University of Southern California Dissertations and Theses

A Description Of Similarity Of Personality Between Selected Groups Of Television Viewers And Certain Television Roles Regularly Viewed By Them

(USC Thesis Other)

A Description Of Similarity Of Personality Between Selected Groups Of Television Viewers And Certain Television Roles Regularly Viewed By Them

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content This dissertation has been
microfilmed exactly as received ® 8-17,037
PERROW, Maxwell Vermilyea, 1927-
A DESCRIPTION OF SIMILARITY OF PERSONALITY
BETWEEN SELECTED GROUPS OF TELEVISION
VIEWERS AND CERTAIN TELEVISION ROLES
REGULARLY VIEWED BY THEM.
University of Southern California, Ph.D., 1968
Social Psychology
University Microfilms, Inc., Ann Arbor, Michigan
A DESCRIPTION OF SIMILARITY OF PERSONALITY BETWEEN
SELECTED GROUPS OF TELEVISION VIEWERS AND CERTAIN
TELEVISION ROLES REGULARLY VIEWED BY THEM
by
Maxwell Vermilyea Perrow
A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(Communication)
June 1968
UNIVERSITY O F SOUTHERN CALIFORNIA
THE GRADUATE SCHOO L
UNIVERSITY PARK
LO S ANGELES. CALI FORNIA 9 0 0 0 7
This dissertation, written by
.....
under the direction of l u D i s s e r t a t i o n Com
mittee, and approved by all its members, has
been presented to and accepted by the Graduate
School, in partial fulfillment of requirements
for the degree of
D O C T O R O F P H I L O S O P H Y
O
Dean
Date....Jun e,.... . 1968. . ,
DISSERTATION COMMITTEE
TABLE OF CONTENTS
CHAPTER PAGE
I. THE PURPOSE AND PROBLEM............... 1
The Purpose.......................... 2
The Problem.......................... 9
Similarity or Difference of
Personality Profiles ........... 10
Similarities or Differences of
Personal Data................... 11
Definitions.......................... 13
Hypotheses.......................... 16
Similarity or Difference of
Personality Profiles ........... 16
Similarities or Differences of
Personal Data................... 17
Assumptions.......................... 19
Limitations.......................... 21
Organization of Remainder of the
Study.............................. 23
II. REVIEW OF THE LITERATURE............. 24
Television Viewing: A Para-Social
Situation.......................... 25
Perception of Persons: An Important
Element in Television Viewing . . . 28
Personality: A Central Determinant
in the Perception of Persons . . . 34
CHAPTER
III.
IV.
PAGE
Functional and Dysfunctional Values
of Television Viewing . . . . . . . 38
Identification: An Important Element
of Television Viewing............. 44
Program Selection: A Result of
Role-Viewer Identification .... 52
Summary............................... 57
METHODS AND PROCEDURES ......... 60
Methods.............................. 60
Procedures.......................... 61
Collection of D a t a ................ 61
Treatment of Data.................. 80
SIMILARITY AND DIFFERENCE BETWEEN THE
PERSONALITY PROFILES OF TELEVISION
ROLES AND THEIR VIEWERS............. 96
Most-Liked Roles .................... 97
Least-Liked Roles ...................... 100
Granny, a Role Rated as Most Liked by
One Group and Rated a Least-Liked
Role by Another Group................ 102
Summary................................. 106
SEX, AGE, LEVEL OF EDUCATION, AND
STANDARD OF LIVING OF VIEWERS AND
OF R O L E S ............................... 108
i
i
|
I
! CHAPTER
VI.
iv
PAGE
Se x.................................. 109
A g e .................................. Ill
Level of Education ; ............... 115
Standard of Living ......... 120
Summary.............................. 124
PERSONALITY TRAITS THAT MOST
CONSISTENTLY CHARACTERIZED MOST-LIKED
ROLES, LEAST-LIKED ROLES, AND VIEWERS
OF THESE R O L E S ...................... 127
Factor Q3, Most Differentiating
Trait.............................. 128
Personality Traits Appearing Most
Consistently in Most-Liked Roles . 132
Personality Traits Appearing Most
Consistently in Least-Liked Roles . 133
Personality Traits on Which Roles and
Their Viewers Were Most Alike . . . 136
Personality Traits on Which Roles and
Viewers Differed Most............. 140
Personality Traits Characterizing
Viewer Groups of Granny Clampett:
One Which Rated Her a Most-Liked
Role and One Which Rated Her a
Least-Liked R o l e ................. 147
V
CHAPTER PAGE
i
Summary.......................... 149
VII. SUMMARY, CONCLUSIONS, IMPLICATIONS,
AND RECOMMENDATIONS............. 151
Summary.......................... 151
Conclusions and Implications .... 174
Recommendations for Further Study . . 186
BIBLIOGRAPHY . .............................. 191
APPENDIXES
A. PERSONAL INFORMATION S H E E T ....... 200
B. TELEVISION ROLE PERSONALITY
QUESTIONNAIRE (COVER SHEET).... 202
C. TELEVISION VIEWING D I A R Y ......... 205
D. TELEVISION ROLE PREFERENCE
QUESTIONNAIRE................... 208
E. SIXTEEN PERSONALITY FACTOR TEST
PROFILES (ROLES AND VIEWERS) .... 213
LIST OF TABLES
TABLE
I.
II.
III.
IV.
V.
VI.
VII.
VIII.
IX.
Distribution of Sex of Sample .........
Distribution of Age of Sample .........
Distribution of Standard of Living
of Sample ............................
Distribution of Level of Education
of Sample . . . ......................
Number of Persons Who Judged Each Role
on the 16 P. F. Questionnaire ....
Coefficients of Pattern Similarity
between the Mean Personality Profiles
of Most-Liked Roles and Their Viewers
Coefficients of Pattern Similarity
between the Mean Personality Profiles
of Least-Liked Roles and Their
Viewers ..............................
Coefficients of Pattern Similarity
between Viewers of Granny, Most
Liked, and Granny, Least Liked, and
Their Perceptions of Granny, Most
Liked, and Granny, Least Liked . . .
Proportion of Males and Proportion of
Females in Viewer Groups ...........
PAGE
63
63
64
64
65
99
101
104
110
TABLE
X.
XI.
XII.
XIII.
XIV.
XV.
XVI.
XVII.
Product Moment Correlations between
Personal Data of Viewers and of
Roles Viewed .......................
Distribution of Ages of Viewer Groups
and of Role Groups .................
Distribution of Level of Education of
Viewer Groups and of Role Groups . .
Distribution of Standard of Living of
Viewer Groups and of Role Groups . .
Outstanding Personality Traits of
Most-Liked Roles and of Least-Liked
Roles............ ..................
Personality Traits on Which Roles and
Viewers Differed No More Than One
. Sten ................................
Personality Traits on Which Roles and
Viewers Differed More Than Three
Stens ................................
Outstanding Personality Traits of
Viewers Who Rated Granny a Most-Liked
Role and Viewers Who Rated Granny a
Least-Liked Role ...................
PAGE
112
i
114 !
i
j
117 I
122 i
j
I
129
j
i
137
141
148
CHAPTER I
THE PURPOSE AND PROBLEM
Some elements in the viewing of television may
operate to attract viewers to certain television
programs and at the same time cause them to be repelled
by other programs. Such elements might include story,
setting, interrelationship of roles, clothing, or
mannerisms.
As several writers have reported, identification
seems to be an important element in people's perception
of roles that appear in the mass media.* Few of these
studies have employed the use of quantitative measure
ment to establish the extent 'to which the viewer iden
tifies himself with the roles in those programs that he
views. Use of such a measurement might lend valuable
support to conclusions already reported in the litera-
Rudolf Arnheim, "The World of the Daytime
Serial," in Radio Research. 1942-1943. edited by Paul
F. Lazarsfeld and Frank N. Stanton (Mew York: Duell,
Sloan and Pearce, 1944), pp. 55-57; Paul G. Cressey,
"The Motion Picture Experience as Modified by Social
Background and Personality," American Sociological
Review, 3:4:519, August, 1938; F. E. Emery, "Psycho
logical Effects of the Western Film: A Study in Tele
vision Viewing: II. The Experimental Study," Human
Relations. 12:3:222, 1959; Leo A. Handel, Hollywood
Looks al~Its Audience: A Report of Film Audience
Researc5 (Urbana: The University of Illinois Press,
1966), p. 145; Muzafer Sherif and Hadley Cantril, The
Psychology of Ego-Involvements: Social Attitudes and
Identifications (New York: John Wiley and Sons,
Incorporated, 1947), p. 351.
ture, and might also provide some new insights into the
relationships of personalities of viewers and viewers'
selections of particular television programs.
I. THE PURPOSE
The purpose of this study was to provide further
understanding of the importance of the personality of
the viewer of television and his attraction to certain
regularly viewed television roles, and thereby to
certain programs. It contains information on some of
the relationships that existed between television
viewers and television roles that were regularly viewed
by these viewers. More specifically, the study was
designed to help to determine if there was not greater
similarity of personality between television viewers
and the perceived personality of specific television
roles that they liked most them there was between the
personality of television viewers and the perceived
personality of television roles that they liked least.
Riley and Riley, among other authorities on
communication, stated that more attention should be
given to the recipient in the communication situation
than to the message or the environment in which the
message is delivered. They suggested:
3
. . . the traditional view does not take fully into
account ongoing processes of social interaction of
which the single communicative act is merely one
component. Nor does it take into full account those
psychological processes which, although they may be
going on within the individual recipient quite apart
from any particular communication, may nevertheless
markedly affect his reaction to it.22 Extensions
of this view in both the sociological and the
psychological directions seem necessary if the
mass-communications process is to be explained more
adequately or its outcome predicted more accurately.
22
As Lazarsfeld and Merton pointed out in 1943,
"Propaganda will not produce the expected response
unless its content corresponds to the psychological
wants of the audience" (Robert K. Merton, Social
Theory and Social Structure. rev., Free Press, 1957,
p. 519). A few years later, Berelson expanded on
this point:
. . . Effects upon the audience do not follow
directly from and in correspondence with the
intent of the communicator or the content of
the communication. The predispositions of the
reader or listener are deeply involved in the
situation, and may operate to block or modify
the intended effect or even to set up a boom
erang effect.
(Bernard Berelson, "Communications and Public
Opinion," in Wilbur Schramm [ed.], Communications
in Modern Society.^University of Illinois Press,
1946, pp. 183-84.)
It was becoming apparent that television programs
were viewed more by distinct, well-defined groups than
by audiences of a general make-up.
^John W. Riley, Jr. and Matilda White Riley,
"Mass Communication and the Social System," in Sociology
Today: Problems and Prospects, edited by Robert K.
Merton, Leonard Broom, and Leonard S. Cottrell, Jr.
under the auspices of the American Sociological Society
(New York: Basic Books, Incorporated, 1959), p. 543.
According to Glick and Levy, a broadcaster
should know who the viewers of his programs are, what
their relationships to the programs are, why they tend
to favor them, and the particular kinds of gratifica
tions they seek from them. The authors suggested that
in the future it would be most meaningful to conceive
of a "core audience”— a group of people who respond
3
fully and reliably to each program. Earlier in a
study of motion picture audiences it was reported that
"most stars do not gain support equally from all groups
of theatergoers, but appeal particularly to certain
segments of the population."
In an early report on radio listening, Stanton
said:
It isn't enough to know what programs are heard
and preferred. We want to know why they are
listened to and liked, and, furthermore, we want
to quantify their influence.5
Ira 0. Glick, Sidney J. Levy, and others,
Living with Television (Chicago: Aldine Publishing
Company, 1962), p. 38.
4
Sherif and Cantril, loc. cit., citing Audience
Research Institute, Incorporated."Increasing Profits
with Continuous Audience Research (New York: Am. Book,
1941), reprinted by permission of G. H. Gallup.
5
Frank N. Stanton, "Psychological Research in
the Field of Radio Listening," in Educational Broad
casting. 1936: Proceedings of the First National
J
5
In another early study of radio listening Beville
stressed that "... the educational or political broad
caster should be deeply concerned with the composition
of his audience as a factor which seriously affects the
g
achievement of his objectives."
In 1963 Broadcasting magazine carried comment on
an article by Murdoch and others to the effect that
"the diversification of interests among the public is
having a more drastic effect on mass media than most
7
people realize." According to the writer in Broad
casting. radio almost died in the early 1950's. Then,
realizing it could not compete with television, radio
began to specialize its programing. This article
recorded that many people were turning from television
in growing protest against the kind of programing which
5 (continued)
Conference on Educational Broadcasting. Held in Wash
ington. D.C., on December 10. 11. and 12. 19567 edited
by C. S.~Marsh~TChicago: The University of Chicago
Press, 1937), p. 373.
6H. M. Beville, Jr., "The ABCD's of Radio
Audiences," The Public Opinion Quarterly. 4:2:206,
June, 1940.
7
Lawrence C. Murdoch, Jr., and others, "Mass
Media and the Gulliver Fallacy," Business Review.
October, 1962, published by Federal Reserve Bank of
Philadelphia, as reported in "Is the Mass Audience
a Thing of the Past?" Broadcasting. 64:2:42-43,
January 14, 1963.
6
I
i
j seemed to appeal to a single set of tastes and inter
ests.®
Murdoch was alleged to have predicted that radio
stations which appealed to selected audiences would
Q
continue in good financial health.
It was hoped that the present study might help
producers of television programs to understand better
the relationships that exist between the roles in their
programs and the viewers who do or do not like to view
these roles. Also, it was hoped that the data reported
might be of value to producers who want to reach a
specific, well-defined audience— viewers who will view
their programs with regularity.
While it was recognized that other elements in
the viewing situation, such as availability of a pro
gram and manifest characteristics of a program, influ
ence viewing behavior, it was believed that the dynamic
personal interaction that takes place between viewers
and roles viewed by them is of primary importance in
understanding why viewers tend to select certain pro
grams for regular viewing and not others. --------
If the findings of this study permit reliable
and valid prediction of the selection of programs, they
8Ibid., p. 43 9Ibid.
should be of value to the following groups that seek
to build and retain regular viewing audiences:
1. Producers, writers, and directors of tele
vision programs who wish to build into their
programs roles that will interest specific
audiences and attract viewers to these
programs.
2. Sponsors of television programs who wish to
identify specific viewing groups which might
have interest in the products or services of
these sponsors.
The findings should also be of value to:
1. Students of social behavior who are inter
ested in the use that persons make of
television in their effort to secure rein
forcement of personal values and character
structure.
2. Students of the use of mass media who are
searching for quantitative data to serve
as evidence for the identification that is
thought to exist between the user of mass
media and those roles in mass media to which
the user is particularly attracted.
Prior reports on use of mass media suggested
that persons were attracted to roles in which they
perceived certain similarities of personality or of
circumstance to their own personalities or circum
stances.^ Little in this vein of research was reported
in studies of television viewing. With the exception
of those by Emery and Fearing, such studies as were
reported did not describe quantitatively the full range
of elements in personality which may be operative in
the viewing situation.
The present study was designed to provide quanti
tative measurement of the degree of similarity or dis
similarity between the personalities of the viewers and
the perceived personalities of roles viewed by them.
Personality was defined as one's unique pattern of
traits. A trait was described as any distinguishable,
relatively enduring way in which one individual differs
from another.
Cattell's personality inventory distinguishes
sixteen first-order trait factors.*1 Through use of
this instrument certain similarities and dissimilarities
between personality profiles were described. These
^Arnheim, oj>. cit., p. 56; Handel, loc. cit.;
Sherif and Cantril, loc. cit.; Emery, loc. cit.
1:1IPAT 16 P. F. (Form A, 1962 Edition. The
Institute for Personality and Ability Testing, 1602-04
Coronado Drive, Champaign, Illinois).
trait factors were taken to be relatively enduring
patterns of organization of personality which provided
information of more lasting value than directly inter
preted statements of opinion.
II. THE PROBLEM
This study was designed to provide information
on the following questions:
1. Is there greater similarity or difference
between the personality profiles of tele
vision viewers and the perceived personality
profiles of regularly-viewed, most-liked
television roles than there is between the
personality profiles of television viewers
and the perceived personality profiles of
least-liked television roles?
2. Is there greater similarity or difference
between television viewers and their per
ception of most-liked television roles than
there is between television viewers and
their perception of least-liked television
roles with respect to age, level of educa
tion, standard of living, and sex?
The first question, that of similarity or dif
ference between personality profiles, comprised the
10
major hypotheses of the study, while the second
question, about personal data, comprised the minor
hypotheses.
To provide a framework for statistical examina
tion of these data, the following specific questions
were structured.
Similarity or Difference
of Personality Profiles
Is there a significant similarity or difference
between:
1. The mean personality profile of a group of
viewers and the mean personality profile of
a role as perceived by these viewers and
rated by them as a most-liked role?
2. The mean personality profile of a group of
viewers and the mean personality profile of
a role as perceived by these viewers and
rated by them as a least-liked role?
3. The mean personality profile of a group of
viewers who rated a particular role as a
most-liked role and the mean personality
profile of a group of viewers who rated
the same role as a least-liked role?
11
Similarities or Differences
of Personal Data
Sex. Is there a significant similarity or
difference between:
1. The sex of the viewers and the sex of a
most-liked role judged by these viewers?
2. The sex of the viewers and the sex of a
least-liked role judged by these viewers?
3. The sex of the viewers who rated a particular
role as a most-liked role and the sex of the
viewers who rated the same role as a least-
liked role?
Age. Is there a significant similarity or
difference between:
1. The age of the viewers and the age of a
most-liked role as perceived by these
viewers?
2. The age of the viewers and the age of a
least-liked role as perceived by these
viewers?
3. The mean age of a group of viewers who
rated a particular role as a most-liked
role and the mean age of a group of viewers
who rated the same role as a least-liked
role?
Level of education. Is there a significant
similarity or difference between:
1. The level of education of a group of viewers
who rated a role as a most-liked role and.
the level of education of this role as per
ceived by these viewers?
2. The level of education of a group of viewers
who rated a role as a least-liked role and
the level of education of this role as per
ceived by these viewers?
3. The level of education of a group of viewers
who rated a role as a most-liked role and
the level of education of a group of viewers
who rated the same role as a least-liked
role?
Standard of living. Is there a significant
similarity or difference between:
1. The median annual family dollar income of
a group of viewers who rated a role as a
most-liked role and the standard of living
of this role as rated by these viewers on
a five point scale?
2. The median annual family dollar income of
a group of viewers who rated a role as a
least-liked role and the standard of living
of this role as rated by these viewers on
a five point scale?
3. The median annual family dollar income of
a group of viewers who rated a role as a
most-liked role and the median annual family
dollar income of a group of viewers who
rated the same role as a least-liked role?
III. DEFINITIONS
A viewer personality profile consisted of that
pattern of scores obtained after a subject answered
questions in the Sixteen Personality Factor Question-
12
naire (16 P. F.)» Form A, with respect to his own
preferences and opinions.
A role personality profile consisted of that
pattern of scores obtained after a subject answered
questions in the 16 P. F.., Form A, as he thought they
would be answered by a particular television role.
A personality profile was a profile of scores.
A profile of scores was the pattern of sixteen
different scores of traits which represented the
responses of one person.
12Ibid.
14
The score of a trait was the sum of weighted
values of responses to one of sixteen combinations
of statements in the 16 P. F.
A trait was any distinguishable and relatively
enduring way in which one individual differed from
another and was represented by any one of sixteen
separate scores obtained on the 16 P. F.
A viewer mean personality profile was a pattern
of means of scores from the profiles of several persons
who had judged a single role.
A role mean personality profile was a pattern
of means of scores from the profiles obtained from
several judgments of a single role.
A most-liked television role was one of ten
roles chosen by a viewer from a list of best-liked
roles and listed in order of preference as most liked,
next most liked, etc.
A least-liked television role was one of ten
roles chosen by a viewer from a list of less-liked
roles and listed in order of preference as least liked,
next least liked, etc. The terms "most liked" and
"least liked" did not mean that these roles were
absolutely liked or absolutely disliked, but were
only relative terms.
15
A regularly featured role was a role that was
listed on a station's program data sheet as a starring
role or a featured role, or was so designated by the
station's program department personnel.
Similarity of personality profiles was the degree
of likeness between two sets of profile scores. This
likeness was represented by a positive correlation
coefficient of pattern similarity (r ).
Difference between personality profiles was the
degree of dissimilarity between two sets of profile
scores. This difference was represented by a negative
correlation coefficient of pattern similarity (rp)*
Age of viewer was that age stated by the subject
on the Personal Information sheet (Appendix A).
Age of role was that estimate of the age of a
television role stated by the viewer on the Television
Role Personality Questionnaire for a particular role
(Appendix B).
Level of education of a viewer was the last
year of schooling completed as stated by the viewer
on the Personal Information sheet.
Level of education of a role was that estimate
of the highest grade in school completed by a specific
television role as stated by the viewer on the Tele
vision Role Personality Questionnaire for a particular
role.
Standard of living of a viewer was the median
annual dollar income for a family as stated on the
Personal Information sheet.
Standard of living of a role was that estimate
of a role's standard of living as stated by a viewer
on the Television Role Personality Questionnaire by
checking one of five categories: very low, rather low,
average, rather high, very high.
IV. HYPOTHESES
To satisfy the problem, the following null
hypotheses were posited:
Similarity or Difference
of Personality Profiles
1. The rp (coefficient of pattern similarity)
obtained between the mean personality
profile of a group of viewers and the mean
personality profile of a role which was
rated by these viewers as most liked will
be zero.
2. The r^ obtained between the mean personality
profile of a group of viewers and the mean
personality profile of a role which was
rated by these viewers as least liked will
be zero.
3. The r^ obtained between the mean personality
profile of a group of viewers who rated a
particular role as most liked and the mean
personality profile of a group of viewers
who rated the same role as least liked will
i
be zero.
i
Similarities or Differences
of Personal Data
Sex:
1. The r (product-moment coefficient of correla
tion) obtained between the sex of a group of
!
viewers and the sex of a role which was rated •
by these viewers as most liked will be zero.
2. The r obtained between the sex of a group of
viewers and the sex of a role which was rated !
by these viewers as least liked will be zero, j
i
3. The r obtained between the sex of a group of |
viewers who rated a particular role as most
liked and the sex of a group of viewers who
rated the same role as least liked will be
zero.
18
Age;
1. The r obtained between the ages of a group
of viewers and the perceived age of a role
which was rated by these viewers as most
liked will be zero.
2. The r obtained between the ages of a group
of viewers and the perceived age of a role
which was rated by these viewers as least
liked will be zero.
3. The r obtained between the mean ages of a
group of viewers who rated a particular
role as most liked and the mean ages of a
group of viewers who rated the same role
as least liked will be zero.
Education:
1. The r obtained between the level of education
of a group of viewers and the perceived level
of education of a role which was rated by
these viewers as most liked will be zero.
2. The r obtained between the level of education |
of a group of viewers and the perceived level
of education of a role which was rated by
these viewers as least liked will be zero.
3. The r obtained between the level of education
of a group of viewers who rated a particular
role as most liked and the level of education
of a group of viewers who rated the same role
as least liked will be zero.
Standard of living:
1. The r obtained between the standards of
living of a group of viewers and the standard
of living of a role which was rated by these
viewers as most liked will be zero.
2. The r obtained between the standards of
living of a group of viewers and the standard
of living of a role which was rated by these
viewers as least liked will be zero.
3. The r obtained between the standards of
living of a group of viewers who rated a
particular role as most liked and the
standard of living of a group of viewers
who rated the same role as least liked will
be zero.
V. ASSUMPTIONS
It was assumed that:
1. Subjects did view those programs which they
entered in the Television Viewing Diary
(Appendix C).
20
2. Subjects did prefer those roles listed as
most liked roles in the Television Role
Preference Questionnaire (Appendix D).
3. Subjects liked least those roles which were
listed as least liked roles in the Television
Role Preference Questionnaire.
4. If a subject listed a program (in the Tele
vision Viewing Diary) for at least two of
the three viewing weeks he did view with
some regularity those roles which regularly
appeared in that program.
5. A subject could distinguish between most-
liked roles and least-liked roles, although
it was not assumed that there was a constant
distance between the ranks which they
assigned to the roles.
6. If a role remained unchecked on the Tele
vision Role Preference Questionnaire the
viewer was either unfamiliar with the role,
had no opinion on the role, or had an opinion
so weak that he could not check either
preference.
7. Each viewer had some concept of ways in
which a role might react to various life
situations and for this reason could imagine
how a particular role would answer a per
sonality inventory questionnaire.
VI. LIMITATIONS
The following limitations were placed upon the
data collected:
1. The population sample was limited to couples
who were aged approximately twenty to fifty
years, lived in the greater Los Angeles
area, and were participants in the young
adult Mariner's program of the United
Presbyterian Church, U.S.A.
2. Television diary data were limited to three
weeks of viewing time recorded over a period
of six weeks, with a one-week interval
between each week of recorded viewing.
3. Only those television roles which were listed
as having been viewed during at least two*,of
the three diary-keeping weeks and were listed
as most liked in the Television Role Prefer- >
ence Questionnaire were included in the study '
as most-liked roles. However, any role that
was listed by the viewer in the Television
Role Preference Questionnaire as a least-liked
role could be admitted for judgment, regard-
less of how many times it had been viewed
during the three weeks of diary keeping.
It was believed that viewers would not view
with any regularity those roles which were
listed by them as least-liked roles.
Roles included for study were chosen from
those television programs that (a) were
viewed between 6:00 P.M. and 12:00 P.M., and
(b) contained roles that appeared each week
or nearly each week on programs of a dramatic
fiction format.
While variables such as program setting and
type of action undoubtedly helped to account
for the regular viewing of particular pro
grams by some viewers, it was believed that
the perceived personality of the roles which
they viewed was of major importance to the
repeated viewing of programs. The only
other variables included for observation
were sex, age, level of education, and
standard of living.
Those persons who did not return all six
mailings exhibited an abnormal degree of
tenacity and therefore could not necessarily
be considered representative of the general
population.
23
VII. ORGANIZATION OF REMAINDER
OF THE STUDY
Chapter II presents a review of the literature.
Chapter III presents the operational plan of
the study including methods, procedures, and techniques
used in the collection and treatment of data.
Chapters IV, V, and VI present findings of the
study. Chapter IV reports similarities and differences
found between the personality profiles of television
roles and their viewers. Chapter V presents findings
on the personal variables of sex, age, level of educa
tion, and standard of living for television roles and
their viewers. Chapter VI discusses personality traits
that most consistently characterized most-liked roles,
least-liked roles, and viewers of these roles.
Chapter VII includes a summary of the entire
study together with conclusions, implications, and
recommendations.
CHAPTER II
REVIEW OF THE LITERATURE
Literature from several different but related
fields of study was searched to provide the foundation
upon which the hypotheses of this study were based.
Studies that were reviewed were drawn not only from
the field of communication, i.e., radio, television,
motion pictures, and mass media in general, but also
from the fields of psychology and sociology, with
particular attention given to studies in perception
and in the definition and measurement of personality.
Information gleaned from the above sources was
divided into six parts:
1. Television viewing: a para-social situation
2. Perception of persons: an important element
in television viewing
3. Personality: a central determinant in the
perception of persons
4. Functional and dysfunctional values of
television viewing
5. Identification: an important element of
television viewing
6. Program selection: a result of role-viewer
identification
25
I. TELEVISION VIEWING: A
PARA-SOCIAL SITUATION
Freidson, among others, contended that perception
and selection of particular content in mass media were
not adequately explained through examination of indi
vidual behavior, but only as the act of selection was
observed and analyzed as a type of social action.1
As a part of one's social environment television
viewing provides a type of face-to-face encounter
between the viewer and television roles that are viewed.
In this situation the viewer may have opportunity to
observe the full range of physical, emotional, and
social elements that are transmitted from the television
screen and to form perceptual judgments about the roles
that are being viewed.
This kind of interpersonal relationship between
viewer and television performer was called a para-social
relationship by Horton and Wohl. They claimed:
One of the striking characteristics of the new
mass media— radio, television, and the movies— is
that they give the illusion of face-to-face rela
tionship with the performer. The conditions of
Eliot Freidson, "Communications Research and
the Concept of the Mass," American Sociological Review.
18:3:315, June, 1953.
26
response to the performer are analogous to those
in a primary group.2
This personal involvement of the spectator with
roles perceived in various mass media has been reported
by students of mass media audience research for many
years.
In 1935 Cantril and Allport reported:
The immediacy and reality of the radio voice
make it a quasi-personal stimulus. It is responded
to in~much the same way as are voices in natural
life.
In 1937 Cantril made a study of the radio program
"America's Town Meeting of the Air" and reported: "A
radio broadcast must be regarded as a social stimulus
since it arouses in the listener definite social
behavior."4 He claimed, however, that "the radio
situation is less complete than the face-to-face
situation.
o
Donald Horton and R. Richard Wohl, "Mass Com
munication and Para-Social Interaction: Observations
on Intimacy at a Distance," Psychiatry, 19:3:215,
August, 1956.
3Hadley Cantril and Gordon W. Allport, The
Psychology of Radio (New York: Harper and Brothers
Publishers, 1935), p. 259.
4Hadley Cantril, "A Comparative Study of Radio
and Face-to-Face Stimulus Situations," The Journal of
Social Psychology, 8:4:452, November, 1937.
5Ibid.
In 1938, reporting study of the motion picture
experience as modified by social background and per
sonality, Cressey stated:
. . . the cinema serves chiefly to set up imagina
tive states. In these, imaginative participation
takes the place of social participation and
identification becomes the means by which a
semblance of vitality and substance can be dis
covered in the movies.6
In 1947, in a study of the influence of motion
pictures on the attitude and behavior of the audience,
Fearing reported:
. . . motion pictures afford an opportunity for
the expression of the basic meanings inherent in
the relationships of human beings to each other,
to their environment, and to the society of which
they are a part. This is not limited to a passive
reflection of those meanings, but may be a dynamic
and creative interpretation.?
Although the concept of the para-social nature
of television viewing as proposed by Horton and Wohl
is not new, it has provided a logical foundation on
which to stand as a new look is given at the relation
ship between the personality of television viewers
6
Paul G. Cressey, "The Motion Picture Experience |
as Modified by Social Background and Personality,"
American Sociological Review, 3:4:519, August, 1938.
7
Franklin Fearing, "Influence of the Movies on
Attitudes and Behavior," The Motion Picture Industry,
edited by Gordon S. Watkins, The Annals of the American
Academy of Political and Social Science, 254:79,
November, 1947.
28
and the perceived personality of television roles
viewed by them.
In this study personality was conceived of not
only as an index to certain unique characteristics
of the individual, but also as a residue of socially
acquired values and behavioral patterns of response.
Thus, when a person observes another's selection of
mass media content and its relationships to personality,
one observes both the unique attributes of the other
individual as they affect his selection of content and
also broader socially conditioned responses.
II. PERCEPTION OF PERSONS: AN IMPORTANT
ELEMENT IN TELEVISION VIEWING
In television viewing people tend to perceive
television roles much as they would perceive persons
in real life. This is especially true of roles that
appear in the context of a life situation dramatic
format.
Person perception was defined by Shrauger and
Altrocchi as the "attribution of psychological char
acteristics (e.g., traits, intentions, emotions) to
Q
other people . . . ."
Q
Sid Shrauger and John Altrocchi, "The Person
ality of the Perceiver as a Factor in Person Percep
tion," Psychological Bulletin. 62:5:290, November,
1964.
29
Tagiuri and Petrullo reported:
. . . when we speak of person perception or of
knowledge of persons, we refer mostly to the
observations we make about intentions, attitudes,
emotions, ideas, abilities, purposes, traits— «
events that are, so to speak, inside the person.
Person perception is special . . . in that the
similarity between perceiver and perceived is
greater than in any other instance. . . . This
may be viewed as empathy or projection or whatnot.
Whatever it is called, it is critical in the sense
that the perceiver's repertory of categorizing
responses is somehow limited by his own personal
or vicarious experience as a person.10
In other words, people can be confronted with
the same person in the same context, and yet tend to
perceive that person differently. Consideration of
this idea suggested need for study of the perceiver
.in television viewing to determine what elements most
influence his perception of television roles.
Gage and Cronbach, in a study of interpersonal
perception, reported:
. . . in the bulk of research to date, social
perception as measured is a process dominated
far more by what the Judge brings to it than
by what he takes in during it. His favorability
Renato Tagiuri and Luigi Petrullo (editors),
Person Perception and Interpersonal Behavior (Stanford:
Stanford University Press, 1958), p. x.
^ Ibid., p. xi.
toward the Other, before or after he observes
the Other, and his implicit personality theory,
formed by his experiences prior to his interaction,,
with the Other, seem to determine his perceptions.
Riley and Flowerman reported that different
groups of children, although exposed to the same kind
of content, often put different interpretations on it,
liked it for different reasons, and extracted different
12
values from it.
Sherif reported that social factors may cause
persons of different cultures to react differently
although the persons are faced with the same stimulus
situation.*3
Murphy claimed:
. . . wherever our needs differ we literally see
differently. Much of the process of individual
perception depends on the force of past wants,
the person's need to disentangle and restructure
N. L. Gage and Lee J. Cronbach, "Conceptual
and Methodological Problems in Interpersonal Percep
tion," The Psychological Review, 62:6:420, November,
1955.
12
Matilda White Riley and Samuel H. Flowerman,
"Group Relations as a Variable in Communications
Research," American Sociological Review, 16:2:174,
April, 195i:
17
Muzafer Sherif, A Study of Some Social Factors
in Perception, Archives of Psychology, edited by R. S.
Woodworth, 27:187:52, July, 1935.
in terms of the situations with which he has had
to cope.
Stating Murphy's ideas another way, ffilner said
that the perception of objects is a function of motiva
tional characteristics of the perceiver as well as a
function of the absolute characteristics of the
stimulus. He concluded that attitudinal organization
of the individual is an important determiner of social
15
reactions which are largely perceptual at their core.
According to Bruner, perception takes place as
a three-step cycle which begins with an expectancy or
hypothesis.
Any given hypothesis results from the arousal of
central cognitive and motivational processes by
preceding environmental states of affairs.
The second analytic step in the perceiving
process is the input of information from the
environment (which environment includes the
stimulus complex brought to us by distance
receptors and by the somatic senses).
14
Gardner Murphy, Personality: A Biosocial
Approach to Origins and Structure (New York: Harper
and Brothers Publishers, 1947), pp. 338-339.
1 5
Daniel M. Wilner, "Attitude as a Determinant
of Perception in the Mass Media of Communication:
Reactions to the Motion Picture, Home of the Brave"
(unpublished Doctoral dissertation, University of
California at Los Angeles, October, 1950), pp. 86-92.
32
The third step in the cycle is a checking or
confirmation procedure. Input information is
confirmatory to or congruent with the operative
hypothesis, or it is in varying degree infirming
or incongruous. If confirmation does not occur,
the hypothesis shifts in a direction partly
determined by internal or personological or
experiential factors and partly on the basis of
feedback from the learning which occurred in the
immediately preceding, partly unsuccessful
information-checking cycle.
Postman and Bruner, in conjunction with others,
defined perception as "goal-directed behavior." They
said:
The goal of perception . . . is the construction
of a meaningful behavioral environment— an environ
ment congruent with "reality" on the one hand and
the needs and dispositions of the organism on the
other.
To construct his perceptual world man requires an
elaborate sequence of hypotheses through which he
seeks out, accepts, and rejects what is percep
tually demanded by his needs, by his reality
situation, and by his past experience. . . . He
learns to eliminate from his perceptual field
what is extraneous to him and to encompass what
is important . . . .I7
| g
Jerome S. Bruner, "Personality Dynamics and
the Process of Perceiving," in Perception: An Approach
to Personality, by Robert R. Blake, Glenn V. Ramsey,
and others (New York: The Ronald Press Company, 1951),
pp. 123-124.
17
Leo Postman and Jerome S. Bruner, "Perception
under Stress," The Psychological Review, 55:6:314,
November, 1948, citing the following: Jerome S. Bruner
and C. C. Goodman, "Value and Need as Organizing
Factors in Perception," Journal of Abnormal and Social
Psychology. 42:1:33-44, January,~T947; Jerome S. Bruner
33
Tagiuri and Petrullo reported that there are
three major elements which influence the process of
person perception: ". . . the situation in which the
person to be judged is embedded. . . . the person,
apart from the situation. . . . the perceiver him
self."18
The several studies cited thus far in the review
of literature indicated that the concept of personality
was an important element in the perception of persons
and would be a useful tool for study of the perception
of television roles by their viewers.
17 (continued)
and Leo Postman, "Tension and Tension Release as
Organizing Factors in Perception," Journal of Per
sonality, 15:4:300-308, June, 1947; Jerome S. Bruner
and Leo Postman, "Symbolic Value as an Organizing
Factor in Perception," The Journal of Social Psy
chology , 27:Second Half:203-208, May, 1948; Jerome
S. Bruner and Leo Postman, "Emotional Selectivity in
Perception and Reaction," Journal of Personality,
16:1:69-77, September, 1947; Eugenia Hanfmann, Morris
I. Stein, and Jerome S. Bruner, "Personality Factors
in the Temporal Development of Perceptual Organiza
tion," The American Psychologist. 2:8:284-285, August,
1947; Leo Postman, Jerome S. Bruner, and Elliott
McGinnies, "Personal Values as Selective Factors in
Perception," The Journal of Abnormal and Social
Psychology, 43:142-154, 1948.
1 f t
Tagiuri and Petrullo, op. cit., pp. xiii-xiv.
34
III. PERSONALITY: A CENTRAL DETERMINANT
IN THE PERCEPTION OF PERSONS
Description of personality provides a kind of
global index with which the nature of a person is
revealed. With this index it is possible for one
person to make certain judgments about the values and
general response patterns of another. The study of
personality has grown in importance throughout the
last decade, and has passed through several historical
phases.
According to Cattell, systematizing of human
knowledge about personality falls into three historical
phases: (1) the literary and philosophical; (2) the
stage of organized observation and theorization, or
proto-clinical; and (3) the quantitative and experi
mental, which began just prior to the twentieth
century.
The quantitative and experimental phase of
personality study splits into two distinct parts:
(1) univariate experiments, such as the Pavlovian
behavioral experiments in Russia; and (2) multi
variate experiments which found their origin with
^Raymond B. Cattell, The Scientific Analysis
of Personality (Baltimore: Penguin Books, Incorporated,
X965), pp. 13-14.
35
20
Sir Francis Galton and Karl Pearson.
In 1950 Harsh and Schrickel reported:
The earlier view of personality as a kind of
substance with inhering properties or traits is
giving way to a dynamic conception wherein per
sonality is viewed not as a thing, entity, or
substance but as a complex of functions of
interactive forces which may be studied within
various abstractive frames of reference—
physical, chemical, hormic, neurological,
psychological, social, and cultural.21
A definition of personality involves both
integration and uniqueness, according to Rosenzweig.
In studying personality the psychologist is
. . . interested in knowing how the individual
in the expression of his needs and in his rela
tionships with other individuals functions as a
recognizable unit that possesses certain dis
tinctive traits, drives, attitudes and habits,
and attains or fails to attain an adjustment to
himself and to his environment.22
Earlier psychologists described personality with
23
long lists of numerous traits, while more recent
20Ibid., pp. 19-20.
21Charles M. Harsh and H. G. Schrickel,
Personality: Development and Assessment (New York:
The Ronald Press Company, 1950), pp. 357-358.
22
Saul Rosenzweig, "Available Methods for
Studying Personality," The Journal of Psychology.
28:Second Half:345, October, 1949.
23
Gordon W. Allport and Henry S. Odbert,
Trait-Names: A Psycho-lexical Study. Psychological
Monographs. edited by Joseph Peterson and others,
47:1(Whole No. 211):38-171, 1936.
36
students of factor analysis attempted to measure
similarities between people by reducing the number of
traits to a minimum. Two names stand out in the
research literature of personality through factor
analysis— Guilford and Cattell.
Guilford defined personality as one's "unique
24
pattern of traits," and defined trait as "any dis
tinguishable « relatively enduring way in which one
individual differs from others.
Traits can be divided into classes such as
needs, interests, attitudes, temperament, aptitudes,
26
physiology and morphology.
According to Guilford a trait concept should
refer to some demonstrable unity in personality, be
as exact as possible, and be capable of integration
27
into a general theory of personality. Guilford
suggested:
The list of trait concepts should be an
economical one. We should aim at a minimum
number that will serve our purposes.
24J. P. Guilford, Personality (New York: McGraw-
Hill Book Company, Incorporated, 1959), p. 5.
25Ibid., p. 6. 26Ibid., p. 7, fig. 1.1.
27Ibid.. pp. 86-87.
37
. . • the list shall provide comprehensive coverage
of the phenomena. No significant aspect of person
ality*'' should be slighted . . . in following the
urge to be economical.
. . . there should be as much general agreement as
possible to the list of concepts.
Because it was the purpose of this study to use
measurements that permitted maximum objectivity, and
because the current trend in study of personality
stemmed principally from factor analytic methods,
personality was defined as one's unique pattern of
traits.
In recent years several methods have been sug
gested for extracting personality traits. Rosenzweig
suggested three methods that might be used for investi
gating personality: subjective, objective, and pro-
29
jective. He claimed that the subjective inventory
questionnaire, while sometimes having questionable
validity, probably provides a wider scope of information
30
about personality than any other method.
Cattell suggested that in obtaining some concept
of the "personality sphere," three methods may be used
28Ibid., pp. 87-88.
29
Rosenzweig, op. cit. , p. 346.
30Ibid., p. 350.
38
to observe and record behavior:
. . . (1) By ratings made by observers on the
frequency and intensity with which specific kinds
of behaviour occur in the people they observe.
(2) By questionnaires which are answered by the
person himself, from his own self-observation and
introspection. (3) By objective tests, i.e.
miniature situations se^uiTTor-a person to react
to, in which he does not really know on what
aspect of his behaviour he is being scored ....
In an earlier study Cattell stated:
. . . the discovery of true functional unities in
the total personality requires (a) the use of a
multifactor analysis of the centroid type . . .
and, (b) a mode of rotation after factorization
which will permit each and every factor to have,
if necessary, negative as well as positive load
ings and some obliqueness between factors.32
The Cattell 16 P. F.33 was selected for use in
this study for reasons stated in Chapter III.
IV. FUNCTIONAL AND DYSFUNCTIONAL VALUES
OF TELEVISION VIEWING
Viewing of television serves specific functions
for the viewer. It is used to meet basic personal
31
Cattell, 0£. cit., pp. 60-61.
32
Raymond B. Cattell, "Oblique, Second Order,
and Cooperative Factors in Personality Analysis," The
Journal of General Psychology. 36:6, 1947, citing
G. H. Thomson, The Factorial Analysis of Human Ability
(New York: Longmans, Green, 1939).
33IPAT 16 P. F. (Form A, 1962 Edition. The
Institute for Personality and Ability Testing, 1602-04
Coronado Drive, Champaign, Illinois).
needs such as ego reinforcement and value reinforcement.
Viewing also can provide a type of diversion and
momentary escape from mental or emotional pressures.
Personal needs and values tend to shape the viewer's
perception of what he sees and hears.
In 1956 Pearlin reported that television was
used primarily as an "escape" mechanism with which
persons sought to flee realities and pressures of daily
living.3* In a later study Pearlin reported that while
the mass media "might have long-run dysfunctions," they
do serve a safety-valve function which offers temporary
35
diversion from personal stresses. Katz and Foulkes
objected to the assumption that escapist patterns of
involvement with the media are necessarily dysfunc
tional.33
According to the headnote to an article by
Katz and Foulkes:
34
Leonard Irving Pearlin, "The Social and
Psychological Setting of Communications Behavior: An
Analysis of Television Viewing" (unpublished Doctoral
dissertation, Columbia University, New York, 1956).
35
Leonard I. Pearlin, "Social and Personal
Stresses and Escape Television Viewing," The Public
Opinion Quarterly. 23:2:259, Summer, 1959.
36Elihu Katz and David Foulkes, "On the Use of
the Mass Media as 'Escape': Clarification of a Concept,"
The Public Opinion Quarterly. 26:3:387, Fall, 1962.
40
| . . . there is great need to know what people do
with the media, what uses they make of what the
media now give them, what satisfactions they
enjoy, and, . . . what part the media play in
their personal l i v e s .37
In 1947 Fearing reported on the influence of
38
motion pictures on attitude and behavior. Later he
; extended these ideas to include television as well.
He reported:
Recent research on the so-called "social" factors
in perception has revealed the importance of the
total need-motivational-value-belief system of the
individual as partial determiners of the percept,
or, in fact, any type of human cognition.39
He further reported:
A wealth of experimental evidence indicates . . .
that in all situations the perceptual response is
a resultant of forces from within the organism
acting on the stimulus material itself.40
Under conditions of stress, the individual seeks
meaning in some form of structured experience.
. . . The structurizations . . . may be in the
37Ibid., p. 377.
38
Fearing, oj). cit., pp. 78-79.
39
Franklin Fearing, "Social Impact of the Mass
Media of Communication," Mass Media and Education,
edited by Nelson B. Henry, The Fiffy-Third Yearbook
of the National Society for the Study~"of Education,
Part II (Chicago: The Society, 1954), p. 172.
40Ibid., p. 175.
41
form of affirmations of known and trusted values.
Or he may seek substitute outlets for his frustra
tions, his inhibited aggressions, and his feelings
of inadequacy or social helplessness.41
Fearing claimed, "Whatever form the response to
content may take, covert or overt, it is always dynamic
in the sense that it serves the social-psychological
42
needs of the interpreter."
In 1960 Runciman reported:
. . . a person's tastes control the kind of enter
tainment he chooses. The audience member selects
from the television's offerings those which he
believes to be harmonious with his likes.43
In a study of the stereotyped Western story,
Barker reported that a boy "utilizes the 'western' to
anticipate difficulties, to experiment, and to seek
socially acceptable and tolerable answers to his most
44
urgent and pressing unconscious emotional conflicts."
In a study of the television viewing of children,
Himmelweit reported that differences in program prefer
ence, although slight, "suggest links between people's
41Ibid.. p. 187. 42Ibid., p. 191.
43Alexander P. Runciman, "A Stratification Study
of Television Programs," Sociology and Social Research.
44:4:261, March-April, 19< j07
A A
Warren J. Barker, "The Stereotyped Western
Story: Its Latent Meaning and Psychoeconomic Function,"
The Psychoanalytic Quarterly. 24:2:278, 1955.
42
viewing tastes, leisure interest, attitudes and personal
needs.1,45
In a study of the daytime listening of 100 women,
Herzog found that radio afforded them "an emotional
release. . . . a wishful remodelling of the listener's
'drudgery.' . . . [and] an ideology and recipes for
adjustment."4®
Klapper claimed that studies of use-and-
gratification in mass media research must be pursued
with greater vigor. These studies "must consider not
only the observed use, but the consequences of that
47
use . . . ." He further claimed:
. . . the stipulation of consequences must be
made on the basis of scientific observation or
. . . in terms of hypotheses that may thereafter
be put to scientific test.48
Hilde T. Himmelweit, "A Theoretical Framework
for the Consideration of the Effects of Television: A
British Report," The Uses of Television, edited by
Lotte Bailyn, The Journal of Social Issues, 18:2:25,
1962.
4 6
Herta Herzog, "On Borrowed Experience: An
Analysis of Listening to Daytime Sketches," Studies
in Philosophy and Social Science, 9:1:69, 1941.
4^Joseph T. Klapper, "Mass Communication
Research: An Old Road Resurveyed," The Public Opinion
Quarterly, 27:4:520, Winter, 1963.
48Ibid.
43
Functional analysis should indicate for whom a
program is functional or nonfunctional or dysfunc-
49
tional. Wright suggested that the functional value
of mass media might be ascertained by asking the
following question:
What are the (1) manifest and (2) latent (3)
functions and (4) dysfunctions of mass communicated
(5) surveillance (news) (6) correlation (editorial
activity) (7) cultural transmission (8) entertain
ment for the (9) society (10) subgroups (11) indi
vidual (12) cultural systems?5®
in a study of the television program "The Real
McCoys," Lichty used Wright's model of mass media
functions to describe the functions that this program
51
served for its audiences. He reported several uses
that listeners made of the program, including the
52
reinforcement of values and meanings.
In the present study effort has been made to
analyze the functional values of television viewing
49Ibid.. p. 521.
50
Charles R. Wright, "Functional Analysis and
Mass Communication," The Public Opinion Quarterly.
24:4:610, Winter, 196(T”
1 51
Lawrence W. Lichty, "'The Real McCoys' and
It's CsicJ Audience: A Functional Analysis," Journal
of Broadcasting. 9:2:158-159, 163-164, Spring, 1965.
52Ibid., p. 161.
among a specific group of viewers. It was believed
that the viewing of a role could be either a reinforcing
and gratifying experience for the viewer or it could
be a disagreeable and disturbing experience. The writer
sought to determine, through quantitative measurement
of the personality of the viewer and the viewer's per
ception of the personalities of the roles that he
viewed, whether or not similarity of personality between
viewer and role was associated with the liking of a
role and therefore whether or not the liking of a role
had functional value for the viewer.
V. IDENTIFICATION: AN IMPORTANT
ELEMENT OF TELEVISION VIEWING
Studies of audience research on radio, tele
vision, and motion pictures have described kinds of
identification that take place between members of the
audience and roles that are listened to or viewed.
Through identification with these roles the viewer or
listener receives reinforcement of his value system
or vicarious fulfillment of certain wishes which cannot
be fulfilled in real life.
Cressey claimed that identification could take
three forms:
45
[1] Projection, described by Kimball Young as the
"thrusting upon others imaginarily" . • . "quali
ties which we ourselves possess," .... t23
Intro.iection . . . an intermittent imaginative
adjustment involving momentary loss of both social
orientation and self orientation and is found
among those who experience identification in a
"more complete" form. . . . [33 Displacement . . .
a partial substitution of certain personalities
and values of one's own social world for the
characters and objects in the screen milieu while
continuing, as oneself, to experience imaginative
participation in the screen action. It is a
derived imaginative adjustment, used quasiconsciously
as a means of enhancing the film's affective sig
nificance, and, in contrast to introjection,
involves only social disorientation, not self
disorientation.53
Identification was defined by Drever as
. . . a process by which an individual, uncon
sciously or partially so, as a result of an
emotional tie, behaves, or imagines himself
behaving, as if he were the person with whom
the tie exists.54
Identification was defined by Healy, Bronner,
and Bowers as "the unconscious molding of a person's
own Ego after the fashion of one that has been taken
55
as a model." Symonds provided two definitions of j
53
Cressey, op. cit., p. 520, citing K. Young,
Social Psychology. 135-136, New York, 1930.
54
James Drever, A Dictionary of Psychology
(London: Penguin Books,”1953), p. 125.
55William Healy, Augusta F. Bronner, and Anna
Mae Bowers, The Structure and Meaning of Psychoanalysis;
As Related to Personality and BehaviorTNew York:
Alfred A. Knopf, Incorporated, 1938), p. 240.
46
! the term identification, but reported that "modeling
of oneself in thought, feeling or action after another
person is the most common use of the term." Dymond
stated, "Identification appears to be a very special
kind of role taking; one that is more lasting, less
frequent, and more emotional than is implied in the
57
term empathy."
Identification has long been recognized as an
important motivational force in a person's use of mass
media. In 1944 Arnheim, in a study of the radio daytime
serial, reported, "The listener's evaluation of the plot
and the characters involved will largely depend on whom
she identifies herself with."®® He reported that
resemblance to the role invites identification by the
listener.5®
E C
Percival M. Symonds, Dynamic Psychology
(New York: Appleton-Century-Crofts, Incorporated,
1949), p. 240.
57
Rosalind F. Dymond, "Personality and Empathy,"
Journal of Consulting Psychology. 14:5:343, October,
1950:
5®Rudolf Arnheim, "The World of the Daytime
Serial," in Radio Research. 1942-1943. edited by
Paul F. Lazarsfeld and Frank N. Stanton (New York:
Duell, Sloan and Pearce, 1944), p. 55.
59Ibid.. p. 56.
Schneiderman suggested "that people perceive
each other in terms of their general attitudes,
including their feelings of identification with one
another.
In 1959 Emery reported an extensive theoretical
and experimental analysis of television viewing,
through use of a Western film, with a group of forty-
61
three schoolboys. He found that "what takes place
in the film will, via the self-cum-hero, directly
communicate with the more central regions of the
62
personality (his primary concerns and needs), . . . ."
He stated:
. . . experience in the viewing situation is most
likely to communicate with and influence those
unconscious psychological processes associated
with the permanent ego of a p e r s o n .63
He also suggested:
. . . while there is a general attraction to the
viewing situation there appear to be differences
in the psychological needs or tensions that are
catered for by different programmes, . . . .6*
60
Leo Schneiderman, "Social Perception as a
Function of Identification," The Journal of Psychology.
37:161, 1954.
gl
F. E. Emery, "Psychological Effects of the
Western Film: A~ Study in-Television Viewing: I. The
Theoretical Study: Working Hypotheses on the Psychology
of Television," Human Relations. 12:3:196, 1959.
62Ibid.. p. 204. 63Ibid., p. 208.
64Ibid., p. 198.
48
In the experimental phase of his study Emery
found:
. . . a viewer will tend to identify with the
character most like his own ego-ideal (which
might beeither the sort of person he really 6g
thinks he is or the person that he could become).
He suggested that "personality factors will affect
fifi
both choice of heroes and of themes."
Fearing stated that when a person responds to
a perceptual stimulus, "he is actually seeking meaning
ful organizations— meaningful, that is, in terms of its
congruence with his existing interests, needs and
motives.
Handel, reporting on a study conducted by the
Motion Picture Research Bureau in New York City in 1947,
said that self-identification was most often the reason
that moviegoers tended to prefer players of their own
68
sex.
Sherif and Cantril made a similar statement:
®5F. E. Emery, "Psychological Effects of the
Western Film: A Study in Television Viewing: II. The
Experimental Study," Human Relations, 12:3:222, 1959.
66Ibid.. p. 223.
®7Fearing, "Social Impact . . ."p. 172.
68
Leo A. Handel, Hollywood Looks at Its Audience:
A Report of Film Audience Research (Urbanal The Uni
versity of Illinois Press, 1950), p. 145.
49
. . . the enormous appeal of motion pictures
is due in no small part to the many vicarious
satisfactions they provide.69
. . . individuals choose as their favorite movie
stars people with whom they can most easily
identify themselves— persons of the same sex,
of comparable age, and who tend to be cast in
roles that represent a person of their income
group.70
As early as 1942, in a study of the response of
children to fictional characters, Foulds reported,
"The child's fictional choices are predictable since
they cohere with other observable aspects of his
71
personality.1 1
According to Sherif and Sargent:
When one enjoys a movie, a radio drama or a novel,
one projects himself into the situation and lives
it vicariously through a kind of identification.
. . . the reader or listener is likely also to
CQ
Muzafer Sherif and Hadley Cantril, The
Psychology of Ego-Inyolvements: Social Attitudes
and Identifications (New York; John Wiley find Sons,
Incorporated, 1947), p. 350.
70Ibid., p. 351.
71Graham Foulds, "The Child's Response to
Fictional Characters and Its Relationship to Per
sonality Traits," Character and Personality. 11:75,
September, 1942-June, 1943.
50
identify himself with a leading character in
the plot . . . .72
Although identification takes place between
viewers and television roles that are viewed, Emery
stated that processes associated with the permanent
ego of a person are "partly protected from change by
the selection and identification with the progra*nmes
that best fit the [viewer's] existing ego structure
„73
• • • •
Cressey conceived of "the cinema, not as a
unilateral social 'force,' but as a reciprocal inter-
74
relationship of screen and spectator . . . ."
Through imaginative participation, identification,
random reflection, phantasy before and after
cinema attendance, and through the impact of prior
interests and values, the cinema experience is
redefined in many ways and may affect the patron
in forms only incidentally associated with film
content.75
This attempt on the part of the viewer to
identify with roles and yet to maintain a "psychic
Muzafer Sherif and S. Stansfeld Sargent, "Ego-
Involvement and the Mass Media," Mass Media: Content,
Function, and Measipement, edited by Franklin Fearing,
The Journal of Social Issues, 3:3:12, Summer, 1947.
7**Emery, "Psychological Effects ..." Part I,
p. 208.
Cressey, op. cit., p. 522.
75Ibid.
distance" between himself and the role viewed is a
problem for producers and directors. A film must be
real enough to be plausible and thereby give identifi
cation and pleasure, yet it must not be so real as to
bring pain and self-consciousness to the viewer. It
must attempt to maintain an optimum psychic distance.
Bauer, elaborating upon a theme suggested by
i
Hovland, stated that in the viewing situation an i
audience will select "for interaction those persons |
who will reinforce its beliefs."77 I
Lazarsfeld claimed, "In all the fields that have
been touched by communications research the self-
78
selection of audiences plays a considerable role."
i
Studies reviewed in this section indicated that
i
I
identification was an important element in television |
viewing. They suggested that television viewing
76Ibid., p. 520.
77
Raymond A. Bauer, "The Initiative of the
Audience," Journal of Advertising Research. 3:2:2,
June, 1963, citing Carl I. Hovland, “Reconciling
Conflicting Results derived from Experimental and
Survey Studies of Attitude Change," American
Psychologist, 14:1:8-17, January, 195&.
78
Paul F. Lazarsfeld, "Communication Research
and the Social Psychologist," in Current Trends in
Social Psychology, by Wayne Dennis”and others
(Pittsburgh: University of Pittsburgh Press, 1948),
p• 242.
functions as a reinforcement of personal values for
the viewer. The present study was designed to examine
the identification that takes place between the viewer
and roles viewed to determine the degree to which this
identification was associated with the liking of certain
roles and the disliking of others. It differs from
previous studies in that its purpose was to obtain a
comprehensive quantitative measurement of identifica
tion through use of a personality inventory question
naire .
VI. PROGRAM SELECTION: A RESULT
OF ROLE-VIEWER IDENTIFICATION
Students of the mass media and those who use
television as a medium through which to address the
public have long been interested in identifying those
elements in television viewing which serve to determine
a person's selection or rejection of specific television
programs for viewing.
According to Friedson, this selection in the
use of mass media is the result of a wide variety of
forces. Members of an audience select their mass
communication content under much pressure and guidance
79
from their experience as members of social groups.
79
Friedson, oj>. cit., p. 315.
i Friedson stated that selection appears to arise out
of "the stimulation of organized social processes
rather than merely the individual's personal inter-
Selection of programs for viewing seems to be
related to the value structure that is revealed in
the personality of the viewer. This structure is the
result of his social and psychological environment,
as well as his own constitution. In keeping with one's
own value structure, a person tends to move towards
i
1
those persons in his environment in whom he perceives i
i
a value system similar to his own.
|
Persons tend to "like" most those persons with
whom they identify most strongly. The value system
i
of the perceiver and the perceived value system of the |
stimulus person are important aspects of the perceptual
situation. j
Newcomb claimed:
. . . we acquire favorable or unfavorable attitudes
toward persons as we are rewarded or punished by j
them, and . . . the principles of contiguity, of j
reciprocal reward, and of complementarity have to j
do with the conditions under which rewards are most
probable.81 !
80Ibid.
Theodore M. Newcomb, "The Prediction of inter
personal Attraction," The American Psychologist.
11:11:577, November, 1956.
54
j According to Newcomb, "attraction [toward a person]
and perceived similarity of attitude tend to maintain
0 9
a constant relationship . . .
Berkowitz and Goranson, in a study of social
1
perception, reported probability that if a person likes
another highly, he will tend to minimize differences
83
between himself and the person to whom he is attracted. :
In a study of attitude toward the communicator, j
i !
perception of communication content, and attitude
change, Kelman and Eagly found
. . . a marked tendency to misperceive the message '
of a negative communicator by displacing it away
from one's own position .... this tendency was
shown to be a direct function of the strength of !
negative feeling toward the communicator. Simi
larly, tendency to displace the message of a
positive communicator toward one's own position
. . . was shown to be a direct function of the
strength of positive feeling.84
Wyburn and others concluded:
82Ibid., p. 579. j
]
83
Leonard Berkowitz and Richard E. Goranson,
"Motivational and Judgmental Determinants of Social
Perception," Journal of Abnormal and Social Psychology.
69:3:299, September, 1964. j
84
Herbert C. Kelman and Alice H. Eagly,
"Attitude Toward the Communicator, Perception of
Communication Content, and Attitude Change," Journal
of Personality and Social Psychology, 1:1:63, January,
19617
55
. . . our perceptions of other people and objects
are affected by our emotional sensitivity to
minute modes of behaviour or qualities often
imperceptible except to ourselves. Our emotional
attitudes to other people . . . often lead to an
exaggeration of some qualities and a diminution
of others .... We perceive in other persons
. . . , various attitudes or intentions favourable
or inimical to ourselves . . . .85
Fensterheim and Tresselt showed twenty-four
photographs to twenty-eight male college students and
asked them to rate each picture on a like-dislike
scale. The students were also asked to complete the
Allport-Vernon Study of Values test, and from a set
of six statements, adapted from the Allport-Vernon
test manual, "to select the statement which 'best'
described each stimulus photograph and the statement
86
which 'least' described each." Finally the students
were asked to rank the six statements as they applied
to themselves. Fensterheim and Tresselt reported:
The closer the value system projected into the
stimuli [photograph] resembled the value system
of a subject, the greater was the liking. The
less the value system projected into a stimulus
photograph resembled that of the S, the greater
was that photograph d i s l i k e d .87 ”
G. M. Wyburn, R. W. Pickford, and R. J. Hirst,
Human Senses and Perception, edited by G. M. Wyburn
(Toronto: University of Toronto Press, 1964), p. 218.
86
Herbert Fensterheim and M. E. Tresselt, "The
Influence of Value Systems on the Perception of People,"
The Journal of Abnormal and Social Psychology. 48:1:94,
J anuary, 1953•
87Ibid.. p. 97.
56
The association between the two variables of
this experiment [seven step like-dislike scale
find Allport-Vernon Study of Values] may come from
an over-all impression of the personality of the
stimulus that had been formed by the Ss. . . .
The like-dislike rating . . . may be an index of
the amount of identification existing between S
and the stimulus.88 “
Horton and Wohl, in discussing the para-social
climate of television viewing, stated that it is
imperative that a viewer accept "the explicit and
implicit terms which define the situation and the
89
action to be carried out in the program." Failure
to accept these terms means that the viewer rejects
his own proffered role in the viewing situation and
therefore cannot enter into the performance itself.
They conjectured that "the probability of rejection
of the proffered role will be greater the less closely
90
the spectator 'fits' the role prescription."
Studies reviewed above revealed that much
research has been directed toward the study of
identification in the viewing of television. Such
88Ibid., p. 98.
89Horton and Wohl, oj>. cit•, p. 220.
90Ibid.
57
studies have drawn attention to personality as an
important element in one's selection of programs for
viewing.
However, none of these studies provided a
satisfactory quantitative description of similarity
or dissimilarity between the personalities of viewers
and the personalities of television roles as a
determiner of the viewer's selection of programs.
It was to help fill this seeming need that the
present study was devised and executed.
VII. SUMMARY
The present study resulted from the conviction
that similarity of personality between the viewer and
television roles viewed by him was an important element
in his selection of programs for viewing. More
specificallyx the study was designed to determine if
there was not greater similarity between the person
ality of a television viewer and the perceived per
sonality of a television role rated by him as a
most-liked role than there was between the personality
of a television viewer and the perceived personality
of a television role rated by him as a least-liked
role.
58
To facilitate the measurement of similarity
or dissimilarity between the personalities of viewers
and roles most liked, and of viewers and roles least
liked, three major hypotheses were posited. These
hypotheses were designed to determine whether or not
the similarities and differences observed between the
personality profiles of the viewers and those of the
roles perceived by them were significantly different
from zero. Several minor hypotheses were posited.
These hypotheses provided statistical tests with which
.to measure the personal variables of sex, age, educa
tion, and standard of living.
A review of literature in the areas of audience
research, psychology, and sociology suggested that
television viewing takes place as a para-social
situation in which the viewer enters into a quasi
face-to-face encounter with roles selected for viewing.
The viewer's selection of roles was attributed to the
fact that the viewer tends to identify with roles
whose appearances are perceived to be similar to his
own. This identification can serve either as rein
forcement of the viewer's present value system or as
a type of wish fulfillment, a vicarious experience
that is perhaps unobtainable in real life.
59
Although several studies of mass media reported
that viewers were most attracted to roles bearing
similarities to themselves, none of these studies
provided a quantitative description of personality in
which objective measurements were used to examine the
relationship of the viewer to roles that were selected
for viewing.
This study was proposed in the belief that
personality was an important single index for use in
the measurement of similarity or dissimilarity between
viewers of television and roles viewed by them. It
was believed that personality profiles reflected not
only the constitutional characteristics of the viewer
but also value structures and behavioral response
patterns which are largely a product of his environ
ment. It was believed that such a study might help
to provide needed information about the motivation of
viewers in their selection of television programs for
regular viewing.
| CHAPTER III
I
I
METHODS AND PROCEDURES
I. METHODS
The problem of the present study was to show the
kind and degree of relationships that existed between
the personality profiles of certain television viewers
and the perceived personality profiles of certain tele
vision roles viewed by them.
!
Gage and Cronbach, in an article on conceptual
1
and methodological problems in interpersonal perception, j
listed three variables in the measurement of perceptual
: !
judgment: "(a) the Judge's self-description, (b) the
" " " I
Other's self-description, and (c) the Judge's prediction j
i
of b."1 |
!
The present study purported to measure item a !
above, the viewer's self-description, and item c, the |
i
I
viewer's judgment of a television role, and to compare j
these measurements for similarity or difference.
This was not an experimental study in that no
effort was made to measure certain variables while
holding other variables constant. Rather, it was an j
N. L. Gage and Lee J. Cronbach, "Conceptual and
Methodological Problems in Interpersonal Perception,"
The Psychological Review, 62:6:415, November, 1955.
61
attempt to measure certain elements in television
viewing as they existed in a somewhat natural, rela
tively uncontrolled situation. It was a descriptive
study which set out to describe the relationships that
pertained between the mean personality profiles of
certain groups of viewers and the personality profiles
of certain television roles as they were perceived by
these viewers.
II. PROCEDURES
Collection of Data
Description of the sample. The original sample
for this study consisted of 1,755 couples, members of
the Mariner program of the United Presbyterian Church,
U.S.A., who resided in the greater Los Angeles area.
Subjects were drawn from groups located in twenty-one
different communities.
This was not a random sample and findings could
not be generalized to apply to the population at large.
The problem of the study was to determine whether or
not certain consistent patterns of association could
be observed between television viewers and certain
roles regularly viewed by them. It was believed that
any sample which possessed somewhat homogeneous char
acteristics of age, education, religious affiliation,
62
and standard of living could serve to reveal such
associations. It was believed that if these patterns
of association became apparent the study might later
be repeated using either a random sample or some type
of stratified sample.
A large volume of information was requested of
subjects through six separate mailings, and because
it was collected over a period of approximately two
and one-half months it made extensive demands upon the
subjects. Therefore, from an initial sample of 1,755
«
couples, the final sample resulted in only 206 persons— ;
127 females and seventy-nine males (Table I). Ninety-
I
four per cent of the subjects were between twenty and
fifty-nine years of age, with 42 per cent between thirty ;
|
and thirty-nine years of age (Table II). j
i
The sample for this study reflected an upper j
j
socio-economic status, with 46 per cent earning $10,000
j
or more annual combined family income (Table III). The !
sample also showed a high educational attainment, with
68 per cent having finished one or more years of college
(Table IV).
Judgments were made on twelve different tele
vision roles. Each role was judged by a single group
with the exception of Granny, which was judged by two |
different groups. Table V shows the number of persons
63
TABLE I
DISTRIBUTION OF SEX OF SAMPLE
Sex Freauency Per Cent
Male 79 38
Female 127 62
Totals 206 100
TABLE II
DISTRIBUTION OF AGE OF SAMPLE
Ages Freauency Per Cent
20-29 24 12
30-39 84 42
40-49 46 23
50-59 34 17
60-69 8 4
70-79 4 2
Totals 200 100
64
TABLE III
DISTRIBUTION OF STANDARD OF LIVING OF SAMPLE
Annual Combined
Family Income Frequency Per Cent
Under 4,000 4 1
4,000-5,999 11 6
6,000-7,999 53 27
8,000-9,999 40 20
10,000 and over 92 46
Totals 200 100
TABLE IV
DISTRIBUTION OF LEVEL OF EDUCATION OF SAMPLE
Years of School
Completed Frequency Per Cent
None
1-8
9-12
College (1 or more
0
3
55
years) 135
0
2
30
68
Totals 193 100
65
TABLE V
NUMBER OF PERSONS WHO JUDGED EACH ROLE
ON THE 16 P. F. QUESTIONNAIRE
Roles Judged
Number of Persons
Judging Role
Most-Liked Roles
Dick Van Dyke* 39
Perry Mason 27
John Kimble 21
Ben Cartwright* 17
John Novak 15
Andy Griffith 14
Granny (Beverly Hillbillies)* 14
Least-Liked Roles
Abbott (Abbott and Costello) 15
Ben Casey 13
Eddy Haskell (Leave It to Beaver) 11
Danny Williams 9
Paladin 9
Granny 6
Total 210
*These roles contained one less judgment on
personal data than shown here.
| who judged each role.
i
Two factors lent reliability to the measurements
provided by this sample:
1. Size— beginning with 1,755 couples, after
six mailings there remained 206 persons.
Each of these persons completed all six
questionnaires and qualified in all respects
2
to be included in the study. Effort was j
made to have no group smaller than ten;
however, this was not possible for three
groups which had judged least-liked roles.
i
The small n for least-liked role groups
I
resulted from the fact that it was more
difficult for subjects to select ten least- !
liked roles than it was for them to select
ten most-liked roles. Subjects tended to
leave the "dislike" columns blank or to list
less than ten least-liked roles.
2. Homogeneity— the sample consisted of couples
who were members of the Mariner program of
the United Presbyterian Church, U.S.A.
While the religious affiliation of members i
of the sample limited breadth of application j
o
See page 85.
67
| of the findings, the homogeneity of the
data increased reliability.
Because the data were recorded by subjects in
their homes during normal viewing hours, validity was
i
increased. The home viewing situation constituted a
| more unstructured setting for the gathering of data
1 than that of the laboratory. No attempt was made to
manipulate the environment or conditions of viewing,
i
i with the exception that subjects were asked to keep
television viewing diaries and to complete question-
| naires.
Determination of television roles regularly
viewed by each group. Before it could be determined
which television roles were viewed regularly, it was
! necessary to determine which television programs were
i viewed regularly. Subjects were asked to complete a
Television Viewing Diary (Appendix C) for three separate
: weeks of viewing. The Television Viewing Diary was
I patterned after a diary used by the American Research
|
j Bureau of Montebello, California. They were asked to
| 9
record all programs viewed between 6:00 P.M. and 12:00
3
midnight for each of three weeks. There was a lapse
The three weeks of 1964 in which the Television
Viewing Diaries were kept were March 16-22, March 30-
April 6, and April 13-19. Television program schedules
68
of one week between each week of viewing. Each diary
provided space for the viewer to write in station
channel, time and name of program, and date and time
entry was made.
If a program was recorded by a subject in at
least two of the three viewing diaries, that program
was considered to have been viewed regularly by that
subject. Any role featured regularly in a regularly
viewed program was considered to be a regularly viewed
role.
The use of no fewer than three Television View
ing Diaries increased the reliability of measurement
of the "regularity" of a subject's viewing of programs.
As an additional check of reliability of the
recording of programs in the three viewing diaries,
the recorded entries of twenty-one programs, which
contained roles judged by viewers, were compared with
4
the listings printed in TV Guides of the same dates.
Twenty-one diaries were chosen through use of a table
3 (continued)
for these three weeks of viewing can be found in the
following issues of TV Guide: 12:11, March 14, 1964;
12:12, March 21, 1964; 12:13, March 28, 1964; 12:14,
April 4, 1964; 12:15, April 11, 1964, and 12:16,
April 18, 1964.
4Ibid.
5
of random numbers. No error was found in the twenty-
one diary entries.
Validity was increased through use of an open-
ended diary.
Determination of most-liked roles and least-liked
roles. After the viewing diary was returned, subjects
received a Television Role Preference Questionnaire
(Appendix D) on which were listed all roles regularly
featured in dramatic life situation programs regularly
broadcast by the seven VHP (Very High Frequency) tele
vision stations in the Los Angeles area.
Subjects were asked to check those roles that
they "particularly liked," those roles that they
"particularly disliked," and to leave unchecked those
roles with which they were unfamiliar or about which
they had no opinion. From those roles checked as
particularly liked they were asked to list, in order
of preference, ten roles, i.e., most liked, next most
5
The random count' began in column 6, row 35 and
ended in column 5, row 45. The count moved from left
to right along the rows embracing three digits at a
time. The table used for the count was Table L, j
"Random Numbers," in John G. Peatman, Introduction to
Applied Statistics (New York: Harper and Row, Publishers,!
Incorporated, 1963), pp. 424-425, citing J. G. Peatman
and Roy Schafer, "A Table of random numbers,from selec
tive service numbers," Journal of Psychology. 14:295-
305, 1942.
70
| liked, etc. From those roles checked as particularly-
disliked they were asked to list, in order of disliking,
ten least-liked roles, i.e., least liked, next least !
liked, etc.
Reliability was increased in that subjects were
assisted in their selection of ten most-liked roles and
ten least-liked roles by narrowing their range of choice,
that is, they were asked to check all roles that they
i
particularly liked and all roles that they particularly
disliked before they were asked to list, in order of
preference, the ten roles that they liked most and the
t
ten roles that they liked least.
Validity was increased in that the subjects were
I
asked to leave unchecked those roles about which they
had no particular opinion or with which they were not
familiar. With this option it was believed that those
roles checked would more accurately reflect the sub- !
jects* polarized opinions.
Validity was further increased by providing
subjects with the opportunity to write in names of
roles that did not appear on the list. This eliminated i
j
the pressure of a forced choice response.
Only those roles from programs which had been
i
recorded as viewed in at least two of the three viewing
diaries were included in the study for judgment as
71
most-liked roles. This increased the probability that
subjects actually did like these roles and were not
listing them as liked roles because of social desir
ability or for some other reason.
Validity was not as great for the listing of
least-liked roles as it was for the listing of most-
liked roles in that any first ranked, least-liked role
was included for study even if it had not been viewed
once during the three weeks of diary keeping.
Recording of Television Viewing Diary data.
Television programs viewed by each subject during the
three weeks specified for viewing were coded and key
punched into IBM cards and verified. This was per
formed by Data Key, Canoga Park, California— a pro
fessional key punch organization. Diary viewing data
were key punched into Card 1.
Recording of Television Role Preference data.
The ten most-liked roles and the ten least-liked roles,
listed by each subject in order of preference, were
coded and key punched into another IBM card and verified.
These data were key punched into Card 2.
From these cards a computer program was developed
which provided the following information:
72
1. A list of the most-liked roles as designated
by each subject in order of liking, and the
number of times each program in which the
role appeared was viewed.
2. A list of the least-liked roles as designated
by each subject in order of disliking, and
the number of times each program in which the
role appeared was viewed.
Selection of television roles to be judged. All
most-liked roles that were recorded by each subject in
the viewing diary in two of the three weeks of diary
keeping were listed by rank from one through ten. The
highest ranking choice was given a rank of one, the next
highest choice two, etc.
i
All least-liked roles recorded by each subject 1
were listed in similar manner. It was not anticipated j
i
that subjects would view least-liked roles with any
regularity, except when least-liked roles appeared in
television programs with most-liked roles, therefore
j
the criterion of being listed at least twice in the
viewing diary was waived for least-liked roles. In
fact, it was not necessary for subjects to have viewed
least-liked roles even once during the three weeks of
television viewing; roles may have been viewed at a
time other than during the three-week period.
73
Each viewer group consisted of persons who had
listed the same role as a most-liked role, or viewers
who had listed the same role as a least-liked role.
Two criteria were used in forming the role
judgment groups:
1. Effort was made to form groups of no less
than ten.
2. Wherever possible those roles which were
ranked at some point between first and
fifth choice, in order of preference, were
assigned to viewers for judgment.
Two other groups of viewers were formed: one
rated Granny a most-liked role, and the other rated
her a least-liked role. To form these groups it was
necessary to make selections from all ranks of one
through ten, thus forming as large groups as possible.
Using the above method of selection, maximum
sized groups were formed. This helped to increase the
reliability of the measurements. In only three groups
of the thirteen did n fall below ten. Use of electronic
computers for the processing of data helped to reduce
errors in the handling of data and in calculations, and
this increased reliability.
By choosing for judgment only high ranking roles,
as ranked in the Television Role Preference Question-
74
, naire, it was thought that these roles would represent
extremes between most-liked and least-liked roles as
listed by each viewer. This did not apply to the two
groups, one of which rated Granny a most-liked role
i and the other of which rated her a least-liked role,
in that they may have ranked Granny at any point from
one through ten.
Selection of a personality inventory question
naire . To aid in the selection of a test instrument
with which to measure personality variables of sample
subjects and their perception of television roles, a
search was made of the literature on psychology.
Based on the anticipated circumstances for the col
lection of data, i.e., by mail, and the desire to
procure a comprehensive, quantitative description of
personality, it was decided that the test selected
for use in this study should possess as many of the
following characteristics as possible. It should:
1. Be multidimensional.
2. Be developed through use of factor analysis
with items drawn from several previous
widely used tests.
3. Be in use long enough to determine its
reliability and establish a fairly stable
set of norms.
75
4. Be designed for use by adults.
5. Be a self-administering, paper and pencil
test.
6. Require approximately forty-five minutes
to complete, therefore contain from 150
to 200 items.
7. Measure a large enough number of person
ality traits to provide a wholistic view
of personality.
8. Produce trait concepts comparable to
current clinical concepts.
9. Provide alternative responses of "don't
know" or "neither," i.e., should not be
"forced choice."
The following multidimensional tests, designed
for adult subjects, were found to be available and
were rated on the above criteria. A test was given
one point for each criterion met. These tests are
listed below and ordered from those which met the
largest number of criteria to those which met the
smallest number:
Number of
Test Criteria Met
Cattell and Stice Sixteen Personality
Factor Questionnaire 9
Guilford Series 9
Number of
Test Criteria Met
Thurstone Temperament Schedule 8
Activities Index 7
Cornell Index 7
Adjustment Inventory 6
Minnesota Multiphasic Personality
Inventory 6
Bernreuter Personality Inventory 6
California Test of Personality 5
Edward's Test of Personality 5
Allport-Vernon-Lindsey Study of Values 5
Only the 16 P. F. and the Guilford Series met
all nine criteria. Upon investigation it was found
that the Guilford tests were not available for use
through the mail but could be administered only in
person, therefore they were not feasible for use in
this study. Cattell's 16 P. F. was selected for use.
It had several outstanding qualifications:
1. The items in the 16 P. F. were secured from
a population of questionnaire items which
were developed from a complete review of
"questionnaire, opinionnaire, interest and
g
value scales," and from "evidence of
6Raymond B. Cattell, "Validation and Intensifi
cation of the Sixteen Personality Factor Questionnaire,
77
entirely new personality factors, from non-
questionnaire sources.”
2. Form A of the 16 P. F., the form used in
O
this study, contained 187 items. These
items survived three successive factor
analyses and continued to have significant
g
validity against the factors. In the
development of the 16 P, F., three factor
analytic studies were conducted with both
student and adult population groups: first
study— 370 twenty-year-old undergraduates,
men and women;*9 second study— 295 under
graduates, men and women;** third study—
6 (continued)
Journal of Clinical Psychology. 12:3:206, July, 1956,
citing R. B. Cattell. The Description and Measurement j
of Personality (New York: World Book Company, 1946). j
7Ibid. I
Q
Raymond B. Cattell and Glen F. Stice, Handbook
for the Sixteen Personality Factor Questionnaire: "The
1<> P. F. Test7ir Forms A, B, and C (1957 Edition, with
1562 Supplementation; Champaign,“Illinois: The Institute
for Personality and Ability Testing, 1957), p. 3.
9Ibid.. p. 4.
*9Raymond B. Cattell, "The Main Personality Fac- j
tors in Questionnaire, Self-Estimate Material," The j
Journal of Social Psychology, 31:First Half:3, February, :
1935: |
**Cattell, "Validation and Intensification ..."
pp. 208-209. 1
78
12
227 Air Force men and 181 undergraduates.
3. Each of the 16 P. F. factors had been
validated against ratings in real-life
13
situations, beyond the questionnaire realm.
In development of the 16 P. F. effort was made
to get "the highest possible reliability for a small
14
number of items." Reliability coefficients took
three major forms: consistency, equivalence, and
stability. "The consistency coefficients (split-half,
corrected to full length), obtained on the largest
15
single sample (450 young adult males),” ranged from
•71 to .93.
Procurement of 16 P. F. scores for television
roles. The 16 P. F. was used to obtain both person
ality profile scores for viewers and personality
profile scores for television roles that were judged
by these viewers. Form A of the 16 P. F. was used
for both.
When the 16 P. F. was used as an instrument
with which to collect personality data on roles, the
12Ibid.
1*1
Cattell and Stice, 0£. cit., p. 2.
14Ibid., p. 3. 15Ibid., p. 4.
79
front sheet of the questionnaire was covered with
another sheet entitled "Television Role Personality
Questionnaire" (Appendix B).
Each subject was asked to complete a Television
Role Personality Questionnaire for one specified most-
liked role or one specified least-liked role, and to
answer the questions as he thought that particular
television role might answer. A few subjects were
asked to complete questionnaires for both a most-liked
role and a least-liked role. Also, viewers were asked
to estimate the age, level of education, and standard
of living of the role and to place these data on the
cover sheet of the Television Role Personality Ques
tionnaire.
Procurement of 16 P. F. scores for television
viewers. After the Television Role Personality Ques
tionnaire was returned each subject received a 16 P. F.
to complete, with himself as the subject.
The reliability of profile comparisons was
increased through use of Form A of the 16 P. F. for
both judgment of television roles and judgment of
subjects.
By asking subjects to judge roles first and
self later, with a subsequent mailing, it was felt
that subjects would not be so likely to project their
80
self-image into the role judgments as they might have
been if they had had prior knowledge of the 16 P. F.
(Form A). Only after the 16 P. F. role judgment had
been returned was the self-evaluation questionnaire
mailed to a subject. This helped to insure greater
validity in the judgment of roles.
Treatment of Data
16
The following kinds of data were treated:
1. The 16 P. F. responses of each viewer,
which formed a personality profile of
sixteen factor scores.
2. The 16 P. F. responses ascribed to each
role by a viewer, which formed a person
ality profile of sixteen factor scores.
3. Personal data of each viewer, including
sex, age, level of education, and annual
combined family income.
4. Personal data of each role as judged by a'
viewer, including sex, age, level of
education, and standard of living.
16
The raw score data will be deposited in the
archives of the Doheny Library at the University of
Southern California for public use as of January 1,
1969. Until that date information concerning the
data is available from the writer at 4042 Vest 59th
Place, Los Angeles, California 90043.
81
The 16 P. F. coded response data were keypunched
into IBM cards and verified by a separate operator by
Key Data Systems, Incorporated, Canoga Park, California. 1
Data from each questionnaire were recorded on three
consecutive cards. Each column of a card contained
the response to a single item. This simplified the
keypunching and tended to minimize possible errors of
recording.
Personal data on each subject were punched into
a single IBM card and the personal data on each role
were punched into another single IBM card. These cards
were keypunched by Data World Corporation of Canoga
Park, California, and were verified by a separate
operator.
The treatment of data consisted of (1) editing, ;
(2) obtaining raw scores, and (3) testing hypotheses. |
Editing data. It was necessary to determine !
the treatment that should be given to "no response" j
scores in the 16 P. F. returns. Each item in the
questionnaire provided opportunity for a multiple
choice response with three options: A, middle response, |
or B. Each option was assigned a weight of 0, 1, or 2. j
These weights were the same as those used in the scoring |
82
stencils of the 16 P. F., Form A.- 1 7
All sets of responses contributing to a single
trait score were so balanced that approximately one-
half of all responses fell on answer A and one-half on
answer B. Response A and response B were assigned
weights of either 0 or 2, while middle responses
received a weight of 1.
To know whether to assign a "no response" item
a value of 1, neutral value, or a value of 0, which
would lend weight to response A, it was necessary to
determine whether there would be any significant dif
ference in final raw scores, for a combined set of
profiles, if "no response" items were assigned a value
of 1 rather than a value of 0.
To test this, a complete set of role judgments
for Kimble was used. The Kimble role cards were
selected because they contained more "no response"
items than any other set of cards for a role group.
An IBM computer program was designed to deter
mine :
1. The weighted score for each response in
each questionnaire.
17
Cattell and Stice, oj). cit., p. 6.
83
2. The sum of the weighted scores which were
combined to form the respective trait
scores in each questionnaire.
3. An average profile for all sets of scores
combined. These average "raw score"
profiles were obtained in two ways: in one
set "no response" items received a weight
of 1, and in the other set "no response"
items received a weight of 0.
A Chi Square test was made between the resultant
set of profiles. The difference between the two
resulting profiles proved to be not significantly
different from zero (P = .99).
It was decided to use a weight of 1 for "no
response" items as this seemed to represent a neutral
"I don't know" or "I can't decide" type of response.
Next it was necessary to determine what treat
ment should be given to "no response" items in the
personal data questionnaire returns. "No response"
items in the Personal Information sheets (Appendix A)
were given a code number of "9" on the IBM personal
data cards. In computing personal data scores of "9"
were omitted from the calculations, and the final
scores resulted only from items to which a response
was given.
It was also necessary to determine how to treat
cases in which both husband and wife judged the same
role. The question arose as to whether or not judgments
of a husband and a wife who had judged the same role
should be retained in the final sample for testing the
major hypotheses. It was believed that their scores
might be more highly correlated than those of indi
viduals who were not closely related.
It was found that of those judgments made on |
most-liked roles and those made on least-liked roles !
equal proportions were made by husband and wife pairs. |
A proportion of .10 of the total judgments made on
!
most-liked roles was the result of husband and wife
& I
pairs who had judged the same roles, and the same pro
portion pertained to the judgments of least-liked
roles made by husbands and wives.
It was decided to leave the joint judgments in
the final sample as any bias which might occur would i
j
exist equally for both most-liked role judgments and j
for least-liked role judgments and would not impair
the validity of the findings on the major hypotheses. |
It was also believed that the additional judgments
thus retained in the final sample would increase the
I
overall reliability of the measurements.
Next, it was necessary to determine what treat
ment should be given to "no response" personal data
items. It was decided that Personal Information sheets
should be retained in the final sample if as many as
three of the four data questions listed therein were
completed. Comparisons of personal data were made only
on those scores which were available both on viewers
and on roles judged by viewers.
Finally, it was determined that for a subject
to be included in the final sample it was necessary
that he have returned a 16 P. F. questionnaire for
himself and one for each role judged by him, and a
Personal Information sheet for himself and one for
each role judged by him.
Obtaining raw scores. Raw scores were obtained
for each of the sixteen personality factors of each
viewer and of each role judged. Through use of an
IBM Fortran IV computer program, scores were weighted
according to Cattell's assigned weights as recorded
18
on the 16 P. F. scoring stencil, Form A. Weighted
scores for each item response-were summed to provide
a single "raw" score for each of the sixteen traits.
A raw score was defined as the sum of weighted responses
18ibid.
86
which combined to constitute a single trait.
Raw scores were also obtained on four kinds of
personal data:
1. Sex— scores were divided into two categories
and were coded as follows:
Code
Male 0
Female 1
No response 9
2. Age— actual ages were recorded for each
subject and for each role judgment. "No
response" items were coded as "9."
3. Level of education— scores for viewers and
for roles were coded as follows:
Code
No schooling 1
Elementary (1-8) 2
High school (9-12) 3
College ( 1 year or more) 4
No response 9
4. Standard of living— viewers were scored on
level of annual family dollar income, while
roles were scored on a standard of living
scale as judged by the viewer:
Scale Used for
Viewer Annual Judging Standard
Family Income Code of Living of Roles
Under 4,000
4.000-5,999
6.000-7,999
8.000-9,999
10,000 and over
No response
1 Very low
2 Rather low
3 Average
4 Rather high
5 Very high
9 No response
87
Testing the hypotheses. Personality profiles
of viewers and of roles were tested for similarity
and for difference. This test was used with:
1. Viewer groups and most-liked role groups.
2. Viewer groups and least-liked role groups.
3. Two groups of viewers, one of which judged
a role as most liked and another which
judged the same role as least liked.
Helmstadter stated that the assessment of !
similarity between two profiles is essentially "deter
mining the degree of agreement between the two sets
19
of numbers of which the profiles are constituted." !
i
According to Cronbach and Gleser, the dimensions |
< \
of elevation, scatter, and shape should be considered
when assessing similarity between profiles:
Elevation is the mean of all scores for a given
person. Scatter is the square root of the sum
of squares of the individual's deviation scores
about his own mean .... Shape is the residual !
information in the score set after equating pro- j
files for both elevation and scatter.20
19
Gerald C. Helmstadter, "An Empirical Comparison j
of Methods for Estimating Profile Similarity," Educa
tional and Psychological Measurement, 17:1:72, Spring,
1957.
20
Lee J. Cronbach and Goldine C. Gleser, "Assess
ing Similarity between Profiles," Psychological
Bulletin, 50:6:460, November, 1953.
88
! They stated:
In general, it appears undesirable to eliminate
elevation unless the investigator can interpret it
definitely as representing individual differences
in a quality which he does not wish to take into
account in his similarity measure.21
This opinion was alleged to be held also by Cattell
and by du Mas.22
Cronbach and Gleser listed three formulas which
measure the similarity of profiles in k space, i.e.,
these tests include measurements of elevation, shape, j
i
and scatter. These statistics are D, a measure of
distance which was advocated by Osgood, Suci, Cronbach,
i
and Gleser; CRL, Pearson's coefficient of racial like- j
ness; and rp, Cattell's coefficient of pattern simi
larity.23 |
Coefficient r is similar to D, except that it j
P i
24 I
provides an index which ranges from +1 to -1. In
developing this index Cattell thought that it should
be "1 when the two profiles are perfectly alike . . . , j
21Ibid., p. 464.
22Ibid.. p. 463, citing R. B. Cattell, "r^ and
Other Coefficients of Pattern Similarity," Psycho-
metrika, 14:4:279-298, December, 1949, and F. M. du Mas, j
"On the Interpretation of Personality Profiles,"
Journal of Clinical Psychology. 3:1:57-65, January,
194*?.
23Ibid., p. 462, see Table 1. 24Ibid., p. 462.
89
o
. . . 0 when 'sum of d 1 equals the chance expectation
of difference, and . . . -1 when the differences are
as great as they can be."2®
The distance measure D is difficult to interpret
because it has no fixed range of scores. The CRL
proved unsatisfactory in the original research for
which it was developed and has given unreasonable
results in subsequent research. The r^ coefficient
of pattern similarity was selected for use in this
study because it had a fixed range of scores similar
to that of the Pearson r, a measure to which researchers |
27 ^
were accustomed.
Horn devised a table in which are set forth the
values of r at various levels of confidence with two
P
28
through thirty degrees of freedom. This table was j
.... « . ,, I
I
I
OK
Raymond B. Cattell, "r^ and Other Coefficients |
of Pattern Similarity," Psychometrika. 14:4:287, J
December, 1949;.
26
Cronbach and Gleser, oj>. cit•, pp. 461-462.
2^Cattell, "r and Other . . ."p. 285.
_J2
I
O Q i
J. L. Horn, "Significance Tests for Use with
r and Related Profile Statistics." Educational and i
_R ...
Psychological Measurement« 21:2:367, Summer, 1961.
90
obtained in this study were significantly different
from zero.
The following steps were taken to obtain the r^
coefficient of pattern similarity:
1. Raw scores were converted into standard
scores.
X—M
z = Standard Score Formula
s
where X = a raw score
M = mean of all raw scores within a single
group on a specific factor
s -WSISLJL. standard Deviation of combined
“ n 26 groups (total sample popula
tion)
where x = deviation of a score from the mean
of the sample
n = size of the sample
2. Using standard scores, the coefficient of
pattern similarity was obtained with the
following formula:
2
r = ov I eSS ^2 Coefficient of pattern
p 2k + sum d* similarity®®
where k = median Chi Square value for 16 degrees
of freedom (i.e., 15.338)
d = difference between the profile.score
of a viewer and the profile score of
a role
29Cattell, "r and Other . . ."p. 292
91
In addition to a test for determining similarity
or difference between personality profiles, tests were
selected with which to measure the similarities and
differences between personal data. Tests of similarity
were used with personal data of:
1. Viewer groups and most-liked role groups.
2. Viewer groups and least-liked role groups.
3. Two groups of viewers, one of which rated
a role as a most-liked role and the other
of which rated the same role as a least-
liked role.
A Pearson coefficient of correlation (r) was
obtained between the following variables:
1. Ages of viewers in each group and the per
ceived age of the role as judged by each
corresponding group.
2. Levels of education of the viewers in each
group and the perceived level of education
of the role as judged by each corresponding
group.
3. Annual combined family income of the viewers
in each group and the perceived standard of
living of a role as judged on a five point
scale by viewers in each corresponding group.
A Fortran IV computer program routine entitled
"Plotting Program" was used to compute the product
moment coefficient of correlation on items one, two,
and three above. This library program was made avail
able by the Computer Sciences Laboratory of the
30
, University of Southern California.
A table of coefficients of correlation, adapted
by Guilford from a table by Wallace and Snedecor, was
used to determine whether or not the resultant r's
31
differed significantly from zero.
The reliability of the product moment correla
tions obtained from personal data was in general quite
limited. This resulted from the small numbers of data
in certain groups and from the fact that only three of
the r's were found to be significantly different from
zero at the .05 level of confidence.
SoComp: A Series of Automatic Statistical j
Programs (University of Southern California: Computer
Sciences Laboratory, 1010 West Jefferson Boulevard,
Los Angeles, California 90007, January, 1967), p. 155.
31J. P. Guilford, "Table D. Coefficients of
Correlation and t Ratios Significant at the .05 Level !
(Roman type) and-at the .01 Level (Bold-faced type)
for Varying Degrees of Freedom," Fundamental Statistics
in Psychology and Education (Third Edition; New tfork:
McGraw-Hill Book Company, Incorporated, 1956), pp. 538-
539, citing H. A. Wallace and G. W. Snedecor, Correla
tion and Machine Calculation. 1931.
The data on sex were dichotomous and did not
lend themselves to analysis with the "Plotting Program."
These data were converted to proportions so as to 1
compare the distribution of the sex of viewers with j
the sex of roles judged by them. I
i
Finally, a test of difference was selected for I
I
use with personal data. This test was used with:
i
1. Viewer groups and most-liked role groups.
!
2. Viewer groups and least-liked role groups. |
3. Two groups of viewers, one of which rated
a role as a most-liked role and the other
of which rated the same role as a least- j
liked role.
Differences of personal data were computed with
the use of a Fortran IV computer program routine j
I
i
entitled "Two-Way Table Program," made available by
the program library of the Computer Sciences Laboratory
* * 1 0 |
of the University of Southern California.
O |
Initially it was decided that a Chi Square (X )
i
would be the most suitable test with which to measure |
these differences, but upon inspection of the data it !
i
was found that unless the data were coarsely grouped,
there were many cells that contained frequencies of
i
________________________________________________________________________ I
j
32
SoComp; A Series of Automatic . . . p. 131.
94
less than five. For this reason original score fre
quencies were used to compare differences between the
t
personal data of sex, age, education, and standard of i
living. !
i
Reliability checks on computer programed data.
In checking the original documents against input cards
for both the 16 P. F. data and the personal data it
was found that one or two cards in five sets of group
i
data had been placed in the wrong set. Corrections |
were made and the programs were run again. It was- j
found that these corrections in input data increased |
the r correlations of each group in which a correction
P
i
was made.
Sixty-four items on the IBM cards were checked
against responses on the original 16 P. F. questionnaire
documents and no error was found. This indicated that
the keypunching of original raw score data onto the
cards was highly accurate.
Twenty items of the program which computed the
group mean profile averages were checked with the use j
of a desk calculator and no error was found. |
Output for the correlation of pattern similarity |
(r ) program was verified step by step with the use of
P
a desk calculator and no error was found in any series
i
of twenty items that were checked.
Desk calculations were made to verify the output
of the program routine "Plotting Program." A computa
tion was made for the product moment correlation that
was printed out on age, level of education, and standard
of living. The computed product moment correlations
corresponded to those in the program print out data.
CHAPTER IV
SIMILARITY AND DIFFERENCE BETWEEN THE
PERSONALITY PROFILES OF TELEVISION
ROLES AND THEIR VIEWERS
In the review of literature it was reported that
certain elements in television viewing served to attract ;
viewers to some television programs but not to others,
and that personality was an important element in tele
vision viewing. It was hypothesized that viewers would
be reliably attracted to those television roles whose
personalities were perceived to be similar to their own, ;
and that they would tend to dislike television roles
whose personalities were perceived to be different from
their own. j
After completion of three separate copies of a |
Television Viewing Diary and a Television Role Preferencej
Questionnaire, thirteen groups of viewers were formed. j
All members of each group had rated the same television
role as either a most-liked role or a least-liked role.
Next, personality inventories were completed by !
the viewers, one for themselves and one for each tele
vision role judged by them. A coefficient of pattern j
similarity (r ) was obtained on the mean personality j
P
profile of each viewer group and the corresponding mean
personality profile of the role judged by that group
97
to determine the degree of similarity or dissimilarity
that existed between them.
This chapter will report findings of similarity
and of difference between personality profiles, as
stated in the major questions of the study. To make
possible the use of statistical tests, a null hypothesis
was posited for each question.
1. MOST-LIKED ROLES
Question 1. Is there a significant similarity or
difference between the mean personality
profile of a group of viewers and the
mean personality profile of a role as
perceived by these viewers and rated by
them as a most-liked role?
Hypothesis 1. The r(coefficient of pattern similar
ity) obtained between the mean person
ality profile of a group of viewers and
the mean personality profile of a role
which was rated by these viewers as
most liked will be zero.
Findings. Null hypothesis 1 was rejected in
three of seven cases. There was a positive correlation
between all sets of mean personality profiles of viewer
98
groups and the mean personality profiles of roles as
perceived by these viewers (Table VI).
One of the seven sets of profiles had an r^
correlation that was significantly different from
zero at the .02 level of confidence, one at the .05
level of confidence, and one at almost the .10 level
of confidence. Other groups showed positive correla
tions of .2190, .1873, and .1381, respectively. These
correlations were not significantly different from
zero. One correlation showed a score of almost zero.
Interpretations. A definite similarity of
personality was found between viewers and roles rated
by them as most liked. The positive correlations
obtained on each of seven pairs of profiles provided
quantitative support to findings of previous research
which reported that identification was an important
element in peoples' perception of roles that appear
in the mass media.'1 '
The fact that two sets of profiles correlated
positively at better than the .05 level of confidence,
and one more at almost the .10 level of confidence,
indicated that this tendency toward identification
with most-liked roles was more than a casual relation-
■^See Chapter I, page 1, footnote 1.
99
TABLE VI
COEFFICIENTS OF PATTERN SIMILARITY BETWEEN
THE MEAN PERSONALITY PROFILES OF MOST-
LIKED ROLES AND THEIR VIEWERS
Roles r
P
P=
Cartwright .3795 Almost .02*
Mason .3233 .05
Kimble .2398 Almost .10**
Van Dyke .2190
Griffith .1873
Novak .1381
Granny (most liked) .0384
*.02 = .397
**.10 = .245
100
ship.
II. LEAST-LIKED ROLES
Is there a significant similarity or
difference between the mean personality
profile of a group of viewers and the
mean personality profile of a role as
perceived by these viewers and rated by
them as a least-liked role?
The r^ obtained between the mean person
ality profile of a group of viewers and
the mean personality profile of a role
which was rated by these viewers as
least liked will be zero.
Findings. Null hypothesis 2 was rejected in
three of six cases. There was a negative correlation
between five of the six sets of mean personality pro
files of viewers and the mean personality profiles of
least-liked roles as perceived by these viewers
(Table VII).
One of the six correlations was significantly
different from zero at the .01 level of confidence,
one at the .05 level of confidence, and one at the .10
level of confidence. Two correlations, while not
i
significantly different from zero, were negative, and
Question 2.
Hypothesis 2.
101
TABLE VII
COEFFICIENTS OF PATTERN SIMILARITY BETWEEN
THE MEAN PERSONALITY PROFILES OF LEAST-
LIKED ROLES AND THEIR VIEWERS
Roles
rp
P=
Williams -.4160
H
O
•
Granny (least liked) -.2615 Almost .05*
Haskell (Leave It to
Beaver)
-.1922 Almost .10**
Casey -.1225
Paladin -.0411
Abbott .1222
*.05 = -.263
**.10 - -.211
102
one was positive.
The highest correlations were on Williams
(-.4160) and on Granny, least liked (-.2615).
Interpretations. In contrast to the correlations
between viewers and most-liked roles, which were all
positive, the correlations between viewers and least-
liked roles were all negative, except one. This pro
vided support for the test hypothesis that there was
greater similarity between the personality profiles of
viewers and the perceived personality profiles of roles
rated by them as most liked than there was between the
personality profiles of viewers and the perceived per
sonality profiles of roles rated by them as least liked.
III. GRANNY, A ROLE RATED AS MOST LIKED
BY ONE GROUP AND RATED A LEAST-LIKED
ROLE BY ANOTHER GROUP
Question 3. Is there a significant similarity or j
difference between the mean personality
profile of a group of viewers who rated
a particular role as a most-liked role '
and the mean personality profile of a
group of viewers who rated the same
role as a least-liked role?
Hypothesis 3. The r obtained between the mean person-
Mr
ality profile of a group of viewers who
rated a particular role as most liked
and the mean personality profile of a
group of viewers who rated the same
role as least liked will be zero.
Findings. Null hypothesis 3 was mildly, affirmed,
although a negative correlation of only -.1867 was found j
between the mean personality profile of viewers who
rated Granny as a most-liked role and the mean person
ality profile of viewers who rated Granny as a least- j
j
liked role (Table VIII). This correlation was not
i
statistically significant, but it did indicate a lack
of similarity between the personality profiles of the
two groups. The correlation between the two sets of
perceptions of the role of Granny as judged by the two
groups of viewers was positive (.2445) at almost the
.10 level of confidence.
Null hypothesis 3 could not be rejected for
statistical reasons, but from the strength and direc
tion of the correlation between the two viewer groups'
i
mean profiles (-.1867) there was reason to believe J
that they were sufficiently dissimilar to give mild
support to the experimental hypothesis, i.e., that
there would be a significant difference between the
104
TABLE VIII
COEFFICIENTS OF PATTERN SIMILARITY BETWEEN
VIEWERS OF GRANNY, MOST LIKED, AND
GRANNY, LEAST LIKED, AND THEIR
PERCEPTIONS OF GRANNY, MOST
LIKED, AND GRANNY, LEAST LIKED
Groups r P=
___________________________________________________p
Viewers of Granny, most liked,
and of Granny, least liked* -.1867
Role of Granny, most liked,
and of Granny, least liked** .2445 Almost .10***
♦Correlation between two groups of viewers, one
of which rated Granny as a most-liked role and the
other of which rated her as a least-liked role.
♦♦Correlation between viewers' perceptions of
the role Granny by two groups of viewers, one of which
rated Granny as a most-liked role and the other of
which rated her as a least-liked role.
***.10 level of confidence = .245.
105
profiles of the two groups of viewers, one of which
rated Granny as a most-liked role and the other of
which rated her as a least-liked role.
Interpretations. While the mean personality
profiles of the viewers of Granny correlated negatively,
the groups' perceptions of Granny as a television role
correlated positively (.2445). This observation added
validity to use of the 16 P. F. questionnaire as an
objective instrument for use in the measurement of
personality. It indicated that the viewers' own per- j
j
sonality profiles were not projected unduly into their j
perception of the personality of the role of Granny.
i
Although the two groups' mean personality profiles were j
different, their judgments of Granny were similar. In !
other words, although the viewer groups differed in
personality, their perceptions of the television role
of Granny were similar. This also gave some indication
that the subjects' judgments of self and their judgment j
I
of a television role were not as highly correlated as i
might be suspected.
Because of the small n for the viewers of Granny
least liked, it was considered hazardous to make any
interpretation based upon these data.
IV. SUMMARY
106
In Chapter IV data were gathered and analyzed
in an effort to determine whether or not there was
greater similarity of personality between television
viewers and the perceived personality of specific
television roles most liked by them than there was
between the personality of television viewers and the
perceived personality of television roles least liked
by them.
The findings were affirmative. The correlations
of pattern similarity which served as tests of simi
larity provided conclusive evidence that there was a
greater similarity between the perceived personality
profiles of most-liked roles and their viewers than
there was between the perceived personality profiles
of least-liked roles and their viewers.
All seven of the comparisons that were made
between viewers and most-liked roles correlated
positively. One of the r 's was significantly dif-
Xr
ferent from zero at the .02 level of confidence, one
at the .05 level of confidence, and one at almost
the .10 level of confidence.
Five of the six comparisons that were made
between viewers and least-liked roles correlated
107
negatively. One of these r fs was significantly dif-
Xr
ferent from zero at the .01 level of confidence, one
at almost the .05 level of confidence, and one at
almost the .10 level of confidence.
A higher positive correlation was found to exist
between Granny most liked and her viewers (.0384) than
between Granny least liked and her viewers (-.2615).
Although the viewers of Granny were themselves different
in personality profile (r = -.1867), their perceptions
Xr
of Granny tended to be similar (rp = .2445).
CHAPTER V
SEX, AGE, LEVEL OF EDUCATION, AND STANDARD
OF LIVING OF VIEWERS AND OF ROLES
The main problem of the present study was to
compare the personality profiles of television viewers
with the perceived personality profiles of television
roles viewed by them. It was believed that it would
also be useful to compare certain personal data of the
viewers with their perception of the television roles
on these same data.
The personal data presented in this chapter
were not derived from a sample representative of the
general population, nor was the study primarily
designed to make comparisons on personal data but
rather to compare personality profiles.
Originally certain tests were designated to be
used for measurement of similarities and of differences
of the viewers and their perception of roles with
respect to four personal characteristics. However,
fragmentation of the sample into thirteen groups of
viewers reduced the n for each group to such a point
that valid conclusions could not be drawn from the
hypotheses as they were originally stated. For this
reason findings on the personal variables of age, sex,
level of education, and,standard of living were
109
reported here only to provide additional information
on these viewers and their perception of certain
television roles.
I. SEX
Findings
Because the number of viewers in three of the
groups was less than ten (Table IX), sex data were
reported in the form of proportions rather than in
coefficients of correlation. Figures were carried
to one place beyond the decimal point.
In the total sample population a proportion
of .62 was female and .38 was male. These proportions
remained fairly constant for each group with the
exception of the viewers of Ben Casey. The viewers
who rated Ben Casey a least-liked role had a larger
proportion of males (.6) than of females (.4), which
seemed to indicate that the role Ben Casey was less
appealing to men than to women.
Among most-liked and regularly-viewed roles the
ones with the largest proportion of male viewers were
John Novak and Granny, each with a proportion of .5
male viewers and .5 female.
Among the viewers who rated Granny a least-liked
role a larger proportion were females (.7) than males
TABLE IX
PROPORTION OF MALES AND PROPORTION OF FEMALES IN VIEWER GROUPS
Roles Judged
Female
Viewers
Male
Viewers Total
Proportion
Female
Proportion
Male
Most liked
Kimble 13 8 21 .6 .4
Van Dyke 25 13 38 .7 .3
Mason 15 12 27 .6 .4
Griffith 10 4 14 .7 .3
Cartwright 11 5 16 .7 .3
Novak 7 7 14 .5 .5
Least liked
Haskell 8 3 11 .7 .3
Williams 6 3 9 .7 .3
Abbott 10 5 15 .7 .3
Paladin 6 3 9 .7 .3
Casey 5 8 13 .4 .6
Granny
Most liked 7 6 13 .5 .5
Least liked 4 2 6 .7 .3
Totals
Proportions
127
.62
79
.38
206
Ill
(.3), while among those who rated her a most-liked
role there was an equal proportion of females and
males. However, the n for Granny least liked was too
i
small to permit valid conclusions.
Interpretations I
Findings on the relation of sex to the selection
of television roles for viewing did not reveal any
definite trends. The proportion of male (.4) and
j
female (.6) viewers selecting each role was fairly
consistent with the proportion of males and females in j
the total sample. Only the role of Ben Casey differed j
i
from the general trend. More males (.6) than females
1
(.4) were among the viewers who rated him a least-liked
role. This might have indicated that the role of Casey j
was particularly distasteful to male viewers. j
I
II. AGE
Findings j
i
Age data provided a higher number of positive
correlations between roles and viewers than any other
personal data variables (Table X).
1
I
Most-liked roles. Two of the six most-liked 1
roles and their viewers exhibited correlations on age i
that were significantly different from zero. They !
112
TABLE X
PRODUCT MOMENT CORRELATIONS BETWEEN PERSONAL
DATA OF VIEWERS AND OF ROLES VIEWED
Gtouds
Product Moment Correlations
Age
Standard
of Living
Level of
Education
Most liked
Kimble -.408 -.297 .000
Van Dyke -.261 .161 -.136
Mason .118 -.143 -.114
Griffith .365 .107 -.152
Cartwright • 532 ♦ .224 .152
Novak • 583 ♦ -.142 .439
Least liked
-
Haskell .361 .380 -.218
Williams .105 .000 .000
Abbott -.346 -.092 .264
Paladin -.501 -.474 -.135
Casey .264 -.013 -.289
Granny
Most liked -.265 -.139 -.512
Least liked -.493 -.525 -.875+
♦Significant at the ,05 Level of Confidence.
I
113
were Novak (.583) and Cartwright (.532) (Table X).
Griffith and Mason correlated positively but
insignificantly with their viewers, and Kimble and
Van Dyke showed a negative correlation on age with
their viewers.
Least-liked roles. Among the least-liked roles
Haskell, Williams, and Casey correlated positively
with their viewers on age, but at an insignificant
level. Paladin and Abbott correlated negatively with
their viewers, also at an insignificant level.
Age distribution of viewers. Mason and Novak
seemed to appeal to a wider and more even distribution
of age groups than did other most-liked roles (Table
XI). Van Dyke seemed to appeal to a more specific age
group— between thirty and thirty-nine years of age—
than did other most-liked roles. Griffith also
appealed to a wide age range of viewers. Those who
viewed Griffith formed an even distribution of persons
between forty and sixty-nine years of age.
A role most liked and least liked— Granny
Clampett. Granny, most liked, was regularly viewed
by persons in a wide range of age groups. Three of
the thirteen viewers were in the twenty to twenty-
nine age range; three, thirty to thirty-nine; two,
TABLE
DISTRIBUTION OF AGES OF VIEWER
Kimble Van Dyke Mason
Aee Category
Viewer
Groin)
Role
Grow)
Viewer
Group
Role
Group
Viewer R
Group G
20-29 3 0 3 2 2
30-39 10 11 22 31 13
40-49 5 9 5 2 7
50-59 2 0 4 0 4
60-69 0 0 1 0 0
70-79 0 0 0 0 1
80-89 0 0 0 0 0
Totals 20 35 27
Least-Liked Roles
Haskell* Williams Abbott
Viewer Role
Group Group
Viewer
Group
Role Viewer Role
Group Group Group
20-29 0 9 2 0 4 0
30-39 5 1 3 1 4 1
40-49 3 0 3 8 5 1
50-59 2 0 1 0 2 6
60-69 0 0 0 0 0 7
70-79 0 0 0 0 0 0
80-89 0 0 0 0 0 0
Totals 10 9 15
" ‘ Haskell role age categories differ from other categories in that
His categories read as follows: 14.5-25, 25.1-35, 35.1-46, 46.1-57, 57.1
TABLE XI
ttON OF AGES OF VIEWER GROUPS AND OF ROLE GROUPS
114
i
+»
0 0
£
Liked Roles
Dyke Mason Griffith Cartwright Novak
? Role Viewer Role Viewer Role Viewer Role Viewer Role
Group G t o u d Group Group Group Group Group Group Group
2 2 0 0 0 1 0 4 10
31 13 2 1 8 5 0 4 4
2 7 22 4 6 5 1 1 0
0 4 3 4 0 3 11 3 0
0 0 0 4 0 1 3 0 0
0 1 0 1 0 0 0 2 0
0 0 0 0 0 0 0 0 0
35 27 14 15 14
Least-Liked Roles Granny
ims Abbott Paladin Casey Most Liked Least Liked
%ole
iroup
Viewer
Group
Role
Group
Viewer
Group
Role
Group
Viewer
Group
Role
Group
Viewer
Group
Role
Group
Viewer
Group
Role
Group
0 4 0 1 0 1 0 3 0 0 0
1 4 1 5 0 8 12 3 0 1 0
8 5 1 1 7 3 1 2 0 2 1
0 2 6 2 2 1 0 4 1 2 0
0 0 7 0 0 0 0 1 5 1 1
0 0 0 0 0 0 0 0 6 0 2
0 0 0 0 0 0 0 0 1 0 2
15 9 13 13 6
ler categories in that he portrayed the role of a teenager.
55.1-46, 46.1-57, 57.1-68, 68.1-78, 78.1-89.
115
forty to forty-nine; four, fifty to fifty-nine; and
one, sixty to sixty-nine.
Age did not seem to be a factor of similarity
between the viewer group and its perception of the
role of Granny, most liked; instead, age correlated
negatively (-.265) (Table X). The n for the Granny
least liked group was too small to afford valid con
clusions.
Interpretations
Some roles tended to appeal to a wide age range
of viewers (Mason, Novak, Griffith), while others
appealed to persons in a more narrow range of ages
(Van Dyke). There was no clear tendency for the age
of most-liked roles to be more correlated with the
ages of their viewers than for least-liked roles to
be correlated on ages with their viewers. However,
age provided a larger number of positive correlations
between viewers and roles than did any other personal
variable.
III. LEVEL OF EDUCATION
Findings
Most-liked roles. Of the most-liked roles,
Novak and Cartwright correlated positively with their
viewers on level of education (Table X). Three of
116
the most-liked roles correlated negatively with their
viewers on level of education (Van Dyke, Mason, and
Griffith), and one exhibited a zero correlation (Kimble).
Only between Novak and his viewers was there a
fairly strong correlation (.439) on level of education.
Of the fourteen viewers of Novak, ten had completed at
least one year of college (Table XII)• Thirteen of the
fourteen persons in this group perceived Novak as having
completed at least one year of college.
Of most-liked roles, there was a greater variety
of levels of education among the viewers of Ben Cart
wright than among other most-liked roles (Table XII).
Among the Cartwright viewers two had completed an
elementary education, five had completed high school,
and seven had finished one or more years of college.
It was interesting to note that both Ben Cartwright
and his viewers scored as less intelligent on the 16
P. F. questionnaire (Appendix E).
Least-liked roles. Of five least-liked roles,
only Abbott correlated positively (.264) on level of
education with his viewers (Table X). Three least-
liked roles correlated negatively (Casey, -.289;
Haskell, -.218; Paladin, -.135), and Williams showed
a zero correlation.
TABLE XII
DISTRIBUTION OF LEVEL OF EDUCATION OF VIEWE
Number of Mo
Level of Education
Kimble Van Dyke Mason
Viewer
Group
Role
Group
Viewer
Group
Role
Group
viewer Role
Group Group
No Schooling 0 0 0 0 0 0
Elementary (1-8)
0 0 0 0 0 0
High School (9-12) 3 0 9 2 5 8
College (1 Year or more) 14 17 27 34 21 18
Totals
17 36 26
Number of Least-Liked Roles
Haskell Williams Abbott Pal
Viewer Role
Group Group
Viewer Role
Group Group
Viewer R61eo.
Group Group
Viewer
Group
No Schooling 0 0 0 0 0 0 0
Elementary (1-8) 0 0 0 0 0 2 0
High School (9-12) 3 8 0 7 6 11 4
College (1 Year or more) 7 2 9 2 9 2 4
Totals 10 9 15
117
TABLE XII
fEL OF EDUCATION OF VIEWER GROUPS AND OF ROLE GROUPS
Number of Most--Liked Roles
rke Mason Griffith Cartwright Novak
lole
iroup
Viewer
Group
Role
Group
Viewer
Group
Role
Group
Viewer
Group
Role
Group
Viewer
Group
Role
Group
0 0 0 0 0 0 0 0 0
0 0 0 0 0 2 3 0 0
2 5 8 3 11 5 8 4 1
34 21 18 11 3 7 3 10 13
26 14 14 14
>er of Least-Liked Roles Granny
Abbott Paladin Casey Most Liked Least Liked
%
*
ip
Viewer
Group
ROleo,
Group
Viewer
Group
Role
Group
Viewer
Group
Role
Group
Viewer
Group
Role
Group
Viewer
Group
Role
Group
) 0 0 0 0 0 0 0 5 0 1
) 0 2 0 3 0 0 1 4 0 3
7 6 11 4 1 5 1 6 3 2 1
>
4 9 2 4 4 6 10 6 1 4 1
15 8 11 13 6
| The strongest correlation on level of education
i among the least-liked roles and their viewers was
1 I
exhibited by Casey, who correlated negatively (-.289).
; This probably resulted from the fact that five of the
six Casey viewers had completed only a high school
i
education whereas Casey, in the role of a doctor, was
given a high rating on education.
All nine of the viewers who rated Danny Williams j
a least-liked role were of college level (Table XII).
Among the viewers of other least-liked roles there was
a more even distribution between high school graduates
i
and persons of college level than among the viewers of
Williams.
A most-liked and least-liked role— Granny
. |
! Clampett. There was a significantly negative correla- j
tion (-.875) between the level of education of those I
viewers who rated Granny a least-liked role and their !
perception of Granny (Table X)• There was also a |
strong negative correlation (-.512) on level of educa
tion between Granny, most liked, and her viewers.
The viewers of Granny, most liked, exhibited a
i
wider range of levels of education than those who j
rated her a least-liked role (Table XII). One viewer j
of Granny, most liked, had only an elementary education; j
j
six had a high school education; and six had at least
119
one year of college education. Ttoo-thirds of the
viewers who rated Granny a least-liked role had com
pleted one or more years of college. However, the
n for Granny, least liked, was only six, and findings
could not be considered valid.
The viewers of Granny attributed many levels
of education to her. Two-thirds of the viewers of
Granny, most liked, felt that she had no more than
an elementary level of education.
Interpretations
Level of education did not seem to offer a clear
index for determining the role preferences of viewers.
However, some roles tended to attract persons of a
similar level of education as that portrayed in the
role. Novak, the high school teacher, correlated
positively with his viewers on level of education. Ten
of the fourteen viewers of Novak had completed one or
more years of college. In contrast, the viewers of
Cartwright, although correlating only mildly with the
role Cartwright, exhibited a broader range of levels of
education. Perhaps the Western role tends to attract
persons of varying intellectual levels, while more
contemporary subjects appeal to more specific educa
tional levels.
IV. STANDARD OF LIVING
120
Findings
The standard of living of an individual can be
estimated with such factors as income, level of educa
tion, place of residence, social rank, etc. However,
for this study the term "standard of living" of a
viewer was defined simply as "annual combined family
income," in the belief that income was an important
ingredient in the determination of a person's standard
of living.
Standard of living" of a role was established
by the viewer, who rated the role on a five point
scale which ranged from a designation of "very high"
to "very low."'*'
There were no statistically significant corre
lations between the standard of living of viewer
groups and that perceived for the roles (Table X).
The highest positive correlation of standard of living
was found between Eddie Haskell, a least-liked role,
and his viewers (.380). The highest negative correla
tion was found between Paladin, also a least-liked
role, and his viewers (-.474).
^See Chapter III, page 86.
Most-liked roles. Among the most-liked roles
there were no large correlations on standard of living,
either positive or negative. The largest positive
correlation was between Cartwright and his viewers
(.224). The largest negative correlation was between
Kimble and his viewers (-.297).
The standard of living of the viewers of Andy
Griffith ranged more broadly and more evenly from low
to high than did the viewer group of any other most-
liked role (Table XIII). There was an even distribution
of two persons in each of the first four categories
beginning with very low, and five in the highest cate
gory. This indicated that the role of Andy Griffith
appealed to persons of all economic levels. Cartwright
and Novak also appealed to a broad range of viewers
with respect to their standard of living.
Conversely, about two-thirds of the viewers of
Dick Van Dyke were in the top standard of living cate
gory.
Least-liked roles. Among the viewers of least-
liked roles the widest distribution of standard of
living was exhibited by the viewers of Abbott, who
seemed to appeal to persons of varying economic levels
ranging from $4,000 to over $10,000 combined annual
family income.
TABLE XIII
DISTRIBUTION OF STANDARD OF LIVING OF VIEWER
Number of Mos1
Kimble Van Dyke Mason
Viewer Role Viewer Role viewer Role
Standard of Livine* Group Group Group Group Group Group
Very Low
0 1 0 0 0 0
Rather Low 1 8 1 0 3 0
Average 7 4 6 8 9 0
Rather High 8 7 6 27 3 18
Very High 4 0 24 2 12 9
Totals 20 37 27
Number of Least-Liked Roles
Haskell Williams Abbott Palac
Viewer Role
Group Group
Viewer Role
Group Group
Viewer Role
Group Group
Viewer
Group
Very Low 0 0 0 0 0 0 0
Rather Low 0 1 0 0 1 1 0
Average 3 4 2 0 5 7 1
Rather High 2 5 2 8 1 5 3
Very High 5 0 4 0 7 1 5
Totals 10 8 14
c
« ■
♦Viewer standard of living was rated as follows: Under $4,000
4.000-5,999
6.000-7,999
8.000-9,999
10,000 and over
Role standard of living was rated as follows: very low, rather low, aver
122
TABLE XIII
NDARD OF LIVING OF VIEWER GROUPS AND OF ROLE GROUPS
Number of Most--Liked Roles
ke Mason Griffith Cartwriffht Novak
ole
roup
Viewer Role
Group Group
Viewer
Group
Role
Group
Viewer
Group
Role
Group
Viewer
Group
Role
Group
0 0 0 2 0 1 0 1 0
0 3 0 2 0 0 0 2 1
8 9 0 2 6 2 2 3 12
27 3 18 2 6 5 9 2 1
2 12 9 5 1 8 5 6 0
27 13 16 14
er of Least-Liked Roles Granny
Abbott Paladin Casey Most Liked Least Liked
Viewer
0 Group
Role
Group
Viewer
Group
Role
Group
Viewer
Group
Role
Group
Viewer
Group
Role
Group
Viewer
Group
Role
Group
0 0 0 0 0 0 0 0 0 2
1 1 0 1 0 1 1 4 0 2
5 7 1 1 7 8 5 6 1 0
1 5 3 5 3 4 2 1 1 2
7 1 5 2 3 0 5 2 4 0
14 9 13 13 6
: Under $4,000
4.000-5,999
6.000-7,999
8.000-9,999
10,000 and over
rery low, rather low, average, rather high, very high.
123
In contrast to the viewers of Abbott, over half
of the viewers of Ben Casey were in the $6,000 to
$8,000 "average" standard of living category.
A role most liked and least liked— Granny
Clampett. Four of the six viewers who rated Granny a
least-liked role were in the "very high" standard of
living category. Only a proportion of .4 of the
viewers who ranked Granny a most-liked role were in
this category.
Viewers who rated Granny a least-liked role
tended to assign her a lower rank on standard of
living than those who rated her a most-liked role.
That is, those viewers who rated Granny as a least-
liked role tended to be in an upper standard of living
category and tended to rate Granny low on standard of
living, whereas those who rated Granny a most-liked
role represented a wider range of standard of living
and assigned Granny a broader range of scores on
standard of living.
Interpretations
Standard of living, as measured in this study,
did not provide any significant correlations between
viewers and a role, and therefore did not serve as a
useful index to determine whether or not the liking
124
or disliking of a role is associated with one's
standard of living.
However, it seemed clear that some roles did
appeal to a much broader range of viewers on standard
of living than did other roles. The viewers of Andy
Griffith varied more broadly and more evenly from low
to high than did other viewer groups. This seemed to
indicate that Griffith appealed to persons of all
economic levels. Conversely, Van Dyke appealed mostly
to persons in the top standard of living category.
A strange phenomenon occurred in the rating of
Granny on standard of living. Although she portrayed
the role of a millionaire in the show "Beverly Hill
billies," she was rated by her viewers from very low
to very high on standard of living. Those who rated
her a least-liked role tended to give her a lower
rating than those who rated her a most-liked role.
Perhaps viewers took into consideration such factors
as social habits, education, manners, etc. when they
rated Granny on standard of living.
V. SUMMARY
Although the present study was not designed
primarily to provide descriptive data on personal
variables, but rather to make comparisons between
125
personality profiles, it was thought that data on sex,
age, level of education, and standard of living might
be useful in determining what personal elements tended
to draw viewers to view specific roles.
The proportion of males and females who selected
the various roles as either most liked or as least
liked did not vary appreciably from group to group,
with the exception of Ben Casey. Most groups of
viewers had proportions of .6 females and .4 males.
Conversely, viewers of Casey had a proportion of .4
females and .6 males.
Of the four personal data variables, there were
more positive correlations between roles and viewers
on age than on any other variable. Among viewers of
most-liked roles, two of the six groups showed a
positive correlation on age that was significantly
different from zero. Griffith, Novak, and Cartwright
were viewed by a wider range of age groups than other
most-liked roles. Granny, most liked, was viewed by
a wider range of age groups than was Granny, least
liked. However, there was a negative correlation
between Granny and both viewer groups on the variable
of age.
Only between Novak and his viewers did there
exist a rather strong correlation (.439) on level of
126
education. Ten of the fourteen Novak viewers had
completed some college, and thirteen rated Novak as
a college graduate.
Of the least-liked roles, Ben Casey showed a
substantial although statistically insignificant
correlation with his viewers on level of education
(-.289). Five of the eleven viewers of Casey had
completed a high school education, while six had
completed one or more years of college.
The viewers of Ben Cartwright showed the
widest range of levels of education of all most-liked
roles. His viewers had levels of education that
ranged from elementary school through college. In
contrast, all nine of the viewers of Danny Williams,
a least-liked role, had completed one or more years
of college.
There were no significant correlations between
the standard of living of viewer groups and that
perceived for roles. The viewers of Andy Griffith
exhibited a broader range of standard of living than
did any other group.
CHAPTER VI
PERSONALITY TRAITS THAT MOST CONSISTENTLY
CHARACTERIZED MOST-LIKED ROLES, LEAST-
LIKED ROLES, AND VIEWERS OF THESE ROLES .
Analysis of findings in the present study
revealed that certain trait factors in the 16 P. F.
profiles seemed to distinguish rather consistently
between most-liked roles and least-liked roles, and
between roles and their viewers.
!
To better understand these traits, which seemed j
|
to have a consistent bearing upon viewing behavior and j
possibly on the selection of roles for viewing, they |
were examined in some detail, and the descriptions
j
presented in this chapter. Only the profiles of those j
roles which showed a significant correlation with the
j
profile of their viewer group were included for com
parison.
An outstanding personality trait was defined j
as any trait that deviated one or more stens either |
i
above or below the average range of sten scores five
and six.
The material reported in this chapter, which j
provided a psychological interprdtation of the various
sixteen personality factors, was taken entirely from the
Handbook for the Sixteen Personality Factor Question-
128
! naire.1
I. FACTOR Q3, MOST DIFFERENTIATING TRAIT
With all three of the most-liked roles factor
Q3+, high self-concept control, was an outstanding
trait, and with all three least-liked roles Q3-, low
integration, was an outstanding trait (Table XIV).
Factor Q3+ represents the tendency for an indi
vidual to respond in ways that are socially approved.
It indicates that a person has a clear self image and
an exacting will power. He exercises vigorous self-
control and is highly self-disciplined. Factor Q3+
is an important ingredient in the mature personality.
It portrays one who has crystallized a self image that
is acceptable to himself; one who responds to his
environment in ways that are acceptable both to him
self and to others.
Factor Q3-, low integration, represents the
opposite pole. This kind of person has not yet deter
mined clearly just who he is. He has a vague concept
of self, or he may have a self image that he cannot
^■Raymond B. Cattell and Glen F. Stice, Handbook
for the Sixteen Personality Factor Questionnaire: 1 1 The
16 P. F. TestT17 Forms A, B, and C (1957 Edition, with
1962 Supplementation; Champaign, Illinois: The Institute
for Personality and Ability Testing, 1957), pp. 10-19.
TABLE XIV
OUTSTANDING PERSONALITY TRAITS OF MOST-LIKED ROLES AND OF LEAST-LIKED ROLES*
Totals Per Row
Trait Most--Liked Roles Least--Liked Roles Most Most Least Least
Factors Kimble Mason Cartwright Granny Haskell Williams Liked+ Liked*- Liked-f Liked-
A
_
+ 1 1
B
—
+ 1 1
C +
— — —
1 1 2
E + + 2
F
—
+ + + 1 3
G +
— —
1 2
H
—
+ + 1 2
I +
— —
+ 1 1 2
L + + + 3
M + 1
N + +
— -
2 2
0
_
+ 1 1
Q1
+
— — 1 2
Q2 + +
—
2 1
03 + + + — — . — 3 3
04
+ 3 1
*Only those roles whose profiles showed a significant correlation, positive
or negative, with that of their viewer group are included in this table.
(-) = A trait sten score of four or less on the 16 P. F. test profile.
J
(+) = A trait sten score of seven or more on the 16 P. F. test profile.
129
accept. He often is rejected by others because his
behavior is not acceptable to them. He lacks will
power, responds erratically to circumstances, and
generally is careless about conforming to social norms.
This type of person is rejected by many because he is
unpredictable and tends to negate, in his own being,
the values and standards by which the majority live.
The three most-liked roles which exhibited Q3+
as an outstanding trait were Richard Kimble, Perry
Mason, and Ben Cartwright. These three roles repre
sented three different kinds of persons: Richard Kimble,
a fugitive, intelligent, living by his wits; Perry
Mason, an ingenious sleuth, fighting for justice and
society; and Ben Cartwright, a pioneer land owner of
the early West, crusading for law and order, bringing
stability and civilization to a wild frontier. Mason
i
and Cartwright had one trait in common— they knew who
they were. They had come to terms with society and,
having found what society expected of them, were deter
mined to use their lives to strengthen that society. I
Having found the necessary reinforcement of ego from
social approval, fhey showed an inner strength and
self-control. Because of this others felt secure in
their presence and did not fear sudden bursts of
erratic behavior. They appeared as trusted, dependable
131
individuals.
The three least-liked roles which exhibited
factor Q3- as an outstanding trait were Granny Clampett,
Eddie Haskell, and Danny Williams. All three of these
roles appeared to be unpredictable in their behavior.
No one could be certain how they would react at any
given moment. Granny was a comedy character in a most
unlikely situation (a hillbilly in Beverly Hills) and j
i
existed more in fantasy than in reality. Eddie Haskell
!
was an obnoxious teenager, a nuisance in most respects.
He was not trusted either by his peer group or by
I
adults. He had no clear self image and no firm set of j
internal standards with which to gauge his behavior.
Danny Williams was always emotionally upset. He kept
his family and friends in turmoil. He was constantly
getting mixed up with people in most unlikely situa- !
!
tions. He exercised little control over his behavior j
and thus kept himself and those around him upset. |
i
Viewers tended to like most those roles that |
I
appeared to know who they were and where they were
going. They seemed to admire responsible citizens who
brought order, justice, and understanding to the com-
!
munities of which they were a part. They liked depend- 1
able persons who exercised self-control and therefore
posed no threat to those around them. In short, it
seemed that the viewers liked most those roles which
could anchor the viewers' own inner resolutions— roles
that portrayed the kinds of persons who strengthen and
further the goals of society. Viewers did not like
roles that portrayed irresponsible, unstable types of
persons.
II. PERSONALITY TRAITS APPEARING MOST
CONSISTENTLY IN MOST-LIKED ROLES
The following outstanding personality traits
appeared in all three of the most-liked roles in this
study:
Q3+ — controlled, socially precise, self-
disciplined, compulsive (High self-
concept control)
Q4- — relaxed, tranquil, torpid, unfrustrated
(Low urgic tension)
These outstanding traits appeared in two of the three
most-liked roles:
N+ — shrewd, calculating, worldly, penetrating
(Shrewdness)
Q2+ — self-sufficient, prefers own decisions,
resourceful (Self-sufficiency)
133
j
Factor Q4-« An Important
Trait in the Selection
of Most-Liked Roles
The same three roles that exhibited trait Q3+,
strong self-control, were outstanding also on trait
Q4-, relaxedness. This kind of person is essentially
unfrustrated, able to resolve inner tensions without
allowing them to build up into major anxieties. The
attraction of viewers to roles with these traits seemed
to indicate that viewers do not like roles which, by
their instability and high tension, pose as an emotional
threat to them.
i
III. PERSONALITY TRAITS APPEARING MOST
CONSISTENTLY IN LEAST-LIKED ROLES
All three of the least-liked roles exhibited the
following outstanding traits:
F+ — happy-go-lucky, heedless, gay,
enthusiastic (Surgency) i
j
L+ — suspicious, self-opinionated, hard j
to fool (Protension)
Q3- — casual, careless of protocol, untidy,
I
follows own urges (Low integration)
i
These outstanding traits appeared in two of the three
least-liked roles: I
134
i
C- — affected by feelings, emotionally less
stable, easily upset (Lower ego strength)
E+ — assertive, independent, aggressive, !
stubborn (Dominance)
G- — expedient, a law to himself, by-passes
obligations (Weaker super ego strength)
H+ — venturesome, socially bold, uninhibited,
spontaneous (Parmia)
I- — tough-minded, self-reliant, realistic,
no-nonsense (Harria)
N- — forthright, artless, natural, sentimental
(Artlessness)
Ql— — conservative, respecting established
ideas, tolerant of traditional diffi-
j
culties (Conservatism)
|
Factors L+ and F+, Important I
Traits for Distinguishing
j
Least-Liked Roles
Trait factor F+. All three least-liked roles
rated highly on trait factor F+. This trait describes
a person who is happy-go-lucky and enthusiastic. It
is one of the most important components in extraversion, !
and a second-order composite of seyeral source traits. j
It is closely related to the elation-depression scale
135
along which psychotics swing from one extreme to the
other.
The interpretation of this trait, as to whether
it is pointing to a most-liked role or to a least-liked
role, seems to be dependent upon what other traits are
outstanding in that same personality profile. For
example, an extrovert can be a joy to others or he can
be a bore. A happy-go-lucky attitude can be an escape
from facing reality or it can be a sign that the person
does not take himself too seriously and is flexible
and able to adjust to changing circumstances.
Therefore, to use this trait to predict whether
or not a role would be a most-liked or a least-liked
role requires the observation of other traits in the
personality of that role.
A combination of traits F+ (happy-go-lucky),
Q3+ (controlled), and Q4- (relaxed) would possibly
indicate a most-liked role for this sample group, while
a combination of traits F+ (happy-go-lucky) and Q3-
(low integration, casual, careless of protocol) would
lead one to suspect that this was a least-liked role.
Add to these the other traits given which tend to
characterize most-liked and least-liked roles and you
would have an indication as to whether or not the F+
trait represented a healthy or an unhealthy happy-go-
136
lucky attitude.
Trait factor L+. Three of three least-liked
roles rated highly on this factor. Not one of the
most-liked roles rated highly on this factor. This
seemed to indicate that factor L+ was a highly disliked
personality trait and roles which rated highly on this
factor were disliked by this sample of viewers.
Factor L+ (Protension) shows itself clearly in
abnormal populations. Although its presence does not
necessarily imply a disorder, this type of personality
often is unpopular. The protensive person shows a
high inner tension which often takes the form of a
feeling of social insecurity, together with a compen
satory behavior and projection.
In group dynamics experiments the protensive
person is rated unpopular. Occupationally, engineers,
mechanics, nurses, and psychiatric technicians are low
on this factor, while executives and policemen are
above average.
IV. PERSONALITY TRAITS ON WHICH ROLES
AND THEIR VIEWERS WERE MOST ALIKE
Table XV lists the traits on which the roles
and their viewers scored most closely. The traits
TABLE XV
PERSONALITY TRAITS ON WHICH ROLES AND VIEWERS DIFFERED NO MORE THAN ONE STEN
Totals Per Row
Trait Most-Liked Roles Least-Liked Roles Most Most Least Least
Factors Kimble Mason Cartwright Granny Haskell Williams Liked* Liked-- Liked* Liked-
A - + - + 1 2 1
B + — + 1 1 1
C
E
— —
2
F
— —
2
G + + 2
H
—
1
I + + 1 1
L
— —
2
M
—
1
N
0 + + 2
Q1
Q2 - - 2
Q3 + + + -
3 1
04 + 1
(-) = Trait sten score of four or less on the 16 P. F. test profile.
(+) = Trait sten score of seven or more on the 16 P. F. test profile.
138
listed were those on which the profile score of the
role and the profile of the viewer differed no more
than one sten.
Most-Liked Roles
There was only one trait factor— Q3+— on which
all three of the three most-liked roles had scores
which differed no more than one sten. This factor
represented an area in which the viewers' perception
of the role was consistently similar to the viewers'
own personality profile.
Persons who score highly on factor Q3+ exhibit
strong self-control, persistence, conscientiousness,
and consideration of others. All of these qualities
are strongly desirable in our society. Kimble, Mason,
and Cartwright held trait factor Q3+ in common with
their viewers. At this point, apparently, they pro
vided reinforcement for the personal value structure
of their viewers. Kimble and Cartwright shared traits
A- (reserved, detached, cool) and trait E- (humble,
mild, conforming) with their viewers. Mason and
Cartwright held in common with their viewers trait
F- (sober, prudent, serious) and trait G+ (expedient,
a law to himself). Kimble and Mason shared with
their viewers trait L- (trusting, adaptable, free of
jealousy).
Least-Liked Roles
Two of the three least-liked roles showed a
similarity with their viewers on trait factors 0+ and
Q2- (Table XV).
Granny and Haskell, both least-liked roles,
held factor 0+ in common with their viewers.
Persons who rate positively on this factor are
generally timid, insecure, worrying, anxious, and
depressed. This kind of person often feels over
fatigued by exciting situations, is easily downhearted
and remorseful, feels that people are not as moral as
they should be, is inclined to piety, and prefers
books and quiet interest to people and noise. Factor
0+ is one of the largest factors in anxiety and tends
to be generally high in neurotics and psychotics.
It is possible that viewers who rated positively
on factor 0+ were high anxiety persons who did not like
to observe roles that were of the same anxietous nature.
They might have preferred, rather, to view roles that
were placid and confident, i.e., roles that rated
highly on factor 0-.
Those viewers who rated Granny as a most-liked
role rated her as average on factor 0 and were them
selves in the normal range on this factor. But those
who rated Granny as a least-liked role rated her
140
highly on factor 0+ and scored similarly on this factor
themselves* This tended to substantiate the contention
that viewers who were themselves apprehensive did not
like to view roles that they perceived also to be
apprehensive, as this would tend to increase their
own anxieties.
Viewers tended to like roles that were relaxed
and self-confident— high on Q3+— and to dislike roles
that were anxietous and uncertain— high on trait 0+.
Factor Q2- is one of the major factors in
introversion. Persons scoring negatively on factor
Q2 tend to go with the crowd rather than stand alone.
They value social approval and tend to be conventional
and fashionable.
Both Haskell and Williams were similar to their
viewers on trait factor Q2-. Like their viewers they
tended to be group-dependent, followers rather than
leaders.
V. PERSONALITY TRAITS ON WHICH ROLES
AND VIEWERS DIFFERED MOST
There were many more differences between the
personality traits of least-liked roles and their
viewers than there were between most-liked roles and
their viewers (Table XVI). There were twenty-one
TABLE XVI
PERSONALITY TRAITS ON WHICH ROLES AND VIEWERS DIFFERED MORE THAN THREE STENS
Trait
Factors
Most-Liked Roles Least-Liked Roles Like Patterns
Mason Kimble Cartwright Haskell Williams Granny Most Liked Least Liked
A +- 1
C -+ -+ 2
E +- +- 2
F -+ +- +- 1 2
G -+ -+ 2
H +- +- 2
I -+ -+ -+ 1 2
L +- +- 2
N -+ -+ 2
Q1
Q2 +- +- -+ 2 1
Q3 -+ -+ 2
04 +- 1
The sign +- indicates that the role and the viewer scores for a given
factor had a deviation of four or more sten scores.
The sign designating the role appears first (+), followed by the sign
designating the viewer (-)„
The sign (+) indicates a score that deviates towards the higher end of
the sten scale and a sign of (-) indicates a score that deviates towards a
sten score of one.
141
142
instances among least-liked roles and their viewers
in which the sten scores on a given trait differed more
than three stens, while there were only four such
instances among most-liked roles and their viewers.
This observation substantiated earlier findings which
revealed a negative correlation between the profiles
of least-liked roles and their viewers in five of six
cases, and a positive correlation between the profiles
of most-liked roles and their viewers in all seven
cases.^
Most-Liked Roles
Factor Q2. Most-liked roles differed from their
viewers on factor Q2 in only two of three cases. Not
too much importance could be attached to this finding.
On factor Q2 the roles of Perry Mason and
Richard Kimble rated positively, indicating that they
were perceived to be self-sufficient and resourceful,
while the scores of their viewers on factor Q2 fell
I
!
within the normal range. As both Mason and Kimble
i
represented stereotypes of undercovermen, or detectives, j
I
it was easy to understand why their viewers rated
them extremely on factor Q2+, which indicates self- !
sufficiency and resourcefulness.
2
See Chapter IV, pages 99 and 101.
143
Least-Liked Roles
Many more differences were exhibited between
least-liked roles and their viewers than between most-
liked roles and their viewers.
Factor E. All three least-liked roles scored
highly on the positive end of the E scale. Williams
and Granny, least-liked, differed significantly from
their viewers on this factor. While the roles rated
positively, indicating a high degree of assertiveness
and independence, their viewers fell within the average
range. Haskell had a high positive score also on
factor E, and his viewers scored slightly above average
in the same direction on this factor. The least-liked
roles differed from their viewers most consistently in
that they were assertive while their viewers exhibited
humility.
Factor E+ (assertiveness) loads highly on fac
tors described as dominance, independent mindedness,
sternness, solemnity, unconventionality, toughness,
and tendency toward getting attention. None of the
most-liked roles differed greatly from their viewers
on this factor, but Kimble and Cartwright rated
similarly to their viewers on the negative pole of
factor E (humility).
144
Factor C. On this factor the least-liked roles
of Haskell and Williams were rated negatively, and
their viewers scored positively. These roles were seen
to be emotionally less stable than were their viewers.
Factor F. On this factor Haskell and Williams
were rated positively, while their viewers scored
negatively. While the roles were perceived to be
happy-go-lucky, their viewers were more sober and
prudent.
Kimble, a most-liked role, and his viewers
scored in an opposite direction. Kimble was perceived
as sober, while his viewers were more happy-go-lucky.
Factor G. On this factor Haskell and Williams
were seen to be expedient, a law to themselves, while
their viewers were more conscientious and rule-bound.
A negative rating on factor G, such as was received
by these two least-liked roles, indicated weak super
ego strength.
Factor H. On this factor Williams, and Granny
least liked, were rated positively, while their viewers
scored negatively. The least-liked roles were seen to
be venturesome and uninhibited, while their viewers
tended to be shy and restrained.
145
Persons who score highly on H+ are thick-skinned,
adventurous, and responsive and have an overt interest
in the opposite sex. Descriptions loading negatively
on factor H are shyness, retiring in face of the opposite
sex, aloofness with a tendency to be embittered.
One of the principal characteristics of the
viewers of Granny, least liked, was shyness. Therefore,
it seemed that an encounter with a person or a role as
thick-skinned or adventurous as Granny would not provide
reinforcement for these viewers, but a threat, and would
tend to destroy any sense of security which they might
have. For this reason viewers, such as those who rated
highly on shyness, would most probably dislike a role
which rated highly on H+, venturesomeness.
i
Factor I. Haskell, and Granny least liked,
scored negatively on this factor, indicating that they
were perceived to be tough-minded and self-reliant
while their viewers were more tender-minded and depen- j
i
dent. Interestingly, Cartwright, a most-liked role,
scored in a similar fashion on this factor with his
viewers. It is easy to see how this self-reliance
could be attributed to the capable manager of the
Ponderosa ranch.
146
Factor L. On this factor Granny, least liked,
and Williams were rated positively, while their viewers
scored negatively. Granny and Williams were seen to be
suspicious and self-opinionated, while their viewers
were more trusting and free of jealousy.
Factor N. Williams, and Granny least liked,
rated negatively on factor N (forthright, artless),
while their viewers scored positively (shrewd, cal
culating) .
Persons who score negatively on factor N are
described as "socially clumsy," vague, sentimental
minded, gregarious, simple in tastes, and lacking in
self-insight. On the opposite end of the scale are
persons who are polished, socially alert, exact,
calculating of mind, and emotionally disciplined.
Factor Q3. On factor Q3 Haskell and Williams,
least-liked roles, were rated negatively and their
viewers scored positively. These least-liked roles
were seen to be casual and careless of protocol, while
their viewers were controlled, socially precise, and
self-disciplined.
On the nine factors described in this section,
two of the three least-liked roles differed in a
similar direction from the sten score of their viewers
by more than three stens.
The least-liked roles were perceived to be out
going, emotionally less stable, assertive, happy-go-
lucky, expedient, venturesome, suspicious, forthright,
and careless of protocol. The viewers of these roles
tended to be more reserved, emotionally stable, humble,
sober, conscientious, shy, tender-minded, trusting,
shrewd, and self-disciplined. In other words, viewers
who were quiet and conservative did not like to view
roles which were gregarious, assertive, and socially
undisciplined.
VI. PERSONALITY TRAITS CHARACTERIZING VIEWER GROUPS
OF GRANNY CLAMPETT: ONE WHICH RATED HER A MOST-LIKED
ROLE AND ONE WHICH RATED HER A LEAST-LIKED ROLE
Similarities
Both those viewers who rated Granny a most-liked
role and those who rated her a least-liked role scored
negatively on factors E, H, and L. That is, both
groups were characterized as being humble, shy, and
trusting (Table XVII).
Differences
Traits B and Q3 distinguished most clearly
between the two groups of viewers. Those who rated
Granny as a most-liked role appeared to be slightly
148
TABLE XVII
OUTSTANDING PERSONALITY TRAITS OF VIEWERS WHO
RATED GRANNY A MOST-LIKED ROLE AND VIEWERS
WHO RATED GRANNY A LEAST-LIKED ROLE
Devia Devia
Viewers Rating tion Viewers Rating tion
Trait Granny a Most- from Granny a Least- from
Factors Liked Role Mean4 1 Liked Role Mean*
A
_
3
B
—
1 + 1
C
—
2
E
—
1
—
2
F
—
2
G + 2
H
—
1
—
4
I
L
—
1
—
2
M + 2
N + 2
0 + 2
Q1
Q2
Q3 + 1
tm
2
04 + 2
♦The number of sten scores that a factor score
deviates from the average range of stens five and six.
(-) = Trait sten score of four or less on the
.16 P. F. test profile.
(+) = Trait sten score of seven or more on the
16 P. F. test profile.
Factors on which groups varied in the same
direction:
E — Humility
H — Shyness
L — Trustingness
Factors on which the groups varied in opposite
directions:
B Most-liked viewer group B-, less intelligent.
Least-liked viewer group B+, more intelligent.
Q3 Most-liked viewer group Q3+, more controlled)
socially precise.
Least-liked viewer group Q3-, more relaxed,
tranquil. '
149
less intelligent and more casual than the group which
rated Granny a least-liked role. The latter group
appeared to he more intelligent and self-disciplined.
VII. SUMMARY
Trait factor Q3 distinguished consistently
between most-liked roles and least-liked roles. Kimble,
Mason, and Cartwright, most-liked roles, were rated
positively on factor Q3, indicating high self-concept
formation, while Granny, Haskell, and Williams rated
negatively, revealing a low self-integration.
Trait factor Q4-, relaxedness, was an important
trait for most-liked roles. The same three most-liked
roles that exhibited Q3+, high self-concept, were out
standing also on trait Q4-*
Trait factors F+, happy-go-lucky; L+, suspicious-
ness; and Q3_, casualness appeared as outstanding traits ;
in all three least-liked roles.
|
There were many more differences between the j
personality traits of least-liked roles and their
viewers than there were between most-liked roles and
their viewers. This tended to reinforce the findings
1
on the initial hypothesis, i.e., there is greater
similarity between the personality profiles of viewers
and the perceived personalities of most-liked roles
than between personality profiles of viewers and the
perceived personalities of least-liked roles.
CHAPTER VII
SUMMARY, CONCLUSIONS, IMPLICATIONS,
AND RECOMMENDATIONS
I. SUMMARY
Problem and Purposes
The problem of this study was to provide
quantitative measurement of the degree of identifica
tion or lack of identification that takes place
between viewers of television and particular television
roles which they rate as most-liked roles and view
regularly, and certain television roles which they
rate as least-liked roles and view infrequently.
Some purposes of such a measurement might be to
lend valuable support to conclusions already reported
j
in the literature, and to provide some new insights
into the relationships of personalities of viewers and
viewers' selections of particular television programs. j
I
The findings of this study should be of value
to (1) producers, writers, and directors of television j
programs who seek to build regularly viewing audiences j
for their programs; (2) sponsors of programs who wish
to direct their programs to specific target audiences;
(3) students of social behavior who wish to know more
of how people use the viewing of television for ego-
reinforcement; and (4) students of mass media who seek
quantitative evidence in support of existing theories
on identification in use of mass media.
The problem was to determine if there was not
greater similarity between the personalities of tele
vision viewers and the perceived personalities of
specific television roles that they liked most than
there was between the personalities of television
viewers and the perceived personalities of television
roles that they liked least.
To facilitate the measurement of similarity or
dissimilarity between viewers and most-liked roles
and between viewers and least-liked roles, two major
questions were proposed:
1. Was there greater similarity or difference
between the personality profiles of tele
vision viewers and the perceived personality
profiles of regularly-viewed, most-liked
television roles than there was between the
personality profiles of television viewers
and the perceived personality profiles of
least-liked television roles?
2. Was there greater similarity or difference
between the television viewers and their
perception of most-liked television roles
153
than there was between television viewers
and their perception of least-liked tele
vision roles with respect to age, level of
education, standard of living, and sex?
Review of Literature
Literature from the fields of radio, television,
motion pictures, general mass media, psychology, and
sociology was reviewed and organized in six areas: (1)
television viewing: a para-social situation, (2) per
ception of persons: an important element in television
viewing, (3) personality: a central determinant in the
perception of persons, (4) functional and dysfunctional
values of television viewing, (5) identification: an
important function of television viewing, and (6)
program selection: a result of role-viewer identifica
tion.
The selection and perception of particular
content in mass media were considered to be types of
social action.1 As a part of one's social environment
television provides a type of face-to-face encounter
o
which Horton and Wohl called a para-social situation.
^Chapter II, page 25.
^Chapter II, pages 25-26.
154
Personal involvement of the spectator with roles per
ceived in various mass media has been reported by
3
students of mass media for many years.
The concept of the para-social nature of the
viewing of television provided the foundation for this
study, the problem of which was to quantify the kind
and degree of relationship that obtains between the
personality of the viewer and the personality of the
role as perceived by that viewer.
In television viewing persons tend to perceive
television roles much as they would perceive persons
in real life. And one's perception of another person
is conditioned and limited not only by the nature of
the person being perceived, but by the personal experi
ences that the perceiver brings to the perceiving
4
situation. According to Tagiuri and Petrullo, one's
perception of another is influenced by the situation,
the person being perceived, and the perceiver himself.
Studies reviewed to this point indicated that
personality was an important element in the perception
^Chapter II, pages 26-27.
4
Chapter II, pages 28-31.
5Chapter II, page 33.
155
of persons and would provide a useful tool for the
study of television roles as perceived by their viewers.
The systematizing of human knowledge about
personality falls into three historical phases: (1)
the literary and philosophical, (2) the proto-clinical,
and (3) the quantitative and experimental. The third
phase can be divided into two parts: earlier univariate
experiments and later multivariate experiments.
Guilford and Cattell have contributed sub
stantially to the study of personality through the use
of factor analysis. Guilford's definition of person-
ality as one's "unique pattern of traits" was adopted
for this study, and Cattell's Sixteen Personality
Factor Questionnaire was selected as the test instru
ment with which to measure the personalities of the
viewers and their perceptions of certain television
roles. As Rosenzweig claimed, the subjective inventory
questionnaire, while sometimes having questionable
validity, probably provides a wider scope of informa-
g
tion about personality than any other instrument.
g
Chapter II, page 34.
7
Chapter II, page 36.
Q
Chapter II, page 37.
Cattell's 16 P. F. questionnaire distinguishes
sixteen first order trait factors. It was thought
that these factors represented relatively enduring
patterns of organization of personality which provided
information of more lasting value than those descrip
tions of personality that were derived from mere state
ments of opinion.
Students of mass media have reported a need to
know what part television viewing plays in the personal
Q
lives of people. Viewing of television can serve
different functions for the viewer, and, as Klapper
stated, functional analysis should reveal for whom a
program is functional, nonfunctional, or dysfunc
tional.^ Pearlin said that the "escape" element in
the viewing of television might serve as a "safety
valve" for the viewer.11 Fearing reported that viewing
of television can be used to affirm one's own value
12
structure or as a substitute outlet for frustrations.
Through quantitative measurement of the per
sonality of the viewer and the viewer's perception of
o
Chapter II, page 40.
^Chapter II, page 43.
11Chapter II, page 39.
12Chapter II, pages 38-39.
157
the personalities of the roles that he viewed, the
present study attempted to determine whether or not
similarity of personality between the viewer and the
role was associated with the liking of a role and
therefore whether or not the liking of a role had
functional value for the viewer through identification
and reinforcement of personal values.
Identification has long been recognized as an
important motivational force in a person's use of mass
media. In 1944 Arnheim, reporting a study of radio
daytime listening, claimed that the resemblance of a
listener to a role invited identification by the
13
listener. In 1959 Emery reported an extensive
analysis of television viewing, and concluded that a
viewer will identify with that role which is most like
the viewer's self-concept or that which he would like
to be.14
Other studies have stressed the importance of
identification in explaining a person's selection in
15
the use of mass media. Lazarsfeld claimed, "In all
"^Chapter II, page 46.
14Chapter II, page 48.
15Chapter II, pages 48-52.
158
the fields that have been touched by communications
research the self-selection of audiences plays a
16
considerable role."
Selection of television programs for viewing
seems to be related to the value structure that is
revealed in the personality of the viewer. In keeping
with one's own value structure, a person tends to move
toward those persons in his environment in whom he
perceives a value system similar to his own. Persons
tend to "like" most those persons with whom they
identify most strongly. According to Berkowitz and
Goranson, if a person likes another highly he will
probably tend to minimize differences between himself
17
and that person to whom he is attracted.
Studies reviewed show that much research has
been directed toward the study of identification in
the viewing of television. Such studies have drawn
attention to personality as an important element in
one's selection of programs for viewing.
This study differs from prior studies in that
effort was made to obtain a comprehensive quantitative
measurement of identification. Through use of a
*6Chapter II, page 51.
^Chapter II, page 54.
159
personality inventory questionnaire there was obtained
a personality profile for the viewer and one for his
perception of the television role. The problem was
to demonstrate the degree of similarity or dissimilarity
between these profiles.
Methods and Procedures:
Collection of Data
This was a descriptive study, the problem of
which was to show the kind and degree of relationships
that pertained between the personality profiles of
certain television viewers find the perceived person
ality profiles of certain television roles viewed by
them.
Starting with a sample of 1,755 couples, after
six separate mailings the final sample for the study
consisted of 206 persons, members of the Mariner program
of the United Presbyterian Church, U.S.A., who resided
in the greater Los Angeles area. It was not a random
sample, but because the purpose of the study was to
observe certain patterns of association between tele
vision viewers and the roles viewed by them it was
believed that any sample which was somewhat homogeneous
with respect to age, education, and standard of living
would serve to reveal such associations. This sample
met these criteria.
160
The sample of 206 persons was divided into
thirteen groups, each of which made judgments on a
single role. Seven groups provided judgments on most-
liked roles and six groups provided judgments on least-
liked roles.
Each subject was asked to keep a Television
Viewing Diary for three separate weeks with an interval
of one week between each viewing week. After all three
diaries were returned the viewers received a Role
Preference Questionnaire on which were listed all
regularly appearing roles in each television program
that was broadcast between 6:00 and 12:00 P.M. during
the three weeks of diary keeping. Subjects were asked j
to list in order of priority the ten roles which they
!
liked most, i.e., most liked, next most liked, etc. j
I
and the ten roles which they liked least, i.e., least |
liked, next least liked, etc.
Selections were made so that groups of approxi- j
i
mately ten or more were asked to judge the same role,
i
I
either as a least-liked role or as a most-liked role. j
Subjects were sent a copy of the Sixteen Per- j
1
sonality Factor Questionnaire and asked to check each
t
j
question as he believed a particular role (designated
j
on the front sheet of the questionnaire) might answer.
The resultant role personality profiles were taken as
that subject's perception of that role.
After the role personality inventory was returned
another 16 P. F. (the same form as was completed for
the role) was sent to the subject and he was instructed
to complete this one, giving his own responses to each
item.
Personal data on age, sex, education, and stand
ard of living were secured for each subject. Also,
each subject was asked to estimate the age, level of
education, and approximate standard of living for each
role assigned to him.
The following data were keypunched into IBM
cards and verified by a separate operator:
1. The 16 P. F. responses of each viewer, which
formed a personality profile of sixteen
factor scores.
2. The 16 P. F. responses of each role as
judged by a viewer, which formed a person
ality profile of sixteen factor scores.
3. Personal data on each viewer, including sex,
age, education, and annual combined family
income•
4. Personal data on each role as judged by a
viewer, including sex, age, education, and
standard of living.
162 |
Raw scores on each of the 16 P. F. factors were
obtained for each viewer and for each role as judged
by that viewer. Through use of an IBM Fortran IV
computer program, scores were weighted according to
Cattell's assigned weights.
Measurements of similarity or difference between j
personality profiles were secured through use of
Cattell's rp coefficient of pattern similarity
19
formula. The Tp is a distance measure, the index of
which is transformed to provide a scale similar to the
Pearson r. Statistical significance of the Tp correla
tions was determined through use of a table derived
by Horn.^
Data for sex, age, level of education, and
standard of living were coded and raw scores obtained !
j
for each.
i
Comparisons of similarity or difference between
i
16 P. F. profile data and between personal data were |
■ |
obtained on (1) viewer groups and most-liked role
groups, (2) viewer groups and least-liked role groups,
and (3) two groups of viewers, one of which rated
1A I
Chapter III, page 85. j
l Q
Chapter III, page 90. j
^Chapter III, page 89.
163 |
Granny as a most-liked role and the other of which
rated her as a least-liked role.
Measurements of similarity for age, education,
and standard of living were computed through use of
i
the Pearson r. A t test was used to determine whether
or not these correlations were significantly different
from zero.
Initially,* differences between personal data
were to be determined by use of Chi Square; however,
this was not feasible because too many cells contained
less than five frequencies. For this reason, propor
tions were used to report differences of personal data.
Reliability of findings. Throughout the study
effort was made to assure maximum reliability. The
final sample for the study contained 206 persons and
was homogeneous, thus lending stability to the findings.
In determining the most-liked roles that were regularly
viewed, three separate diaries were kept by each sub
ject. It was necessary for a role to be entered in at
least two of the three diaries to be included for study.
In determining which roles were most liked and which
were least liked, the subjects were assisted in narrow
ing their ranges of choice by being asked to check all |
roles that they particularly liked and those that they j
i
particularly disliked before checking the ten most-liked !
164 ;
roles and the ten least-liked roles. In the selection
of television roles to be Judged by groups of viewers,
effort was made to form groups of no fewer than ten !
viewers, but one group of viewers contained nine per
sons and one group contained six persons. In develop
ment of the 16 P. F., tests of consistency, equivalence,
and stability were used to assure an adequate reliabil
ity for test items. Form A of the 16 P. F. was used
as a measure of personality for both viewers and their
perceptions of roles.
All data that were processed by computer were
i
recomputed in part or in whole with the use of a desk
calculator to ascertain that the data printed out for
each stage of the program were computed accurately.
The reliability of the product moment correla- j
* i
tions obtained from the personal data was limited. j
i
i
Validity of findings. Steps were taken through
out the study to assure maximum validity. The data in
the Television Viewing Diaries were recorded in a
natural home setting. The Television Viewing Diary
was open-ended, providing a maximum freedom in choice
of programs. In determining most-liked roles and
least-liked roles, the viewer was asked to omit roles
with which he was unfamiliar or on which he had no
opinion, thus assuring a more accurate reflection of
165 |
his opinion on the roles. He was provided opportunity
to write in roles not appearing on the list. For a
most-liked role to be assigned to a viewer for judgment
the role must have been recorded in two of the three
Television Viewing Diaries kept by that subject. A
least-liked role need not have been viewed even once
because it was assumed that a viewer would not view
frequently a least-liked role. In the selection of
television roles to be judged by a viewer, effort was
made to use only first ranked most-liked or least-liked
I
roles.
In development of the 16 P. F., each of the
i
sixteen factors was validated against ratings in real-
I
life situations. The Form A of the 1957 edition of
the 16 P. F. was the result of three successive factor
analyses. In securing the 16 P. F. profiles the viewer |
completed his own questionnaire only after he had j
i
completed and returned a 16 P. F. role questionnaire.
This increased the probability that his familiarity
i
with the statements of the 16 P. F., Form A, would not
influence his perceptions of the role and insured a
greater validity for both his perception of the role j
and for his own profile.
Findings
Similarities and differences between
personality profiles of roles and of
their viewers.
The problem was to determine if there was not
greater similarity of personality between television
viewers and the perceived personality of specific
television roles most liked by them than there was
between the personality of television viewers and the
perceived personality of television roles least liked
by them.
The findings were affirmative. The correlation
of pattern similarity which served as an index of
similarity or difference between profiles provided
conclusive evidence that there was a greater similarity
between the perceived personality profiles of most-liked
roles and the personality profiles of their viewers
than there was between the perceived personality pro
files of least-liked roles and the personality profiles
of their viewers.
All seven of the comparisons that were made
between the personality profiles of viewer groups and
the perceived personality profiles of most-liked roles
correlated positively. One of the **p's was signifi
cantly different from zero at the .02 level of con-
167
fidence, another at the .05 level of confidence, and
one at almost the .10 level of confidence.
Conversely, it was found that five of six com
parisons that were made between the personality profiles
of viewer groups and the perceived personality profiles
of least-liked roles correlated negatively. One of
these Tp's was significantly different from zero at the
.01 level of confidence, another at the .05 level of
confidence, and one at almost the .10 level of confi
dence.
When comparisons were made between the person
ality profiles of viewers who rated Granny a most-liked
role and those who rated her a least-liked role a more
positive correlation was found between Granny, most
liked, and her viewers (.0384) than between Granny,
least liked, and her viewers (-.2615). In fact, the
correlation between Granny and those who rated her a
least-liked role was significantly different from zero
at almost the .05 level of confidence. Although the
personality profiles ot the two viewer groups, one
which liked Granny as a role and one which disliked
Granny, correlated negatively (-.1867), their per
ceptions of Granny as a role were similar, correlating
.2445, significant at almost the .10 level of confidence.
168
Similarities and differences between
viewer groups and roles with respect
to sex, age, education, and standard 1
of living.
Although this study was not designed primarily
to provide descriptive data on personal variables, but
rather to make comparisons between personality profiles,
it was felt that data on sex, age, level of education,
and standard of living might be of value in developing
a better understanding of those elements of a personal
nature which tend to be associated with a viewer's
selection of specific roles for viewing.
The proportion of males and females who selected
the various roles as either most liked or as least
liked did not vary appreciably from group to group, |
with the exception of Ben Casey. Most viewer groups
consisted of a proportion of approximately .6 females
and .4 males. Viewers of Casey consisted of a propor- |
i
tion of .4 females and .6 males.
Of all personal data variables, age provided
I
the largest number of positive correlations between j
i
roles and viewers. Among viewers of most-liked roles,
two of the six groups correlated positively with roles
at a o05 level of confidence. There was a high correla
tion on age between the roles of Cartwright and Novak
169
and their viewer group.
Novak also correlated rather highly (.439) with
his viewer group on level of education. Thirteen of
the fourteen viewers of Novak had completed one or more
years of college and ten of the fourteen viewers rated
Novak as a college graduate.
A high negative correlation was exhibited between
Ben Casey and his viewers on level of education (-.289).;
Five of the eleven viewers of Casey had completed only
a high school level of education.
Granny, most liked, appealed to a wider age j
range of viewers than did Granny, least liked. However,
i
there was a negative correlation between Granny and
both of her groups of viewers on age.
The viewers of Ben Cartwright showed the widest
range of level of education of all most-liked roles.
Viewers of Cartwright had levels of education that j
ranged from elementary school through college. All j
nine of the viewers of Danny Williams, a least-liked
role, had completed one or more years of college.
There were no significant correlations between
!
the standard of living of viewer groups and that per-
I
ceived for roles. The viewers of Andy Griffith
exhibited a broader range of standard of living than
any other viewer group. The viewers of Novak and
170
Cartwright also exhibited a broad range of standard
of living. '
Personality traits that distinguished
most clearly between most-liked roles
and least-liked roles, and between
roles and their viewers.
Factor Q3. This trait distinguished most con
sistently between most-liked roles and least-liked j
roles. All three of the most-liked roles were rated
positively by their viewers on this trait, while all
three least-liked roles were rated negatively by their
viewers on trait factor Q3. A positive factor Q3+
represents high self-concept formation, and a Q3-
stands for low integration of self. Kimble, Mason, j
i
and Cartwright, most-liked roles, received high positive !
i
scores on Q3, and Granny, Haskell, and Williams, least- i
I
liked roles, received high negative scores. |
It appeared that viewers liked most those roles |
I
which gave reinforcement to their inner resolutions— |
roles portraying the kind of person who strengthens
and furthers the goals of society, and did not like j
roles that portrayed irresponsible, unstable types of
persons.
171
Traits appearing consistently in most-liked
roles. Trait factors Q3+, strong self-concept, and
Q4-, relaxedness, appeared in all three of the most-
liked roles, which correlated significantly with their
viewer groups.
Trait Q4-, relaxedness, appeared to be an
important trait in most-liked roles. The same three
most-liked roles that exhibited trait Q3+, strong self-
concept, were outstanding also on trait Q4-.
Traits N+, shrewdness, and Q2+, self-sufficiency,
appeared in two of the four most-liked roles.
Traits appearing consistently in least-liked
roles. Trait factors F+, happy-go-lucky; L+, sus
picious; and Q3-, casual, careless of protocol,
appeared in all three least-liked roles, which corre
lated significantly with their viewer group.
All three least-liked roles were rated highly
on factor F+. This trait describes one who is happy-
go-lucky and enthusiastic. Perhaps this trait alone
could not be used to predict whether or not a role
would be liked or disliked by this sample group, but
when it appears in combination with other traits
certain conclusions might be drawn. For example, the
presence of trait factor F+, happy-go-lucky, plus
trait factor Q3-, low integration, casual, careless
172
of protocol, could indicate that this type of role
would be disliked by this sample group.
Factor L+, suspicious nature, was an outstanding
trait for least-liked roles only. For these viewers, ’
factor L+ appeared to be a highly disliked personality
trait.
Traits which appeared in two of the three least-
liked roles were C-, affected by feelings; E+, asser- j
s
tive; G-, expedient; H+, venturesome; I-, tough-minded; i
i
N-, forthright; and Q1-, conservative.
i
i
Traits on which roles and viewers were most
alike. All three most-liked roles and their viewers !
— ...........
rated strongly on factor Q3+ (self-control). Two of j
three roles also scored similarly with their viewer
group on factors A- (reservedness), E- (humility),
F- (sobriety), G+ (expediency), and L- (trustingness).
Two of three least-liked roles scored highly
with their viewers on trait factors 0+ (apprehensive-
i
ness) and Q2- (group dependency). It was believed
that persons who rated high on the anxiety factor 0+
would not like to view roles which exhibited this
[
trait, as this would only increase the viewer's j
anxieties.
173
Traits on which roles and viewers differed
most. There were many more differences between the
personality traits of least-liked roles and their
viewers than there were between most-liked roles and
their viewers.
Most-liked roles and their viewers differed
more than three sten scores only on factor Q2. The
roles of Mason and Kimble were seen as highly self-
sufficient and resourceful) while the scores of their
viewers fell within the normal range.
Least-liked roles and their viewers differed
on many factors. In all three cases viewers and least-
!
liked roles differed on factor E. The roles were rated ;
highly (assertive) while the viewers scored negatively
(humility). This appeared to be a significant trait |
in the determination of the viewers' dislike for these |
i
i
roles.
I
Other traits on which the viewers and least-
liked roles differed were factor A, on which the roles i
9 !
were rated positively (outgoing) and the viewers
negatively (reserved); factor C, on which the roles !
were rated negatively (emotionally less stable) and
the viewers positively (more stable); factor F, on
which the roles were rated positively (happy-go-lucky)
and the viewers negatively (sober); factor G, on which
174
the roles were rated negatively (expedient) and the
viewers positively (conscientious); factor H, on which
the roles were rated positively (venturesome) and the
viewers negatively (shy); factor L, on which the roles
were rated positively (suspicious) and the viewers
negatively (trusting); factor N, on which the roles
were rated negatively (forthright) and the viewers
positively (shrewd); and factor Q3, on which the roles
were rated negatively (casual) and theaYi©wers posi
tively (controlled).
Trait B (intelligence) distinguished most
clearly between the two groups of viewers, one of
which rated Granny a most-liked role and one of which
rated her a least-liked role. Those who rated her a
i
most-liked role appeared to exhibit slightly less
I
intelligence than those who rated her a least-liked
role.
II. CONCLUSIONS AND IMPLICATIONS j
!
Conclusions one and two are related to hypotheses !
21 i
one, two, and three. Conclusion three is supple- |
i
mentary and is not related to an initial hypothesis,
but to the findings in Chapter VI. If the assumptions
21Chapter I, pages 16-17.
175
22
made for this study were valid, especially with
regard to the data recorded in the Television Viewing
Diaries, in the Personal Information sheets, and in
the Television Role Preference Questionnaires, the
following conclusions can be made.
1. Adult persons tend to like most and to view
most frequently those television roles whose
personalities are perceived to be similar to s
the viewers' own personalities.
In all seven comparisons made between the mean
personality profiles of viewer groups and the perceived j
I
personality profiles of roles regularly viewed by them j
i
and rated as most-liked roles, there was a positive j
23
correlation. One correlation was significantly dif-
i
ferent from zero at the .02 level of confidence, !
1 i
another at the .05 level of confidence, and one at
almost the .10 level of confidence. The frequency
of viewing of most-liked roles was established by the j
requirement that a role must have been entered in at j
|
least two of the three Television Viewing Diaries to |
i
24 I
have been included for study as a most-liked role. j
22
Chapter I, pages 19-21. j
23Chapter IV, page 99.
24Chapter III, pages 70-71.
176
Conversely, it was concluded that adult persons
tend to like least and to view infrequently those tele
vision roles whose personalities are perceived to be
dissimilar to their own. j
In five of six comparisons that were made between :
the mean personality profiles of viewer groups and the
perceived personality profiles of roles rated by them
as least-liked roles and infrequently viewed by them,
there was a negative correlation.2® One correlation !
was significantly different from zero at the .01 level
of confidence, one at the .05 level of confidence, and
one at the .10 level of confidence. I
I
In addition, the personality profiles of viewers
and those of least-liked roles differed substantially |
on twenty-one traits, while the profiles of most-liked !
I
roles and their viewers differed greatly on only four ■
traits.2®
i
In the analysis of data for selection of roles j
t
to be judged, it was found that least-liked roles were j
seldom entered in more than one of the three Television
Viewing Diaries, and frequently were not listed as
having been viewed even once. Because it was believed
25Chapter IV, page 101.
2®Chapter VI, pages 140-141.
177
that viewers would view infrequently those roles that
they liked least, it was not required that a least-
liked role be viewed even once during the three weeks
27
of diary keeping to be included for study.
This first conclusion has obvious implications
for the director, producer, or sponsor of television
programs who wants to attract an audience that will
view his programs with regularity.
First, it means that the programer ought to be
aware of the personality traits that are perceived in
the role characterizations of his leading players. He
ought to know how these roles are being perceived by
whom.
Second, it means that he ought to become more
familiar with the personality characteristics of those
groups of viewers that he wishes to attract to his
programs.
From the results of this study it would appear
that group personality profiles are a most useful
holistic index with which to describe the nature of a
viewing group, the group's perceptions of certain
television roles, and the degree of similarity or
dissimilarity between them, and therefore the group's
27Chapter III, page 72.
178
attractions or lack of attractions to these programs.
Knowing this, and based upon the first conclusion of
this study, a programer should be able to construct
role characterizations that will attract specific groups
of viewers that will become regular viewers of his
program.
For example, if a producer of a television
program wants to structure a role which would appeal
to the viewers in the sample group of this study,
implications are that he should (1) include in the
role characterization those traits on which the most-
liked roles and their viewers revealed the greatest
similarity: Q3+, strong self-control; A-, reservedness;
E-, humility; F-, sobriety; G+, expediency; and L-,
trustingness; and (2) avoid those traits on which
least-liked roles and their viewers differed most:
E+, assertiveness; C-, less emotional stability; F+,
happy-go-luckyness; G-, expediency; H+, uninhibitedness;
I-, toughmindedness; L+, self-opinionatedness; N-,
I
artlessness, forthrightness; and Q3-, carelessness of j
protocol. |
i
2. Viewers who differ in their liking or dis
liking of a particular role will tend to
differ in their respective personality
characteristics.
179
Two groups of viewers viewed the role of Granny.
One group rated her a most-liked role and the other
rated her a least-liked role. The mean personality
profiles of these two viewer groups showed a mild
28
negative correlation of -.1867. This correlation
i
was not significantly different from zero, but when
contrasted to the significant, positive correlation
j i
(.2445) obtained between the perception of these two
groups on the role of Granny it lent mild support to j
the rejection of hypothesis three.
i
It was also concluded that viewers can differ
i
from each other in their liking or disliking of a
particular role and still have a similar perception
“ i
of the role.
|
One group of viewers rated Granny a most-liked
role and another rated her a least-liked role, yet
i
their perception of Granny as a role was quite similar, j
29
showing a positive correlation of .2445. This |
i
correlation was significantly different from zero j
i
at almost the .10 level of confidence. In other words, j
I
although the respective profiles of two viewer groups
I
differ, and although the viewers differ in their liking
2®Chapter IV, page 104. j
22Chapter IV, page 104.
180
of a particular role, it is possible for them to have
a perception of that role which is quite similar.
This conclusion provides further verification
of the fact that the liking or disliking of a role has
reference to basic personality traits of viewers as
well as to the personality traits of the role. In
this instance two groups of viewers perceived Granny
in a similar manner, but one group liked her and one
did not. The difference lay more between the mean
personality profiles of the two groups of viewers than
between their perceptions of the role of Granny.
This indicated that a broadcaster should make
every effort to be as aware of the personality char
acteristics of his potential audience as he is of the
role which he is creating for them to view. The dif
ferences in his prospective viewers' personality
characteristics will, according to these findings,
strongly affect their liking or disliking of any
particular television role.
3. Certain trait factors of personality, as
determined through use of the Sixteen
^Personality Factor Questionnaire, tend to
distinguish more clearly between most-liked
roles and least-liked roles than do others.
181 |
Of the sixteen trait factors in the 16 P 0 F.
inventory, trait factor Q3 distinguished most clearly
|
between most-liked roles and least-liked roles. All
most-liked roles that correlated significantly with j
their viewer groups exhibited a high positive score on
factor Q3. All least-liked roles that correlated j
...significantly with their viewer groups exhibited a |
30
high negative score on factor Q3. A positive Q3 j
rating indicates high self-concept control and a nega- j
tive rating indicates low self-integration. j
I
Trait factor Q4- (relaxedness) was a distinguish-j
ing trait factor of most-liked roles. All three most- j
i
liked roles exhibited this trait as an outstanding i
31 1
personality trait.
Trait factor F+ (happy-go-lucky) and trait
factor L+ (suspiciousness) were distinguishing factors
i
of least-liked roles. All three least-liked roles
32
exhibited this as an outstanding personality trait.
It appeared that viewers react more strongly
to certain personality characteristics than to others,
^Chapter VI, page 128.
^Chapter VI, page 132.
32Chapter VI, page 134.
182
and consequently score them more extremely either
toward the positive end of the scale or the negative
end. The factors listed above appeared to have caused
i
extreme reactions from the viewers.
It may be that only factors such as these need
be employed in a questionnaire to be used for pretesting
target audience response to certain television roles,
and hence, programs. If a short questionnaire could
be developed, one which included only those factors j
which distinguish most clearly between a viewer's liking
or disliking of a role, the time and cost of testing
could be minimized. This would be an important step
toward the development of a useful instrument with
which pilot television programs could be tested with
their intended target audience.
Although the evidence was not conclusive, there
was definite indication that some personality trait
factors seem to appeal to a broad group of viewers while
other trait factors appeal only to smaller groups within |
this larger audience. In other words, some traits seem }
to be common denominators that appeal to a broad range j
i
of persons, while other traits appeal only to specific
subgroups of viewers.
j
For example, trait factor Q3+ was an outstanding j
trait in all three most-liked roles and Q3- was an
183 |
33
outstanding trait in all three least-liked roles.
This trait appeared to be a common denominator trait
i
for this population.
However, although Van Dyke and Kimble were both
most-liked roles and both correlated rather highly
with their viewer groups (Kimble .2398, Van Dyke .2190),
they seemed to appeal to different groups.
Van Dyke was rated positively on factors A
(outgoing), F (happy-go-lucky), H (venturesome), and j
negatively on factor Q2 (a "joiner")*
Kimble was rated as being just the opposite.
He was rated negatively on factors A (reserved), F j
I
(sober), H (shy), and positively on factor Q2 (self- j
sufficiency). j
i
Factors A, F, H, and Q2 seemed to distinguish
between two subgroups within a broader group of viewers
that responded favorably to trait factor Q3. j
!
If the producer wanted to appeal to either of
these subgroups it would be necessary for him to build
into his leading roles specific kinds of traits. If
he wanted to appeal to a broad target audience, he
would then include in his role characterizations traits
that appealed to the total sample, and he would minimize
33Chapter VI, page 128.
184 |
use of traits that appealed to subgroups only.
In a recent article on television viewing it
was reported that "at a time when corporations are
tightening their advertising budgets, many sponsors are
seeking the same kind of 'selective' audiences that
34
they look for in magazines." According to this
report, "the trend seems to be toward more demographic
35
analysis and away from numbers alone." It appears
that the ability to determine specific subgroups
within the general population that might be attracted
to specific programs for regular viewing could be of
l
great value to broadcasters and sponsors who want to
expose their products to specific audiences at a
minimum of cost.
i
i
Summary Implication j
The findings of this study have made it possible j
i
to conclude that similarity of personality is an impor- ;
tant ingredient of television viewing which helps to
suggest the attractiveness of a television role to a
viewer and whether or not he is likely to view that
role with regularity. It was concluded that the
^"Honor Without Profit," Time, 90:23:81,
December 8, 1967.
55Ibid., p. 82.
185
personality of the viewer was as important to this
selection of roles for viewing as was the personality
of the television role. And finally, it was reported
that certain personality traits in the 16 P. F. tended
to distinguish more clearly between most-liked roles
and least-liked roles, and between viewers of these
roles, than did other traits.
All of the above conclusions seemed to indicate
that the broadcaster should give more attention to the
behavioral aspects of his potential viewers, with
special regard to their personality characteristics.
These findings underscore the fact that it is the
viewer who decides what he wants to view, and it is
the viewer whose perception of a role largely deter
mines whether or not he will view that television role
with any regularity.
This study provided quantitative support to
many previous studies which concluded, from logic and
surmise, that identification was an important element
in a viewer's selection of content of mass media. It
could be concluded that a viewer likes a role and is
attracted to it because his own personality and the
personality of the role are indeed similar. On the
other hand, it could be concluded that a viewer tends
to perceive the personality of the role as similar to
his own because he likes the role. Support for either
kind of conclusion awaits further study.
Now there remains the task of creating further
quantitative tests and indexes which can yield further
information on these concepts, measures which are easy
to apply with sample groups; relatively inexpensive
tests which will yield measurements of high reliability
and validity.
III. RECOMMENDATIONS FOR FURTHER STUDY
The following studies should be of value in
extending and improving the methods, techniques, and
findings of the present study.
1. Study is needed to construct a personality
questionnaire instrument that comprises only those
factors which distinguish most consistently between
most-liked roles and least-liked roles, and viewers of
these roles. In the present study certain trait factors,
such as Q3 (low integration versus high self-concept
control), have shown promise for this kind of pre
diction.
The sample for such a study should be random.
The problem would be to develop a questionnaire that
provides maximum prediction on most-liked and least-
liked roles, with use of a minimum number of items.
The short Form C version of the 16 P. F. might be
useful.
2. Study is needed for the development of test
items which, together with a personality inventory,
could be weighted and the results combined to provide
a single index which would indicate the probability of
the liking or disliking of a particular role by a
specific group of viewers.
Items to be considered, along with an index of
personality, might be age, sex, and standard of living.
Standard of living might comprise subscores on factors
such as education, income, occupation, place of resi-
!
dence, and group memberships in the community.
This battery of questions should be minimal in
i
length and difficulty, producing a single index that j
has maximum power to predict whether or not a particular j
l
group of viewers are apt to like or to dislike a par
ticular television role. j
3. The present study should be replicated with J
i
a larger and random sample of viewers. In future use j
of the 16 P. F. for pretesting the desirability of
television roles for specific viewer groups, it would
i
be helpful to know whether the same traits which seem
to distinguish clearly between most-liked roles and j
least-liked roles in this study continued to make clear
188
and consistent delineations with test scores produced
from a larger and more random sample.
The present study indicated that similarity of
personality is an important element in a person's
selection of roles for regular viewing. Use of a
random sample of viewers would help to indicate which
personality traits contribute most to this liking or
disliking of roles for the general population.
4. This study should be replicated using a
sample comprising both children and adults. It is
possible that children, in the viewing of television,
tend to seek models rather than mirrors. While in
this study it was found that adults tended to like
most those roles which appeared to reflect their own
values and behavior patterns, in their selection of
television programs children might be seeking adult
models after which they can pattern their ideas and
behavior in their effort to become accepted in the
adult world.
If this were found to be true, programers who
wanted to attract children to their programs would
structure television roles in a different way than
they are structuring for adult viewers.
5. Study is needed to determine whether or
not certain sets of personality traits tend to attract
subgroups of viewers within the general population
while other traits have universal appeal. The findings
of the present study implied that this was true.
It might be useful to continue use of the
16 P. F. but with a larger and random sample, to
discover which traits have universal appeal and which
subsets of traits tend to be associated with certain
roles and with certain groups of viewers.
Findings from such study would be of value to
the producer or the sponsor who desires to present a
program that will appeal to a specific part of the
viewing public, rather than to the general public.
|
6. In replication of this study it is suggested ;
that the following steps be taken to further insure the
|
validity of the findings: (a) Personal Information |
|
sheets should be enclosed with both the first and the |
final mailing so that a validity check can be made on
these data, and (b) it would be useful if "participant
observation" or other observation of diary keeping
could be established for a sample of the Television
Viewing Diaries to verify the validity of the entries.
BIBLIOGRAPHY
I
I
BIBLIOGRAPHY
A. BOOKS
Cantril, Hadley, and Gordon W. Allport. The Psychology
of Radio. New York: Harner and Brothers Publishers,
1935.
Cattell, Raymond B. The Scientific Analysis of Per
sonality. Baltimore: Penguin Books, Incorporated,
15657
, and Glen F. Stice. Handbook for the Sixteen
Personality Factor Questionnaire: hThe l6 P. F.
Test,” Forms A, B, and C. 1957 Edition, with 1962
Supplementation. Champaign, Illinois: The Institute
for Personality and Ability Testing, 1957.
Drever, James. A Dictionary of Psychology. London:
Penguin Books, Incorporated, 1953.
Glick, Ira 0., Sidney J. Levy, and others. Living with
Television. Chicago: Aldine Publishing Company,
1962.
Guilford, J. P. Fundamental Statistics in Psychology
and Education. Third edition. New York: McGraw-
Hill Book Company, Incorporated, 1956.
_______ . Personality. New York: McGraw-Hill Book
Company, Incorporated, 1959.
Handel, Leo A. Hollywood Looks at Its Audience: A
Report of Film Audience Research. Urbana: The
University of Illinois Press, 1950.
Harsh, Charles M., and H. G. Schrickel. Personality:
Development and Assessment. New York: The Ronald
Press Company, 1950.
Healy, William, Augusta F. Bronner, and Anna Mae Bowers.
The Structure and Meaning of Psychoanalysis: As
Related to Personality and Behavior. New York:
Alfred A. Knopf, Incorporated, 1938.
Murphy, Gardner. Personality: A Biosocial Approach
to Origins and StructureT New York: Harper and
Brothers Publishers, 1947.
192
Peatman, John G. Introduction to Applied Statistics.
New York: Harper and Row, Publishers, Incorporated,
1963.
Sherif, Muzafer, and Hadley Cantril. The Psychology
of Ego-Involvements: Social Attitudes and Identi
fications. New York: John Wiley and Sons,
Incorporated, 1947.
Symonds, Percival M. Dynamic Psychology. New York:
Appleton-Century-Crofts, Incorporated, 1949.
Tagiuri, Renato, and Luigi Petrullo (editors). Person
Perception and Interpersonal Behavior. Stanford:
Stanford University Press, 1956.
- Wyburn, G. M., R. W. Pickford, and R. J. Hirst. Human
Senses and Perception. Edited by G. M. Wyburn.
Toronto: University of Toronto Press, 1964.
B. PUBLICATIONS OF LEARNED SOCIETIES
Fearing, Franklin. "Social Impact of the Mass Media
of Communication," Mass Media and Education.
Edited by Nelson B. Henry. The Fifty-Third
Yearbook of the National Society for the Study
of Education, Part III Chicago: The Society,
1954. Pp. 165-191.
IPAT 16 P. F. Form A, 1962 Edition. The Institute
for Personality and Ability Testing, 1602-04
Coronado Drive, Champaign, Illinois.
i
SoComp: A Series of Automatic Statistical Programs.
University of Southern California: Computer
Sciences Laboratory, 1010 West Jefferson Boulevard,
Los Angeles, California 90007, January, 1967.
C. PERIODICALS
Allport, Gordon W., and Henry S. Odbert. Trait-Names:
A Psycho-lexical Study. Psychological Monographs.
Edited by Joseph Peterson and others,47:1 (Whole
No. 211):1-171, 1936.
193
Barker, Warren J. "The Stereotyped Western Story: Its
Latent Meaning and Psychoeconomic Function," The
Psychoanalytic Quarterly, 24:2:270-280, 1955.
1
Bauer, Raymond A. "The Initiative of the Audience,"
Journal of Advertising Research, 3:2:2-7, June,
1963.
Berkowitz, Leonard, and Richard E. Goranson. "Motiva
tional and Judgmental Determinants of Social
Perception," Journal of Abnormal and Social
Psychology, 69:3:296-302, September, 1964.
Beville, H. M., Jr. "The ABCD's of Radio Audiences,"
The Public Opinion Quarterly. 4:2:195-206, June,
1940.
Cantril, Hadley. "A Comparative Study.of Radio and
Face-to-Face Stimulus Situations," The Journal of
Social Psychology, 8:4:443-458, November, 1937.
Cattell, Raymond B. "Oblique, Second Order, and
Cooperative Factors in Personality Analysis,"
The Journal of General Psychology, 36:3-22, 1947.
. "rp and Other Coefficients of Pattern
Similarity," Psychometrika, 14:4:279-298, December,
1949.
_____ . "Validation and Intensification of the
Sixteen Personality Factor Questionnaire," Journal
of Clinical Psychology, 12:3:205-214, July, 1956.
_______. "The Main Personality Factors in Question
naire, Self-Estimate Material," The Journal of
Social Psychology. 31:First Half:3-38, February,
1950.
Cressey, Paul G. "The Motion Picture Experience as
Modified by Social Background and Personality,"
American Sociological Review, 3:4:516-525, August, |
1938.
i
Cronbach, Lee J., and Goldine C. Gleser. "Assessing
Similarity between Profiles," Psychological
Bulletin. 50:6:456-473, November, 1953.
Dymond, Rosalind F. "Personality and Empathy," Journal
o* Consulting Psychology. 14:5:343-350, October,
1950.
194
Emery, F. E. "Psychological Effects of the Western
Film: A Study in Television Viewing: I. The
Theoretical Study: Working Hypotheses on the
Psychology of Television," Human Relations,
12:3:195-213, 1959.
_______. "Psychological Effects of the Western Film:
A Study in Television Viewing: II. The Experimental
Study," Human Relations, 12:3:215-232, 1959.
Fearing, Franklin. "Influence of the Movies on Atti
tudes and Behavior," The Motion Picture Industry.
Edited by Gordon S. Watkins. The Annals~~of the
American Academy of Political and Social Science,
£54:70-79, November, 1947.
Fensterheim, Herbert, and M. E. Tresselt. "The
Influence of Value Systems on the Perception of
People," The Journal of Abnormal and Social
Psychology. 48:1:93-957 January, 1953.
Foulds, Graham. "The Child's Response to Fictional
Characters and Its Relationship to Personality
Traits," Character and Personality 11:64-75,
September, 1942-June, 1943.
Freidson, Eliot. "Communications Research and the
Concept of the Mass," American Sociological Review,
18:3:313-317, June, 19357
Gage, N. L., and Lee J. Cronbach. "Conceptual and
Methodological Problems in Interpersonal Per
ception," The Psychological Review, 62:6:411-422,
November, 1955.
Helmstadter, Gerald C. "An Empirical Comparison of
Methods for Estimating Profile Similarity,"
Educational and Psychological Measurement.
17:1:71-62, Spring, 1957.
Herzog, Herta. "On Borrowed Experience: An Analysis
of Listening to Daytime Sketches," Studies in
Philosophy and Social Science, 9:1:65-95,1941.
Himmelweit, Hilde T. "A Theoretical Framework for
the Consideration of the Effects of Television:
A British Report," The Uses of Television. Edited
by Lotte Bailyn. The Journal of Social Issues,
18:2:16-28, 1962.
195
"Honor Without Profit," Time, 90:23:81-82, December 8,
1967.
Horn, J. L. "Significance Tests for Use with rp and
Related Profile Statistics," Educational ana
Psychological Measurement. 21:2:363-37b, Summer,
1961.
Horton, Donald, and R. Richard Wohl. "Mass Communica
tion and Para-Social Interaction: Observations on
Intimacy at a Distance," Psychiatry, 19:3:215-229,
August, 1956.
Katz, Elihu, and David Foulkes. "On the Use of the
Mass Media as 'Escape': Clarification of a Concept,"
The Public Opinion Quarterly, 26:3:377-388, Fall,
1962.
i •
Kelman, Herbert C., and Alice H. Eagly. "Attitude
Toward the Communicator, Perception of Communica
tion Content, and Attitude Change," Journal of
Personality and Social Psychology, 1:1:63-78,
January, 1965. |
Klapper, Joseph T. ""Mass Communication Research: An
Old Road Resurveyed," The Public Opinion Quarterly,
27:4:515-527, Winter, 1963.
Lichty, Lawrence W. "'The Real McCoys' and It's [sic]
Audience: A Functional Analysis," Journal of
Broadcasting, 9:2:157-166, Spring, 1965.
Murdoch, Lawrence C., Jr., and others. "Mass Media
and the Gulliver Fallacy," Business Review,
October, 1962,. published by Federal Reserve Bank
of Philadelphia, as reported in "Is the Mass
Audience a Thing of the Past?" Broadcasting,
64:2:42-45, January 14, 1963.
Newcomb, Theodore M. "The Prediction of Interpersonal
Attraction," The American Psychologist, 11:11:575-
586, November, 1956.
Pearlin, Leonard I. "Social and Personal Stress and
Escape Television Viewing," The Public Opinion
Quarterly, 23:2:255-259, Summer, 1959.
196
Postman, Leo, and Jerome S. Bruner. "Perception under
Stress," The Psychological Review, 55:6:314-323,
November, 1948.
Riley, Matilda White, and Samuel H. Flowerman. "Group
Relations as a Variable in Communications Research,"
American Sociological Review, 16:2:174-180, April,
1951.
Rosenzweig, Saul. "Available Methods for Studying
Personality," The Journal of Psychology, 28:Second
Half:345-368, October, 1949.
Runciman, Alexander P. "A Stratification Study of
Television Programs," Sociology and Social Research.
44:4:257-261, March-April, 1960.
Schneiderman, Leo. "Social Perception as a Function
of Identification," The Journal of Psychology.
37:155-162, 1954.
Sherif, Muzafer. A Study of Some Social Factors in
Perception. Archives of Psychology. Edited by
R. S. Woodworth, 27:187:5-60, July, 1935.
_______ , and S. Stansfeld Sargent. "Ego-Involvement
and the Mass Media," Mass Media: Content. Function.
and Measurement. Edited by Franklin Fearing. The
Journal of Social Issues. 3:3:8-16, Summer, 1947.
Shrauger, Sid, and John Altrocchi. "The Personality of
the Perceiver as a Factor in Person Perception,"
Psychological Bulletin, 62:5:289-308, November,
1964.
i
TV Guide, 12:11, March 14, 1964; 12:12, March 21, 1964;
“T27I3, March 28, 1964; 12:14, April 4, 1964; 12:15, i
April 11, 1964; 12:16, April 18, 1964.
Wright, Charles R. "Functional Analysis and Mass j
Communication," The Public Opinion Quarterly i
24:4:605-620, Winter, 1960.
197
D. ARTICLES IN COLLECTIONS
Arnheim, Rudolf. "The World of the Daytime Serial,"
in Radio Research. 1942-1945. Edited by Paul F.
Lazarsfeld and Frank N. Stanton. New York:
Duell, Sloan and Pearce, 1944. Pp. 34-85.
Bruner, Jerome S. "Personality Dynamics and the
Process of Perceiving," in Perception: An Approach
to Personality. By Robert R. Blake, Glenn V.
Ramsey, and others. New York: The Ronald Press
Company, 1951. Pp. 121-147.
Lazarsfeld, Paul F. "Communication Research and the
Social Psychologist," in Current Trends in Social
Psychology. By Wayne Dennis and others.
Pittsburgh: University of Pittsburgh Press,
1948. Pp. 218-273.
Riley, John W. , Jr. , and Matilda White Riley. "Mass
Communication and the Social System," in Sociology
Today: Problems and Prospects. Edited by Robert
K. Merton, Leonard Broom, and Leonard S. Cottrell,
Jr. under the auspices of the American Sociological
Society. New York: Basic Books, Incorporated,
1959. Pp. 537-578.
Stanton, Frank N. "Psychological Research in the j
Field of Radio Listening," in Educational Broad
casting, 1956: Proceedings of the First National j
Conference on Educational Broadcasting, Held in
Washington, D.C., on December 10, 11, and 12,
1956. Edited by C. S. Marsh. Chicago: The
University of Chicago Press, 1937. Pp. 365-377.
i
E. UNPUBLISHED MATERIALS
Pearlin, Leonard Irving. "The Social and Psychological
Setting of Communications Behavior: An Analysis
of Television Viewing." Unpublished Doctoral dis
sertation, Columbia University, New York, 1956.
Wilner, Daniel M. "Attitude as a Determinant of
Perception in the Mass Media of Communication
Reactions to the Motion Picture, Home of the
Brave." Unpublished Doctoral dissertation,
University of California at Los Angeles,
October, 1950.
APPENDIX A
PERSONAL INFORMATION SHEET
APPENDIX A
200
PERSONAL INFORMATION
(Note: No personal information will be reported except
as it is combined with all other subjects and
reported in summary form.)
PLEASE FILL IN ALL BLANKS. THANK YOU.
1. NAME: Mr. and Mrs._________________________________
ADDRESS: ________
2. HUSBAND: Occupation? Age?
Circle last year of scnool completed:
Elementary 1 2 3 4 5 6 7 8
High School 1 2 3 4
College 1 2 3 4 (more)
3. WIFE: Occupation? Age?
Circle last year of school completed:
Elementary 1 2 3 4 5 6 7 8
High School 1 2 3 4
College 1 2 3 4 (more)
4. BOTH: Family annual combined income: (check one)
Under $4,000 [ ]
4.000 to 5,999 C 3
6.000 to 7,999 .[ 3
8.000 to 9,999 C 3
10,000 and over C 3
How many children have you living at home?
How many: Under 5
5 to 14
15 to 19__________
20 and over__________
How many television sets do you have
in your home?________________
How many are in working order?
APPENDIX B
TELEVISION ROLE PERSONALITY QUESTIONNAIRE
(COVER SHEET)
202
APPENDIX B
TELEVISION ROLE PERSONALITY QUESTIONNAIRE
(COVER SHEET)
WHAT TO DO:
1. You are being asked to answer these questions as
you think ______________ would answer them.
Select the answers that you think he (she) would
select, not as you might answer them for yourself.
2. Please record your answers on the answer sheet on
the back cover, not on the questionnaire. Check
the appropriate space opposite the number of the
question.
3. Use the "in between" answers only when absolutely
necessary. Try to select an answer either to the
left or to the right.
4. Please answer every question even if some are
difficult to answer or don't seem to apply.
5. Please check the following answers first:
a. Approximately, when did you last view this
role? Month______ Day______.
b. How old do you think __________ is? _________
c. What do you think his (her) standard of living
is (economic status) considering his (her)
relationship to other people in the show?
(Check one)
Very low Rather low Average Rather high
Very high
d. What do you think is the highest grade he (she)
completed in school? _______________
203
NOW ANSWER THE FOLLOWING EXAMPLES FOR PRACTICE AND
BEGIN ANSWERING THE REST OF THE QUESTIONS BEGINNING
ON THE NEXT PAGE. (Remember to place your answers
on the answer sheet on the back page.)
EXAMPLES:
1. I like to watch team games, (a) yes, (b) occa
sionally, (c) no.
(That is, I think that would say that
he (she) likes to watch team games. (a) yes,
(b) occasionally, (c) no.
2. I prefer people who: (a) are reserved, (b) (are)
in between, (c) make friends quickly.
(That is, I think that ____________ would say that
he (she) prefers people who: (a) are reserved,
(b) (are) in between, (c) make friends quickly.
UPON COMPLETION OF ALL QUESTIONS PLEASE RETURN BOTH
THE QUESTIONS AND THE ANSWER SHEETS AS SOON AS POSSIBLE.
THANK YOU.
APPENDIX C
TELEVISION VIEWING DIARY
APPENDIX C
205
TELEVISION VIEWING DIARY
MRS
Name of Viewer
March 30, 1964
Start viewing on this date
WHAT TO DO: 1.
2.
3.
TIME
4.
5.
6.
7.
Husband and wife each keep individual
viewing diaries.
Please avoid watching television either
more or less than you normally do.
Record only viewing between 6 PM and
12 midnight.
Begin your diary recording on Monday
and conclude on Sunday.
Use a separate sheet for each day of
the viewing week. (Starting date
above.)
Please mail the diary to me on the
Monday following completion. Use the
enclosed stamped return envelope.
Please record your program viewing as
soon as possible after the program is
over, noting, in the far right-hand
column, the time and date you recorded
your viewing.
THANK YOU
STATION
When was this
viewing recorded
From To Channel NAME OF PROGRAM Time Date
(Example)
6 6:30 4 Evening News 6:30 3/24
HAVE YOU FORGOTTEN TO RECORD ANY VIEWING? USE OTHER
SIDE IF NECESSARY. START TOMORROW'S VIEWING ON THE
NEXT PAGE, PLEASE.
206
(The original diary included one page for each day of
* TELEVISION VIEWING DIARY
Ufhon l i f f l f i fh*i a
TIME STATION viewing recorded
From To Channel NAME OF PROGRAM Time Date
HAVE YOU FORGOTTEN TO RECORD ANY VIEWING? USE OTHER
SIDE IF NECESSARY. START TOMORROW'S VIEWING ON THE
NEXT PAGE, PLEASE.
APPENDIX D
TELEVISION ROLE PREFERENCE QUESTIONNAIRE
208
T E L E V I S I O N ’ R O L E P R E F E R E N C E Q U E S T I O N N A I R E
/AIL
(liusbard and wife EACH fill out a SEPARATE QUESTIOEHAIEE, plscsej
W H A T T O D O : E e l c w a r e l i s t e d a n u m b e r o f p r o g r a m s p r e s e n t e d o n L o s A n g e l e s T e l e v i s i o n
S t a t i o n s . U n d e r e a c h p r o g r a m t i t l e a r e l i s t e d s e m e o f t h e r o l e s , o r c h a r a c t e r s ,
r e g u l a r l y a p p e a r i n g o n t h a t p r o g r a m . S c a n d o w n t h e l i s t .
1 . . I f y o u p a r t i c u l a r l y LIKE a c e r t a i n r o l e , p l a c e a c h e c k m a r k i n C o l t e r . A .
2 . I f y o u p a r t i c u l a r l y DISLIKE a r o l e , c h e c k C o l u m n B .
3 . I f y o u n e i t h e r l i k e n o r d i s l i k e t h e r o l e , o r a r e u n f a m i l i a r w i t h i t ,
j u s t l e a v e t h e l i n e b l a n k .
4 . I f t h e r o l e t h a t y o u w a n t t o c h e c k i s n o t l i s t e d , w r i t e i t i n t h e c o l u m n
j u s t l e f t o f t h e p r o g r a m t i t l e .
A_
Roles
Best
Liked
3
Holes
Least
Liked
- ! jl
Roles\ House j
Best j Least j
! i
i A b b o t £ C o s t e l l o
* * *
B i l l D a n a S h o w
- . V i ’ . i ' c j i * : I ’ . * ; ' ;
i A b b o t J o s e J i m e n e z i
1 C o s t e l l o H o t e l M a n a g e r
1
t
| A d v e n t u r e s i n P a r a d i s e
A A A
F e l l o w B e l l H o d
i
I
I
B o n a n z a
i
C l a y B a k e r B a n C a r t w r i g h t !
i A n d v G r i f f i t h S h o w
A A A
A d a m C a r t w r i g h t i "
; A n d y T a y l o r H o s s C a r t w r i g h t i
i . B a r n e y F i f e L i t t l e J o e C a r t w r i g h t i
1 O p i e B r e a k i n g P o i n t
A A A | A A A
j A u n t E e e D r . M a c T h o m p s o n
1 G o m e r P y l e D r . W i l l R a y m e r
A n n S o u t h e r n S h o w
A s h ' t
B u r k e ' s L a w
|
K a t y O ' C o o n a r A m o s B u r k e . . . . i
J a s o n M a c a u l e v T i m T i l s o n
I
I O l i v e S m i t h S g t . H a r t 1
i | F l o r a M a c a u l a y C a l l M r . D . * * * 1 * * *
I l A r r e s t a n d T r i a l
. ' . . ' . A
A A A M r . D i a m o n d 1
1 I S g t . A n d e r s o n C h e c k m a t e A A A ! A A A
1 1 J o n n E g a n D o n C o r e y i
: j B e n C a s e y i t i t i t A A A J e d S i l l s :
j 1 D r . B e n C a s e v D r . C a r l H y a t t 1
' | D r . Z o r b a i C h e v e n n e A A A 1 A A A
j i D r . G r a h a m 1 C h e y e n n e B o d i e
! ' N i c k I C c m b a t A A A
! | N u r s e W i l l s L t . H a n l e v
• | D r . H o f f m a n S g t . S a u n d e r s
1
i I f i e v e r l y H i l l b i l l i e s isit A K i r b v
1 i J e d C l a m p e t t ' C a g e
1 G r a n n y L i t t l e i o h n
!
!- - - - - - - E l l y M a y D a n n y T h c m a s S h e w A A A 1 A A A
J e t h r o D a n n v W i l l i a m s
C o u s i n P e a r l K a t h y W i l l i a m s
1
M r . D r y s d a l e R u s t y
M i s s H a t h a w a y L i n d a
M r s . D r y s d a l e D e f e n d e r s A A A j , A A A
L a w r e n c e P r e s t o n
!
K e n n e t h P r e s t o n
!
209
1 A_
Roles
Most
Liked
B_
Roles
Least
Liked
A j 2'
Roles-. Roles j
Most 1 Least ■
Liked j Liked i
T h e D e p u t y
* * *
i t i t i t
G a l l a n t M e n
. . « • . . .
ititit : ‘ . r . r . : ■
M a r s h a l l F r v I
C o n l e y W r i g h t
I
j D e p u t y M c C o r d j
C a p t a i n B e n e d i c t i
! i ^ e s t r y i t i t i t H i t - : :
P r i v a t e D 1 A n g e l o
l ;
D e s t r v
G r i n d l
D i c k V a n D y k e S h o w i t i t i t i t i t i t
G r i n c i l i
R o b P e t r i e
A n s o n F o s t e r i
S a l l y R o g e r s
G u n s m o k e
i t i t i t • i t i t i t ' <
u u d a y S o r r e l l
M a r s h a l l D i l l c n
. . . . ,
i ;
L a u r a P e t r i e
C h e s t e r
i ;
i
R i t c h i e P e t r i e
D o c A d a m
i
M e l C o o l e y
K i t t y
! i
D o n n a R e e d S h e w
i t i t i t ‘ . ' i i t
Q u i n t A s o e r 1 !
_ D o a n a S t o n e
H a w a i i a n E v e
A l e x S t o n e
C r i c k e t B l a k e i
D a v e K e l s e y
T o n L o n a k a i i
L M i d g e K e l s e y
T r a c v S t e e l e
! i
J e f f
G r e g M a c k e n z i e
• i !
T r i c h a
P h i l i p B a r t o n
i
D r . K i l d a r e
i t i t i t f t * *
K i m
!
D r . K i l d a r e
H a v e G u n . W i l l T r a v e l
i t i t i t | i t i t i t 1
D r . G i l l e s D i e
P a l a d i n
E a s t S i d e , W e s t S i d e
i t i t i t i t i t i t
H a z e l
i t i t i t i t i t i t :
N e i l B r o c k
H a z e l
-
F r i e d a H e c h l i n g e r G e o r g e B a x t e r
J a n e F o s t e r I D o r o t h v B a x t e r ! 1
8 7 t h P r e c i n c t
i t i t f t i t i t i t
H a r o l d
i :
S t e v e C a r e l l a J o e y B i s h o p S h o w
i t i t i t i i t i t i t i
R o g e r H a v i l l a n d
J o e v B a r n e s I
E e r t K L i n g
E i l l i e B a m e s i ;
M e v e r M e v e r
L a r r v
T e d d y C a r e l l a H i l d a
E l e v e n t h H o u r
i t i t i t i t i t i t
L a r a m i e
i t i t i t | i t i t i t '
D r . S t a r k e
J e s s H a r p e r
I
D r . G r a h a m
S l i m S h e r m a n
E m p i r e i t i t i t i t i t i t
D a i s y C o o p a r
I
J i m R e d i g o
■ M i k e W i l l i a m s
i
C o n n i e G a r r e t
L a w m a n
i t i t i t i t i s i t j
L u c i a G a r r e t t D a n T r o o p
T a l G a r r e t
J o h n n v M c K a v
•
E n s i g n O ' T o o l e
i t i t i t i t i i i t
L i b v
E n s i g n O ' T o o l e L a s s i e
i t i t i t 1 i t i t i t
C o m m a n d e r S t o n e r T i m m v
;
C h i e f N e l s o n 1 R u t h M a r t i n
L t . S t . J o h n P a u l M a r t i n
i
F a r m e r ' s D a u g h t e r
i t i t i t i t i i i t
L e a v e i t t o E e a v e r
i t i t i t * * * ]
K a t y H o l s t r u m W a l l v
!
C o n g r e s s m a n M o r e l y B e a v e r
1
A g a t h a M o r e l y W a r d C l e a v e r
I
S t e v e J u n e C l e a v e r
I
D a n n y E d d i e H a s k e l l
I
' " T h e F u g i t i v e
i t i t i t i t i t i t
T h e L i e u t e n a n t
i t i t i t i t i t i t \
D r . K i n i b l e L t . B i l l R i c e
I
L t . G e r a r d C a o t . R a v R a m a e e i
1 L u k e | L u c y S n o w
i t i t i t i t i t i t
' . ' o v 1
210
! ■
i
A
Roles
Rent
Liked
_ §
Roles
Leant
Liked
Roles
Bent
Liked
5
Roles
L * .
Liked
i
1 ! M a v e r i c k
i t i t i t i t i t i t
P a t t v D u k e S h e w
iV.'ii
1 B r e t M a v e r i c k
P a t t v L a n e
i
j B a r t M a v e r i c k
K a t h v L a n e
i *
i j K e a i c i t i i i t AAA
R i c h a r d
i
i ; D r . M a x S t y n e r
M a r t i n
; ;
; K c H a l e 1 s N a v y
i ' i i t AAA
N a t a l i e
1
! ' C u i n t M c H a l e
R o s s
! C a p t . B i n g h a m t o n
P e r r y M a s o n j
f t f t f t j
i i t n s . P a r k e r
P e r r y M a s o n
; ! u r u f c e r
D e l l a S t r e e t
■ F u i x
P a u l D r a k e
i
! i M i k e h a m m e r i t i t i t A A A
H a m i l t o n B u r g e r
i
1 M a k e H a m m e r
L t . T r a p p
. . . . . j
! M r . N o v a k AAA AAA
L t . A n d e r s o n
i
j o n n N o v a k
P e t e r G u n n i
, A l b e r t V a r . e
P e t e r G u n n
• 1
: J e a n P s g a r o
L t . J a c o b y
i i M a s t e r D a A * i'i i t i t i t
E d i e
i
1 I W i l b e r H o s t
P e t t i c o a t J u n c t i o n
ftiWi
.J . A
C a r o l p e s t
K a t e B r a d l e y 1
K o g e r A d d i s o n
U n c l e J o i
K a y A d d i s o n
B i l l i e J o
1
. M y F a v o r i t e M a r t i a n
A iVA" ' A A A
E o b b i e J o
I !
! ! U n c l e M a r t i n
^ B a t t v J o
; i T i m O ' H a r o
T h e R e b e l
AAA
; ! M y T h r e e S o n s AAA AAA
J i m B u c k l e y ............ ' i
; S t e v e D o u g l a s
J o h n n v Y u m a !
B u b O ' C a s e y
R i o c o r d
AAA AAA
M i k e
T e d M c K e e v e r
R o b b i e
R o u t e 6 6 AAA
i
! C h i n
T o d S t i l e s
1
!
1 I S a l l y M o r r i s o n
L i n e C a s e
;
■ i T n i e
M r s . B a r b e n
i N a k e d C i t y AAA AAA
S a m B e n e d i c t
AAA
A d a m F l i n t
S a m B e n e d i c t 1 !
L t . P a r k e r
' H a n F T a v l o r 1
1 A r c a r o
S t a g e c o a c h W e s t
a A n
j.,1, |
*. . . i
I M e w B r e e d »1tAA AAA
S i m o n K a n e
j
| i m i c e A d a m s
L u k e P e r r v I
| ! V i n c e C a v e l l i
D a v i d K a n e
i
i
! i J o e H u d d l e s t o n
S t o n e v B u r k e
AAA AAA |
1 i P e t e G a r c i a
S t o n e B u r k e
1
! 1 C a n t . G r e g o r y
S u p a r f c o t
AAA
;---------------
N e w P h i l S i l v e r s S h e w
AAA A A A
T o m B r e w s t e r
h a r r y G r a f t o n !
| 7 7 S u n s e t S t r i p
AA A V .'v ';:’ ;
i i ' i ' r . e N u r s e s
AAA AAA
! S t u a r t B a i l e v
1 i L i s T h o r p e
K b o k i
1 ' G a i l L u c a s
1 J e f f S p e n c e r t
j : 0 u r M i s s B r o o k s
AAA AAA
! S u r f s i d e 6
A A A
1
1 i M i s s B r o c k s
i S a n d y W i n d f i e l d I I
: M r . C o n k l i n K e n M a d i s o n 1
' O z z i e a n d H a r r i e t
AAA AAA
D a p h n e D u t t o n 1 I
i O z z i e N e l s o n T a r g e t t h e C o r r u p t o r s 1
. . . . —
h a r r i e t N e l s o n P a u l M a r i n o l . i i
r - . v s r J
1 .T a c k F l o o d 1 '
211
\
1
1
1
1
Roles
Best
Liked
B
Roles
Least
Liked
A
Roles
Best
Liked
Roles j
Least !
Liked ;
; (
, T a - e i e H o u s t o n
V J a g o n T r a i n
it Mi
1
*
i ? e . T ; D l e H o u s t o n
C h r i s H a l e
u e c r g e T a g g a r t \
C o o p e r S m i t h
. - r . a u n t o u c h a b l e s i V . ' f j V j ’ n V . V C h a r l i e W e b s t e r
i
I ! R o b e r t S t a c k
B i l l H a w k s
. . . . i
L e e H o b s o n
E a m a b y W e s t
i
t - n r i c o } < o s s i
D u k e S h a n n o n
i
; : T h e V i r g i n i a n
i V v ' n * ;
W i d e - C o u n t r y
it it ft t ’l t ' i i ' ; I
} ! J u d g e G a r t h I
M i t c h G u t h r i e i
l T h e V i r g i n i a n j
A n d y G u t h r i e 1
. T r a m o a s j
I i S t e v e j
• i
B e t s y G a r t h | !
i N ' p T Z : IN LISTING THE FOLLOWING POLES IN ORDER,, YOU MAI WANT TO USE A PENCILJ
IN CASE you CHANGE YOUR MIND AS YOU GO ALONG.
F r o m t h e r o l e s y o u c h e c k e d i n C o l u m n
_ A , p i c k t e n t h a t y o u l i k e BEST
a n d l i s t t h e m b e l c w i n o r d e r , b e
g i n n i n g w i t h ( 1 ) m o s t l i k e d , ( 2 )
n e x t m o s t l i k e d , e t c .
PLEASE FILL IN ALL SPACES
MOST LIKED ROLES
Proaram
1.
2.
3.
A
S.
-
.......
7.
3.
9.
10.
F r o m t h e r o l e s y o u c h e c k e d i n C o l u m n j
B , p i c k t e n t h a t y o u l i k e LEAST
and l i s t t h e m b e l c w i n o r d e r , b e - !
g i n n i n g w i t h ( 1 ) l e a s t l i k e d , ( 2 ) ]
n e x t l e a s t l i k e d , e t c . 1
PLEASE FILL IN ALL SPACES !
LEAST LIKED ROLES
1
Proarcm ;
■ !
1. .
1
1
. !
2. i
3.
i
l
s
4.
!
S.
i
i
i
6.
!
j
i
7.
«
8.
j
9.
•
i
10.
APPENDIX E
SIXTEEN PERSONALITY FACTOR TEST PROFILES
(ROLES AND VIEWERS)
16 P.F. TEST PROFILE
FACTOR
Stan
dard
Scon
LOW SCORE
DESCRIPTION
A
RESERVED, DETACHED, CRITICAL, COOL
(Sizothymia)
B
LESS INTELLIGENT, CONCRETE-THINKING
(Low er scholastic m ental capacity)
C
AFFECTED BY FEELINGS, EMOTIONALLY
LESS STABLE, EASILY UPSET
(Low er e g o strength)
E
HUMBLE, MILD, OBEDIENT, CONFORMING
(Subm tssiveness)
F
SOBER, PRUDENT, SERIOUS, TACITURN
(D esurgency)
G
EXPEDIENT, A LAW TO HIMSELF,
BY-PASSES OBLIGATIONS
(W eaker su p e re g o strength)
H
SHY, RESTRAINED, DIFFIDENT, TIMID
(Threctia)
I
TOUGH-MINDED, SELF-RELIANT,
REALISTIC, NO-NONSENSE
(Harria)
L
TRUSTING, ADAPTABLE, FREE OF JEALOUSY,
EASY TO GET ON WITH
(Alaxia)
M
PRACTICAL CAREFUL, CONVENTIONAL,
REGULATED BY EXTERNAL REALITIES,
PROPER lPr»xem l§)
N
FORTHRIGHT, NATURAL, ARTLESS,
SENTIMENTAL
(A rtlessness)
0
PLACID. SELF-ASSURED, CONFIDENT, SERENE
(U ntroubled adequacy)
Q .
CONSERVATIVE, RESPECTING ESTABLISHED
IDEAS, TOLERANT OF TRADITIONAL
DIFFICULTIES (C onservatism )
Q .
CROUP-DEPENDENT, A "JO IN ER" AND
SOUND FOLLOWER
(G roup adherence)
Q ,
CASUAL CARELESS OF PROTOCOL, UNTIDY,
FOLLOWS OW N URGES
(le w integration)
Q ,
RELAXED, TRANQUIL, TORPID,
UNFRUSTRATCD
(Low crg ic tension)
STANDARD TEN SCORE (STEN)
->AveraQe^»
O 1963. Institute for Personality
and Ability Testing.
A ttc n o f 1 3 3 4 5 6 7
b y a b o u t 2.3% 4.4% 9.2% 15.0% 19.1% 19.1% 15.03
HIGH SCORE
DESCRIPTION
OUTGOING, WARMHEARTED, EASY-GOING,
PARTICIPATING
(Cyclothymia)
MORE INTEUIGENT, ABSTRACT-THINKING,
BRIGHT
(H igher scholastic m ental capacity)
EMOTIONALLY STABLE, FACES REALITY,
CALM
(H igher e g o strength)
ASSERTIVE, INDEPENDENT, AGGRESSIVE,
STUBBORN
(Dominance)
HAPPY-GO-LUCKY, HEEDLESS, GAY,
ENTHUSIASTIC
(Surgency)
CONSCIENTIOUS, PERSEVERING, STAID,
RULE-fiOUNO
(S tronger su p ereg o strength)
VENTURESOME, SOCIALLY BOLD,
UNINHIBITED, SPONTANEOUS
(Parmia)
TENDER-MINDED, DEPENDENT,
OVER-PROTECTED, SENSITIVE
(Premsia)
SUSPICIOUS, SELF-OPINIONATED,
HARD TO FOOL
(Protension)
IMAGINATIVE, WRAPPED UP IN INNER
URGENCIES, CARELESS OF PRACTICAL
(Autia) MATTERS, BOHEMIAN
SHREWD, CALCULATING, WORLDLY,
PENETRATING
(S hrew dness).
APPREHENSIVE, WORRYING, DEPRESSIVE,
TROUBLED
(Guilt proneness)
EXPERIMENTING, CRITICAL, LIBERAL,
ANALYTICAL, FREE-THINKING
(Radicalism)
SELF-SUFFICIENT. PREFERS OW N
DECISIONS, RESOURCEFUL
(Self-sufficiency)
CONTROLLED, SOCIALLY-PRECISE.
SELF-DISCIPLINED. COMPULSIVE
(High sclf-ccncept control)
TCNSE, DRIVEN. OVERWROUGHT,
FRETFUL
;-*igh crcjic tension)
, o b tain ed
• f cduffs
9
H-
3
oq
l —'
©
O
H-
► S
o
t-
©
n a
O H-
© 3
©
to
H
01
PERSONALITY PROFILE OF PERRY MASON
16 P.F. TEST PROFILE
214
PERSONALITY PROFILE OF RICHARD KIMBLE
AND THAT OF HIS VIEWERS
KEY: Circle = Viewer
Triangle = Role
UJZ
E O
or;
x g
od
S o . S
z £
o C
= u =
X £ fc.
t O 5
5^0
S . z
o*? ici
= 1 ih 2 & 2 ^ J j
= 3 0 > S
U J
z &
<5 S
Ou.
zo
z . «
> 9
16 P.F. TEST PROFILE
215
PERSONALITY PROFILE OF BEN CARTWRIGHT
AND THAT OF HIS VIEWERS
KEY: Circle « Viewer
Triangle « Role
a o
< o
9 °
Jo
5 5 “ Z 8
> « c
§2s
O k ! O ^ o O
3 X
< * 5
*5
►
2-
Stt ? is
n S ^
9 >; S 9 5
O 5
< D ;
2 <
O z
U “
U i
oo
2
z W o « o o
2 < ci
O '
16 P.F. TEST PROFILE
FACTOR
Stan
dard
Scora
LOW SCORE
DESCRIPTION
A
RESERVED, DETACHED, CRITICAL, COOL
(Sizolhymia)
B
LESS INTELLIGENT, CONCRETE-THINKING
(lo w e r scholastic m ental capacity)
C
AFFECTED BY FEELINGS, EMOTIONALLY
LESS STABLE, EASILY UPSET
(Lower e g o strength)
E
HUMBLE, MILD, OBEDIENT, CONFORMING
(Subm issivenest)
F
SOBER, PRUDENT, SERIOUS. TACITURN
(D esurgency)
G
EXPEDIENT, A LAW TO HIMSELF,
BY-PASSES OBLIGATIONS
(W eaker su p ereg o strength)
H
SHT, RESTRAINED, DIFFIDENT, TIMID
(Threctia)
I
TOUGH-MINDED, SELF-RELIANT,
REALISTIC, NO-NONSENSE
(Harria)
L
TRUSTING, ADAPTABLE, FREE OF JEALOUSY,
EASY TO GET O N WITH
(Alaxia)
M
PRACTICAL, CAREFUL, CONVENTIONAL,
REGULATED BY EXTERNAL REALITIES,
PROPER (Praxernia)
N
FORTHRIGHT, NATURAL, ARTLESS,
SENTIMENTAL
(A rtlessness)
0
PLACID, SELF-ASSURED, CONFIDENT, SERENE
(U ntroubled adequacy)
Q .
CONSERVATIVE, RESPECTING ESTABLISHED
IDEAS, TOLERANT OF TRADITIONAL
DIFFICULTIES (C onservatism )
Q *
GROUP-DEPENDENT, A "JOINER” AND
SOUND FOLLOWER
(G roup odherence)
Q .
CASUAL, CARELESS OF PROTOCOL, UNTIDY.
FOLLOWS OW N URGES
(lo w integration)
Q a
RELA- 0 . TRANQUIL, TORPID,
UNFRUSTRA1ED
(lo w erg ic tension)
O 1963. Institute for Personality
and Ability Testing.
STANDARD TEN SCORE (STEN)
A v e r a g e ^ *
3 4 5 6 7 8
A ile n o f 1 2
b y ab ou t 23% 4.4%
3 4 5 6 7
93% 15.0% 19.1% 19.1% 15.0%
HIGH SCORE
DESCRIPTION
OUTGOING, WARMHEARTED, EASY-GOING,
PARTICIPATING
(Cyclothymia)
MORE INTELLIGENT, ABSTRACT-THINKING,
BRIGHT
(Higher scholastic m ental o p a c ity )
EMOTIONALLY STABLE, FACES REALITY,
CALM
(H igher e g o strength)
ASSERTIVE. INDEPENDENT, AGGRESSIVE,
STUBBORN
(Dominance)
HAPPY-GO-LUCKY, HEEDLESS, GAY,
ENTHUSIASTIC
(Surgency)
CONSCIENTIOUS, PERSEVERING, STAID,
RULE-BOUND
(Stronger superego ttrcngA)
VENTURESOME, SOCIALLY SOLD,
UNINHIBITED, SPONTANEOUS
(Parmia)
TENDEX-MINDED, DEPENDENT,
OVER-PROTECTED, SENSITIVE
(Premsia)
SUSPICIOUS, SELF-OPINIONATED,
HARD TO FOOL
(Protension)
IMAGINATIVE, WRAPPED UP IN INNER
URGENCIES. CARELESS OF PRACTICAL
(Autia) MATTERS. BOHEMIAN
SHREWD. CALCULATING, WORLDLY,
PENETRATING
(Shrew dness)
APPREHENSIVE. WORRYING, DEPRESSIVE,
TROUBLED
(Guilt proneness)
EXPERIMENTING, CRITICAL. LIBERAL,
ANALYTICAL, FREE-THINKING
(Radicalism)
SELF-SUFFICIENT, PREFERS OWN
DECISIONS. RESOURCEFUL
(Self-sufficiency)
CONTROLLED, SOClAllY-fRECISE.
SELF-DISCIPLINED, COMPULSIVE
(High self-concept contrr
TENSE, DRIVEN, OVERWROUGHT,
FRETFUL
(High t-rgic tension)
ft o b ta in e d
o f ■ t .l t t
h n
W<
O H-
H ©
© i
©
> 1
>
Z
o
C O
H
O)
PERSONALITY PROFILE OF DANNY WILLIAMS
16 P.F. TEST PROFILE
217
PERSONALITY PROFILE OF EDDIE HASKELL
AND THAT OF HIS VIEWERS
KEY: Circle » Viewer
Triangle * Role
U1Z
CEO
V ) 3
e t rr
Sa s -
00.1 *2
o u o ' !
x a
& O 5
6*
l U U f V
vt O £
rz
ui <
<
3 v i n £ «
|5f is
9 > a P a
5 o
l _ o e
Z 3
a < 5
O “ I
S o
9?,
S6
z <
O z
u «
j5
2^
o o
z 6
81
< O
^ t/> z W
>25
16 P.F. TEST PROFILE
218
PERSONALITY PROFILE OF GRANNY, MOST-LIKED ROLE,
AND THAT OF HER VIEWERS
KEY: Circle « Viewer
Triangle « Role
11 | 5*
O K
ill m 2<
< I O i
° s
9*
O u.
z o
B O
16 P.F. TEST PROFILE
FACTOR
Stan
dard
Score
LOW SCORE
DESCRIPTION
A
RESERVED, DETACHED, CRITICAL, COOL
(Sizothymia)
B
LESS INTELLIGENT. CONCRETE-THINKING
(Lower scholastic m en tal capacity)
C
AFFECTED BY FEELINGS, EMOTIONALLY
LESS STABLE, EASILY UPSET
(lo w e r e g o strength)
E
HUMBLE, MILD, OBEDIENT. CONFORMING
(Subm issiveness)
F
SOBER, PRUDENT, SERIOUS, TACITURN
(D esurgeney)
G
EXPEDIENT. A LAW TO HIMSELF,
BY-PASSES OBLIGATIONS
(W eaker su p e re g o strength)
H
SHY. RESTRAINED, DIFFIDENT, TIMID
(Threctia)
I
TOUGH-MINDED, SELF-RELIANT,
REALISTIC, NO-NONSENSE
(Harria)
L
TRUSTING, ADAPTABLE. FREE OF JEALOUSY,
EASY TO GET ON WITH
(Alaxia)
M
PRACTICAL. CAREFUL. CONVENTIONAL,
REGULATED BY EXTERNAL REALITIES,
PROPER (Praxernia)
N
FORTHRIGHT, NATURAL. ARTLESS,
SENTIMENTAL
(A rtlessness)
0
PLACID. SELF-ASSURED, CONFIDENT, SERENE
(U ntroubled adequacy)
Q .
CONSERVATIVE. RESPECTING ESTABLISHED
IDEAS, TOLERANT OF TRADITIONAL
DIFFICULTIES (C onservatism )
Q >
GROUP-DEPENDENT, A "JO IN E R" AND
SOUND FOLLOWER
(G roup adherence)
Q .
CASUAL. CARELESS OF PROTOCOL. UNTIDY.
FOLLOWS OW N URGES
(lo w integration)
0 4
RELAXEO, TRANQUIL, TORPID.
UNrRUSTRATED
(lo w crgic tension)
© 1963. Institute for Personality
and Ability Testing.
STANDARD TEN SCORE (STEN)
«>• Average
12 34 5 6 7 8 9
-Ir-lr-jr-t-rt— t— h-*— *-
A s te n o f I 2
b y a b o u t 2J% 4.4%
6 7
19.1% 15.0% 9.2%
HIGH SCORE
DESCRIPTION
OUTGOING. WARMHEARTED. EASY-GOING,
PARTICIPATING
(Cyclothym ia)
MORE INTELLIGENT. ABSTRACT-THINKING.
BRIGHT
(H igher scholastic m ental capacity)
EMOTIONALLY STABLE. FACES REALITY.
CALM
(H igher e g o strength)
ASSERTIVE. INDEPENDENT. AGGRESSIVE.
STUBBORN
(Dominance)
HAPPY-GO-LUCKY. HEEDLESS, GAY,
ENTHUSIASTIC
(Surgency)
CONSCIENTIOUS, PERSEVERING, STAID,
RULE-BOUND
(S tronger su p e re g o strength)
VENTURESOME, SOCIALLY BOLD,
UNINHIBITED, SPONTANEOUS
(Parmia)
TENDER-MINDED, DEPENDENT,
OVER-PROTECTED, SENSITIVE
(Premsia)
SUSPICIOUS, SELF-OPINIONATED,
HARD TO FOOL
(Protension)
IMAGINATIVE. WRAPPED UP IN INNER
URGENCIES. CARELESS OF PRACTICAL
(A utia) MATTERS, BOHEMIAN
SHREWD, CALCULATING, WORLOLY,
PENETRATING
(Shrew dness)
APPREHENSIVE, WORRYING. DEPRESSIVE.
TROUBLED
(G uilt prcneness)
EXPERIMENTING, CRITICAL, LIBERAL,
ANALYTICAL. FREE-THINKING
(Radicalism)
SELF-SUFFICIENT, PREFERS OW N
DECISIONS. RESOURCEFUL
(Self-sufficiency)
CONTROLLED, SOCIALLY-PRECISE,
SELF DISCIPLINED, COMPULSIVE
.High sclf-ccnept ion»rol)
TENSE. D51VLN. C V iO K O U G H T ,
i R t l f U l
( H ' . j h crgic tcn-ien)
i$ o b ta in e d
o f a d u lts
e e
H- H
B&
W ©
M
A
N I I
O H-
H ©
© <
©
t o
M
<©
PERSONALITY PROFILE OF GRANNY, LEAST-LIKED ROLE
16 P.F. TEST PROFILE
220
PERSONALITY PROFILE OF VIEWERS OF GRANNY, MOST-LIKED
ROLE, AND VIEWERS OF GRANNY, LEAST-LIKED ROLE
KEY: Circle
Triangle
Mdstebiked
Le&st Liked
UJZ
< 5
y z
o * o .
So
o o
o £
£ 9
H i ili.
2 < 1 J « ¥
O a. i i S at £
a x . 5
ii?
SrSl
3a
5 i t I 9 O .2 , E " .9
! 2&1Z3S, o“S
its?
i s ? Z 3 « *
9> *
o ? Z P
u j Z
“ o
J Z <
So
u. tu
o o
< 5 2
9-9 z <
O z
u “
z o
gs
Coo
y U J o t
< 0 6 Q .
U 013V J
© 1963. Institute X o r Personality
and Ability Testing.
16 P.F. TEST PROFILE
221
PERSONALITY PROFILE OF ROLE OF GRANNY, MOST-LIKED,
AND ROLE OF GRANNY, LEAST-LIKED
KEY: Circle - Most Liked
Triangle « Least Liked
UIZ
C O
oc
i g
%Q
Z q
g O a
O “
So
5 - 2
i o _
K of .5
o g e
Z > J J
“ o &
z 2
O U
0 0.0
5*
Z u l
oS w Z * ■
>
< (A i
9*5
Q O
Z 3
2 s -
J 8
§1
P'S
<5 2 * 3 M
86
2 S
o z
u “ w u j
o o
z o
u z
a . <
£
< < at
Sis
f i O O
y " «
O | 5
t " 1
>92
dO-LOVd
This dissertation has been
microfilmed exactly as received 6 8-17,037
PERROW, Maxwell V erm ilyea, 1927-
A DESCRIPTION OF SIMILARITY OF PERSONALITY
BETW EEN SELECTED GROUPS OF TELEVISION
VIEWERS AND CERTAIN TELEVISION ROLES
REGULARLY VIEWED BY THEM.
U niversity of Southern C alifornia, Ph.D ., 1968
Social Psychology
University Microfilms, Inc., Ann Arbor, Michigan
A DESCRIPTION OF SIMILARITY OF PERSONALITY BETWEEN
SELECTED GROUPS OF TELEVISION VIEWERS AND CERTAIN
TELEVISION ROLES REGULARLY VIEWED BY THEM
by
Maxwell Vermilyea Perrow
A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(Communication)
June 1968
ui'dvc.naii i u r mc.ni'i ^nuirum xm
TH E GRADUATE SCHOOL
UNIVERSITY PARK
LOS ANGELES, CA LIFO RN IA 9 0 0 0 7
This dissertation, written by
M M E L L . M M I M M . P E M Q W .....
under the direction of hX$.... Dissertation Com
mittee, and approved by all its members, has
been presented to and accepted by the Graduate
School, in partial fulfillment of requirements
for the degree of
D O C T O R O F P H I L O S O P H Y
O
Dean
Date Jyne/....1?68
DISSERTATION COMM!
Chairman
(Ur
TABLE OF CONTENTS
CHAPTER PAGE
I. THE PURPOSE AND PROBLEM............... 1
The Purpose.......................... 2
The Problem.......................... 9
Similarity or Difference of
Personality Profiles ........... 10
Similarities or Differences of
Personal Data............ 11
Definitions.......................... 13
Hypotheses.......................... 16
Similarity or Difference of
Personality Profiles ........... 16
Similarities or Differences of
Personal Data................... 17
Assumptions.......................... 19
Limitations.......................... 21
Organization of Remainder of the
Study.............................. 23
II. REVIEW OF THE LITERATURE............. 24
Television Viewing: A Para-Social
Situation.......................... 25
Perception of Persons: An Important
Element in Television Viewing . . . 28
Personality: A Central Determinant
in the Perception of Persons . . . 34
CHAPTER
III.
/
*
IV.
iii
PAGE
Functional and Dysfunctional Values
of Television Viewing . . . . . . . 38
Identification: An Important Element
of Television Viewing............. 44
Program Selection: A Result of
Role-Viewer Identification .... 52
Summary............................... 57
METHODS AND PROCEDURES ......... 60
Methods............................... 60
Procedures........................... 61
Collection of D a t a ................ 61
Treatment of Data.................. 80
SIMILARITY AND DIFFERENCE BETWEEN THE
PERSONALITY PROFILES OF TELEVISION
ROLES AND THEIR VIEWERS............. 96
Most-Liked R o l e s .................... 97
Least-Liked Roles.................... 100
Granny, a Role Rated as Most Liked by
One Group and Rated a Least-Liked
Role by Another Group............. 102
Summary............................... 106
SEX, AGE, LEVEL OF EDUCATION, AND
STANDARD OF LIVING OF VIEWERS AND
OF R O L E S ............................. 108
I CHAPTER
VI.
iv !
i
PAGE !
Se x ................................... 109
A g e .............................. Ill j
Level of Education ; ........... 115 !
Standard of Living ......... 120 !
i
Summary.............................. 124
PERSONALITY TRAITS THAT MOST
I
CONSISTENTLY CHARACTERIZED MOST-LIKED
ROLES, LEAST-LIKED ROLES, AND VIEWERS
OF THESE R O L E S .................. 127 !
i
Factor Q3, Most Differentiating
Trait.......................... 128
i
Personality Traits Appearing Most
Consistently in Most-Liked Roles . 132
Personality Traits Appearing Most
Consistently in Least-Liked Roles • 133
Personality Traits on Which Roles and
Their Viewers Were Most Alike . . . 136
Personality Traits on Which Roles and
Viewers Differed Most......... 140
Personality Traits Characterizing
Viewer Groups of Granny Clampett:
One Which Rated Her a Most-Liked
Role and One Which Rated Her a
Least-Liked R o l e ............. 147
V
CHAPTER PAGE
i
Summary........................... 149
VII. SUMMARY, CONCLUSIONS, IMPLICATIONS,
AND RECOMMENDATIONS............. 151
Summary........................... 151
Conclusions and Implications .... 174
Recommendations Tor Further Study . . 186
BIBLIOGRAPHY . . ............................... 191
APPENDIXES
A. PERSONAL INFORMATION S H E E T ....... 200
B. TELEVISION ROLE PERSONALITY
QUESTIONNAIRE (COVER SHEET)..... 202
C. TELEVISION VIEWING D I A R Y .......... 205
D. TELEVISION ROLE PREFERENCE
QUESTIONNAIRE ........................ 208
E. SIXTEEN PERSONALITY FACTOR TEST
PROFILES (ROLES AND VIEWERS) .... 213
LIST OF TABLES
TABLE PAGE
I. Distribution of Sex of Sample......... 63
II. Distribution of Age of Sample......... 63
III. Distribution of Standard of Living
of Sample............................ 64
IV. Distribution of Level of Education
of Sample............................ 64
V. Number of Persons Who Judged Each Role
on the 16 P. F. Questionnaire .... 65
VI. Coefficients of Pattern Similarity
between the Mean Personality Profiles
of Most-Liked Roles and Their Viewers 99
VII. Coefficients of Pattern Similarity
between the Mean Personality Profiles
of Least-Liked Roles and Their
Viewers....................... 101
VIII. Coefficients of Pattern Similarity
between Viewers of Granny, Most
Liked, and Granny, Least Liked, and
Their Perceptions of Granny, Most
Liked, and Granny, Least Liked . . . 104
IX. Proportion of Males and Proportion of
Females in Viewer Groups... 110
TABLE
X.
XI.
i
XII.
XIII.
XIV.
XV.
XVI.
XVII.
Product Moment Correlations between
Personal Data of Viewers and of
Roles Viewed ........................
Distribution of Ages of Viewer Groups
and of Role Groups .................
Distribution of Level of Education of
Viewer Groups and of Role Groups . .
Distribution of Standard of Living of
Viewer Groups and of Role Groups . •
Outstanding Personality Traits of
Most-Liked Roles and of Least-Liked
Roles............. ..................
Personality Traits on Which Roles and
Viewers Differed No More Than One
, S t e n .................................
Personality Traits on Which Roles and
Viewers Differed More Than Three
Stens .................................
Outstanding Personality Traits of
Viewers Who Rated Granny a Most-Liked
Role and Viewers Who Rated Granny a
Least-Liked Role ...................
vii
PAGE
i
112
114
1 1 7 !
I
122 |
i
i
i
129
i
i
137
141
148
CHAPTER I
THE PURPOSE AND PROBLEM
Some elements in the viewing of television may
operate to attract viewers to certain television
programs and at the same time cause them to be repelled
by other programs. Such elements might include story,
setting, interrelationship of roles, clothing, or
mannerisms.
As several writers have reported, identification
seems to be an important element in people's perception
of roles that appear in the mass media.* Few of these
studies have employed the use of quantitative measure
ment to establish the extent -to which the viewer iden
tifies himself with the roles in those programs that he
views. Use of such a measurement might lend valuable
support to conclusions already reported in the litera-
Rudolf Arnheim, "The World of the Daytime
Serial," in Radio Research. 1942-1943. edited by Paul
F. Lazarsfeld and Frank N. Stanton (New York: Duell,
Sloan and Pearce, 1944), pp. 55-57; Paul G. Cressey,
"The Motion Picture Experience as Modified by Social
Background and Personality," American Sociological
Review. 3:4:519, August, 1938; F. E. Emery, k , Psycho-
logical Effects of the Western Film: A Study in Tele
vision Viewing: II. The Experimental Study," Human
Relations. 12:3:222, 1959; Leo A. Handel, Hollywood
Looks af* ~Its Audience: A Report of Film Audience
Research (Urbana: The University of Illinois Press,
195b), p. 145; Muzafer Sherif and Hadley Cantril, The
Psychology of Ego-Involvements: Social Attitudes and
Identifications (New York: John Wiley and Sons,
Incorporated, 1947), p. 351.
ture, and might also provide some new insights into the
relationships of personalities of viewers and viewers'
selections of particular television programs.
I. THE PURPOSE
The purpose of this study was to provide further
understanding of the importance of the personality of
the viewer of television and his attraction to certain
regularly viewed television roles, and thereby to
certain programs. It contains information on some of
the relationships that existed between television
viewers and television roles that were regularly viewed
by these viewers. More specifically, the study was
designed to help to determine if there was not greater
similarity of personality between television viewers
and the perceived personality of specific television
roles that they liked most than there was between the
personality of television viewers and the perceived
personality of television roles that they liked least.
Riley and Riley, among other authorities on
communication, stated that more attention should be
given to the recipient in the communication situation
than to the message or the environment in which the
message is delivered. They suggested:
3
. . . the traditional view does not take fully into
account ongoing processes of social interaction of
which the single communicative act is merely one
component. Nor does it take into full account those
psychological processes which, although they may be
going on within the individual recipient quite apart
from any particular communication, may nevertheless
markedly affect his reaction to it.22 Extensions
of this view in both the sociological and the
psychological directions seem necessary if the
mass-communications process is to be explained more
adequately or its outcome predicted more accurately.
22
As Lazarsfeld and Merton pointed out in 1943,
"Propaganda will not produce the expected response
unless its content corresponds to the psychological
wants of the audience" (Robert K. Merton, Social
Theory and Social Structure, rev., Free Press, 1957,
p. 519). A few years later, Berelson expanded on
this point:
. . • Effects upon the audience do not follow
directly from and in correspondence with the
intent of the communicator or the content of
the communication. The predispositions of the
reader or listener are deeply involved in the
situation, and may operate to block or modify
the intended effect or even to set up a boom
erang effect.
(Bernard Berelson, "Communications and Public
Opinion," in Wilbur Schramm [ed. ], Communications
in Modern Society .nUniversity of Illinois Press,
1946, pp. 1&3-84.)
It was becoming apparent that television programs
were viewed more by distinct, well-defined groups than
by audiences of a general make-up.
2John W. Riley, Jr. and Matilda White Riley,
"Mass Communication and the Social System," in Sociology
Today: Problems and Prospects, edited by Robert K,
Merton, Leonard Broom, and Leonard S. Cottrell, Jr.
under the auspices of the American Sociological Society
(New York: Basic Books, Incorporated, 1959), p. 543.
According to Glick and Levy, a broadcaster
should know who the viewers of his programs are, what
their relationships to the programs are, why they tend
to favor them, and the particular kinds of gratifica
tions they seek from them. The authors suggested that
in the future it would be most meaningful to conceive
of a ’ ’core audience"— a group of people who respond
fully and reliably to each program. Earlier in a
study of motion picture audiences it was reported that
"most stars do not gain support equally from all groups
of theatergoers, but appeal particularly to certain
segments of the population."
In an early report on radio listening, Stanton
said:
It isn't enough to know what programs are heard
and preferred. We want to know why they are
listened to and liked, and, furthermore, we want
to quantify their influence.5
Ira 0. Glick, Sidney J. Levy, and others,
Living with Television (Chicago: Aldine Publishing
Company, 1962), p. 38.
4Sherif and Cantril, loc. cit., citing Audience
Research Institute, Incorporated."Increasing Profits
with Continuous Audience Research (New York: Am. Book,
1941), reprinted by permission of G. H. Gallup.
5
Frank N. Stanton, "Psychological Research in
the Field of Radio Listening," in Educational Broad
casting. 1936: Proceedings of the First NationaT
In another early study of radio listening Beville
stressed that "... the educational or political broad
caster should be deeply concerned with the composition
of his audience as a factor which seriously affects the
g
achievement of his objectives."
In 1963 Broadcasting magazine carried comment on
an article by Murdoch and others to the effect that
"the diversification of interests among the public is
having a more drastic effect on mass media than most
n
people realize." According to the writer in Broad
casting. radio almost died in the early 1950's. Then,
realizing it could not compete with television, radio
began to specialize its programing. This article
recorded that many people were turning from television,
in growing protest against the kind of programing which
5 (continued)
Conference on Educational Broadcasting. Held in Wash
ington , D.C., on December 10. 11. and 12. 19367 edited
by C. S.~Marsh~TChicago: The University of Chicago
Press, 1937), p. 373.
6H. M. Beville, Jr., "The ABCD's of Radio
Audiences," The Public Opinion Quarterly. 4:2:206,
June, 1940.
n
Lawrence C. Murdoch, Jr., and others, "Mass
Media and the Gulliver Fallacy," Business Review.
October, 1962, published by Federal Reserve Bank of
Philadelphia, as reported in "Is the Mass Audience
a Thing of the Past?" Broadcasting. 64:2:42-43,
January 14, 1963.
6
seemed to appeal to a single set of tastes and inter
ests.®
Murdoch was alleged to have predicted that radio
stations which appealed to selected audiences would
Q
continue in good financial health.
It was hoped that the present study might help
producers of television programs to understand better
the relationships that exist between the roles in their
programs and the viewers who do or do not like to view
these roles. Also, it was hoped that the data reported
might be of value to producers who want to reach a
specific, well-defined audience— viewers who will view
their programs with regularity.
While it was recognized that other elements in
the viewing situation, such as availability of a pro
gram and manifest characteristics of a program, influ
ence viewing behavior, it was believed that the dynamic
personal interaction that takes place between viewers
and roles viewed by them is of primary importance in
understanding why viewers tend to select certain pro
grams for regular viewing and not others. — ■
If the findings of this study permit reliable
and valid prediction of the selection of programs, they
8Ibid., p. 43 9Ibid.
I should be of value to the following groups that seek
to build and retain regular viewing audiences:
1. Producers, writers, and directors of tele
vision programs who wish to build into their
programs roles that will interest specific
audiences and attract viewers to these
programs.
2. Sponsors of television programs who wish to
identify specific viewing groups which might
have interest in the products or services of
these sponsors.
The findings should also be of value to:
1. Students of social behavior who are inter
ested in the use that persons make of
television in their effort to secure rein
forcement of personal values and character
structure.
2. Students of the use of mass media who are
searching for quantitative data to serve
as evidence for the identification that is
thought to exist between the user of mass
media and those roles in mass media to which
the user is particularly attracted.
Prior reports on use of mass media suggested
that persons were attracted to roles in which they
perceived certain similarities of personality or of
circumstance to their own personalities or circum
stances.^ Little in this vein of research was reported
in studies of television viewing. With the exception
of those by Emery and Fearing, such studies as were
reported did not describe quantitatively the full range
of elements in personality which may be operative in
the viewing situation.
The present study was designed to provide quanti
tative measurement of the degree of similarity or dis
similarity between the personalities of the viewers and
the perceived personalities of roles viewed by them.
Personality was defined as one's unique pattern of
traits. A trait was described as any distinguishable,
relatively enduring way in which one individual differs
from another.
Cattell's personality inventory distinguishes
sixteen first-order trait factors.*1 Through use of
this instrument certain similarities and dissimilarities
between personality profiles were described. These
*®Arnheim, op. cit., p. 56; Handel, loc. cit.;
Sherif and Cantril, loc. cit.; Emery, loc. cit.
113PAT 16 P. F. (Form A, 1962 Edition. The
Institute for Personality and Ability Testing, 1602-04
Coronado Drive, Champaign, Illinois).
trait factors were taken to be relatively enduring
patterns of organization of personality which provided
information of more lasting value than directly inter
preted statements of opinion.
II. THE PROBLEM
This study was designed to provide information
on the following questions:
1. Is there greater similarity or difference
between the personality profiles of tele
vision viewers and the perceived personality
profiles of regularly-viewed, most-liked
television roles than there is between the
personality profiles of television viewers
and the perceived personality profiles of
least-liked television roles? 1
2. Is there greater similarity or difference
between television viewers and their per
ception of most-liked television roles than
there is between television viewers and
their perception of least-liked television
roles with respect to age, level of educa
tion, standard of living, and sex?
The first question, that of similarity or dif
ference between personality profiles, comprised the
1 0
major hypotheses of the study, while the second
question, about personal data, comprised the minor
hypotheses.
To provide a framework for statistical examina
tion of these data, the following specific questions
were structured. /
Similarity or Difference
of Personality Profiles
Is there a significant similarity or difference
between:
1. The mean personality profile of a group of
viewers and the mean personality profile of
a role as perceived by these viewers and
rated by them as a most-liked role?
2. The mean personality profile of a group of
viewers and the mean personality profile of
a role as perceived by these viewers and
rated by them as a least-liked role?
3. The mean personality profile of a group of
viewers who rated a particular role as a
most-liked role and the mean personality
profile of a group of viewers who rated
the same role as a least-liked role?

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Effects Of Video Tape Feedback Versus Discussion Session Feedback On Group Interaction, Self Awareness And Behavioral Change Among Group Psychotherapy Participants

PDF

An Application Of A Two-Stage 'Attention' Model To Concept Formation In The Mentally Retarded

PDF

A History Of The Canadian Radio League: 1930-1936

PDF

The Influence Of The Control Of Personal Set Upon Prediction By Factored Tests Of Temperament And Interest

PDF

The Development And Validation Of A Proverbs Test For The Selection Of Supervisors

PDF

A Criticism Of News Broadcasting And Regulatory Issues In The Richards Case

PDF

An Experimental Study Of Aggression Habit Patterns And Sex Role

PDF

A Description Of The Organization, Production Process, And Production Environment Of A Dramatic Television Series

PDF

The role of cues in the arousal of anxiety

PDF

A Study Of The Effectiveness Of Teaching Methods Of Study To Selected High School Freshmen

PDF

A Study Of The Effects Of Generalized Expectancies Upon Accuracy Of Interpersonal Perception

PDF

A Study Of The Relationship Of Temperament Variables To The Ability To Make Certain Judgments Of Emotional Behavior

PDF

The Effects Of Making Social Desirability Judgments On Personality Inventory Scores Of Schizophrenics

PDF

A Study Of Recall Set Utilizing Early Childhood Recollections And Recent Recollections

PDF

Perceptual Defense And Somatization: A Comparison Of The Perceptual Thresholds Of Obese And Peptic Ulcer Patients

PDF

International Broadcasting And Its Societal Environments: A Test Of Hypotheses

PDF

A History Of Roman Catholic Church Policies Regarding Commercial Radio And Television Broadcasting In The United States, 1920 Through 1961.

PDF

Behavioral (Social) Intelligence: A Factor Analysis

PDF

Delinquency As A Function Of Intrafamily Relationships

PDF

A Study Of Affective Sets: The Effects Of Family And Non-Family Verbal Contexts On Word-Need Stimuli In A Word Association Experiment With Reference To Pleasant And Emotional Tones Of Associated...

Asset Metadata

Creator Perrow, Maxwell Vermilyea (author)

Core Title A Description Of Similarity Of Personality Between Selected Groups Of Television Viewers And Certain Television Roles Regularly Viewed By Them

Contributor Digitized by ProQuest (provenance)

Degree Doctor of Philosophy

Degree Program Communication

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag OAI-PMH Harvest,psychology, social

Language English

Advisor Harwood, Kenneth A. (committee chair), Jacobs, Alfred (committee member), Smith, Don C. (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c18-630410

Unique identifier UC11361243

Identifier 6817037.pdf (filename),usctheses-c18-630410 (legacy record id)

Legacy Identifier 6817037.pdf

Dmrecord 630410

Document Type Dissertation

Rights Perrow, Maxwell Vermilyea

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA