Group Title: investigation of interpersonal understanding
Title: An investigation of interpersonal understanding
CITATION PDF VIEWER THUMBNAILS PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00098303/00001
 Material Information
Title: An investigation of interpersonal understanding between disruptive and nondisruptive adolescents and their parents
Physical Description: xiii, 136 leaves : ; 28cm.
Language: English
Creator: Colbert, Cecilia Bierley, 1943-
Publication Date: 1975
Copyright Date: 1975
 Subjects
Subject: Interpersonal relations   ( lcsh )
Parent and child   ( lcsh )
Adolescent psychology   ( lcsh )
Foundations of Education thesis Ph. D   ( lcsh )
Dissertations, Academic -- Foundations of Education -- UF   ( lcsh )
Genre: bibliography   ( marcgt )
non-fiction   ( marcgt )
 Notes
Thesis: Thesis--University of Florida.
Bibliography: Bibliography: leaves 126-134.
Statement of Responsibility: by Cecilia Bierley Colbert.
General Note: Typescript.
General Note: Vita.
 Record Information
Bibliographic ID: UF00098303
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: alephbibnum - 000163572
oclc - 02754540
notis - AAS9929

Downloads

This item has the following downloads:

investigationofi00colb ( PDF )


Full Text

















AN INVESTIGATION OF INTERPERSONAL UNDERSTANDING BETWEEN
DISRUPTIVE AND NONDISRUPTIVE ADOLESCENTS
AND THEIR PARENTS







By

CECILIA BIERLEY COLBERT


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

1975















ACKNOWLEDGMENTS


It is impossible to adequately thank all of the many

people who had a part in the preparation of this study. Even

though they cannot all be mentioned here, the contributions

of the author's friends and teachers in making this study

possible are deeply appreciated.

Dr. Donald L. Avila, chairman of the author's super-

visory committee, provided abundant support, guidance, and

encouragement. His consistent efforts to help make the author's

doctoral studies a meaningful and pleasant experience are

gratefully acknowledged.

Dr. Walter A. Busby, committee member, invested much

energy and love in helping the author grow professionally and

personally. His commitment to students, his sensitivity to

the feelings and interests of others, and his active efforts

to make humanistic education a reality are highly appreciated.

Dr. Chester E. Tillman, committee member, will be fondly

remembered for his understanding counsel, his warm sense of

humor, and for his patience and genuine concern in helping

the author cope with each phase of her doctoral program. The








many hours he invested in conversations with the author helped

her considerably in expanding her personal and educational

horizons.

Dr. Mary H. McCaulley, committee member, was responsible

for helping the author develop a special appreciation for

Jungian theory and for the Myers-Briggs Type Indicator. Her

warm personal qualities, her ability to provide positive sup-

port at just the right moments, and her influence as an out-

standing teacher have been an inspiration to the author.

Dr. Vynce A. Hines, committee member, made an invaluable

contribution to the statistical conceptualization of this

study. He provided guidance throughout the preparation of

the study and opened the author's perceptual field to the com-

plexities and value of research in education.

Dr. A. Garr Cranney, Dr. Janet Larsen, Mrs. Arden Goettling,

and the author's fellow graduate students at the Reading and

Study Skills Center of the University of Florida have given

special meaning to the author's life and to her graduate studies.

They have shared the author's low points and peak experiences

of the past four years with much love and concern, and their

loyalty and constant faith in the author's ability are deeply

appreciated.

Special thanks are extended to the staff and families of

the P. K. Yonge Laboratory School who made this study possible.

iii








Special recognition is given to Dr. J. B. Hodges, Dr. Catherine

Longstreth, Dr. Hellen Guttinger, Dr. Nancy Baldwin, Miss

Barbara Dalsheimer, and Mr. J. B. Hannum for the time they

invested and for their cooperation in handling the details

related to the selection of subjects for this study.

Mr. Alan Kirby and Mr. Randy Rist provided invaluable

assistance and encouragement during the collection of the data.

Their hard work and dependability are much appreciated.

Mrs. Shirley Guerry and Mr. Richard Kainz of the Univer-

sity of Florida's Typology Laboratory and Mr. William Ingram

offered crucial help with the technical details and computer

programming necessary for the analysis of the data. The author

is most grateful for their cooperation.

The author also extends a special note of appreciation to

her family for the love and encouragement they have offered

throughout her doctoral studies. She further recognizes with

gratitude the enthusiasm, patience, and gentleness of her son,

Scott, and the joy he has given to her life.

Finally, with much love the author recognizes her special

friends, Miss Mary Ganikos, Mr. Alejo Vada, and Dr. and Mrs.

Warren F. Jones, who were always available to share her experi-

ences and who never failed to offer warmth, understanding, and

inspiration at the times when they were needed most.















TABLE OF CONTENTS


Page
ACKNOWLEDGMENTS. ....................................... ii

LIST OF TABLES........................................ viii

LIST OF FIGURES .......................................... ix

ABSTRACT.... .... .. .................................. x

CHAPTER

I. INTRODUCTION.................................... 1

Nature of the Study.......................... 3

Definition of Terms.......................... 5

Significance of the Study..... .............. 6

Statement of Research Hypotheses............. 8

II. REVIEW OF THE LITERATURE...................... 11

Societal Trends in Parent-Child Inter-
actions ................................. 12

Theoretical Implications: Communication
and Empathy................................. 14

Parent-Child Research: Empathy and
Communication............................... 17

Parent-Child Research: Adjustment, Behavior,
Discipline, Birth Order, and Social Class. 24

Introduction to Psychological Type........... 35

Methodological Problems..................... 40








CHAPTER Page

III. DESIGN OF THE STUDY........................... 46

Subjects .................................... 46

Instrumentation.............................. 50

Procedures and Data Collection.............. 53

Data Processing and Analysis................ 56

Operational Definitions for Measures of
Understanding .............................. 57

Restatement of Hypotheses in Operational
Form .................................... 60

IV. RESULTS....................................... 63

Tests for Homogeneity of Variance........... 63

Evaluation of Hypotheses.................... 64

Evaluation of Demographic Data.............. 67

Additional Findings: MBTI Data.............. 71

V. DISCUSSION.................................... 78

Demographic Variables....................... 78

Interpersonal Understanding................. 80

Assumptions of Similarity ................... 82

Additional Observations on MBTI Data......... 84

Suggestions for Further Research............. 89

Summary and Conclusions...................... 91

APPENDICES

A. FAMILY INVENTORY FORM ......................... 96

B. INTRODUCTORY LETTER............................ 98








APPENDICES Page

C. INFORMED CONSENT SHEET......................... 100

D. MODIFIED INSTRUCTIONS FOR THE MYERS-BRIGGS
TYPE INDICATOR (F): CHILD FOR MOTHER........ 102

E. MODIFIED INSTRUCTIONS FOR THE MYERS-BRIGGS
TYPE INDICATOR (F): CHILD FOR FATHER........ 103

F. MODIFIED INSTRUCTIONS FOR THE MYERS-BRIGGS
TYPE INDICATOR (F): PARENTS FOR CHILD....... 104

G. EVALUATION OF DEMOGRAPHIC DATA............... 105

H. MBTI TYPE TABLES FOR SELF-REPORTS AND
PREDICTED TYPES............................... 112

REFERENCES............................................ 126

BIOGRAPHICAL SKETCH................................... 135















LIST OF TABLES


TABLE Page

1 Summary of Analysis of Variance for Measures of
Overall Warranted Assumptions (OWA)............. 65

2 Scheffe Tests of Overall Warranted Assumptions:
Hypothesis 1.................................... 66

3 Summary of Analysis of Variance for Measures of
Overall Assumed Similarity (OAS)................ 68

4 Scheffe' Tests of Overall Assumed Similarity:
Hypothesis 2..................................... 69


viii















LIST OF FIGURES


FIGURE Page

1. Four preferences are scored to arrive at a
person's type.................................. 37

2. Type table: dominant processes................ 39








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

AN INVESTIGATION OF INTERPERSONAL UNDERSTANDING BETWEEN
DISRUPTIVE AND NONDISRUPTIVE ADOLESCENTS
AND THEIR PARENTS

By

Cecilia Bierley Colbert

August, 1975

Chairman: Donald L. Avila
Major Department: Foundations of Education

The purpose of this study was to determine whether there

is a difference in the level of interpersonal understanding

between disruptive and nondisruptive adolescents and their

parents. A response predicting paradigm was employed wherein

disruptive adolescents (N=20) and their parents (N=40) and

nondisruptive adolescents (N=21) and their parents (N=42)

completed the Myers-Briggs Type Indicator (MBTI) to describe

themselves and also to predict each other's responses.

All students participating in the study were enrolled

in either ninth, tenth, eleventh, or twelfth grade at the

P. K. Yonge Laboratory School of the University of Florida.

A disruptive student was operationally defined as a high

school student who was removed from his or her learning

environment two or more times during the 1974-75 school year

by the teacher in charge of the learning environment for the

purpose of disciplinary action.








A nondisruptive student was operationally defined as a

high school student who was never removed from his or her

learning environment during the 1974-75 school year for dis-

ciplinary action and whose behavior was rated by the school

counselors as exemplary and highly facilitative for his or

her own learning and the learning of others.

Two hypotheses were tested by comparing eight combina-

tions of the predicted responses made by adolescents for their

parents and the predicted responses made by parents for their

children. Data were subjected to an analysis of variance

followed by the Scheffe method where significant F's were

found.

Hypothesis 1 was designed to test for differences in

levels of interpersonal understanding as determined by the

number of correct predictions made in the response predicting

tasks. None of the eight comparisons of predicted responses

were found to reach a significant level of difference. An

analysis of overall main effects, however, did show that the

nondisruptive adolescents and their parents as a group did

have a significantly higher (p .017) level of interpersonal

understanding than did the disruptive adolescents and their

parents as a group.

Hypothesis 2 was designed to test for differences in the

degree to which different subjects assume similarity between

xi








themselves and another subject. High frequencies of assumed

similarity were considered to indicate a lack of interpersonal

understanding. None of the eight comparisons of assumptions

of similarity were found to reach a significant level of

difference.

Demographic data and data generated by the responses to

the MBTI were also analyzed for significant differences between

the families of disruptive children and the families of non-

disruptive children.

The following conclusions were reached:

(1) The results of this study do not appear to support

the use of measures of overall warranted assumptions for

distinguishing between disruptive and nondisruptive adoles-

cents and their parents.

(2) The results of this study do not support the use of

measures of overall assumed similarity for distinguishing

between disruptive and nondisruptive adolescents and their

parents.

(3) Personality type dimensions as measured by the MBTI

are potentially useful for explaining differences in inter-

personal understanding between disruptive and nondisruptive

adolescents and their parents.

(4) Demographic data suggest that the children partici-

pating in this study come from a predominantly middle class

xii








population and may be somewhat atypical of a sample of children

that would be drawn from a public high school.

(5) Further research on interpersonal understanding between

disruptive and nondisruptive adolescents and their parents

should make use of large samples and should include various

combinations of variables known to be associated with disruptive

behavior in the school.


xiii














CHAPTER I

INTRODUCTION


Disruptive behavior in the schools is certainly not a

new problem, but educators are still searching for effective

means of helping disruptive students and of reducing the edu-

cational losses caused by disruption in the classroom. Because

of increasing difficulties with disruptive behavior in the

Florida schools, a Governor's Task Force on Disrupted Youth

was commissioned to investigate the problem. The findings of

this Task Force verified the increase in the number of disrup-

tive acts in the schools and also supported the conclusion

that more students than ever before are involved in behavior

that is detrimental to the educational process (Rollin, 1973).

Research studies have identified as many as 87 variables

that are related to disruptive behavior in school children.

Among these are age, sex, race, socioeconomic factors, academic

achievement, IQ, family size, and the marital status and educa-

tional level of parents (Branch, 1974; Feldhusen, 1973; Hagstrom

& Gardner, 1969; New York State Education Department, 1972;

Rollin, 1973; Teitelbaum, 1970). Unfortunately, these are






2

variables that either cannot be changed (e.g., race, age, and

sex) or that fall outside the jurisdiction of the school (e.g.,

parental divorce, family income, and social class).

School personnel have tried a variety of approaches for

handling disruptive behavior. Among these have been suspension

from school, corporal punishment, detention, operant condi-

tioning techniques, and group or individual counseling. Research

on these processes has been inconclusive and in many cases has

shown these methods to have been virtually ineffective (Branch,

1974). Even though there are reports that behavior has been

significantly changed in special classes designed for "social

adjustment" (Becker, 1969; O'Leary & Becker, 1967; Pedrini &

Pedrini, 1972; Walker et al., 1968), student behavior may

become even more disruptive after the child returns to his

regular classroom (Walker et al., 1972).

There is also increasing evidence (Carkhuff, 1971) that

behavioral problems and even psychopathology result when there

is a lack of facilitative conditions in a parent-child relation-

ship. Thus, when low levels of empathic understanding, warmth,

and genuineness exist between a parent and child, the personal

growth of the child and/or parent is likely to be impeded, and

nonproductive behavioral patterns may develop. On the other

hand, it has been shown that when empathic understanding exists

between psychologically adequate people, it allows them "to








communicate with depth and intensity of feeling"(Combs, 1959,

p. 256).

If a relationship could be found between student disrup-

tion and the level of empathic understanding the disruptive

students have with their parents, educators might be able to

devise special programs for helping these parents and their

children improve their interpersonal understanding and their

communication within the family. If levels of empathic under-

standing within a family do affect disruptive behavior, then

the improvement of family interactions might lead to a reduc-

tion of detrimental behavior in the schools.


Nature of the Study


The purpose of this study was to determine whether there

is a difference in the level of interpersonal understanding

between disruptive and nondisruptive adolescents and their

parents. The purpose was implemented by focusing on the rela-

tionship between a child's behavior in school and the level

of his empathic interactions with his parents.

The population involved included both male and female

high school students and both of their natural parents. The

family interactions of students who had engaged in disruptive

behavior at school were contrasted with the family interactions

of students who had never created disciplinary problems in the








classroom and whose behavior at school had been exemplary.

Levels of interpersonal understanding in each family were

measured by the Interpersonal Perception Method, a projective

technique which was designed by Dymond (1949; 1950). This

method is explained in depth in Chapter III.

For the present study, various combinations of predic-

tions made by each child, his mother, and his father on the

Myers-Briggs Type Indicator were examined in terms of two major

facets of interpersonal perception theory. One assumption is

based on the degree to which a subject correctly predicts the

responses of another subject, and the second assumption is

based on the degree to which one subject incorrectly assumes

that another subject's responses were similar to his own.

Thus, the study was designed to answer two basic questions.

First, is there a difference in the degree to which disruptive

adolescents and their parents and nondisruptive adolescents

and their parents are correctly able to predict each other?

And secondly, is there a difference in the degree to which

disruptive adolescents and their parents and nondisruptive

adolescents and their parents assume that others are similar

to themselves and thus show a lack of acceptance of true

interpersonal differences?

The study also included an analysis of certain demo-

graphic variables that previous research has shown to be








associated with the problem of disruptive behavior in the

schools. These included the ages, birth order, religious

preference, and grades failed in school for both the students

and the parents as well as the sex and grade level of the

adolescents. Data on the parents included educational and

income levels, previous marriages, number of children in the

family, and the number of children living in the home.


Definition of Terms


A high school student was operationally defined as a

student enrolled in the ninth, tenth, eleventh, or twelfth

grades of the P. K. Yonge Laboratory School, Gainesville,

Florida, during the 1974-75 school year.

A disruptive student was operationally defined as a high

school student who was removed from his or her learning envi-

ronment two or more times during the 1974-75 school year by

the teacher in charge of the learning environment for the

purpose of disciplinary action.

A nondisruptive student was operationally defined as a

high school student who was never removed from his or her

learning environment during the 1974-75 school year for dis-

ciplinary action and whose behavior was rated by the school

counselors as exemplary and highly facilitative for his or

her own learning and the learning of others.








Interpersonal understanding between subjects was opera-

tionally defined as the number of items on the Myers-Briggs

Type Indicator for which one subject correctly predicted the

self-report of another subject.

Assumption of similarity between subjects was operation-

ally defined as the number of items on the Myers-Briggs Type

Indicator for which one subject correctly or incorrectly pre-

dicted that the self-report of another subject was like his

own self-report.


Significance of the Study


Studies dealing with the interaction effects of parents

with their children have been numerous and have measured many

different single variables and combinations of variables.

Traditionally, such research has focused on preschool and

elementary age children and their relationship with their

mother. The role of the father in the home has been virtually

ignored, and researchers have frequently depended solely on

reports given by the mother. These reports, according to

Yarrow (1963), often based on retrospective opinions have

required responses by the mother that were biased by her own

feelings, values, and practices, that included false but

socially acceptable remarks, and that were based on modal

("what you usually do") rather than current or actual behaviors.








A review of the literature revealed no study that used

the same combination of variables that were used in this

research. For this study, the focus was on the older child

of high school age, and data were collected on the mother and

the father as well as the child. The inclusion of the father

added an important dimension to this study--a dimension recom-

mended as valuable by several researchers in the conclusions

of their own fatherless studies (Longstreth & Rice, 1964;

Medinnus & Curtis, 1963; Osofsky & Oldfield, 1971; Radke, 1946).

Another common trend in parent-child research has been to

study some dimension either in terms of a child's perceptions

of his parents or in terms of a parent's perceptions of the

child. This study combined both approaches and measured the

current perceptions of both the child and the parent.

In the review of the literature, no studies were found

which related disruptive behavior to interpersonal under-

standing between high school students and their parents.

Studies dealing with disruption have been concentrated on

young children, on counseling techniques, on methods for

behavior modification, or on extremely disruptive behaviors

(e.g., rioting, picketing, or physical violence). This study

deals with milder but perhaps more typical types of disruptive

behavior that are encountered regularly in American high

schools. It also attempts to relate disruptive behavior at






8
school to relationships the child experiences at home. Since

a substantial part of the child's time is spent in the home,

school personnel cannot afford to ignore the learning experi-

ences and the developing of interpersonal skills that take

place away from the school environment but that display them-

selves in behavior and interpersonal relationships at school.

Many recent studies have shown positive changes in

parent-child relationships after communication training ses-

sions for parents, counselor-parent interviews in the school,

or home visits by school personnel (Duncan & Fitzgerald, 1969;

Erickson, 1973; Ginsberg, 1971; Hanley, 1974; Larson, 1972;

Lillibridge, 1972; Steam, 1971; Toshiharu, 1972; Wunderlin,

1973). There is no evidence that these methods have been

specifically used to improve the family relationships of dis-

ruptive students. However, if poor family communication or

lack of interpersonal understanding can be established as

variables that are common to students who engage in disruptive

behavior, perhaps some of the already successful training

methods in interpersonal skills can be more readily provided

for this particular population.


Statement of Research Hypotheses


Hypothesis 1. There is no difference between the level

of understanding which disruptive adolescents and their parents








have for each other and the level of understanding which non-

disruptive adolescents and their parents have for each other.

Hypothesis 2. There is no difference between the number

of assumptions of similarity which disruptive adolescents and

their parents make for each other and the number of assumptions

of similarity which nondisruptive adolescents and their parents

make for each other.

Both of the hypotheses for this study are stated in the

null form and will be tested by comparing the following eight

pairs of predicted scores: (a) disruptive adolescents for

mother with nondisruptive adolescents for mother, (b) disrup-

tive adolescents for father with nondisruptive adolescents

for father, (c) mothers of disruptive adolescents for child

with mothers of nondisruptive adolescents for child, (d)

fathers of disruptive adolescents for child with fathers of

nondisruptive adolescents for child, (e) disruptive adoles-

cents for mother with disruptive adolescents for father, (f)

nondisruptive adolescents for mother with nondisruptive adoles-

cents for father, (g) mothers of disruptive adolescents for

child with fathers of disruptive adolescents for child, and

(h) mothers of nondisruptive adolescents for child with

fathers of nondisruptive adolescents for child.

The first hypothesis is designed to answer the following

question: Is there a difference in the degree to which








disruptive adolescents and their parents and nondisruptive

adolescents and their parents are able to show empathic under-

standing for each other?

The second hypothesis is designed to answer the following

question: Is there a difference in the degree to which dis-

ruptive adolescents and their parents and nondisruptive

adolescents and their parents focus on themselves and thus

show a lack of acceptance of true interpersonal differences?

This chapter has included an introduction to the problem

and a discussion of the nature and of the significance of the

problem. Chapter II consists of a review of the literature

related to the present area of investigation. The method and

hypotheses are described in Chapter III; the results of the

study are presented in Chapter IV; and the discussion, con-

clusions, and summary are contained in Chapter V.















CHAPTER II

REVIEW OF THE LITERATURE


The review of the literature will be presented in six

major divisions. Since psychological research on parent-

child relationships during the last four decades has strongly

reflected the societal attitudes about child rearing that

existed at that time, the first section will describe the

evolutionary trends in the philosophy of parent-child inter-

actions. The second section will present the theoretical

implications of empathy and communication as elements in

family interactions. These theories are directly related to

empathic understanding which is the primary focus of this

study, and they will be followed by a third division which

will include research studies on empathy and/or communication

between parents and their children.

As was mentioned in Chapter I, the review of the litera-

ture revealed no studies that employed the same combination

of variables that were used in this study. There have been

investigations, however, dealing with the effects of parent-

child relationships or perceptions on personality adjustment,








behavior, discipline, birth order, and social class that do

have implications for this study. This research will be

described in the fourth section.

The final two presentations will be issues related to

the methodology used in this study. First, the theory of

psychological type on which the Myers-Briggs Type Indicator

is based will be introduced, and secondly, methodological

problems in the use of the Interpersonal Perception Method

will be explained.


Societal Trends in Parent-Child Interactions


Society's view of parent-child relationships tends to

reflect the philosophy of child rearing prevailing at any one

time. This is important to consider in a review of the litera-

ture on parent-child research since studies typically have

been designed to test the contemporary modes of family inter-

action. It is also important to note that the present study

reflects the current emphasis on humanistic psychology and on

the importance of empathy and communication in the parent-

child relationship.

Diana Baumrind (1966), describing major trends in parental

discipline in the United States, identified the ideal home of

the 1940s and 1950s as one which provided unlimited acceptance

of the child's needs. This trend was the result of the







popularity of the psychoanalytic approach to psychology and

led to an era of prolonged breast feeding, self-demand sched-

ules, and milder toilet training. This societal concern for

lenient discipline was reflected in Spock's 1946 book, Common

Sense Book of Baby and Child Care.

The late 1950s and early 1960s showed a reverse in dis-

ciplinary measures. Research had not supported the idea that

restraints on the child produced harmful effects. Spock

revised his book and his point of view in 1957. Baumrind

(1966) quoted him as saying,

Since then (1946) a great change in attitude has
occurred and nowadays there seems to be more chance
of a conscientious parent's getting into trouble
with permissiveness than strictness. (p. 888)

Literature for parents in this period claimed that it was

necessary to reinstitute parental control and to make definite

efforts to inculcate parental ideals and standards into the

child.

This trend toward strictness seems to have been short-

lived, however. In the last ten years, humanistic psychology

has come to the fore and with it an emphasis on the under-

standing and acceptance of individual differences. Baumrind

(1966) concluded that the effects of such strict adult

authority as inhibiting, neurotogenic, and indefensibly

unethical are undeniably supported by the work of such






14

humanistic psychologists as Abraham Maslow (1962), A. S. Neill

(1960), and Carl Rogers (1961).

Adhering to the principles of humanistic psychology, Haim

Ginott (1965; 1969) and Thomas Gordon (1970) have had a marked

effect on parent-child relationships. Both of these men

approach parents with the need for open and honest communica-

tion with their children. Their philosophies of child rearing

are similar and are modeled after Carl Rogers' (1961) ideas

of warmth, empathy, acceptance, and unconditional positive

regard in therapy. From this point of view, the parent is to

be a friend to the child, is to understand the personal meaning

a situation has for the child, and is to resolve problems in

such a way that the solutions are mutually agreeable to both

the parent and the child. This viewpoint which is prevalent

among contemporary child psychologists represents an extreme

change from the previous trend in parent-child relationships

whereby rules were arbitrarily selected and strictly enforced

by the parent alone.


Theoretical Implications: Communication and Empathy


Communication

Bienvenu (1969) and Combs (1959) have defined communica-

tion as the process a person uses to exchange feelings and

meanings in his attempts to better understand the perceptual








field of another. They have also agreed that communication

is a primary element in family interactions and an essential

tool for interpersonal problem solving. Combs (1959) further

stressed that a family's ability to communicate in everyday

interactions is probably the most profound of all influences

on the healthy development of a child's definition of self.

Communication failure as Bienvenu (1969) indicated is

probably the major problem confronted by contemporary parents

and their children. Rogers (1961) took the position, however,

that it is not a hopeless problem and "that breakdowns in

communication can be avoided . by creating a situation in

which each of the different parents come to understand the

other from the other's point of view" (p. 336). This is

essentially what Ginott (1965; 1969) and Gordon (1970) have

attempted to accomplish by teaching parents the skills of

nonevaluative listening and honest communication.


Empathy

The ability to sense the feelings and personal meanings

of another person and to experience these reactions in the

attitude of "being in the other's shoes" is called empathic

understanding (Rogers, 1961; 1971). This process not only

involves understanding another person, but also it involves

understanding the person from his own point of view in the

absence of an evaluation or judgment of him.








Empathy is not considered a therapeutic technique but

rather an interpersonal skill possessed in varying degrees by

everyone (Katz, 1963). It is, however, an essential ingre-

dient for producing change in a therapeutic relationship.

According to Rogers,

. if the therapist provides a relationship
in which he is (a) genuine, internally consistent;
(b) acceptant, prizing the client as a person of
worth; (c) empathically understanding of the
client's private world of feelings and attitudes,
then certain changes occur in the client. Some
of these changes are: the client becomes (a)
more realistic in his self-perceptions; (b) more
confident and self-directing; (c) more positively
valued by himself; (d) less likely to repress
elements of his experience; (e) more mature,
socialized, and adaptive in his behavior; (f)
less upset by stress and quicker to recover from
it; (g) more like the healthy, integrated, well-
functioning person in his personality structure.
(1961, p. 375)

Considering the positive personal changes generated by therapy,

it is important to note that Rogers also stated that the

"therapeutic relationship is only a special instance of

interpersonal relationships in general and that the same law-

fulness governs all such relationships" (1961, p. 2). This

is in agreement with Carkhuff's (1971) observation that

counseling conditions have direct relevance for parent-child

relationships and with Combs' (1959) statement that empathy

"is an important factor in communication and in effective

human relationships" (p. 236).








Parent-Child Research: Empathy and Communication


Empathy

Abraham (1966) investigated how objectively parents can

view their children. In studying parents' perceptions of

their children, he observed that parents assume that they

have an accurate understanding of their children's attitudes,

interests, thoughts, and abilities simply because they have

lived with their children since they were infants. He con-

cluded, however, that the ability to understand children is

a skill which does not come automatically but is learned over

time and in varying degrees by different parents.

Pursuing the idea of differences in the ability to

demonstrate empathic understanding, Lindner (1972) designed

a study very similar to the present study and also based on

Dymond's (1949; 1950) Response Predicting Paradigm (RPP).

His instrument for implementing the RPP was also the Myers-*

Briggs Type Indicator, but his subjects (N=42) were married

couples. Unlike this study, he focused on the relationship

between Jungian type dimensions (measured by the MBTI),

marital happiness (measured by the Locke-Wallace Marital

Adjustment Test), and interpersonal understanding and had

each of his subjects complete the MBTI under three conditions

--a self-report, a prediction of their mate's self-report,








and a prediction of how their mate predicted them. In terms

of type dimensions, he found that introverts were better pre-

dictors of their mates' predictions of them. Significant

results were also found between marital happiness and feelings

of being understood and between marital happiness and inter-

mediate values of personality similarity. There was also a

significant tendency for extraverts to assume that others are

similar to themselves. Thus, like Abraham (1966), Lindner

concluded that different people do possess in differing degrees

the ability to demonstrate empathic understanding for another

person.

Bellante (1970) investigated the possibility of dif-

ferences in empathic ability between lower and middle class

adolescent girls (N=141) and adolescent boys (N=152). His

results showed that the girls were more empathic than the

boys (p .01) and that middle class (but not lower class)

girls were more empathic (p t .01) than boys of the same

socioeconomic level. Without controlling for other variables,

he found that the degree of empathy was related to the level

of self-concept (p .001). On the other hand, he found no

support for a relationship between empathy and the variables

of educational performance, age, scholastic aptitude, birth

order, or size of family.

Two studies (Bangs, 1969; Wakefield, 1966) in which the






19

parents predicted the responses of their adolescent child on

a personality inventory revealed that mothers were more accu-

rate or more empathic than were fathers. The hypotheses tested

by Bangs (1969) and Wakefield (1966) were not substantiated,

but Wakefield discovered that the mothers in his group used

more projection (i.e., assuming similarity between themselves

and their sons and daughters), that the fathers used more

stereotyping, and that both parents were more accurate in

their understanding of the child's problems with finances,

living conditions, and employment than they were with social-

psychological relationships. A third study by Collins et al.

(1975) lends support to these conclusions. Their research was

conducted on 327 adolescents ages 17-19 and both of their

parents. Each adolescent completed the Mooney Problem Check

List for himself, and each parent completed it as they thought

their child would answer. Product-moment correlations revealed

that same-sex parents had greater awareness of their child

than did opposite sex parents and that the highest levels of

empathic understanding were reached by mothers in describing

their daughters. Both parents were ill-informed on their

child's problems with courtship, sex, and marriage but were

more aware of the child's (especially the daughter's) problems

in the home.

Using scores from the Minnesota Multiphasic Personality


1








Inventory (MMPI), Irving (1965) divided his sample of 139

adolescents into two groups of adjusted and maladjusted and

compared the groups on measures of parental empathy. He dis-

covered that both of the parents of the well-adjusted adoles-

cents were in fact more empathic and were perceived by their

children as more empathic. Furthermore, there was a highly

significant difference between the two types of adjustment

for both the child's perception of the empathy of the mother

and the actual empathy of the father.

Similar findings resulted from two studies of normal

adolescents. Siskind (1972), studying 24 father-son pairs,

found that fathers who made warm and empathic affectual

responses to their sons had children with better task perfor-

mance, better self-esteem, and more mature perceptions of

their relationship with their father than did fathers with

low levels of empathy. The sons receiving empathic under-

standing from their father also had significantly better

relationships with their mother. Melton (1971) focused on

parental expressiveness (i.e., sharing love and affection,

understanding and accepting, consoling and comforting, etc.)

as perceived by 261 high school juniors. He found that both

the boys and the girls regardless of their social class saw

their mothers as more expressive than their fathers. Upper

and middle class girls and lower class boys rated fathers






21

higher on expressiveness than did the other adolescents in the

study. Parents with only one child received the highest ratings,

and mothers with part-time employment were rated higher than

mothers who were unemployed or employed full time.


Communication

In examining the interpersonal relationships of mothers,

fathers, sons, and daughters, researchers have frequently

reached the conclusion that girls have better communication

with both of their parents than do boys and that communication

between mothers and daughters is more effective than communica-

tion between fathers and daughters (Beaubien, 1970; Berdie

et al., 1970; Crow, 1956; Larson, 1970). Data collected by

Berdie et al. (1970) and Larson (1970) led them both to con-

clude that the better communication between mothers and daugh-

ters may be the result of a societal pattern in this country

for mothers and daughters to spend more time together and to

participate in more joint activities than do fathers and sons

or fathers and daughters.

Even as high levels of empathy in parent-child relation-

ships are associated with higher levels of personal adjustment

and achievement, so are high levels of communication associated

with positive factors. Berdie et al. (1970) administered four

inventories to 309 adolescents and their parents and found






22

that adolescents (and especially girls) from the families with

the highest levels of communication gave more positive responses

about themselves. Likewise, high parent-child agreement on

communication factors was related to high academic achievement

among the boys.

Beaubien (1970) and Marshall and Miller (1971) studied

affective reactions in terms of family communication styles

The sample for the latter study was very small (N=6 families),

but the investigators attempted to compare families with

normal sons with families with delinquent sons. They concluded

that the communications between delinquent children and their

parents showed a decided lack of understanding of connotative

meanings that was not observed in the communications between

normal children and their parents. Furthermore, the parents

of delinquent sons engaged in unilateral rather than two-way

conversations with their child and frequently misunderstood

the child's statements about emotions. Beaubien (1970)

administered the Adolescent-Parent Checklist to 82 males and

females and both of their parents and scored them according

to categories of adolescent communication action, parent

reaction, adolescent satisfaction, and parent satisfaction.

She concluded that there was a significant tendency for low

adolescent communication when paired with low parent reaction

to result in low adolescent satisfaction.






23

Other studies have examined social variables in relation

to parent-adolescent communication factors. Using the Parent-

Adolescent Communication Inventory (PACI), Love (1970) did a

comprehensive study of 1,578 adolescents on 13 social vari-

ables. He found a significant relationship (p- .05) between

communication levels and a combination of education and occu-

pation of the father, income source, socioeconomic status,

and education of the mother even though the degree of associa-

tion for each of the variables with the total score was quite

low. No significance was found for the variables of age,

grade placement, sex, race, birth order, religion, number of

siblings, or maternal employment status. Love concluded,

however, that all 13 of the variables viewed singly or in

combination were poor predictors of the level of parent-

adolescent communication.

Larson's (1970) study of 296 eleventh grade students in

which the PACI was also used supported no differences in com-

munication patterns between white and black families when the

father's occupational status was controlled, but he did find

a positive association between the father's occupational status

and the child's willingness to disclose to him (but not to the

mother). Beaubien (1970) found that the occupational class of

the father affected only father-daughter communications with

the daughters of fathers with high occupational status scoring








higher on communication than the daughters of fathers with

low occupational status.


Parent-Child Research: Adjustment, Behavior
Discipline, Birth Order, and Social Class


Personality Adjustment

Psychologists have studied the effects of parent-child

relationships and the effects of the adolescent's perceptions

of his parents on the emotional development and the personality

adjustment of the child. In nearly all of the research of

this nature, significant results have been found for some

variable of family interaction.

Reuter (1969) examined the father-son relationship of

late adolescent males and discovered that the availability of

the father as well as a nurturing and nonrejecting attitude

on the part of the father were directly related to the per-

sonality adjustment of the sons. Based on the California

Psychological Inventory and an adjective check list, his data

showed that the most mature boys and those with the most posi-

tive self-concepts had fathers who were highly available,

positively involved, and highly or moderately nurturant.

Murrell (1971), in dividing a group of younger, normal boys

into low, medium, and high groups on the basis of the child's

social acceptance and achievement effort scores, found that








the medium group showed the most stability in family inter-

action patterns. The families of this group were sufficiently

organized to make fast family decisions and had family members

who paid attention to each other, talked to each other, and

shared and communicated their interests. The boys in the

medium group were less "driven" as compared with the boys in

the high group who overcompensated in order to achieve. The

most important aspect of the study was the conclusion that

families do consistently follow stable patterns of interaction.

Children identified as having personality problems have

been shown to have parents who also have problems. In one

study (Adams & Sarason, 1963), high school students and their

parents were given four anxiety scales, and parents with high

anxiety tended to have children with high anxiety. This was

particularly true when the scores for the daughters were com-

pared with scores for the mothers. The mean scores for

children of both sexes were more highly correlated with the

scores of the mothers, and mothers were found to be more

anxious than fathers. Socioeconomic factors also correlated

highly with anxiety scores with fathers holding professional

positions being significantly less anxious than the other

parents. Medinnus and Curtis (1963) also found support for

a hypothesis that stated that there is a significant positive

relationship between a mother's acceptance of herself and a






26

child's self-acceptance. Likewise, in a study of maladjusted

and adjusted children, Peterson et al. (1959) concluded that

the parents of problem children were themselves less well

adjusted and sociable, experienced more disciplinary conten-

tion, and were more autocratic than were the parents of

adjusted children. This study led to the conclusion that

conduct problems were related to an autocratic attitude and

lack of concern on the part of the father and to symptoms of

maladjustment in the mother.

Children identified as having personality problems have

also been shown to have different perceptions of their parents

than do well-adjusted children. Rode (1971) concluded that

alienated adolescents, regardless of sex, age, or intelligence,

see parental behavior in a consistently different way than do

less alienated adolescents. The youths with problems viewed

their fathers as hostile and rejecting and viewed their mothers

ambivalently as either possessive or rejecting. Kemp (1965)

worked with three groups who were divided on the basis of

their scores on the California Test of Personality and the

Mooney Problem Check List into few problems, medium problems,

and many problems. Adolescents with few and many problems

felt that they were better adjusted than their parents, and

the personal adjustment of the youths with few problems was

in fact closer to their self-perceptions than to their parents'

perceptions of them.








Behavior

Bandura and Walters (1959) did a comprehensive study of

family interrelationships associated with adolescent aggres-

sion that has become a classic in parent-child research.

They identified a group of 52 aggressive boys ages 14-17 from

intact homes and whose intelligence was average or above and

compared them to a control group which was matched on as many

variables as possible. Their data were derived from exten-

sive parental interviews and psychological testing, and their

findings were quite similar to other studies which have already

been reviewed.

First, it was found that the fathers in the control group

spent much more time in affectionate interaction with their

sons than did the fathers of the aggressive boys. The latter

showed less warmth and positive response to their children,

and in turn their sons sought their help on few matters and

had little emotional dependence on them. Nearly all of the

families with aggressive children had consistent breaks in

emotional ties between father and son, and the fathers demon-

strated themselves to be hostile and aggressive models.

Patterns of rejection were also prevalent in the mother-

son relationship of aggressive boys but to a lesser degree

than with the fathers. Aggressive boys did show some depen-

dence on their mothers but considerably less than did the








boys in the control group. Mothers in the control group

encouraged their sons to seek help (p .05), used reasoning

with their children (p .001), felt little hostility towards

their sons (p ~ .01), and were less likely (p .05) to punish

their children for seeking help.

Bandura and Walters (1959) concluded that defective parent-

child relationships often reflect the parents' inability to

establish close relationships outside the home and their tend-

ency to exhibit less warmth for and more hostility for their

mate. These parental problems consequently result in conditions

unfavorable for the child's identification with the parents and

for his development of conscience.

Studies of children who are aggressive and who create

discipline problems in the school have born out Bandura and

Walters' (1959) conclusions. Aggressive-disruptive children

were found to receive inappropriate paternal discipline and

inadequate maternal supervision (Feldhusen et al., 1973) and

to receive less love and control from their parents as well

as to show less identification with their parents (Longstreth

and Rice, 1964). Sybouts (1967) added the information that

the frequency and the degree of severity of disruptive behavior

in the schools increased as levels of home disruption were

varied from intact families to families with stepparents to

families that were broken by separation, divorce, or death.








The same family syndromes were identified in another study

(Morrow & Wilson, 1961) which contrasted high-achieving high

school boys with under-achieving boys. In this case, the

parents of high-achievers as opposed to the parents of under-

achievers engaged in more sharing of activities and ideas with

their sons, were more affectionate and approving, and were

less restrictive and severe.


Discipline

Roe (1971) states that the majority of research on parent-

child interactions relate in some way to two global, bi-polar

dimensions. The first of these focuses on the style of

parental discipline ranging from extreme authoritarianism to

extreme permissiveness while the second one deals with parental

attitudes and behavior on a continuum from highly accepting

and loving to highly rejecting and nonaccepting.

Radke (1946) is quoted throughout the literature for her

review of research on parent-child relationships. She begins

with a classic study done in 1939 by Lewin, Lippitt, and White.

In an attempt to show that the home atmosphere has a profound

effect on the socialization of children, they had parents

respond to their children under certain positive or negative

conditions and were able to show within a matter of days

drastic shifts in the children's behavior. In two clinical








studies also reviewed by Radke (1946), Knight in 1933 found

that submissive children were more likely than aggressive

children to come from harmonious homes, and Karlin in 1930

found that in case studies of neurotic children there was a

predominance of marital discord, parental neuroticism, faulty

discipline, and unwholesome parent-child relationships.

Quoting eight studies dealing with children who were

highly jealous of their siblings or who were classified as

delinquent, Radke (1946) shows common personality character-

istics among the parents. These include greater over-solicit-

ousness, neglect, inconsistent disciplinary measures, parental

maladjustment, and the provision of fewer play and social

opportunities for their children.

Watson (1957) did a major work comparing the personality

differences of children with strict parents and children with

permissive parents. He easily found a sample of strict parents

but had much difficulty finding parents he considered to be

genuinely permissive. He quoted a study by Whiting and Child

who had studied 47 cultures and found only two as severe on

the child as the American WASP family and none more severe

than the American WASP family. Watson (1957) compared the

children from these two groups on nine dimensions. On inde-

pendence-dependence he found a marked tendency for children

from permissive homes to be more independent in their behavior






31

outside the home. On the socialization-ego-centrism dimension,

he again found the children of permissive parents to have the

advantage. They were better socialized and more effective in

their cooperation with others. In the area of persistence-

discouragement, the permissively raised children seemed to

make a better effort to solve problems and to continue their

intellectual attacks on problems longer (though not persisting

indefinitely against improbable odds). When studying creativity-

conformity, Watson (1957) found that 33 percent of his sample

of permissively raised children were creative as compared with

2 percent of the children raised with strict standards. There

was also a statistically significant difference on friendliness-

hostility. Children from the strict group were more hostile,

and children from the permissive group showed more positive

feelings toward others. Four dimensions were found to be non-

significant: self-control versus emotional disintegration,

energy versus passivity, security versus anxiety, and happiness

versus sadness.

Jackson (1967) found that mothers suggest more coercive

methods of discipline than do fathers and that mothers vacil-

lated more between mild and severe methods of discipline than

did the fathers. He suggests that mothers may dilute what

they consider to be aggressive, nonfeminine behavior by com-

bining severe and mild approaches to child rearing.






32

In punishment situations, according to Morgan and Gaier

(1957), mothers tend to be more extrapunitive and children

more intropunitive. The experimenters observed a tendency of

people to overestimate the amount of ego-defense in the other

person and found that frustrating obstacles are more important

to the child than to the mother. They also concluded that

children do not as often feel that a solution has been found

for a problem as much as the mother does.

Baumrind (1966) goes beyond the traditional discipline

dichotomy of strict versus permissive discipline to describe

a third prototype, authoritative discipline. Authoritative

discipline requires a verbal give and take between the parent

and the child. The parent directs the child's activities in

a rational manner and shares with the child the reasoning

behind his policies. The parent recognizes the child's inter-

ests and solicits his objections but still uses firm control

with the child when necessary. Baumrind (1973) followed up

her ideas about permissive, authoritarian, and authoritative

parents with two formal research projects. Each time she

concluded that authoritative discipline produced in girls

more purposive, dominant, and achievement-oriented behavior

and produced in boys improvement on all of the indices of

social responsibility used in the studies.

The Berkeley Growth Study (Bayley & Schaefer, 1960a)






33

provides interesting longitudinal data. The children in this

study were normal, white children born in 1928 and 1929 who

were observed and evaluated from birth to adulthood. Descrip-

tions of maternal behavior were converted to objective scores,

and the researchers looked for the maternal characteristics

which stayed consistent over time and which seemed to be the

most influential in the development of the child's personality.

On the love-hostility dimension over a 10-year period, maternal

behaviors had a correlation coefficient of .68 whereas maternal

behaviors related to control-autonomy correlated at .26. There

was a significant difference between education and type of

discipline with the more educated mothers allowing their chil-

dren more autonomy. In comparing the behavior of the child

with that of the mother, it was found that girls were more

consistent in their scores over time than were boys but that

boys' behavior stayed more like their mother's behavior over

time. Girls developed better when treated consistently, and

boys developed better when they received love at an early age

and autonomy at later ages.


Birth Order and Social Class

Two variables frequently analyzed in terms of parent-

child interactions and that may have relevance for the two

groups in this study are birth order and social class. Studies
a


I






34

on birth order have produced conflicting results but have demon-

strated a few characteristics common to particular ordinal

positions within the family. For instance, the first born

child has often been shown to be highly oriented to intellectual

achievement, to be more dependent on the parents, and to be

more likely to affiliate with others (McClure, 1971; Sampson,

1965; Shrader & Leventhal, 1968). This cannot be accepted as

an established fact, however, since data also exist to show

that first borns are more self-directed, less parent-oriented,

and less academically oriented (Barlelt, 1972). Nonetheless,

first borns frequently reach higher levels of academic achieve-

ment, make better grades, and compose a greater percentage of

college populations than do their younger siblings (McClure,

1971).

In spite of the conflict in data, there do seem to be

definite socializing advantages for last born children because

of the role models provided by older siblings and because of

the additional experience of the parents in child rearing

(Lasko, 1954; Sears, 1950). Nonetheless, many researchers

conclude that birth order as a factor in parent-child relation-

ships should be studied only as an interacting variable and

not as a single independent variable (Bradley & Sanborn, 1969;

Douvan & Adelson, 1966; McClure, 1971; Schacter, 1963).

Research over the past twenty-five years has shown middle






35

class parents to be more accepting and more equalitarian in

their relationship to the child than lower class parents

(Bayley & Schaefer, 1960b; Bronfenbrenner, 1958; Mass, 1951;

Rosen, 1959). Two studies, one with a large sample of lower

and middle class boys (Rosen, 1964) and one with a large sample

of high school seniors (Osborn, 1971), showed that adolescents'

perceptions of their parents differed according to the youth's

social class in such a way that middle class parents were

viewed by their children as more competent, more emotionally

secure, and more interested in their child's performance.

Rosen (1964) found the greatest differences between the lower

and middle class child's parental perceptions on the variables

of acceptance and support. Osborn (1971) concluded that the

students in his study had attitudes and aspirations comparable

to the educational achievements of their same-sex parents and

that subsequently the students' perceptions of their parents

affected their efforts to succeed academically.


Introduction to Psychological Type


The Myers-Briggs Type Indicator (MBTI) was developed to

implement the Swiss psychologist Carl Jung's theory of person-

ality type (1923). The theory is based on the idea that

people differ from each other in a systematic and orderly

manner in the ways they perceive and judge the world in which








they live. Perception in this sense is defined as the way

people become aware or take in data about things, events,

people, or ideas, Judgment, on the other hand, refers to the

way people make decisions or come to conclusions about the

things they perceive (Myers, 1962).

According to Jung (1923), psychological types come from

inner predispositions which are evident even in babies and

young children. It is important to note, however, that the

term "type" represents a preference for a way of doing some-

thing and is not a rigid, self-inclusive or binding category

or box into which people are sorted.

This means that a type describes how a person pre-
fers to use his processes of perception and judgment.
While he prefers in most activities to act 'true to
type,' it is possible and often necessary to 'go
against the grain.' (McCaulley & Natter, 1974, p. 96)

The impetus of the theory lies in the assumption that

people who differ in their preferred processes for judgment

and perception will also show differences in their behaviors,

needs, interests, motivations, reactions, etc. Assuming this

to be true, the MBTI purports to measure from a subject's

self-report the direction of his preferences for perception

and judgment. Research studies may then relate this data to

practical situations (e.g., psychotherapy, vocational choice,

interpersonal relationships, learning styles, etc.).

The MBTI consists of four separate indices which are each






37

composed of dichotomous dimensions. Each person is assumed to

have a preference for one dimension from each of the four cate-

gories, and his type score reflects these four preferences.

These preferences which combine into sixteen different psycho-

logical types are described in Figure 1.




E (--Does the person's interest flow mainly to the --- I

EXTRAVERSION outer world inner world of INTROVERSION
of actions, objects and concepts and ideas
persons


S Does the person prefer to perceive --- N

SENSING the immediate, the possibil- INTUITION
real solid facts of ities, meanings and
experience relationships of experience


T-Does the person prefer to make judgments or decisions- F

THINKING objectively and subjectively and FEELING
impersonally, analyzing personally, weighing values
facts and ordering them and the importance of
in terms of cause and choices for oneself and
effect other people


J Does the person prefer to live ) P

JUDGING in a planned, in a flexible, PERCEIVING
orderly way, aiming to spontaneous way, aiming
regulate and control to understand and adapt
events to events


Figure 1. Four preferences are scored to arrive at a
person's type. (Reprinted from McCaulley & Natter, 1974, p. 97)








The El Index

The extravert focuses his perception and judgment on

people and things and tends to be oriented to the outer world.

The introvert is primarily concerned with his inner world and

focuses his perception and judgment on ideas and concepts.


The SN Index

This index is designed to measure two modes of perceiving.

The sensing type becomes aware or perceives directly through

his five senses and tends to be present-oriented and interested

in facts, details, and concrete experiences. The intuitive

type perceives indirectly through his unconscious and is

interested in possibilities, meanings, relationships, and

abstract experiences.


The TF Index

This index is designed to measure two modes of judging.

Thinking types make their decisions objectively and tend to

be logical, analytical, and matter-of-fact. Feeling types

make their decisions based on their value systems and sub-

jectively weigh the values of alternatives.


The JP Index

Judging types rely on a judging process (T or F) in

dealing with the outer world and prefer a systematic, planned,






39

and orderly way of living. Perceptive types rely on a percep-

tive process (S or N) for coping with the outer world and

prefer a spontaneous and flexible way of living.


Dominant and Auxiliary Processes

The perceptive (SN) and judging (TF) processes have the

strongest influence on the personality and provide stability

for the personality. For each person one of the two processes

dominates and unifies his life while the auxiliary process

supports and aids the dominant mode of functioning. The JP

index identifies the process a person uses in dealing with the

outer world. For the extravert this process is his favorite

or dominant process, and for the introvert the process most

often used in his outer behavior is his auxiliary process.

Dominant processes for each type are identified in Figure 2.



ST SF NF NT


I--J


I--P


E--P


E--J


Figure 2. Type table: dominant processes. (Reprinted
from Myers, 1970, p. 3.)


ISTJ ISFJ INFJ INTJ


ISTP ISFP INFP INTP


ESTP ESFP ENFP ENTP


ESTJ ESFJ ENFJ ENTJ








Methodological Problems


Interpersonal Perception Method

A procedure called the "response predicting paradigm"

(Brown, 1965) or the "interpersonal perception method (IPM)"

(Laing et al., 1966) has been used in studies of interpersonal

perception to quantify variables such as social sensitivity,

empathy, understanding of others, etc. The typical process

involves the prediction by a judge (J) of how another person

(0) will respond on a personality inventory. The degree to

which J is correct in predicting O's actual responses is

considered to be an indication of the accuracy of some inter-

personal perception skill (e.g., empathy, understanding

others, etc.).

Gage and Cronbach (1955) have done a comprehensive

critique of methodological and conceptual problems pertaining

to response predicting paradigms. Their foremost criticism

of the studies in which this procedure was used was of the

inadequacy of operational definitions and face validity. Many

writers did not carefully specify the variables they purported

to measure and erroneously generalized their findings to unre-

lated situations. Gage and Cronbach attempted to identify

sources of error within the system by breaking it down into

measurable components. These included elevation of scores,








social desirability of responses, degree of acquaintance of

subjects, effects of real similarity, and effects of warranted

and unwarranted assumed similarity.

Elevation component. Problems of elevation occur in

studies where subjects must rate items which have a range of

numerical values (e.g., -3 through 0 to +3). Answers may tend

to reflect individual bias with some subjects tending to select

high numbers and others low numbers. This problem will not

apply to this study since the Type Indicator is a forced-choice

inventory.

Social desirability. As is true of other psychological

testing, the interpersonal perception method may use a ques-

tioning format which may tempt a subject to describe himself

or another in a manner he considers socially desirable rather

than realistically accurate. This should not be a problem for

this study, however, since according to the manual for the

Myers-Briggs Type Indicator (Myers, 1962), items for the instru-

ment were selected so that they would appeal to the appropriate

types or where this was not possible were differentially

weighted to control for social desirability.

Acquaintance of subjects. Criticisms pertaining to the

degree of acquaintance between subjects were directed primarily

at studies which used groups of subjects who had had little

previous contact with each other. Again this should not be a






42

source of error for this study since the subjects are adoles-

cents and parents who live in the same home and who have

regular contact with each other.

Similarity effects. Gage and Cronbach (1955) describe

various implications related to real similarity and assumed

similarity between subjects. Given the study where we have

(a) Subject A's self-description, (b) Subject B's self-

description, and (c) Subject A's prediction of Subject B,

the responses to any item on the test will have the following

three dimensions:

RS (real similarity): a = b

AS (assumed similarity): a = c

ACC (accuracy): b = c

Only two of these three are independent relations.
That is, when two of these relations are known,
the third may be inferred. Thus, if AS and RS on
an item are scored 1 (denoting agreement), ACC
must be 1. Scores for the three relational vari-
ables are obtained by summing the values obtained
on single items. Any score may be considered a
resultant of the other two. (p. 415)

Gage and Cronbach (1955) also report that people who tend

to assume that others are similar to themselves tend to do so

to the same degree over all items. Other evidence (Lundy,

1956) supports the contention that a subject's AS score

reflects his general attitude toward people. In this case,

people who pay more attention to themselves than to others

receive higher AS scores.





43

Accuracy has a simple definition, but it is difficult to

determine whether a person makes an accurate prediction because

he can in some way understand the similarity between himself

and another or because he assumes the other is like himself.

Gage and Cronbach (1955) suggest that the issue can be resolved

by identifying more elemental components of similarity.

It is important to note that these elemental components

will be the source of data for the statistical analyses for

this study. Hypothesis 1 on levels of interpersonal under-

standing will be analyzed from data resulting from a summation

of the two measures on warranted (i.e., correct) assumptions

of both similarity and dissimilarity. Hypothesis 2 on assump-

tions of similarity will be analyzed from data resulting from

a summation of scores on warranted assumed similarity and

unwarranted (i.e., incorrect) assumed similarity. Gage and

Cronbach (1955, p. 416) use the following schema to demonstrate

the four possible components of each test item:

a b a= b
Real Dissimilarity Real Similarity
(RD) (RS)


a = c
Assumed
Similarity
(AS)

a c
Assumed
Dissimilarity
(AD)


Unwarranted Warranted
Assumed Assumed
Similarity (UAS) Similarity (WAS)
a = c # b a = b = c

Warranted Unwarranted
Assumed Assumed
Dissimilarity (WAD) Dissimilarity (UAD)
a b = c a = b c

a = judge's self-description
b = other's self-description
c = judge's prediction






44

Lindner (1972) points out several practical implications

related to scores high on assumed similarity. In dealing with

subjects highly familiar with each other as in the case of

parents and children, warranted assumed similarity may produce

no dissonance in the relationship if the subjects are in fact

similar. However, in real-life situations a subject making

unwarranted assumptions about similarity may have to modify

his perceptions when his interactions with another fail to

support his false assumptions. On the other hand, investiga-

tors should be cautious in their conclusions about subjects

demonstrating strong tendencies for assuming similarity. When

pairs of subjects are high on real similarity, excessive assump-

tions of similarity by one or both subjects will indicate a

high level of interpersonal understanding. On the other hand,

subjects who are in fact very dissimilar but who make exces-

sive assumptions of similarity will appear to be low on inter-

personal understanding.

Gage and Cronbach (1955) have devised a correction for-

mula for estimating whether a particular subject is assuming

similarity or is in fact making accurate predictions. They

indicate, however, that the problem of assumptions is impor-

tant when the subjects have known each other only a short time

and becomes much less of an issue when subjects are highly

familiar with each other as in the case of parents and their






45

children. Nonetheless, Lindner (1972) concludes ". that

answering empirical questions in this area requires extreme

care in specification, collection, and analysis of data" (p. 22).















CHAPTER III

DESIGN OF THE STUDY


This study employed an interpersonal perception method

to investigate the relationship between the levels of under-

standing between disruptive adolescents and their mothers and

fathers and the levels of understanding between nondisruptive

adolescents and their mothers and fathers. A group of dis-

ruptive high school students and their parents were compared

to a group of nondisruptive students and their parents on

measures of real similarity, real dissimilarity, warranted

and unwarranted assumptions of similarity, and warranted and

unwarranted assumptions of dissimilarity. In addition, the

factors of disruption and nondisruption were related to

demographic data.


Subjects


Description and Selection

Two groupings of adolescents (N=41) and both of the

natural parents (N=82) with whom these adolescents live were

used in the study. One group was composed of disruptive






47

adolescents (N=20) with a mean age of 16.25 and their natural

parents (N=40). The mothers in this group had a mean age of

43.05, and the fathers had a mean age of 46.4. The second

group was composed of nondisruptive adolescents (N=21) with a

mean age of 15.9 and their natural parents (N=42). Mothers

of nondisruptive adolescents had a mean age of 45.0 as com-

pared to their husbands' mean age of 46.9. All of the students

participating in the study were enrolled at the P. K. Yonge

Laboratory School of the University of Florida.

Subjects were limited to a high school population (i.e.,

those students completing grades 9, 10, 11, and 12 at the time

of the study). The disruptive group was composed of 17 white

students, 3 black students, 15 males, and 5 females. All of

the nondisruptive group was white, and it was comprised of

12 males and 9 females. The distribution by grade for the

disruptive group was as follows: ninth grade--4 males and

1 female; tenth grade--3 males and 1 female; eleventh grade--

5 males and 1 female; and twelfth grade--3 males and 2 females.

The nondisruptive group was distributed as follows: ninth

grade--2 males and 2 females; tenth grade--4 males and 4

females; eleventh grade--3 males and 1 female; and twelfth

grade--3 males and 2 females.

The ages of the adolescents ranged from 14 years to 19

years. This age group was selected to minimize the emotional






48

or behavioral problems associated with prepuberty and to con-

trol for behavioral and maturity differences between middle

school and high school students. In addition, since the Type

Indicator is standardized for seventh grade students (age 12),

the use of an older sample (age 14+) made allowance for educa-

tional (i.e., reading) deficiencies assumed to exist among

younger students (especially among disruptive students).

The sample was also restricted to families where both

the natural mother and the natural father are living in the

home. This provision allowed for a study of differences

between parents and their natural children without the con-

founding social and emotional variables that would have

resulted from the inclusion of stepparents or other adult

figures in the home.

Since the P. K. Yonge Laboratory School is a research-

oriented school, the total student body was selected to

approximate, on variables of race and socioeconomic factors,

the distribution of students in the local public school

system. Subjects for the study were drawn from a total popu-

lation of 360 high school students (90 students in each grade

9-12) at P. K. Yonge and were identified by the counselors in

the Pupil Personnel Department of the school.

The counselors were instructed to identify disruptive and

nondisruptive students who live with their natural parents.






49

Disruptive students were defined as those whose behavior inter-

fered with their own learning and the learning of others, whereas

nondisruptive students were defined as those whose behavior was

highly facilitative for their own learning and the learning of

others.

On a more objective basis, disruptive students had to have

been removed from their learning environment two or more times

during the 1974-75 school year for disciplinary action. Typi-

cal behaviors leading to removal from the classroom included

physical or verbal aggressiveness towards others, making exces-

sive noise in the classroom, leaving the assigned work area,

coming to class intoxicated or high on drugs, refusing to

participate in class activities, and encouraging disruptive

acting out behaviors in others.

Nondisruptive students, on the other hand, had no history

of disciplinary action requiring removal from the classroom.

They were selected on the basis of the following behavioral

criteria: exemplary behavior on a consistent basis, refraining

from physical aggression toward others, positive or facilitative

verbal interactions with others, maintaining appropriate noise

levels, remaining in the assigned work area, having no history

of coming to class intoxicated or high on drugs, cooperative

participation in class activities, and encouraging productive

behaviors in others.








Instrumentation


Family Inventory Form

The Family Inventory Form (see Appendix A) was a set of

questions designed specifically for this study. The questions

provided a standard format for acquiring demographic data on

each of the families participating in the study. These data

were supplemented for each subject with additional information

requested on the Type Indicator answer sheets (i.e., date of

birth, occupation, and sex).


Myers-Briggs Type Indicator (Form F)

The Myers-Briggs Type Indicator, Form F, is a self-

administered and untimed personality inventory. The forced-

choice, self-report questionnaire consists of 166 items and

purports to measure the four following dichotomous preferences

which were described in Chapter II: extraversion-introversion

(E-I), sensing-intuition (S-N), thinking-feeling (T-F), and

judgment-perception (J-P). This inventory is standardized for

use with persons age 12 to adult, and the data for reliability

and validity are reported in the test manual (Myers, 1962) and

by McCaulley and Natter (1974).

Each of the four bipolar indices of the MBTI are scored

separately with the resulting "preference score" consisting

of a letter identifying the direction of the preference (e.g.,






51

"E" indicates a preference for extraversion) and a number indi-

cating the strength of the preference. When all four indices

have been scored, the preference score may be simplified by

reducing it to a "type formula." A type formula consists

simply of the four letters designating the direction of prefer-

ence on each index (e.g., "ENTJ").


The Interpersonal Perception Method

For the purpose of this study (i.e., to measure under-

standing between a parent and a child), the first and third

steps in Dymond's Response Predicting Paradigm (1949; 1950)

were sufficient. For each pair of subjects (i.e., a child and

his mother or a child and his father), both subjects described

themselves and predicted how the other subject described himself.

The Myers-Briggs Type Indicator was the tool for this

study for implementing the interpersonal perception method.

Each child in the study answered the questions on the MBTI to

describe himself, to predict how his mother described herself,

and to predict how his father described himself. Each parent

in the study answered the MBTI questions to describe himself

or herself and to predict how the child described himself.

Dymond (1949) was the first investigator to devise a

response predicting paradigm for measuring empathy. She

slightly modified her original instrument (Dymond, 1950),





52

and it is this basic revised model that has been most often

used in recent studies of empathy and other interpersonal

variables. She operationally defines empathy in terms of the

following four processes: (a) Subject A rates himself on a

personality inventory; (b) Subject A on the basis of his own

perceptions of Subject B rates Subject B on the same inventory;

(c) Subject A predicts how Subject B will rate himself; and

(d) Subject A predicts how Subject B will rate him (Subject A).

Reliability data for Dymond's instrument were split-half

correlations yielding a reliability coefficient of +.82. Valid-

ity was determined by a judge's rating technique, an empathic

index based on the Thematic Apperception Test, and subject

reports on their introspective processes (Dymond, 1950).

Lindgren and Robinson (1953) evaluated Dymond's paradigm

through research of their own and further expanded the reli-

ability and validity data.

Laing et al. (1966) devised the Interpersonal Perception

Method (IPM). Their approach was similar to Dymond's method

but deleted the second step in the Dymond procedure in which

Subject A on the basis of his own perception of Subject B

rated Subject B on the inventory. Thus, Dymond's operational

definition of empathy becomes more comprehensive than Laing's.

Whereas Laing defines empathy as a subject's ability to deter-

mine how another person views himself, Dymond includes Laing's








measure of empathy plus the additional possibility of con-

trasting Subject A's own view of Subject B with Subject A's

prediction of how Subject B will describe himself. Both the

Dymond and the Laing procedures, by contrasting a subject's

(e.g., Subject A) view of himself with his prediction of how

the other (e.g., Subject B) sees him (i.e., Subject A), allow

for a measure of "feeling understood."


Procedures and Data Collection


After the subjects were identified by the counselors in

the Pupil Personnel Department of the P. K. Yonge Laboratory

School, an introductory letter (see Appendix B) was mailed to

the parents of 28 disruptive students and to the parents of

34 nondisruptive students. The letter was followed by a

personal contact with the parents in the form of a telephone

call. The personal contact served the following purposes:

(a) to encourage the families to participate in the study,

(b) to answer questions pertaining to the study, (c) to sched-

ule family members for the necessary testing sessions, (d) to

invite family members to optional feedback sessions during

which the examiner explained basic type theory and interpreted

the MBTI scores, (e) to survey the reasons why some families

refused to participate in the study, and (f) to assure the

families that all of their scores and personal data would

remain anonymous.






54

The investigator was able to contact 27 families in the

disruptive group and 33 families in the nondisruptive group.

Initially, 24 families in the disruptive group and 25 families

in the nondisruptive group agreed to participate in the study.

In the final analysis, 20 families with disruptive children

and 21 families with nondisruptive children actually completed

the required testing. This represented a dropout rate of 25.9

percent for the disruptive group and 36.4 percent for the non-

disruptive group. Reasons for declining included illness in

the family, absence of a family member due to a business trip

or vacation, time commitments to other research projects, over-

extension of the family in other activities, and in one case a

prior family decision to withdraw from all research. In only

three families was there a direct refusal to discuss the possi-

bility of participating in the study. In two of these families

the father (but not the mother and the child) refused to partici-

pate, and in one family the mother (but not the father and the

child) was unwilling to be a subject for research.

Each family agreeing to participate in the study was con-

tacted individually by one of three examiners. The examiners

gave standard directions for each part of the testing and

explained the use of the answer sheets, the Family Inventory

Form, and the Informed Consent sheets (see Appendix C). The

testing materials were then left with each family so that the






55

family members could complete the forms at a time convenient

to them. When the families had completed all of the answer

sheets, the materials were returned to the examiners.


Testing of Adolescents

Each adolescent (disruptive and nondisruptive) was given

a copy of the Myers-Briggs Type Indicator (MBTI) and three

answer sheets separately labeled "child for self," "child for

mother," and "child for father." Following the standard direc-

tions appearing on the back of the machine-scored answer sheets,

subjects described themselves on the "child for self" answer

sheet. On the "child for mother" sheet, each subject using

a modified set of instructions (see Appendix D) described the

way he believed his mother would answer the MBTI questions.

A second set of modified instructions (see Appendix E) were

followed for completing the "child for father" answer sheet.

Under this condition, the child predicted the way his father

would describe himself when he answered the Type Indicator

questions. Thus, after reading each MBTI item, the adoles-

cents recorded responses on three answer sheets according to

three different sets of directions.


Testing of Parents

Each parent in the study was given a copy of the Myers-

Briggs Type Indicator and two answer sheets separately labeled








either "mother for self" and "mother for child" or "father

for self" and "father for child." Following the standard

directions on the back of the machine-scored answer sheets,

the parents described themselves on the sheets labeled "mother

for self" and "father for self." On the "mother for child"

and "father for child" answer sheets, the parents followed a

modified set of instructions (see Appendix F) for describing

the way they believed their child would answer the MBTI

questions. Thus, after reading each MBTI item, the parents

recorded responses on two answer sheets according to two dif-

ferent sets of directions. The parents also worked together

to complete the Family Inventory Form.


Provision for Poor Readers

The examiners were prepared to read the research materials

aloud to any subject whose reading level was too low for reading

alone. However, only two families were identified in which one

parent had only an elementary education. One examiner read

the test materials aloud for both of these parents and assisted

them in the completion of their answer sheets.


Data Processing and Analysis


All answer sheets for the MBTI were machine scored, and

all statistical analyses were done by computer using a Fortran








program written specifically for this study. Each of the

hypotheses was subjected to analysis of variance followed by

Scheff'es method where significant F's were found. The Scheffe'

tests employed a correction formula to control for increased

probability of significant differences by chance resulting

from multiple comparisons which required the use of the same

mean scores in various combinations.


Operational Definitions for Measures of Understanding


The following operational definitions will be employed

in the statistical analyses of the hypotheses for this study:

(1) Child's self-report (CSR) is composed of the individ-

ual responses each child in the study made on each of the 166

items of the Myers-Briggs Type Indicator when describing the

way he naturally feels or acts.

(2) Mother's self-report (MSR) is composed of the individ-

ual responses each mother in the study made on each of the 166

items of the MBTI when describing how she naturally feels or

acts.

(3) Father's self-report (FSR) is composed of the individ-

ual responses each father in the study made on each of the 166

items of the MBTI when describing how he naturally feels or

acts.

(4) Child's prediction of mother (CPM) is composed of






58
the individual responses each child in the study made on each

of the 166 items of the MBTI when predicting his mother's self-

report (MSR).

(5) Child's prediction of father (CPF) is composed of

the individual responses each child in the study made on each

of the 166 items of the MBTI when predicting his father's self-

report (FSR).

(6) Mother's prediction of child (MPC) is composed of

the individual responses each mother in the study made on each

of the 166 items of the MBTI when predicting her child's self-

report (CSR).

(7) Father's prediction of child (FPC) is composed of

the individual responses each father in the study made on each

of the 166 items of the MBTI when predicting his child's self-

report (CSR).

(8) Real similarity (RS) is the total number of MBTI

items on which a child and one of his parents made the same

response on their self-reports (i.e., CSR compared with MSR

or CSR compared with FSR).

(9) Real dissimilarity (RD) is the total number of MBTI

items on which a child and one of his parents made different

responses on their self-reports (i.e., CSR compared with MSR

or CSR compared with FSR).

The following four operational definitions involve a






59

comparison of one subject's self-report with his prediction of

another subject's self-report. There are four possible com-

parisons of this nature: child's self-report (CSR) with child's

prediction of mother (CPM); child's self-report (CSR) with

child's prediction of father (CPF); mother's self-report (MSR)

with mother's prediction of child (MPC); and father's self-

report (FSR) with father's prediction of child (FPC).

(10) Warranted assumed similarity (WAS) is the total

number of MBTI items for which Subject A correctly predicts

that both he and Subject B made the same response on their

individual self-reports.

(11) Warranted assumed dissimilarity (WAD) is the total

number of MBTI items for which Subject A correctly predicts

that both he and Subject B made different responses on their

individual self-reports.

(12) Unwarranted assumed similarity (UAS) is the total

number of MBTI items for which Subject A incorrectly predicts

that both he and Subject B made the same responses on their

individual self-reports.

(13) Unwarranted assumed dissimilarity (UAD) is the total

number of MBTI items for which Subject A incorrectly predicts

that both he and Subject B made different responses on their

individual self-reports.

The final two operational definitions are developed from








all of the preceding definitions and will provide the basis

for the statistical analysis of the hypotheses for this study.

(14) Overall warranted assumptions (OWA) is the sum of

warranted assumed similarity (WAS) and warranted assumed dis-

similarity (WAD).

(15) Overall assumed similarity (OAS) is the sum of

warranted assumed similarity (WAS) and unwarranted assumed

similarity (UAS).


Restatement of Hypotheses in Operational Form


Since having empathic understanding for another person

requires the ability to recognize both similarities and dif-

ferences between oneself and another person, it is assumed

that a person who has a high level of understanding of another

person will be able to make more overall warranted assumptions

(OWA) of both similarity and dissimilarity about that person.

The following null hypothesis was formulated to test for dif-

ferences between the levels of interpersonal understanding

which exist between disruptive and nondisruptive adolescents

and their parents. It will be tested by comparing the following

eight pairs of mean scores for overall warranted assumptions

(OWA) which were determined by comparing one subject's pre-

diction of another (e.g., CPM) with the other subject's self-

report (e.g., MSR): (a) mean OWA score of disruptive adolescents








for mother with mean OWA score of nondisruptive adolescents

for mother; (b) mean OWA score of disruptive adolescents for

father with mean OWA score of nondisruptive adolescents for

father; (c) mean OWA score of mothers of disruptive adoles-

cents for child with mean OWA score of mothers of nondisruptive

adolescents for child; (d) mean OWA score of fathers of dis-

ruptive adolescents with mean OWA score of fathers of non-

disruptive adolescents for child; (e) mean OWA score of

disruptive adolescents for mother with mean OWA score of

disruptive adolescents for father; (f) mean OWA score of non-

disruptive adolescents for mother with mean OWA score of non-

disruptive adolescents for fathers; (g) mean OWA score of

mothers of disruptive adolescents for child with mean OWA

score of fathers of disruptive adolescents for child; and (h)

mean OWA score of mothers of nondisruptive adolescents for

child with mean OWA score of fathers of nondisruptive adoles-

cents for child.

Hypothesis 1. There is no difference between the mean

OWA scores of disruptive adolescents and their parents and

the mean OWA scores of nondisruptive adolescents and their

parents.

The following hypothesis also stated in the null form was

formulated to test the assumption that persons with lower levels

of empathic understanding of another person will make more






62

assumptions of similarity between themselves and the other

person. Using the same eight combinations employed in Hypoth-

esis 1, Hypothesis 2 will be tested by comparing mean scores

for overall assumed similarity (OAS) which like the mean OWA

scores were determined by contrasting one subject's prediction

of another (e.g., CPM) with the other subject's self-report

(e.g., MSR).

Hypothesis 2. There is no difference between the mean

OAS scores of disruptive adolescents and their parents and the

mean OAS scores of nondisruptive adolescents and their parents.














CHAPTER IV

RESULTS


Computer analysis of the data was conducted in terms of

main effects and interaction of main effects. Main effects

included the following four categories for both the disruptive

adolescents and their parents and the nondisruptive adoles-

cents and their parents: (a) child's prediction of mother

(CPM), (b) child's prediction of father (CPF), (c) mother's

prediction of child (MPC), and (d) father's prediction of

child (FPC). Analysis of variance was followed by the Scheffe

method where significant F's were found, and all categories

were tested for homogeneity of variance.


Tests for Homogeneity of Variance


Each of the eight categories of main effects (four for

subjects in the disruptive group and four for subjects in the

nondisruptive group) were tested for homogeneity of variance

on all of the following measures: (a) real similarity (RS),

(b) real dissimilarity (RD), (c) warranted assumed similarity

(WAS), (d) warranted assumed dissimilarity (WAD), (e) unwarranted






64

assumed similarity (UAS), (f) unwarranted assumed dissimilarity

(UAD), (g) overall warranted assumptions (OWA), and (h) overall

assumed similarity (OAS). No significant differences in vari-

ance were found, and all group means fell within the 95 percent

confidence interval for expected means.


Evaluation of Hypotheses


Hypothesis 1

Hypothesis 1 was designed to test for differences between

the levels of interpersonal understanding (OWA) which exist

between disruptive and nondisruptive adolescents and their

parents and states that there is no difference between the

mean OWA scores of disruptive adolescents and their parents

and the mean OWA scores of nondisruptive adolescents and their

parents. The eight pairs of interactions described in Chapter

III were tested, and none of the comparisons revealed a

significant difference. A summary of the analysis of vari-

ance is presented in Table 1, and the mean OWA scores, standard

deviations, ranges, degrees of freedom, and the difference

between means used in the Scheffe test for each comparison are

reported in Table 2.

An overall analysis of variance for the combined main

effects did show a significant difference (p .017) between

the disruptive adolescents and their parents as a group and












0

cd

C H r000 r- 0
.r-l

*H u OO OC



ooco o
-H







M



U3

Q It 01 0 0L
-Q) C-r- Cl
0 oo C)l n m -









C0 l
OM O








f0 44
) 4d d0
0 0


4 0 0

0 :4- 0 -






H ( o c
C- C H 0
4-l U e








En C 4 CO i
LO C -

0 cnt
S- 0 -l 00 0
0i EQ
i -o e CM Cn





CO 4
0 0

0 0

I e O
0 *H



Cd -p -H


4-1 cd
0 ID3 ) 4
> a a)

0 4- 4J -



00
-H 0 aU)t
toal o m
Cd CD
z H K












S ON 0



N 0 C'!
0 oe -oo
o .- N'


a) 0




0 a



(a) S
co 0




do0<


r-l








Q)l
0






0
.,A
40



CO














oi

0a
E- p
C












4 I
44-


4-)
Ln










rJ2
'0)
>4-
a)


Ml


O 0, c0' CO CO 0 0
0000 c








co co a
I I I


O O




m mo
' o c a to


o o in
0 0 -l
0N CO
o) c4 oc
















0) '0 i0 n




S 4 0 0r4

: 0 -H -4 4
CL Cc C- r



ci Pi 0
C5, *0l*rl CD a 04r
440 4 '4 4-q )4A a i







III l I l
-1 "o l 0 U U
0c0 0 04
















0
C -
0U 0 r *H.H4-J 4-l




*" o "' "' 4 i


u z


c) o) )


I






67

the nondisruptive adolescents and their parents as a group on

measures of OWA. The nondisruptive children and their parents

were more accurate than the disruptive children and their

parents in predicting each other's responses and as a group

show higher levels of interpersonal understanding.


Hypothesis 2

Hypothesis 2 was formulated to test the assumption that

persons with lower levels of empathic understanding of another

person will make more assumptions of similarity (OAS) between

themselves and the other person and states that there is no dif-

ference between the mean OAS scores of disruptive adolescents

and their parents and the mean OAS scores of nondisruptive

adolescents and their parents. The same eight comparisons

employed for Hypothesis 1 were tested for Hypothesis 2, and

none of the comparisons revealed a significant difference. A

summary of the analysis of variance is presented in Table 3,

and the mean OAS scores, standard deviations, degrees of

freedom, and the difference between means used in the Scheffe

test for each comparison are reported in Table 4.


Evaluation of Demographic Data


Means for all data collected on the Family Inventory

Form (see Appendix A) and for the ages of all subjects and








68



u
ci)
Cd

*,A 0 1 . I







cr- -4 c-i


a)r O) 00 l O
TrO












Q u eq (r-f- r CA
CM 00QC)n C) U

00




.n
Cd





















,-0 0 .0.
0 ,-I' c

















O -r4
ci m C O




c o C4 i 0 C C




< CO Di> ON




0 0>


cc 0 -P 40

0 1 4 0


0 ( *
Q e u





c 0 4 P
0 U)
L H1 E C4 C
Qr P I cc








C

Ce 0 /
-'-4 ZC/ Cr








cc )xCici C
rO- ci '
0, CL) d
Ce i C
H a p










a)
C) C
Sr-H


r Pq
fml


m E
aU 0


Q) 4
Q H


0,

Ci


0 0
o Q\
0o -
0 -
CO i


C o LO




) a)

0 0
-, C-4 -- 1 .. ..0
O Lr CM CC CC CI
r- 0 00 0

a) a)

C') CM) a- )-

Co hrl ii


C.I

U)

CO


4-J
0





-,







C/)
i--I





cc
*rl










i-i
HI





4-1
CO


a.
nj

rU
>








4-1
44


SUO U
u ->a


u u


h U
rL, u


HU 1
" S U


69





























Lfc N
mL CM4 r
S- NOC



II II II


Ca 4j CI)


4J 4J 4J









. ... 0 T O O
0 0 0UO O *H .r
4 4 O L 00 H -.

44L 4-4 IT) 0 C0 u *
0L 00 * -

4- 4-4 0 0

)O- -




II II II - 4J-)
S4 4J 4Jl L) 0










'0 O d T C * H44
u 0 H






II II II II
p4 i C PL P-4




('1 (*H -1 4-J






0 0
C 3
.!../
kFPP


U H


C Z






70

the sex of all children which were recorded on the MBTI answer

sheets were computed for disruptive adolescents and their

parents and for nondisruptive adolescents and their parents.

The means for each group on each demographic variable were

tested for significant differences either by t-tests or by

tests for significant differences in proportion. These figures

are reported in Appendix G.

Three educational variables, one income variable, and two

birth order variables reached significant levels of difference.

A significantly higher proportion (p .01) of both the mothers

and fathers of disruptive adolescents were found to have only

a high school education or less. On the other hand, a signifi-

cantly higher proportion (pl .02) of the fathers of nondisrup-

tive adolescents as compared to the fathers of disruptive

adolescents had earned an undergraduate college degree.

Birth order variables for the children, mothers, and

fathers were compared in three ways. The first analysis

divided each group into only child, first child, middle child,

or last child. The second analysis divided each group by the

number of siblings, and the third analysis combined first,

middle, or last birth position with the number of siblings.

The group of nondisruptive adolescents included a signifi-

cantly higher proportion (p .05) of children with three

siblings. No significant differences were found in the first






71

two analyses of birth order variables for the fathers, but in

the third analysis which combined birth position with number

of siblings, it was found that the group of fathers of non-

disruptive adolescents included a significantly higher pro-

portion (p .05) of last children with one sibling.

One other variable, total family income over $30,000,

reached a probability level of .05. Nine families of nondis-

ruptive children as compared to three families of disruptive

children fell into this category.


Additional Findings: MBTI Data


All answer sheets for the Myers-Briggs Type Indicator

were machine scored for type following standard scoring proce-

dures to determine preference scores for both the self-reports

and the predicted responses. Distributions of types for sub-

groups within the study were further analyzed by a computer

program which uses the Chi Square statistic for determining

significant differences in the proportion of subjects selecting

each of the sixteen type categories and a standard series of

combinations of the types. Three sets of comparisons were

made by contrasting the disruptive and nondisruptive groups

on the self-reports of the children, the self-reports of the

mothers, and the self-reports of the fathers. Eight other

comparisons were made by contrasting MBTI type distributions






72

for subjects from the disruptive group and for subjects from

the nondisruptive group in the following four combinations:

(a) CPM with MSR, (b) CPF with FSR, (c) MPC with CSR, and (d)

FPC with CSR. The fourteen type tables generated by these

comparisons are presented in Appendix H.


Self-Reports

No significant differences were found by comparing the

type distributions of the self-reports of disruptive and non-

disruptive adolescents. However, there were substantial

differences on the extraversion-introversion index. Seventy

percent of the disruptive children chose introversion as

compared to 47.62 percent of the nondisruptive children.

The self-reports of the mothers of disruptive children

and the mothers of nondisruptive children differed signifi-

cantly on several dimensions. Disruptive children had

significantly more mothers who preferred thinking (p .01)

and particularly the tough-minded thinking-judging types

(pt .05) and fewer who preferred feeling (p .01) and particu-

larly intuition-feeling (p .05) and feeling-judging (p .05).

All of the mothers of nondisruptive children preferred

feeling (p .01) and particularly intuition-feeling (p .05)

and feeling-judging (p .05). None preferred thinking (pd .01).

or were in the tough-minded thinking-judging type (p .05).






73

In addition to the difference between the two groups of mothers

on the T-F index where 100 percent of the mothers of nondisrup-

tive adolescents selected feeling as compared to 70 percent of

the mothers of disruptive adolescents, smaller differences

were evident on the S-N and the J-P indices with more of the

mothers of disruptive adolescents preferring sensing (70 per-

cent as compared to 52.38 percent) and with fewer preferring

the "executive" judging attitude (55 percent as compared to

71.43 percent).

The two groups of fathers did not differ significantly

on any measure related to their self-reports. Forty percent

of the fathers of disruptive children as compared to 9.52 per-

cent of the fathers of nondisruptive children fell into the

preference type ISTJ. Like the mothers, more fathers of dis-

ruptive students as compared to the fathers of nondisruptive

students preferred sensing (75 percent as compared to 57.14

percent) and thinking (70 percent as compared to 57.14 percent).

Four out of five fathers in both groups reported a preference

for the organized, systematic judging attitude.


Parental Predictions of Children

It is necessary to be cautious when comparing the pre-

dicted scores for a group with the group's self-reports since

the predicted scores are not being analyzed individually for






74

each family. Therefore, comparisons of this nature can only

reflect an overall tendency of a group to overestimate or under-

estimate in a particular direction.

It appears that there was an overall tendency for the

parents to be more accurate in predicting their children's

responses than the children were in predicting their parents'

responses on the MBTI items. No significant differences were

found when comparing the predictions of the mothers and fathers

of disruptive adolescents and of the fathers of nondisruptive

adolescents with their children's self-reports. There was,

however, a tendency for the mothers of nondisruptive children

to underestimate (p .05) the proportion of their children who

were introverted types in the perceptive attitude (IP).


Children's Predictions of Parents

Surprisingly, disruptive adolescents appear to have been

more accurate in predicting their parents' responses than were

the nondisruptive adolescents. No significant differences were

found when comparing the disruptive adolescents' predictions

of their fathers with their fathers' self-reports. Disruptive

adolescents, however, did have an overall tendency to see

their mothers as more sensing-thinking (p .05), less intuitive-

perceptive (p! .95), and less feeling-perceptive (p .05) than

the mothers reported themselves to be.






75

Nondisruptive adolescents seem to have a more accurate under-

standing of their fathers than of their mothers. They described

their mothers as having significantly more sensing (p .01)

and thinking types (p .01) than appeared in the mothers' self-

reports and significantly more of the subgroups of the following

types: sensing-thinking (p! .05), sensing-judging (p! .05),

thinking-judging (pt .05), and extraversion-sensing (p' .001).

In addition, the nondisruptive adolescents predicted more of

the decisive extraverted-judging types (p .05) than occurred

in their mothers' self-reports. When predicted types for the

main eight MBTI dimensions approach significance (e.g the

extraversion and judging types predicted by nondisruptive

adolescents for their mothers), a combination of these pre-

dicted dimensions will reach significant levels of difference

(e.g., the EJ types predicted by nondisruptive adolescents

for their mothers).

The nondisruptive adolescents predicted their fathers'

types more accurately than their mothers' types. In the main

preferences, they overestimated the number of thinking types

(p! .05); and in the subgroups of the types, they overestimated

sensing-thinking (p= .01) and sensing-perception (p .05).

There seems to be a common tendency for children and

mothers to overestimate on the extraversion index. On extra-

version the nondisruptive adolescents overestimated their








mothers by 28.57 percent and their fathers by 19.05 percent.

On the same index, disruptive adolescents overestimated their

mothers by 10.0 percent and their fathers by 20.0 percent.

The mothers of disruptive adolescents predicted a proportion

of extraversion 25 percent higher than their children's self-

reports, and the mothers of nondisruptive adolescents over-

estimated the extraversion of their children by 14.29 percent.

The same pattern was not evident in the fathers' predictions

of the children. The proportion of extraversion predicted by

the fathers of disruptive adolescents was the same as the pro-

portion of their children's self-reports, and the proportion

of extraversion predicted by the fathers of nondisruptive

adolescents for their children was underestimated by 9.52

percent.

In the predicted types made by disruptive and nondisrup-

tive children and their parents for the subgroup of extra-

version-sensing (ES), only the fathers of nondisruptive

children in predicting their children's self-reports did not

overestimate. Mothers of disruptive students saw their

children as 20 percent more ES, and the fathers of disruptive

students saw their children as 10 percent more ES than their

self-reports. The mothers of nondisruptive students over-

estimated their children's self-reports for ES by 9.52 percent,

but the proportion of ES which the fathers of nondisruptive






77

students predicted for their children was the same as appeared

in the children's self-reports. In predicting their fathers

on ES, disruptive youths overestimated by 20 percent, and non-

disruptive youths overestimated by 19.05 percent. The most

extreme differences were in the predictions of ES made by the

children for their mothers. Disruptive adolescents saw their

mothers as 25 percent more ES than their self-reports, but

nondisruptive adolescents overestimated their mothers' prefer-

ence for ES by 52.38 percent (p .001).














CHAPTER V

DISCUSSION


Demographic Variables


Considering the previous research on disruptive behavior,

it is surprising that so few of the demographic variables in

this study reached significant levels of difference between

the families of disruptive youths and the families of nondis-

ruptive youths. The absence of such differences may possibly

be attributed to the nature of the laboratory school from which

the samples were drawn since a large number of the children

from both the disruptive and nondisruptive groups have parents

who are associated with the University of Florida or who hold

other professional positions in the community. The small

representation of black families (15 percent of the disruptive

group) in the study also suggests that the sample drawn from

the laboratory school may be somewhat atypical of a sample

that would be drawn from a public high school.

Fifty-five percent of the fathers and 30 percent of the

mothers of disruptive children do have at least one college

degree. However, the parents of nondisruptive children have

78






79

substantially higher educational levels with all of the fathers

and 66.7 percent of the mothers having one or more college

degrees. Nonetheless, the finding in this study of a signifi-

cantly greater proportion of parents of disruptive adolescents

with only a high school education or less is not unusual. The

fact that a family income of over $30,000 reached a significant

level in favor of the families of nondisruptive adolescents

reflects the trend for this group of parents to be better

educated. Combined measures for education and family income

for all subjects indicate that the families participating in

this study come from a predominantly middle class population.

If the adolescents being studied use their parents for

behavioral models and if their attitudes reflect the interests

and values of their parents, it would be possible to assume

that nondisruptive students because of the educational achieve-

ments of their parents take education more seriously and value

academic success more than disruptive students. If this is

true, nondisruptive students would be more likely to adapt

their behavior at school to standards considered exemplary or

facilitative by school personnel. On the other hand, if the

parents of disruptive students value education less than do

the parents of nondisruptive students, their attitudes toward

school may be different, and they may have fewer expectations

of how their children should behave or perform in school. If








this is the case, the parents of disruptive adolescents may

not respond as negatively when reports of their children's

acting out behaviors come to their attention, or they may be

more prone to consider the disruptive behaviors as a passing

phase or even as cute.

In over one hundred comparisons of demographic data, it

can be expected that several comparisons will be significantly

different due to chance. Research on birth order has typically

produced unstable results and offers no support for the birth

factors found to be significant for nondisruptive children

and their fathers. The fact that the nondisruptive children

were more likely to have three siblings and that their fathers

were more likely to be the younger of two children can prob-

ably be attributed to chance since the sample size was small

and since the birth data were divided into several different

categories for analysis. Certainly much significance should

not be attached to this information without further extensive

research.


Interpersonal Understanding


Even though none of the eight comparisons of measures

of overall warranted assumptions (OWA) revealed a significant

difference, all of the comparisons of the disruptive group

with the nondisruptive group were in the expected direction








with nondisruptive adolescents and their parents being more

accurate in the response predicting task than the disruptive

adolescents and their parents. The overall analysis of vari-

ance of main effects which showed the nondisruptive children

and their parents as a group to be significantly more accurate

on measures of OWA than the disruptive children and their

parents as a group suggests the existence of higher levels of

interpersonal understanding within the families of nondisrup-

tive students even though the specific comparisons of children,

mothers, and fathers do not show any pairing of subjects to

have higher interpersonal understanding than another. Further

research on OWA seems to be necessary before either accepting

or rejecting the OWA measures as a useful means of distin-

guishing between disruptive and nondisruptive adolescents and

their parents. The use of a small sample size and the fact

that the two groups were highly similar on demographic vari-

ables may have contributed to the present outcome. Therefore,

studies correcting the research weaknesses inherent in this

study may produce more practical and meaningful results.

Previous research requiring a response predicting task

for mothers and fathers has consistently shown mothers to have

a better understanding of their children. In this study, a

comparison of MPC and FPC for the parents of nondisruptive

children and for the parents of disruptive children also






82

showed the mothers (and especially the mothers of disruptive

children) to have a slightly greater understanding of their

children than did the fathers.

Differences for both the disruptive and nondisruptive

adolescents between CPM and CPF were very small. Nondisrup-

tive adolescents, however, were substantially more accurate

than disruptive adolescents in predicting the responses of

their fathers. It is possible to speculate that nondisruptive

children as compared to disruptive children may have a more

positive relationship with their fathers which may in turn

lead to higher levels of interpersonal understanding between

them. Further research in this area would be worthwhile

especially if attention is paid to the quality and quantity

of father-child interactions.


Assumptions of Similarity


The analyses for Hypothesis 2 revealed no significant

differences in the eight comparisons of measures of overall

assumed similarity (OAS). The OAS measure for distinguishing

between disruptive and nondisruptive adolescents and their

parents would seem to be of little value. The differences

between the means for all variables except for MPC and FPC

were quite small, and even the MPC and FPC measures were well

within the 95 percent confidence interval for expected means.






83

Even though the disruptive children assumed more similarity

between themselves and their parents than did the nondisruptive

children, both of the parents of nondisruptive adolescents saw

their children as more like themselves than did the parents

of disruptive adolescents with the mothers assuming similarity

to a higher degree than did the fathers. The two groups of

mothers and of fathers did not differ significantly on the RS

measure obtained by comparing their self-reports with the self-

reports of their children. Therefore, the assumptions of

similarity made by the parents of nondisruptive adolescents

cannot be attributed to a high degree of real similarity

between themselves and their children.

When warranted assumptions for both similarity and dis-

similarity are combined and when unwarranted assumptions for

both similarity and dissimilarity are combined, there is

little difference between the mean scores for the two groups

of mothers and between the mean scores for the two groups of

fathers. Mothers and fathers of nondisruptive adolescents,

however, made more warranted and unwarranted assumptions of

similarity, and the mothers and fathers of disruptive adoles-

cents made more warranted and more unwarranted assumptions of

dissimilarity. Thus, even though the differences are small,

there appears to be a slight tendency for the parents of non-

disruptive youths incorrectly to see their children as being






84

like themselves and a tendency for the parents of disruptive

youths to be more prone incorrectly to see their children as

different from themselves.

Wakefield (1966) discovered a tendency among the mothers

in his study to use more projection and for the fathers to use

more stereotyping when predicting the responses of their teen-

age children. If projection is a factor in this study, there

is little reason to believe that it would not appear among all

mothers and not just among the mothers of nondisruptive adoles-

cents. It would not be expected to appear as a significant

factor in the predictions made by the fathers.


Additional Observations on MBTI Data


It is unusual for all of the mothers of nondisruptive

adolescents to have described themselves on the MBTI as

feeling (F) types. Even though 70 percent of the mothers

of disruptive adolescents also selected F, 30 percent of

that group did describe themselves as thinking (T) types.

As a result, the mothers of disruptive children were signifi-

cantly more T, whereas the mothers of nondisruptive children

were significantly more F, NF, and FJ The number of mothers

of nondisruptive youths describing themselves as judging (J)

types was 16.43 percent higher than the mothers in the other

group but did not reach a significant level of difference.






85

Feeling types (F) tend to handle situations with personal

warmth and to be interested in human relationships and in the

human side of issues, whereas thinking types (T) tend to be

skeptical, critical, interested in the objective truth, and

prone to handle situations with impersonal analysis. Con-

sidering type theory, it is possible to speculate that the

mothers of disruptive children may use their thinking skills

to see their children and to relate to their children more

objectively and impersonally. On the other hand, the mothers

of nondisruptive adolescents, lacking a preference for T, may

see their children and relate to their children on a highly

subjective basis and with personal warmth and a sincere

interest in their children's feelings, values, and interests.

Their subjectivity may very well be a prime factor in their

tendency to assume that their children are like themselves.

Katz (1963), in his discussion of empathy, mentioned a

tendency for people who have positive feelings for each other

(e.g., friends) to assume that their loved one has interests

and feelings similar to their own. If the exemplary behavior

that nondisruptive adolescents exhibit at school carries over

to their behavior in the home, their parents may experience

a parent-child relationship with little dissension and much

mutual cooperation. Positive interactions with a child






86

coupled with a highly subjective and emotional mode of relating

to the child, may contribute to the tendency of the mothers of

nondisruptive children to view their children as similar to

themselves.

The tendency of disruptive adolescents as a group, their

mothers as a group, and their fathers as a group to be more

accurate in predicting MBTI type preferences than the non-

disruptive adolescents and their parents and the tendency of

nondisruptive adolescents as a group to be more accurate in

predicting the type preferences of their fathers as compared

to predicting the preferences of their mothers may possibly

be due to the behavioral characteristics associated with

sensing, thinking, and perhaps judging.

In type theory, ST types are typically practical and

matter-of-fact, are more interested in things than in human

relationships, and analyze situations impersonally. Judging

types (J) are prone to be systematic and orderly. When

faulty type development occurs, STJ types tend to be rigid

and authoritarian. Since the S, T, and J dimensions in

various combinations are more frequent among disruptive

adolescents and their parents, the parent-child interactions

for these subjects may possibly be more objective and imper-

sonal and subsequently more realistic. Also, since the

mothers of nondisruptive adolescents prefer a more subjective




University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs