r .t’ ti . ‘ “pgv’”q.-muM.—.uav-— clip”.-

\ "4

THE EFFECTS OF MESSAGE INTENSITY 0N
EVALUATIONS OF THE SOURCE,
THE MESSAGE AND THE TOPIC

THESIS FOR THE DEGREE OF M. A.
MICHIGAN STATE UNIVERSITY

- ‘ 1mm JOSEPH KOOHEVAR
r i 19 6 7

 

 

I I! um; Imam I III I :21: In gym; I m; I ’

 

I‘t' “ '
§-«~I' ‘
< ; .4'” A!"
u" y 7: if? a” .273 gt- ,1
fl ’ ‘
. . -~ ,- ’C‘ t": I“)
_ [a . ‘_\(-"“ )‘9-‘V
£31k AL 5.4.: 2 "3G ‘
‘ I {11" {Fat
TI ""

I“

J T
A
a}: '-
1 —’ 'n ' —
(.mm-‘-“

 

 

ABSTRACT
THE EFFECTS OF MESSAGE INTENSITY ON

EVALUATIONS OF THE SOURCE, THE MESSAGE
AND THE TOPIC

by John Joseph Kochevar

This study investigated the effects of variations in message
intensity on receiver attitudes toward the message, the tapic, and the
source of a persuasive message. Message intensity was conceptually
defined as the degree to which various linguistic assertions in a
message deviated from neutrality. It was operationalized by the
substitution of previously rated high and low intensity verbs, adverbs,
and adjectives into a basic message. The difference in judged
intensity between the high and low intensity message conditions was
significant at the .01 level.. Hypothesis 1, which predicted that the
high intensity message would lead to more attitude change than the low
intensity message, was not supported. Hypothesis 2, which predicted
that the high intensity message would lead to more polarized evaluations
of the message source, was not supported. Hypothesis 3, which predicted
that a high intensity message would lead to more polarized evaluations
of message characteristics, was not supported. Some evidence indicated
that the relationship between manipulated intensity and the dependent
variables was not a direct one, and may have been confOunded.by both

extraneous variables and degree of perceived intensity.

 

THE EFFECTS OF MESSAGE INTENSITY ON
EVALUATIONS OF THE SOURCE, THE MESSAGE

AND THE TOPIC

By

John Joseph Kochevar

A THESIS

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

MASTER OF ARTS
Department of Communication

1967

 

Accepted by the faculty of the Department
of Communication, College of Communication Arts,
Michigan State University, in partial fulfillment of

the requirements for the Master of Arts degree.

1

I a
I I
. // \/,
I)? (1' 1' “(II( / I"I(C("’A if
/ 'J A

Director of Thesis {j

 

 

ACKNOWLEDGMENTS

The writer gratefully acknowledges his adviser, Dr. Bradley S.
Greenberg, not only for much needed advice and criticism, but also for
the gentle prodding without which this thesis may never have been
attempted, let alone completed. The writer also wishes to thank the
other members of his committee, Dr. Erwin Bettinghaus and Dr. Gerald
Miller.

Special thanks to Mrs. Shirley Sherman for.typing this thesis
and keeping track of the many details required for its final preparation.

Finally, the writer extends his deepest gratitude to his wife

and friends without whose help he could not have gotten along.

ii

 

TABLE OF CONTENTS

CHAPTER
I INTRODUCTION . . . . . . . . . . .

Prior Research . . . . . . . . .
Hypotheses . . . . . . . . . .

II mTHOD Q o . O O O O O O O O O O 0

Overview . . . . . . . . . . . .
Independent Variable . . . . . .
Perception of Message Intensity
Dependent Variables . . . . . .
Message . . . . . . . . . . . .
Subjects . . . . . . . . . . . .
Design . . . . . . . . . . . .
Procedure . . . . . . . . . . .

III RESULTS 0 O O O O O O O O O O O 0

Ratings of Message Intensity . .
Attitudes Toward the Topic . . .
Source Evaluation . . . . . . .
Message Evaluations . . . . . .

IV DISCUSSION . . . . . . . . . . . .
Summary . . . . . . . . . . . .

Conclusions . . . . . . . . . .
Discussion . . . . . . . . . .

Implications and Research Extensions

BIBLIOGRAPHY O O O O O O O O O O O O O ‘ O

.111

M

12

12
12
13
1a
15
16
16
17

19

19
19
22
2M

29
29
29
30
35

#0

 

LIST OF TABLES

Ratings of Message Intensity . . . . .
Attitudes Toward Toothbrushing . . . .
Mean Deviations of Source Evaluations

Source Evaluations - Absolute Scores .
Mean Deviations of Message Evaluations

Message Evaluations - Absolute Scores

iv

23

25

26

28

 

 

APPENDIX A:

APPENDIX B:

APPENDIX C:

APPENDIX D:

LIST OF APPENDICES

HIGH INTENSITY BOOKLET

LOW INTENSITY BOOKLET

CONTROL BOOKLET

ATTITUDE MEASURES

u2

52

62

72

 

CHAPTER I

INTRODUCTION

In 1955, Osgood, Saporta and Nunnally (1955) reported a new content
analytic technique called evaluative assertion analysis. This technique
yielded reliable judgments of the meaning intensity of various verbs and
adverbs. Using some of these previously rated adverbs and verbs, McEwen
(1967) was able to improve attitude change predictions from Osgood and
Tannenbaum's (1955) congruity model. Instead of simply allowing the value
of the assertion to be associative (positive) or disassociative (negative)
as Osgood and Tannenbaum had done, McEwen manipulated the degree of
assertion intensity. Using this modification, he demonstrated that when
an assertion was of the same degree of favorableness toward the tapic,
the congruity model predicted attitude change toward the topic more
accurately than when the assertion was of a different intensity. Variations
in assertion intensity did not improve prediction of source evaluations.

This present study extended McEwen's procedures to a somewhat more
complex situation. Specifically, high and low intensity assertions were
substituted into a basic persuasive message to create high and low intensity
persuasive messages. Receiver reactions to these variations in message
intensity were then measured not in terms of the source-concept dyad but
in terms of specific attitudes toward the message, the topic, and various

qualities of the source.

 

Message intensity is here defined as the degree to which the
assertions in a message deviate from neutrality. This conceptualization
is similar to Bowers' (1963) definition of "language intensity" as
"the quality of language which indicates the degree to which the Speakers
attitude toward a concept deviates from neutrality" (p. 3H5). In each
definition, intensity is regarded as a subjective element, with the baseline
neutrality a function of each individual judge. Of course, the concept of
neutrality refers only to evaluative relations. A completely neutral
assertion would imply no relation at all -- a psycholinguistically impossible

situation.

Prior Research

 

Early content analytic studies conceptualized intensity in terms of
"emotionalism" and "sentimentalization" (Waples 8 Berelson, lQul) or
referred to the "strength" or "excitement value" (Jacob, 19u2) with which
the communication was made. Except for the technique of "Evaluative
Assertion Analysis" developed by Osgood, Saporta, and Nunnally (1955),
most content analytic approaches were used to analyze Specific types of

war-time propaganda and cannot be generalized to other content because of

. the confines of categbry specification.

For example, Naples and Berelson (1941) compiled a list of "emotional"
terms applying to war-time propaganda and analyzed the relative frequency of
such terms per unit of content. For the analysis of atrocity propaganda,
Jacob (1942) developed a "scale of atrociousness," based on group judgments

of the abnormality of specific acts or events.

 

Wright and Nelson (1939) analyzed American newspaper opinion toward
China and Japan by having judges rate newspaper statements according to
the amount of hostility they perceived. From these judgments, the investi-
gators developed a "hostility scale" and rated relative intensity according
to the number of statements occurring at different points on this scale.

Osgood, Saporta, and Nunnally (1955) also used judges' ratings to
assess intensity but their method was generalizable to many types of content.
(higinally, their "Evaluative Assertion Analysis" was developed as an
indirect method for measuring communicator attitudes. Osgood (1959) stated,
Immever, that: "it might yield predictions as to the effects on receivers
letter than inferences as to the intentions of sources" (p. 52). Following
'Hfis suggestion, this experiment employed verbs and adverbs previously
rated by judges using evaluative assertion analysis to generate high and low
intensity messages.

The assertion was the principal linguistic construction-used.by
Osgood, Saporta, and Nunnally to index intensity. It consisted of an actor
(usually a noun) associated with, or dissociated from, a complement via a
verbal connector. An example of an assertion would be the simple sentence:
"Bill is a student." Here, "Bill" is the actor, "student" the complement,
and "is" the verbal connector. By masking the significant concepts in the
message, coders were able to judge the intensity of these connectors and
their modifiers with high reliability and.not be influenced.by their
attitudes toward the concept. From this analysis, Osgood (1959) has listed
a number of connectors and modifiers which vary in the intensity or degree

to which they associate or dissassociate actor and complement. For example,

 

strong intensity was carried by the verb "to be" (X is a.y); weak intensity
was carried by connectors which imply only possible or hypothetical
relation between actor and complement (x may commit, might agree with,
ought to join y). High intensity was implied by the indexing adverbs
"absolutely, definitely, and positively; low intensity by slightly,
occasionally, and somewhat."

Experimental investigation of the effects of differing intensity
messages by Bowers is more relevant to this study than are earlier content
analytic approaches. Bowers (1963) examined the effects of high and low
persuasive Speeches on subjects differing in degree of social introversion.
The high and low intensity Speeches were generated in fbur steps:

1. Bowers removed from four persuasive Speeches all the words

which, in his opinion, indicated an attitude toward the topic
of each Speech. I

2. The words and phrases "indicating attitudes" were replaced
by numbered blanks.

3. Judges rated the intensity of a series of words and phrases
(also chosen by Bowers) which could.be appropriately
substituted for the numbered.blanks.

u. Four high intensity Speeches were produced by sub-
stituting words with the highest intensity ratings into
all the blanks.‘ The low intensity Speeches were created by

substituting all low intensity. words into the blanks.

\

 

 

 

Subjects' attitudes toward the topics and Speakers were measured before

and after exposure to the persuasive speeches.

The following six hypotheses were tested:

1.

3.

H.

5.

An interaction of language intensity in Speeches and social
introversion levels in listeners results in a higher degree
of persuasibility among introverts measured by attitude
change toward the concepts, than can be accounted for’by
the two effects considered separately (Results: No
significant interaction).

Introverted listeners to argumentative Speeches change

more in their attitudes toward the advocated concepts than do
those in a middle group; likewise, the middle group is more
persuasible than are the extroverts. (Results: opposite to
those predicted. A significant F was found for the main
effects of social introversion and the difference between

the introvert and extrovert groups was significant at .05.
Extroverts, however, changed more than introverts.)

Speeches using highly intense language produce greater attitude
change toward concepts than do those using language of low
intensity. (Results: the only significant F was found for
the interaction between intensity and direction of intended
influence.) - '

An interaction of language intensity in speeches and social
introversion levels in Speakers results in a higher degree .
of persuasibility among introverts, as measured by attitude
change toward the Speakers, than can be accounted for by the
two effects considered separately. (Results: no significant
interaction.)

Introvert listeners to argumentative Speeches change more in
their attitudes toward Speakers than do those in a middle group;
likewise, the middle group is influenced more than are the
extroverts. (Results: analysis revealed an interaction
between social introversion and combination of speeches.
Secondary analysis revealed that there were shifts in attitudes
for only two particular sources. The middle group changed
positively toward two sources but introverts and extroverts
changed negatively. The difference between introverts and
extroverts was not Significant.)

 

6. Speeches using highly intense language produce greater
attitude change toward Speakers than do those using language
of low intensity. (Results: No significant effect due to
intensity variation.)

Thus, most of the statistical tests of these hypotheses either failed
to reach significance or were the reverse of what was predicted. The
only significant relationship involving language intensity was the inter-
action between language intensity and direction of intended influence.
Specifically, low intensity Speeches caused more attitude change than
high intensity speeches for those Speeches attacking concepts favored
by the receivers. Intensity did not have any effect on attitudes toward
speakers who were labeled as a well known college president, a Democratic
senator, a prominent home economist, and a university president.

Later,» Bowers (1964) examined some of the correlates of intensity
by analyzing the words used to generate different intensity speeches in
his earlier study. Four variables were positively related to judgments
of intensity: number of syllables; obscurity; presence of qualifiers;
and metaphorical quality.

A correlation (biserial r = 1.0) between metaphorical quality and
judged intensity led. to another experiment, this time. using only metaphors
as an intensity manipulation (Bowers 8 Osborn, 1966). The investigators
compared the effects of persuasive Speeches concluding with intense
metaphors to speeches which concluded literally. For each Speech, the
metaphorical conclusion resulted in more attitude change toward both

speaker and topic. (Attitude change toward Speakers varied in direction

depending on the Specific Speaker and topic.

 

 

 

rem ,M-

The results of Bowers' two experimental studies were not very clear.
First, his experimental designs were too complicated and led to results
which could not be interpreted unambiguously. In his first experiment,
for example, four different topics, four different Speakers and two
different directions of advocacy were all varied at the same time.
Similarly, varying topics and Speakers in his last experiment resulted in
a complex interaction that could not be clearly explained. Second, Bowers
did not obtain subject ratings of the intensity of the messages. Without
such a check, one cannot be sure that the observed effects were the result
of intensity manipulations. Third, because Bowers alone chose the words
and phrases which indicated attitudes, as well as those to be substituted,
it would be difficult for other investigators to replicate his messages.
Finally, the use of recorded speeches may have introduced confounding
Variables into Bowers’ intensity manipulations. Intensity in Spoken
language may be indexed by variables such as volume and inflection as
Well as Specific content. There is no evidence that these variables were
cOntrolled or that subjects were instructed to pay closer attention to
céntent.

This present study used written messages with only one topic, one
s“Durce, and one direction of advocacy. Furthermore, the effect of the
intensity manipulation‘ on dependent variables was directly assessed.
Pinally, words found by evaluative assertion analysis to differ in intensity
“ere used to generate high and low intensity messages. This procedure can
be easily replicated by other investigators. These four changes -- written

Messages, simplified design, direct measurement of the independent variables,

 

and reproducible procedures -- allowed fairly direct tests of the ex-

perimental hypotheses described below.

Hypotheses

 

The variable "message intensity" was conceptualized in terms of the
degree of deviation from neutrality. It was operationalized by the
substitution of adjectives, adverbs and verbs previously rated as relatively
high and low in intensity into a basic message. Three effects were examined:
1) persuasive impact of the message; 2) evaluation of the source of the
message; and 3) evaluation of certain message characteristics.

The hypotheses tested were:

Hypothesis 1: A high intensity message will result in more

attitude change in the direction advocated than
will a low intensity message.

 

In this experiment, the high intensity persuasive message made
Stronger assertions about the topic and advocated more definite changes than the
low intensity message. To the extent that subjects perceive these re-
lationships as they read their message, they should also perceive the
position of the high intensity message as more discrepant from their
beliefs than the low intensity message. In this particular case, it is
predicted that the greater discrepancy will result in mere attitude
change toward the position advocated by the message.

Hovland's (l959) generalizations about the relation between amount
of change advocated.and actual change support this first hypothesis. He
noted that under conditions where there is some ambiguity about the

credibility of the communicator, the greater the attempt at change, the

-‘_

 

higher the resistance. On the other hand, with reSpected communicators,
the greater the discrepancy between the subject's position and the one
advocated, the greater the change. Hovland also noted that for
relatively noninvolving issues where expert Opinion is highly relevant
such as health topics and science, advocacy of positions quite discrepant
from the individual's own position will have a marked effect. These

_ generalizations were supported in studies by Hovland and Pritzker (1957),
Goldberg (1951+), and Zimbardo (1960).

The source of this message was identified as a writer for a major
metropolitan daily newSpaper. While this source was not of the highest
possible credibility, it was believed that this amount of credibility
would be high enough ‘to induce attitude change and still avoid in-
credulity effects. Furthermore, using a moderately credible source allowed
subjects some latitude for evaluation and avoided ceiling effects some-
times caused by high credible sources.

This eXperiment used a message attacking the merits of tooth
brushing. Previous studies (e.g., McGuire, 1961) have shown this message
to be highly persuasive. While generalizability to other topics is some-
what hindered by use' of a cultural truism, the advantages of using such
a message were felt to be greater than the disadvantages. For example,
the anti-toothbrushing message closely fit Hovland's generalization. It
was a health topic, and relatively noninvolving. Under these conditions,
differences in discrepancy Should be enhanced. That is, subjects

reSponding to the low intensity message would find it less discrepant

 

 

10

with their beliefs than subjects in the high intensity condition. Because
the toothbrushing topic would be relatively non-involving, attitude change
should vary directly with discrepancy.

Hypothesis 2: A high intensity message will result in

more polarized evaluations of the source's
credibility than will a low intensity message.

Recent research indicated that people often form impressions of
communicators from their messages. Miller and Hengll (196“), for
example, feund that as the number of verbal nonfluencies increased, ratings
of the speaker became less favorable. In another experiment, Razinsky
(1967) found that as errors in spelling, punctuation and grammar increased,
source evaluation decreased.

While this evidence demonstrated that people form impressions of
communicators from their messages, it is more difficult to predict the
favorableness of these evaluations. Experiments by Bowers 8 Osborn
(1966) and Eagly and Manis (1966) showed that attacking persuasive messages
caused attitude change in the direction advocated but negative changes
toward the source of the message. Because the message used in this proposed
experiment will be an attack on toothbrushing, a concept favorably valued
by the public, it would seem logical that similar effects should be found.
UnfOrtunately, certain other characteristics of the experimental topic
make this situation more complex.

McGuire's attacks on cultural truisms such as toothbrushing were
highly effective because most subjects were poorly prepared to defend their

beliefs about such topics. Under these conditions, then, it is difficult

 

11

to Specify exactly what prediSpositions subjects have or how they will
react to a message with such "unusual" arguments. Thus, a more general
hypothesis was postulated concerning the consequent development of
attitudes toward the source: if a person is attacked by a message, he will
regard the source of the message negatively; if a person is supported,
he will evaluate positively; if a person is strongly supported, he will
evaluate a message source very positively. Because of the possible
occurrence of this variety of reSponses, it seemed more reasonable to pre-
dict that a high intensity message would lead to more polarized changes,
regardless of direction, than would a low intensity message.

Hypothesis 3: A high intensity message will lead to more

polarized evaluations of message characteristics
than will a low intensity message.

 

Razinsky (1967) found that as the number’of spelling punctuation
and grammatical errors increased, the evaluation of other message
characteristics such as logic, quality, and style also decreased.

The particular message characteristics of interest in this experi-
ment are perceived logical quality, informational quality, clarity, and
opinionated language. Because assertion intensity is not as clearly
linked to a positive or negative evaluation as are errors, it seemed
reasonable to predict only more polarized responses to these variables.
A high intensity message will make a greater impression on subjects and
thus lead to more intense evaluations; however, these evaluations could

as easily be in one direction as another.

 

CHAPTER II

METHOD

Overview

This experiment used an after-only design with subjects randomly
assigned to treatments. Each subject was given a booklet containing an
introduction, a persuasive message, and two pages of scales tapping
perceptions of the message source and message characteristics. The
message, a persuasive attack on the merits of toothbrushing was identified
as a recent article from a major metropolitan daily newspaper. One
control group of subjects received a message about flying saucers instead
of the experimental message. Their responses were used to determine
pretest attitudes toward toothbrushing. After reading their particular
message, subjects were handed a second booklet in which they indicated
their attitudes toward a variety of topics, including toothbrushing.

After completing this task, the experiment was explained.

Independent Variable

 

Message intensity was manipulated by substituting verbs, adverbs,
and adjectives into the basic message. The verbs and adverbs were
selected on the basis of earlier evaluations by judges (Osgood, Saporta,
and Nunnally, 1956). Only adjectives that implied different intensities

such as "most" and "some" were used. Specifically, for the high intensity

l2

13

message, 12 assertions which implied possible or hypothetical relations
such as "constant gum irritation an result in infection" were changed

to unqualified present tense: "Constant gum irritation results in
infection." In addition, the adverbs "absolutely, definitely, and
positively," were individually added to three different assertions. In

the low intensity message, the qualifiers "may" and "might" were added

to twelve assertions and the adverbs "slightly, occasionally, and somewhat"
inserted in eight different assertions. For example, here are excerpts

from the high and low intensity messages:

High Intensity Low Intensity
"...too frequent brushing causes "...too frequent brushing may
.gum infections and even mouth occasionally cause gum infections-
cancer, and increases rather and even mouth cancer, and may
than diminishes the amount of slightly increase rather than
tooth decay." diminish the amount of tooth
decay."

A total of 30 words were deleted from the basic message to produce
the high intensity message. Twenty-four high intensity alterations were
made. In the low intensity message, twenty-six words were deleted; thirty-
eight words were added. In most cases the basic message was altered at the
same position forlboth the high and low intensity messages. The complete

messages are in Appendix A.

Perception of Message Intensity
Subjects' perceptions of message intensity were tapped with a single
item that had been previously shown to differentiate between high and low

intensity messages. (McEwen, 1967). This item asked "How strongly

 

14

would you say that the author of this article felt about excessive tooth-
brushing?" Subjects rated intensity on a seven-point scale that ranged

from "very Strongly opposed" to "very strongly in favor."

Dependent Variables

 

Three dependent variables were examined in this study.

(1) Subjects' attitudes toward toothbrughing were measured with
four 15-point agree-disagree statements. The four toothbrushing items
were included among eight other opinion statements relating to health
practices and unidentified flying objects. A score of 15 indicated
maximum disagreement with the opinion statement (all positively stated).
Summing across the four scales then, a maximum score of 60 indicated complete
cagreement with the position advocated by the experimental message, and a
score of u indicated maximum disagreement.

The feur attitude statements were (as numbered in the test booklet):

1. Everyone should.brush his teeth after every meal if at all
possible.

II. There are almost no disadvantages to regular and frequent tooth-
brushing. I

7. Frequent toothbrushing is a very healthy practice.

10. The best way to prevent tooth decay is to brush one's teeth
frequently.

(2) Perceived credibility of source as a function of message intensity
was separated into three dimensions, each represented by two seven-point
rating scales. The source's safety dimension was measured on scales of

trustworthy-untrustworthy and fair-unfair; the competence dimension on

 

IS

scales of competent-incompetent and experienced-inexperienced; the
dynamism dimension on scales of active-passive and timid-bold.

(3) Generalization between message intensity and four message
characteristics was measured with the following items (as numbered in
the test booklet):

3. In general, how would you rate the quality of information in this
article? (four response categories: very good to very bad)

u. How would you rate the logic of the arguments presented? (four
response categories: very good to very bad)

5. How would you rate the clearness of writing in this article?
(four response categories: very clear to very vague)

6. In general, how opinionated.was the language used in this article?

(four reSponse categories: very Opinionated to not opinionated at all)

Message

The basic message used in this study was developed by McGuire (1961)
for his studies on resistance to attitude change. This message, a SOC-word
attack on the merits of toothbrushing was chosen for two reasons. First,
previous researdh (McGuire, 1961) showed that most subjects had highly
positive attitudes toward toothbrushing. Under the assumption of similar
prior attitudes, there was no necessity for a pre-test and the possible
sensitization it entailed.' Thus, the experimental design was simplified by
use of this topic. Second, assuming that subjects are seldom aware of
arguments against toothbrushing, it fellowed that they would not be prepared

to defend their beliefS. Under these conditions, Hovland’s (1959)

 

16

generalizations seem to be the most applicable: the more discrepant an
argument, the more likely it is to produce the advocated amount of change.
If the experimental rationale was correct, the high intensity (most
discrepant) message would produce the greatest changes.

A message attacking a cultural truism is somewhat limited in
.generalizability because such attacks occur so seldomly. However, the
simplification of the experimental design made possible by use of this
t0pic, as well as the advantage it gave the independent variable, out-

weighed its disadvantages.

Subjects -

Four different groups of subjects were used in this experiment.
Two of the groups were undergraduate classes in Business writing at
Michigan State UniverSity. One class contained 20 subjects and the other
had 37. The other two groups were undergraduate Speech classes. There
were eight subjects in one group and 10 in the other. Altogether 75
subjects participated in this experiment. After coding, four questionnaires
had to be rejected. All these questionnaires contained two or more blank
response categories and two of them contained written messages indicating
that the subjects had deliberately tried to mislead the experimenter.
After discarding these questionnaires, there were 29 subjects in the high
intensity condition, 26 in the low intensity condition and 16 in the

control group.

Design

An after-only design was used with a different message group serving

17

as a control to examine the persuasive effect of the toothbrushing message.
Since the differential effects of intensity were of interest, and previous
studies (McGuire, 1961) indicated relatively similar attitudes toward
toothbrushing, subjects were not pretested. It was assumed that individual
differences among subjects, including initial attitude toward tooth-
brushing and familiarity with the arguments presented would be controlled
through random assignment of subjects to treatments. Examination of the

control group's attitude scores tended to support this assumption.

Procedure

The author was introduced in each class as a graduate student from
the Department of Communication. 0n the cover of each booklet, the study
was identified as "...part of a project to examine and hopefully improve
upon various aspects of newspaper and magazine articles that report new
developments and discoveries in the fields of science and medicine."
Subsequent discussion with the subjects indicated that this guise was only
partially successful. Some of the subjects did not believe that the
article in the booklet had actually appeared in a newspaper. ‘

The experimenter distributed to each class member a booklet containing
either a high intensity, low intensity or control message. Distribution
was random in each class.

To insure that the subjects would read the message thoroughly, the
cover Sheet instructed them to read the message carefully and underline

each main point. The cover sheet was immediately followed by the

experimental message, the message intensity item, Six source credibility

18

scales, and four message evaluation items. When subjects had completed
this section, they were told a second booklet would be distributed and
were asked not to refer back to the first section..

Part II consisted of 12 Opinion statements (including four tooth-
brushing items). After the subjects completed this section, the test
booklets were collected and the true purpose of the experiment was ex-

plained.

 

CHAPTER III

RESULTS

The experimental results have been divided into four sections.
The first section reports measurements of perceived message intensity.
The second section reports differences in toothbrushing attitudes among
the two experimental groups and the control group. The third section
reports mean deviations (or polarization) of source evaluations as well

as absolute ratings. The fourth section reports message evaluations.

Ratings of Message Intensity

 

The manipulation of message intensity was successful. Subjects in
the two experimental conditions differed significantly in their ratings Of
how strongly the source felt about the message topic. 'These results are
represented by the mean perceived intensity ratings in Table l. A lower
absolute value indicates greater perceived message intensity. A t-test
for uncorrelated means resulted in a significant difference between the
two groups (PԤ.01, two-tailed). Thus, the high intensity message was

rated as significantly more intense than the low intensity message.

Attitudes Toward the Topic

 

A comparison of the experimental groups with the control group
indicated that the exPerimental messages led to more negative attitudes

toward the topic of toothbrushing. The two experimental messages, however,

19

 

20

did not lead to different amounts of attitude change as predicted.by

 

Table 1. Ratings of Message Intensity

 

 

 

 

Treatments
High Intensity Message Low Intensity Message
(n=29) (n=26)
I: 1.72 Ti: 2.73
6 = .9u 4 - 1.75
t = 2.67*
*p (.01

 

the experimental hypothesis. These results are reported in Table 2.

Each value was obtained by first summing across each subject's scores

on four different attitude scales and then averaging the composite scores
to»find.the group means. It was possible to have attitude scores ranging
from u (maximally in favor of toothbrushing) to 60 (maximally opposed to
toothbrushing). In actuality, scores for the control group, who did not
receive an experimental message, ranged.from u to 60. In the low intensity
experimental condition, scores ranged from 8 to 53, and in the high

intensity condition they ranged from 6 to 55.

 

21

Table 2. Attitudes Toward Toothbrushing

 

High Intensity Message

Mean Attitudes Toward Topic

Treatments

Low Intensity Message Control (No Message)

 

 

(n = 29) (n = 26) (n = 16)

Ta = 29.72 Yb :2 28.81 Tc = 17.19

6 = 11.99 6 = 11.96 6 = 13.58
tab = .292* tac = 3.09% tbc = 2.81**

Significance Test

 

 

 

Source Of variance Sums Of Squares df Means of Squares F Values
Between Columns 1,826 2 913 5.8u**
Within 10,613 ' 68 156.1
Total 12,939 70

critical F (P = .05) is 3.15
*NOSO. -

aMp < .01

22

.A single classification analysis of variance revealed that the
three groups were significantly different.* T-tests for uncorrelated
means showed that each experimental group was significantly different
from the control group.* The difference between the means of the high
and low intensity grOUps was in the predicted direction, that is, the
high intensity group had more negative attitudes than the low intensity
groups, but it did not achieve statistical significance. Thus the
experimental messages led to significantly less favorable attitudes
toward toothbrushing,.but did not have differential effects as predicted

by the experimental hypothesis.

Source Evaluation

 

Comparisons of the subjects' source evaluations failed to reveal
any significant differences between the two experimental conditions. Two
comparisons were made. First, to test the hypothesis that a high
intensity message would lead to more polarized reSponses than a low
intensity message, all responses were scored as absolute deviations from
neutrality. The deviations (range: 0-3) were averaged across each pair
of scales representing a credibility dimension. Each subject had three
scores representing the average deviation of his competence, safety,
and dynamism ratings. Group means were computed from these scores and
compared using t-tests for uncorrelated means, as reported in Table 3.

Secondly, ratings were also scored on an absolute scale ranging from one

 

*p < .01

23

Table 3. Mean Deviations of Source Evaluations

 

a) Competence (Scales: Competent - incompetent; experienced - inexperienced)

 

 

 

Treatments
High Intensity Message Low Intensity Message
(n=29) .‘ (n=26) ‘
“SE: 1.19 3? = 1.17
4 = .652 4 = .631: t = .114*

b) Safety (Scales: Trustworthy - untrustworthy; fair - unfair)

 

 

Treatments
High Intensity Message Low Intensity Message
(n=29) (n=26)
36 = 1.00 'x‘ = 1.10‘

 

4 = .541 d = .l-I86 t= .725*

c) Dynamism (Scales: Active - passive; timid - bold)

 

 

Treatments
High Intensity Message Low Intensity Message
(n=29) (n=26)
32‘: 1.111 3?: 1.33

 

4 4 = .784 t = .9176

I
O
O)
i...
\I

 

*N.S.

 

26

(most positive) to seven (least positive). These results, also averaged
for each dimension of source credibility, are in Table 9.

There were no significant differences between mean deviations of
source evaluations for any dimension of source credibility. Similarly,
none of the differences in the absolute scores reached significance. In
the case of the dynamism dimension, subjects in the high intensity condition
rated the source as more dynamic than those subjects in the low intensity
condition but the t-test did not reach a conventional level of significance
(I%<.20, two-tailed). Thus, source credibility evaluations by the two

experimental groups did not differ significantly, either in terms Of mean

deviations from neutrality or in absolute scores.

Message Evaluations

 

Comparisons of message evaluations failed to reveal any significant
differences between the experimental conditions. Again, two separate
comparisons were made. To test the experimental hypothesis that the high
intensity message would lead to more polarized reSponSes than the low
intensity message, each message evaluation item was scored in terms of
deviation from neutrality. There were four reSponse categories for each
item; the extreme categories were assigned a weight of two, and the
remaining categories a weight of one. These results are summarized by
item in Table 5. To evaluate the direction of the responses, items were
scored from one to four. In terms of informational quality, logic and

clarity, a low score indicated a more favorable evaluation of the message

 

 

25

Table 4. Source Evaluations - Absolute Scores

 

a) Competence (Scales: Competent - incompetent; experienced -

 

 

 

 

 

 

 

 

 

inexperienced)
Treatments
High Intensity message Low Intensity message
(n=29) (n=26)
3?: 4.02 32': 3.67
t = 1.16*
6 = 1.12 6. = 1.12
1)) Safety (Scales: trustworthy - untrustworthy; fair-unfair)
Treatments
High Intensity message \ Low Intensity message
__ (n=29) (n=26)
51': 4.28 32': 3.98
t " .85‘P"
4 = 1.01 4 = 1.24
c) Dynamism (Scales: active - passive; timid - bold)
Treatments
High Intensity message Low Intensity message
_, (n=29) (n=26)
'2': 2.76 32': 3.23
' = l.59**

‘ - .987 _ 6 =1.20

AN.S.
**p .2

 

 

 

26

Table 5. Mean Deviations of Message Evaluations

 

3. INFORMATIONAL QUALITY

 

 

 

 

 

 

 

TREATMENTS
High Intensity Message Low Intensity Message
(n=29) (n=26)
Ti: 1.10 32': 1.08
t = 0.260
6 = 0.304 4 = 0.266
'4. LOGIC
TREATMENTS
High Intensity Message Low Intensity Message
(n=29) (n=26)
3?: 1.14 3?: 1.15
t = 0.105
6 = 0.344 6 = 0.360
'5.. CLARITY
TREATMENTS
High Intensity Message Low Intensity Message
(n=29) . (n=26)
34‘: 1.34 'x‘= 1.23
t = 0.909
4 = 0.475 6 = 0.422
6. OPINIONATED LANGUAGE
TREATMENTS
High Intensity Message Low Intensity Message
(n=29) (n=26)
3?: 1.38 3?: 1.35
t = 0.231

 

6 = 0.485 6 = 0.475

 

27

than did a high score. For the last item concerning opinionated language,
a low score indicated that the subjects thought the language was more
opinionated. These results are summarized in Table 6.

T-tests for uncorrelated means failed to reach significance for any
of the message evaluation variables. There was a slight difference
between'the two groups on absolute evaluations of clarity (pt .20, two-
tailed). Thus, message evaluations from high and low intensity ex-

;perimental conditions were not significantly different either in terms of

polarization or absolute scores.

28

Table 6. Message Evaluations - Absolute Scores

 

3. Informational quality

 

 

 

 

 

 

 

 

 

 

 

 

Treatments
High intensity message Low intensity message
(n=29) (n=26)
35:2.45 232.31
t = .82A
4 = .674 6 = .601
k. Logic
Treatments
High intensity message Low intensity message
(n=29) ' (n=26)
Y:2.07 i=2.23
t = .920
4 = .583 6 = .697
5. Clarity
Treatments
High intensity message I Low intensity message
(n=29) (n=26)
X=1.76 76:2.00
' t = 1.36*
4 = .524 4: .679
6. Opinionated Language
Treatments
High intensity message Low intensity message
(n=29) (n=26)
'x'=1.76 . X=1.89
t = .622
6. - .677 6 - .751

p(.2

CHAPTER IV

DISCUSSION

Summary

This Study investigated the effects Of variations in message in-
tensity on receiver attitudes toward the message, the topic, and the
source of a persuasive message. Message intensity was conceptually
defined as the degree to which the assertions in a message deviated
from neutrality. It was Operationalized by the substitution Of previously
rated high and low intensity verbs, adverbs, and adjectives into a basic
message. Effects Of the independent variable were measured on scales which
tapped: l) Attitude change; 2) perceptions of the source along safety,
competence, and dynamism dimensions; 3) perceptions of four message attri-

butes: logic, informational quality, clarity, and language Opinionation.

Conclusions

 

Hypothesis 1: While the message was shown to be persuasive, there
was no significant difference between the high and low intensity versions.
Thus, Hypothesis 1, which predicted that the high intensity message would
lead to more attitude change than the low intensity message, was not
supported.

Hypothesis 2: There were no significant differences in source
evaluations between treatment groups. Thus, Hypothesis 2, which predicted
that the high intensity message would lead to more polarized evaluations

of the message source, was not supported. There was some evidence that

29

30

the dimension of dynamism was more reSponsive to the intensity manipu-
lations than were the safety and competence dimensions. The difference
between the two groups, however, did not reach conventional levels of
significance (p < .2) .

Hypothesis 3: Judgments of logical quality, informational quality,
clarity, and Opinionated language were not significantly different between
treatment groups. Thus, Hypothesis 3, which predicted that a high
intensity message would lead to more polarized evaluations of message
characteristics, was not supported. There was a slight difference between
the two groups on absolute judgments Of clarity, but this difference did

not reach conventional levels of significance (p< .2).

Discussion

 

There are a number of possible reasons why the experimental
hypotheses were not confirmed. The first, and simplest reason, would be
that the intensity manipulation was not effective. Evidence for this
explanation is somewhat contradictory. The comparison between treatment
_groups on the measure tapping perceived intensity was highly significant;
however, both means were on the high intensity side of the scale. Dis-
cussion with the subjects after the experiment indicated that those in
the high intensity condition felt their message was rather "extreme,
while subjects in the low intensity group felt their message was "weak"
because it did not draw what they referred to as "definite conclusions."
On the other>hand, 14 out of 26 subjects in the low intensity condition
perceived the source as being "very strongly opposed", or "quite Opposed"

to toothbrushing. Thus, there remains the possibility that the experimental

31

manipulations alone were not adequate enough to produce different per-
ceptions of intensity.

Some evidence, in the form of produce moment correlations indicated
that the relationship between intensity and the dependent variables may
not be a direct one. While the correlation between intensity judgments
and attitude scores was not significant, two of the three source
evaluation dimensions were related to perceived intensity rather than
manipulated intensity.

(1) Neither deviation or absolute scores on the competence
dimension were significantly correlated with perceived
intensity for the high or low intensity conditions.

(2) Absolute safety evaluations were significantly correlated
with intensity judgments (High I. r = -.338, t = l.76*,

Low I. r =-.341, t = l.78*). Safety deviation scores
were not significantly correlated (High I. r = 1.01;
Low I. r = .015).

(3) Absolute dynamism scores were significantly correlated
with intensity judgments (High I. r = .394, t = 2.23**;

Low I. r = 3.62, t = 1.90*), but correlations with
deviation scores also failed to reach significance

 

* P‘£.05, critical t 05 = 1.71 (one tailed)
9“" P< .01, critical t.01 2,05 (one tailed)

32

Correlations were also calculated between intensity judgments and
evaluations of the various message characteristics. Only the correlations
between intensity and opinionated language proved to be significant.
Absolute Opinionated language scores were significant for both the

high and low intensity messages (High I. r = .329, t = l.7l*, Low I. r =
.328, t = l.70*). For the deviation scores, only the high intensity
condition showed a significant correlation (High I. r = -.371+, t = 1.98;
low I. r =.-.258, t = 1.10).

Failure to find a correlation between intensity judgments and
attitude scores could indicate that the intensity manipulation was not
strong enough. There is more reason to believe, however, that lack of
a relationship resulted from the use of an unusual topic.

Razinsky'(l967) also drew this conclusion after failing to find any
between-group variations in attitudes induced by Spelling, punctuation and
. grammatical errors in a message attacking toothbrushing (the same as
used in this experimentl He commented that the message itself may have
been entirely too effective: "...the cultural truisms high susceptibility
to change allows any reasonable attack to be successful. The inherent
susceptibility might interact with the potential persuasiveness of the
attack, leading to rather large changes in attitudes. The effect of
the...(independent variable)...would easily be cancelled by the strong

persuasive potential of the Situation." (p. 29) Razinsky's reasoning

 

*P: .05, critical 12.05 = 1.71 (one tailed)
**P< .01, critical t 01 2.05 (one tailed)

33

contradicts the rationale for using a toothbrushing attack in this
experiment. Originally, it was reasoned that toothbrughing attitudes would
be most differentially affected by high and low intensity attacks. Now,
because the difference between groups actually failed to reach significance,
Razinsky's argument must be given more weight.

Linking an unusual topic to a moderately credible source may also
have interfered with the effect of the intensity manipulation. As
indicated previously, the ploy used to dnguise the real nature of the
experiment may not have been effective. Some subjects commented that
they did not believe the message really appeared in a major metropolitan
daily newspaper. This reaction to the source induction may have re-
sulted in confOunding influences and raised the variance of the attitude
scores. That is, some subjects may have been responding as if the
attack were a "real" message; some may have tried to remain unaffected
and responded as if they had never read the message; and still others
may have resented the attempt to influence their Opinions and reacted
by indicating extreme favorability toward the topic.

BefOre they can be accepted, these conjectures require somewhat
more substantial evidence than the large attitude score variance fOund
in this experiment. However, there remains a strong possibility that
reaction to the unusual topic may have been a confounding variable in
the perceived intensity-attitude relationship.

Significant correlations between perceived intensity and source

evaluations would seem to indicate that subjects in both experimental

31+

conditions who felt the message was more intense also thought the
message source was less safe, but more dynamic. Somewhat similar results
were found by Razinsky (1967).

Razinsky showed that judgments of source credibility did not vary
with increasing numbers of errors in grammar, Spelling and punctuation,
unless attention to the errors was emphasized. Directed attention to
errors elicited greater differences across treatment groups, eSpecially
when responses were analyzed according to the number of errors identified
by the subjects. It may be that the same is true of assertion intensity
manipulations. For some subjects, the assertions were not sufficiently
strong enough alone to act as indexing cues fOr evaluation of the source.
Still, some subjects apparently did find the message intense and evaluated
the source accordingly, at least on the safety and dynamism dimensions.
These variations could be a result of the intensity manipulations them-
selves, but it is also possible that the differences between the subjects
who did perceive the intensity manipulation and those who did not may
be a function Of an intervening personality variable. The implications
of this reasoning will be discussed later in this chapter.

Of the four message characteristics examined in this study, only
Opinionated language was significantly related to judged intensity. Here,
those subjects who found the message intense also felt the language was
more opinionated. Apparently subjects had other criteria besides
assertion intensity for judging logic, clarity, and infOrmational quality.

Failure to find many significant relationships may have resulted

from the particular choice of items used to tap message perceptions.

35

There was no apriori rationale for the choice of items and it is quite
probable that there are other dimensions Of message evaluations which
would have been more affected by variations in message intensity. The

implications of this reasoning will be discussed in the next section.

Implications and Research Extensions

 

Although none of the experimental hypotheses were supported, there
is some reason to believe that the effects of the intensity manipulations
were confounded by extraneous variables. Therefore, rather than discard
the notion of message intensity, further research should be designed to
clarify the relationships between intensity, attitude, source evaluations,
and perceived message evaluations. A number of research extensions can
be suggested.

This experiment used only one rather unusual topic. To increase
generalizability, several different topics should be chosen. A wider
range of topics would permit comparisons to be made with the results
Of prior experiments by Bowers (1963) and McEwen (1967) and would provide
useful information for professional communicators. A wider range of
intensity variations would also increase generalizability. This ex-
periment used only one message from each of the more extreme ends of the
intensity continuum. A large number of messages varying from high to
medium to low intensity could be created by varying the number of insertions
and deletions of high and low intensity words. In this manner, the effects
of intensity on attitudes, source evaluations, and message evaluations

could be examined to determine if they were continuous and linear.

36

As another research extension, subjects could be cued to variations
in intensity. If they were told that the messages they were about to
read were written by people who either felt strongly or weakly about
their topic, they would be able to concentrate on intensity cues. By
comparing the responses of subjects reSponding under these conditions
to subjects who had not been cued, it would be possible to determine if
intensity manipulations alone were strong enough to induce changes. As
in the case of Razinsky's (1967) message errors, it might be that subjects
require extra cues befOre they can perceive and respond to message in-
tensity manipulations.

Up to this point, an attempt has been made to predict subject's
behavior solely in terms of reactions to variations in message intensity.
In the future, an alternative approach might be to predict reactions to
variations in message intensity according to individual differences in
the personalities of the message receivers. For example, it may be
that certain subjects are more reSponsive to variations in intensity than
are other subjects. As a personality trait, this would be similar to
what Baron (1965) called "differentiation". People who are "differentiators"
as compared to others whom Baron calls "assimilators" are better able to
discriminate and remember separate details. "Assimilators", on the other
hand, tend to ignore details or lump them all tOgether. Here, assimilation-
differentiation could be postulated as an intervening variable. Persons
classified as differentiators would be more aware of intensity manipulations
and respond accordingly. Conversely, assimilators would be expected to

ignore most variations in intensity. If assimulation-differentiation is

37

related to perception of variations in intensity, as an intervening
variable it would reduce within-variance and allow the effects Of
intensity to be defined more clearly.

Another variable that seems relevant to message intensity is
Sherif, Sherif and Nebergall's (1966) notion of assimilation-contrast.
Briefly, contrast can be defined in the following manner: when a subject
evaluates a position advocated by a communication, the greater the
distance between the subject's own stand and the position advocated, the
greater will be the subject's displacement of that position away from his
own. Assimilation can be defined as: when there is only a small dis-
crepancy between the subject's position and the position advocated by
the communication, the subject will displace the communication toward his
own stand. In other words, when a subject contrasts, he judges the
communication as more distant from his own stand than it objectively is;
the converse is true for assimilation. In contrast to Baron's
assimilation-differentiation, which is topic free, the Sherif, Sherif
and Nebergall notion of assimilation and contrast only applies to tOpics
which are ego-involving for the receiver.

In terms of intensity, if a person is ego involved with, and highly
in favor of the position advocated by a communication, a high intensity
message will be closer to his position and will have a greater chance of
being assimilated. A low intensity message would be judged as somewhat
farther away by an extremely favorable person, and may be contrasted.

For the person who is extremely unfavorable toward the position advocated

by the message, the high intensity messsge would be most discrepant and

38

would be contrasted, whereas the low intensity message would not be as
discrepant and would stand a greater chance of being assimilated. Thus,
having pre-test measures Of subject's attitudes toward the topic would
allow detailed predictions to be made of a subject's responses to
variations in intensity.

The final research extensions suggested by this experiment are of a
more general nature and not directly related to message intensity. There
seems to be some need for a better conceptualization of those characteristics
of a written message which audiences use to evaluate the source of a
message. Miller and Hewgill (1964) have already shown that non-fluence
in spoken communications will cause adverse ratings of some dimensions of
speaker credibility. Eagly and Manis (1966) demonstrated that the tOpic
of a written message led to different evaluations of the source, depending
on the subjects' attitudes toward the topic. These manipulations, how-
ever, are rather Obvious and can be easily perceived.by audiences. In
the case of spoken messages, subjects may be able to make much more
subtle evaluations of a speaker because they constantly make such dis-
tinctions in everyday life. The personal characteristics of the sources
Of written messages, however, are seldom given along with their messages
and pe0ple have less practice associating stylistic cues with personality
traits. For this reason, variations in written messages have not been
too successful to date in inducing variations in source evaluations.

For example, Razinsky (1967) found that errors in a message did
not affect evaluations of the source unless attention was directed to

errors. Even then, only evaluations of the sources' competence and

39

safety were affected. McEwen (1967) did not find any difference in source
evaluations when he manipulated assertion intensity. Similarly, Bowers
(1963) reported failure to find any variations in source evaluations re-
sulting from his intensity manipulation. These results are all supported
by Osgood and Tannenbaum's (1955) general finding that it is more difficult
to change evaluations of a source than it is to change evaluations Of a
topic.

0n the basis of this evidence, it would seem desirable to devise a
series of experiments to discover what message characteristics subjects
can recognize, and which of these characteristics make a difference in
source evaluations. One method would be to ask subjects to evaluate
a message with a series Of open ended questions. For example, one might
ask: "What did this message tell you about the person who wrote it?"
Responses could then be content analyzed to find what kinds of destinctions
were made. After analyzing the distinctions, operational methods could
be developed-for emphasizing the linguistic differences subjects noticed.
For example, a particular element could be simply repeated at different
points in the message; or, Specifically defined variations in a single
element could be inserted throughout the message. Different messages
agenerated using this procedure could then be tested under relatively
realistic communication situations. Until such procedures are utilized,
studies of the effects of stylistic manipulations will have to continue
searching for meaningful independent and dependent variables in a rather

unsystematic manner.

BIBLIOGRAPHY

Baron, R.M. Cognitive basis of attitude change as a function of
motivational, stylistic, and stimulus factors. Journal of
Personality and Social PsychOIOgy, 1965, 2, 219-231.

 

 

Bowers, J.W. Some correlates of language intensity. Quarterly Journal
of Speech, 50, 1964, 415-420.

 

Bowers, J.W. Language intensity, social introversion, and attitude
change. Speech Monographs, 1963, 32, 345-352.

 

Bowers, J.W. 8 Osborn, M.M. Attitudinal effects of selected types of
concluding metaphors in persuasive Speeches. Speech Monographs,
1966, pp, 146-155.

 

Eagly, A.H. 8 Manis, M. Evaluation of message and communicator as a
function of involvement. Journal of Personality and Social
PsychOIOgy, 1966, 3, 483-485.

 

Goldberg, S.C. Three situational determinants of confOrmity to social
norms. Journal of Abnormal and Social Psychology, 1954, 49,
325—329.

Hovland, C.I. Reconciling conflicting results derived from experimental
and survey studies of attitude change, American Psychologist, 1959,
1’4, 8-170

 

Hovland, C.I. 5 Pritzker, H.A. Extent of opinion change as a function of
amount of change advocated. Journal of Abnormal and Social
Psychology, 1957, 54, 257-261. -

 

 

Jacob, P.E. Atrocity prOpaganda, in Childs, H.L. and Witton, J.B.
(eds.) Propaganda by Short Wave, Princeton University Press, 1942,
211-259.

 

McEwen, W. The effects of assertion intensity on the congruity principle.
Unpublished Master's Thesis, Michigan State University, 1967.

McGuire, W.J. Resistance to persuasion conferred by active and passive
prior refutation of the same and alternative counter-arguments.
Journal of Abnormal and Social Psychology, 1961, 63, 326-332.

 

Miller, G.R. 8 Hewgill, M.A. The effect of variations in nonfluency on
audience ratings of source credibility. Quarterly Journal of Speech,
1964, :2, 36-44.

40

41

Osgood, C.E. The representational model and relevant research methods,
in Pool, I. (ed.) Trends in Content Analysis, Urbana: University of
Illinois Press, 1959, 33-38.

 

Osgood, C.E., Saporta, S., and Nunnally, J.S. Evaluative Assertion Analysis,
Littera, 1956, 33 47-102.

Osgood, C.E. 6 Tannenbaum, P.H. The principle of congruity in the pre-
diction of attitude change. Psychological Review, 1955, 62, 42-55.

 

Razinsky, E.L. The effects of message deterioration on message and source
evaluation, attitude and information gain. Unpublished Master's
Thesis, Michigan State University, 1967. .

Sherif, M., Sherif, C.W. 6 Nebergall, R.E. Attitude and Attitude Chang_,
Philadelphia: W.B. Saunders CO., 1965.

 

Waples, D. 6 Berelson, B. What the voters were told: an essay in content
analysis. Graduate Library School, University of Chicago, 1941
(mimeographed).

Wright, 0. 5 Nelson, C.J. American attitudes toward Japan and China,
1937-38._I_’.0.Q., 1939, 3, 546-580.

Zimbardo, P.G. Involvement and communication discrepancy as determinants
of Opinion change. Unpublished doctoral dissertation, Yale University,
1959. .

APPENDIX C

CONTROL BOOKLET

College of Communication Arts
Michigan State University
August, 1967

Science and Health News Study

This study is part of a project to examine and hopefully improve upon
various aspects of newspaper and magazine articles that report new develop-
ments and discoveries in the fields of science and medicine.

In this class, and other University classes, we are asking students
to read and evaluate a series of articles originally published in a major
metropolitan daily newspaper. Your reaction to these articles will help
us find out how effectively or ineffectively each particular science or
medical writer was able to convey his ideas. The article in this booklet
has been reproduced exactly as it was published.

Specifically, we would like you to read the attached article very
carefully. As you read the article, please underline the main points
made by the author. Underline as many words, clauses, or sentences as
you think represent each main point.- At the end Of the article are a
few items which will enable you to tell us what you think of the article.

We very much appreciate your helping us in this study.

John Kochevar
Project Director

ggadfithis message carefully and underline all main points

A

 

Some Dangers of Excessive Tooth Brushing
By Richard Thorpe

 

Many people brush their teeth more or less automatically after each meal
without realizing that of late, medical reports_may,be calling this procedure
into question. Recent medical and biological studies indicate the beneficial
effects of constant tooth brushing probably have been exaggerated. Furthermore,
it has been partially demonstrated that a number of bad effects_may result from
brushing teeth often. Constant gum irritation occasionally can result in infec-
tion and even mouth cancer. Also, brushing teeth so frequently tends,to_pu§h
hack the gums and expose the non-enameled parts of the teeth to decay. Hence,.
medical authorities are beginning_£9 urge that instead of brushing our teeth so
frequently, we take other measures to_improve dental health, such as a better
diet. Let us review some of this recent evidence demonstrating that constant
tooth brushing mayinot do any great amount of good and.migh£ do some harm.

The most undesirable effect of tooth brushing is the damage it occasionally
causes to the gums.: All of us may have noticed that when we brush our teeth,
we sometimes cause our gums to bleed.” Such bleeding, obviously,_g§y indicate
some degree of gum injury. These injuries, besides the physical damage they
cause, somewhat increase the likelihood of infection._§gme_doctors generally
believe that many serious gum infections may_result from accidental injury to
the gums inflicted during tooth brushing. Furthermore, repeated injuries of the
gums caused by constant tooth brushing might, even when each of these injuries
is only slight, produce mouth cancer. Also, frequent brushing may_actually in-
orgase rather than decrease- the amount of tooth decay by exposing the unprotected
areas of the teeth to the decay-causing bacteria. _Nature has given our teeth a
very good protection: the enamel sheath. This sheath covers only the exposed

portions of the teeth: there is almost no enamel under the portions

In 1963 in Illinois, a UFO followed a car carrying two passengers as it
twisted and turned along a country road. The UFO flew behind the car for a
while, then zipped in front of it and zoomed over the car at great speed. Air
Force officials called the object the planet Jupiter. They did not explain how
the planet could make 90 degree turns and pass a moving car from the opposite
direction.

Many other events like these could be described. All of them increase
the likelihood that some intelligent being is controlling the craft.

Next, consider this evidence. In 1957, a Brazilian fisherman watched a
UFO dive toward the ocean, then race straight up for~a few seconds until it
exploded with a blinding flash. Pieces of the object washed ashore near him.
The fisherman took these pieces to scientists and told them his story. The
scientists concluded that the object could not have been a meteor because
meteors do not travel straight up. The men had the fragments chemically
examined. The pieces were pure magnesium. Pure magnesium is not found on
earth nor can it be manufactured by known processes. Obviously, the metal
had to come from some place other than earth. The Air Force has made no
comment on this evidence even though it has been given pieces of the metal
for analysis.

Flying saucers cannot be mass hallucinations. They have been seen by
radar and by the human eye at the same time--radar cannot detect an hallucina-
tion . They cannot be weather balloons or planets or meteors. These objects
do not maneuver as UFO's have been seen to maneuver. All the evidence points
to the fact that UFO's are real objects controlled by some intelligent beings
from a world with a technology perhaps more advanced than our own. To con-
tinue to insist that UFO's do not exist or are just "natural phenomena" is

not a rational conclusion.
GO TO NEXT PAGE

Q. -. a“.

PART I

Now that you have read the article, and underlined the main points, we
are interested in what you thought of the article and the person who wrote

it. '

1. How strongly would you say that the author of this article felt about the
topic he was discussing. (CIRCLE one)

VERY ‘Q‘UI’T'E Stream "" ’ SLIGHTL‘? f'QUITE ' VERY
STRONGLY QPPQSED- OPPOSED NEUTRAL IN FAVOR IN FAVOR STRONGLY
OBEOSED. ' IILEAYOB

 

 

2. Now, we would like to know what you think of the author of this article.

a. How competent or incompetent do you think he is?
Competent , Incompetent
very very

b. How trustworthy or untrustworthy do you think he is?
Untrustworthy Trustworthy

very ' very

 

c. How active or passive do you think he is?
Active Passive
very very

d. How experienced or inexperienced do you think he is?
Inexperienced A Experienced
very very

e. How fair or unfair do you think he is?
Unfair Fair.
very very

f. How bold or timid do you think he is?
Timid Bold

very very
3. In general, how would you rate the quality of the information in this article

very good
good
bad
very bad

 

 

4. How would you rate the logic of the arguments presented?

very good

good

bad
__ very bad

5. How would you rate the clearness of writing in this article?

very clear
clear
vague
very vague

 

6. In general, how opinionated was the language used in this article?

very opinionated
opinionated

not very opinionated
not opinionated at all

 

.-— .‘

‘- ----

APPENDIX D

ATTITUDE MEASURES

PART II

Editors and reporters are very concerned over the possibility that their
news articles may be affected by their own opinions. They also would like to
know if~¢he opinions of their readers have any effect on reader interest or
understanding. To determine this, we would like you to indicate your own per-
sonal opinions toward the topic of the article you just read. In this section,
please indicate your own opinions, at this time, toward each item.

Each statement below will be accompanied by a scale. You are asked to in-
dicate the extent of your agreement or disagreement with the statement. You
may find yourself agreeing strongly with some of the statements, and disagreeing
just as strongly with others. Some of the statements deal directly with the
subject matter of the article you just read. Others deal with articles being
used in other parts of this study. \

Each statement is accompanied by a scale. You are asked to
indicate on this scale the extent of your agreement with the
statement. The following is an example:

At the present time in the U.S., life expectancy is greater for people
living in rural areas than for those in urban areas.

Illlll: lILII IIII
I Definitely I’ Mildly [’ Neutral I Mildly I Definitely!
Disagree Disagree Agree Agree

Notice that the scale has five categories ("Definitely Disagree," etc.)
and that each of these categories has three divisions. You are asked to
indicate your agreement with the statement by marking an "X" in whichever of
these divisions best shows your opinion. For example, suppose that you agree
completely with the above statement, without reservation. In this case, you
would put your "X" over in the place to the extreme right, as shown below:

Li I l J] III I 1! glLXl
I Definitely I— Mildly I Neutral I Mildly I Definitely—7
Disagree Disagree Agree Agree

If you strongly agree with the statement, with some slight reservations,
you may want to put the "X" down toward the left end of the "Definitely Agree"
category, as follows:

IrlLIIIIIJII .XIII

 

I Definitely I Mildly ‘1 Neutral 'I Mildly I Definitely I
Disagree Disagree Agree Agree

If, on the other hand, you completely disagree with the statement. with-
out reservation, you should mark an "X" in the space at the extreme left, as
follows: .

 

IX: 4 l 1 l l [J l I i : ll I
I'Definitely I Mildly I Neutral I Mildl I Definitely I
Disagree Disagree Agree Agree

We want your personal opinion on each statement, as you think at this moment.

1. Everyone should brush his teeth after every meal if at all possible.

IT?III lIII II

 

 

I 1 .
I Definitely I Mildly 3 Neutral I Mildly I Definitely
‘ Disagree ~ Disagree Agree Agree

2. Most unidentified flying objects could easily be explained as mass hallu-
cinations or illusions by those who see them.

 

 

 

 

gr I I I L L I I I I I I I
Definitely I» Mildly I Neutral I Mildly Definitely
Agree Agree ‘ Disagree Disagree

3. Life, as we know it, has probably developed on other planets in the
universe, in addition to Earth.

IIIIIILIJIII III
I Definitely 5 Mildly I Neutral I Mildly DefinitelyI
Agree Agree Disagree Disagree

——.~-.-———

4. There are almost no disadvantages to regular and frequent tooth-brushing.

I I I i I I : E I gJ L I I I

' I
u - i t r I
Definitely I Mildly f» Neutral I Mildly I Definitelyj

Agree ”‘ ' Agree Disagree Disagree

5. Even though one may not have any reason for suspecting TB, it is a good
idea to have regular chest X-ray examinations.

 

I
a ; I
ILIIJJIIIIIIIIII
I Definitely I Mildly ; Neutral I Mildly I Definiteffl
Agree Agree Disagree Disagree

6. One of the greatest single advances in the history of medical science was
the discovery of penicillin.

 

IJILII III5II} III
| Definitely Mildly : Neutral I Mildly ' DefinitelyI
Disagree ’ Disagree ' Agree 5 Agree

7. Frequent tooth-brushing is a very healthy practice.

, I

IIIIII EILIII; I I:

I Definitely ‘I’ Mildly I Neutral. I Mildly I Definitely?
Agree ' Agree ' Disagree 8 Disagree

8. The Air Force is withholding information about unidentified flying objects
from the public.

III II ;IIJII: i II
I’Definitely Mildly 2 Neutral I Mildly DefinitelyI
Agree - Agree ' Disagree I Disagree

._-.._.r—. -

9. All things considered, getting an annual chest X-ray for detecting TB is a
very wise praCtice.

I I I I I I I I ‘ I I I ' I

l ; I I I r r L fl.

I‘ Definitely ‘ Mildly I Neutral I Mildly ‘- Definite lyI
Disagree ' Disagree ' Agree Agree

10. The best way to prevent tooth decay is to brush one's teeth frequently.

I I d ; I I 2 I I ' i I '+ I I it
I" Definite ly ' ' Mildly I Neutral I Mildly I De finite lyI
Agree“ ‘ Agree I Disagree I Disagree

ll. Penicillin may well be considered a ”wonder drug", because of the many
advantages to its use.

I I I t I I l I a I I I I I
I Definitely I Mildly I Neutral I Mildly ‘ De finite I]
Disagree ‘ Disagree 1 Agree Agree

12. Most of the sightings of unidentified flying objects are probably wrong
identifications of common objects.

 

I I I I I I I I I I I L I I I
FDefinitely { Mildly I Neutral Mildly I Definitely I
Disagree Disagree I I Agree ' ' Agree

ICHI RQRIES

III/IIIIIIIIIIIIIIIIIIIIIIIII

19