3 . .‘ . .- 'F- “. 3w .5. g‘ 1 _ " I J‘ ' ‘ .' I O ' ‘.
_ ‘ .1-1v'00} Ivl‘, w". ‘ . _ ”0", ', ¥.l. E: a I“ ... . . -l'¢‘.'l.(\....‘,,' .'. . |-.. “,)L'
, u -.~ ......,, . ,_._ o w -' ' '
v at r o? I. \ 3. fi . g - . .‘ ‘_ . I

AN EXPERIMENT ‘usme ROKEAQCH'S BELIEFS
THEORY T0 DETERMINETHE ‘
EFFECTIVENESS or” ADVERTISEMENTS m ATTRACT-1N6
ATTENTION AND ‘CHANGlNG BRAND PREFERENCE

Thesis for thé-Degre'e ofM. A; .‘ v
MICHIGAN'STATE'UNIVERSITY‘. i
JANE MARLEEN '_:SIMMON-S a.
' 1:974 “ ' '

 

 

alias-335535!!! I:

ABSTRACT

AN EXPERIMENT USING ROKEACH'S BELIEFS THEORY TO DETERMINE
THE EFFECTIVENESS OF ADVERTISEMENTS IN ATTRACTING
ATTENTION AND CHANGING BRAND PREFERENCE

BY

Jane Marleen Simmons

The purpose of the study was to determine the effec-
tiveness of magazine advertisements which contained or
violated one of four types of beliefs as defined by Milton
Rokeach. The effectiveness of the magazine advertisements
was tested on the basis of attention and brand perference.

The research design for the study was a 'one group
pretest-posttest with control.‘ The types of beliefs
contained in the advertisements were the independent vari-
ables, and the dependent variables were attention and brand
preference. The experimental and the control groups of 64
females were selected from students enrolled in undergraduate
advertising courses at Michigan State University. The types
of questionnaires used for the experiment were disguised and
structured. The questionnaire for the pretest and the post-
test used a seven point Likert scale to determine brand
preference, and the questionnaire administered only to the

experimental group asked the subjects to rank the advertisements

Jane Marleen Simmons

from one to four on the basis of attention. The data were
processed and analyzed by using a routine CISSR computer
program at Michigan State University. The rankings of the
advertisements on the basis of attention were analyzed by

Chi square. The ratings of the brands for both the experi-

mental and control groups were compared by a t-test between
two correlated means to determine if there was a significant
difference in the ratings of the experimental group after
seeing the advertisements.

The results indicated that the centrality of a belief
used in an advertising appeal, does affect whether or not the
advertisement attracts the attention of the subjects exposed
to it. But exposure to advertisements utilizing one of the
four types of beliefs has no effect on the brand preference

of the subjects.

Accepted by the faculty of the Department of Adver-
tising, College of Communication Arts, Michigan State Uni-

versity, in partial fulfillment of the requirements for the

degree of Master of Arts.

\

v f
./ j
'0 r ’ ‘ I

.r’ I '1) {gr/9 3
_. .' ',' ’ 3'" - .u - . k

Director of Thesis

6'//779f

AN EXPERIMENT USING ROKEACH'S BELIEFS THEORY TO DETERMINE
THE EFFECTIVENESS OF ADVERTISEMENTS IN ATTRACTING
ATTENTION AND CHANGING BRAND PREFERENCE

BY

Jane Marleen Simmons

A THESIS

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

MASTERS OF ARTS
College of Communication Arts

Department of Advertising

1974

ACKNOWLEDGMENTS

My sincerest appreciation goes to Dr. Gordon B.
Miracle for his many contributions in both time and ideas,
and without whose professional guidance, this thesis would
not have been possible. I would also like to thank Dr. John
D. Simpkins for his help with the statistical analysis and
for providing an opportunity to select subjects for the test.
A special thanks is directed to all the females in Advertising
classes who took the time to participate in the test, and to
the Michigan State University Computer Consulting Service
without whose help, a computer analysis of the data would
have been impossible. I would also like to thank my room-
mates for their moral support while this thesis was being

written.

ii

TABLE OF CONTENTS

Page
LIST OF TABLES O O O O O O O O O O O O O 0 iv
LIST OF APPENDICES . . . . . . . . . . . . vii

LIST OF ILLUSTRATIONS . . . . . . . . . . . viii

Chapter
I 0 INTRODUCTION 0 O O O O O O O O O O O l
The Problem . . . . . . . . . . . . 1
Purpose of the Stud . . . . . . . . . 11
General Hypotheses . . . . . . . . . 12

II. METHODOLOGY . . . . . . . . . . . . 22

Research Design . . . . . . . . . . 22
Sample . . . . . . . . . . . . . 26
Questionnaire . . . . . . . . . . . 27
Analysis of the Data . . . . . . . . . 29
Literature Search . . . . . . . . . . 31

III. RESULTS 0 O O O O O O O O O O O I O 32

Attention O O O O O O O O O O O O 32
Brand Preference . . . . . . . . . . 55

IV. SUMMARY OF CONCLUSIONS . . . . . . . . . 69
Attention . . . . . . . . . . . . 69

Brand Preference . . . . . . . . . . 75
APPENDICES . . . . . . . . . . . . . . . 78

BIBLIOGRAPHY C O O O O O O O O O O O O O 112

iii

10.

ll.

12.

l3.

14.

15.

16.

LIST OF TABLES

Continuum of Authority . . . . . . . .
Diagram of the Research Design . . . . .
Frequency of Rankings for Watch Advertisements

Chi Square Values for the Watch Advertisements
Frequency of Rankings for Panty Hose

Advertisements . . . . . . . . . .

Chi Square Values for the Panty Hose
Advertisements . . . . . . . . .
Frequency of Rankings for Liquor
Advertisements . . . . . . . . . .
Chi Square Values for the Liquor

Advertisements . . . . . . . . . .

Frequency of Rankings for Cigarette
Advertisements (A) . . . . . . . .

Chi Square Values for
Advertisements (A)

Frequency of Rankings
Advertisements . .

Chi Square Values for
Advertisements . .

Frequency of Rankings
Advertisements (B)

Chi Square Values for
Advertisements (B)

Compilation of Chi Square Values for the

Cigarette

for Automobile

Automobile

for Cigarette

Cigarette

Advertisements Tested . . . .

24

Ranking of Watch Advertisements and Hierarchy

of Beliefs . . .

iv

Page
10
22
33

34

35

35

36

37

37

38

38

39

40

40

42

44

Table

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.
31.
32.
33.
34.
35.

36.

Central Tendency Measurements for Watch
Advertisements . . . . . . . . . .

Ranking of Panty Hose Advertisements and
Hierarchy of Beliefs . . . . . . . .

Central Tendency Measurements for Panty Hose
Advertisements . . . . . . . . . .

Ranking of Liquor Advertisements and Hierarchy
Of Beliefs O O O O O O O O O O 0

Central Tendency Measurements for Liquor
Advertisements . . . . . . . . . .

Ranking of Cigarette Advertisements (A) and
Hierarchy of Beliefs . . . . . . . .

Central Tendency Measurements for Cigarette
Advertisements (A) . . . . . . .

Ranking of Automobile Advertisements and
Hierarchy of Beliefs . . . . . . . .

Central Tendency Measurements for Automobile
Advertisements . . . . . . . . . .

Ranking of Cigarette Advertisements (B) and
Hierarchy of Beliefs . . . . . . . .

Central Tendency Measurement for Cigarette
Advertisements (B) . . . . . . . .

Compilation of Hierarchy of Beliefs Analysis

Six Type A Advertisements and Reason for First
Ranking . . . . . . . . . . . .

t Values for Watch Brands . . . . . . .
Frequency of Ratings for Watch Brands . . .
Frequency of Ratings for Panty Hose Brands .
t Values for Panty Hose Brands . . . . .
Frequency of Ratings for Liquor Brands . .
t Values for Liquor Brands . . . . . .
Frequency of Ratings for Cigarette Brands (A)

V

Page

44

45

46

47

47

49

49

50

50

52

52

53

54
57
58
59
6O
61
62

63

Table
37.
38.
39.
40.

41.

42..

43.

t Values for

Frequency of

t Values for Automobile Brands

Frequency of

t Values for

Cigarette Brands (A)

Ratings for Automobile Brands .

Ratings for Cigarette Brands (B)

Cigarette Brands (B)

Advertisements for which Null Hypothesis was

Accepted .

Advertisements that did not Conform to Hierarchy

of Beliefs

vi

Page
64
65
66
67

68

7O

73

LIST OF APPENDICES

Appendix Page

A. Advertisements Tested . . . . . . . . . 79

B. Questionnaires . . . . . . . . . . . . 104

vii

Illustration

1. Poor Girl's Rich Watch
2.

3. Rolex . . . .
4. Waltham . . . .
5. Mojud . . . .
6. Hanes . . .

7. Supphose . . .
8. Sears . . . .
9. Early Times . .
10. Chivas Regal . .
ll. Dewars . . . .
12. Seagrams . . .
13. Virginia Slims .
14. Winchester . .
15. Tareyton . . .
16. Salem . . . .
17. Volkswagen . . .
18. MG . . . . .
19. Fiat . . . . .
20. Honda . . . .
21. Benson & Hedges .

LIST OF ILLUSTRATIONS

Are You Paying for Your Watch's Mistakes .

viii

Page

80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100

Illustration Page
22. Camel . . . . . . . . . . . . . . . 101
23. Old Gold . . . . . . . . . . . . . . 102

24. Raleigh . . . . . . . . . . . . . . 103

ix

CHAPTER I

INTRODUCTION

The Problem

 

Many advertisers have attempted to measure the
effectiveness of advertising in order to justify its expense.
This thesis contains a possible means of measuring adver-
tising's effectiveness with particular reference to adver-
tising's ability to attract attention and change brand
preference. It was the purpose of this study to explore
and investigate the effectiveness of specific appeals which
contain the types of beliefs in Milton Rokeach's hierarchy
of beliefs, and to determine whether or not there is a
relationship between the centrality of the belief utilized
in the advertisement and the effectiveness of the advertise-

ment.

Theory
Milton Rokeach defines beliefs as, "inferences made
by an observer about underlying states of expectancy. . . .
Beliefs cannot be directly observed but must be inferred as
best as one can with whatever psychological devices available

from all things the believer says or does."1 A belief system,

 

lMilton Rokeach, Beliefs, Attitudes, and Values: A
Theory of Organization and Change (San Francisco: Jossey-
Bass, Inc., 1968), p. 2.

 

 

1.

as defined by Rokeach, is as "having represented within it
in some organized psychological but not necessarily logical

form, each and every one of a person's countless beliefs

about physical and social reality."2

Rokeach provides a summary of the organization of
beliefs along a central-peripheral dimension in Appendix A
of his book, Beliefs, Attitudes, and Values: A Theory of
Organization and Change, which are:

Type A Beliefs--ca11ed primitive beliefs--are supported
by 100 percent social consensus. Such primitive beliefs
are fundamental, taken-for-granted axioms that are not
subject to controversy because we believe, and we believe
everyone else believes.

Type B Beliefs--which are also extremely resistant to
change. Such beliefs do not depend on social support

or consensus but instead arise from deep personal experi-
ence. Type B beliefs are incontrovertible, and we believe
them regardless of whether anyone else believes them.

Many of these unshakable conceptions are positive ones--
Type 8+ and some are negative ones--Type B-. The positive
ones represent beliefs about what we are capable of, and
the negative ones represent beliefs about what we are
afraid of.

Type C Beliefs--are authority beliefs, beliefs we all
have about which authorities to trust and not to trust.
Many facts of physical and social reality have alterna-
tive interpretations, are socially controversial, or

are not capable of being personally verified or experi-
enced. . . . For these reasons, all men need to identify
with specific authorities . . . to help them to decide
what to believe and what not to believe.

Type D Beliefs--are the beliefs we derive from the
authorities we identify with. Such beliefs can be changed
providing the suggestion for change emanates from one's
authority.

 

21bid.

Type E Beliefs--are inconsequential. If they are

changed, the total system of beliefs is not altered

in any significant way.3
The types of beliefs will be discussed in further detail
later, but it is necessary to elaborate upon the formulation
of Rokeach's theory for clarification.

Rokeach makes three assumptions in the formulation
of his research into the development of the five types of
beliefs which are:

1. Not all beliefs are equally important to the indi-
vidual; beliefs vary along a central-peripheral

dimension.

2. The more central a belief, the more it will resist
change.

3. The more central a belief changed, the more wide-
spread the repercussions in the rest of the belief
system.

From these assumptions, the following hypotheses were

formulated, tested and accepted by Rokeach:

1. Types of beliefs located along a central-peripheral
dimension are functionally distinct.

2. The more central a belief, the more it will resist
change.

3. Changes in central beliefs will produce greater
changes in the rest of the belief system than changes
in less central beliefs.5
Rokeach believes that the importance of a belief to

an individual is the result of connectedness: "the more a

given belief is functionally connected or in communication

 

31bid., pp. 180-181.

4Ibid., p. 3.

51bid., p. 23.

with other beliefs, the more implications and consequences

it has for other beliefs and, therefore, the more central

the belief."6 Rokeach relates four defining assumptions

concerning the criteria of connectedness. They are:

l.

Rokeach.

Existential versus nonexistential beliefs

Beliefs directly concerning one's own existence
and identity in the physical and social world are
assumed to have more functional connections and
consequences for other beliefs than those which
less directly concern one's existence and identity.

Shared versus unshared beliefs about existence and
self-identity

Beliefs concerning existence and self-identity
may be shared or not shared with others. Those
shared with others are assumed to have more functional
connections and consequences for other beliefs than
those not shared with others.

Derived versus underived beliefs

Many beliefs are learned not by direct encounter
with the object of the belief but; indirectly, from
reference persons and groups. Derived beliefs are
assumed to have fewer functional connections and
consequences for other beliefs from which they are
derived.

Beliefs concerning and not concerning matters of
taste

Many beliefs represent more or less arbitrary
matters of taste and are often so perceived by the
individual holding them. Such beliefs are assumed
to have relatively fewer functional connections and
consequences for other beliefs than beliefs that do
not represent arbitrary matters of taste.7

There are three components of a belief according to

They are cognitive, affective, and behavioral, and

he explains these components as "a cognitive component because

it represents a person's knowledge, held with varying degrees,

 

6Ibid., p. 5.

7Ibid.. pp. 5-6.

about what is true or false, good or bad, desirable or un-
desirable; and affective component because under suitable
conditions, the belief is capable of arousing affect of
varying intensity around the object of the belief, around
other objects (individuals or groups), taking a positive or
negative position with respect to the object of the belief,
or around the belief itself when its validity is seriously
questioned, as in an argument, and a behavioral component
because the belief being a response predisposition of
varying threshold must lead to some action when it is suitably
activated."8
Rokeach mentions other means for organizing beliefs
besides in a central-peripheral dimension such as intensity
and verifiability. Intensity of belief, Rokeach believes,
is important in consideration only within each type of belief
rather than as on a whole, and it is considered dubious by
Rokeach to assume that a belief which could be verified is
more important to an individual as opposed to a belief which
cannot.9
It is Rokeach's contention that all beliefs of an
individual except possibly those of taste "are formed and
developed very early in childhood, and undoubtedly the child

first learns them in the context of dealings with his

 

81bid., pp. 113-114.

9Ibid., pp. 12-13.

parents."10 Rokeach further extends his concept of beliefs
in that it "is broader than ideology, containing pre-

ideological as well as ideological beliefs"11

in a system
with an ideology being derived from external authorities
and more or less shared with others and institutionalized.
The types of beliefs are divided into primitive and
non-primitive or authority beliefs. As Rokeach states: "A
person's primitive beliefs represent his 'basic truths'
about physical reality, social reality, and the nature of
the self; they represent a subsystem within the total system

12 Primi-

in which the person has the heaviest commitments."
tive beliefs are those which are learned by direct encounter
with the object or situation of the belief; moreover, "like
all beliefs, conscious or unconscious, they have a personal
aspect: they are rooted in the individual's experience and

in the evidence of his senses. Like all beliefs, they also
have a social aspect with regard to every belief a person
forms, he also forms some notion of how many other people

have the experience and the knowledge necessary to share it
with him and of how close the agreement is among the group."13

Type A beliefs, a person's conception of the self,

are reinforced by unanimous social consensus. They are

 

10Milton Rokeach, The Three Christs of Ypsilanti: A
Psychological Study (New York: Vintage Books, 1964), p. 26.

11

 

Rokeach, Beliefs, Attitudes, and Values, p. 123.

 

lzIbid., p. 6.

l3Rokeach, Three Christs, p. 20.

 

"psychologically incontrovertible because they are rarely,
if ever, experienced as subjects of controversy and therefore
have an axiomatic, taken-for-granted character."l4 If the
validity of this type of belief is challenged, no matter how
impersonal it may be, the effect upon the person is rather
emotional and traumatic since the questioning suggests that
the person is incapable of observing and defining reality
correctly which could result in a Shakedown of all beliefs
connected to this very central belief.

Type B beliefs are also primitive beliefs like Type
A. Type B beliefs do not depend upon universal social con-
sensus like Type A beliefs. Type B beliefs are similar to
Type A beliefs in that they are incontrovertible, learned by
direct encounter with the belief, and involve existence and
self—identity. The Type B beliefs are not shared with others,
and "there are no reference persons or groups outside the

15

self who could controvert such a belief" since "beliefs

that are not shared with others are therefore impervious to

16 As was indicated before,

persuasion or argument by others."
Type B beliefs can be positive to the extent that the person
is capable of viewing himself at one with the world, and

negative to the extent that the person views himself, real

or imagined, not in tune with the things around him.

 

14Rokeach, Beliefs, Attitudes, and Values, p. 6.

15Ibid., p. 8.

16Ibid.

Non-primitive beliefs deal mainly with a person's
authority; it is possible for these beliefs to be
converted because it is possible for the authority to
change. These types of beliefs are generally resistant to
change; however, they are less important and easier to change
than Type A and Type B beliefs. The main difference between
primitive and non-primitive beliefs is that in dealing with
primitive beliefs, the main interest lies in the specific
content of these beliefs. With authority beliefs, it is the
formal content of the belief which is of more interest to
the person in the examination of the beliefs. Rokeach fur-
ther explains: "Non-primitive beliefs are conceived to
develop out of Type A beliefs and to be in a functional
relationship with them. They seem to serve the purpose of
helping the person to round out his picture of the world
realistically and rationally to the extent possible, defen-
sively and irrationally to the extent necessary."17

Type C beliefs are labelled authority beliefs by
Rokeach and are capable of being changed "because the
believer has learned that some of his reference persons and
groups do and some do not share his belief."18 Type C
beliefs are concerned with "not only which authorities could

know but also which authorities would know. . . . The particu-

lar authorities relied on for information differ from one

 

17Ibid., p. 9.

laIbid., p. 10.

person to the next and would depend on learning experiences
within the context of the person's social structure-~family,
class, peer group, religious and political groups, and

19 Edward J. Lessin, in his thesis, "An Investi-

country."

gation of Primitive and Authority Beliefs," points out that

authorities can be either positive or negative. "One accepts

the belief of a positive authority while one rejects the

belief of a negative authority or accepts the contrary

belief."20 He further points out that "an authority

must have some influence on the person's beliefs or

behavior."21 If the person remains indifferent to a

second person's opinions or desires for some kind of

response, then that second person is neither a positive

nor a negative authority. Lessin also presents a con-

tinuum of authority shown in Table 1, "from positive or

accepting, to indifferent, and to negative or rejecting."22
Type D beliefs are labelled derived authority beliefs

by Rokeach and result from the fact that "if we know that a

person believes in a particular authority, we should be able

to deduce many of his other beliefs, those which emanate or

 

lgIbid.

20Edward J. Lessin, "An Investigation of Primitive
and Authority Beliefs" (unpublished M.A. Thesis, Michigan
State University, 1965), p. 10.

ZlIbid., p. 11.

22Ibid.

10

TABLE l.--Continuum of authority.

 

 

 

 

Negative Authority Positive Authority
-3 -2 -l 0 +1 +2 +3
Very Moder- Slight- Indif- Slight- Moder- Very
Reject- ately 1y ferent ly ately Accept-
ing Reject- Reject- Accept- Accept- ing
ing ing , ing ing
derive from the authority he identifies with."23 These types

of beliefs are controvertible and tend to be "derived second
hand through processes of identification with authority rather
than by direct encounter with the object of the belief."24

Type E beliefs are inconsequential beliefs, and they
are considered to be "arbitrary matters of taste."25 These
types of beliefs are considered inconsequential since they
have few if any ties with other beliefs and a change in them
would not result in an uprooting of other beliefs in the

system.

Rokeach's other books, The Open and Closed Mind, The

 

Three Christs of Ypsilanti, and his most recently published

 

book, The Nature of Human Values, all deal with the concept

 

of his belief system; however, they are not as closely

 

23Rokeach, Beliefs, Attitudes, and Values, pp. 10-11.

 

24Ibid., p. 10.

251bid., p. 11.

11

related to the definition of the different types of beliefs

so that they will not be discussed in this study.

Purpose of the Study

 

Rokeach states that advertisements appeal to Type E

26 However, the appeals of most

or inconsequential beliefs.
advertisements contain elements of other types of beliefs.
The purpose of this study is to attempt to determine whether
or not the utilization of a Type A belief in the creative
appeal in the advertisement has a greater effect upon the
reader of the advertisement than a Type 8+ or B- appeal in
the advertisement has a greater effect than Type C, and so
on, conforming to the hierarchy of beliefs in Rokeach's
central-peripheral dimension of beliefs. It is believed
that the more central the belief utilized in the copy plat-
form of the advertisement, the more impact the advertisement
has upon the reader of the advertisement because it is
dealing with beliefs that are a part of or closer to the
self no matter how illogical that the beliefs might be.
Rokeach suggested that, "it is possible to associate
the inconsequential beliefs (Type E) with Type D, or C, or

27 Rokeach points out that adver-

B+, or B-, or A beliefs."
tisers have not employed all of the appeals equally and have

concentrated on associating Type E with Type C and B-

 

26Ibid., p. 183.

27Ibid., p. 184.

12

beliefs. This was as far as he took the relation of his

theory to the field of advertising.

General Hypotheses

 

Perception and Attention

The major purpose of advertising is to communicate a
desired message to the potential target audience so as to
induce a behavioral response. To do this, the target audience
must first notice the advertisement.

There have been numerous theories and writings
published concerning perception and its relation to attention.
The subject is vast, and the psychological literature cannot,
within the scope of this study, be reviewed completely. How—
ever, a discussion of generally accepted knowledge about
perception is necessary in order to utilize the concept in
this study. This is done by presenting a concise meaning of
what is to be measured and why that measurement is considered
most useful for this study.

William H. Ittelson and Hadley Cantril offer a
simplistic definition of perception: "the process by which
a particular person from his particular behavioral center
attributes significances to his immediate environmental

28

situation." Significances can be things, sequences,

 

28William H. Ittelson and Hadley Cantril, Perception:
A Transactional Approach (Garden City, N.Y.: Doubleday &
Company, Inc., 1954), p. 26.

 

13

29

actions, and evaluations. Ittelson and Cantril also point

out that, "We are constantly perceiving simply because it is
an inseparable and necessary part of everything we do."30
Because perception cannot be separated from activity,
researchers in perception have acknowledged that in order
to study perception, one must perceive which poses an inter-
esting paradox.

Ittelson and Cantril present three characteristics
of perception which may be helpful in the definition. The
first characteristic is that perception is a transaction.
"Perceiving never occurs independent of some other activity";31
the second is that perception is unique in that "perception
is always an activity by a unique participant from his own
unique position, providing him with his own unique world of

32 The third characteristic of perception is

experience."
that it is an externalization, or "the things we see and
hear and taste and touch are experienced as existing outside

ourselves and as possessing for themselves the characteristics

which we see in them."33

S. Howard Bartley, in his book, Principles of Per-

 

ception, offers this concept of perception: "The overall

 

291bid., p. 20.

3OIbid., p. 1.

31Ibid., p. 2.

32Ibid., p. 4.

33Ibid.

14

activity of the organism that immediately follows or ac-
companies energistic impingements upon the sense organs,"34
and he presents six relational properties of perception
which are:

l. Symbolism
Perception is symbolic. This is to say that it
is a kind of behavior manifesting an abstract
relation between the organism and its surrounds.
. . . The literal energies convey to the person
something that is not literally inherent in them.

2. Classification
. . . The reaction to the immediate situation is not
as if to a totally isolated and unique one, but
essentially to an example of a class of situations.
. . . This process by which the organism develops
classification is called concept formation.

3. Evaluation
. . . Everything is reacted to in terms of a broad
spectrum at one pole of which is harmful, and at
the other the beneficial and acceptable. In the
middle is the region of indifference.

4. Prognosis and interpretation
In as much as the observor is confronted with
alternatives, the perception may be said to be
prognostic. It is as though it were a bet on the
nature of externality with reference to possible
consequences of action.

5. Internal Self-consistency
Self consistency applies only within perception
and not to the sequence from one perception to
another.

6. Field determinism
Perception is a phenomenon that emerges from a
system of interrelated events, first in the indi-
vidual's surrounds and then withigsthe neuromuscular
system of the individual himself.

 

34S. Howard Bartley, Principles of Perception (New
'York: Harper & Brothers, 1958), p. 22.

35Ibid., pp. 33-36.

15

The fourth and fifth relational properties of per-
ception are especially noteworthy for this study. Prognosis
and interpretation deal with selectivity. The focus of this
study was to determine which advertisement, containing one
of the types of beliefs, the subjects selected to attend.
The fifth relational property, internal self-consistency,
is concerned with object constancy and self constancy which
Rokeach maintains are essentials of primitive beliefs since
they provide a person with the means to define his environ-
ment and his self. Rokeach adds, "Object constancy, more-
over, is not merely a sensory phenomenon as many perception
psychologists have believed. It is a social phenomenon as
well, developed in childhood side by side with person
constancy. The child learns that objects maintain their
identity, and also that other people experience physical
objects as he does."36

From this discussion of perception, it becomes clear
that the measurement desired for this study is not the per-
ception of the advertisements since it is assumed that by
their existence within the subjects' fields of vision, the
advertisements are capable of being perceived. The measure-
ment wanted is the designation of which advertisement draws
the attention of the subjects or which advertisement the
subjects selectively perceive. Magdalen D. Vernon, in an

article entitled, "Perception, Attention, and Consciousness,"

 

36Rokeach, Three Christs, p. 21.

 

16

discusses this selectivity and its application to perception
and relation to attention: "It appears that there are many
degrees or levels of clarity and detail with which we are
aware of our surroundings, varying from a precise and
accurate perception of the environment upon which attention
is concentrated and focused to a vague marginal awareness

"37 It is Vernon's contention

of its less important aspects.
that, "if the eyes are fixated upon a single point in the
field of vision in order that this area may be perceived
in detail, then comparatively little of the surrounding area

38 It is believed that this statement can be

is perceived."
directly applied to the purpose of this study which is to
determine what advertisements the subjects attended to first
with the contention that it was the advertisements containing
elements of Type A beliefs.

The direction and extent of conscious awareness or
attention is affected by certain factors as Vernon indicates
such as: "the general state of alertness or 'vigilance' as
it is sometimes called; by factors of motivation and interest,
by learning and experience, and by features in the environ-

"39

ment itself. This statement relates to Rokeach's contention

that the beliefs which are closest to the individual are the

 

37Magdalen D. Vernon, "Perception, Attention, and
Consciousness," in Attention, ed. by Paul Bakan (Princeton,
N.J.: D. Van Nostrand Company, Inc., 1966), p. 38.

38

 

Ibid.

391bid.

17

beliefs to which the individual will attend. J. Anthony
Deutsch and D. Deutsch, in an article entitled, "Attention:
Some Theoretical Considerations," discuss an experiment in
which the subject is exposed to two auditory messages, one
for each ear, where "if a subject is listening selectively
to one channel and ignoring the other, calling his name on
the rejected channel will on a certain proportion of in-
stances cause him to switch his attention to this channel.
This was explained by assuming that the subject's name had
a higher priority for the filter than to the message to

40 An individual's name is

which he has been attending."
part of identity which is a very central belief and further
supports Rokeach's theory of a central-peripheral dimension
of beliefs and also suggests that attention will be drawn
to more primitive beliefs.

Attention has been defined as a perceptual set that
is "a process which makes one thing seen more readily than

41 It is from this definition and the concept of

another."
perception and its relation to attention as described above
that the following general hypotheses have been formulated

for testing:

 

40J. Anthony Deutsch and D. Deutsch, "Attention:
Some Theoretical Considerations," in Attention, ed. by Paul
Bakan (Princeton, N.J.: D. Van Nostrand Company, Inc.,
1966), p. 211.

41U. S. Department of Health Education and Welfare,
Office of Education Contract Number 3-20-003, Research and
Theory Related to Audio-Visual Transmission, by Robert
Travers (Washington D.C.: Government Printing Office,
1967), P. 228.

 

18

Hypotheses

 

Null Hypothesis
The centrality of a belief used in an advertising
appeal, does not affect whether or not the advertisement
attracts the attention of subjects exposed to it.
Alternate Hypothesis
The centrality of a belief used in an advertising
appeal, does affect whether or not the advertisement
attracts the attention of subjects exposed to it.
Effectiveness as it is used in the hypotheses is
defined as, "adequate to accomplish a purpose; producing

the intended or expected results."42

Opinion Change and Brand Preference
This study was concerned also with the degree to
which the opinions of the subjects towards brand preference
might change after being exposed to the advertisements
containing the different types of beliefs. Rokeach discusses
attitude and opinion change in great detail in his book,
Beliefs, Attitudes, and Values: A Theory of Organization

and Change, and defines attitudes as being, "a relatively

 

enduring organization of interrelated beliefs that describe,
evaluate, and advocate action with respect to an object or
situation, with each belief having cognitive, affective, and

behavioral components."43 Opinion is defined by Rokeach as,

 

42The Random House Dictionary of the English Language,
Rev. ed. (1968), s.v. "effectiveness."

43Rokeach, Beliefs, Attitudes, and Values, p. 132.

19

"a verbal expression of some belief, attitude, or value."44

Rokeach deals with the concept of attitude and behavioral
change and contends that, "whenever a person encounters an
attitude object within a situation, two attitudes, A0 and
AS, are activated; further, a comparison of the relative
impOrtance of these two attitudes is also activated. The
two attitudes are assumed to affect behavior in direct
proportion to their perceived importance with respect to
one another."45
In discussion of research designs, Rokeach states
that, "the classical paradigm employed in the experimental
studies of opinion change--pretest, treatment, posttest--
is not capable of telling us whether an expression of opinion

46 Rokeach

has changed as a result of a particular treatment,"
advocates several posttests to be taken and an analysis of
behavior which is far beyond the means of this study. The
research design to test opinion change is discussed later;
however, it was considered necessary to indicate the type
of change which is measured in the study according to

Rokeach's definition of opinion. The following general

hypotheses have been formulated:

 

44Ibid., pp. 125.

4SIbid., pp. 136-137.

46Ibid., p. 140.

20

Hypotheses

 

Null.Hypothesis

Exposure to advertisements utilizing one of the
four types of beliefs has no effect on the brand
preference of the subjects.

Alternate Hypothesis

Exposure to advertisements utilizing one of the

four types of beliefs has an effect on the brand
preference of the subjects (as expressed by opinion),
especially those advertisements with elements of higher
beliefs, particularly Type A beliefs.

The independent variables were the types of beliefs
contained in the magazine advertisements. Each advertise-
ment in the study contained or violated one of the Types
of beliefs, Type A, B+, B-, C, and D, as defined by Rokeach.
Type E beliefs were not used since the study was concerned
with beliefs used in the copy platform, and few copy platforms
utilize Type E, inconsequential beliefs.

The advertisements used for the testing were 'selective
demand' advertisements which concentrate on increasing market
share by promoting a brand name. As Rokeach indicated:

. . . it could be suggested from a psychological stand-
point that the advertising man has concentrated mainly

on forming and changing Type E beliefs--inconsequential
beliefs--to the extent that his purpose is to meet the
competition, and that he has concentrated mainly on

Type D--derived beliefs--to the extent that his purpose

is to give information. Furthermore, the more competitive
the advertising, the more it seems to address itself to

changing psychologically inconsequential beliefs about
the relative merits of one brand over another.

 

47Ibid., p. 183.

21

The dependent variables were the measurements of
attention and brand preference for the magazine advertise-
ments. The two dependent variables were discussed in the

formulation of the general hypotheses.

CHAPTER II

METHODOLOGY

Research Design
The research design for the study was a 'one group
pretest-posttest with control.‘ The types of beliefs
utilized in magazine advertisements were the independent
variables, and the dependent variables were attention and
brand preference. A diagram of the research design is

presented in Table 2.

TABLE 2.--Diagram of the research design.

 

 

Group Pretest Treatment Posttest
Experimental 01 X 02
Control 01 02

 

The subjects in both the experimental and control groups were
interviewed at the same time on their brand preference for
the pretest. A few days later during scheduled interview
1:hnes, the experimental group received the treatment--
exposure to the advertisements--and were asked to rank the
advemtisements according to which attracted their attention
first, second, third and last. The control group, during

22

23

the scheduled interview, received the posttest. A few days
later, the experimental group received the posttest.

The subjects were assigned to either the control or
experimental group by flipping a coin. Thus, the volunteers
for the experiment became members of either group by chance.

An equal sample size (N=32) for both the experimental and
control group was achieved.

Donald T. Campbell and Julian C. Stanley, in their
book, Experimental and Quasi-Experimental Designs for Research,
consider the 'pretest-posttest with control' to be one of the

48 Testing, maturation, and

history are accounted for by the control group.49 Instru-

three true experimental designs.

mentation is controlled since there is no difference between
the questionnaire for the pretest and the questionnaire for
the posttest. Regression is assumed to occur in both the
experimental and control groups equally, and mortality is
the greatest factor to overcome; however, with repeated
efforts, to ensure that the volunteers for the test took

the posttest, this factor is reduced.

The experiment was conducted by first asking the
subjects in the experimental and control groups to rate their
brand preference for the brands featured in the advertisements

cxnntaining or violating the types of beliefs.

 

48Donald T. Campbell and Julian C. Stanley, Experi-
nmnital and Quasi-Experimental Designs for Research (Chicago:
Rand McNally and Company, 1963), p. 13.

491bid., pp. 13-14.

24

The subjects in the experimental group were asked to
rank on the basis of attention six groups of four advertise-
ments which were in the same or at least similar product
categories. For example, automobile advertisements were
compared with each other and not with cigarette advertisements.
This method was used to prevent ranking on the basis of
product interest or attention qualities instead of on the
basis of the attention qualities of the advertisements to
be tested.

Six different product categories were necessary
because of the inherent limitations of the magazine adver-
tisements available. One of the limitations is that many
advertisers choose to be strong in one medium so that the
kinds of products were limited. For example, cigarette and
liquor products are restricted by law to certain media.
Another limitation was that few advertisers have chosen
to use Type A beliefs in their advertisements. The adver-
tisements used in the test appear in Appendix A.

The rank of 'one' for Type A advertisements was
considered important since the ranking would conform to
the hierarchy of beliefs in Rokeach's central-peripheral
dimension of beliefs. Because of the importance of the
first ranking, it was necessary to determine from the
subjects the reason for the rankings for the Type A adver-
tisements. Direct question probes were used to obtain this

information.

25

All of the six Type A advertisements chosen for the
test violate Type A beliefs in some way. The Poor Girl's
Rich Watch violates a Type A belief by illustrating a foot
with wrist watches on it. If the subjects mentioned this
point, it was concluded that they chose the advertisement
first because of the violation of the Type A belief that it
contains. The Mojud Panty Hose advertisement illustrates
legs growing out of the ground, which is not possible, thus,
if this was mentioned, it was assumed that the subjects
selected the advertisement because of the Type A belief
that it violates. The Early Times advertisement illustrates
a golf club that is bent around a bottle of Early Times so
that any reference to the golf club as the attention getter
would indicate that the subjects had chosen the advertisement
because of Type A beliefs. The Virginia Slims advertisement
illustrates Mount Rushmore with a woman included among the
presidents. Any reference to her inclusion indicated that
the subjects had chosen the advertisement for its Type A
beliefs. The Volkswagen advertisement has a headline
stating: "Today it's bigger than a Cadillac." The "it"
refers to an illustration of a used Volkswagen. The Type A
belief that it violates is that a Volkswagen is obviously
smaller than a Cadillac. If this was mentioned, it was
assumed that the subjects chose the advertisement because
Of the Type A belief. The last advertisement is Benson &

Hedges lOO's which illustrates a green moon man smoking a

26

cigarette and encountering and astronaut. The existence of
the moon man would have to be the object mentioned that
attracted the subjects' attention in order for it to be
attributed to Type A beliefs.

Both the experimental and control group were asked
again to rate their brand preference for the brands featured
in the advertisements containing or violating the types of

beliefs.

Sample

The experimental and the control group of 64 females
were selected from students enrolled in undergraduate
advertising courses at Michigan State University. Since sex
was not one of the independent variables of interest, the
hypotheses could be tested with either males or females as
subjects. It was decided to use females because of one of
the products, panty hose, was appropriate only for females.
Two males were included in the pretest of the questionnaire
to determine if the exclusion of males from the sample was
justified. Both of the males when questioned why they had
given a rank of 'one' to the advertisement that they had
indicated in the 'panty hose' group, reported sexual reasons
for attracting their attention. Both the experimental and
<montrol group were recruited voluntarily, and appointments
for the administration of the questionnaire were set up

individually.

27

Using undergraduate advertising classes as a uni-
verse limits the generalizability of the findings; however,
because the test is an opinion study, the basic requirement
is that the subjects are human beings capable of having
opinions about magazine advertisements.50

It is realized that results from a small sample of
students taking undergraduate advertising classes cannot be
projected to make inference about the major consuming public.
However, if evidence from this study suggests that the two
general alternate hypothesis should be accepted, that is if
there seems to be a relationship between the centrality of
beliefs utilized in the advertisements and their ability to
attract attention and change brand preference, further

research should be done with other appropriate consumer

samples.

Questionnaire

 

The type of questionnaire used for the experiment
was disguised in that the respondents did not know what was
being measured; and therefore, were not biased in their
answers. The questionnaire was structured so as to facili—
tate measurement. (See Appendix B for a copy of the ques-

tionnaire.)

 

50Rokeach also utilizes Michigan State University
students in his studies concerning beliefs. Students were
used in his preliminary study into the organization of
beliefs, in his experimental analysis of belief systems, and
in his race and shared belief experiment performed on campus.

28

The questionnaire for the pretest and the posttest
consisted of two parts, the introduction and the body. The
introduction explained how to complete the questionnaire,
and the body consisted of a seven point Likert scale from
'dislike completely' to 'like completely.‘ Nine product
categories with eight brands within each were included in
the questionnaire to measure brand preference. Four of the
product categories were used as dummies to keep the purpose
of the study disguised.

The pretest questionnaire was administered during
class sessions to females in the undergraduate classes.

The posttest questionnaire was administered to the control
group during the time of the interview. The posttest ques-
tionnaire was administered to the experimental group during
a class session.

The questionnaire for the ranking of the adver-
tisements was administered personally and on individual
basis to the 32 subjects in the experimental group during
the scheduled interview times. The questionnaire consisted
of two parts, an introduction and a body. The introduction
which was read to the subjects explained what was going to
be asked of the subjects, and the body of the questionnaire
consisted of a ranking from 'one' to 'four' on the basis
of attention for the six groups of advertisements. The
questionnaire also included a probing question after each

group of rankings which asked the subjects why they chose

29

the advertisements in the order in which they indicated.
The answer was recorded only if the Type A advertisements
were chosen first. The purpose of the questions was to
determine if the subjects chose the advertisements because
of the Type A beliefs they violated or for some other
reason.

The pretest of the questionnaire consisted of a
sample of ten females and two males. The only alterations
made to the questionnaire as a result of the pretest were
to replace two advertisements because they were ranked on
the basis of graphic considerations rather than on their
types of appeals. The two advertisements replaced were a
Johnnie Walker advertisement replaced by a Chivas Regal
advertisement, and a Datsun advertisement replaced by a MG
advertisement. The Johnnie Walker graphics were a black
starry night dominating the entire advertisement, and the
Datsun graphics were four color while the three advertise-
ments the advertisement was to be compared with were black

and white.

Analysis of the Data

 

The data were processed and analyzed by using a
routine CISSR computer program at Michigan State University.
The analysis provided the frequencies, means, and standard
deviations for the ranking of the advertisements by attention
and for the ratings of the brands. Chi squares, t-tests,

and medians were calculated by hand from the computer data.

30

Attention

The subjects were asked to rank six groups of four
advertisements on the basis of attention. It is assumed
that the frequency of the ranking for each advertisement
would be equal if there was no difference in the attention-
attracting powers of the advertisements. The rankings of
the .advertisements were analyzed by Chi square with the
level of significance for the test at .05 or less.

If the null hypothesis for the Chi square analysis
is rejected, it is still unknown whether or not the ranking
of the advertisements was in accordance with the hierarchy
of beliefs in Rokeach's central-peripheral dimension of
beliefs. A presentation of the rankings in the position
where they are expected to be the highest if they conform
to the hierarchy of beliefs, i.e. a rank of 'one' for Type
A beliefs, is given in tables. The median score, median,
and mean of the ranks were also calculated to determine if
the results conform to the hierarchy of beliefs contained
in Rokeach's central-peripheral dimension of beliefs. The
conformance was determined by inspection.

The probing questions of why the subjects gave a
rank of 'one' to the advertisements which they indicated
were examined to determine if the rankings were because of

Type A beliefs or for some other reason.

31

Brand Preference
The subjects, during the pretest and the posttest,
were asked to rate brands on the basis of how much they
liked or disliked them. The ratings of the brands for both
the experimental and control gangs were compared by a t-test
to determine if there was a significant difference in the
ratings of the experimental group after seeing the adver-

tisements.

Literature Search

 

The literature on advertising's effects on attention
and brand preference is vast. However, a search of the
literature yielded no studies in which Rokeach's beliefs
theory was used as a basis for assessing the effects of

advertising appeals on attention or brand preference.

CHAPTER III

RESULTS

Attention

 

General Hypotheses

Null Hypothesis

 

The centrality of a belief used in an advertising
appeal, does not affect whether or not the advertise-
ment attracts the attention of the subjects exposed
to it.

Alternate Hypothesis

 

The centrality of a belief used in an advertising
appeal, does affect whether or not the advertisement
attracts the attention of the subjects exposed to it.

Attention—Attraction Analysis

A single sample Chi square analysis was performed

on the six groups of four advertisements.

The formula used to compute the Chi square analysis

was:
(0 - E)2

E

O is the observed frequency, and the expected frequency (E)
was determined by dividing the sample size, 32, by the
number of advertisements, four, since it was assumed that

the frequency of the ranking for each advertisement would be

32

33

equal if there was no difference in the attention-attracting
power of the advertisements. The degrees of freedom for
the test are three since there are four columns of adver-
tisements, and it is a single sample Chi square which has
only one row. The level of significance for the test was
set at .05, so that there is a probability of .05 or less
of rejecting the null hypothesis if it is in fact correct.
The sample size for all the tables is 32.

Table 3 presents the frequency of the rankings on
the basis of attention for the 'watch' advertisements which

were tested.

TABLE 3.--Frequency of rankings for watch advertisements.

 

Type Frequency of Ranking
* of For Each Advertisement
Advertisements Belief

Rank of: One Two Three Four

 

Poor Girl's . . . Type A 25 6 O 1
Are you . . . Type B l 13 ll 7
Rolex Type C l 5 8 l7
Waltham Type D 5 7 13 7

 

*
See Appendix A for advertisements identified in this

column.

Table 4 presents the Chi square values calculated
from the observed frequencies in Table 3 for the 'watch'

advertisements.

34

TABLE 4.--Chi square values for the watch advertisements.)

 

 

Advertisements Chi Square Value
Poor Girl's . . . 50.625*

Are you . . . 10.50*
Rolex 22.375*
Waltham 4.5

 

*
An asterisk indicates the values for the advertise-

ments which were significant at the .05 level of probability.

The null hypothesis is rejected for the three advertisements
with asterisks, and the alternate hypothesis is accepted
that the centrality of the belief used made a difference in
the attention-attracting power of each advertisement. The
null hypothesis is accepted for the one advertisement with-
out an asterisk that the type of belief used made no differ-
ence in the attention-attracting power of the advertisement.

Table 5 presents the frequency of the rankings on
the basis of attention for the 'panty hose' advertisements
which were tested.

Table 6 presents the Chi square values calculated
from the observed frequencies in Table 5 for the 'panty
hose' advertisements. The null hypothesis is rejected for
the three advertisements with asterisks, and the alternate
hypothesis is accepted that the centrality of the belief
used made a difference in the attention-attracting power

of each advertisement. The null hypothesis is accepted

TABLE 5.--Frequency of rankings for panty hose advertisements.

 

 

Type Frequency of Ranking

* of For Each Advertisement
Advertisements Belief

Rank of: One Two Three Four

Mojud Type A 22 7 2 1

Hanes' Type B 8 9 ll 4

Supphose Type C l 4 9 18

Sears Type D l 12 10 9

 

*
See Appendix A for advertisements identified in this

column.

TABLE 6.--Chi square vahms for the panty hose advertisements.

 

 

Advertisements Chi Square Value
Mojud 35.2s*
Hanes 3.25
Supphose 20.75*
Sears 8.75*

 

*

An asterisk indicates the values for the advertise-
ments which were significant at the .05 level of probability.

36

for the one advertisement without an asterisk that the type
of belief used made no difference in the attention-attracting
power of the advertisement.

Table 7 presents the frequency of the rankings on
the basis of attention for the 'liquor' advertisements which

were tested.

TABLE 7.--Frequency of rankings for liquor advertisements.

 

 

Type Frequency of Ranking
* of For Each Advertisement
Advertisements Belief
Rank of: One Two Three Four
Early Times Type A l4 l3 4 l
Chivas Regal Type B 16 12 4 O
Dewars Type C 2 4 7 l9
Seagrams Type D 0 3 17 12

 

*
See Appendix A for advertisements identified in this
column.

Table 8 presents the Chi square values calculated
from the observed frequencies in Table 7 for the 'liquor'
advertisements. The null hypothesis is rejected for all
four of the advertisements, and the alternate hypothesis
is accepted that the centrality of the belief used made a
difference in the attention-attracting power of each adver-

tisement.

37

TABLE 8.--Chi square values for the liquor advertisements.

 

 

Advertisements Chi Square Value
Early Times 15.75*
Chivas Regal 20.0*
Dewars 21.75*
Seagrams 23.25*

 

*
An asterisk indicates the values for the advertise-
ments which were significant at the .05 level of probability.

Table 9 presents the frequency of the rankings on the
basis of attention for the 'cigarette' advertisements which

were tested.

TABLE 9.--Frequency of rankings for cigarette advertisements
(A).

 

Type Frequency of Ranking
* of For Each Advertisement

Advertisements Belief
Rank of: One Two Three Four

 

Virginia Slims Type A 7 12 9 4
Winchester Type B 8 8 4 12
Tareyton Type C 8 5 6 12
Salem Type D 9 6 l3 4

 

*
See Appendix A for advertisements identified in this
column.

‘1' li

38

Table 10 presents the Chi square values calculated
from the observed frequencies in Table 9 for the 'cigarette'

advertisements.

TABLE 10.--Chi square values for cigarette advertisements (A).

 

 

Advertisements Chi Square Value
Virginia Slims 4.25
Winchester 4.0
Tareyton 3.625
Salem 5.75

 

The null hypothesis is accepted for all four of the adver-
tisements that the type of belief used made no difference
in the attention-attracting power of the advertisements.
Table 11 presents the frequency of the rankings on
the basis of attention for the 'automobile' advertisements

which were tested.

TABLE ll.-—Frequency of rankings for automobile advertisements.

—_
.___._._.— 0“ . —— —.——.-—-——.- . _._—_

Type Frequency of Ranking
. * of For Each Advertisement
Advertisements Belief

Rank of: One Two Three Four

 

Volkswagen Type A 12 15

MG Type B 20 10

Fiat Type C 0 4 20

Honda Type D 0 3 6 23

*
See Appendix A for advertisements identified in this
column.

39

Table 12 presents the Chi square values calculated
from the observed frequencies in Table 11 for the 'automobile'

advertisements.

TABLE 12.--Chi square values for automobile advertisements.

 

 

Advertisements Chi Square Value
. *
Volkswagen 16.25
*
MG 31.0
*
Fiat 28.0
*
Honda 39.75

 

*
An asterisk indicates the values for the advertise-
ments which were significant at the .05 level of probability.

The null hypothesis is rejected for all four of the adver-
tisements, and the alternate hypothesis is accepted that
the centrality of the belief used made a difference in the
attention-attracting power of the advertisements.

Table 13 presents the frequency of the rankings on
the basis of attention for the 'cigarette' advertisements
which were tested.

Table 14 presents the Chi square values calculated
from the observed frequencies in Table 13 for 'cigarette'
advertisements. The null hypothesis is rejected for all
four of the advertisements, and the alternate hypothesis
is accepted that the centrality of the belief used made a
difference in the attention-attracting power of the adver-

tisements.

40

TABLE l3.--Frequency of rankings for cigarette advertisements

(B).

 

 

 

Type Frequency of Ranking
* of For Each Advertisement
Advertisements Belief
Rank of: One Two Three Four
Benson & Hedges Type A 22 5 5 0
Camel Type B 7 21 2 2
Old Gold Type C 2 4 18 8
Raleigh Type D l 2 7 22
*
See Appendix A for advertisements identified in this
column.

TABLE l4.--Chi square values for cigarette advertisements (B).

 

Advertisements

Chi Square Value

 

Benson & Hedges
Camel
Old Gold

Raleigh

*
34.75

*
30.25
*

19.0
*
24.5

 

*

An asterisk indicates the values for the advertise-
ments which were significant at the .05 level of probability.

41

A compilation of the Chi square values and an
indication of whether or not the null hypothesis for the
Chi square test was accepted appears in Table 15 on page
42.‘ Out of the 24 advertisements which were ranked on
the basis of attention, 18 had Chi square values which
indicated that the type of belief used made a difference
in each advertisement's attention-attracting power. There-
fore, it can be concluded that the type of belief used made
a difference in the attention-attracting power of the adver-

tisements tested.

Hierarchy of Beliefs Analysis

The difference in the rankings on the basis of
attention as demonstrated by the Chi square analysis could
be the result of other elements in the advertisements besides
the beliefs that they contain or violate. Therefore, it is
necessary to determine if the ranking of the advertisements
on the basis of attention conforms to the hierarchy of beliefs
in Rokeach's central-peripheral dimension of beliefs. This
determination was accomplished by testing the following
hypotheses:

Null hypothesis: The ranking of the advertisements

 

on the basis of attention does not conform to the hierarchy
of beliefs in Rokeach's central-peripheral dimension of

beliefs.

Alternate hypothesis: The ranking of the adver-

 

tisements on the basis of attention does conform to the

42

TABLE 15.--Compilation of Chi square values for the 24
‘ advertisements tested.

 

 

 

 

 

 

Advertisements** Chi Square Value
Watches
Poor Girl's . . 50.625*
Are you . . . 10.50*
Rolex 22.375*
Waltham 4.5
Panty hose
.Mojud 35.25*
Hanes 3.25
Supphose 20.75*
Sears 8.75*
Liquor
Early Times 15.75*
Chivas Regal 20.0*
Dewars 21.75*
Seagrams 23.25*
Cigarettes (A)
Virginia Slims 4.25
Winchester 4.0
Tareyton 3.625
Salem 5.75
Automobiles
Volkswagen 16.25*
MG 31.0*
Fiat 28.0*
Honda 39.75*
Cigarettes (B)
Benson & Hedges 34.75*
Camel 30.25*
Old Gold l9.0*
Raleigh 24.5*

 

*

An asterisk indicates the values for the advertisements

which were significant at the .05 level of probability.

**See Appendix A for advertisements identified in this

column.

43

hierarchy of beliefs in Rokeach's central-peripheral dimension
of beliefs.

If the advertisements do conform to the hierarchy
of beliefs, it would be expected that the Type A advertise-
ments would most frequently be ranked first; the Type B
advertisements most frequently ranked second; the Type C
advertisements most frequently ranked third, and the Type D
advertisements would most frequently be ranked fourth.

The calculations of the median score, median, and
mean of the ranks is presented to determine if there is any
conformance to the hierarchy of beliefs in Rokeach's central-
peripheral dimension of beliefs and the ranking of the
advertisements on the basis of attention. The conformance
was determined by inspection.

Table 16 presents the rankings for the 'watch' adver-
tisements taken from Table 3 on page 33 of Chapter III. The
rankings are presented where they would most frequently be
ranked if the advertisements conformed to the hierarchy of
beliefs in Rokeach's central-peripheral dimension of beliefs.

Table 17 presents the median score, median, and mean
of the ranks for the 'watch' advertisements. The null
hypothesis is rejected for the Type A and B advertisements,
and the alternate hypothesis is accepted that the ranking of
the advertisements on the basis of attention does conform to
the hierarchy of beliefs in Rokeach's central-peripheral

dimension of beliefs. The null hypothesis is accepted for

44

TABLE l6.--Ranking of watch advertisements and hierarchy of

 

beliefs.
Type Frequency of Ranking
* of For Each Advertisement
Advertisements Belief

Rank of: One Two Three Four

 

Poor Girl's . . . Type A 25

Are you . . . Type B 13

Rolex Type C 8 (l7)**
Waltham Type D (13)** 7

 

*
See Appendix A for advertisements identified in this

column.
**
The numbers in parentheses indicate the highest

rankings when they did not conform to the hierarchy of beliefs.

TABLE l7.--Central tendency measurements for watch advertise-

 

 

ments.

Type Mean

of Median of
Advertisements Belief Score Median Ranks
Poor Girl's . . . Type A 1 .86 1.2813
Are you . . . Type B 3 3.318 2.75
Rolex Type C 4 4.44 3.2813
Waltham Type D 3 3.192 2.6875

 

45

the Type C and D advertisements that the ranking of the
advertisements on the basis of attention does not conform
to the hierarchy of beliefs in Rokeach's central-peripheral
dimension of beliefs.

Table 18 presents the rankings for the 'panty hose'
advertisements taken from Table 5 on page 35 of Chapter III,
The rankings are presented where they would most frequently
be ranked if the advertisements conformed to the hierarchy
of beliefs in Rokeach's central-peripheral dimension of

beliefs.

TABLE 18.--Ranking of panty hose advertisements and hierarchy

 

of beliefs.
Type Frequency of Ranking
* of For Each Advertisement
Advertisements Belief

Rank of: One Two Three Four

 

Mojud Type A 22

Hanes Type B 9 (ll)**
Supphose Type C 9 (18)**
Sears Type D (12)** 9

 

*
See Appendix A for advertisements identified in this
column.

**
(Nuanumbersin parentheses indicate the highest
rankings when they did not conform to the hierarchy of beliefs.

Table 19 presents the median score, median, and mean

of the ranks for the 'panty hose' advertisements. The null

46

TABLE l9.--Central tendency measurements for panty hose

 

 

advertisements.

Type Mean

of Median . ~. of
Advertisements Belief Score Median Ranks
Mojud Type A 1 .77 1.2813
Hanes Type B 2 1.61 2.3438

.Supphose Type C 4 4.389 3.375
Sears Type D 3 3.2 2.8438

 

hypothesis is rejected for the Type A and B advertisements,
and the alternate hypothesis is accepted that the ranking

of the advertisements on the basis of attention does conform
to the hierarchy of beliefs in Rokeach's central-peripheral
dimension of beliefs. The null hypothesis is accepted for
the Type C and D advertisements that the ranking of the
advertisements on the basis of attention does not conform

to the hierarchy of beliefs in Rokeach's central-peripheral
dimension of beliefs.

Table 20 presents the rankings for the 'liquor'
advertisements taken from Table 7 on page 36 in Chapter III.
The rankings are presented where they would most frequently
be ranked if the advertisements conformed to the hierarchy
of beliefs in Rokeach's central-peripheral dimension of
beliefs.

Table 21 presents the median score, median, and mean

of the ranks for the 'liquor' advertisements. The null

47

TABLE 20.--Ranking of liquor advertisements and hierarchy
of beliefs.

 

Type Frequency of Ranking
of For Each Advertisement
Advertisements* Belief

Rank of: One Two Three Four

 

Early Times Type A 14

Chivas Regal Type B (16)** 12

Dewars Type C 7 (l9)**
Seagrams Type D (l7)* 12

 

*
See Appendix A for advertisements identified in
this column.

**
{Huenumbersin parentheses indicate the highest
rankings when they did not conform to the hierarchy of
beliefs.

TABLE 21.--Centra1 tendency measurements for liquor adver-
tisements.

 

 

Type Mean
of Median of
Advertisements Belief Score Median Ranks
Early Times Type A 2 2.346 1.75
Chivas Regal Type B 2 2.5 1.625
Dewars Type C 4 4.342 3.3438

Seagrams Type D 3 3.42 2.375

 

48

hypothesis is tentatively rejected for the Type A and B
advertisements, and the alternate hypothesis is tentatively
accepted that the ranking of the advertisements on the basis
of attention does conform to the hierarchy of beliefs in
Rokeach's central-peripheral dimension of beliefs. The null
hypothesis is accepted for the Type C and D advertisements
that the ranking of the advertisements on the basis of atten-
tion does not conform to the hierarchy of beliefs in Rokeach's
central-peripheral dimension of beliefs.

Table 22 presents the rankings for the 'cigarette'
advertisements taken from Table 9 on page 37 in Chapter III.
The rankings are presented where they would most frequently
be ranked if the advertisements conformed to the hierarchy
of beliefs in Rokeach's central-peripheral dimension of
beliefs.

Table 23 presents the median score, median, and mean
of the ranks for the 'cigarette' advertisements. The null
hypothesis is accepted for all four of the advertisements
that the ranking of the advertisements on the basis of
attention does not conform to the hierarchy of beliefs in
Rokeach's central-peripheral dimension of beliefs.

Table 24 presents the rankings for the 'automobile'
advertisements taken from Table 11 on page 38 in Chapter
III. The rankings are presented where they would most
frequently be ranked if the advertisements conformed to the
hierarchy of beliefs in Rokeach's central-peripheral dimension

of beliefs.

49

TABLE 22.--Ranking of cigarette advertisements (A) and
hierarchy of beliefs.

 

Type Frequency of Ranking
* of For Each Advertisement
Advertisements Belief

Rank of: One Two Three Four

 

Virginia Slims Type A 7 (12)**

Winchester Type B 8 (12)**
Tareyton Type C 6 (12)**
Salem Type D (l3)** 4

 

* _
See Appendix A for advertisements identified in

this column.

**
The numbers in parentheses indicate the highest
rankings when they did not conform to the hierarchy of
beliefs.

TABLE 23.--Central tendency measurements for cigarette
advertisements (A).

 

 

Type Mean

of Median of
Advertisements Belief Score Median Ranks
Virginia Slims Type A 2 1.75 2.3125
Winchester Type B 3 3.0 2.625
Tareyton -Type C 3 3.167 2.6875

Salem Type D 3 3.42 2.375

 

50

TABLE 24.--Ranking of automobile advertisements and hierarchy
of beliefs.

 

Type Frequency of Ranking
* of For Each Advertisement
Advertisements Belief

Rank of: One Two Three Four

 

**
Volkswagen Type A 12 (15)
*1-
MG Type B (20) 10
Fiat Type C 20
Honda Type D 23

 

*
See Appendix A for advertisements identified in

this column.

**
The numbers in parentheses indicate the highest

rankings when they did not conform to the hierarchy of
beliefs.

Table 25 presents the median score, median, and mean of the

rankszfixrthe 'automobile' advertisements. The null hypothesis

TABLE 25.--Central tendency measurements for automobile

 

 

advertisements.
Type Mean
of Median of
Advertisements Belief Score Median Ranks
Volkswagen Type A 2 2.234 1.8125
MG Type B 1 .7 1.4375
Fiat Type C 3 2.9 3.125

Honda Type D 4 4.196 3.625

 

51

is accepted for the Type A and B advertisements that the
ranking of the advertisements on the basis of attention does
not conform to the hierarchy of beliefs in Rokeach's central-
peripheral dimension of beliefs. The null hypothesis is
rejected for the Type C and D advertisements, and the
alternate hypothesis is accepted that the ranking of the
advertisements on the basis of attention does conform to

the hierarchy of beliefs in Rokeach's central-peripheral
dimension of beliefs.

Table 26 presents the rankings for the 'cigarette'
advertisements taken from Table 13 on page 40‘ in Chapter
III. The rankings are presented where they would be most
frequently ranked if the advertisements conformed to the
hierarchy of beliefs in Rokeach's central-peripheral dimension
of beliefs.

Table 27 presents the median score, median, and mean
of the ranks for the 'cigarette' advertisements. The null
hypothesis is rejected for all four of the advertisements,
and the alternate hypothesis is accepted that the ranking
of the advertisements does conform to the hierarchy of beliefs
in Rokeach's central-peripheral dimension of beliefs.

Table 28 on page 53 presents the compilation of
the hierarchy of beliefs and whether or not the null
hypothesis was accepted or rejected for all of the adver-
tisements. From the data, it can be concluded that the ranking

of the advertisements on the basis of attention does conform

52

TABLE 26.--Ranking of cigarette advertisements (B) and
hierarchy of beliefs.

 

Type Frequency of Ranking
* of For Each Advertisement
Advertisements Belief

Rank of: One Two Three Four

 

BensOn & Hedges Type A 22

Camel Type B 21

Old Gold Type C 18
Raleigh Type D 22

 

*
See Appendix A for advertisements identified in this
column.

TABLE 27.--Central tendency measurements:fimrcigarette
advertisements (B).

 

 

Type Mean

of Median of
Advertisements Belief Score Median Ranks
Benson & Hedges Type A 1 1 1.4688
Camel Type B 2 2.07 1.9688
Old Gold Type C 3 2.94 3.0

Raleigh Type D 4 4.227 3.5625

 

53

TABLE 28.--Compilation of hierarchy of beliefs analysis.

 

 

 

 

 

 

Advertisements* Type of Belief Null Hypothesis
Watches
Poor Girl's Type A Rejected
Are you . . . Type B Rejected
Rolex Type C Accepted
Waltham Type D Accepted
Panty hose
Mojud Type A Rejected
Hanes Type B Rejected
Supphose Type C Accepted
Sears Type D Accepted
Liquor
Early Times Type A Rejected
Chivas Regal Type B Rejected
Dewars Type C Accepted
Seagrams Type D Accepted
Cigarettes (A)
Virginia Slims Type A Accepted
Winchester Type B Accepted
Tareyton Type C Accepted
Salem Type D Accepted
Automobiles
Volkswagen Type A Accepted
MG Type B Accepted
Fiat Type C Rejected
Honda Type D Rejected
Cigarettes (B)
Benson & Hedges Type A Rejected
Camel Type B Rejected
Old Gold Type C Rejected
Raleigh Type D Rejected

 

*

column.

See Appendix A for advertisements identified in this

54

to the hierarchy of beliefs in Rokeach's central-peripheral
dimension of beliefs.
Analysis of Reasons for First Rankings
of Type A Advertisements

When the subjects ranked the advertisements with
Type A beliefs first, they were asked why they did so. It
was recorded whether the ranking of the advertisement was
because of the Type A belief in the advertisements or for
some other reason. Table 29 lists the six Type A advertise-
ments and whether or not the subjects ranked the advertise-

ments first because of the Type A beliefs.

TABLE 29.--Six Type A advertisements and reason for first

 

 

ranking. '
Number of Because of Because of
* First Type A Some Other
Advertisements Rankings Beliefs Reason
Poor Girl's . . . 25; 25 0
Mojud 22 22 0
Early Times 14 14 0
Virginia Slims 7 7 0
Volkswagen 12 l 11
Benson & Hedges 22 21 l

 

* I 0
See Appendix A for advertisements identified in this
column.

55

It is apparent from the table that all of the advertisements,
except for the Volkswagen advertisement, were ranked first
because of the type of belief that they violate. From this
conclusion and on the basis of the data presented in the

Chi square analysis and in the hierarchy of beliefs analysis,
it can be concluded that the centrality of a belief used in
an advertising appeal, does affect whether or not the adver-
tisement attracts the attention of the subjects exposed to

it.

Brand Preference

 

General Hypotheses

Null Hypothesis

 

Exposure to advertisements utilizing one of the
four types of beliefs has no effect on the brand
preference of the subjects.

Alternate Hypothesis

Exposure to advertisements utilizing one of the
four types of beliefs has an effect on the brand
preference of the subjects (as expressed by opinion),
especially those advertisements with elements of
higher beliefs, particularly Type A beliefs.

Testing the Difference Between
Two Correlated Means

The ratings of the brands for both the experimental
and control group are compared by a t-test to determine if

there is a difference in the ratings of the experimental

group after seeing the advertisements.

56

The formula used to determine the difference in the

brand preference was:

XDVN - l

JEZDZ - (ZD)2

 

t:

 

D is the difference of the ratings between the pretest and
the posttest. These differences were computed by taking

the difference of the ratings in the pretest from the

ratings in the posttest for each of the 32 subjects in the
control group and each of the 32 subjects in the experimental
group. For example, if a subject rated one of the brands

as 'neither like nor dislike' in the pretest (a rating of
four), and then rated the same brand 'like completely' in

the posttest (a rating of seven), the difference in the
ratings was recorded as plus three.

The degrees of freedom for the test is 31 (32-1),
and the level of significance is .05. The critical value
for .05 with a sample of over 30 and a two-tail test is 1.96.
The sample size for all the tables is 32.

Table 31 on page 58 presents the frequency of the
ratings for the three watch brands in the pretest and the
posttest, for both the control and experimental groups. The
ratings range from 'liking' to 'disliking.‘

Table 30 presents the values for t for the watch

brands. The null hypothesis is accepted that exposure to

57

TABLE 30.--t values for watch brands.

 

 

 

t Values
1:
Brands Experimental Control
Bulova .366 .21
Rolex 0.0 1.913
Waltham 1.4326 .60375

 

*
See Appendix A for advertisements of brands identi-

fied in this column.

advertisements utilizing one of the four types of beliefs has
no effect on the brand preference of the subjects.

Table 32 on page 59 presents the frequency of the
ratings for the four panty hose brands in the pretest and
the posttest, for both the control and experimental groups.
The ratings range from 'liking' to 'disliking.'

Table 33 presents the values for t for the panty hose
brands. The null hypothesis is accepted for the Supphose
and Hanes brands that exposure to advertisements utilizing
one of the four types of beliefs has no effect on the brand
preference of the subjects. The null hypothesis is rejected
for the Mojud and Sears brands, and the alternate hypothesis
is accepted that exposure to advertisements utilizing one of
the four types of beliefs has an effect on the brand prefer-
ence of the subjects.

Table 34 on page 61 presents the frequency of the

ratings for the four liquor brands in the pretest and the

58

TABLE 31.--Frequency of ratings for watch brands.

 

Dislike Dislike Dislike Neither Like Like Like
Com— fixmr- It Like Nor I\ Some- (11w-

 

 

 

 

Brand pletely ‘what Little Dislike Little what pletely

IHEOWA
Experimental

Pretest 12 14

Posttest 9 5 12
Control

Pretest ll 4 8

Posttest 1 10 2 10
ROLEX
Ebcperinental

Pretest 25 2

Posttest l l 24
Control

Pretest 26

Posttest 21 4
WAETHAM
Experimental

Pretest 26

Posttest 23 4
Control

Pretest 1 26 2 2 1

Posttest 25 3 2 2

 

59

TABLE 32.--Frequency of ratings for panty hose brands.

 

Dislike Dislike Dislike Neither Like Like Like

 

 

 

 

 

(knr Samr- A LflanOr .A Samr-(xmr
Brand pletely what Little Dislike little what pletely
MORE)
Experimntal
Pretest 2 1 27
Posttest 2 2 23 3 2
Contnql
Pretest l 26 2 3
Posttest l l 26
HANES
Experimental
Pretest l 1 15
Posttest 1 13
Control
Pretest 1 l4
Posttest 1 l 1 12
SUPPHOSE
Experimental
Pretest 4 l 18
Posttest 3 4 17
Control
Pretest 2 l 19
Posttest 3 l 2 19 5
SEARS
Experimental
Pretest l 1 2 22
Posttest 2 2 1 16
Control
Pretest 1 21

Posttest 1 1 4 18 4 3

 

60

TABLE 33.--t values for panty hose brands.

 

 

 

t Values
Brands* Experimental Control
Mojud 2.518** .6858
Hanes 1.1588 .95
Supphose .9442 .3340
Sears 2.1968** .5730

 

*
See Appendix A for advertisements of brands identi-
fied in this column.

**
The asterisks indicate the brands which were
significant at the .05 level of probability.

posttest, for both the control and experimental groups. The
ratings range from 'liking' to 'disliking.'

Table 35 presents the values for t for the liquor
brands. The null hypothesis is accepted for the four liquor
brands that exposure to advertisements utilizing one of the
four types of beliefs has no effect on the brand preference
of the subjects.

Table 36 on page .63 presents the frequency of the
ratings for the four cigarette brands in the pretest and the
posttest, for both the control and experimental groups. The
ratings range from 'liking' to 'disliking.'

Table 37 presents the values for t for the cigarette
brands. The null hypothesis is accepted for the four

cigarette brands that exposure to advertisements utilizing

61

TABLE 34.--Frequency of ratings for liquor brands.

 

Dislike Dislike Dislike Neither Like Like Like
Oaw- Smmr A LfianOr .A Samr-<1mr

 

 

 

 

 

Brand pletely' what Little Dislike Little what pletely

EMEN'EWES
Eacperimntal

Pnaxst 4 2 23 2

Posttest 6 2 l9 4 1
ContrOl

Pretest 5 1 23

Posttest 4 l 21 2 3 1
CHIVAS REGAL
Ebcperimntal

Pretest 2 19 3

Posttest 7 l 15
Cbntrol

Pretest 4 17 l 2 8

Posttest 4 16 3
DEWXRS
Beerinental

Pretest 4 1 22

Posttest 5 2 l 16 4 l 3
Control

Pretest 5 1 l9

Posttest 5 3 17 3 2 2
SEAGRAMS
Experimental

Pretest 3 l 9 5 6 8

Posttest 5 7 2 8 10
Cbntrol

Pretest 4 9 3

Posttest 3 l l

 

62

TABLE 35.--t values for liquor brands.

 

 

 

t Values
Brands* Experimental Control
Early Times .1401 1.572
Chivas Regal .2726 1.469
Dewars .1043 .9161
Seagrams .2726 .3569

 

*
See Appendix A for_advertisements of brands identi-

fied in this column.

63

TABLE 36.-~Frequency of ratings for cigarette brands (A).

 

Dislike Dislike Dislike Neither Like Like Like
Cknr- Some- .A Like NOr A. Some- Cbmr
Brand pletely’ what Little Dislike Little what pletely

 

'VIRGINIA.SLIMS

 

 

 

 

Experimental

Pretest 15

Posttest 14 3 3 6 2 1
Control

Pretest 13 l 10

Posttest l7 4 l 5
WHNCHESTER
Experimental

Pretest 19 12

Posttest 20 3 7 2
Control

Pretest 17 2 13

Posttest 21 l 2 8
TAREYTON
Experimental

Pretest 16 2 10 2 2

Posttest l9 3 l 5 2 2
Control

Pretest 15 2 14 l

Posttest l9 3 1 7 2
SALEM
Experimental

Pretest 17 2 10

Posttest l3 1 5 5 l 3
Control

Pretest 13 l l 12

Posttest 15 2 5 3 3

 

64

TABLE 37.--t values for cigarette brands (A).

 

 

 

t Values
Brands* Experimental Control
Virginia Slims .3169 -2.0189
Winchester .37899 .5545
Tareyton -.l348 -.2235
' Salem 1.777 .2297

 

*
See Appendix A for advertisements of brands identi-
fied in this column.

one of the four types of beliefs has no effect on the brand
preference of the subjects.

Table 38 on page 65 presents the frequency of the
ratings for the four automobile brands in the pretest and
the posttest, for both the control and experimental groups.
The ratings range from 'liking' to 'disliking.'

Table 39 presents the values for t for the automobile
brands. The null hypothesis is accepted for the four auto-
mobile brands that exposure to advertisements utilizing one
of the four types of beliefs has no effect on the brand
preference of the subjects.

Table 40 on page 67' presents the frequency of the
ratings for the four cigarette brands in the pretest and
the posttest, for both the control and experimental groups.

The ratings range from 'liking' to 'disliking.'

65

TABLE 38.--Frequency of ratings for automobile brands.

 

Dislike Dislike Dislike Neither Like Like Like
Cknr- Ekner- A. Like NOr A. Some- Cbmr

 

 

 

 

 

Brand pletely what Little Dislike Little what pletely
VOLKSWAGEN
Ekperinental
Pretest ’ 3 2 3 5 11 8
Posttest 3 2 3 6 8 10
CbntrOl
Pretest 2 l 2 4 4 10 9
Posttest 2 2 3 5 8 12
DI;
ERperimental
Pretest 5 6 8 l3
Posttest 4 5 9 14
Control
Pretest 4 6 9 l3
Posttest l 8 5 4 14
FIAT
Ebtperilrental
Pretest l 1 7 7 6 10
Posttest l 2 l 6 7 9 6
Control
Pretest 2 8 8 5 9
Posttest 4 6 9 l3
I-DNDA
Experimental
Pretest 6 4 ll 4 4 3
Posttest 8 2 4 7
Control
Pretest 6 3 l 10 6

Posttest 5 5 2 8

 

66

TABLE 39.--t values for automobile brands.

 

 

 

t Values
Brands* Experimental Control
Volkswagen .928 1.784
MG .3252 1.1317
Fiat .525 1.597
Honda .519 1.256

 

*
See Appendix A for advertisements of brands identi-
fied in this column.

67

TABLE 40.--Frequency of ratings for cigarette brands (B).

 

Dislike Dislike Dislike Neither Like Like Like
Cknr- SbflEF A. Like Nor I; Ekrmr- Cbmr
Brand pletely what Little Dislike Little what pletely

 

BENSON & HEDGES

 

 

 

 

Ecperimantal
Pretest 14 2 9 3 3 1
Posttest 15 1 3 6 3 2 2
Control
Pretest 12 l 11 4 2 2
Posttest 15 3 7
CAMEL
Ebrperimntal
Pretest 19 l 2 8 2
Posttest 18 3 6 3 1 1
Control
Pretest 19 1 1 ll
Posttest 19 2 l 7 2 1
OLD GOLD
Experimental
Pretest 18 2 12
Posttest 18 3 8 2 1
Control
Pretest 16 1 15
Posttest 18 l 2 9 l 1
RALEIGH
Experimental
Pretest 20 l 10 1
Posttest 20 4 7 1
Control
Pretest 16 1 14 1

Posttest 21 1 1 8 1

 

68

Table 41 present the values for t for the cigarette
brands. The null hypothesis is accepted for all four of the
cigarette brands that exposure to advertisements utilizing
one of the four types of beliefs has no effect on the brand

preference of the subjects.

TABLE 41.--t values for cigarette brands (B).

 

 

 

t Values
Brands* Experimental Control
Benson & Hedges 1.281 .2293
Camel -.l459 -.7017
Old Gold .9407 0.0
Raleigh 1.675 -3.772

 

*
See Appendix A for advertisements of brands identi-

fied in this column.

On the basis that only two of the advertisements
tested had significant t values, it is concluded that
exposure to advertisements utilizing one of the four types
of beliefs has no effect on the brand preference of the

subjects.

CHAPTER IV

SUMMARY OF CONCLUSIONS

Attention

 

General Hypotheses

Null Hypothesis

 

The centrality of a belief used in an advertising
appeal, does not affect whether or not the advertise-
ment attracts the attention of the subjects exposed to
it.

Alternate Hypothesis

 

The centrality of a belief used in an advertising
appeal, does affect whether or not the advertisement
attracts the attention of the subjects exposed to it.

On the basis of the evidence presented in Chapter III,
the null hypothesis is rejected and the alternate hypothesis
is accepted. The centrality of a belief used in an adver-
tisement does affect whether or not the advertisement
attracts the attention of subjects, and the effects conform

to the hierarchy of beliefs in Rokeach's central-peripheral

dimension of beliefs.

Attention-Attraction Analysis
Although the evidence is strong (18 of 24 instances)

that the alternate hypothesis should be accepted, it seemed

69

7O

desirable to analyze the other six advertisements to ascertain
if there were any reasonable explanations as to why the
'findings were different. The data on these six advertise-

ments are summarized in Table 42.

TABLE 42.--Advertisements for which the null hypothesis was

 

 

accepted.
Type
* of Group Chi Square

Advertisements Belief Number Value
Virginia Slims Type A 4 4.25
Hanes Type B 2 3.25
Winchester Type B 4 4.0
Tareyton Type C 4 3.625
Waltham Type D l 4.5
Salem Type D 4 5.75

 

*
See Appendix A for advertisements identified in this
column.

In all four of the advertisements in Group four the
centrality of the belief used in the advertisement did not
make any difference in their attention-attracting powers. A
plausible explanation for this result is that the Type A
belief in the Virginia Slims advertisement was more subtle
and required greater observational acuity of the subjects
to spot it in a cursory glance. The type of the product
and the familiarity with the themes of the cigarette

advertisements such as the Tareyton advertisement may also

71

have been a factor in the ranking. Another explanation for
the finding about this group of advertisements which the
author did not realize but which was pointed out by many
subjects during the experiment was that the Salem adver-
tisement utilizing a Type D belief, also violated a Type A
belief since it illustrated a man smoking a cigarette with
his head wet and his body submerged in water which violates
a Type A belief that fire is extinguished by water.

The lack of significant Chi square values for the
Hanes and Waltham advertisements cannot be explained by
using Rokeach's beliefs theory; however, the Chi square
values were significant for the other three advertisements
with which they were ranked, and the advertisements do con-
form, for the most part, with the hierarchy of beliefs.

As already indicated, the type of product and
familiarity with advertising themes were also factors in
attracting the attention of the subject rather than the
appeals in the advertisements. A possible means to avoid
the attraction on the basis of product familiarity would
be to use only one brand and create four advertisements
utilizing one of each of the four beliefs. By using only
one brand, it could result in a more definite determination
of the effectiveness of the advertisements on the basis of
attention and the beliefs which the advertisements utilize.

The findings seem to indicate that the more out-
standing the violation of a Type A belief, the more likely

it will catch the attention of its target market. Product

72

familiarity along with the violation of a Type A belief

seems to be a strong combination in attracting attention.

Hierarchy of Beliefs Analysis

Table 43 presents the advertisements which did not
conform to the hierarchy of beliefs in Rokeach's central-
peripheral dimension of beliefs by their ranking on the
basis of attention.

The Virginia Slims, Winchester, Tareyton, and Salem
advertisements did not conform to the hierarchy of beliefs
in Rokeach's central-peripheral dimension of beliefs for
the same reasons cited for their lack of difference in the
Chi square analysis above.

The Rolex, Supphose, and Dewars advertisements
utilize Type C or non-primitive, authority beliefs as defined
by Rokeach. A possible reason for these advertisements not
being ranked third by the subjects on the basis of attention
could be that the subjects did not view the authority in the
advertisements as authorities for them as persons. This
rejection of the authority could also explain why the Type
D advertisements, Waltham, Sears, and Seagrams, were ranked
above the Type C advertisements.

The Volkswagen advertisement with Type A beliefs was
ranked below the MG advertisement with Type B beliefs. The
Volkswagen advertisement utilized a written violation of a
Type A belief in its headline while the five other Type A

advertisements tested utilized illustrations to violate

73

TABLE 43.--Advertisements that did not conform to hierarchy

 

 

 

 

 

of beliefs.
Type
* of Hypothesized Actual

Advertisements Belief Ranking Ranking
Watches

Rolex Type C

Waltham Type D
Panty hose

Supphose Type C

Sears Type D
Liquor

Dewars Type C

Seagrams Type D 4 3
Cigarettes

Virginia Slims Type A l 2

Winchester Type B 2 3

Tareyton Type C 3 3

Salem Type D 4 3
Automobile

Volkswagen Type A 1 2

MG Type B 2

 

*
See Appendix A for

column.

advertisements identified in this

74

their Type A beliefs. This seems to be a possible reason
for the nonconformance of the Volkswagen advertisement to
the beliefs hierarchy.

Two suggestions for doing additional kinds of
research arise. One is that, if possible, it should be
determined if the authority figures presented in the Type
C advertisements included in the experiment are viewed as
authorities by the subjects participating in the experiment.
Assuring that the authorities actually exist as authorities
for the subjects could be done by making the authorities in
the advertisements universal or by targeting particular
authorities with a series of advertisements to segmented
groups of subjects participating in the experiment.

The second suggestion for further research is to
determine if the advertisements containing or violating the
four types of beliefs would conform to the hierarchy of
beliefs in Rokeach's central-peripheral dimension of beliefs
if the beliefs were presented in written form. From the
ranking of the Volkswagen advertisement, with its written
violation of a Type A belief, the indications suggest that
the ranking might not conform to the hierarchy of beliefs.
A comparison of the effectiveness of the advertisements
between an illustrative presentation of the beliefs and a
written presentation of the beliefs could also be conducted.

The suggestions for selecting appeals for adver-
tisements for this analysis seem to be that authorities

used in advertisements should coincide with the authorities

75

of the readers of the advertisement. Another suggestion,
although further research appears to be necessary for it to
be conclusive, seems to be that an illustrative violation
of a Type A belief has more effect in attracting attention
than a written violation of a Type A belief.
Analysis of Reasons for First Rankings
of Type A Advertisements

The findings for the probing question were presented
in Chapter III. The question dealt with whether or not the
subjects' attention was attracted by the Type A violations
in the advertisements or for some other reason. It seems
that advertisements violating Type A beliefs were most
effective in attracting attention, with the exception of
the Volkswagen advertisement. Only one subject ranked the
Volkswagen advertisement first because of its violation of
a Type A belief.

It might be necessary in further research to include
a probing question for every type of belief to determine if
the ranking for all of the advertisements is because of the
type of belief that the advertisement contains or violates

or for some other reason.

Brand Preference

 

General Hypotheses

Null Hypothesis

 

Exposure to advertisements utilizing one of the
four types of beliefs has no effect on the brand
preference of the subjects.

76

Alternate Hypothesis

 

Exposure to advertisements utilizing one of the
four types of beliefs has an effect on the brand
preference of the subjects (as expressed by Opinion),
especially those advertisements with elements of
higher beliefs, particularly Type A beliefs.
On the basis of the evidence presented in Chapter
III, the null hypothesis is accepted that advertisements
utilizing one of the four types of beliefs do not have an
effect on brand preference.
Testing the Difference Between
Two Correlated Means
Two brands exhibited a more favorable preference by
the subjects in their ratings of the pretest and the posttest.
The difference in the rating for the Mojud Panty Hose brand
with its utilization of Type A beliefs supports the alternate
hypothesis that exposure to an advertisement with Type A
beliefs has an effect on brand preference. The difference
in the rating of the Sears Panty Hose brand could be explained
by the fact that the product was unknown and might possibly
have a poor image with the association of the store, Sears,
and the product, panty hose. The subjects might have developed
a more favorable image of the product upon learning about the
brand by exposure to the advertisement.
Other factors besides advertising often are contri—

butors to brand preference. Some of the factors are: how

the product is used, packaging, design of the product, cost,

77

where the product is sold, and brand name or image.51 One

advertisement by itself cannot be expected to be responsible
to alter brand preference significantly. Advertising is only
a part of an interrelated system of functions designed to
change brand preference. In further research, the beliefs
theory in relation to advertisements should be tested on the
basis of what the advertisement by itself is supposed to
accomplish in the marketing of a product.

The findings gave one indication that advertising
has the ability to convert a poor brand image to a more
favorable image. This ability was demonstrated when subjects
who were exposed to the Sears Panty Hose advertisement were
then more favorably disposed toward that brand. Advertising
also seems to have the ability to change the brand preference
of a relatively unknown brand. Subjects who were exposed to
the Mojud advertisement violating Type A beliefs were more

favorably disposed to that brand in the posttest.

 

51S. Watson Dunn, Advertising; Its Role in Modern
Marketing (New York: Holt Rinehart and Winston, Inc., 1969),
pp. 240-243.

 

 

APPENDICES

78

APPENDIX A

ADVERTISEMENTS TESTED

79

80

The poorgirl’s rich-watch.

 

Designer wrist-watches from the House of Dior. But the on 1y expensive thing about them is (heirlook.

The Dior Contemporary Collection from Bulova.

Models shown left to right: #53016 #53014 #53029 #52030. These and other styles from $100 at fine jewelry
and department stores. 1’E'Bulova Watch Co.. Inc.

81

.1

'fjuimc mm “(Ann’s A

'~ . '45.
O.
.,. I» 401‘, I
s‘ .,.p 5
~—‘\ ,1

\.

.‘ >-
' .

 

 
  
 
 
 
  
 
 
 
 
 
  
   
   

A dishonest watch can moire you gdilty in the eyes ol the law
lazy to the eyes of your boss And sloppy in the eyes of your hiends
To prevent this character assassination, get an AccutronO
‘ watch It doesn't run the run 01 the mill way.

‘ It has a tuning iorlr movement guaranteed to tell
the truth to wuthin a minute a month?’'
. And alter months and months 01 its unwavering
'\ hOnesty, it’ll restore people's laith in y0u.
~ Baton ACCII'I‘IOI

For men and women

Id! ‘25526: UK solid 91:1 ‘9‘" ‘74802 10k 901:! ltllec‘

h-lllflifluwdwfl wh'cim‘yaddomflmf

. non: $100595

'T-mduphg ‘Mll tn od-wcd to this We. i"

memory. 1‘ mud to km 60* tom
mined mbr am- yeov lrovn

dowdndnaONmV‘v/adiCo. "\-

82

 

Red .‘\(l.'|lr is internationally famous for fighting otl «tell seals itlte a miniature submantte hatch The crystal is.
lit'es lle answers alarms iront .‘tll over the world. diamond-cut lot a meticulous match with the case and

l’t‘olmhly his greatest challenge was this Sahara blowout uniquely engineered so that it actually grips more tightly
that sent it geyser ot tlame thundering to .‘t height of Ttttt t‘ect. ttlttlt't‘ pressure ttleep underwater. for instance)

growing SStlJitttLttttt) cubic feet of gas ever y 24 hours. lnstde is at Rolex tttmement that truly deserves all this
'l‘hose who watched this spectacle dubbed it "'l‘he protection Crafted skilltully in time-honored tradition by

l)evtl's Cigarette l-ieltter."Jolm (ilenn was able to obset‘x e skilled Swiss. hands Fitted with a sell-winding rotor that

it from his space capsule Rolex tnyentetl. Submitted Voluntarily to an Official Swiss
It took fiyc lot :4 months ot punctiltous preparation Institute for L'ht‘ononteter 'l‘ests. where each individual

before ~\dair was ready to mine in. penlously close to the movement undergoes two weeks of exhaustive tests

meme tttiL‘t‘ntt. and dcillV place the specially made explostt e before earning the pi i7ed "chronometer'~ rating. The result

cnaree that — boottt.’ ——
blew out the tire much
as you might blow out
a match.

A man like tltis
demands a lot ot‘ him-
self. Hts possessions
mu~.t match his life-
style. That's why
you'll find a Role\
on his wrist.

It looks rugged —
and is. The case is
carved from a solid
ingot at 18 kt. gold ot
stainless steel. The
'l‘nplock winding
crown is firmly
screwed down into
the case so that it

 

l]"l"1'rlllrl~'t u- {N 1‘! trlvlt L"’l'

.4' l .L-f't “Wt tl'u wrkt 7‘21! .4, i.

is a triumph of the
\t'tttchmaker's art:
a timepiece you'll
own with pride: a
timepiece you'll
never have to
pamper.

Ask Red Adair.

ll Illr' lit!" 'IH r' mlur
(Wm'httt‘r'. It’nlcx ll‘rllr'll
(15 .l Im.. 58" Ft]!!!
sleutm. .\.l"' lin‘k. .\'. l'.
lit/I36. Dept. .44.

Owning one is
almost as satisfying

as making one.

Little Smsmasterpteces

g.17.,je‘iiiéls. 50d011§i3§5fif

. . lad jcg'fy‘tibtfltafi '
'Z’u? ’ :4 ,

in

83

l
‘A

n
.1

Sonté':.dramaticaUyt has; - ‘. 5 _“§5,;gs'é:l
Thoroughly female, SMss-watchescmosé’éfrom " .
expansion band‘and bracelet styleszhe “3118?”.03‘,"
ranges ironhtlho ultra. {entinine "to the spo - Lg “
i'Faccmdi-«Sctllf’lured. Each watch-you st .. " has -‘

.a 17 [must Swissrriitgement. Stunnidglittlciw'drlrs'" "

of art,~aren:tythe’ PAndyoo can pick :5 any 011.0 of
these for a tow y_litfle prieo.‘ Most (tithgm i’
finely firmirgm 25 to 59-dtdlars. .

t K r_'.’ ,

"it IVAN at at: ”W “ l " ""
ff no!” on 9' ."°, " t?$’g¥%“{f~o .. " ‘

l’.

3.77] '- CHlCAGQ ' NEWJYWK 'I'GENEVA ', ,_ .

..
' ‘ ' .
r A _ .‘

‘
s

 

:‘ -. i"

84

LWa_. . '
m‘immdtayw-lom o

~ '“A‘ I

{\t
t

'" r l .t
a O “
'"‘.t ,' .')
- ' 7‘” u
. .J
r

\ ,2,- r
' ,. u

t

l

A , Mojud' Parity Hose. .
For legs that come alive in the Spring.

This SpringMojud brings you more st/zhng pantyhose colors. In éxelusne super stretch ( armolon.
()paque. Agtlon 'and Sheer-to-thcsWatst. So, no matter where your fashionsgo. our colors will go with them.
It 5 Mount xColortul way to wttke up your legs thts Spring. MOJUD

\" ‘ , I

 

85

What makes a woman feel it’s worth ' .
$3.00 to wear Hanes Ultra Sheer pantyhose? ‘4
0 s' ,. -' o 0 4*. .-'..j "t-

I
I.
o

J

1. a
,,
i

r‘. "I -
1' o: f ‘3'“

.

Gentlemen prefer Hanes.

Available at litter deputttttertt and uppalel stoic-n ($3.00 suggested retail pure )

 

Newtfilp p-hose

86

    

Shoer-To-Tho-Waist Panty Hose

Now there’ 8 a new Supp- hose Payntv Hose

with ultra sheer, flat knit support.

It looks great with all the newest fashions "

because it’s sheer- to- the- waist.
And because it’s Supp- hose, it gives

you that 9 AM feeling at 5 PM.
For only $4. 95.

- “J t_.__ _

Another tine D'oduct 0' Ka-ywrrflo'h

 

  
  
  

87

.3

:‘e‘f;

wags“
'J’35H‘ on“; .

‘9. ,

n
4 U‘E'I. ‘

 

Panty hose co
u can imagine. tall and extra large.

 
 
   
  

   
 
  
 
 

 
 
 
  
 

     
   
  

 

t, your ankle, Something to fit all t .

[the thigh— because the world is not made up
",8 clings to your leg. of slender models.

L}- e round, and the first time The way Sears, Roebuck and Co. SC

‘ 'nd out what leg really is. it’s made up of nice women who want to feel
is a process so special attractive right down to their toes.
t. Panty hose, stockings and Thi—Top” hosiery,
a fit as proportion. only at Sears. In most larger stores, through
ize our stockings in petite, shapely, the catalog or by telephone, if you call
and statuesque. Catalog Shopping Service.

Sears

Cling-alon hosiery.
@ Sun Roebuck ms (.0 19m The fit to be tried.

 

 

 

88

 

KENTUCKY MIC!" meow WHISIY - M n00: ° EMU "MES DISTILLIIV 00., [WISVILLL KY. Qfloc I!”

89

 

S ’anumm 'qud N - USIHM HDLOOS 030N319 ' BOWO‘MOM 010 “A ll

 
  
  
   

““BP.‘*%=-.

“i‘,
M

g‘

H

r 1' s.

§

_._-_ -
—
..

3:7300055
su.gcor,,

,‘;brcn

Aul‘

~

' .~..
%. ___

90

DEWAR’S PROFILES

(Pronounced Do-ers “White Label")

 

l’lxiNDrO SI? YICH wins” . 31.19 mm - , SIIHINLIY IMPORIS (0 N Lu 1

EVE QUELER

HOME: New York, N.Y.

AGE: 37

PROFESSION: Conductor

HOBBIES: Playing the piano; attending concerts.

MOST MEMORABLE BOOK: Dante's
“Divine Comedy"

LAST ACCOMPLISHMENT: Recently .
conducted the Opera Orchestra of New York
at Carnegie Hall in Bizet'a “Pearl Fishers,”

receiving raves from New York music critics.

QUOTE: “I think all young American
conductors should have the opportunity to come
up the way many European conductors have—
through opera. It gives you fantastic back-
ground and absolute mastery of your materials."

ybite Iab¢l

a. e'.
(HulkM‘ufiav‘

 

AuthentiCJhere are more than e thousand ways

PROFILE: Vlgorous' Chlc' Exalglng' conduCts to blend whisklesln Scotland.buttewareauthentlcenough
With a sure command of her music and her for Dewar's "White Label." The quality standards we set
musicians. down In 1846 have never varied. Into each drop go only

, “ . ,, the finest whlskles from the Highlands. the Lowlands. the
SCOTCH. Dewar 3 White Label Hebrides. Dew ar, 8 new varies.

91

   

America's whiskey
and how to enioy it.

1 _ , It isn’t easy, sneaking one past

      

 

- .V 3- mm. - ”g“ 5 a three-inch goalie.
. ,:* "f: , . 4;, ' So when you do, you ought to
" .- . . celebrate. And a Seven Sour is what

, you ought to celebrate with.
All it takes is lemon juice, sugar,
and of course, Seagram’s 7 Crown.
Give it a shake, dress it with fruit,
”Cg: and you’ve got it made.
. But no matter how you like your
' whiskey—in a sour, with soda, or on
_. the rocks—you’ll like it better if
. ‘ ‘ it’s Seagram's 7 Crown.
- It’s light, smooth, and uniquely
versatile.
That's why, for 26
championship seasons, it’s
the whiskey America has
liked best.

Seagram's7Crown.
It's Amedca’slaeorlte.

......

DlSllllflS CO. I Y C. AKIICJI WHISKEY—A [m S "W.

THE SEVEN SOUR: /

.- ’l' 1)
To the juice of one lemon. add one teaspoon fine sugar, \} I)!
and 1% ounces Seagram’s 7 Crown. Shake with ice. Garnish ' ‘ /
with orange slice and cherry. ‘-7 -—-'

 

THU! / MARCH 1974 I ss

l" l:-

 

 

Wsrning: The Surgeon General Hes Determined
That Cigarette Smoking Is Dangerous toYour Health.

 

 

~.‘ 4
A
.3.
-

Q

.b'ev
.

'.t

i}! ‘f
«is -
‘3 rs

31k

'0

.‘h’

Afllii l

a

Q
a

t a
3 M...

A

 

92

 

 

17mg1'tar1' 1.2 mg nicotine sv. "m. FTC mum

93

3"

.1 ,.
. ‘5:

Winchester

. separates the men...from the boys. l:

    

You're a big boy now. And when your taste grows up, your
smoke should, too. Winchester is a very mature smoke. Ask any
woman. She'll tell you that Winchester is slim and sexy, with a
filtered smoothness. Mild and light. She’ll tell you that Winchester
isn't heavy-handed on aroma. One gentle whiff whispers in her
car: “It's not a cigarette. Nat iust another little cigar. It's a whale
'nother smoke? Very adult. And she'll tell you that when you're
man enough for Winchester, you're man enough for her.

Winchester: It's a whole 'nothor smoke.

. .-.~

20 LITTLE CIGARS.

WINCH TRE _-'

 

l.'- A. it. twill-d;-

94

Y

 

“Sm Warning. The Surgeon General Has Determined

 

'.—‘~‘-‘-. ' W3 P» . .
s' 'r . r/ h . bertet‘as" TharCrgaretteSmokrngls DangeroustoYourHealth.
h

 

Tmemo ’m w ”ing Size: _2I mg. "tar". 1.4 mg. nicotine;
masle “0 Ola IOU mm. 21 mg. "rar". 1.5 mg. nrcmrne, av. per Cigarette, FTC Report Sept. '73

1.

115:;- l -

xx.

*1,
.5

‘.-' h .
45: 3"in

 

95

*0!

° Naturally grown menthol.
° Rich natural ta'bacca taste.
° No harsh, hat taste.

'u
'19."! “

King or Super King

Warning The Surgeon General Has Determined

That Cigarette Smoking ls Dangerous to Your Health ““5- 1933 1,"l8'".l-3m9-m€01m9.
SUPEF} , .19 mg."lar", IA mgthmme. av. per Clqamllla l H; llT'IIlrll SI Pl.‘73.
l A. ‘

 

96

”us-null OI Ant-ca. Inc.

 

Today, it’s bigger than a Cadillac.

ltyou bought a Cadillac Sedan DeVille in '72 and traded it in today,
you'd get back 53% ' of its original sticker price.
llyou bought a basic Beetle in '72 and traded it in today,
you'd get back 78% ‘ of its original sticker price.
Of course, we can't guarantee you'd do as well next year.
But then again, you might even do better.

620

U
Dazed on NADA Oflrra' Used Cor Gudi- TEoI'ern ld hat-l ——.lur‘c 7‘ ICu'H-nt "ado -" who u '9’? II (In price-l.

97

MG. First on the scene.
And still one jump ahead.

O

W"

g, A.

Amt-i1 . ‘

Betore MG. there wasn‘t much ot a
sports car scene in America.

But trom the moment the tirst
MG-TC touched these shores in 1947
the picture changed. So much so that
MG has almost become an American
synonym for sports car.

Today‘s M68 is the SCCA National
Champion in E Production tor the third
year in a row. With that kind ot track
record. you can see why M68 is what
great sports car motoring is all about.

M68 is trim. taut and agile.

The tour-speed gearbox puts your
reflexes in touch with the 1798 c.c.
engine. That's a 4-cylinder powerplant

that's as economical as it is lively.

The rack and pinion steering. the
race-seasoned suspenslon and the
tront disc brakes combine to give you
precise. sure-looted handling and stop-
ping the kind at response that turns
driving pressure into driving pleasure.

You‘ll experience it most where
the roads still touch the edges 01 na-
ture and the air is scented with the
sweet smells ot earth.

But make no mistake. the M68 is
just as much at home on a six-lane ex-
pressway as it is on two-lane blacktop.

The M68 is complete with tull
sports car instrumentation. including

tachometer. trip odometer. and gauges
lor tuel. oil. water and battery. There
are also reclining bucket seats.
wrapped steering wheel. carpeting. Oll
cooler. mag-style wheels and radial-
ply tires.

50 make the scene at your MG
dealer and see why MG is still one jump
ahead. For his name and tor intor-
mation about overseas delivery, call
(800) 447-4700. In Illinois. call as... run;
(800) 322-4400. Calls are toll
tree.

MG. The sports car America i
loved tirst.
BRITISH LEYLAND MOTORS INC LEONIA. N J 07605

. .
.Ll-VLAND

 

98

IN EUROPE.
WHEREAGALLON OFGAS
GOES Eott BETWEEN
mun $1.80.
MORE PEOPLE truv FIATS
THAN Attv otttEtt oAtt.

   

 

      

Prices of regular gasoline in Europe. Denmark. $1.13. Norway. $1.22. West Germany. $1.16, Sweden. $1.15. Greece. $1.80. (Costs and currency
exchange rate as of Li’2/74! l Italy, $l.l4. Bel ium. $1.01. Netherlands. $1.05. Portugal. $1.05. Switzerland. 5.95. Spain. $1.17. England. $.77.
(Costs and currency exchange rate as of 1/25/ 4.”)

'Source: European Embassies. Paris. France. "Source: Institute of Petroleum. London. England.

EHEH

The biggest selling car in Europe.

Overseas delivery arranged through your dealer.

ESOUIRE: MAY 11

 

99

What you're looking at 8'0 U3 results of Mic Humet Spottabuut A3 15.5 LinuvltV-Mt‘rutr) Montego A3 99

Cl‘t‘i rulet Vega Panel Chevrolet Malibu Clam A3 9.9

a gas mileage test pertormed on 1974 cars by the Wm... M. ,5. pm... mm... .3 9..
us. W m Am Tn} tutu Mark ll Sedan A3 I51 Ford Tunno Mi 91‘

 

l-n mi Mark II Wagon A3 152 Burch Centur) Wagon A3 9 7
Th test W m m trip m 1.04.. Mark II Sedan W is 2 Jaguar E Type V-l2 A3 9 7
Chi-\rult't Nova Hatchback A3 152 Bu-di Estate Wilson A3 9.0
m ”uh”- AML‘ Home! Sedan Alt ”.7 Cheinilct Caprice anm A3 9.6
\irlvn l64 A3 ”.5 l.lnCtlln-l\li'lk\lr) Cnumr A3 9 5
it you're h the maket tor a new car. we suggest ”mm M 230 M m M “mm M 9..

you Me u” of m m as follows, Mercedes Benz 230 At It I Uldxttllbtlr Cutlass

Ford Trinm A3 Hll Supreme All 9.5
1' m m a” I” m M W m cat min BM“ lLimrigi Ml Ht" l’i'ntim l,l'.\iillt.\ MI 9.1
Che—tint Sedan A3 I} H Rulls Rl')k1'50l\'t‘f Shaldtm A3 9 .l
' \irlw I64 MS [31 l‘utitmc‘Catalina Ail it 2
2. 0001983 its mm to the car at the Mir; (ireinlin Mt l.'t 2 twin- t..-M.m~ A3 92
AUC Javelin MJ l3 '3 Hillt‘h (ililnd Stunt A3 9|
top of the list. BMW mam ;; i2: (mun A3 9 I
3. M m. Pl) rtmutli \’:ili;rrit [)ll‘drl M3 IL? .5 ()ldxltihllt‘ Ut'ltn 88 Rnytil A3 90
AMC Malndnr A3 [2 1 PtlfiililC Venturu (2T0 A3 8.9
Tris list is behg W by the makers of the Mic \t..i.-.d...\i'..w~ .\:i i2 .r Putin...- vi-nium (m) Mt as
. AMC Lin-Ira A3 I.’ l Clii slri “firm-n A3 8."
car at the top of the list. Partty as a public servroe. or... SM At ll .. ii, 3...”. a... M... .u tit)
Pit rr-Iuth Sillt‘lltil.‘ Ml ll.“ C.idill.it “(Villa :‘3 NH
r\.\lL Hurtit't Ml ll 7 “NHL R211.” A3 8.“
‘ V.“ II ‘ _ Ply tt-thli Silt‘llttr: Ail ll ti l'unti.“ (inind Am A3 8 l"
Honda (.mc Md 29] Chevrolet Vega ltatchbadr A3 I91) M Hem. H...” M w | In Chm mi... up"... CHM A3 3.“
deswafltn “2 Willi?" M4 27‘9 Saab99 LE , M‘ W l Firrtl Ti-ttnii Wagon A l ll ’l ("dun-hilt Vie-tit Cmrwr Ail 8.7
Torch Corolla I20!J brunt M4 27.1 Toyou Martl ll Warren M l9 1 thtuvlt‘l hlrrwry cumin... l"lertu--rd Aft it 7
mus Europa Sufi“ M5 25-2 Al 3 Rommzom GTV MS l9l hlivnlt u \Vagun A] l l. i l’mtttst Tram Am Mi HS
08‘5"" 32") M4 24 9 Renault 1?- N‘dan A3 19 | Ctlttt't‘ SM Mi. i 12 rim“... Lthins sitar. A3 as
Tng Corolla 1200 g‘déll‘l M4 24.8 [10(th QHT M4 '9‘ Ahjntr Cry“: [‘1 I l U Elt'flllbtll ll A3 8,5
V. kswagen 412 Wagon A3 24 6 TVR 25(0 34 Mt l9 l) Chm wk! llllfh‘lld ”d”. 5",” Wagnn A3 35
Chevrolet Vega Hatrhtmck hill 24 6 Volkswagen Ktimhi-lll’ Spiirh Sedan ,\:5 l I ll pm...“ (inind Satan A3 8.1
Lotus Europa MI 24 5 (Mimbust A3 '9 0 l.lttt‘tvlll Mercury 5.1mm. NH r i ll ()ldsrnnbilt' Tun-0.160 A3 8 :t
deswagen Dasha Sedan M4 24.3 Manila 808 Coupe A'l Id 9 ..\\1g J... (4.“ Mt lint Buick Elettrti 225 A3 3.3
“ii-“‘38?“ Dnsher anm A3 23-7 Renault '2 St'di‘“ Mi '5 5 ~\\l\.‘ Amhawidur A3 lint l'uritmc Catalina Safari A3 8.3
Volkswagen Dasher Sit-dart A't 23.3 MOB M-t in 7 \i.i:rl.i RX 3 W'tl'm in; “”4 km“, liitrrwptur A3 8.2
Triumph Spitfire \‘H 23.] Toyota Cort-m SR Suntan Ml 18 1 Fund A] In? l'iumnc (inind Ville A3 8 I
Fm! Pimo W 22.8 Tarot-i Cum-1 SR Satan M It” um. Rx rump.- A3 I“ 7 Mercury Wagon A3 8.!
Dodge (k-It Wagon m 22 8 Volvo l-t5 Mt In t stud. Rx 2 0...... Mt rm; Liner-tn Continental A3 7.9
Dodge Colt Coupe A3 22.7 ()pel Manta Mt In 1.! “mm... Hm; .131) Aft lll6 Maseru“ [20 MS 71‘
Subaru Waxun \H 22 7 ”tn-1 1900 A1! In 2 midi RX 1 thgin Mr K) 1 Pontiac 8mm Illt‘ A3 73
Toyota Corr-ll.) lfitll :x-dnn Mt 22 6 Fur l24 Sum! Sedan M i In 0 Find l’.iiiu~i.i \l5 It) I Chm mm Chmflk Luxunn Mi 7_
Volkswagen C'rrrvrrtihic SA 22 6 Renault l5 Tl. Uruur Mi ”9 Hiiii'k Cz‘iitiir) Tull A3 ID i ()ldsnuhile 98 RWPDC)‘ A3 7.6
“MW 3002 M 22.6 Uriel Manta Luw M I7 9 Hunk LcS‘rlirr A‘s to l Uldsrthbtlr Della as Wagnn A3 7.6
[lodge Colt Curio:- \M 2‘2 3 Full l24 Slint" TC A] '7 5’ (.xtillm‘ Eldiimdn M In t la‘lllhbfflhllll Jainima M5 7.3
MG Mldg“ M3 33 l Full 123 Wagrin Mi ‘7 "‘ MAMA RX 10.1in A} I“ it Igirrtlrrghitti Espndn NS 7.2
Datsun 8210 M 21! 3 Fiat IL‘I Wagon M 17 Wu.” F. T. tr V I2 Mi iii .1 Fm... .1035 (.Th t MS 6.;
Renault 17 (iiii'diiu M“) 3 _' 3 Porsche 9| $4 M?) 17 {i (”mu-lint, Cum“ 4;: Ill .t
Renault l2 Wag-m A"! 1?. 2 Renault I? ll. know M-t IT 3 Chi-x mm ltr'mh UM..."
Audi Fax M1 33 0 \i'ulm l-I‘Z MT} IT .3 (”my At [til i. ., ... ...... 5...... t n ”lard. .a N i. Duh-d
Dodge Cult Wagon Aft 319 Flat IZt‘i Sedan M8 l.’ 3 pom.“- Tm”, \m .~\ g m | '- U -- {WNW Ali-MW rd 5“ a ! h-r \mvl
Hunda CH H. 5‘ 2‘ 3 CN'\ “d!" ‘m' H t' h" u k 5“ l '1 ‘ Ft’fl’Jll l’lth‘ :ll‘ (iT \lJ I” i‘ . ‘l‘;::l"l.."i:':lt:;:"‘ll~ at! v uh ”I” w~l
Sham)? Mt 21 7 Ford Mustang Mt 17.3 ch“ “.1“ 1......“ l’~.t.«.ri- l w...“ m. mud n run imam w
Viillu‘wttgr‘it Kn.- imir 0]!er M l 3 l 7 l’ofscht' 9H 5 MS ”.2 “3‘8”" Ali u, it fiffi‘ffx'vu “'::1:::~“ 'J
Subaru (””9” MI 2! 7 F‘Wd l'rntn AB 17'! PimllJC Vi'trttim A3 99 (m n Mada: aim-urn: rm and I‘d t9.l"~.‘t
Tnyi‘rttt CIWUHJ ltll“) W up.- A l L l l l’c-ugurt 50$ Rut l." ' "
Volkswagen IHl‘ l‘lii ru' M '3! 0 Volvo l-H
Volkswagen Super Bfic'h‘ Mi '3” ‘l Ford Mustnr-g
Toyntn t'oriilln l-vilfifilm A? ENS Lincoln Mi-rcrir i CJUH
Datsun TlO M f‘ll r' l’ma‘he ‘Jll h
llntsuri 610 .‘.’l f’tl Pi Triumph 'l K it
Pint X11“) M I 3‘” 4 Purge-4:101 anti
B'v'iW 200.“ . M l 20 l Plymnuth \'.i"'.1” l",.t‘.-r
Ftill lL’iSfK't'Ji TC ‘vll 2")2 Ford ‘Iimcmk
Fi-rd Mama"; it 1301 Ford l'iiitii“ uzu'
Dnt «in 710 M8 200 M08 (1T
M :iztln 808 0st Mi 2” 0 Datsun L’til‘r'
Chevrolet Veg t P.1‘.-.‘l l’urk‘lw fill 'l~
Expu-w M ”i so it A udi ton
Chevrii'iet Veg) K1: fuck AT 20 lb anti 99 Li:
Lim-rln-Mercui', Ctr-ac". MI 199 at l2! Sport C i. '
()pelMafitaR'il-Iyr MI I98 Dad eDnrt
Linailn-Men‘ur,’ Capo Mt [98 AME‘Urt-mlir~
Datsun 6l0 A'l ISLE Datsun 2602
Alfa Romeo 2000 Berlm.) MB [9.7 Chevrolet Nut.) H itchlnck
Ford Pinto Wagon M4 l9 7 AMC Crrmlrn
Volkswagen Kn ‘ ‘7' 23 Ford Mavenk
(Micrubusl Mt l‘Jb Lincnln-Mera‘sr) Comet

\“ t9" Ame'vcm “darn M)‘ )' C.) I \c

 

100

at."
Q\ '

«o

America's Favorite Cigarette Break.

 

Benson & Hedges
‘ 100’ . ‘

Warning: The Surgeon General Has Determined
That Cigarette Smoking Is Dangerous to Your Health.

 

101

W uld h alk
@ flit/{Iggy agnile frfr‘lvne?

Warning: The Surgeon General Has Determined
That Cigarette Smoking ls Dangerous toYour Health.

1 ‘n R J Reynolds Tn'caccc Comram thatonSalem N C

25 mg "rail 6 mg. nicotine 3V per Cigarette. FlC Reuot FEB 7’3

 

 

102

O 197‘W

Have one of mine.

5

  
   
 

Get hold of
honest taste.

Haven
OldGold.

W \

Warning: The Surgeon General Has Determined Valuable Gift Stars 19mg'te.“l.2m .
That Cigarette Smoking ls Dangerous toYour Health. Coupons. 100. at prcigemtoITGR-thelL'n

 
 
 

103

Introducing
Raleigh Extra Milds

. lowered tar
mild natural fl

H trig. tar
ll l mg. rtrr r )llllt‘-

New Raleigh Extra Milds joins Raleigh
Filter Kings and longs in uttering tree
B&W (nupons. These coupons add up
last tor valuable gifts

like this sterling silver

Duchin table lighter.

For vour tree

Gilt Catalog showing

oter mun gitts,

write: Box l2,

louifiille, Ky. 40201.

Warning; The Surgeon General Has Determined ' ": a ‘ ”4"“: ‘lw '1‘!) ”“l."‘"?f’}‘"fi tr. arraareutt‘v. llL ,
That Cigarette Smoking ls Dangerous toYour Health. ’ ‘" l'rl’m' "”- Wl ' ll”

i_.'l rt»; rrrrrltrr're. 5w ;:o3r ll ; :ll'll" t-l'C lie; rill J. ;' .

 

 

10 / TRUE / MARCH 197‘

 

APPENDIX B

QUESTIONNAIRES

104

 

Page

Code (11213)
Con. or EX. (4)
Name Student No.

 

 

 

Listed below are eight brands in eleven product categories.
For each brand, place an X in the one box which best indicates
how much you like or dislike that brand. There are no right
or wrong answers. Only your Opinion counts.

(1) (2) (3) (4) (5) (6) (7)

DISLIIG} DISLIKE DISLIKE NEITHER LIKE LIKE LIKE
COM- SCME- A LIKE NOR A SCME- CG“!-
AUI’QVDBJLES PIEI'EIX WHAT LITILE DISLIKE LITI‘LE WHAT PIEI'ELY

 

 

 

 

 

 

 

 

\kflksmmen ( 5)
Datsun . ( 6)
Audi . ( 7)
Toyota ( 8)
Ikmda ( 9)
Fiat (10)
MG (11)
SatmuNJ ‘ (12)

 

 

 

 

 

 

 

 

 

DISLIKE DISLIKE DISLIKE NEITHER LIKE LIKE LIKE
COM- SClVlE- A LIKE NOR A SCME- COVI-
SOFI‘ DRINKS PIEI'ELY WHAT LI'ITLE DISLIKE LITTLE WHAT PLETELY

 

 

 

 

 

 

 

 

Coco Cola (13)
Seven Up (14)
Pepsi Cola (15)
Doctor Pepper (16)
Faygo (17)
Tab (18)
Canada Dry (l9)
Mr.Eflbb (20)

 

 

 

 

 

 

 

 

 

105

 

BEER

106

Page

(1) (2) (3) (4) (5) (6) (7)

DISLIKE DISLIKE DISLIKE NEITHER. LIKE ILIKE .LIKE
(IIP- SIDE}- A» LIKE NOR. .A SOME- (33W-
PLETELY ‘WHAT .LITTLE DISLIKE LITTLE WHAT PLETELY

 

Budweiser

 

Pabst

 

Strohs

 

Black Label

 

Blatz

 

Busch

 

Nficheldb

 

Nfillers

 

 

 

 

 

 

 

 

 

BAR SOAP

DISLIKE DISLIKE DISLIKE NEITHER LIKE LIKE LIKE
COMP SOME- .A LIKE NOR. .A SOME- COM?
PLETELY ‘WHAT LITTLE DISLIKE .LITTLE WHAT PLETELY

 

Ivory

 

Safeguard

 

Lifebuoy

 

Camay

 

DOVE

 

Dial

 

Irish Spring

 

Phase III

 

 

 

 

 

 

 

 

 

(21)
(22)
(23)
(24)
(25)
(26)
(27)
(28)

(29)
(30)
(31)
(32)
(33)
(34)
(35)
(36)

 

PANTY HOSE

107

Page

(1) (2) (3) (4) (5) (6) (7)

DISLIKE DISLIKE DISLIKE NEITHER LIKE LIKE LIKE
C(14- SQ/IE- A LIKE DDR A SOME- C114-
PLEI‘ELY MM LITTLE DISLIKE LITTLE WHAT PLEI'ELY

 

Hanes

 

No NOIISGIISG

 

Sears.
Panti-hose

 

‘Ugly'rnxflCIing

 

Leggs

 

Belle-
Sharnexn:

 

.Mojud
Pantirfmxxa

 

Supphose

 

 

 

 

 

 

 

 

 

LIQJOR

DISLIKE DISLIKE DISLIKE NEITHER LIKE LIKE LIKE
C(IVI- SOME- A LIKE NOR A SQIE- CCM-
PIEI'ELY WHAT LITTLE DISLIKE LI'I'I‘LE WHAT PLETELY

 

Old Forester

 

Old Grand Dad

 

Dewars

 

Early’Tfinrfis

 

(flrbmas Regan.

 

JOhnnie
‘walker

 

Seagrams

 

Cutty Sark

 

 

 

 

 

 

 

 

 

(37)
(38)

(39)
(40)

'(41)

(42)

(43)
(44)

(45)
(46)
(47)
(48)
(49)

(50)
(51)
(52)

ND'IORCYCLES

108

Page

DISLIE DISLIE DISLIE NEITHER LIKE LIKE LIE

C(14- SOME- A LIE NOR A SOVIE- CCIVI-
PLEI'ELY WHAT LITIIE DISLIKE LITTLE WHAT PIETELY

 

HOnda

 

Suzuki

 

YamEflua

 

Harley'
Davidson

 

TrirmmflI

 

Rickneux

 

Ctquxrral

 

Kawasaki

 

 

 

 

 

 

 

 

 

CIGARETTES

DISLIKE DISLIE DISLIKE NEITHER LIE LIKE LIE
CCM- SOVIE- A LIE NOR A SCME- CCM-
PIEI'ELY WHAT LITTLE DISLIE LITTLE WHAT PLEI'ELY

 

Benson.&
Hedges 100's

 

Old Gold

 

Raleigh

 

'Virginia
Slims

 

Camel

 

‘Winchester

 

Tareyton

 

Salenl

 

 

 

 

 

 

 

 

 

(53)
(54)

(55)

(56)
(57)
(58)
(59)

(60)

(61)
(62)
(63)

(64)
(65)
(66)
(67)

(68)

 

'109

Page

DISLIKE DISLIKE DISLIKE NEITHER LIKE LIKE LIKE
CIPP- SOME- A. LIKE NOR .A SOME- COMP
WATCHES PLETELY' WHAT LITTLE DISLIKE LITTLE WHAT PLETELY

 

Tfinrnc

 

Bulova

 

Rolexn

 

Ukuxiunn

 

Seiko

 

Butcherer

 

Elgin

 

(Imama

 

 

 

 

 

 

 

 

 

Thank you for completing the questionnaire.

(69)
(70)
(71)
(72)
(73)
(74)
(75)

(76)

 

-110

Page

Code
Name Student No.

 

 

 

I am going to show you a series of four brand advertisements
of the same kind of product. I want you to tell me that if
you were looking through a magazine, which of the four
advertisements do you think would most likely catch your

attention first? Second? Third? And Last? This should
take you only a few seconds to decide. Are there any

 

 

 

 

questions?
Watches

The Poor Girl's Rich Watch (4)
______Are you paying for . . . (5)
_____ Rolex (6)

Waltham (7)
Why did you choose the ads in the order in which you
indicated? (8)
Panty Hose
_____ Mojud Panty Hose (9)
._____ Hanes (10)
_____ Supphose (11)

Sears (12)
Why did you choose the ads in the order in which you
indicated? (13)
Liguor

Early Times (14)
._____ Chivas Regal (15)
_____ Seagram's 7 (16)

Dewars (17)
Why did you choose the ads in the order in which you
indicated? (18)

 

.111

Page _

Cigarettes

 

Virginia Slims
Winchester
Tareyton

Salem

 

Why did you choose the ads in the order in which you
indicated?

Automobiles

 

Volkswagen

MG

Fiat

Honda

 

Why did you choose the ads in the order in which you
indicated?

Cigarettes

 

Benson & Hedges
Camel
Old Gold

Raleigh

 

Why did you choose the ads in the order in which you
indicated?

(19)
(20)
(21)

(22)

(23)

(24)
(25)
(26)

(27)

(28)

(29)

(30)

(31)

(32)

(33)

 

BIBLIOGRAPHY

112

 

BIBLIOGRAPHY

Books

 

Bartley, Howard. Principles of Perception. New York:
Harper & Brothers, 1958. .

Boyd, Harper W., and Westfall, Ralph. Marketing Research:
Text and Cases. 3rd ed. Homewood, 111.: Richard
D. Irwin, Inc., 1972.

 

Campbell, Donald T., and Stanley, Julian C. Experimental
and Quasi-Experimental Designs for Research. Chicago:
Rand McNally & Company, 1963.

 

Dunn, 5. Watson. Advertising: Its Role in Modern Marketing.
New York: Holt Rinehart & Winston, Inc., 1969.

Ittelson, William H., and Cantril, Hadley. Perception: A
Transactional Approach. Garden City, N.Y.: Doubleday
& Company, Inc., 1954.

 

Rokeach, Milton. Beliefs, Attitudes, and Values: A Theory
of Organization and Change. San Francisco: Jossey-
Bass, Inc., 1968.

 

Rokeach, Milton. The Nature of Human Values. New York:
The Free Press, 1973.

 

Rokeach, Milton. The Open and Closed Mind: Investigations
Into the Nature of Belief Systems. New York: Basic
Books, Inc., 1960.

 

 

Rokeach, Milton. The Three Christs of Ypsilanti: A
Psychological Study. New York: Vintage Books,
1964.

 

 

Periodicals

 

Deutsch, J. Anthony, and Deutsch, D. "Attention: Some
Theoretical Considerations." Attention. Edited
by Paul Bakan. Princeton, N.J.: D. Van Nostrand
Company, Inc., 1966.

 

113

 

114

Vernon, Magdalen. "Perception, Attention, and Conscious-
ness." Attention. Edited by Paul Bakan. Princeton,
N.J.: D. Van Nostrand Company, Inc., 1966.

 

Reports

U. S. Department of Health Education and Welfare, Office of
Education, Contract Number 3-20-003. Research and
Theory Related to Audio-Visual Transmission. Edited
by Robert M. W. Travers. Washington D.C.: Govern-
ment Printing Office, 1967.

 

 

Dictionaries

 

The Random House Dictionary_of the English Language. Rev.
Ed. 1968, s.v. "effectiveness."

 

Unpublished Material

 

Lessin, Edward J. "An Investigation of Primitive and
Authority Beliefs." Unpublished M.A. Thesis,
Michigan State University, 1965.

ll llWill)III‘HIHWIHMIIJ llIHIHJIHI'tlltilU ll

860147