‘Hl

 
 

WW

'I
l

ég WWWlWlWHIHlHHI‘

THS

  

" ' f) O ‘ ‘ .4)“ '1!“ I
«~- c—.«-.-\.~/ o , z o
T RAH

llllllllllllllllllllillllllllllllllllllll.llllllll }
3 1293 00558 4820

LIBRARY
Michigan State
University

 

 

 

 

 

 

 

 

This is to certify that the

thesis entitled

COMMUNICATIVE .RECEPTIVITY

AND MEDIA INVOLVEMENT
presented by

MARCIA LEE RIBBLE

has been accepted towards fulfillment
of the requirements for

M.A. degree in COMMUNICATION

 

 

Wm

Major professor

 

Date rm”; 15; In!

0-7539 MS U is an Affirmative Action/Equal Opportunity Institution

 

MSU

 

 

 

RETURNING MATERIALS:
Place in book drop to

 

 

LIBRARIES remove this checkout from

m your record. FINES will
be charged if book is
returned after the date
stamped below.

[122' rig/63?”

 

 

 

COHHUNICATIVE RECEPTIVITY AND HEDIA INVOLVEMENT

'By
Marcia Lee Ribble

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

MASTER OF ARTS

Depart-ent of Communication

1987

 

502 006’XCR

Copyright by Marcia Lee Ribble,

1987.

ABSTRACT
COMMUNICATIVE RECEPTIVITY AND MEDIA INVOLVEMENT
By
Marcia Lee Ribble
This study tested the Companionship Activities and

Emotional Involvement with the Media scales. The
companionship activities scale has five subscales:
print media. visual media, friend, intimate and
impersonal. The emotional involvement scale has two
subscales: emotional behavior and emotional reaction.
Those scales were correlated with McCroskey's
Willingness to communicate scale to test the theory
that shy people turn from human to media companionship
activities without supporting that theory. The study
did find significant correlations between emotional
behavior and the five companionship factors. The
magnitude of those correlations suggests the presence
of intervening variables and additional research is
suggested. Those variables may be embedded within the
communicative receptivity construct measuring the
degree to which one may perceive emotional involvement

with others as harmless to self.

DEDICATION

To Frank, Mary, Dawn, Lee and Vicki who are my

most demanding critics and my staunchest supporters.

iv

ACKNOWLEDGMENTS

Special thanks must go to the intrepid members of
my committee, Doctor Charles Atkin, Doctor Ronald
Tamborini and Doctor Mary Bresnahan. Astute practitioners
of the art of raising baby scientists and scholars, they
have challenged, prodded, provoked, comforted, nurtured
and lived with me through the time I was learning to accept

my limitations without being limited by them.

TABLE OF CONTENTS

Title Page

Abstract

Copyright

Dedication

Table of Contents
Table of Tables
Chapter I Introduction
Chapter II Methods
Chapter III Results
Chapter IV Discussion
References

Tables

vi

ii
iii

iv

vi

20
26
37
44

49

Table 1.

Table 2.

Table 3.

Table 4.

Table 5.

Table 6.

Table 7.

Table 8.

Table 9.

TablelO.

Table 11.

Table 12.

TABLE OF TABLES

Companionship Activity Scale Items

Emotional Involvement with the Media
Items

Willingness to Communicate Scale Items

Factor Analysis of the Companionship
Activities Scale

Factor Analysis of the Emotional
Involvement with the Media Scale

Factor Analysis of the Willingness
to Communicate Scale

Pearson Correlations of the
Companionship Activities Scale

Pearson Correlations of the Emotional
Involvement with the Media Scale

Pearson Correlations of the
Willingness to Communicate Scale

Pearson Correlations of Willingness to
Communicate with the Companionship
Activities Scale

Pearson Correlations of Willingness to
Communicate with Emotional Involvement
with the Media

Pearson Correlations of Companionship

Activities with the Emotional
Involvement with the Media Scale

vii

49
50

51
52

53

54

55

55

56

56

57

57

Table 13.

Table 14.

Pearson Correlations of Media Use with 58
the Companionship Activities, Emotional
Involvement with the Media and Willingness
Willingness to Communicate Scales

Pearson Correlations of Age, Gender 59
and Education with the Companionship
Activities, Emotional Involvement

with the Media and Willingness to
Communicate Scales, and with TV Use Per
Day, Favorite Television Time Slot, and
Radio Use Per Day

viii

CHAPTER I INTRODUCTION

This paper was designed to report the construction
of scales measuring human and media companionship
activity choices and emotional involvement with the
media. The relationship of those scales to one
another and to James McCroskey's (1986) willingness to
communicate scale was then tested. The initial
assumption was that a relationship would be found
between willingness to communicate and choices of
companionship activities. A relationship was predicted
to exist between willingness to communicate and emotional
involvement with the media. It was also thought that a
relationship would exist between choices of human and
media companionship activities and emotional involvement
with the media.

The predicted relationships between the variables
under study werefelt to be due to factors which could
predispose people to select activities for companionship
The three basic factors this study examined are shyness,
companionship need fulfillment and attempts to maintain
comfortable levels of arousal. Based on previous
studies it was felt that those three factors might work

in tandem to influence people's choices of companionship

activities.

Prior media studies had discovered that people
used media for companionship need fulfillment quite
consistently. Rubin (1981) found nine clusters of
media viewing motivations: pass time/habit, arousal,
companionship, program content, relaxation, information,
escape, entertainment and social interaction. In
addition to this group of motivations, Zillmann (1987)
and Zillmann and Bryant (1985) have described a
relationship between selective exposure to the media
and individual attempts to either maintain a
satisfactory level of arousal or to raise or lower the
level of arousal to a more comfortable level.

Shyness has also been found to influence choices
of media exposure. Whether it is called shyness
(Cheek and Buss, 1981), loneliness (Russell, Peplau
and Cutrona, 1980), social—communicative anxiety
(Daly and Stafford, 1985), communication apprehension
(McCroskey, 1985), or communication reticence (Burgoon
and Hale, 1983), or any other names by which the
inability to communicate with others is known, shyness
researchers agree that shyness causes many problems for
shy people. Among the most obvious problems which have
been shown to relate to shyness are alcoholism, social

isolation, fear, and constricted cognitive control.

Kagan and Resnick (1986) suggest that between five and
ten percent of the total population are shy and retreat
from some types of social interaction and move into
others.

Thus, the researcher began to wonder if the reason
that shy people avoid the company of others might also
explain why shy people also use media differently than
nonshy people. Using Zillmann's (1987) explanation of
the hedonic principle that people make choices of media
exposure and other behaviors to control arousal levels
and maintain a comfortable level of arousal, this
researcher asked if it might be possible that shy people
choose media companionship activities rather than human
companionship to reduce higher than normal levels of
arousal. This view is supported by Kagan and Resnick's
(1986) findings that shy children retreat from other
children and have persisting patterns of shy behaviors
over longitudinal study. Both Zillmann (1987) and Kagan
and Resnick (1986) suggest that people whose arousal
levels are uncomfortable will move away from the source
of arousal. For people who are shy the source of arousal
might be other people, even though Cheek and Buss (1981)
described shy—sociable people as longing to be with other
people.

If people who are shy retreat from others to reduce

their levels of arousal, it would seem reasonable to

assume that one of the basic reasons for their doing so
might be an attempt to control the amount of arousal
created by the emotional involvement they would have in
a situation. Therefore an assumption would be made that
as people become more shy it would be increasingly
difficult for them to enter companionship activities
which are highly emotionally involved. It would seem
that if one looked at activities which are chosen for
companionship they would range from those which are
very emotionally involving such as intimacy to those
which are less emotionally involving such as reading a
book or watching television.

Another possible reason for the choices shy people
make of companionship activities might be found in
Mehrabian's (1968) discussion of ambiguity and
schizophrenic children. Mehrabian studied schizophrenic
children and found a strong correlation between those
children and parents who send children ambiguous
messages. His study attempted to find out whether the
children had become schizophrenic in response to their
parents' ambiguous messages. However, the results of
Mehrabian's study indicated that parents of schiZOphrenic
and nonschizophrenic children were not significantly
different in the number of ambiguous messages sent to
their children. Thus, it may be that schiZOphrenic

children are differentially affected by ambiguous

messages, even though both groups receive them. If
ambiguity influences shy behaviors, this influence
might occur as a result of there being more possible
ambiguous situations as intimacy increases and fewer
possible ambiguous communicative events as intimacy
decreases. This study was not designed to test the
possible relationship between the reception of
ambiguous messages and choices of behavior, but it is
a plausible explanation which also deserves testing.

A third possible explanation for why people differ
in the choices they make of activities to fulfill their
need for companionship might reasonably be the familial
process of socialization in which they were raised. It
may well be that some families have interaction patterns
which are close and lovingly intimate, some are friendly
and cordial to one another, and others have interaction
patterns which are cold, distant and even hostile.

Those children who grew up in families which are cold
may not have developed skills which would allow them

to know how to behave in intimate relationships or even
in friendships. Their lack of social skills could be
reflected in a hesitance to attempt social actions for
which they are unprepared. Thus, they may retreat from
forms of social interaction which require that they deal
intimately with others, preferring companionship

activities which are less close, or even moving away from

relating to people and choosing to relate to characters
on television or in books who do not demand response.
It is possible that these people would make the choice
of media consumption for companionship because the media
would allow them to find some sights and sounds of other
people, but people who have no expectations the shy person
is unable to fill. Although this study does not test
socialization processes in their relationship to the issue
of why shy people may consume media or interpersonal
activities to fulfill their need for companionship, it
does look for patterns of selective exposure to media
versus interpersonal activitiesvddxflrwould be consistent
with such an explanation.
Shyness and Media Exposure Patterns

Basic to all three possible explanations for why a
person might choose media consumption in preference to
human companionship is the realization that a possible
continuum exists between choices of human and media
companionship activities. Perhaps shy people are more
likely to prefer less intensely intimate forms of
interaction whether their reason for doing so is a way
of coping with arousal levels, the increased ambiguity
of intimate relationships, or a socialized lack of skills
with which to handle the demands of intimacy. This study
will attempt to tap only the arousal aspect of media use

patterns, to limit potential confounding of variables.

Conceivably, it is reasonable to assume that people

who are shy would choose to consume larger quantities
of media than people who are not shy. Previous studies
of mass media effects have demonstrated that a
relationship does exist between heavier consumption of
media and shyness.

Johnstone (1974) reported a study of 9,000 school
students and the relationship between their use of media
and their inclusion into or exclusion from their peers.
The results of this study showed that television use
patterns decline rapidly in the four years of high school
and that boys who are not socially integrated consume
more media than those who are integrated. According to
Johnstone, although "this does not establish directly
that television viewing functions to reduce tensions
generated in the social world, it does suggest at least
that felt deprivations emanating from the social
environment do lead people to turn to television to seek
out emotional restoration (1974, pp. 41—42)."

Rubin (1981) constructed nine clusters of viewing
motivations and examined the relationship between those
motivations and the type and amounts of media consumed
by people who most strongly chose those categories. The
people who fit into the companionship motivation group

not only watched television, but also appeared to be

strongly attached to it, choosing dramatic narratives.
In this study the link between the use of media for
companionship and a brief description of a preferred
content type is made, but no rationale for why a group
might choose one genre over another is given other than
an assumption from the uses and gratifications paradigm
that viewers were making their choices to fill
pre-existing needs.

Perloff, Quarles and Drutz (1983) asked the question
of whether people who have fewer social interactions
watch more television because they have more time
available or whether television viewing provides viewers
with some sort of psychological alternative to
interpersonal relationships. Perloff, et al.'s sample
of college students were tested to see if a relationship
existed between dating involvement, number of friends,
depression and amounts of television viewing time.
Perloff and his associates discovered that the interaction
between those situational variables and television
viewing was mediated by the respondents' subjective
dispositions, raising the question of how many different
types of personality variables might intervene between
media use patterns and situational variables.

Rubin (1983) found a predictive relationship between

companionship and subsequent viewing levels while failing

to specify content areas which might account for that
relationship, although his study did show a general
liking for television as a correlate of those viewing
levels. Subsequently Rubin examined possible links
between loneliness, parasocial interaction and local
television news viewing. Rubin, Perse and Powell (1985)
thought that lonely people used media due to a sense of
intimacy with television newscasters. Rubin, g£_gl.
did not find any significant relationship between
loneliness and parasocial interaction, but reiterated
previous findings that lonely people watch and respond
to television news differently than nonlonely people.

Austin (1985) utilized the revised UCLA loneliness
scale, correlating results on that scale with media use
patterns. He found no relationship between loneliness
and movie going, magazine reading, book reading and
television viewing. He found a modest inverse
relationship between loneliness, newpaper reading, and
radio use. He suggested that more specific content
based approaches might yield better results.

Reporting on the use of media to satisfy one's
needs for human companionship, Canary (1986) suggested
that loneliness is not one construct but a continuum of
relationships which range from human and media
companionship to a chronic form of loneliness which is

characterized by withdrawal from nearly all interaction.

10

Canary's results did not find an overall difference
between shy and nonshy media gratifications. But he
suggested that the shy might move away from television
to less interactive forms of content such as reading
books. In addition, his study included family as a
variable leading this researcher to wonder if the
inclusion of human relationships as part of a continuum
of companionship satisfaction activities might be more
revealing of the interaction between choices of human
and media companionship activities.

Responding to Zillmann's (1987) prediction of a
relationship between media exposure choices and attempts
by the individual to control his or her levels of arousal
Finn and Core (1987) studied the interaction between
shyness and television viewing motivations. They found
that shyness was correlated with the companionship
viewing motivation about equally for males (r=.28,
p<.001) and females (r=.27, p<.001). Shyness was not
correlated with relaxation, entertainment, or arousal
for either males or females. Information and habit
were correlated at r=.l9, p<.05, and r=.25, p<.01,
respectively, for women only. Pass time was correlated
at r=.22, p<.01 for males and at r=.32, p<.001 for females.
Escape was correlated with shyness at r=.l7, p<.05 for

males and at r=.32, p<.001 for females. Thus, this

ll

researcher felt that companionship was the cleanest
television viewing motivation when correlated with
shyness and that its use would allow the most accurate
results.

Three major problems have been exposed by these
studies. The first problem is a conceptual one. At
least part of the difficulty in finding consistent
results in the relationship between shyness and media
use patterns may result from an assumption that many
methods of measurement are capable of tapping the same
phenomenon. Another difficulty may have come from an
assumption that people withdraw from human interaction
and use media mainly because human interaction is not
available. Although Finn and Core (1987) distinguish
between shyness and loneliness, often the terms are
interchanged as though they measure the same thing.

To measure television usage patterns, researchers have
utilized peer relationships in teenagers, content choices,
self-reports of loneliness, dating patterns, parasocial
interaction with news anchors on television, and
psychological scales. It is conceivable that peer
interactions may represent many different underlying
mechanisms from unpleasant body odors to physical

unattractiveness, and might include such demographic

factors as physical distance from peers. Content choices

12

may result from many different factors. If one is
consuming media with others group decision-making
processes may supercede individual choice. Rubin,

g£_gl. (1985) described a situation in which subjects
chose to reject or failed to choose interpersonal
interaction to relieve loneliness, preferring instead

the choice of media consumption. Prior conceptualizations
of loneliness and shyness were based on the rationale
that people choose to consume media because there are no
available humans to interact with, but when Rubin's
sample was given the choice of interpersonal relationship
to reduce their state of loneliness they were more likely
to choose media instead.

Zillmann's (1987) explanation of the use of media
makes sense in that context. If people are comfortable
using media and uncomfortable with other people, it
would follow that media would be given preference. Thus
a conceptualization of a shy person as one who feels
discomfort with human interaction and more comfortable
with media interaction would seem to fit with recent
study results and to move away from research simply
characterizing loneliness to ask why a person might be
lonely in the first place. That would allow researchers

to test specific reasons for loneliness such as shyness.

A second problem encountered is an ambiguity over

13

why there might be a relationship between shyness and
media use patterns. Little has been done beyond merely
establishing whether or not a form of shyness or
loneliness exists and attempting to relate that shyness
to media exposure. This is true despite Rosengren's
(1974) warning that media use may be based more on
(subjective interaction with content than on generic
choices of media. If people who are shy choose media
interaction over human interaction it is reasonable to
assume that they are finding something in that interaction
which replaces the interaction they might have had with
humans if they did not withdraw from them. Although the
concept of parasocial interaction did not prove
particularly useful as a way of explaining the interaction
of people who are shy with the media, there are other
possible explanations for human interaction with media.
Recalling the previous studies which often found a
relationship between shyness and an attraction to the
media, it would appear that the dramatic content linked
to the use of media for companionship (Rubin, 1981) may
serve to meet specific needs in shy people. Rubin did
not speculate on why that relationship might have been
found, but the more colloquial label "tear jerker" might
contain a clue. If people are moved to tears while

consuming that content it is reasonable to assume that a

14

strong emotional involvement with the content of the

tear jerker is taking place. The content might not be
specifically related to the lives of shy people, but the
thematic content of emotion may be related. A person who
is experiencing the emotion of grief may see a portrayal
of grief tucked inside a story line which does not
exactly mimic the experience of the viewer, but is made
common by the similar emotional quality. A child who is
afraid of parental abandonment could watch Bambi and feel
some degree of relatedness to Bambi's loss of his parents.
Reading a book one might sense that the characters know
the same feelings that human experience and feel close

to those characters. While media content may vary
considerably in its ability to link the viewer to the
content, the process of linkage may be similar for many
viewers.

Recent studies which examine the emotional process
by which viewers may involve themselves with the media
include one carried out by Tamborini, Stiff and Heidel
(1987). Operating under the assumption that viewers of
horror movies will become emotionally involved with that
content, Tamborini and his associates predicted and found
a relationship between emotional involvement with graphic
horror, physical arousal, and later self—reports of

negative affect in empathetic viewers. It may be that

15

people who become physically aroused by content in media
persentations connect these feelings of arousal with
prior occasions when they experienced similar feelings of
arousal, thus tapping into their memories, their present
fears, or any other emotions.

The studies which link media use with arousal have
raised many issues while demonstrating that people do
utilize media to balance hedonic valence. One issue
which is particularly applicable to this study is
Zillmann's (1987) caution that understanding the response
of consumers to media content can be biased by research
expectations and assumptions made by researchers about
that content. If researchers would believe that "only
Sissies cry" in response to media content, that will
influence their perceptions of how one "ought" to respond,
thus limiting the ability to see what is really there.
This is especially relevant in a world which equates male
as tough, unfeeling, nonresponsive, and emotionally
distanced from media most of the time--with the exception
of sports events and political speeches, when emotional
response is almost mandatory.

A demonstration of the effect of that attitude on
behavior patterns linked to media exposure can be found
in Zillmann, Weaver, Mundorf and Aust (1986). Zillmann,
.g£_al. tested the effect that being with a person of the

opposite gender would have on one's behavior when

16

confronted by the fear induced by graphic horror. They
found that both males and females altered their response
to graphic horror when in the presence of persons of the
opposite gender to conform to the societal norms of
male-strong and female—vulnerable behaviors, regardless
of their physiological levels of arousal.

A third problem found in media effects research is
the lack of external validity and generalizability
caused by forced choice experimental procedures. No
matter how much care is taken in an experimental
manipulation, subjects do not necessarily behave as they
would when able to make their own choice of media to
consume. That does not invalidate the work of laboratory
researchers whose valuable insights often contain clues
which lead to deeper understanding of human behavior.
However, it is a caution that researchers should not
accept results of experiments as adequate without the
backing of nonexperimental studies to act as a real world
check on assumptions of validity.

This study was designed to respond to the three
problems of conceptual weakness, lack of a direct link
between shyness and media exposure, as well as an
ambiguity about whether media is chosen instead of
personal interaction with others, and the need for less
controlled studies to provide a real world link to

experimental manipulations. Therefore the researcher

l7

proposes these research hypotheses:
Hypothesis 1. The choice of human companionship
activities is positively related to the willingness
to communicate with others.
Hypothesis 2. The choice of media companionship
activities is inversely related to the willingness to
communicate with others.
Hypothesis 3. The choice of media companionship
activities is positively related to emotional
involvement with the media.
Hypothesis 4. The choice of human companionship
activities is inversely related to emotional
involvement with the media.
Hypothesis 5. Emotional involvement with the media is
inversely related to willingness to communicate.
Hypothesis 6. The use of media is inversely related to
willingness to communicate with others.
Hypothesis 7. The use of media is positively related to
emotional involvement with the media.
Hypothesis 8. The use of media is inversely related to
the choice of human companionship activities.
Hypothesis 9. The use of media is positively related to
the choice of media companionship activities.

It is expected that people who are willing to

communicate (that is those who are not shy) will

18

find it easy to relate to humans for the fulfillment of
their companionship needs and will find, within those
human relationships sufficient emotional involvement

to fulfill their need for emotional involvement. Thus,
it is expected that they will not have the need for
media companionship and will consume less media. It is
also expected that they will not be emotionally involved
with the media. At the same time those who are not
willing to communicate (that is, those who are shy)

are expected to choose media companionship activities,
to consume a greater quantity of media and to be more
emotionally involved with the media they consume.

These relationships are predicted on the basis of
Zillmann's theory of the media exposure usage to achieve
hedonic valence. It is assumed that people who are
willing to communicate with others and who have achieved
a state of hedonic valence through their interpersonal
relationships will not be motivated to use media for
companionship, although they may use it to fill other
needs such as the need for entertainment. One the other
hand, it is assumed that people who are not willing to
communicate with others will not be able to achieve the
state of hedonic valence (that is comfortable levels of

arousal) through their interpersonal relationships and

19

they will therefore use media to fill their need for

companionship.

20

CHAPTER II METHODS

Two-hundred fifteen subjects were recruited from
communication classes at a large midwestern university.
They were free to refuse to participate without penalty
and were given an extra credit option if they chose not
to participate.

The procedure followed was to test the subjects in
their classes. Class members were given a survey which
contained a shyness measure (McCroskey's, 1986,
willingness to communicate scale); a scale to measure
emotional involvement with the media which was
constructed for this study; a scale of activities used
for the purpose of companionship which was also created
for this study; and several demographic measures. After
subjects filled out the surveys they were thanked for
their participation and debriefed.

The use of the scales rather than an experimental
procedure was chosen to enable the researcher to test
the scales. In addition, it was felt that shyness is
a personality trait which cannot be manipulated. ruling

out the utilization of experimental research techniques.

21

Neither can shyness be adequately compared with a
control group. However, it should be cautioned that
test results in themselves should not be seen as a
sufficient reason for making assumptions about
individuals without in—depth interviews to back up the
use of the scales.
The Companionship Activities Scale

The companionship activities scale is composed of
ten items which are combined to form a subscale (human
companionship activities), thirteen media items which
are combined to form a subscale (media companionship
activities), and a single item, pet, which provides an
alternative to the narrow perspective that sees only
human and media interaction as possible fulfillment of
the need for companionship. One of the human items was
designed to tap those activities which one might enjoy
while alone. The assumption is that while a person is
engaged in activities which are intensely involving of
self, the person may have a sense of completeness, ie. of
the self as the chosen companion. See Table 1. for scale
items.
The Emotional Involvement with the Media Scale

The emotional involvement with the media scale is

used as a tool to further investigate the connection

Finn and Core (1987) made between Zillmann's (1987)

22

mood-management hypothesis and the use of media to fulfill
companionship needs. Tamborini, Stiff and Heidel (1987)
have investigated the role of emotional involvement in
determining reactions to media by incorporating measures
of personality need states, media usage patterns and/or
effects, and the arousal response to media which forms

the connection between the need state and its fulfillment.
Tamborini, g£_gl. used both physiological and self-report
measures of emotional involvement. The present study is
meant to extend the self-report measure of emotional
involvement with the media which Tamborini and his
associates described as fictional involvement with
characters and which was measured by such items as "I
really get involved with the feelings and characters in a
novel....If I see something very sad or very beautiful on
television, I sometimes cry because what I see makes me
very happy or very sad." The emotional involvement with
the media scale divides that involvement with the media
into a behavioral component and a reactive component.

The emotional behaviors component of the scale was
created to address more of the possible emotional
resonses one might have when consuming media, including
tears, laughter, and anger. The emotional reaction
measure was designed to tap the social aspect of the

emotional reactions people have to the media. Items link

23

emotional involvement with the media to a willingness to
share one's feelings with others by discussing the
characters and action of media content with others. See
Table 2. for the seven items which comprise the emotional
involvement with the media scale.

The Willingness to Communicate Scale.

McCroskey's (1987) willingness to communicate scaleis
being used as a measure of shyness. The willingness to
communicate scale is expected to further test Canary's
(1986) findings that the shy are actually unwilling to
communicate with others rather than wanting to talk to
others. Canary's work suggests that people who are shy
withdraw from interpersonal interaction to show a
preference for the more solitary forms of media, such as
print and music, in conjunction with the movement away
from television and movies.

Interaction between these variables would support
Zillmann's (1980, 1983, 1987) and Zillmann and Bryant's
(1985) theories of affect dependent responses to
stimuli. The implications of Zillmann's theory for an
understanding of the mechanisms which drive shyness
include an awareness that the fears of the shy could
create sufficient arousal that interpersonal interaction
could be felt as noxious. Thus, the individual might

tend to move away from interpersonal interaction and

24

towards media interaction.
Demographic Variables

The demographic items which were measured consisted
of age, gender, educational level, and several media use
pattern items. Three media use items were measured:
"Hours in an average day watching TV....Favorite TV time
slot....Hours in an average day listening to radio."

The item favorite TV time slot was measured as mornings,
afternoons, prime time and late night television. The
other two items were measured by simply allowing subjects
to estimate the amount of time per day spent consuming
media. Age and time spent consuming media were ratio
scales. The other items were dummy coded.

The gender item was added because Finn and Gore's
(1987) research had indicated that gender might play a
strong role in mediating the results of media effects
research. The amount of media consumed was measured
because one of the basic assumptions of media effects
research is that personality variables affect, not only
the type of media consumed, but also the amount of media
consumed. With shyness in particular, there is an
assumption that shy people consume greater quantities of
media than nonshy populations do. Therefore, the research
assumption is that there is a strong correlation

between the use of media for companionship and the amount

25

of media which is actually consumed.
Statistical Procedures

Factor analyses were computed to examine the
scales for the reliability of items and the presence
of subscales which are orthogonal to one another.

PA 1 and Varimax rotation were used. Unless an item
very clearly loaded on one factor alone (variable 2 of
the companionship scale), minimal factor loadings of
.50 were used as a cut off point in deciding whether
or not to include an item in a factor. Even if an
item loaded at more than .50, if it also loaded highly
on another scale [within .10 of the first factor loading]
it was discarded. As an example, an item which loaded
as .55 on one factor and .48 on another was dropped,
while an item which loaded at .60 on one factor

and .48 on another was included.

After the factor analyses were computed Pearson
Correlations were used to test the possible
relationships between the variables formed through
the factor analysis. Although the relationships of
the scales themselves had been predicted, the
relationships between the subscales could not be
predicted before they were known through the
factor analyses. Thus, the statistical procedures
are presumed to be a test of the hypotheses and a

method of discovering unknown relationships.

26

CHAPTER III RESULTS

Factor analyses were done to detect subscales and
to ensure reliability and orthogonality of the scales
and their subscales. See Tables 4, 5, and 6 for a
report of factor analysis results. Tables 7, 8 and 9
report results of Pearson Correlations computed with
the factor analyzed subscales for the companionship
activities scale, the emotional involvement with the
media scale and the willingness to communicate scale.

The factor analysis of the companionship
activities scale resulted in the creation of five
factors which represented 54.7 percent of the total
variance. Factor 1 represents 23.4 percent of the
variance and is composed of the four print media items,
a music media item and a self as companion item. Thus,
it was named print media since that group was most
represented in the factor loadings. Factor 2 is
comprised of items dealing with friends and others
with whom one is acquainted, but not intimately.

Factor 2 represents 12.1 percent of the variance and is
named friend. Factor 3 accounted for another 7.4

percent of the variance and contained the television

27

and movie items. This factor was named visual media.
Factor 4, which is comprised of the items close friend/
lover and good friends, added 6.2 percent of the

variance and was called intimate. Factor 5 added

5.6 percent of the variance with the items pet and
caregiver. Pets could be assumed to be less close

to people than human interaction, and caregivers such

as doctors are noted for keeping a professional distance
from their patients, so this factor was named impersonal
using Williams' (1984) definition of impersonal
communication as lacking in interpersonal development.
This was felt to be able to include both the person-pet
relationship and the person-caregiver relationship, since
the expectation is that neither would develop on an
interpersonal level within those roles. (The possibility
of exceptions exists, but the factoring would seem to
back up this interpretation because authority figures
were seen as fitting into the friend category where the
potential for development exists, but caregivers did not
load on that factor.

Factoring the emotional involvement with the media
scale resulted in two subscales. The first was called
emotional behavior because the items which loaded on
this factor represented such emotional behaviors as tears
and laughter. This factor was responsible for 42.7

percent of the variance. Factor 2, emotional reaction,

28

added 15 percent of the variance to account for a total
of 57.7 percent of the variance. Emotional reaction was
composed of the items which indicated that viewers who
become emotionally involved with the characters become
angry when those characters are harmed, talk to their
friends about the characters and worry about how things
will work out for the characters.

The willingness to communicate scale factored into
two factors. Factor 1 accounted for 46.1 percent of the
variance and was named stranter because most of the
items indicated willingness to communicate with
strangers. Factor 2 accounted for another 12.8 percent
of the variance for a total of 58.9 percent of the
variance. This factor was named friend because the
greatest number of items which loaded on this factor
involved talking to friends in various circumstances.

Subsequently correlations were computed using the
newly constructed scales and subscales. Several items
in the scales loaded on more than one factor and were
dropped from the correlations to maintain orthogonality.
The results of those correlations confirmed some of the
predicted results and failed to confirm others. It was
expected that the three scales would be correlated with

one another.and they were, however, some of the ways

 

Place Tables 10, 11, and 12 about here.

29

they correlated with one another were somewhat
surprising. The strongest correlations were observed
between the companionship activities scale and the
emotional involvement with the media scale at r=.38,
p-.0001. This relationship was not expected to be
that strong because it was felt that the strength of
the norm of seeking human companionship in the human
companionship items of the companionship scale would
account for the greatest amount of variance and, thus,
that weighting would result in a lowered correlation
with the emotional involvement with the media scale,
despite the expected correlation with the media items
in the companionship scale. Conversely, the modest
relationship between the companionship activities
scale and the willingness to communicate scale of
=.16, p=.008 was much lower than had been expected.
It was assumed that people would use human
companionship if they were able to talk to other
people and that doing so would be given strong
preferential treatment. Quite obviously that
expectation was not met despite the fact that the
scales were positively correlated with one another.

Again, the relationship between willingness to
communicate and emotional involvement with the media
was surprising. Although those scales had been

expected to be inversely related due to the media

30

rather than the human nature of the emotional
involvement with the media items, the observed
relationship was a moderate positive relationship
of r-.21, p=.001. The fact that this was a stronger
relationship than the one between willingness to
communicate and companionship activities was
totally unexpected, but very interesting.

These results were duplicated in the Pearson

Correlations of the media use variables with the

 

Place Table 13. About Here.

 

three scales. Television use was correlated with
the companionship activities scale at r=.17, p=.005
as was expected given the results of previous media
studies suggesting that the use of media for
companionship would result in a higher use of media.
It was not related at all to the emotional involvement
with the media scale or the willingness to communicate
scale. It is not surprising that television use would
not be related to willingness to communicate unless it
was inversely related, but it was expected that it
would be strongly related to emotional involvement
with the media.

There was a modest inverse relationship between
emotional involvement with the media and favorite

viewing time (r=-.12, p=.04) which indicates that there

31

is a slight correlation between emotional involvement
with the media and morning and afternoon television
viewing patterns. Radio use per day was also slightly
correlated with companionship activities at r-.11,
p=.05. See Table 13 for media use correlations.

Age was moderately and inversely related to
companionship activities with r=-.20, p-.002, and
with the emotional involvement with the media scale at
r=-.20, p=.002. However, it was not related to
willingness to communicate or any of the media use
variables.

Gender correlated positively with the companionship
activities scale at r=.12, p=.04. It was also fairly
strongly correlated with the emotional involvement
with the media scale at r=.31, p=.0001. Willingness

to communicate was not influenced by gender. Radio

 

Place Table 14 About Here.

use was related to gender at r=.12, p=.04. And, not
too surprisingly, favorite TV time was related to
gender at r=-.34, =.0001. This is an indication that
women consume more TV in the early times of the day,
and that males consume more TV in the prime time and
late evening hours. Education was only related to one
of the subscales of the emotional involvement with the
media scale, emotional reaction, at r=.12, p=.04.

See Table 14 for the Pearson Correlations of age,

32

gender and education with the other variables.

The real excitement of this study's results lies
in the relationships of the subscales which were created
in the factor analysis. These subscales seem able to
provide explanations for the unexpected relationships
among the scales.

In the relationship between the companionship
activities scale and the willingness to communicate
scale, the predicted correlation between the human
items which loaded on friend and intimate on the
companionship scale were correlated strongly with
the stranger and friend factors of the willingness
to communicate scale, respectively. Evidently the
friend factor of the companionship scale is more
closely allied to the stranger factor of the
willingness to communicate scale at r=.34, p=.0001
than it is to the friend factor of the willingness
to communicate scale at r=.26, p=.0001.

The intimate factor of the companionship activities
scale is related to the friend factor of the
willingness to communicate scale at r=.37, p=.0001.

The relationship with the stranger factor is r=.08, and
it is nonsignificant at p=.117.

Neither of the media factors is related to
willingness to communicate although they were expected

to be inversely related. Nor is the impersonal factor

33

related to any of the willingness to communicate
subscales.

A comparison of the strong correlations between
friend and intimate in the companionship scale and
stranger and friend in the willingness to communicate
scale with the relations between the scales at r=.17
indicates that there is a weakening of that
relationship which is not revealed in correlations
between the other factors. It would be reasonable
if the media factors were inversely related to the
willingness to communicate scale to assume that they
had simply canceled out the effect of the strong
relationships, but this did not occur.

An examination of the relationship between the
emotional involvement with the media scale and the
willingness to communicate scale shows that the
relationship between the scales can be accounted for
by the emotional behavior factor at r=.24, p=.0001.
Emotional behavior is correlated with stranger at
r=.21, p=.0001. It is correlated with friend at
r=.22, p=.0001.

The correlation of the companionship factors with
the emotional involvement factors again yielded some
surprises. It was not expected that emotional

involvement would translate to emotional involvement

34

on an interpersonal level. Print media was related to
emotional behavior at r=.28, p=.0001 and emotional
reaction at r-.l4, p=.02. Visual media relates to
emotional behavior at r=.28, p=.0001 and to emotional
reaction at r-.29, p-.0001.

The inverse relationships which had been predicted
with the other variables failed to materialize. Friend
was correlated with emotional behavior at r=.21, p=.001
and with emotional behavior at r=.l9, p=.002. Intimate
was correlated with emotional behavior at r=.2l, p=.001
and with emotional reaction at r=.12, p=.04. Impersonal
was also related to emotional behavior at r=.16, p=.009
and to emotional reaction at r=.11, p=.05. Even the
weaker associations were significant. The strongest
associations were invariably those with emotional
behavior.

Interestingly, the relationships of the subscales
of the emotional involvement and companionship activity
scales does not translate to an equivalent relationship
with media use patterns. TV use per day is correlated
with print media at r=.13, p=.03 and with visual media
at r=.28, p=.0001. Correlations with friend, intimate
and impersonal are .02, .03 and -.01 respectively.
Radio use per day is correlated with emotional behavior
at r=.16, p=.01 and favorite TV time is related

inversely to emotional reaction at -.18, p=.004.

35

Willingness to communicate is not related to any
of the media use items. The subscale friend is weakly
related to gender at r-.11, p=.05. but not to age or
education. Stranger is not related to any of them.

The companionship activities subscales provide
some more insight. Friend is correlated with age at
r=-.12, p.04. Although a weak correlation, this would
be interesting if it indicates that one's friendships
decrease with age or that the criteria for naming
another friend change. The inverse correlations with
age hold also for visual media at r=—.22, p=.001, with
intimate at r=—.20, p=.002 and with impersonal at
r=—.12, p=.04. Those inverse relationships continue to
hold with emotional behavior at r=-.18, p=.003 and with
emotional reaction at r=-.16, p=.008.

Gender influences the use of print and visual
media for companionship with correlations of r=.12,
p=.04 and r=.14, p=.02 respectively. It also influences
the choice of intimate relationships for companionship
with a correlation of r=.21, p=.001. Gender is also
related to emotional behavior at r=.33, p=.0001 and
to emotional reaction at r=.22, p=.001.

The media factors were far more explanatory than
had been expected. The print and visual media factors

accounted for 30.8 percent of the variance explained

36

in total and 56 percent of the variance accounted for
by the five factors of the companionship activities
scale. If shyness is a predictor of media use patterns
this study shows that the relationship is not a direct
one since TV use and radio use and the media factors

of the companionship scale were not inversely related

to the willingness to communicate scale.

37

CHAPTER IV DISCUSSION

People who are emotionally involved with other people
are also emotionally involved with the media and consume
more media. The results of this study dispute a long
cherished media effects belief that people who are shy
and lonely will turn to using the media to replace
human companionship. This study suggests that the use of
media is not a replacement for human companionship, but
merely an alternative source of human companionship most
often utilized as part of a continuum between deep
intimacy and print media. At all levels of that
continuum, the common denominator is emotional
involvement and particularly emotional behavior if the
use of interaction across the continuum is to be
for the purpose of companionship.

The findings support Zillmann's (1987) theory
which asserts that people will use media to achieve a
personally comfortable level of hedonic valence through
their emotional involvement with media content. This
level of research seems capable of exposing
a process of emotional involvement which may be

generalizable, while allowing for a broad range of

38

idiosyncratic choice-making from among multiple Options
of media content. The researcher contends that on an
affective level of involvement it makes little difference
if the choice of content is the soap opera or the Monday
night football game if the individual consuming that
media content becomes emotionally involved with the
content. The possibility exists, and should be further
tested, that when people feel emotional involvement with
media content they feel the other humans portrayed by
that media content as really hgmgn even if the portrayal
is fictional, and that they feel the connectedness of

a companionship which allows the expression of their
feelings. This evidently may be able to take place in
both males and females despite socialization rules

which limit that expression of emotion. Saarni (1979)
discusses that socialization process and notes that
preschool children are already socialized to know which
emotions may be expressed publicly and under what
conditions. Rubin's (1987) work with soap opera
involvement has demonstrated that many soap operanviewers
do become involved with the characters represented in
the soap opera on a parasocial level. And Tamborini

and his associates (1987) have shown a link between
emotional involvement and empathic reactions to the

content of graphic horror.

39

An increasing body of study is pointing the way
toward an understanding of human involvement with the
media as simply another form of human involvement.

The emotional behavior factor is highly significantly
associated with virtually every form of human
interaction whether that interaction is interpersonal

or mediated. If we can cry or laugh, yell or get angry,
when exposed to media content or human interaction,
something in us becomes connected with something which
is real and which exists in that content. The same
emotional involvement associated with choosing to be
with people is also associated with watching a football
game, listening to the radio, or reading a novel. The
fact that emotional involvement is so broadly associated
with choices of companionship activities is even more
impressive when it is understood that the emotional
involvement scale was constructed to measure emotional
involvement with the media, not with people.

The research question then becomes, not whether
personality variables influence specific content choices,
although that is still an important question in its own
right, but on a much broader scale what are the factors
that either promote or fail to promote emotional
interaction with other humans through interpersonal and

mediated emotional involvement. The magnitude of the

40

results suggest the presence of intervening variables
which may influence whether or not people allow
themselves to become emotionally involved with human
companions and with media companions. One probability

is that human beings read cues from the environment

which indicate the possibility of negative feedback

from others for expressing their emotions. Saarni (1979)
clearly stated that our emotions are self-controlled

and masked to avoid feelings of embarassment, derision

or other negative feedback. Zillmann, Weaver, Mundorf
and Aust (1986) revealed the masking process which
Zillmann describes as response override in their study

of gender-socialized reactions to the frightening content
of graphic horror.

This study is not capable of claiming a causal link
between emotional involvement, choices of companionship
activities and willingness to communicate, however
Anderson (1987) does suggest that "the finding of an
association can be a material element in the development
of a causal explanation." The fact that the correlations
are not .50 and higher suggests that emotional involvement
is not the only factor which is influencing the choice of
companionship activities. Therefore, further research
into other possible factors which influence emotional

involvement and choices of companionship activities is

41

suggested. Additionally, further conceptual work on
defining shyness (Cf. Kagan, 1984, pp. 65-70, on
inhibition, and Kagan, 1978, 1969, for additional
information.) is suggested. Given Saarni's (1979)
contention that children are taught to withhold their
emotional selves as a self-protective measure, it makes
sense to assume that all people are somewhat inhibited
from self-expression. People who are more inhibited

than others may not be different than others in being
able to process environmental stimuli on a sensory

level, but their internalizing of those cues may be
somewhat more sensitive to cues which all people use to
protect themselves from being singled out, laughed at,
embarassed, derided, yelled at, put down, or threatened.
If the fear of reactions from others which make us
uncomfortable is universal it may be one of the universal
processes which define us as human. The fear of negative
reactions from others may be a process which could hold
true for all people and across time as the definition of
what constitutes a negative reaction from others changes
but the process remains intact. It would also hold true
despite differing individual sensitivity to environmental
cues. Therefore it might appear behaviorally as multiple
differing reactions to the same stimuli. (Cf. Ribble,
1987, unpublished paper, for a discussion of differing

reactions to a fear stimulus.)

42

From this research a definition of companionship
must include emotional involvement. If one combines
awareness of the process of emotional masking or
response override within one's definition of
companionship one definition would be that companionship
selection is a choice of an activity within which one
is free to express one's real feelings.

The nearly universal use of the masking of self-
expression to protect one's emotional self from
negative interactions with others suggests also that
our conceptualization of communication may be too narrow
when it is limited to asserting control over our
environment. It appears from this research that it is
possible that the communication of companionship can only
take place in an environment where the need for self-
control or other control is not required. Fisher (1987)
states: "It is one of the supreme ironies of psychology
that the self, the quality that makes an individual a
unique human being, develops through and is probably
impossible to attain without social interaction--
communication with other human beings." This search
for companionship, for that situation in which one
can honestly be self with another, may be one of the
compelling human activities. It may, even more strongly

than the social interaction for physical survival,

43

be the factor that makes humans social beings
dependent upon communication with others not just
for physical survival, but for the survival of the

psychological self.

The term communicative receptivity was chosen as
indicative of the construct which is being explored,
since it appears that one's choice of whether or not
to seek companionship, ie. to be willing to unmask
one's real self, may be influenced by one's perceptions
that receiving and responding to emotional content
from human or media companions will not result in
harm to oneself. That perception may be contained
either within the content itself or in attendant
situational reality. Further research is suggested in

this area.

44

REFERENCES

Anderson, J. A. (1987). Communication research: Issues

and methods. New York: McGraw-Hill

 

Austin, B. A. (1985). Loneliness and use of six mass
media among college students. Psychological
Reports, 56, 323-327.

Burgoon, J. K. & Hale, J. L. (1983). A research note
on the dimension of communication reticence.
Communication Quarterly, 31, 238-248.

Canary, D. J. (1986). Uses of media for the lonely.
Paper presented at the annual convention of the
International Communication Association, May,
Chicago, Illinois.

Cheek, J. M. & Buss, A. H. (1981). Shyness and
sociability. Journal of Personality and Social
Psychology, 41 (2), 330-339.

Daly, J. A. & Stafford, L. (1985). Correlates and
consequences of social-communicative anxiety.

In Daly, J. A. & McCroskey, J. C. (Eds.), Avoiding

communication: Shyness, reticence and communication

 

apprehension. Beverly Hills: Sage.

 

45

Finn, S. & Gore, M. B. (1987). Loneliness, shyness and
social support. Paper presented at the annual
convention of the International Communication
Association, May, Chicago, Illinois.

Fisher, B. A. (1987). Interpersonal communication:
Pragmatics of human relationships. New York:
Random House.

Johnstone, J. W. C. (1974). Social integration and
mass media use among adolescents: A case study.

In Blumler, J. G. & Katz, E. (Eds.), The uses of
mass communication: current perspectives on
gratifications research. Beverly Hills: Sage.

Kagan, J. (1984). The nature of the child. New York:
Basic Books.

Kagan, J. (1978). The growth of the child. New York:
Norton.

Kagan, J. (1969). Personality development. In

Janis, I. L. (Ed.), Personality: Dynamics,
developmentJ and assessment. New York: Harcourt.

Kagan, J. & Resnick, S. (1986). New research shows
inhibited behavior may be inherited. Lansing
State Journal, October 3. 1986. (Gannett News Service)

McCroskey, J. C. (1986). An introduction to rhetorical
communication. 5th Edition. Englewood Cliffs, NJ:
Prentice-Hall.

McCroskey, J. C. (1985). The communication apprehension

46

perspective. In Daly, J. A. & McCroskey, J. C.
(Eds.), Avoiding communication: Shyness, reticence,
and communication apprehension. Beverly Hills:
Sage.

McCroskey, J. C. & McCroskey, L. L. (1986).

Predictors of willingness to communicate:
Implications for screening and remediation. Paper
presented at the annual convention of the
International Communication Association, May
Chicago, Ill.

Mehrabian, A. (1968). Communication without words.
Psychology Today, 2 (4), 53-55.

Perloff, R. M., Quarles, R. C. & Drutz, M. (1983).
Loneliness, depression and uses of television.
Journalism Quarterly, 60, 352-357.

Ribble, M. (unpublished paper). Visual stimuli and
the perception of fear. Michigan State University.

Rubin, A. M. (1983). Television uses and gratifications:
The interactions of viewing patternStnuimotivations.
Journal of Broadcasting, 21 (1), 37-51.

Rubin, A. M. (1981). An examination of television
viewing motivations. Communication Research,

8 (2), 141-165.

Rubin, A. M., Perse, E. M. & Powell, R. A. (1985).

Loneliness, parasocial interaction and local

television news viewing. Human Communication

 

47

Research, lz_(2), 155—180.

Rubin, A. M. & Perse, E, M. (1987). Audience activity
and soap opera involvement: A uses and effects
investigation. Human Communication Research, 14
(2), 246-268.

Russell, D., Peplau, L. A. & Cutrona, C. E. (1980).
The revised UCLA loneliness scale: Concurrent
and discriminant validity evidence. Journal
of Personality and Social Psychology, 32 (4),
472—480.

Saarni, C. (1979). When not to show what you feel:
Children's understanding of relations between
emotional experience and expressive behavior.
Paper presented at the biennial meeting of the
Society for Research in Child Development, March,
SanFransisco.

Tamborini, R., Stiff, J. & Heidel, C. (1987). Empathy
and emotional reactions to graphic horror. Paper
presented at the annual convention of the
International Communication Association, May,
Montreal, Canada.

Williams, F. (1984). The new communications. Belmont,
CA: Wadsworth.

Zillmann, D. (1987). Mood-management: Using
entertainment to full advantage. In Donohew, L.,

Syper, H. & Higgins, T. (Eds.), Communication,

48

social cognition and affect. Hillsdale: Erlbaum.

Zillmann, D. (1983). Arousal and aggression. In Geen,
R. G. & Donnerstein, E. I. (Eds.), Aggression:
Theoretical and empirical views. New York:

Academic Press.

Zillmann, D. (1980). Anatomy of suspense. In
Tannenbaum, P. H. (Ed.), The entertainment functions
of television. Hillsdale: Erlbaum.

Zillmann, D. & Bryant, J. (1986). Exploring the
entertainment experience. In Zillmann, D. &

Bryant, J. (Eds.), Perspectives on media effects.
Hillsdale: Erlbaum.

Zillmann, D. & Bryant, J. (1985). Affect, mood, and
emotion as determinants of selective exposure. In
Zillmann, D. & Bryant, J. (Eds.), Selective Exposure
to communication. Hillsdale: Erlbaum.

Zillmann, D., Weaver, J., Mundorf, H. & Aust, C. (1986).
Effect of an opposite-gender companion's affect to

horror on distress, delight, and attraction.

Journal of Personality and Social Psychology, 51,

586—594.

49

TABLES

49

Table 1. Companionship Activity Scale Items.

Talking with my closest friend of lover.

Reading works of fiction.

Talking with good friends.

Watching TV sitcoms.

Talking with acquaintances.

Listening to my own records and tapes.

. Visiting with parents, siblings or relatives.

8. Watching TV news.

9. Talking socially with coworkers, classmates or
teammates.

10. Going to movies.

11. Watching VCR movies.

12. Talking informally in small groups of friends.

13. Reading the news in papers.

14. Watching TV dramas.

15. Talking at meetings.

16. Listening to D-jays.

17. Reading feature stories.

18. Reading magazines.

19. Spending time with self.

20. Listening to music on the radio.

21. Talking with care-givers.

22. Time spent with a pet.

23. Talking with authority figures.

24. Watching TV soaps.

NOMfiri-I

These items were scored on a scale from
0- never use for companionship to
lO-always use for companionship.

50

Table 2. Emotional Involvement with the Media Items.

1.

I cry when I am watching TV/movies, reading or
listening to music.

I sometimes yell at the TV, radio or newspaper when
something dumb is said.

I allow myself to become emotionally involved with

the characters or action.

I laugh out loud when I am watching TV/movies, reading
or listening to music.

I get angry when one of my favorite characters is hurt
or mistreated.

I talk to my friends about what is going on in the
lives of my favorite characters.

I sometimes worry about what is going to happen next
to characters I like in serials.

These items were scored on a scale from
0- never to 10- always.

 

51

Table 3. Willingness to Communicate Scale Items.

*

G

Present a talk to a group of acquaintances.
Talk with an acquaintance while standing in line.
a

Talk in a large meeting of friends.
*

‘Omfla‘ut‘wNH
sasssssss

Talk in a small group of strangers.

Talk with a friend while standing in line.
10.*

11.Talk in a large meeting of acquaintances.
12.Ta1k with a stranger while standing in line.
130*

14.Present a talk to a group of friends.
15.Talk in a small group of acquaintances.
160*

17.Talk in a large meeting of strangers.

18.*

19.Talk in a small group of friends.
20.Present a talk to a group of acquaintances.

This scale is copyrighted by James McCroskey (c 1985).
McCroskey and Baer (1985) report an internal alpha of .92.
McCroskey and McCroskey report an internal alpha of .91
(1986). ‘

These items were scored on a scale from 0- never to
10-always.

Starred items were filler items not used to compute
the scale.

52

Table4~ Factor Analysis of the Companionship Activities
Scale.

Factor 1 Factor 2 Factor 3 Factor 4 Factor 5

Var 1 .06058 -.O3415 .03917 .76483* .00424
Var 2 .42988* .07611 .12139 -.08552 .10241
Var 3 .06175 .47992 .14139 .60125 -.10170
Var 4 .12050 -.15879 .66482* .13999 .18188
Var 5 .11216 .70919* -.02411 .14219 .00355
Var 6 .38234 .15102 .31909 .25733 .02379
Var 7 -.O4688 .38105 .03350 .44423 .33275
Var 8 .30360 .19255 .52716* -.21512 -.05303
Var 9 .07146 .79607* .04125 .18455 -.08248
Var 10 .23886 .21895 .72760* .02980 -.08074
Var 11 .09480 .19361 .76544* .00731 -.Ol689

Var 12 -.O7038 .70569* .20219 .19136 .09745
Var 13 .64631* .11717 .35019 -.O9006 -.12263
Var 14 .30562 -.12839 .64492* .16851 .21349
Var 15 .18107 .64683* .07841 —.25653 .14153
Var 16 .41293 .39859 .33484 .04988 -.14248
Var 17 .73573* .17836 .15250 -.O9858 .01813
Var 18 .80950* .00652 .25328 .05272 .00677
Var 19 .68175* -.05783 -.12452 .15031 .15488
Var 20 .63475* -.03213 .27614 .33369 .12583
Var 21 -.11575 .37173 .18280 -.O4039 .58499*
Var 22 .23270 .08641 -.10224 -.02206 .64544*
Var 23 .18042 .66720* -.O7216 -.24759 .29240
Var 24 .07016 -.15648 .46470 .16552 .55075

*Starred items were used to construct the five subscales
of the companionship activities scale: print media,

visual media, friend, intimate and impersonal. Items
6, 7, 16, and 24 were eliminated from the scale since

those items loaded on more than one factor.

53

Table 5. Factor Analysis of the Emotional Involvement
with the Media Scale.

Factor 1 Factor 2
Var 1 .53654* .33160
Var 2 .54118* .14033
Var 3 .74110* .00815
Var 4 .80633* .00815
Var 5 .43725 .652745
Var 6 .22415 .76594*
Var 7 .04434 .88799*

*Starred items were utilized to form Factor 1 emotional
behavior and Factor 2 emotional reaction.

Table 6. Factor Analysis of the Willingness to Communicate

Scale.

Var
Var
Var
Var
Var
Var
Var
Var
Var
Var
Var
Var

*Starred items were combined to form Factor 1 Stranger

Factor 1

.77892*
.25707
.33659
.79770*
-.14145
.55763
.57122*
.46954
.53598
.84257*
.12684
.66363*

and Factor 2 Friend.

54

Factor 2

.07366
.51558*
.69867*
.12767
.77166*
.59057
.25204
.54251
.62267
.11030
.78757*
.47111

55

Table 7. Pearson Correlations of the Companionship
Activities Scale.

1

2.
Print
Media

3.
Friend

4.
Visual
Media

5.
Intimate

6.

Impersonal

Table 8. Pearson Correlations
With the Media Scale.

Emotional
Involvement

Emotional
Behavior

Emotional
Reaction

All correlations are significant at p<.OOOl.

Companionship 1.00

.78

.0001

.62

.0001

.75

.0001

.34

.0001

.47

.0001

.0601

1.00
.21
.001
51
.17
.006

.15
.012

Emotional
Involvement

1.

.0601

.0001

.19
.003

24

33

1.00
.19
.002

.16
.008

5 6
1.00

.07 1.00
.166

of the Emotional Involvement

Emotional
Behavior

.00

.52

Emotional
Reaction

1.00

56

Table 9. Pearson Correlations of the Willingness to
Communicate Scale.

Willingness to Stranger Friend

Communicate
Willingness to 1.00
Communicate
Stranger .94 1.00
Friend .75 .48 1.00

All correlations are significant at p<.0001.

Table 10. Pearson Correlations of Willingness to
Communicate with the Companionship Activities Scale.

Willingness to Stranger Friend

Communicate

Companionship .17 .15 .16
.005 .015 .009

Print .08 .09 .03
Media .115 .093 .299
Friend .35 .34 .26
.0001 .0001 .0001

Visual -.O4 -.05 .01
Media .283 .197 .436
Intimate .21 .08 .37
.001 .117 .0001

Impersonal .03 -.OO .07
.352 .478 .142

57

Table 11. Pearson Correlations of Willingness to
Communicate with Emotional Involvement with the Media.

Willingness to Stranger Friend

Communicate
Emotional .19 .17 .16
Involvement .002 .006 .009
Emotional .24 .21 .22
Behavior .0001 .001 .001
Emotional .09 .09 .06
Reaction .092 .089 .209

Table 12. Pearson Correlations of Companionship Activities
with the Emotional Involvement with the Media Scale.

Emotional Emotional Emotional

Involvement Behavior Reaction
Companionship .38 .37 .29
Activities .0001 .0001 .0001
Print .25 .28 .14
Media .0001 .0001 .018
Friend .23 .21 .19
.0001 .001 .002
Visual .33 .28 .29
Media .0001 .0001 .0001
Intimate .19 .21 .12
.003 .001 .039
Impersonal .16 .16 .11
.009 .009 .047

58

Table 13. Pearson Correlations of Media Use with the
Companionship Activities, Emotional Involvement with the
Media and Willingness to Communicate Scales.

Television Use Favorite TV Radio Use
Per Day Viewing Time Per Day
Companionship .17* -.08 .11
Activities .005 .134 .052
Print .13* —.10 .09
Media .030 .065 .089
Friend .02 .01 .05
.409 .435 .228
Visual .28* -.05 .07
Media .0001 .237 .151
Intimate .03 —.O4 .06
.350 .261 .185
Impersonal -.01 -.05 .09
.451 .242 .106
Emotional .06 -.12* .09
Involvement .177 .037 .085
Emotional .04 —.04 .16*
Behavior .264 .292 .011
Emotional .07 -.18* .00
Reaction .158 .004 .481
Willingness to .05 -.05 .02
Communicate .218 .241 .368
Stranger .03 —.05 .03
.323 .233 .315
Friend .08 - 03 -.00

.134 .346 .476

59

Table 14. Pearson Correlations of Age, Gender and
Education with the Companionship Activities, Emotional
Involvement with the Media and Willingness to
Communicate Scales, and with TV Use Per Day, Favorite
Television Time Slot, and Radio Use Per Day.

Age Gender Education
Companionship -.20* .12* -.00
Activities .002 .036 .474
Print -.06 .12* -.03
Media .189 .041 .317
Friend -.12* -.O6 .04
.038 .208 .272
Visual -.22* .14* -.01
Media .001 .016 .426
Intimate -.20* .21* -.O4
.002 .001 .259
Impersonal -.12* .07 .02
.042 .168 .388
Emotional -.20* .31* .04
Involvement .002 .0001 .289
Emotional -.18* .33* —.05
Behavior .003 .0001 .247
Emotional —.16* .22* .12*
Reaction .008 .001 .041
Willingness to .05 .04 .06
Communicate .221 .294 .200
Stranger .07 -.Ol .02
.148 .441 .368
Friend -.00 .11* .10
.487 .048 .067
Television Use -.08 -.O6 .01
Per Day .117 .200 .449
Favorite TV Time .03 -.34* -.11
.347 .0001 .061
Radio Use Per Day -.10 .12* -.00
.979 .042 .439

"lllllllllllllllt