J NWWWWWI 1—

31293 010

 

 

{133091 5;) 0

ABSTRACT

AN EXPERIMENTAL STUDY OF THE EFFECTS
OF COMMUNICATION MEDIA ON
SOURCE CREDIBILITY

By Ray E. Weisenborn

This study investigated the main effects of four
communication media (printed message, oral, visual, and com-
bination) on terminal source credibility. The study also
attempted to discover the additive nature of oral and visual
media stimuli in the production of terminal source credibil-
ity.

Hypotheses tested were: (1) There is no difference
between speaker images formulated for the same communicator
employing either printed message, oral, visual, or combin-
ation communication media. (2) Independently presented oral
communication stimuli and visual communication stimuli pro—
duce the same speaker image as when they are presented si-
multaneously.

A speech pretested for attitude neutrality was
filmed by a speaker unknown to the experimental Ss. The
speaker employed four different oral-physical delivery var-
iations. Introduction of the speaker to all media groups

was constant. Printed message 83 read the message and then

Ray E. Weisenborn

completed semantic differential source credibility scales.
Other media Ss received their stimulus, read the message and
completed the scales.

Hypothesis one was rejected in the dynamism
(p.‘=.0005) and authoritativeness (p.<:.025) source credi-
bility dimensions. Results were not conclusive for the char-
acter dimension because of a non-significant F-ratio (F =
.453, p. <.715).

To test hypothesis two, the effects of the printed
message were removed from the remaining communication media.
Summed differential dimension means for printed message were
removed from corresponding dimensions in the other medias.
The equation CMB-M = (O-M) + (V-M) was produced. This hypo-
thesis was rejected in the dynamism (p. <.02) and authorita-
tiveness (p. <.05) dimensions.

Results of this study indicated: (1) There are dif-
ferences in perceived credibility attributable to the media
through which the message is presented. (2) In two source
credibility dimensions oral and visual communication media
stimuli-are interactive, not additive, in producing terminal
source credibility. (3) The character dimension of source
credibility, if relevant at all, has neither an additive or

interactive effect on terminal source credibility.

Accepted by the faculty of the Department of Speech and Theatre,
College of Comunication Arts, Michigan State University, in partial

fulfillment of the requirements for the Doctor of Philosophy degree.

55 5.5 5
Wector of Thesis //

Guidance Comittee: flflé , Chairman
/.‘ " L‘AXL.‘ “Q. s

4.. AM-ll in -’. V’)/(
I. ‘ CUM

AN EXPERIMENTAL STUDY OF THE EFFECTS
OF COMMUNICATION MEDIA ON

SOURCE CREDIBILITY

BY
I, ix
Ray El Weisenborn

A THESIS

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

DOCTOR OF PHILOSOPHY
Department of Speech and Theatre

1968

ACKNOWLEDGEMENTS

The generation and completion of this writing rests
on many shoulders:

to Dr. James McCroskey's guidance and encouragement,

to Drs. Jerry Anderson, Thomas Conner, David Ralph and
Gordon Thomas' assistance,

to the patience of my colleagues in graduate school,
especially Charles Ertle and James Gardiner,

and most of all, to my wife Donna, for two years of gui-

dance, encouragement, assistance and patience.

ii

TABLE OF CONTENTS

Page
ACKNOWLEDGEMENTS . . . . . . . . . . . . . . . . . . . ii
LIST OF TABLES . . . . . . . . . . . . . . . . . . . . v
LIST OF FIGURES . . . . . . . . . . . . . . . . . . . vi

LIST OF APPENDICES . . . . . . . . . . . . . . . . . . vii

Chapter

H

I. INTRODUCTION . . . . . . . . . . . . . . . .

Genesis of the Research Problem . . . . . . 3
Review of the Literature . . . . . . . . . 5
Ethos or Credibility of the Source 6
Nonverbal Communication of the
Source . . . . . . . . .
Preliminary Studies . . . . . .
Preliminary Study One . . .
Experimental procedure .
Results and implications
Preliminary Study Two . . .
Experimental procedure .
Results and implications
Conclusions . . . . . . . . . .
Hypotheses to be Investigated

C C O O O O O O O
O O O O O O O O O O
O O O O O O O O O O
C O O O O O O O O O
O O O O O O O O O O
H
0‘

II 0 RESEARCH DESIGN 0 O O O O O O O O O O O O O O 22

Hypotheses Investigated
Operational Definitions
Experimental Group . . .
Test Instrument . . . .
Subject Matter . . . . .
Stimulus Preparation . .
Stimulus Presentation .
Data Collection and Analy

l

.. . .. . ..
.. . .. . ..
o o o o o ‘0 O o
.. . .. . ..
.. . .. . ..
.. . .. . ..

a:

01

SS

iii

Chapter
III. REPORTING OF RESULTS . . . . . . . . . . . .

Effects of Media on Source Credibility .

The Additive Nature of Oral and Visual
Communication Stimuli . . . . . . . .

Secondary Measures . . . . . . . . . . .

IV. CONCLUSIONS, SUBJECTIVE OBSERVATIONS, IMPLICA-

TIONS FOR FUTURE RESEARCH AND SUMMARY . .

Conclusions . . . . . . . . . . . . . . .
Analysis of Experimental Group
Composition . . . . .
Analysis of Media Effects
Dynamism . . . . . . .
Authoritativeness . . .
Conclusions . . . . . .
Analysis of Oral-Visual Communication
Media Stimuli Additivity . . . . .
Potential Confounding Variables to
be Considered . . . . . . .
Subjective Observations . . . . .
Implications for Future Research
Summary . . . . . . . . . . . . .

REFERENCES 0 O O O O O O I O O O O O O O I I O O O O

APPENDICES O O O O O O O O O O O O O O O I O O O O O

VITA O O O O O O O O O I O O O O O O O O O O O O O 0

iv

Page
32
32
35
38
40
40
40
42
42
43
43
44
44

49
51

55
59
73

Table
1.

LIST OF TABLES

Dynamism and Authoritativeness Semantic
Differential Cell Comparisons and
Results 0 O O O O O O O I O O O I O O O

Dynamism, Character, and Authoritativeness
Semantic Differential Mean Difference
Scores and Results . . . . . . . . . .

Dynamism, Character, and Authoritativeness
Semantic Differential Means Adjusted to
Constant Message Factor . . . . . . . .

Chi-square Values and Probabilities for
for Secondary Measures . . . . . . . .

Page

34

36

37

39

LIST OF FIGURES
Figure Page

1. Type of Treatment Administered to
Experimental and Control Groups . . . . . . 26

vi

LIST OF APPENDICES

APPENDIX Page
A SEMANTIC DIFFERENTIAL EXPLANATION . . . . . . 59

B SEMANTIC DIFFERENTIAL IN CREDIBILITY

DIMENSIONS 0 O C O O O O O O I O O O O O O 60
C RESEARCH TEST BOOKLET . . . . . . . . . . . . 62
D INTRODUCTION OF EXPERIMENTAL SPEAKER . . . . 67

E ORAL-PHYSICAL DELIVERY STIMULI
DESCRIPTIONS . . . . . . . . . . . . . . . 68

F ANALYSIS OF VARIANCE RESULTS . . . . . . . . 71

vii

CHAPTER I
INTRODUCTION

The study of rhetoric has produced diverse informa-
tion regarding the art of oral discourse. Scholars in the
field have synthesized this material and attempted to in-
struct peOple in the correct, or most effective, method of
Speaking. Researchers have tested aspects of Aristotle's
Rhetoric and proposed new theories about public speaking.
This research has investigated such elements as organiza-
tion, forms of support, reasoning, and the effect of the
source on the impact of the message.

In regard to this last area, concern has not been
solely confined to public address research. The televised
debates during the 1960 Presidential campaign stimulated
numerous research studies. Many of these studies deal with
public attitudes toward Nixon and Kennedy as they were per-
ceived through the television media (Kraus, 1962). In al-
most all instances the investigations at some time concerned
themselves Specifically with the impact of the speaker's
television image on viewers.

The concept of ethos, or source credibility, cen-

ters on the perceived speaker, However, most of the studies

which have concerned themselves with persuasive discourse
from the viewpoint of source credibility have neglected one
of the fundamental aspects of the oral communication pro-
cess. Studies have apparently not been concerned with many
of the effects of a communicator's nonverbal communication.

If we are to believe the conclusions of researchers
who have worked with political oratory, the impact of the
communicator's personality and visual appearance must be
considered as primary factors (Rosenthal, 1964). In the
post-analysis of the 1960 debates "the leaders of both par-
ties seemed to believe that personality and image-making
are the most powerful forces in swaying the electorate"
(Kraus, 1962, p. 162).

The late President Kennedy realized early in his
career that he must be alert to the influence of tele-
vision's visual image (Neubauer, 1967). Many persons felt
that the turning point in Kennedy's race for the Presidency
"came because of the favorable image he projected during
his televised debates with Richard Nixon" (p. 89). "We
wouldn't have had a chance without that gadget," Kennedy
said. "It was television more than anything else that
turned the tide" (p. 89). Research specialists like F.
Clifton White, an advisor for men such as John Kennedy,
discover and create discernable images for an individual.
For a television appearance questions must be asked and

answered. What camera angle suits him best? In what manner

of dress should he be attired? What mannerisms should he
use? White, as a participant on a television panel, stated
that the image must be carried from the television screen
to public campaign appearances (White, 1967).

It is evident that attention is being directed
toward nonverbal speaker images in the broadest sense by
mass media and public relations research. However, for the
most part, the studies provide semantic differential pro-
file images of different political speakers which merely
reveal which politician seems to create the most favorable
image and do not tap the rational for the favorableness.
The same appears to be true in almost all of the current
image research.

The assumption of this writer was, that the verbal

 

and nonverbal stimuli of a communicator are interacting

factors of importance in the production of source credibil-

 

ity.
Genesis of the Research Problem

The image thesis of political oratory states that
candidates and their agents create certain public images of
a man, and, the candidate who casts the most favorable
image gets the most favorable vote (Katz, 1966). Politi—
cians recognize the importance of the image thesis. For
instance, Senator Mark Hatfield (R-Ore.) has pointed out

many facets of the image thesis which need to be considered

by theorist and practioner.
Is it the image the candidate is trying to establish
through the image-building techniques of public rela-
tions firms, promotional campaigns and publicity, and
some of the other things that go with what we call pub-
lic relations? Or, is it the impressions that people
themselves gain from either a casual or uncontrived re-
lationship they have had with the candidate? Not one
the candidate has established, but one they have made
seeing the candidate in a non-political environment
making non-political remarks or comments. I think lots
of things in politics may not be the factual as much as
they may be the impressions that cause people to react
in one way or another (1967).

This writer had previously been interested in deter-
mining the validity of the Senator's final inference. Prior
to interviewing Senator Hatfield, the writer had conducted
survey research in Oregon during August, 1967 (Weisenborn,
1967). This research investigated several basic aspects of
the image thesis through the use of selected demographics
and semantic differential means plotted in profile form.
Summed semantic differential means revealed that there were
differences, for instance, between Democrats' and Republi-
cans' images of Hatfield. Senator Wayne Morse (D-Ore.) was
included as a response stimuli in the survey. Data analysis
showed that numerous demographic groups displayed signifi-
cantly different attitudes toward the two men.

This survey raised more questions than it answered.
This writer became increasingly convinced that the image
thesis was of considerable potential significance in the

study of source credibility. However, a major question

was generated by the research. What factors constitute

the basic elements of the image thesis? In order to obtain
information relevant to this question the writer surveyed
the literature which seemed applicable to an investigation

of speaker images.

Review of the Literature

 

Professional speakers rarely speak to audiences
trained in speech criticism. Their hearers are usually
laymen in this regard. This notion becomes quite meaning-
ful when one realizes that in Brilhart's study (1967) lis-
teners untrained in speech tended to evaluate speakers more
accurately than their messages. That is, there was a sig-
nificantly higher correlation between listener's ratings of
several speaker's credibility than there was between ratings
of message elements. Brilhart's findings warrant further
investigation of the relationship between perceptual ten-
dencies and the focus of listeners.

Giedt (1958) reported data showing that personality
ratings can be altered by raters noting such small manner-
isms as averting eyes or touching the face. The personal-
ities Giedt speaks of are those of image, and may not ne-
cessarily reflect those of the true personality. Thus, an
actor portraying a specific state of neurosis can be ana-
lyzed more accurately than the neurotic simply because he
will present a patterned series of traits for that condi-

tion. The neurotic would not make such fine distinctions,

i.e., his actions would be individualistic. Even though we
may study words and movements, it is difficult to see what

a person is really like (Giedt, 1958).

Ethos or Credibility of the
Source

Personality may be difficult to assess, and in the
same vein, so is the credibility of a communicator. Rosen-
thal (1966) states that oral persuasion is that phenomenon
which "involves a Speaker generating visual and aural cues
which evoke neural and chemical responses from the listen-
er" (p. 115). He also presents the following dichotomy of
the persuasive act:

When a speaker delivers a speech with the intent of in-
fluencing the behavior of his auditors in a particular
direction, we may conceive of the communication er s2
as presenting two distinct objects as potential oci of
listener reaction: . . . Either the image or the mes-
sage will ultimately emerge as the dominating agent of
value activation for the listener and this will deter-
mine the basic character of the persuasive act (p. 119%

McCroskey (1968) divides ethos or source credibil-
ity into three basic categories.

Initial ethos is the image the audience has of a com-
municator before the inception of communication. . . .
Derived ethos is the image of the communicator produced
in tfie minds of the respondents as a result of the com-
munication, including content, presentation and the
circumstances in which the communication occurs. ESE?
minal ethos is the image the audience has of the com-
municator at the completion of the communication and is
produced by the interaction of initial and derived
ethos (pp. 58-59).

The dimensions McCroskey prescribes are similar to

the types of credibility discussed by Berlo, Lemert and

Mertz (1961). The terms "ethos", "speaker image", and
"source credibility" all refer to the same concept. This
writing shall employ the latter term in its discourse.
Source credibility has at least two dimensions, perceived
authoritativeness and character. Within these dimensions,
appearance of the source can affect initial credibility in
oral communication (McCroskey, 1967).

McCroskey (1967) also states that since derived
credibility is produced by the act of communicating, it
seems probable that internal evidence would have a substan-
tial effect upon it. In the same vein, it seems probable
that different channels of communication would have a sub-
stantial effect upon derived credibility.

In the credibility study by Haiman (1949) signifi-
cant differences between mean shifts of opinion were asso-
ciated with who the speaker was thought to be. Haiman
found that college groups tended to place a high premium on
a speaker's competence and that the prestige of a speaker,
as conveyed to a student audience through previous knowledge
of the speaker's reputation and through the chairman's in—
troduction, does influence the effectiveness of his per-
suasion.

It is known that vocal cues are important in the
formation of speaker images (Winthrop, 1956 and Harms,
1960), but evidence regarding the value of visual percep-

tions to source credibility is minimal. The total scope of

source credibility is definitely an area which needs more
investigation. Miller and Hewgill (1965), after doing re-
search in this area, asserted "that further attention to
the physical and vocal cues affecting audience reaction to
a speaker is warranted" (p. 43).1

Research on the effects of delivery on source cre-
dibility could remove all cognitive content from the aural
message. This has been done in audiology research, but its
effectiveness and application seem limited in the area of
general speech research (Sherman, 1954). The main problem
arising from utilization of such a methodology is that con-
tent continuity is lacking as the technique is based on
playing tape recorded speeches backwards. Another study
(Davitz and Davitz, 1959) investigated the ability of Ss to
communicate with content-free speech. By having their ex-
perimental Ss orally project specific feelings by reciting
portions of the alphabet, these researchers found that Ss
could express non-coqnitive content with oral expression
alone. The applicability of the study to the concept of
source credibility is, however, somewhat limited. Research
reported above highlights the importance of the visually
perceived speaker and the Davitz and Davitz study was only
concerned with content-free speech in terms of a communica-

tor's orality.

 

1The reader is referred to Greenberg (1966) and
King (1966).

Nonverbal Communication of
the Source

 

The scientific investigation of observable human
communication stimuli is known as the study of kinesics.
Birdwhistle (1955) defines this science as:

The systematic study of how human beings communicate by
body movements, including facial expressions, shifts in
posture, and formalized gestures (p. 11).

Basic elements in the study of kinesics are kines,
allokines and kinemes. Kines are the smallest movement
positions of a part of the body. Allokines are the differ-
ent ranges of position possible for the kinemic movements.
Kinemes are the various classes of allokines. Although
they are seemingly minute, allokinic movements can be very
important. President Kennedy, for instance, would often
extend his right arm, index finger extended, in a gesture
that made him appear to dominate a conversation (Neubauer,
1967).

With training in kinesics it is possible, by only
hearing a person speak, to determine what simple movements
he is making. Paradoxically, the average person is obli-
vious to the extent to which he communicates nonverbally.
Hildreth (1954) points out that many people who use body
movements which contradict their actual words, reveal that,
consciously or unconsciously, they do not believe what

they say.

 

2The reader is referred to Reusch (1964).

10

In regard to speaker images, kinesics and the aural
aspects of human communication are combined. This joining
becomes the end result of a speaker's delivery. We must
condition ourselves, as Aristotle did, to the fact that
delivery does affect results, especially through its impact
on attitudes and emotions. Thus, delivery cannot be disre-
garded in nonverbal communication research.3 However, this
writing is not concerned with the intracacies of aural com-
munication; but rather, with the importance of the visually
perceived speaker.

The term "nonverbal" has connotations which differ
from this writer's conception of silent communication.
Kinesics speaks only of human physical actions, disregard-
ing whether or not that stimulus is simultaneously produc-
ing another communication stimulus, either of a primary or
secondary nature. In other words, kinesics assumes the
physical stimulus to be the singular primary stimulus while
in fact, it may be a secondary, incidental stimulus.
Throughout the remainder of this chapter, this writer will
continue to review the literature in terms of the defined
denotation of "nonverbal" communication. However, in set-
ting forth the research hypotheses and presenting the de-

sign, this writer will employ the term "visual" in the

 

3The reader is referred to Woolbert (1920), Jer-
sild (1928), Ehrensberger (1960) and Nadeau (1964).

ll

sense normally attributed to "nonverbal" communication.

Senator Hatfield (1967) has recognized the impor-
tance of delivery and highlighted the need to analyze its
physical aspects. He says:

Can you divorce completely the so—called impressions
and images from the substantive part of the candidate
as to what he says? In other words, can you separate
what he says from how he says it? He may say some-
thing in a very clever way that has been devised for
him by a ghostwriter, speechwriter, or public relations
firm. But still, there must be, I believe, a certain
analysis of what he is saying as well as how he is
saying it, how he verbally, and physically, presents
his message (1967).

Interpretation of nonverbal cues, state Thomkins
and Izard (1965), is based, to a large extent, on the na-
ture of the judgement task and increased exposure of judges
to nonverbal stimuli. Their series of socio-economic ex-
periments were set up to probe the ability of untrained,
unprompted judges to relate nonverbal cues to instances of
spontaneous behavior. In one series of these experiments,
judges matched, with 65 per cent accuracy, nonverbal cues
to specific mental and physical states. Conversely, Giedt
(1955) found in the prediction of patient responses to un-
completed sentences that nonverbal cues resulted in a poor-
er than chance correlation. That is, judges had a poorer
than chance correlation when they matched a verbal or
printed message stimuli to a visual stimulus.

A general conclusion from the Thomkins and Izard

(1965) research is that from still and motion pictures,

12

without any sound involved, judges cannot make accurate
statements about true emotional states of interview Ss.
Most information conveyed by still pictures alone is ambig-
uous and judges find it difficult to assess a Ss' true per-
sonality. However, when the test 53 presented pseudo-per-
sonality traits, the judges could correctly identify them.
Such an instance would be similar to the previously men-
tioned one of an actor portraying a state of neurosis.

In attempts to obtain personality ratings and to
identify emotional states, Williams and Sundene (1965) dup—
licated the Giedt methodology (1958). These experimenters
implemented three basic stimuli: visual expression, vocal
expression, and a combination of the two. Although their
findings do not relate to this discussion, their experi-
ments are mentioned because of their manner of testing.

The use of verbal, nonverbal, and combination stimuli has
also been implemented by Greenhill (1957) in communication
research. Further discussion of this method of inducing
experimental stimuli appears in the Preliminary Studies
section of this chapter.

There are many obstacles which must be overcome by
researchers who wish to study the relationships between
nonverbal communication and source credibility. Although
Eckman (1964) states that there is no effective unit of
measurement for kinesics, Krestinger (1962) has developed

an electro-magnetic movement meter which reliably measures

13

gross bodily movement and levels of audience interest. The
Krestinger meter is simply a recording device, and Weaver
and Weaver (1965) specify problems remaining to be overcome.

All of this suggests that, although vocal and visual
cues have the capacity of generating meaning in the
listener, the items of information cannot be expressed
in bits. Moreover, this conclusion seems even more
obvious for visual than vocal cues. Even if it were
possible to record the visual postures and movements

of the talking person and to use the probabilities of
the occurrence of each signal as the basis of conver-
sation into bits, the interpretation of these final
bits would be difficult, if not impossible. Surely,
many of these visual signals would be as vital as the
words, and the remainder would lie on a continuum some-
where between the two extremes. It is difficult to
imagine a way to program a computer or anything else to
do what the nervous system seems to do--to select only
the relevant visual cues for compilation, or to record
all movements and postures and weigh them according to
their relatedness (p. 439).

The studies cited above illustrate the problems
connected with research involving picture cue stimuli.
Basically, in terms of still pictures, a judge's rating
correlations will increase with the increase of cue stimuli
(Giedt, 1958). Research Seems to indicate that only with
the use of movie film will more reliable nonverbal commun-
ication research results be obtained. As Eckman (1964)
states:

In experiments [in] which nonverbal behavior was pre-
sented as a communicative stimulus, and an observer's
response to that stimulus was measured, still pictures,
rather than motion pictures, eliminated any cues from
sequence or movement patterns which are present to the
observer (p. 301).

All of the aforementioned studies represent just a

beginning of the task of learning how important the oral,

l4

and more specifically, the visual image of a speaker is to
the communicative act. Research in the image-making of the
political speaker and its importance has been noted. Stud-
ies in the field of psychiatry indicate the importance of
nonverbal personality research and current communication
research has isolated and defined numerous factors of
source credibility. To date however, this writer is aware
of no research attempting to determine the relationship be-
tween oral and visual communication stimuli as they relate

to source credibility.

Preliminary78tudies

Previously mentioned research conducted by this
writer led him to believe that one of the more significant
problem areas in communication research needing further
investigation was that of a source's nonverbal communica-
tion. Lack of knowledge in this area led to preliminary
research which asked: "Is there a difference between
speaker images formulated by 'trained' judges receiving
visual or oral communication stimuli? Is there a differ-
ence between speaker images formulated by 'trained' and
'naive' judges receiving visual or oral communication

stimuli?"

Preliminary Study One

 

To answer the first of these questions this writer

15

conducted experimental research at Michigan State Univer-
sity in November, 1967. Ss were students enrolled in Com-
munication 320, Nonverbal Communication. The null hypothe-
sis to be tested was: "There is no difference between the
speaker images of Ss sensitized to nonverbal communication
theory who receive either oral or visual communication
stimuli." The term "sensitized" was employed as the SS
were in the process of learning more about nonverbal theory
and could not be qualified as being "trained." However, as
they had been exposed to more theory than a population with
no background in nonverbal communication theory, the study
equated them to trained SS. Speaker image was operation-
ally defined as, "summed semantic differential means."

Experimental procedure.--The experimental cells and
stimuli in preliminary study one were: (1) Combination--
playing of sound movie film; (2) Visual—-playing of movie
film without sound; (3) Oral--p1aying a tape recording of
the movie film soundtrack; (4) Control--completion of the
test instrument with no stimulus. These test stimuli were
implemented on the basis of previously cited research by
Williams and Sundene (1965), Giedt (1958), and Greenhill
(1957). United States Senators from Oregon, Wayne Morse
and Mark Hatfield, were the speakers presented in separate,
three minute film clips.

Testing procedure was identical for all cells. Af-

ter being told they were being asked to participate in a

16

study developing a speech evaluation form and having a se-
mantic differential explained to them (Appendix A), SS
viewed the first film clip. Immediately afterwards they
completed a semantic differential of eighteen source cre-
dibility scale items (McCroskey, 1966b, Berlo, 25. 21.,
1961, and Appendix B). This procedure was repeated for
the second film clip. A factor analysis was obtained from
the collected data and dimension means were obtained by
summing across the scales in each dimension. Means were
obtained on three dimensions for each of the stimuli.

Results and implications.-—From the main hypothesis

 

several sub-hypotheses were developed. Two null hypotheses
were rejected at the .01 level of confidence. (1) It was
found that there was a difference between the oral and
Visual communication stimuli for Hatfield, and (2) that
there were differences between the visual and combination
communication stimuli for Morse. Preliminary study one in-
dicated that there was a trend in the direction of differ-
ences in speaker images of Ss sensitized to the theory of

a specific type of communication media.

The question then raised by this writer was whether
or not the speaker images of judges trained in speech would
be different from those of judges with no training in
speech. Specifically, the question was: "Are there dif-
ferences between all communication media stimuli speaker

images formulated by trained and naive judges?" That is,

17

would the naive judges' image from a combination communica-
tion media stimulus differ significantly from a trained
judges' corresponding image, and would this also hold true

for oral and visual communication media stimuli?

Preliminaryistudy Two

 

A second study was conducted in an attempt to an-
swer the question generated by preliminary study one. The
importance of the second study was the verification of the
assumption that trained and naive judges' speaker images,
derived from the same communication media stimuli, are sig-
nificantly different.

Experimental procedure.--Students enrolled in Mich—

 

igan State University during Winter Term, 1968 were employed
as experimental Ss in the second study. They were arbi-
trarily placed into the same stimulus treatment cells im-
plemented in preliminary study one. The SS were divided
into trained and naive groups. The trained Ss (N = 172)
were students enrolled in upper division speech courses who
had completed a performance course in speech on the college
level. The naive Ss (N=228 were students enrolled in dif-
ferent sections of the basic speech performance course.
Naive Ss participated in the research prior to any oral
performance in the classroom. Stimulus presentation was
identical to preliminary study one. Again, means were

summed across the scales for each of the three credibility

18

dimensions. 2 tests for independent samples were run be-
tween the different cells in each group. Stated as null
hypotheses in equation form, the tests for each Senator
were:

Trained Naive

combination = combination
visual = visual
oral = oral
control = control

Results and implications.--Between the first three

 

of the above four hypotheses, there were no image differ-

ences for Morse. The fourth hypothesis for Morse was sig-
nificant at the .10 level. Only one test achieved signi-

ficance at the .10 level for Hatfield, that being the vis-
ual hypothesis.

This writer's interpretation of the results of pre-
liminary study two produced three assumptions. (1) Within
both trained and naive groups, the varying degree of famil-
iarity to the experimental speakers may have biased the
results. The conclusion of the Haiman study (1949), which
showed that ethos is affected by who the speaker was
thought to be, were supported by preliminary study two.

(2) By using two speakers in a single study, the effects

of sequence interaction on respondents' images is difficult
to control. Preliminary study two did not attempt to con-
trol this variable and this factor may have biased the re-

sults. (3) Message content may bias respondents' images.

19

The Hatfield message was presented in printed form to thir-
teen Michigan State University freshmen and they were asked
to complete a semantic differential after reading the mes-
sage. The differential items were: like-dislike, agree-
disagree, good-bad, honest-dishonest, informed-uninformed
and valuable-worthless. Hatfield was not identified as the
source of the message. The possible range of summed scores
on the differential was 6-42, the hypothetical mean being
24. The mean for the test group was 28.6 with a variance
of 1.87. In this instance message content produced a
summed mean considerably above the presumed neutral point.4
While failing to reject its main null hypothesis,
that there are no differences between trained and naive
judges' images, the study raised additional questions. Is
there a difference between images of a Speaker formulated

by printed message, oral, visual, and combination communi-

cation media stimuli?

Conclusions

 

The previously stated initial assumption of this

writer was: "The verbal and nonverbal stimuli of a commun—

 

4McCroskey, et. al. (1968b), state: "In short, the

assumption that the fiIdpSInt on individual bipolar scales
represents the neutral point of attitude was demonstrated
to be unacceptaBIe. . . . Thus, it appears that although
the midpoint of a single bipolar may not represent the
neutral point in attitude, the midpoint of the range of
possible summated scores across Six bipolar scales does
represent the neutral point in attitude" (p. 245).

20

icator are interacting factors of importance in the produc-

tion of a source credibility."

After conducting survey research, reviewing liter-

ature pertinent to the study of speaker images and conduc-

ting two preliminary experimental research investigations,

this writer synthesized several conclusions regarding re-

search in the area of Speaker images. These are:

l.

Listeners untrained in speech evaluated speakers
more accurately than their messages.

Physical movement is important in assessing Speak-
er image.

Speaker image is produced by both the heard and
seen speaker.

Appearance of the source can affect initial credi-
bility.

Prior knowledge about a communicator can affect
speaker image.

Judges need a continuum of stimuli, such as movie
film, to evaluate the image of a communicator if
that communicator is not observed live.

Messages must be controlled to evaluate speaker
images.

In speaker image research it is not necessary to
employ judges sensitized to specific types of com-
munication media.

In image research judges may be trained or naive

21

in speech.

For all practical purposes, the terms ethos, source
credibility, and speaker image are synonymous. From the
writing above it is obvious that a great deal is known about
this area of theory. The conclusions stated above attest
to that fact. However, a final conclusion cannot be drawn
regarding the relationship between the different types of
communication stimuli and their interactions.

In order to arrive at such a conclusion this writer

conducted the main study underlying this report.

Hypotheses to be Investigated

Based upon the above writing and conclusions, the
following hypotheses were formulated:

1. There is no difference between speaker images for-
mulated for the same communicator employing either printed
message, oral, visual, or combination communication stimuli.

2. Independently presented oral communication stimuli
and visual communication stimuli produce the same speaker
image as when they are presented simultaneously.

The research design employed to test these hypothe-
ses is presented in Chapter II. The results of the hypo-
theses tests are found in.Chapter III. Chapter IV presents
an analysis and evaluation of those results and suggests
implications for further research in the area of speaker

images.

CHAPTER II

RESEARCH DESIGN

In brief, this research involved the presentation
of twelve experimental stimuli to groups of college stu-
dents. A speech pretested for attitude neutrality was
filmed by a speaker unknown to the experimental subjects.
The speaker used four different oral-physical delivery
variations. Twelve groups each received one of these four
stimuli and a thirteenth group served as a control group.
The Ss in all cells had the speaker introduced to them.
The control group read the Speaker's printed message and
then completed a semantic differential on credibility di-
mensions. SS in the remaining cells were divided into
three communication media groups -— oral, visual and com-
bination stimuli. These SS received the particular stimu-
lus, read the message and then completed the differential.

Summed means for the three credibility dimensions,
i.e., dynamism, character and authoritativeness, were ob-
tained for each cell. An analysis of variance between the
media conditions was computed and individual cell compari-
sons were made when the analysis of variance resulted in a

Significant F-ratio. Also, an equation constructed by this

22

23

writer was tested by means of Efratios to determine whether
or not oral and visual communication media stimuli addi-
tively produce terminal credibility. This equation is dis-
cussed further in the "Data Collection and Analysis" por-
tion of this chapter.

From the results of all statistical analysis of

this study certain conclusions and inferences were drawn.

Hypotheses Investigated

 

The following hypotheses were investigated in the
experimental research:

1. There is no difference between speaker images for-
mulated for the same communicator employing either printed
message, oral, visual, or combination communication stimuli.

2. Independently presented oral communication stimuli
and visual communication stimuli produce the same speaker
image as when they are presented simultaneously.

Basically, the hypotheses asked two questions. Do
audience members react differently to a Speaker when they
receive his message through different communication media?
Secondly, are oral and visual communication stimuli addi-

tive agents in the production of terminal credibility?

Operational Definitions
Prior to presenting further details of the research
design the following operational definitions are offered.

Speaker image--Individually summed semantic differ-

 

24

ential scores for dynamism, character, and authoritative-
ness.

Dynamism--Six bipolar semantic differential items
measuring the dynamicness of a communicator (see Appendix
B).

Character--Six bipolar semantic differential items

 

measuring the character of a communicator (see Appendix B).

Authoritativeness--Six bipolar semantic differen-

 

tial items measuring the authoritativeness of a communica-
tor (see Appendix B).

Initial credibility--Perceptions of the experimen-

 

tal 55 based upon the credibility of a source prior to ex—
posure to a communicative stimuli. To obtain a near neutral
level of initial credibility, three introductions were pre-
tested using the semantic differential character and author-
itativeness dimensions described in preliminary study two.
The selected introduction's score was 26.2. All three in-
troductions were above the presumed neutral point of the
measure (24)--26.2, 26.9, and 28.4. The experimental
speaker introduction is contained in Appendix D.

Derived credibility--The image of a communicator

 

in the minds of the respondents which was formed by the act
of communicating.

Terminal credibilitye-Numeric scores obtained by

 

summing scores of six bipolar semantic differential items

for each of the credibility dimensions.

25

Printed message stimuli--A message presented as a
stimulus in the form of a printed speech.

Oral stimuli--A message presented as a stimulus in

 

the form of a tape recorded speech.

Visual stimuli--A message presented as a stimulus

 

in the form of a silent motion picture film.
Combination stimuli--A message presented as a
stimulus in the form of a sound motion picture film.

Active physical--Use of appendage gestures and body

 

movement in public speaking.
Inactive physical--Lack of appendage gestures and
minimal body movement in public speaking.

Fluent ora1--Smooth flowing, articulate, and force-

 

ful delivery.
Non-fluent oral--Uneven flowing, inarticulate, and

hesitant delivery.

Experimental Group

 

Test SS for this study were students at Michigan
State University enrolled in Public Speaking, Voice and
Articulation, and Justin Morrill College's "Inquiry and
Expression" classes. Figure 1 presents the testing

schedule of the research.

Test Instrument

 

The test instrument was a three page booklet. Page

one required the respondent to indicate his sex, year in

26

 

 

 

HMOflmanm o>wuomcw .Hmno pcmsam «sowumcflneoo ma oouv om\m USU

Doommmfi pawn maco maoum “Houucoo mm ooum mH\m Hoa

HMDfimmEQ m>fluom .Hmuo ucmsam coaumcwnfioo mm ooum ma\m Hod

HMOflmhnm m>wuomcw .HmHOIucode «HMDmH> wa mane ¢H\m moa

Hmowmmsm w>auom .HMHOIHSDSHM Lawnmfi> NH ooum vH\m Hoa

HMDflmmnm o>wuom.Jmuo ucosHMIcoc «ammo wa ooum ¢H\m Hoa

Doommos pooh maco maoum “Houucoo ha ovuma va\m Hoa

HMDflmmnm o>wuooaw .Hmuo “GDDHMISOG “Homo NH ovuma va\m HoH

mmmwwofi coon MHSO msoum «Houucoo ma omuoa ¢H\m Hod

amoflmhsm o>wuomcfl .Hmuo ucodam «Homo ma omuoa ¢H\m Hoa

mommmofi Doom maso mnoum “Houucou «a ooum va\m Hoa

Hmoflmanm m>fiuom .Hmuo Damsam «Hugo ma ooum «axm Hoa

Hmowmmnm m>fluom .Hmuo ucoDHMIcoc «aoflumsflnsoo ma mmum MH\m Hoa

HMOflmmnm m>fluomcfl .Hmuo ucosHMIco: «coaumswneou ¢H omua ma\m Hoa

HMDHmazm m>flpom .Hmuo ucmsamncoc «Hmsmfl> ma mvum ma\m Hod

Hmowmhsm m>wuomcw .Hmuo uSDSHMISoc «anamfl> «a oaum ma\m Hod
mumpwHom cowuwpcou 2 (mafia mumo onusou

,mDHdEHum H. II. Haoo I»

 

 

museum Houuaoo can Hmucoafluomxm on pououmwcwapm ucofiumouu mo onus

H HMDme

27
school and number of college speech courses taken or in
progress. The text of the speech from the experimental
film was on page two. Page three was an eighteen item
semantic differential (McCroskey, 1966c and Berlo, 33. 31.,

1961). Appendix C contains a sample test booklet.

Subject Matter

A speech concerning legal and blackmarket adoption
of babies was constructed by this writer. To test for
potential attitude bias, twenty-four undergraduate students
read the message and then completed the semantic differen-
tial described in preliminary study two. With a hypotheti-
cal neutral point of 24, the message scored 23.8. Appendix

C contains the speech employed.

Stimulus Preparation

A non-teaching graduate student in the Department
of Speech and Theatre at Michigan State University was em-
ployed as the speaker in the research.

This writer and the experimental speaker spent
several hours practicing the delivery of four oral-physical
speaking conditions. These conditions were filmed at Mich—
igan State University's Instructional Media Center under
the direction of Dr. Edward McCoy.

The hypotheses of this study were primarily con-
cerned with the main effects of communication media, not

different delivery conditions. However, to strengthen the

28

generalizations made from data analysis, four different
delivery conditions were employed by the experimental
speaker. This writer felt that by having varying degrees
of delivery proficiency presented by the experimental
Speaker in the different communication media, the results
of the research would not be overly influenced by a Speci-
fic value laden delivery style. Thus, four delivery condi-
tions were presented as experimental sub-groups within each
communication media: active physical-fluent oral, inactive
physical-fluent oral, inactive physical-non-fluent oral and
active physical-non-fluent oral.

After the filming, four graduate assistants from
the Department of Speech and Theatre each viewed a single
condition and attempted to objectively describe its delivery
factors. The delivery conditions and the assistants' des-

criptions of them are contained in Appendix G.

Stimulus Presentation

 

During the sixth week of Spring Term, 1968, 253
Michigan State University students participated in the ex-
periment (224 after random exclusion). The schedule of
testing is presented in Figure l (p. 26). All of the Ss
were told that they were going to participate in an exper-
iment attempting to create a new speaker evaluation form.
Then, the semantic differential was explained to them

(Appendix A). The experimental speaker was introduced to

29

the Ss via the pre-tested introduction (Appendix D). In
the oral, visual, and combination conditions the SS were
told to complete the first page of the test booklet and
then stop. The Ss were told what specific condition stim-
ulus they were going to receive, and the stimulus was pre-
sented. SS were then told to turn the page and read the
printed message and then complete the semantic differen-
tial. In the printed message condition SS received the
same orientation to the study, had the semantic differen-
tial explained, were introduced to the Speaker and then

told to proceed through the entire test booklet.

Data Collection and Analysis

The data were coded on computer cards to facilitate
scoring and statistical computations. Within each media.
condition chi-square scores were obtained for "year in
school" (range of 1-4) and "college speech courses taken
or in progress" (range of 1-6). Means for the credibility
dimensions of dynamism, character and authoritativeness
(each with a range of 6-42) were obtained for each media.
Oral, visual, and combination media groups were equalized
in size by random exclusion of Ss. The base of equaliza-
tion was the number of SS in the smallest delivery sub-
group within each media. An analysis of variance for all
medias was computed for the dependent variables—-dynamism,

character and authoritativeness.

30

To test hypothesis 1, all differences between con-
ditions were tested with the E statistic when an overall
F-ratio reached significance. Tests of significance were
run between the printed message media (control) and the.
remaining three media conditions, and, independently be-
tween the oral, visual, and combination communication
medias.

Hypothesis 2, that the oral and visual communica-
tion conditions additively equal the combination condition
did not employ the normal 5 statistic. Rather, testing im-
plemented an equation developed by this writer. This
writer held the following assumption: "'X' topic, plus
prior knowledge of a source, plus message content, plus
oral communication stimulus, plus visual communication
stimulus, equals combination communication stimulus, or,
terminal credibility." Employing the symbol codes for
these factors, the following equation was produced:

Topic + PK + M + O + V = CMB (TC)

The reader will recall that in stimulus presenta-
tion all Ss in oral, visual, and combination conditions
read the printed message after receiving the primary stim-
ulus. Thus, the effect of the message theoretically can be
removed from these media conditions. This was done for
hypothesis 2 by subtracting the printed message means for
each credibility dimension from the corresponding credibil-

ity dimensions in the other media conditions. Thus, the

31

effect of the message upon the Ss' responses to the seman-
tic differential became theoretically constant.

Hypothesis 2 postulated that combination communica-
tion stimulus Speaker images equal oral plus visual commun-
ication stimuli speaker images. Following that thesis,
oral equals combination stimuli minus visual stimuli and
visual media equals combination stimuli minus oral stimuli.
When the constant factor is removed from each condition,
the following equations are produced:

1. (CMB - M) = (V - M) + (0 - M)

2. (V 1 M) (CMB - M) - (O - M)

3. (O - M) (CMB - M) - (V - M)

The difference between the means for the credibil—
ity dimensions in Hypothesis 2 were tested for significance

by means of the following equation:

 

 

 

 

_ + _. _ _
D(O-M) D(v-M) D(CMB-M)
t =
MSw MSw
No+Nv + NCMB

 

2

CHAPTER III

REPORTING OF RESULTS

Chapter I presented two hypotheses that were tested
in this experimental study. This chapter reports the re-
sults of the study under three headings: (1) Effects of
stimulus conditions on source credibility. (2) The addi-
tive nature of oral and visual communication stimuli.

(3) Secondary measures.

Effects of Media on Source Credibility

Hypothesis 1: There is no difference between
speaker images formulated for the same communicator employ-
ing either printed message, oral, visual, or combination
communication stimuli. The hypothesis was rejected in two
of the three credibility dimensions. Results of the analy—
sis of the dynamism dimension scores indicated that there
were differences among the conditions (F = 7.97, p.< .0005);
results of the analysis of the authoritativeness dimension
scores indicated that there were differences among the con—
ditions (F = 3.17, p.< .025); but the results of the anal-
ysis of the character scores did not show that there were

differences among the conditions (F = .453, p.> .715).

32

33

The differences between conditions were tested with
the E statistic in the dynamism and authoritativeness di-
mensions as follows: printed message to oral, visual, and
combination medias; oral to visual and combination medias;
visual to combination media. Table 1 illustrates the re-
sults of these tests.

In both dimensions the oral and combination stimuli
resulted in means that were closer to the hypothetical neu-
tral point than were the printed message and visual commun-
ication stimulus means. The data seem to indicate that a
communicator may be perceived as being more dynamic and
authoritative when the stimulus is only a printed message
or a visually perceived speaker. This assumption appears
to be substantiated most firmly in the dynamism cross con-
dition comparisons. Of the six tests of comparison, only
the test between the printed message condition and visual
condition did not reach the required level of significance.

The authoritativeness dimension of source credibil-
ity produced the same result with regard to the test be-
tween the printed message condition to the visual condition
test. However, the reader Should observe that the oral to
combination and visual to combination tests produced simi-
lar results.

Although it is not possible to statistically com-
pare separate factors that are a product of reliable factor

analysis (Berlo, 33. 31., 1961 and McCroskey, 1966c), ob-

34

TABLE 1

Dynamism and Authoritativeness Semantic Differential Cell
Comparisons and Results

 

 

 

 

 

 

 

 

 

 

 

 

A; Dimension Means
Group N —Dynam1sm Authorita-
tiveness
Printed Message 72 19.27 15.94
Oral Media 48 21.39 19.18
Visual Media 48 19.16 16.50
Combination Media 56 24.42 18.10
A Dynamism
5' 2. Significance.
Printed to Oral 2.12 1.726 .05
Printed to Visual .11 .089 .10
Printed to Combination 5.15 4.386 .0005
Oral to Visual 2.23 1.657 .05
Oral to Combination 3.03 2.337 .01
Visual to Combination 5.26 4.058 .0005
Authoritativeness
U 2 Significance
Printed to Oral 3.24 2.002 .025
Printed to Visual .56 .346 .10
Printed to Combination 2.16 1.110 .05
Oral to Visual 2.68 2.106 .025
Oral to Combination 1.08 .880 .10
Visual to Combination 1.60 1.300 .10

 

 

 

servation of the differences

teresting.

 

between the media means is in-

The data seem to indicate, speculatively, that

the dynamism dimension is more sensitive to stimulus vari-

35

ation than is authoritativeness. Differences on the former
dimension failed to achieve significance in only one com--
parison while half of the differences on the latter dimen—
sion failed to do so. And yet, in all corollary instances,
the means of the dynamism conditions are closer to the pre—
sumed hypothetical neutral point than are those of the
authoritativeness dimension.
The Additive Nature of Oral and Visual
Communication Stimuli

Hypothesis 2: Independently presented oral commun-
ication stimuli and visual communication stimuli produce
the same speaker image as when they are presented simul-
taneously. The hypothesis was rejected on two of the three
credibility dimensions. The test of the dynamism scores
indicated (p.< .02) that oral and visual communication
stimuli were not additive and the test of the authorita-
tiveness scores indicated (p.< .05) that oral and visual
communication stimuli are not additive. The test of the
character scores failed to reject the hypothesis of addi-
tivity, though the observed difference approached signifi-
cance (p.‘<.20). Table 2 presents the main results of
tests on hypothesis 2.

The discussion of hypothesis 1 pointed out that
the printed message and visual communication media stimuli

means were consistently further from the hypothetical

36

TABLE 2

Dynamism, Character, and Authoritativeness Semantic Differ-
ential Mean Difference Scores and Results

 

 

 

 

Statistic
Dimension 5' Error 2 Signifi-
(OfAVx)-CMBx*term cance.
Dynamism 3.14 1.30 2.415 .02
Character 1.32 1.01 1.307 .10
Authoritativeness 2.64 1.23 2.147 .05

 

 

 

 

 

neutral point than were the oral and combination stimuli
means. Recalling the equation: CMB-M=(O-M)+(V—M), indi-
vidual media results for each dimension prove interesting.
Table 3 illustrates the stimuli mean differences for each
dimension after the constant factor of printed message has
been removed.

The data in Table 3 clearly indicate that for the
dynamism and authoritativeness dimensions of source credi-
bility, oral and visual stimuli are interactive, not addi—
tive agents. Because there was not a significant F-ratio
produced by the analysis of variance for the character di-
mension, we may not conclude that the stimuli are either

additive or interactive in this dimension. No significant

relationship between oral and visual communication media
stimuli as they influence the character of a communicator

was observed in this study.

37

TABLE 3

Dynamism, Character, and Authoritativeness.Semantic Differ-
ential Means Adjusted to Constant Message Factor

 

 

 

 

___ Dimensions
Condition DynamISm Char— Authorita-
acter tiveness
Printed Message 19.27 18.12 15.94
Oral Media minus Message - 2.12 .29 3.24
Visual Media minus Message .11 1.10 .56
Combination Media minus
Message ' 5.15 .07 2.16
Adjusted Oi—Vi- - 2.01** 1.39 4.80*

 

 

 

 

*Significant at the .05 level or beyond.

**Significant at the .05 level or beyond.

It should be noted that on the dynamism and author-
itativeness dimensions the combination stimuli consistent-
ly produces a mean further from the presumed hypothetical
neutral point than do the combined independent oral and
visual stimuli. The addition of the communicator's inde-
pendent oral stimulus to his independent visual stimulus
image produced a poorer total image than did his combina-
tion stimulus image alone. The visual stimulus appears to
be an interactive agent tending to off-set the negative
effect of the oral stimulus, but is not powerful enough to
have a significant effect.’

Results on the character dimension of source credi-

bility indicate the exact opposite effects for the inter-

38

action between the oral and visual stimuli. But, it must
be remembered that a significant F-ratio was not achieved

on this condition.

Secondary Measures

 

SS were asked to indicate their sex, year in
school, and number of college speech courses taken or in
progress on the test booklet. These secondary measures'
were employed to ascertain whether or not the results of
statistical tests may have been influenced by biased ex-
perimental group composition. It was desired that the sex
ratio be as close to 1:1 as possible, that the Ss be varied
in their year in school, and that the number of speech
courses they had taken was minimal. Table 4 reports the
Chi-square values for these factors.

The data obtained from Table 4 shows that the over-
all sex ratio was not significantly different. The between
media and year in school Chi-square scores also indicated
no significant differences. In the number of college
speech courses taken or in progress, Chi-square analysis
showed that there was a significant difference between the
number of Ss who had had only one course and those who had
had two or more courses across experimental conditions.

Interpretation of the above results and implica-
tions for further research are presented in the succeeding

chapter.

39

TABLE 4

Chi-square Values and Probabilities for Secondary Measures

 

 

 

 

 

 

Condition Frequencies: Expected/Observed
Variables Message Oral Visual Combin-
ation
Sex
Male 31.5/33 20.5/16 20.5/20 24.4/29
Female 40.5/39 27.0/32 27.0/28 31.5/27
x2=3.24-
Year in School
Freshman» 34.0/32 22.7/26 22.7/23 26.5/25
Sophomore 13.5/12 9.0/ 6 9.0/14 10.5/10
Junior 18.6/25 12.4/13 12.4/ 8 14.5/13
Senior 5.7/ 3 3.4/ 3 3.4/ 3 44.5/ 8
x2=13.7o
Speech Courses
One 35.6/63 36.6/34 36.6/31 34.7/45
Two or more 16.3/ 9 10.9/14 10.9/17 12.7/11
x2=22.42*

 

 

 

 

 

*Significant at the .01 level or beyond.

CHAPTER IV

CONCLUSIONS, SUBJECTIVE OBSERVATIONS,
IMPLICATIONS FOR FUTURE RESEARCH

AND SUMMARY

In the previous chapters we reviewed the literature
pertinent to a study of speaker images and reported the de-
sign of an experimental study and the results of that
study. It is the purpose of this chapter to: (1) present
this writer's objective analysis of those results, (2)
offer subjective comments pertinent to the objective re-
sults of the study, (3) discuss some implications for fu-
ture research in the field of speaker-image-media-research,

and (4) summarize the complete project.

COnclusions

 

Analysis of Experimental
Group Composition

 

Chi-square tests were computed for the data from
the secondary measures to ascertain whether or not the
composition of the experimental groups may have biased the
results of the study. It was hoped that there be no dif-
ference in the ratio of men to women among the various ex-

perimental cells in the study. Of the total number of Ss

40

41

in the study, 43.6 per cent were men and 56.4 per cent were
women. Chi-square analysis indicated no more than chance'
variation in the sex ratio among the various treatment con—
ditions. It was hoped that the Ss' year in school would be
relatively diverse and this end result was also achieved.
Preliminary study two, reported in Chapter I, suggested
that there is no difference between Speaker images formu-
lated by trained and by naive judges. However, it was
hoped that the Ss would have little experience in perfor-
mance speech courses on the college level. The differences
in the number of speech courses taken or in progress on the
college level among the experimental conditions were sig-
nificant (xzp.=<.01). However, it must be remembered that
the experiment was run little more than half-way through a
ten week term and thus, the majority (77%) of the subjects
had not completed one college speech performance course.
The reader can obtain a fairly comprehensive picture of the
composition of the experimental groups by referring to
Table 4 in Chapter III.

The results of the Chi-square computations suggest
that the study was not biased by the composition of the
experimental cells. The study was apparently not influen-
ced by an overloading of males or females and the ratio of
Ss from each academic year was constant throughout the
experimental cells within each communication media stimuli.

Though there was a substantially different proportion of

42

"naive" to "trained" subjects employed in the research, the-
previous experiences of this writer lead him to-believe
that this factor had little, if any, effect on the results

of the present study.

Analysis of Media Effects

Results of tftests between the different communica-
tion stimuli conditions were applicable for only the dyna-
miSm and authoritativeness dimensions of source credibility.
Respectively, these dimensions had F-ratios of 7.97,
p.‘<.0005 and 3.17, p.<:.025. The character dimension of
source credibility had an F-ratio of .453, p.> .715. In-
dividually, and comparatively, the tested dimensions pro-
duced interesting results.

Qynamism.--The only cross—communication stimuli

 

test which did not produce an acceptable level of signifi-
cance was the printed message to visual test. The oral
communication stimulus was significantly different from
printed message (p.‘<.05), visual (p.<:.05) and combination
(p.‘<.01). The visual stimulus condition was significantly
different from combination (p.< .0005) and combination was
significantly different from printed message (p. <.0005).
With five of the six tests producing significant differ-
ences in the dynamism dimension of source credibility, it
is readily apparent that communicators must realize that
different communication stimuli can differentially affect

one aspeCt of their terminal credibility.

43

Authoritativeness.--The cross-communication media
stimuli tests in this dimension of source credibility did
not produce acceptable levels of Significant differences in
the printed message to visual, oral to combination, and
visual to combination stimuli conditions. Tests which did
indicate significant differences were those between oral to
printed message (p. < .025), oral to visual (p.< .025), and
printed message to combination (p. <.05). Though only half
of the tests in the authoritativeness dimension of source
credibility produced significant Erratios, this writer feels
that the admonition in the discussion of the dynamism di-
mension is again appropriate. Communicators must realize
that different communication stimuli can differentially
affect aspects of terminal credibility.

Conclusions.--Although statistical tests between
source credibility dimensions are of dubious validity, some
observations are appropriate. More of the dynamism dimen-
sion cross-communication media stimuli tests achieved
levels of significance than did those on the authoritaé
tiveness dimension. These results suggest that on the
whole, all media stimuli affect dynamism more than they do
authoritativeness in terms of a communicator's source cre-

dibility.

44

Analysis of Oral-Visual
Communication~Media
Stimuli Additivity

 

 

 

Employing the basic formula CMB-M = (O-M)+(V-M)
discussed in Chapter II, the following conclusions were
drawn: (1) Oral and visual communication media stimuli are
not additive in producing terminal credibility in the dyna-
mism and authoritativeness dimensions of source credibili-
ty. (2) Oral and visual communication media stimuli, if
relevant at all, may be additive in producing terminal
credibility in the character dimension of source credibil-
ity.

Potential Confounding Vari-
abIeS to be Considered

 

 

Several factors should be considered when viewing
the results and conclusions of this experimental study.

The introduction used to introduce the experimen-
tal Speaker to the Ss may have biased the results of the
study. This is not likely, however, as the introduction
was pretested for attitude neutrality and scored 26.2 with
a hypothetical neutral point of 24. Also, if we can assume
that Ss' knowledge about a speaker is reflected by Ss on a
semantic differential in the character dimension of source
credibility, then the introduction did not bias the results
as the analysis of variance was not significant in that di-

mension (p.< .715).

45

Delivery types may have been overloaded in one
style for one.or all of the media and this may have biased
the results of the study. This factor should have been in-
operative as random exclusion of Ss used the smallest num—
ber of Ss within each delivery type as its base. Thus, for
the printed message condition there was no random exclusion
as all SS received the same stimulus. Each delivery style
cell was equal in size to all other experimental cells
within its particular treatment condition.

The number of times anvS received the experimental
speaker's message may have biased the results of the study.
Recall that the effect of the message was theoretically re-
moved from the oral, visual, and combination stimuli con—
ditions once. However, even after this constant factor had
been removed, a message residual may have been retained.
For instance, in the oral stimulus condition the Ss heard
and then read the experimental message, and_yet, the con-
stant factor was removed only once. The same situation
would apply to the combination stimulus condition. Only in
the visual stimulus condition was constant factor removal
equal to number of message inputs. The results of this
study are not purported to be in any sense definitive be—
cause of this potential confounding variable.

Even though the experimental message employed in
this research was pretested for neutrality, the content may

have made the experimental Speaker appear to be very author-

46

itative. This effect was not considered in secondary
measures testing and it may have biased the results of the

study.

Subjective Observations

 

As Marshall MacLuen has said, "The media is the
message." That statement and the results of this study
would seem to have some far-reaching implications for the
politician, businessman, educator, and layman alike. The
aforementioned image thesis which Senator Hatfield said
needs more investigating, needs just that. Vance Packard

in The Hidden Persuaders tells us that businessmen-are con-

 

cerned about their communicator images as advertisers. The
advent of educational television has given educators a man-
date to consider the images they may foster. Laymen must
realize that what they read, see, and hear creates in their
minds particular communicator images. With more knowledge
of the "whys" and "hows" of media and delivery interaction,
it is likely that communicators will be able to more.pro-
fitably alter their images than they have previously
thought possible and/or plausable.

The conclusions of this study, in the writer's
opinion, seem to have significant ramifications for com-
municators. One seldom thinks of the different effects his
communication may have on receivers when that message is

presented via different communication media. Results of

47

this study lead this writer to believe that caution, or at
least consideration, should be exercised by a communicator
intending to utilize different communication media.

The results of our tests of hypothesis 2 suggest
that if options were available, a communicator might find
it more advantageous only to be seen then to have his mes-
sage read. From this point of view, politicians should
distribute picture-leaflets rather than printed speeches.
Following the results relating to hypothesis 2 further, it
would seem that communicators might discover a more favor-
able image if they had their messages read, rather than
heard; were seen, rather than heard when communicating; and
heard, rather than having their messages read and then ber
ing simultaneously seen and heard delivering the same mes-
sage.

The significant results relating to hypothesis 1,
that terminal source credibility is affected differentially
by different communication media stimuli, might be attrib-
uted to the effect of the combining of several different
delivery stimuli within each media. This possibility does
exist and this writer would speculate that data obtained
from cross-communication media tests which had employed
distinctly different delivery styles would produce results
showing significant interactions between media types and
delivery styles. For instance, if a communicator's only

stimulus was oral, and he possessed "good" oral and "poor"

48

physical delivery traits, this writer believes that his
auditors might be able to subconsciously perceive the phys-
ical aspects. Conversely, if the communicator were only
seen, and had "good" physical and "poor" oral delivery
traits, the oral factors might somehow influence receivers'
reactions toward him. In other words, visual communication
media stimuli-may carry with it latent cues which have been
generated by oral delivery. This writer feels that the
same would hold true for the oral stimuli--in some manner
it carries with it cues related to physical activity.

Because of the differences between media and the
apparent interaction of oral and visual communication media
stimuli, this writer believes that delivery will have an
interactive effect on types of communication media. He
would also speculate that specific different gestures and
oral inflections have effects quite different than styles
which have few gestures and limited oral inflection.

Oral and visual communication stimuli seem to be
interactive rather than additive in producing terminal
credibility. This writer would question why the oral media
stimulus becomes a negative factor when combined with a
visual media stimulus. Also, why is the off-setting inter-
active effect of the visual media stimulus so minimal?

Even though the assumption of this writer, that behavioral
communication research is erring by not using the visual

stimuli more in its investigations, has apparently been

49

rejected by this study, he still feels that the visual
media stimulus is more important than this study has shown.
There are too many evident physical cues to poor oral de-
livery for them to be sublimated. Physiological reactions
to oral dysfluencies are quite common--casting the eyes up-
ward or snapping one's fingers to recall the "right" word,
an abrupt checking of notes when a word has been mispro-
nounced, biting the lip or flipping the hands during a
thought pause--the list seems endless. Surely these are
cues which make an impression in the minds of receivers,
even if they have only viewed a communicator.

Subjective contemplations on the results of this
study are intriguing. This writer suggests that answers
to some of the media and delivery variables discussed above
should be isolated and identified. This writer believes
that it would be possible to transgress beyond subjectivity
and benefit communicators by discovering what types of fac-
tors precipitate the oral-visual communication stimuli in-

teraction.
Implications for Future Research

The experimental study, results, and conclusions
reported above provide some relevant data for the study of
the effects of different communication stimuli upon the
credibility of a communicator. Two questions were tenti-

tively answered with the following statements: (1) There

so

are differences in perceived communicator credibility at-
tributable to the media through which the stimulus is pre-
sented. (2) In at least two source credibility dimensions,
dynamism and authoritativeness, oral and visual communica-
tion stimuli are not additive in producing terminal credi-
bility.

There is a need for studies of cumulative replica-
tion. For instance, the study of media effects on source
credibility would profit by knowing if the different media
produce constant effects and if media and delivery produce
constant effects.

Several profitable studies could be designed. Such
research might:

1. Study the interaction between different media stim-
uli and delivery styles.

2. Study the interaction between different media stim-
uli and specific delivery features, such as gestures and
oral inflections.

3. Study the interaction between different media stim-
uli, delivery styles and amount of message removal.

4. Study the interaction between different media stim-
uli, delivery styles and message removal when two speakers
are used as experimental stimuli. The possibilities might
include tests of matched and cross-matched experimental
speaker cells.

The four research possibilities stated above would

51

investigate some of the most basic aspects of communication
media and delivery interaction. Another realm could be
opened with the inclusion-of other variables, such as one-
side, two-side arguments in the messages, credibility man-

ipulation, climactic order and similar types of variables.

Summary

Richard Nixon's defeat in the 1960 Presidential
election is commonly attributed to the poor image he pre-
sented in the televised debates with John Kennedy. Whe-
ther or not one accepts this notion, he must realize that;
many political leaders "believe that personality and image-
making are the most powerful forces in swaying the elector-
ate" (Kraus, 1962, p. 162).

Behavioral research in persuasive communication
accepts the position that either the image or the message
will ultimately emerge as the dominant agent of value acti-
vation for the listener. Whichever factor dominates de-
termines the basic nature of the persuasive act. The re-
sults of the study of kinesics suggest that the nonverbal
communication of a source may reveal subconscious motiva-
tions.

This writer's review of the literature pertinent to
the study of speaker images led him to believe that repor-
ted research in the field did not consider two major areas:

(1) Behavioral research employs different communication

52

medias to present stimuli for testing hypotheses regarding
various communication variables without knowing the intra-
cacies of the effects of those medias. (2) Communication
research commonly employs only the oral communication media
stimulus to test hypotheses regarding content and delivery.
This writer felt that in doing so, research was neglecting
the important factor of the communicator's visual communi-
cation.

This writer conducted preliminary experimental re-
search and the major study reported in this writing to
examine the two areas stated above. Two hypotheses were
tested in the major experimental study: (1) There is no
difference between speaker images formulated for the same
communicator employing either printed message, oral, vis-
ual, or combination communication stimuli. (2) Independent-
ly presented oral communication stimuli and visual commun-
ication stimuli produce the same speaker image as when they
are presented simultaneously.

Four delivery styles were presented to Ss through
four communication media--printed message, oral, visual,
and combination of oral and visual. Semantic differential
measures on three source credibility dimensions, dynamism,
character, and authoritativeness, were obtained from exper-
imental Ss after they were introduced to the experimental»
Speaker and had received his message. Chi-square scores

were obtained from the Ss for certain demographic factors.

53

Dependent variable data was subjected to analysis of vari-
ance. Tftests for between media effects were computed
where significant F-ratios were obtained. To test the ad-
ditive nature of oral and visual communication stimuli, t:
tests were also computed.

Results of these analyses suggested the following
generalizations: (1) The composition of the experimental
groups did not appear to bias the results of the study.

(2) Favorability of speaker credibility on the authorita-
tiveness and dynamism dimensions varies with the media
through which a receiver receives a communication stimulus.
(3) In two of the source credibility dimensions, dynamism
and authoritativeness, oral and visual communication media
stimuli are interactive, not additive, agents in the pro-
duction of terminal credibility. (4) On the character di-
mension of source credibility, oral and visual communica-
tion media stimuli, if applicable at all, appear to be nei-
ther additive or interactive in the production of terminal
credibility.

The experimental study was designed to discover
any differences between Speaker images attributable to
varying the media of stimulation and the additive or inter-
active nature of oral and visual communication stimuli.

At present there is an awareness of the different
oral, visual, and nonverbal communicator images. Previous

research had investigated other types of images, from po-

54

litical profiles to psychological personality traits. Re-
search on ethos and studies in the field of psychiatry
suggest the importance of the verbal and nonverbal images.
However, no research had attempted to determine the rela-
tive differences between these images in their relationships
to terminal credibility. It would appear to this writer
that the information gathered by this study was-needed for
communication research to promulgate more effective theory.

This study was designed to answer some questions
in this regard. It does not purport to be all-inclusive.
The writer recognizes that it is but one of the steps nec-
essary to obtain information to answer the questions which
will arise in connection with media-speaker image research.
Only cumulative studies will provide firm answers to the
basic problems approached in this study.

By design the study was limited to testing college
students under favorable conditions. Thus, the results may
not be widely generalized until similar studies employing
different populations support the findings of this re-
search. It is the hope of this writer that more refined
research will be conducted into the variables related to

nonverbal communication and source credibility.

REFERENCES
Allport, G. W. and Cantrill, H. Judging personality from
voice. Journal of Social Psychology, 1934, 5, 37-55.

Benham, T. Polling for a presidential candidate. Public
Qpinion_guarter1y, 1965, 29, 185-199.

Berelson, B. and Janowitz, M. Reader in Public Opinion and
Communication, New York: The Free Press, 1966.

 

Berlo, D. K., Lemert, J. B., and Mertz, R. J. Dimensions
for evaluating the acceptability of message sources.
Unpublished paper based on a paper presented at the
igzich Association of America.Convention, New York,

Birdwhistle, R. L. Background to Kinesics. Etc., 1955,
13’ 10-180

Brilhart, B. L. Speaker message perception and attitude
change as a function of field independence. S eech
Monographs, 1967, 34, 222-223. Doctoral dissertation
abstract, Pennsylvania State University, 1966.

 

Davitz, Joel R. and Lois Jean. The communication of feel-
ings by content—free speech. Journal of Communication,
1959, 9, 6-13.

Eckman, P. Body position, facial expression and verbal be-
havior during interviews. Journal of Abnormal Psych-
ology, 1964, 68, 295-301.

Ehrensberger, R. An experimental study of the relative
effectiveness of certain forms of emphasis in public
speaking. Speech Monographs, 1960, 27, 94-111.

Gardiner, J. C. An experimental study of the use of selec-
ted forms of evidence in effecting attitude change.
Unpublished Master's thesis, University of Nebraska,
1966.

Greenburg, B. S. and Miller, G. R. The effects of low

credible sources on message acceptance. Speech Mono-
graphs, 1966, 33, 127-136.

55

56

Greenhill, L. P. Application of sound motion pictures to
research in clinical psychology. Audio-Visual Commun-
ication Review, 1955, 5, 529-539.

 

 

Giedt, F. H. Cues associated with accurate and inaccurate
interview impressions. Psychiatry, 1958, 21, 405-409.

 

. Visual, content and auditory cues in interview-
ing. Journal of ConsultingAPsychology, 1955, 19, 407-
416.

 

Haiman, F. S. An experimental study of the effects of
ethos in public Speaking. Speech Monographs, 1949, 16,
191-202.

 

Harms, L. S. Social judgments of status cues in language.
Speech Monographs, 1960, 27, BL -Ph.D. dissertation ab-
stract, Ohio State University, 1959.

Hatfield, M. 0. Personal interview by Ray E. Weisenborn,
November 17, 1967.

Hildreth, R. A. An experimental study of audiences' abil-
ity to distinguish between Sincere and insincere
speeches. Speech Monographs, 1954, 21, 146. Doctoral
dissertation abstract, UnIVersity of Southern Califor-
nia, 1953.

 

Jersild, A. Modes of emphasis in public speaking. Journal
of Applied Psychology, 1928, 12, 611-620.

King, T. R. An experimental study of the effects of ethos
upon the immediate and delayed recall of information.
Central States Speech Journal, 1966, 17, 22-28.

Kraus, S. The Great Debates. Bloomington: Indiana Uni-
versity Press, 1962.

 

Krestinger, E. A. An experimental study of gross bodily
movement as an index to audience interest. Speech
Monographs, 1952, 19, 244-248.

 

Levonian, E. The use of film in opinion measurement.
Audio-Visual Communication Review, 1962, 10, 250-254.

McCroskey, J. C. An Introduction to Rhetorical Communica-
tion. New Jersey: Prentice Hall, 1968a.

, Pritchard, V. O. and Arnold, W. E., Attitude in-
tensity and the semantic differential. Public Opinion
Quarterly, 1967-1968b, 4, 242-245.

 

57

, and Dunham, R. E. Ethos: a confounding factor
in communication research. Speech Monographs, 1966a,
33, 456-463.

. Experimental studies in the effects of ethos and
eVIdence in persuasive communication. Unpublished Doc-
toral dissertation, Pennsylvania State University,
1966b, (Abstract: Speech Monographs, 1967, 34, 254).

. Studies of the effects of evidence in persuasive
communication. Speech Communication Research Labora-
tory, Michigan State University, 1967.

. Scales for the measurement of ethos. Speech
Monographs, 1966c, 33, 65-72.

McGrath, J. and M. The effects of partisanship on the per-
ceptions of political figures. Public Opinion Quarter-.
1.x, 1962' 26' 236-2480

Michael, W. and Crawford, C. C. An experiment in judging
intelligence by the voice. Journal of Educational
Psychology, 1925, 18, 107-114.

 

Miller, G. R. and Hewgill, M. A. The effect of variations
in non-fluency on audience ratings of source credibil-
ity. Quarterly Journal of Speech, 1964, 50, 36-44.

Nadeau, R. Delivery in ancient times. Quarterly Journal
of Speech, 1964, 50, 50-63.

 

Neubauer, J. The camera and JFK. Popular Photography,
1967, 61, 88-102.

Reusch, Jurgen and Kees. Nonverbal Communication. Berke-
ley and Los Angeles: University of California Press,
1954.

Rosenthal, P. I. Ethos in the presidential campaign of
1960: a study of the basic persuasive process of the
Kennedy-Nixon televised debates. Speech Monogra hs,
1964, 31, 257. Doctoral dissertation abstract, Uni-
versity of California, 1963.

. The concept of ethos and the structure of per-
suasion. Speech Monographs, 1966, 33, 114-126.

Sapir, E. Speech as a personality trait. American Journal
of Sociology, 1927, 32, 892-905.

 

58

Sherman, D. The merits of backward playing of connected
speech in the scaling of voice quality disorders.
Journal of Speech and Hearing Disorders, 1954, 19,
312-321.

Sigel, R. The effects of partisanship on the perceptions
of political figures. Public Opinion Quarterly, 1964,
28, 483-496.

Thomkins, S. and Izard, C. E. (eds.). Affect, Cognition
and Personality. New York: Springer, 1965.

 

Weaver, C. H. and G. L. Information theory and the measure-
ment of meaning. Speech Monographs, 1965, 32, 244-248.

Weisenborn, R. E. An exploratory study of speaker images.
Unpublished research paper, Department of Sociology,
Michigan State University, December, 1967.

White, F. C. The political image makers. Public Broad-
casting Laboratory. Panel discussion televiSed’On
November 19, 1967.

 

Winthrop, H. Effects of personal qualities on one-way
communication. Psychological Reports, 1956, 324-334.

Williams, F. and Sundene, B. Dimensions of recognition:
visual vs. vocal expression of emotion. Audio-Visual
Communication Review, 1965, 13, 44-52.

APPENDIX A
SEMANTIC DIFFERENTIAL EXPLANATION

The semantic differential is an attitude measuring
instrument. It consists of a concept, such as "Michigan
State University," and below the concept are two columns of
adjectives separated by seven spaces or blanks. The adjec-
tives are bi-polan’in other words, they have opposite mean-
ings, such as, "good" and "bad". You are to indicate how
you feel about the concept for each pair of adjectives.

The middle, or fourth space is neutral, and you should
place an."x" there is you feel neither adjective applies
to the concept or if they both apply equally. When you-
feel that an adjective does apply to the concept, indicate
your feeling by placing an "x" toward or away from the ad-
jective on the scale. For instance, with the concept
"Michigan State University" and the bi-polars "large-
small" you would probably place your "x" in one of the
three spaces between the word "large" and the scale's

neutral point.

59

Aggressive
Emphatic
Forceful

Bold
Active

Energetic

Unselfish
Nice
Friendly
Honest
Pleasant

Virtuous

APPENDIX B

SEMANTIC DIFFERENTIAL IN

CREDIBILITY DIMENSIONS

Dynamism

 

 

 

60

Meek
Hesitant
Forceless
Timid
Passive

Tired

Selfish
Awful
Unfriendly
Dishonest
Unpleasant

Sinful

61

Authoritativeness

 

 

Informed :_:_:_:_:__:_:_: Uninformed
Qualified :___;___5___}___}___:___:___: Unqualified
Reliable :__:_:__:__:__:__:___: Unreliable
Valuable -___;___;___fi___- : : : Worthless
Intelligent : : : : : : : : Unintelligent
Expert : : : : : : : : Inexpert

 

APPENDIX C

RESEARCH TEST BOOKLET

Male Female

Indicate academic standing:

(circle) Frosh. Soph. Junior Senior

Indicate courses taken or in progress:

Basic Speech

Group Discussion
Communication course(s)
Oral Interpretation
Argumentation

Persuasion

62

63

Speech by Mr. Robinson

A few years ago childless couples hopefully turned
to adoption agencies only to be turned away because of the
strict regulations which resulted from the shortage of a-
doptable children. Many of these couples refused to take
a polite "no" for an answer and searched elsewhere for a
child. The result has been a boom in black market babies.
The pressure of black markets has forced the adoption agen-
cies to retreat. They have been widely criticized -- for
piling on red tape, snooping, prying, and for "playing God"
in selecting those who will receive babies.

The demand for young infants has put pressure on
adoption agencies and made it necessary for them to alter
their restrictions. They now realize that couples who will
adopt children independently are willing to take a number
of risks in order to obtain children. These couples not
only take risks by breaking the law, but they also chance
the possibility of thwarting a child's personality. Cou-
ples who want intelligent and responsive chidren must wait
until the child is old enough for psychological tests to
be effective. Granted, an agency is responsible for trying
to find a child for prospective adOptive parents. But,
their main focus is on finding a family for a child rather
than a child for a family. It must be remembered that

those who would sell an infant on the black market take

64

neither the parent-to-be or child into consideration.
Their main aim is the turning of a dollar. Couples who
wish to adopt a child must protect themselves.

It is the job of the agencies to provide homeless
children with good homes and reliable parents. Social
workers must be in constant touch with prospective parents.
The more a person can reveal about the way he functions in
his work with his family and friends, the better the social
worker can predict how he will function as a parent. And,
adOptive parents have a great responsibility. A good adop-
tive parent is someone who can love and enjoy children, is
reasonably happily married, can face life realistically, is
mature, stable, and yet flexible enough to roll with the
punches. A good adoptive parent.must have an extra dimen-
sion. He must be able to accept that his child was born to
someone else, and tolerate the often painful but inalter-
able facts of the child's heredity and background. The
latter points can be adjusted much more readily by adoption
agencies. They have the facilities to match the "best"
child with the "best" parents.

Until recently there has always been a demand for
more babies. But the tables have turned, and now for the
first time in thirty years there are more children to be
adopted than there are people to adopt them. Many factors
contribute to this fact; the rate of illegitimacy has been

on the rise, pe0ple seem to be marrying at a younger age

65

and having "unwanted" children, medical science has made.
infertility a thing of the past. To correct the "parent
gap" adoption agencies will have to remodel their adoptive
laws. If they do not, black market sales of babies are-
sure to skyrocket. They must remove religious barriers,
expand agency facilities, place children at an earlier
age, place more than one child in the same family, concern
themselves less with matching physical characteristics and
insisting on a substantial bank account or ownership of a
home as requirements for adoptive parents. Adoption is a
lifetime decision. A valid adoption decree creates legal
relationships which are the same as if the child had been
born to his new parents. Only legal agencies can safeguard

child and family in this regard.

Aggressive
Hesitant
Forceful

Timid
Active
Tired
Unselfish
Awful
Friendly
Dishonest
Pleasant
Sinful
Informed
Unqualified
Reliable
Worthless
Intelligent

Inexpert

66

ROBINSON AS A PUBLIC SPEAKER

 

 

 

MR.

 

 

 

Meek
Emphatic
Forceless
Bold
Passive
Energetic-
Selfish
Nice
Unfriendly
Honest
Unpleasant
Virtuous
Uninformed
Qualified
Unreliable
Valuable
Unintelligent

Expert

APPENDIX D
INTRODUCTION OF EXPERIMENTAL SPEAKER

Mr. Michael Robinson is a native of Bloomington, Indiana.
Upon graduating from high school he attended Bowling Green
University for one year. During the next four years he

was employed by a youth center in Wisconsin. Mr. Robinson
was then an activities director for the YMCA in Racine,
Wisconsin. He held the job at the YMCA for six years and
for the next three years was on a work-study program for
non-degree holding personnel in Wisconsin's Welfare Depart-
ment. Mr. Robinson was transferred to the Social Work main
office for five years and for the past two years has been a

social worker assistant.

67

APPENDIX E

ORAL-PHYSICAL DELIVERY STIMULI DESCRIPTIONS

Fluent-Oral-Active Physical -- This Speaker's vocal
quality exhibits rich resonance, with lengthy pauses be-
tween sentences. The volume is moderate, and tends to drop
low at sentence endings. The articulation is clear. The
speaker uses many "half gestures" with either hand (some--
times both)--jabbing, pointing, and chopping. He divideS‘
his time between looking at his notes and looking at his
audience. He rocks from side to side, and his facial ex-
pression is always serious.

Fluent Oral-Inactive Physical -- The speaker uses

 

very little variety in all delivery qualities. Vocal em-
phasis patterns, gestures and bodily movements are repeti-
tive and constant throughout the speech. The gestures used
are small expressive hand movements with the arms remaining
in a bent position, seldom being removed from the lecturn.
Bodily expression consists of small nervous head and shoul-
der movements. The speaker appears to be very sincere.

His facial expression and vocal tone contribute to this
image. Throughout the Speech the speaker relies heavily

on his notes and often looks down--eye contact is minimal.

68

69

Non-fluent Oral-Inactive Physical -- Throughout the
speech the speaker was largely confined to notes. At
points where there was an attempt to get away from them and
engage in eye contact, loss of train of thought often re-
sulted. Little in the way of gestures; one major gesture
with the right hand. The left hand used to graSp the lec-
turn throughout the speech. Many non-fluencies noted: they
seemed to consist of (1) loss of place in notes, thus mix-
up of wording, or (2) search for the appropriate word,
lending some indication that he was attempting to be extem-
poraneous. Speaking voice largely uninteresting. Tone
tended toward a single pitch level with the inflection pat-
tern varied little. Toward the end of the speech the Speak-
er seemed to be more relaxed, have more inflection and an
extemporaneous tone.

Non-fluent Oral-Active Physical -- The speaker used

 

frequent hand gestures accompanied by frequent body move-
ments from side to side. Eye contact with the audience was
infrequent. The speaker had his eyes on his manuscript
during the major portion of the speech. Several times dur-
ing the speech the speaker lost his place while Speaking
and became verbally non-fluent, injecting vocal pauses--
"uhm's"--until he again found his place and went on with
the speech. The Speaker's voice followed a pattern of be-
ing loud, then soft. The speaker's verbal delivery also

followed a very broken up pattern of intonation by pausing

 

70

in the middle of sentences as well as between sentences of
a Single thought pattern. The speaker mispronounced sever-
al words as well as not vocalizing the endings of several

other words.

 

 

APPENDIX F

ANALYSIS OF VARIANCE RESULTS

 

J
r

 

 

 

Dynamism
Category Mean Frequency Sum of Standard
Squares Deviation
Printed Message 19.27 72 29028.0 5.654
Oral Media 21.39 48 25139.0 8.206
Visual Media 19.16 48 19652.0 6.553
Combination Media 24.42 56 35530.0 6.196
(over-all) 20.99 224 109349.0 6.896
Source of Variance Mean F Sta- Approx. Significancev
Square tistic Probab. of F Statistic
Between Categories 346.896 7.977 .0005
Within Categories 43.483
Character
Category Mean Frequency Sum of Standard
, Squares Deviation
Printed Message 18.12 72 25063.0 4.456
Oral Media 19.22 48 19567.0 6.220
Visual Media 18.41 48 17286.0 4.625
Combination Media 18.48 56 20715.0 5.369
(over-all) 18.51 224 82631.0 5.124
Source of Variance Mean F Sta- Approx. Significance
Square tistic Probab. of F Statistic
Between Categories 11.985 0.453 .715
Within Categories 26.454

 

71

72

 

 

 

 

Authoritativeness

Category Mean Frequency Sum of Standard
Squares Deviation

Printed Message 15.94 72 20366.0 5.388
Oral Media 19.18 48 20493.0 7.747
Visual Media 16.50 48 14414.0 5.351
Combination Media 18.10 56 20692.0 6.510
(over-all) 17.29 224 75965.0 6.328
Source of Variance Mean F Sta- Approx. Significance

Square tistic Probab. of F Statistic

Between Categories 123.504 3.174 .025
Within Categories 38.911

 

 

VITA

Ray Edward Weisenborn was born on July 10, 1941 in
Portland, Oregon. After being graduated from Battle Ground
High School, Battle Ground, Washington in June, 1959, he
attended Pacific Lutheran University, Tacoma, Washington
for one year. The Associate of Arts degree was granted to
him by Clark College, Vancouver, Washington in June, 1961.

During two years of study at Portland State College,
Portland, Oregon, he was active in student activities and
the Speech and Drama Departments. The Bachelor of Science
degree was awarded him in June, 1963. During the following
year he taught English, speech, and drama at Sutherlin High
School, Sutherlin, Oregon.

Prior to beginning graduate studies at Eastern New
Mexico University, Portales, New Mexico, he was a resident
actor in that institution's Corrall Playhouse. The Master
of Arts degree was awarded to him in Speech and Drama by
that institution in June, 1965. During the academic year
1965-66 he was an Instructor in the Speech and Drama De-
partment at Clark College.

Mr. Weisenborn began doctoral studies at Michigan
State University in September, 1966. He was a graduate

teaching assistant during the 1966-67 academic year and

73

 

74

during 1967-68 was the Administrative Assistant for the
Speech 101 program, under the direction of Dr. David C.
Ralph. Mr. Weisenborn was also an Instructor in Justin
Morrill College during the 1967-68 academic year.

Mr. Weisenborn is a member of the dramatic honor-
aries Junior Collegiate Players and Alpha Psi Omega and the
speech honorary Pi Kappa Delta. He is an active member of
the Speech Association of America, National Education Asso-
ciation, Association of Higher Education, American Associa-
tion of University Professors and the Benevolent and Pro-
tective Order of Elks.

Mr. Weisenborn is married to the former Donna Marie

Graves of Vancouver, Washington.