were retained for theoretical reasons even, on occasion,
when their part-whole correlations were less than the informal
cutting point for retention of items. Items which factored
in more than one subscale were deleted from those scales in
which they had the least significant relationship.
Step Four: Combination of subscales.
Fourteen subscales were developed through the above procedure.
Using the subscales as variables, a correlation matrix was
generated to provide an index of the internal consistency
to be expected when subscales were combined.
Step Five: Testing of subscales and general construct
level scales among subpopulations.
In order to furnish a detailed understanding of the relationships
between items within scales which would aid in the interpretation
of results as well as indicate the limits of scale reliability,
an analysis of homogeneity of covariance and of individual
item means as a percentage of scale means within selected
subpopulations was performed. (Warheit et al., 1973: 6-9)
In short, Warheit et al. (1973: 9) state that the construction of
subscales ". .. involved a complex interaction of both theoretical
and statistical considerations."
Comprehensive Phobia Scale
The Comprehensive Phobia Scale contains the following ten
"Do you have any strong fears about any of the following:
1. Driving or riding in a car?"
2. Being in crowds?"
4. Fear of bodily harm?"
5. Being hexed, witched, or 'evil eye'?"
6. Any particular animals or insects?"
7. High places?"
8. Closed in places?"
9. Thunder and lightning?"
10. The dark?"
Each subject was asked to respond with a "yes" or "no" for each
of the ten items listed above. For each affirmative response, the
subject was given a score of "4'; for each negative response, the subject
scored "O". The respondent's total scale score was computed by summing
the coded responses. Hence, the Comprehensive Phobia Scale has a range
of 40 points, from 0 to 40 inclusive. The scale was tested for internal
consistency by use of Cronbach's Alpha. The coefficient of scalability
was found to be .65; it ranged from .54 to .68 over the four race-sex
groups (white male, black male, white female, black female) (Warheit
et al., 1973: 18).
On the validity of using subjective reports for fear estimates,
Rachman (1968: 26) has concluded that subjects provide estimates which
do seem to indicate a ". . reliable but gross discrimination between
people who are frightened of a particular stimulus and those who are
fearless." The phobia index does not measure the intensity of
fear present nor the degree of loss of function when the subject is
confronted with the aversive stimuli.
Phobia Scale -- Personal
The Phobia Scale -- Personal is one of the two subscales
comprising the Comprehensive Phobia Scale. It consists of questions 1
through 5 of the Comprehensive Phobia Scale as listed above. Scoring
remains identical to that already outlined; the range of potential
scores, however, varies between 0 and 20.
Phobia Scale.-- Situational
The Phobia Scale -- Situational is the second subscale which,
together with the personal phobia subscale, comprises the Comprehensive
Phobia Scale. Again, the range of scores that an individual might
receive is from 0 to 20. All coding of responses and computations is
identical to that already stated above. The Phobia Scale -- Situational
consists of questions numbered 6 through 10 on the Comprehensive Phobia
At times during the analysis of the data, responses made for
each Item of the phobia scale will be considered individually. Possible
responses are two, "yes" and "no", as previously indicated.
Measurement of Race
During the interview, the interviewer noted the subject's
race and recorded it as white, black, or other. Those reported as
"other" were to be specified on the schedule. Of the total sample,
only 12 respondents fell into the last category, and 9 of those were
identified as Oriental. For the purpose of statistical analysis
and presentation of the data, these 12 individuals are combined with
the white subjects.
Measurement of Subject's Sex
The subject's sex, male or female, was recorded by the
Measurement of Age
Age was ascertained by asking each respondent, "What was your
age on your last birthday?" The answer received was written down by
the interviewer. For this dissertation, the subjects' ages have
been recorded into seven groups: 17 to 24; 25 to 34; 35-to 44;
45 to 54; 55 to 64; 65 to 74; and 75 or over.
Measurement of Educational Attainment
Each subject was asked, "What was the last year of school you
completed?" The following categories were used to code responses: no
formal schooling; less than fourth grade; fifth to eighth grade;
some high school; high school graduate; trade or business school
graduate; college graduate; some post graduate work; and post graduate
Measurement of Father's Educational Attainment
Each subject was asked, "What was the last year of school
completed by your father/stepfather?" The same categories were used
here to code the respondent's answer as were used to code the subject's
level of educational attainment.
Measurement of Marital Status
The subject's marital status at the time of the interview was
ascertained by asking the question, "What is your present marital
status?" Possible responses included: single, married, widowed,
separated, divorced, and common law married. For this dissertation,
the seven persons indicating they had common law marriages are combined
with those who indicated "married."
Measurement of Type of Residence Area
Subjects' places of residence were coded into three categories:
farm, rural-nonfarm, and in-town. In this dissertation, the first
two response categories are combined into a new category, rural, and
the latter is referred to as urban.
Measurement of Socioeconomic Status
The socioeconomic status level was determined by following the
method suggested by the United States Bureau of the Census (1967).
Holzer (1973: 23-24) explains the procedure:
Scores were calculated for the component items,
education . (of the subject and family income),
by computing cumulative percentage distributions of
the component items and by then assigning the midpoint of
the distribution for each category of education or family
income as the score for that category. 'he occupational
categories, for occupation of the chief income recipient,
were ranked on the basis of the combined average level of
scores on Income and education for the given category.
Subsequently, a cumulative percentage distribution for the
ranked occupational categories was obtained and the midpoint
of the distribution for each category was taken as its
score. From these scores, the socioeconomic status
score, a simple average of the individual's education,
family income, and occupation scores was calculated.
The ranks of these variables were based on the sample of 1645
subjects and not on the national census data. Therefore, It was
". .. insured that a respondent's SES score would be determined by
comparisons with the educational, income, and occupational structure
of the specific population to which he belongs" (Warheit et al., 1972: 6).
For ease in the presentation of the data, socioeconomic status has been
recorded into five categories: 0 to 19 (low); 20 to 39; 40 to 59;
60 to 79; and 80 to 100 (high).
A number of additional items will be presented in the ensuing
analysis which are not given space at this time to state how they are
operationalized. Informational items of relevance concerning these
items are furnished In the text of Chapter 5. The items presented
above are included at this time because they are either dependent
variables serving as the focus of analysis or because they are independent
variables used routinely throughout the analysis as controls.
Statistical Techniques Used in the Study
Although strict hypothesis testing will not be reported in
this investigation, the data will be subjected to statistical analysis
and levels of significance will be reported for those who desire such
an interpretation of the data. The major inferential techniques to
be used and their results reported, when appropriate, are the Chi
Square Test of Independence, Kendall's Rank Order Correlation
Coefficient Tauc, gamma for a proportional reduction-in-error
interpretation of ordinal data, one-way analysis of variance, and
stepwise multiple regression.
The procedure to be used when imposing controls on the data
will be to generate separate tables for the categories of the
appropriate variables and compute the various statistics. The data
have been coded, punched on I.B.M. cards and stored on tape. All
statistical analyses will be conducted at the Northeast Florida
Regional Data Center located on the campus of the University of
On the following pages are presented the findings of this
investigation. Results dealing with the Comprehensive Phobia
Scale as the dependent variable will be presented initially,
followed by the Personal Phobia Scale, the Situational Phobia
Scale, and concluding with a look at the individual items comprising
the phobia scales.
Comprehensive Phobia Scale
Among the random sample representing the general population
of Alachua County, Florida, 67.4 percent (n = 1109) of the subjects
responded that they were afraid of at least one item or situation
included on the Comprehensive Phobia Scale (see Table 1). The mean
scale score was 6.6; the median was 4.7.
Of the male respondents, 54.5 percent (n = 401) admitted
having at least one fear included among the ten scale items. Of
the females in the sample, 77.9 percent (n = 708) had a scale score
of 4 or higher. The Chi Square Test of Independence performed on
these data indicates that scores received on the phobia scale were
not independent of sex; the null hypothesis of independence between
scale score and sex is rejected at the .001 level of confidence (see
Table 1. Frequency Distribution of Responses to Comprehensive
Scale Absolute Frequency
Score Frequency (Percent)
0 536 32.6%
4 418 25.4
8 277 16.8
12 164 10.0
16 121 7.4
20 65 4.0
24 32 1.9
28 22 1.3
32 7 0.4
36 2 0.1
40 1 0.1
Mean = 6.6 Standard Error = 0.17
Median = 4.7 Standard Deviation = 7.02
Mode = 0.0 Variance = 49.3
Table 2. Frequency Distribution of Responses to Comprehensive
Phobia Scale by Sex.
Score Male Female
0 45.5% (335) 22.1% (201)
4 26.5 (195) 24.5 (223)
8 13.7 (101) 19.4 (176)
12 6.5 (48) 12.8 (116)
16 4.8 (35) 9.4 ( 86)
20 2.3 (17) 5.3 ( 48)
24 0.4 ( 3) 3.2 ( 29)
28 0.3 ( 2) 2.2 ( 20)
32 0.0 ( 0) 0.8 ( 7)
36 0.0 ( 0) 0.2 ( 2)
40 0.0 ( 0) 0.1 ( 1)
100.0o T(73T TT.o T- o9
Chi Square = 149.5; df = 10; p / .001
From Table 3, one may observe that 62.5 percent (n = 799)
of the white subjects indicated possession of at least one phobia
included among the scale items. Among the black subjects, 84.7
percent (n = 310) had a scale score of 4 or higher. With a Chi
Square of 170.2 (df = 10; p / .001), a statement of independence
between scale score and race could not be made.
Table 3. Frequency Distribution of Responses to Comprehensive
Phobia Scale by Race.
Score White Black
0 37.5% (480) 15.3% ( 56)
4 26.0 (333) 23.2 ( 85)
8 16.7 (214) 17.2 ( 63)
12 9.5 (122) 11.5 ( 42)
16 6.0 ( 77) 12.0 ( 44)
20 2.4 ( 30) 9.6 (35)
24 1.2 ( 15) 4.6 ( 17)
28 0.4 ( 5) 4.6 (17)
32 0.1 ( 1) 1.7 ( 6)
36 0.1 ( 1) 0.3 ( 1)
40 0.1 ( 1) 0.0 ( 0)
700.06% (127-9 100.0% T3765
Chi Square = 170.2; df = 10; p / .001
A positive, but relatively weak, association was found also
for the relationship between scale score and age of the subject
(see Table 4) with a Tauc value of .05 (p / .001) and a gamma coefficient
of .07. Similarly, a relationship was observed between scale score
and the subject's level of educational attainment (see Table 9) with
a Tau of -.17 (p / .001) and a gamma coefficient of -.22.
Table 4. Frequency Distribution of Responses to Comprehensive
Phobia Scale by Age.
Scale Age (Years)
Score 17-24 25-34 35-44 45-54 55-64 65-74 75-91
0 33.4% 37.0" 31.7% 33.1% 27.1% 27.6% 33.2%
(129) (130) ( 82) ( 80) ( 49) ( 42) ( 24)
4 26.9 26.5 27.8 20.7 26.5 24.4 18.1
(104) ( 93) ( 72) ( 50) ( 48) ( 37) ( 13)
8 16.8 16.2 14.7 17.3 16.0 21.7 18.1
( 65) ( 57) ( 38) ( 42) ( 29) ( 33) ( 13)
12 8.5 8.6 11.2 12.4 12.1 7.9 11.1
( 33) ( 30) (29) ( 30) ( 22) ( 12) ( 8)
16 7.0 4.8 6.2 7.4 10.5 10.5 11.1
( 27) ( 17) (16) ( 18) ( 19) (16) ( 8)
20 3.4 3.7 4.6 4.1 4.4 4.6 2.8
( 13) ( 13) (12) ( 10) ( 8) ( 7) ( 2)
24 1.8 2.3 1.1 2.1 1.7 2.6 2.8
( 7) ( 8) ( 3) ( 5) ( 3) ( 4) ( 2)
28 1.6 0.9 0.4 2.1 1.7 0.7 2.8
( 6) ( 3) ( 1) ( 5) ( 3) ( 1) ( 2)
32 0.3 0.0 1.5 0.8 0.0 0.0 0.0
( 1) ( 0) ( 4) ( 2) ( 0) ( 0) ( 0)
36 0.0 0.0 0.8 0.0 0.0 0.0 0.0
( 0) ( 0) ( 2) ( 0) ( 0) ( 0) ( 0)
40 0.3 0.0 0.0 0.0 0.0 0.0 0.0
( 1) ( 0) ( 0) ( 0) ( 0) ( ) ( 0)
1oo.0% 100-70-7 T3= Too1 5 T 100.0% 10077 o7
(386) (351) (259) (242) (181) (152) ( 72)
Tau = .05; p / .001
Gamma = .07
Family income and the subject's socioeconomic status level
were also correlated negatively with scores on the Comprehensive Phobia
Scale (see Table 5 and Table 6, respectively). The respective Tauc
values for the two sets of data were -.14 (p / .001) and -.19 (p / .001);
the gamma coefficients were -.18 and -.24.
Table 5. Frequency Distribution of Responses to Comprehensive
Phobia Scale by Family Income
Scale $0S- $3000- $6000- $10,000- $15,000
Score $2999 $5999 $9999 $14,999 or more
0 25.4% 29.4% 31.0% 38.2% 43.8%
( 71) ( 99) (106) (104) ( 99)
4 21.5 28.0 28.3 24.6 27.0
( 60) ( 94) ( 97) ( 67) ( 61)
8 18.6 14.9 19.6 17.3 14.2
( 52) ( 50) ( 67) ( 47) ( 32)
12 11.8 9.5 9.1 9.6 9.3
( 33) ( 32) ( 31) ( 26) ( 21)
16 9.0 8.6 5.0 7.4 4.4
( 25) ( 29) ( 17) ( 20) ( 10)
20 6.5 4.5 4.4 1.8 0.4
( 18) ( 15) ( 15) ( 5) ( 1)
24 3.2 2.4 1.7 0.7 0.4
( 9) ( 8) ( 6) ( 2) ( 1)
28 1.4 2.1 0.9 0.4 0.4
( 4) ( 7) ( 3) ( 1) ( 1)
32 2.2 0.3 0.0 0.0 0.0
( 6) ( 1) ( 0) ( 0) ( 0)
36 0.4 0.0 0.0 0.0 0.0
( 1) ( 0) ( 0) ( 0) ( 0)
40 0.0 0.3 0.0 0.0 0.0
( 0) ( 1) ( 0) ( 0) ( 0)
lioo.o0 Tdo oo.% 1 i- oo.o%
(279) (336) (342) (272) (226)
Tau = -.14; p / .001
Gamma = -.18
Table 6. Frequency Distribution of Responses to Comprehensive
Phobia Scale by Socioeconomic Status.
Score 0-19 2
12 14.1 1
Tau = -.19; p / .001
Gamma = -.24
Gamma = -.24
Phobia scale scores were dependent upon the subject's marital
status (see Table 7), and there was a positive association between
scale scores and the subject's reported degree of childhood happiness
(see Table 8). Both relationships were significant beyond the .001
Table 7. Frequency Distribution of Responses to Comprehensive
Phobia Scale by Marital Status
= 97.47; df = 40;
= .08; p / .001
Widowed Separated Divorced
p / .001
Table 8. Frequency Distribution of Responses to Comprehensive
Phobia Scale by Reported Degrees of Childhood Happiness.
Scale Childhood Happiness
Score Mostly Happy Mixed Happy Mostly Unhappy
0 35.1% (481) 22.5% ( 40) 15.4% ( 14)
4 25.9 (356) 24.7 ( 44) 17.5 ( 16)
8 16.6 (228) 16.9 ( 30) 19.8 ( 18)
12 9.2 (126) 11.2 ( 20) 19.8 ( 18)
16 6.3 ( 87) 12.9 ( 23) 12.1 ( 11)
20 3.5 ( 48) 6.7 (12) 5.5 ( 5)
24 1.7 ( 23) 2.2 ( 4) 5.5 ( 5)
28 1.2 ( 16) 1.7 ( 3) 3.3 ( 3)
32 0.4 ( 6) 0.6 ( 1) 0.0 ( 0)
36 0.0 ( 0) 0.6 ( 1) 1.1 ( 1)
40 0.1 ( 1) 0.0 ( 0) 0.0 ( 0)
iTW (T1372) 0. TTT T0.-0a (T91)
Tau = .10; p / .001
Gamma = .29
Preliminary Analysis of Results
Comprehensive phobia by sex
In the previous chapter, it was suggested that females would be
characterized by a higher phobia rate than would be males. The
explanation given then is consistent with the data: females do report
having a substantially higher number of phobias than do males. Whereas
more than 70 percent of the males possess less than two phobias, fewer
than 50 percent of the females can make this claim.
Women have generally been socialized into accepting a less
assertive, more dependent role than have men. Being afraid of snakes
or spiders is not a threat to a woman's femininity, but it is commonly
viewed as a threat to a man's masculinity. Consequently, during the
socialization process, males are discouraged by their significant
others from exhibiting fear. In contrast, girls are often rewarded
for exhibiting the same behavior for which boys are countermanded.
As one's commitment to the traditional male sex role increases,
one would expect that person to become less phobic; as one's
commitment to the traditional female sex role increases, one would
expect that person to become more phobic. Consequently, persons
conforming to the traditional male sex role will be expected to have
lower rates of self-reported phobic neurosis than persons conforming
to the traditional female sex role. Unfortunately, the data in the
present study do not allow this to be evaluated since a sex role
identification scale was not administered to the subjects.
Comprehensive phobia by race
.. It was suggested earlier that responses to the Comprehensive
Phobia Scale would vary when controls are implemented for the subject's
race. It was also suggested that one of the reasons for this
hypothesized variance was due to the different social experiences of
the races in the United States. Specifically, since blacks have a
long history of oppression and are continually frustrated by the
encounterment of prejudice and institutionalized discrimination, it is
thought that they would have higher phobia rates than whites. This
is confirmed by the data as reported in Table 3; 61.5 percent of
the blacks, as contrasted to 36.5 percent of the whites, admit having
at least two of the fears included on the scale.
It is unclear at this point, however, whether the difference
observed between the races is real or a function of social class
membership. An attempt to explicate the relationship will be made
Comprehensive phobia by age
The association between scale scores and the subject's age,
although statistically significant beyond the .001 level, is hardly
Impressive (Tau = .05; gamma = .07). Earlier, it was suggested
that age and the rate of self-reported phobic neurosis would
approximate a U-shaped curve: during the early childhood years, the
phobia rate would be relatively high, declining as the children
progress through the various levels of formal education, reaching
a low during the years they are attempting to "get established in
life" and raising a family, and sharply rising as their retirement
The data in this study do not allow an analysis of phobia
rates among persons under age 17 years, as they were excluded from
the sampling frame. From Table 4, however, it does appear that the
slight positive association observed is due to an increase in phobia
responses for the cohorts 55-64 and 65-74 years of age -- the
retirement years where adjustments must be made to new life styles.
The accuracy of the theory on the acquisition of phobic
neurosis by age groups is not established at this point, but additional
analysis will soon be undertaken in an attempt to explicate the
Comprehensive phobia by educational attainment
As noted in Table 9, there is a relatively strong inverse
relationship between phobia rates by educational attainment of the
c L.. .. -
< o I- .r,\ L O .OO m o C- o- oo o0 oo oo
- E *
' C 0 0 N CO LA 0 0 0 0 0 0 Oo
..- -. ..- |o
U c o- coo LC-) !- oL c. oo co oo oo 0 co
V] U -7 )m *C' -J I.
> 0- cj 0r o o 0 o o"
C c o
CL C7 C; C; C7 7O OO OO
0 *IL C0\ C -C A 0000 0 00 00 00 o00
Ef OE co oo
- E- .0 . *.
L C 0- 0 0 0 0 0
to F .
c 0 L p s c' 00 w m 0L 00 00 00 0010
I s t 0 0 0 Co
C) ^- C) -' 30 C OC)\ C')LA -LA I-0 C!- 00 00 PC) 00 o
o; CO 0 CO 0 0 0 0
J 3 U r A 00 : -n 0C 0 0
r u 04 ff% n -
0 b -
c C) .-- re' 00 ,s' ,-- -- 00 00 -ci 00 0o o o
0 0 o r- o C'. 0\ 0 0 o CO 0 0 o o
.0 I C 0o -3) CO C') '.0 0 .3) C ) .0 0
IC 0 *- rI -cs cc r -
I- EvX Vo
subjects. The relationship is statistically significant beyond the
.001 level and conforms to the theoretical explanation offered in
Chapter 3 as well as that offered in the section above on phobia
score by the age of the subject.
It will be recalled that phobias are traditionally defined as
irrational fears, and that one of the functions of the educational
institution in this country is the replacement of blind faith and
nonrational action with rational thought and modes of action. If
the institution is fulfilling this objective, then it is to be
expected that the phobia rate should decrease as the level of formal
education increases. The data confirmed that the relationship
between phobia rates and educational attainment was in the predicted
direction, at the .001 level of significance.
Comprehensive phobia by family income
It would appear reasonable to hypothesize that as one becomes
increasingly financially successful, he would develop a more favorable
self-image and have greater confidence in his abilities. Concomitant
with this would occur the feeling that one is in total control of the
situation, or, at least, that he has greater control over things which
affect him than does someone who has not been financially successful,
and, perhaps, characterized by feelings of fatalism. While the
relationship between phobia rates and fatalistic outlooks on life
will be examined later, a preview might be made by looking at the
relationship between phobia scores and family income.
As reported earlier, the observed relationship is statistically
significant (p / .001) and in the predicted direction with a Tauc
coefficient of -.14 and a gamma coefficient of -.13 (see Table 5).
Consequently, tentative support is provided for the relationship
between fatalistic attitudes and the tendency to have high phobia
Comprehensive phobia by socioeconomic status
Since socioeconomic status is generally viewed as a composite
of several different factors, such as income and education, one would
expect to observe a strong inverse relationship between phobia scores
and socioeconomic status levels, knowing that inverse relationships
exist between phobia scores with income and education. Table 6
provides confirmation. The explanation offered to account for the
relationship is identical to the accounts offered for the observed
relationships between phobia scores with subjects' education levels
and family incomes.
First order controls. Implementing controls to account for the
subject's sex does not substantially alter the observed zero order
correlations (see Table 10). Controlling for race, however, the
gamma coefficient is reduced from -.24 to -.15 for whites and to -.11
for blacks; the Tau coefficient is reduced from -.19 (p L .001)
to -.11 (p / .001) for whites and to -.08 (p / .08) for blacks. It
does appear, therefore, that while race contributes to the observed
zero order correlation between phobia scale scores and socioeconomic
status, it does not account for the entire relationship.
Table 10. Gamma and Tauc Coefficients for the Relationship Between
Comprehensive Phobia Scale Scores and Socioeconomic Status
Controlling for Race, Sex.
N Gamma Tau, p
Zero Order 1645 -.24 -.19 .001
Male 736 -.25 -.17 .001
Female 909 -.23 -.19 .001
White 1279 -.15 -.11 .001
Black 366 -.11 -.08 .01
Comprehensive phobia by marital status
A number of equally plausible hypotheses could be presented
concerning the relationship between phobia scores and the subjects'
present marital statuses. Since no one hypothesis seems to stand out
among the alternatives, a look at the data would appear to be in
From Table 7, it is apparent that persons widowed and separated
have higher rates of phobic neurosis. An explanation consistent with
this observation and with the observation that persons single and
married have the lowest rates of phobic neurosis is that the former
have undergone or are undergoing a relatively traumatic period in their
lives. It does not appear unreasonable to suggest that such persons,
in an emotionally taxing situation, become more neurotic as a result
of their experience. And, of course, as the psychoanalyst would
quickly point out, the emergence of phobic symptomatology during
such life crisis events is not unusual.
A harder relationship to explain is why divorced persons have
a lower phobia rate than either the separated or the widowed. One
explanation accounting for the difference between the divorced and
the separated is that the former are finished with the trauma of
court proceedings. To account for the differences between the
divorced and the widowed, however, is more difficult. It might be
a function of both age and sex, rather than a function of marital
status. For example, the widowed are generally disproportionately
female because of the female's longer life span, and, as noted earlier,
females tend to have higher rates of phobic neurosis than do men.
The divorced, however, are probably more equally divided between men
and women since divorce generally occurs earlier in one's life than
does widowment. Consequently, the inclusion of more males into this
section of the sample would "counterinfluence" the tendency of
females to have higher phobia rates.
Comprehensive phobia by childhood happiness
It would appear reasonable to suggest that the happier a
person's childhood was, the less neurotic that person would be as an
adult. Since possession of phobia is one type of neurotic behavior
pattern, one would expect to observe a higher phobia rate among persons
who perceive their developmental years to be unsatisfactory.
Table 8 provides some data with which to examine this hypothesis.
By asking subjects to describe on a three-point scale ("mostly
happy," "mixed happy," "mostly unhappy") their childhoods, one can
get some indication of the subjects' perceptions of their early
childhood and their satisfaction with them. Kendall's Tauc was
calculated to be .10 (p / .001) for the set of data in Table 8.
Gamma, however, was found to be .29. Thus, using the latter statistic,
one can explain 29 percent of the variance in phobia scores by having
knowledge of the subject's perception of his childhood happiness.
In other words, by knowing whether the subject perceived his childhood
as happy, mixed, or unhappy, one can reduce the number of errors made
in predicting the phobia scale score without such knowledge by 29
First order controls. As seen in Table 11, the implementation
of controls for the subject's sex and race, considered separately,
did not substantially alter the zero order correlation coefficients.
Neither sex nor race appear to be an intervening variable in the
observed relationship between phobia score and childhood happiness.
Second order controls. Controlling for the sex and race of
the subject simultaneously, however, one notes the relationship
between phobia score and childhood happiness essentially dissipates
for black males while the coefficients remain relatively large for
the other three sex-race categories. Consequently, while knowledge
of the childhood happiness of black females and all whites allows
a reduction in the number of errors made in predicting phobia scale
scores from 28 to 33 percent, prediction of black males' phobia
scores can be reduced only 8 percent.
Table 11. Gamma and Tauc Coefficients for the Relationship Between
Comprehensive Phobia Scale Scores and Childhood Happiness
Controlling for Race, Sex, and Race-Sex.
N Gamma Tau, p
Zero Order 1641 .29 .10 .001
Male 734 .25 .07 .005
Female 907 .29 .11 .001
White 1275 .31 .10 .001
Black 366 .23 .10 .005
White Males 581 .28 .07 .01
Black Males 153 .08 .03 N.S.
White Females 694 .30 .11 .001
Black Females 213 .33 .14 .001
Comprehensive phobia by
satisfaction with retirement
The zero order correlation coefficients for the relationship
between phobia scores and satisfaction with retirement were .18 (gamma)
and .12 (Tauc, p / .05). Although the number of subjects is relatively
small (n = 89), a relationship does exist between phobia scores and
satisfaction with retirement when the latter is coded into two groups:
satisfied and unsatisfied.
First order controls. As seen in Table 12, both sex and race
serve as intervening variables. For males, the zero order gamma
Table 12. Gamma and Tauc Coefficients for the Relationship Between
Comprehensive Phobia Scale Scores and Satisfaction with
Retirement Controlling for Race, Sex, and Race-Sex.
N Gamma Tau, p
Zero Order 89 .18 .12 .05
Male 57 .16 .10 N.S.
Female 32 .50 .21 .05
White 65 -.03 -.01 N.S.
Black 24 .33 .46 .;O
White Males 43 .05 .03 N.S.
Black Males 14 .28 .20 N.S.
White Females 22 .41 .06 N.S.
Black Females 10 .76 .52 N.S.
coefficient drops to .16; for females, it rises to .50; for whites,
it drops to -.03; for blacks, it rises to .33. The Tauc values
follow a similar pattern with statistically significant relationships
being observed only for females (p / .05) and for blacks (p / .01).
Second order controls. Attempting to explicate the relationship,
gamma values ranged from .05 to .76 for the four race-sex groups and
Tauc values ranged from .03 to .52. Of the latter statistics, none was
significant at the .05 level due principally to the small numbers
of cases in the four categories.
As suggested earlier in this chapter, it was predicted that
persons who were satisfied with their retirements would have lower
phobia rates than those who were unsatisfied. The zero order
correlation coefficients appear to confirm this. Applying first
order controls, however, showed that the relationship was altered
when taking account of the subject's race; the relationship approached
zero for whites, but remained strong for blacks. Nevertheless,
the small numbers of subjects preclude the possibility of making any
firm statements after the implementation of second order controls.
Comprehensive phobia by place of residence,
attendance at church, and membership in
By dividing the subjects into two groups according to their
place of residence, a relationship was predicted between phobia
scores and rural-urban residence. It was thought that since urban
residence is characterized by a more diversified culture base and
by feelings of anonymity, crowded conditions, and the relative lack
of primary ties, urbanites would have higher scale scores than would
rural residents. As can be seen in Table 13, the zero order
correlation coefficient was near zero and not statistically significant.
There does exist the possibility, however, that urban life in
Gainesville, Florida, does not represent enough of a contrast to the
local rural life to cause a significant difference in reported phobia
It was also thought that persons who attend church regularly
would be characterized by a higher prevalence of phobia because of
their belief in the supernatural. As also seen in Table 13, the
relationship was not significant; both gamma and Tauc were
computed to be -.02.
It was also believed that persons who are more active in club
activities would have a lower rate of phobic neurosis than would
persons who are less active. The data provide support with a gamma
coefficient of .17 and a Tauc coefficient of .13 ( p / .001 ).
Table 13. Gamma and Tauc Coefficients for the Relationships Between
Comprehensive Phobia Scale Scores and the Major
Variables N Gamma Tau, p
Race 1645 .44 .25 .001
Sex 1645 .42 .33 .001
Age 1643 .07 .05 .001
Education 1641 -.22 -.17 .001
Father's Education 1159 -.15 -.11 .001
Childhood Happiness 1641 .29 .10 .001
Residence 1642 -.05 -.03 N.S.
faction 89 .19 .11 N.S.
Church Attendance 1644 -.02 -.02 N.S.
Club Memberships 1642 .17 .13 .001
Leighton Scale 1645 .46 .25 .001
Family Income 1455 -.18 -.19 .001
SES 1645 -.24 -.19 .001
Comprehensive phobia by subjects'
responses to social psychiatric
Since phobias are behavior patterns defined as neurotic, one would
expect phobia scale scores to vary directly with other indices of
psychoneuroticism. In Table 14 are listed 13 items measuring
Table 14. Gamma and Tauc Coefficients for the Relationships Between
Comprehensive Phobia Scale Scores and Social Psychiatric
S Symptomatology Questions.
Variables N Gamma Tau, p
Feels in Good Spirits 1644 .31 .10 .001
Wonders if Things
SWorthwhile 1629 .30 .20 .001
Worry Results in
Physical Strain 1629 .33 .11 .001
Reason 1612 .34 .09 .001
Frequency of Worry/
Nervousness 1611 .32 .22 .001
Things Don't Turn
Out Right 1618 .23 .17 .001
Blames Self if Things
Go Wrong 1550 .08 .06 .001
Frequency of Crying
Spells 1641 .39 .19 .001
Don't Enjoy Things
in Life 1623 .34 .17 .001
Feels Alone and
Helpless 1637 .44 .20 .001
People Don't Care
Today 1622 .37 .17 .001
Feels That Life is
Hopeless 1635 .46 .16 .001
Future Outlook is
Poor 1592 .26 .14 .001
Each item was crosstabulated with the Comprehensive
Phobia Scale scores, and the zero order correlation results reported
in Table 14 along with the levels of statistical significance for the
Tauc values. As expected, scale scores were positively correlated
with each item.
Comprehensive phobia by subjects' responses
to questions tapping fatalistic outlooks on life
The relationship between phobia scores and fatalistic attitudes
is examined and reported in Table 15. As expected, the relationship
is positive and statistically significant beyond the .001 level for
Table 15. Gamma and Tauc Coefficients for the Relationships Between
Comprehensive Phobia Scale Scores and Questions Tapping
Variables N Gamma Taur p
Lot of Average Man
Is Getting Worse 1586 .26 .18 .001
Unfair to Bring Child-
ren into World 1595 .25 .15 .001
One Must Live for
Today 1636 .33 .21 .001
Can't Count on
Anyone 1624 .34 .22 .001
Writing to Officials
Is Useless 1567 .24 .17 .001
People Today Are
Powerless 1600 .30 .19 .001
The explanation for this observation can probably be divided
into two parts. First, fatalistic attitudes occur more often among
persons who exhibit neurotic behaviors; phobic behavior is neurotic;
therefore, one would expect a direct correlation between phobic
neurosis and fatalistic outlooks on life. Second, persons characterized
by social oppression, that is, whose life chances are reduced by
structural arrangements, may place greater reliance on supernatural
events to explain their social situations rather than attribute their
lack of success to personal inadequacies. Such persons would be
expected to have higher rates of fear concerning supernatural phenomena,
and this would be reflected in higher phobia scale scores.
Prediction of Comprehensive Phobia Scale Scores
By Race, Sex, Age, and Socioeconomic Status
To complete the analysis, the data were subjected to stepwise
multiple regression in an attempt to determine the best predictor
variables sequence. Race, sex, age, and socioeconomic status were
the variables used in the procedure.
As seen in Table 16, race accounts for almost 9 percent of the
total variance in phobia scores. With the addition of sex, the next
most important variable in the variable list, 16 percent of the
variance could be explained. Socioeconomic status accounted for
an additional one percent; age contributed almost nothing to the
regression equation. The four variables together explained almost
18 percent of the total variance.
Table 16. Prediction of Comprehensive Phobia Scale Scores by
Race, Sex, Age, and Socioeconomic Status.
Independent Simple Multiple R
Variables R R Sauare Beta
Race .29764 .29764 .08859 .23616
Sex .28053 .40383 .16308 .26285
SES -.25490 .41737 .17420 -.12982
Age .05373 .41865 .17527 -.03443
Constant = -1.28186
Personal Phobia Subscale
Of the subjects in the sample, 31.3 percent (n = 512) responded
that they were afraid of at least one item included on the Personal
Phobia Subscale. The Personal Phobia Subscale, it will be recalled,
consists of the following fears: bodily harm, evil eye, strangers,
crowds, and driving or riding in a car. The mean response was a
scale score of 1.9; both median and modal responses were 0 (see
Table 17. Frequency Distribution of Responses to Personal Phobia
Scale Absolute Frequency
Score Frequency (Percent)
0 1133 68.9%
4 322 19.6
8 127 7.7
12 42 2.5
16 19 1.2
20 2 0.1
Mean = 1.9 Standard Error = 0.08
Median = 0.0 Standard Deviation = 3.40
Mode = 0.0 Variance = 11.5
Of the male subjects, 22.3 percent (n = 164) admitted having
at least one fear included among the five scale items. Of the females
in the sample, 38.3 percent (n = 348) had a scale score of 4 or
higher. The Chi Square Test of Independence performed on these data
indicated that scores received on the phobia scale were not independent
of sex; the null hypothesis of independence between scale score
and sex was rejected at the .001 level of confidence (see Table 18).
From Table 19, one may observe that 27.1 percent (n = 346) of
the white subjects indicated possession of at least one phobia included
among the scale items. Among the black people in the sample, 45.4
Table 18. Frequency Distribution of Responses to
Subscale by Sex.
Score Mal e Female
0 77.7% (572) 61.7% (561)
4 15.1 (111) 23.2 (211)
8 4.7 ( 35) 10.1 ( 92)
12 1.8 ( 13) 3.2 ( 29)
16 0.7 ( 5) 1.6 ( 14)
20 0.0 ( 0) 0.2 ( 2)
100.0% (73) 1500.0% (9097
Chi Square = 51.5; df = 5; p / .001
percent (n = 166) had a scale score of 4 or higher. With a Chi Square
of 90.9 (df = 5; p / .001), a statement of independence between scale
scores and race could not be made.
Table 19. Frequency Distribution of Responses to Personal Phobia
Subscale by Race.
Score Wh i te Black
0 73.0% (933) 54.6% (200)
4 19.2 (246) 20.8 ( 76)
8 5.8 (74) 14.5 (53)
12 1.4 ( 18) 6.5 ( 24)
16 0.5 ( 7) 3.3 (12)
20 0.1 ( 1) 0.3 ( 1)
1oo.o0% (12T"7 700.70 (3"6)
Chi Square = 90.9; df = 5; p / .001
There appears to be no relationship between personal phobia
scores and the subject's age (see Table 20) as both rank order coefficients
were computed to be -.01. An inverse relationship was observed,
Table 20. Frequency Distribution of Responses to Personal Phobia
Subscale by Age.
Scale Age (Years)
Score 17-24 25-34 35-44 45-54 55-64 65-74 75-91
0 64.0% 71.2% 73.7% 73.2% 68.0% 67.8% 56.9%
(247) (250) (191) (177) (123) (103) ( 41)
4 24.6 17.1 15.8 13.6 22.1 23.7 23.6
(95) ( 60) ( 41) ( 33) ( 40) ( 36) ( 17)
8 6.5 8.8 6.6 8.7 7.2 7.2 12.5
(25) ( 31) ( 17) ( 21) ( 13) ( 11) ( 9)
12 3.3 2.9 0.8 3.3 2.2 0.0 5.6
( 13) ( 10) ( 2) ( 8) ( 4) ( 0) ( 4)
16 1.3 0.0 3.1 0.8 0.5 1.3 1.4
( 5) ( 0) ( 8) ( 2) ( 1) ( 2) ( 1)
20 0.3 0.0 0.0 0.4 0.0 0.0 0.0
( 1) ( 0) ( 0) ( 1) ( 0) ( 0) ( 0)
10.0% 1000.0% 1007.0-% 70070% 14007.00 100.0. 1700.04
(386) (351) (259) (242) (181) (152) ( 72)
however, between scale score and the subject's level of educational
attainment (see Table 25) with a Tauc of -.11 (p / .001) and a gamma
coefficient of -.23.
Family income and the subject's socioeconomic status level
were also correlated negatively with scores on the Personal Phobia
Subscale (see Table 21 and Table 22). The respective Tauc values for
the two sets of data are -.12 (p / .001) and -.14 (p .001); the
gamma coefficients are -.27 and -.31.
Personal phobia scale score-s were dependent upon the subject's
marital status (see Table 23); there was also a positive association
between scale score and the subject's reported degree of childhood
happiness (see Table 24). Both relationships were significant beyond
the .001 level.
Table 21. Frequency Distribution of Responses to Personal Phobia
Subscale by Family Income.
Scale $0- $3000- $6000- $10,000- $15,000
Score $2999 $5999 $9999 $14,999 or more
0 54.8% 68.7% 72.2% 74.3% 83.2%
(153) (231) (247) (202) (188)
4 28.3 17.3 17.2 19.5 12.9
(79) ( 58) ( 59) ( 53) ( 29)
8 10.8 8.3 7.9 5.1 3.1
(30) ( 28) ( 27) ( 14) ( 7)
12 2.5 4.2 1.8 1.1 0.4
( 7) ( 14) ( 6) ( 3) ( 1)
16 3.2 1.2 0.9 0.0 0.4
( 9) ( 4) ( 3) ( 0) ( 1)
20 0.4 0.3 0.0 0.0 0.0
( 1) ( 1) ( 0) ( 0) ( 0)
o0.0i 1 o00.0o 0 Too.70 Ta7o.
(279) (336) (342) (272) (226)
Tau = -.12; p / .001
Gamma = -.27
Preliminary Analysis of Results
Personal phobia by sex
The general pattern which emerged during the analysis of responses
made to the Comprehensive Phobia Scale remained when considering only
those items which factored into the Personal Phobia Subscale. Women
continued to be characterized by a higher proportion of phobias than men.
Table 22. Frequency Distribution of Responses to
Subscale by Socioeconomic Status.
Scale Socioeconomic Status
Score 0-19 20-39 40-59 60-79 80-100
0 49.5% 63.8% 69.1% 77.2% 82.1%
(105) (242) (331) (271) (184)
4 27.3 20.1 19.4 18.5 13.4
( 58) ( 76) ( 93) ( 65) ( 30)
8 12.3 11.3 8.4 3.4 2.7
(26) (43) ( 40) ( 12) ( 6)
12 5.2 3.7 2.3 0.9 1.3
( 11) ( 14) ( 11) ( 3) ( 3)
16 5.2 1.1 0.6 0.0 0.5
( 11) ( 4) ( 3) ( 0) ( 1)
20 0.5 0.0 0.2 0.0 0.0
( 1) ( 0) ( 1) ( ) ( 0)
100.0 / 100.07 ooo/ 1T00.o0 T3007
(212) (379) (479) (351) (224)
Tauc = -.14; p / .001
Gamma = -.31
There would appear to be no reason to alter the theory of phobia
acquisition, as presented in earlier pages, on the basis of analysis
of these data.
Personal phobia by race
Once again, the pattern established in the previous section
with the Comprehensive Phobia Scale is repeated: blacks are
disproportionately characterized by a higher prevalence of phobic
neurosis than are whites, and the relationship is statistically
Table 23. Frequency Distribution of Responses to Personal Phobia
Subscale by Marital Status.
Scale Marital Status
Score Single Married Widowed Separated Divorced
0 62.7% 73.1% 61.2% 44.9% 73.5%
(185) (738) (104) (31) ( 75)
4 25.4 17.4 23.5 24.6 14.7
(75) (175) ( 40) ( 17) ( 15)
8 6.8 6.7 10.0 18.8 8.8
(20) ( 68) ( 17) ( 13) ( 9)
12 3.4 1.8 3.5 10.2 1.0
(10) ( 18) ( 6) ( 7) ( 1)
16 1.7 0.9 1.8 1.5 1.0
( 5) ( 9) ( 3) ( 1) ( 1)
20 0.0 0.1 0.0 0.0 1.0
( 0) ( 1) ( 0) ( 0) ( 1)
10I 0. O0. 1007 10-077 1""0"0.0""-
(295) (1009) (170) ( 69) (102)
Chi Square = 66.19; df 20; p / .001
Tauc = .01; N.S..
Gamma = .04
Table 24. Frequency Distribution of Responses to Personal Phobia
Subscale by Reported Degrees of Childhood Happiness.
Scale Childhood Happiness
Score Mostly Happy Mixed Happy Mostly Unhappy
0 72.1% (989) 53.9% ( 96) 49.4% ( 45)
4 18.3 (251) 24.7 ( 44) 28.6 ( 26)
8 6.6 (91) 14.0 ( 25) 12.1 ( 11)
12 2.0 (28) 5.1 ( 9) 5.5 ( 5)
16 0.8 (11) 2.3 ( 4) 4.4 ( 4)
20 0.2 (2) 0.0 0) 0.0 ( 0)
100.0% (1372) 100.0% T(1'7y 100.0% 91)
Tauc .09; p / .001
Gamma = .36