Hill”Hill"lllllllllilllllllllllIll!Ulllllllillllllllllllll

1293 10454-8866

 

MSU

LIBRARIES

V

 

m

 

 

RETURNING MATERIALS:
Place in book drop to
remove this checkout from

your record.

FINES wi11

 

be charged if book is
returned after the date
stamped be1ow.

 

Ma’SL

fl‘t’ f/ 2

W-A4'M

{333%

I‘J

~‘ 3.. L}-
m,3:) L UL

F

:Hi ‘ ._ 'fif‘“ '

‘3. x n
. \ ‘\’ 1' (1. I .1"
I" ' 4 -' ." 1M!

5

 

‘53 (I:

J
8:12;»

M‘GS 31:33

1

,...4I
.' . $1.. 2'», l '1 ~
11:") ' (i Q’J‘Jv‘

 

 

_A333

3
v - :. ~..~2 k

S.

f /
f; a. L) N Ouu‘v‘ k. 7

) ff / /

DEGEPTIVE ADVERTISING:
AN EXPERIMENTAL EVALUATION
OF AN ATTITUDE CHANGE APPROACH
TO DETECTING DECEPTION
AND
MEASURING THE EFFECT OF CORRECTIVE ADVERTISING

BY

Karen Margaret Clark

A THESIS

Submitted to

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

MASTER OF ARTS

Department of Human Environment and Design

1981

ABSTRACT
DECEPTIVE ADVERTISING:
AN EXPERIMENTAL EVALUATION
OF AN ATTITUDE CHANGE APPROACH
TO DETECTING DECEPTION

AND
MEASURING THE EFFECT OF CORRECTIVE ADVERTISING

By
Karen Margaret Clark

The research objectives identified for this study were: (1) to
empirically test an attitude change approach to detecting deception in
advertising; and (2) to measure the effectiveness of a corrective
advertisement on eliminating the effects of previous deception.

Both the Fishbein attitude model and the extended behavioral
intention model were used in an experimental design to test for possible
deception contained in advertisements of three product categories: mouth-
wash, pantyhose and footwear. The experiment took the form of a two-
Stage non-equivalent control group design. Subjects consisted of 147
high school students in two Michigan cities. Pretest-posttest gain
scores for the control and experimental groups were analyzed using a t-test
of independence. No significant differences in gain scores were found
for either the attitude or behavioral intention model.

TWo beliefs from each product class were selected for further
investigation. Mean scores of the brand under investigation were
compared with an average of the mean brand belief scores of the other

three brands. This technique identified deception for three beliefs.

ACKNOWLEDGMENTS

The writer would like to express her appreciation to a few of the
many individuals who assisted her throughout the master's program. A
special thanks is offered to Dr. Barbara Stowe, Chairperson of Human
Environment and Design, who offered encouragement throughout the program.
Gratitude is expressed to my committee members: Dawn Pysarchik, Instruc-
tor of Human Environment and Design; Dr. Dennis Keefe, Assistant Pro-
fessor of Family Ecology, and Dr. Bruce Vanden Bergh, Assistant Professor
of Advertising. Their encouragement, comments, suggestions, and willing—
ness to work within a limited time frame were greatly appreciated. A
special acknowledgement is given to Dr. Brenda Witter, Assistant Professor
of Human Environment and Design, not only for her assistance in the study's
conception, her guidance in details of data collection and analysis, and
her valuable suggestions for the written manuscript, but especially for
her continued encouragement and emotional support through the master's
program.

Sincerest appreciation should also be given to the fellow graduate
students in Room 404 for their encouragement, especially to Mary Beth Cope
Robb and Maureen Sweeney Henry for their assistance in data collection.
Thank you to the high school students who participated in this study,
as well as to their teachers, who so graciously accepted our presence in

their classrooms.

ii

But, most of all, deepest appreciation and a big hug are offered to
Debbie for her emotional support, encouragement, and patience with a mother
who missed school activities and served 30-minute meals during much of

the past two years. Thank you, sweetheart.

iii

TABLE OF CONTENTS

CHAPTER PAGE
I. INTRODUCTION . . . . . . . . . . . . . . . . . l
Justification . . . . . . . 2
Statement of the Problem . 4
Research Objectives . . . . . . . . . . 4

II. REVIEW OF LITERATURE . . . . . . . . . . . . . . . . . 5
The Federal Trade Commission and Deceptive Advertising . 5
Authority to Regulate Deceptive Advertising . . 5

Development and Use of Corrective Advertising . . 6

Use of Corrective Advertising in Warner-Lambert Case 7

Use of Evidence to Determine Deception 8

Defining Deception. A Behavioral Research Approach 9

The Fishbein Attitude Model . 14

The Fishbein Behavioral Intention Model 15
Empirical Research Findings 17
Deceptive Advertising . 17

Corrective Advertising 20

III. METHODSANDPROCEDURES..................... 24
The Products . . . . . . . . . . . . . . . . . . . . . . . . . 24

The Evaluative Criteria . . . . . . . . . . . . . . . . . . . 25
The Questionnaire . . . . . . . . . . . . . . . . . . . . . . 25
The Sample . . . . . . . . . . . . . . . . . . . . . . . . . 29
The Research Design . . . . . . . . . . . . . . . . . . . . . 30
The Advertising Treatment . . . . . . . . . . . . . . . . 32
Hypotheses and Statistical Treatment . . . . . . . . . . . . . 33
IV. RESULTS AND DISCUSSION . . . . . . . . . . . . . . . . . . . . 35
Analysis of the Fishbein Models . . . . . . . . . . . . . . . 35
The Sample . . . . . . . . . . . . . . . . . . . . . . . . 36
Mouthwash Group . . . . . . . . . . . . . . . . . . . . . 37
Footwear Group . . . . . . . . . . . . . . . . . . . . . 41
Pantyhose Group . . . . . . . . . . . . . . . . . . . . 44
Summary of Demographic Information . . . . . . . . . . . 44
Testing the Hypotheses . . . . . . . . . . . . . . . . . . . 47
Stage I: Detecting Deception . . . . . . . . 47
Stage II. Measuring the Effectiveness of Corrective
Advertising . . . . . . . . . . . . . . . . . . . . 51
Normative Belief Technique . . . . . . . . . . . . . 52
The Affect Component of the Fishbein Model . . . . . . . . . . 62
V. SUMMARY, LIMITATIONS AND RECOMMENDATIONS . . . . . . . . . . . . 73
Summary . . . . . . . . . . . . . . . . . . . . . . 73
Limitations and Recommendations for Further Study . . . . . . 75

iv

CHAPTER PAGE

LIST OF REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
APPENDICES . . . . . . . . . . . . . . . . . . . . 82
Appendix A . . . . . . . 82
Appendix B . . . . . 83
Appendix C . . . . 95
Appendix D . . . . . 96
Appendix E . . . . 111

TABLE

1. Final Evaluative Criteria by Source . . . . . . . . . . . .

2. Fishbein Components Operationalized in Questionnaire

3. Directional Rating of Variables .

4. Mouthwash Group--Summary of Demographic Information .

5. Mouthwash Group--Comparison of Control and Experimental Groups

6. Footwear Group--Summary of Demographic Information

7. Footwear Group--Comparison of Control and Experimental Groups .

8. Pantyhose Group-—Summary of Demographic Information .

9. Pantyhose Group——Comparison of Control and Experimental Groups

10. State I: Attitude Change and Behavioral Intention Change Gain
Scores for Mouthwash Brands . . . . . . . . .

11. Stage I: Attitude Change and Behavioral Intention Change Gain
Scores for Footwear Brands . . . . . . . . . . . .

12. Stage I: Attitude Change and Behavioral Intention Change Gain
Scores for Pantyhose Brands . . . . . . . . . . . .

13. Stage II: Attitude Change and Behavioral Intention Change Gain
Scores for Mouthwash Brands .

14. Stage II: Attitude Change and Behavioral Intention Change Gain
Scores for Footwear Brands . . . . . . . . . . . . . . . .

15. Stage II: Attitude Change and Behavioral Intention Change Gain
Scores for Pantyhose Brands . . . . . . . . . . . .

16. Mean Brand Belief Scores for Selected Attributes

17. Mean Affect Scores for Selected Attributes

18. Multiple Regression Analysis of Fishbein Attitude Model . .

19. Multiple Regression Analysis of Fishbein Behavioral Intention Model .

LIST OF TABLES

vi

PAGE

26
28
29
38
40
42
43
45

46

48

49

50

53

54

55
56

65

.111

.113

LIST OF FIGURES

FIGURE
1. Two-Stage Non—Equivalent Control Group Design . . . . . . . . . . .
2. Normative Belief Technique--Belief that Mouthwash Prevents Colds
3. Normative Belief Technique--Belief that Mouthwash Kills Germs .
4. Normative Belief Technique--Belief that Footwear Gives Leg Exercise .
5. Normative Belief Technique--Belief that Footwear Has Health
Value for Feet . . . . . . . . . . . . . . . . . . . . . . .
6. Normative Belief Technique--Belief that Pantyhose Has Sex Appeal
7. Normative Belief Technique-—Belief that Pantyhose Has Texture

One Likes .

vii

0

PAGE

31

57

59

CHAPTER I
INTRODUCTION

Since man first discovered that goods could be bartered, the
seller has sought to get the highest price for his product. This has
sometimes resulted in the use of deception for the purpose of gaining
a higher price for a product or a larger share of the market than
would otherwise be obtained. For many centuries, the marketplace
operated under a philosophy of "caveat emptor." In modern times, this
philosophy has gradually changed to one more concerned with protection
0f the consumer.

In 1914, the Federal Trade Commission was created to regulate
Commerce. For many years, the Commission sought to protect the con—
Sumer from the use of outright lies in advertisanents. Since 1960,
however, the Commission has gradually become more concerned with
advertisements which seem to give false impressions, even though the
Claims may be literally true (Armstrong and Russ, 1975).

While the FTC has seemed to expand its concern with deception,
it has also devised methods to correct effects of previous deception
and to insure the truthfulness of claims made to the consumer. The
Commission has initiated an advertising substantiation program in
which substantiating data for product claims must be publicly disclosed.
Affirmative disclosure requires specific types of information to be
disclosed in all advertising of a certain product class. Counter
advertising has been required in certain cases, as in the provision for
anti-smoking commercials to counter cigarette advertisements. Correc-
tive advertising has been required in some cases in order to eliminate

1

the effect of past deception on future consumer decision-making.

The increased concern with deception and the increasing efforts
to remedy its effects are dependent on the supposition that deception
in advertising can be detected. The FTC has_generally relied on an
EPFUit.ixs.,§}2£§9§_9§’ -in.,-vhis=h thsiudgemenfipf Pic members is used 35°
detect deception in advertising. However, in recent years, advertisers
(Dillon, 1973), market researchers (Pollay, 1969; Haefner and Permut,
1974; Bernacchi, 1974; Wilkie and Gardner, 1974; Armstrong and Russ,
1975; Gardner, 1975), lawyers (Gellhorn, 1969; Mann and Gurol, 1978),
and even a former Commissioner of the Federal Trade Commission (Jones,
1971), have written of the need for aniobjective methodology for detec-

ting deception in advertising.
I. JUSTIFICATION

The existence of a consumer-based research procedure to determine
deception would be of benefit to regulators, advertisers, and to con-
sumers. It would aid regulators such as the Federal Trade Commission
in its selection of cases, prosecution, and in determination of remedies
(Armstrong and Russ, 1975; Mann and Gurol, 1978). It has been suggested
that consumer research could be incorporated into the existing FTC
advertising substantiation program, making research data available on
certain product categories, such as those in which public health and
safety are involved (Rotfeld and Reid, 1977). Consumer research would
aid in determining remedies such as cease and desist orders, affirmative
disclosure, counter advertising, and corrective advertising, since the
actual existence of deception as well as deceptive characteristics must

be determined before these methods can be used effectively (Armstrong and

Russ, 1975).

In addition, behavioral science research could aid the FTC in
determining whether or not what the advertiser considers to be puffery
is actually deceptive (Preston and Johnson, 1973). The law has gener-
ally seen puffery as a statement of mere opinion which no reasonable
consumer would rely upon in purchase decisions. However, there has been
little attempt to substantiate this position. Consumer research
evidence would help to demonstrate whether or not consumers do indeed
treat puffery as fact and if they are subject to being deceived by it.

Since, as behaviorally defined, deception is dependent on the
perception of the receiver, it will continue to exist in advertisements
regardless of the intent of the advertiser. For the FTC to use this
procedure, it would be necessary, therefore, to establish certain stan-
dards of allowable deception. These standards might allow differing
amounts of deception to occur in different product classes.

Advertising agencies could use a consumer-based research procedure
to pre-screen questionable advertisements and to provide evidence at the
time cases are being considered for litigation. For advertisers who
routinely do copy testing to measure the effectiveness of their adver-
tisements, such data could be gathered at a minimal cost. A bonus to
the advertiser would be the elimination of vague or misdirected
messages that interfer with effective advertising. The further elim-
ination of deception would also promote a more favorable impression of
advertising by the consumer.

If advertising can become more informative by the elimination of
deception, the consumer would benefit by gaining a better basis for

purchase decision-making. In addition, rather than relying on value

judgements and emotional suspicions, the consumer would have objective

evidence on which to base his complaints (Armstrong and Russ, 1975).
II. STATEMENT OF THE PROBLEM

There is a need for a consumer research methodology which can be
used as the basis for identifying deceptive advertising and for evalu-

ating the effect of corrective advertising.
III. REASERCH OBJECTIVES

This research was designed to:

(l) empirically test an attitude change approach to detecting deception
in advertising.

(2) measure the effectiveness of a corrective advertisement on elimi-
nating the effects of previous deception.

The authority and actions of the Federal Trade Commission in
regulating deceptive advertising and proposed methodologies for defin-
ing deception are presented in the following chapter. .Attitude theory
will be discussed, and two models of potential use for the FTC will
be introduced. The chapter will conclude with a discussion of the use

of attitude theory in assessing deceptive and corrective advertising.

CHAPTER II

REVIEW OF LITERATURE

I. THE FEDERAL TRADE COMMISSION AND DECEPTIVE ADVERTISING

Authority to Regulate Deceptive Advertising

 

In 1914, the Federal Trade Commission Act created the Federal
Trade Commission. This Commission was formed not to deal with
advertising regulation, but instead for the enforcement of antitrust
laws (Pelgro, 1978). Using its authority to prevent "unfair methods
of competition in commerce" (FTC act of 1914, ch. 311 § 5, 38 Stat.
719), the Commission immediately began to exercise jurisdiction over
advertising. This action continued until 1931, when the Supreme Court,
in FTC vs. Radam Co. (238 U. S. 643, 1931), severely restricted the
Commission's authority in this area. In this case, a manufacturer of
an "obesity cure" was ordered to stop advertising the product as "safe"
and "effective" when the Commission found that the preparation could
not be used safely except under medical direction and advice. The
Supreme Court provided a landmark ruling that forbade the FTC to pro—
hibit false and deceptive advertising unless it could show that there
was a harmful effect upon the competitors of the advertiser, regardless
of whether or not consumers were harmed by the advertising. Congress
responded to this ruling by passing the Wheeler-Lea Act of 1938,
which extended the Commission's authority to preventing "unfair or
deceptive acts or practices in commerce" (Wheeler-Lea Act of 1938,

3, 15, U. S. C. § 45, a, 1, 1970).

Development and Use of Corrective Advertising

 

/

In writing the Wheeler-Lea Act, Congress chose to use the general
words "deceptive" and "unfair" rather than to define deception by the
enumeration of proscribed acts. Thus, the Commission was given the
power to determine deception on a case-by-case basis. Although provid-
ing a workable framework for FTC control of deception, this decision
prevented businessmen from being certain that specific acts were unlaw-
ful until ruled so by the Commission. To prevent due process concerns,
Congress decided against the imposition of civil or criminal liabilities
for violation of the act, in favor of the issuance of cease and desist
orders.

Exclusive use of cease and desist orders to deter deception con-
tained two fundamental inadequacies: (1) provision of delay profits,
and (2) failure to eliminate persistent false impressions. Delay pro-
fits refers to profits gained from a deceptive advertising campaign
during the period of sometimes several years between the initial FTC
investigation and the issuance of a cease and desist order. By the
time the order is finalized, the deceptive advertising campaign may
have run its natural course and the guilty party has gained a profit
based on deception and paid only attorney fees. Secondly, the tradit-
ional remedies do nothing to eliminate the false impression that
persists after the deceptive advertisement has ceased. It is theorized
that consumers could continue making purchase decisions based on im-
pressions created by deceptive advertisements long after the deceptive
advertisement has been withdrawnv/

Corrective advertising is one method recently utilized by the

FTC to eliminate delay profits and persistent deception. Most simply

stated, corrective advertising is "advertising designed to inform con-
sumers that previous advertising by the company was deceptive" (Mann
and Gurol, 1978, p. 79). The exact wording and length of exposure of
this advertising is generally stipulated in a consent decree.

Corrective advertising was first formally proposed by SOUP
(Students Opposing Unfair Practices), a group of George Washington law
students who sought to intervene in a FTC action against the Campbell
Soup Company (77 FTC 664, 1970). This case concerned the soup company's
practice of putting marbles in soup shown in television commercials,
giving the soup a deceptively rich appearance. Although the Commission
declined to use corrective advertising in that case, it reserved the
right to use the remedy in future cases. SOUP again sought the correc-
tive advertising remedy in a second case, Firestone Tire and Rubber
Company, with the same results (81 FTC 398, 426, 1972). The FTC has
subsequently utilized corrective advertising in a number of cases as
part of the consent order. The most recent was signed by STP on
February 9, 1978, as part of a settlement of a civil penalty action
for violation of a 1976 FTC cease and desist order.

The federal court first acknowledged the FTC's authority to
order corrective advertising in the Listerine mouthwash case (Warner-
Lambert Co. vs. FTC, 562 F. 2d 749, DC Cir. 1977, cert denied, 98 S.
Ct. 1575, 1978). Since Listerine will be used in this study, a brief

description of the case follows.

Use of Corrective Advertising in Warner-Lambert Case

 

Listerine had been represented as a mouthwash product beneficial

in certain aspects for colds, cold symptoms, and sore throats ever

since its introduction to the market in 1879. On December 9, 1975,
the Commission ordered Warner-Lambert to cease and desist from
alleging that Listerine would cure or lessen the severity of colds or
sore thorats. It was also stipulated that the next ten million dollars
of Listerine advertising contain the disclosure that, "Contrary to prior
advertising, Listerine will not help prevent colds or sore throats or
lessen their severity" (Warner-Lambert Co. vs. FTC, 86 FTC 1398, 1513-
15, 1975).

When Warner—Lambert sought review of this order, the court found
the following evidence that supported the Commission's findings: (1)
ingredients in Listerine are not present in sufficient quantities to
have any therapeutic effect; (2) it is impossible for Listerine to
reach the critical areas of the body in sufficient concentration during
gargling; (3) even if significant quantities of the active ingredients
were to reach the critical sites, they would not interfere with the
activities of the virus because they could not penetrate the tissue cells;
(4) the ability of Listerine to kill germs by the millions on contact
is of no medical significance in the treatment of colds or sore throats,
because colds are caused by viruses (Warner-Lambert Co. vs. FTC, 652
F. 2d at 753—54). Although the court omitted the confessional preamble
of the corrective message ("Contrary to prior advertising,"), the court
upheld the requirement that the corrective advertisement continue until
a sum had been expended on Listerine advertising equal to the average

annual advertising budget for the period of April 1962 to March 1972.

<Use of Evidence to Determine Deception)

The Federal Trade Commission and the courts have used various

information techniques and types of evidence in their litigation pro-
ceedings. These include: (1) internal evidence, such as the intuitive
approach, dictionary definitions and expert testimony, and, occasionally,
(2) external evidence, such as consumer testimony and the results of
consumer surveys (Brandt and Preston, 1977).

Some of these sources of information have inherent problems.
The use of the intuitive approach, in which advertisements are judged
on the basis of the previous experience and expertise of FTC members
rather than on any outside objective evidence, creates a potential bias.
The use of dictionary definitions to determine the meaning of advertising
copy does not take into consideration the actual meaning of the words
to the FTC, the advertiser, or to the consumer. Expert testimony may
be biased towards the advertiser, the consumer, or the FTC, depending
on the source of the expertise.

Several authors have indicated that the second type of evidence,
external evidence, provides a more accurate indication of the existence
of deception (Gelhorn, 1969; La Rue, 1971; Gardner, 1975). This type
of evidence may prove useful, especially when it is based on the behav-
ioral effect on the consumer. Though a recent study (Brandt and Preston,
1977) indicates that the use of external evidence has increased greatly
in the past twenty years, it continues to "supplement rather than to
supplant the Commission's expertise" (Crown Central, Dkt. 8851, trr 20,

790, 1974).'
II. DEFINING DECEPTION: A BEHAVIORAL RESEARCH APPROACH

In 1971, Commissioner Mary Gardiner Jones, in an address before

the Second Annual Conference of the Association for Consumer Research,

lO

spoke to the need for behavioral research in the area of consumer
protection:

Consumer protection lawyers are just beginning to become
aware of the need for obtaining a more thorough understanding of
the behavior of the consumers they are attempting to protect...
There is no established framework for operationalizing this
interdisciplinary process and much work, therefore, must be
done first in familiarizing the lawyers with the potentalities
and capabilities, and second, in developing appropriate method-
ologies to come to grips most effectively with problems which at
this point he may not even recognize he has. (1971, p. 3)
“During the past decade, many marketing researchers have explored

the possibility of forming a consumer-based definition of deception and
developing a research methodology for detecting deception in advertising.
This work has borrowed heavily from the disciplines of communication
and behavioral psychologyy.

Communication theory has proposed that the communication system
in its entirety, rather than in isolated parts, must be considered in
understanding the message (Weaver, 1949). Thus, to study only the
message itself, without considering the receiver, leads to misunder-
standing the communication.

Based on this communication theory, Gardner offered the thesis 7
that deception takes place not solely by virtue of the advertisement,§
”but also includes the way the advertising message is perceived by the}
receive£j(Gardner, 1975; 1976). Thus, an advertisement could be liter-
ally true in and of itself, but deceptive because of the impact it has
on the receiver (Gardner, 1976). For example, suppose the results of
a research study indicated that consumers associate the color blue with
cleanliness. An enterprising manufacturer might color a portion of

the detergent crystals in each box blue and advertise a product contain-

ing "blue cleaning crystals." Even though no additional claims were

11

made for these crystals, the advertisement would be deceptive if consumers
perceived the blue crystals as containing extra cleaning agents.

A consideration of the perception process adds to an understand-
ing of this receiver-centered approach to studying deceptionq/ Gardner
(1976), reviewed this process as it applies to consumer perceptions of
advertisements. The received stimuli are categorized according to
learned relationships, although all stimuli received by the individual
are not categorized into perception of an advertisement for two reasons:
QED perceptions depend on the advertiser's skillful use of various
physical characteristics such as figure-ground, frequency of presenta-
tion, stimulus intensity, movement and change, and number of objects
in the advertisement; 62) personal factors such as emotions, wants,
and span of apprehension result in differing perceptions of the same
advertisement by different individuals and at different points in time
by the same individual. In addition to interference in the catego-
ization process, another phenomenon which affects perception is pre-
dictive verdicality, a process which permits one to "go beyond" the
information presented in the advertisement and predict what properties
to anticipate or expect.

Considering the legal realities of regulation and the receiver—
centered theory of deception, it is proposed that the most useful
definition of deception would be one which would integrate a receiver—
focused approach with the legal definition of deception. Gardner offered
such a definition:

If an advertisement (or advertising campaign) leaves the average

consumer within some reasonable market segment with an impress-

ion(s) and/or belief(s) different from what would normally be
expected if the average consumer within that market segment had
reasonable knowledge and that (impression(s) and/or belief(s) is

factually untrue or potentially misleading, then deception is
said to exist. (1976, p. 7)

12

This definition has certain important characteristics: (1) it
makes the economic assumption of adequate information for the average
consumer; (2) it focuses on the receiver, rather than the act of
deceiving; (3) it defines the deception within certain specific market
segments; (4) it retains the legal aspect of "factually untrue or
potentially misleading."

Consistent with this behavioral definition of deception, Gardner
(1976) categorized two main types of deception: (l) the "unconscionable

" and (2) the claim-fact discrepancy type. The "unconscionable lie"

lie
refers to an outright false statement. This type of deception is one
which is relatively easy to determine by simply finding whether or not
there is any reasonable evidence to support the claim.

The claim-fact discrepancy, the second main category of deception,
occurs in cases in which qualification must be added in order for the
receiver to properly categorize the information. This type of deception
may occur when the receiver adds his own learned information to incomplete
information presented in the advertisement, resulting in an incorrect
impression. It is also present in those conditions in which an advertise-
ment does not specify the circumstances under which the claim is correct.
Likewise, pre-emptive advertising, in which a false uniqueness claim is
made, would also be considered a type of claim-fact discrepancy. This
second type of deception cannot be detected by simple comparison with
scientific evidence. Another approach, a behavioral one, is needed to
detect the claim—fact discrepancyy/f

To determine this type of deception, Gardner (1976) suggested a
research approach based on the assumption that deception of this type

invloves consumer attitudes. He suggested the use of a multi-attribute

13

model based on work by Fishbein (1967) to serve as a basis for
studying deception. The Fishbein model will be discussed in the
following section.

Sales strategies usually attempt to change existing beliefs
about a product and brand, to introduce a new belief, or to make
an existing belief more salient; therefore Gardner (1976) hypothesized
that deception could occur if advertisements influence the probability
of a particular belief being associated with an attitude toward a brand,
or influence the evaluation of a particular belief associated with a
brand.

Gardner suggested three possible techniques to be used as screen—
ing devices in indicating deception: (l) the Normative Belief Technique,
(2) the Consumer Impression Technique, and (3) the Expectation Screening
Procedure. The Normative Belief Technique is based on the assumption
that product attributes can be classified into two categories: functional
and nonfunctional. Functional attributes are objective ones such as
wear, design, performance, and guarantees, while nonfunctional attributes
are nonobjective attributes such as style and appearance. An advertise—
ment could be judged deceptive for at least two reasons: (1) if con-
sumers rated the probability of a functional attribute being associated
with a product, when, in fact, the product did not contain that attribute,
or contained it only in insignificant amounts, or (2) if functional pro-
duct attributes received a higher evaluation than that of the product
class norm. Using the Consumer Impression Technique, consumers would
be shown advertisements, then asked to state what they felt the
advertisements were telling them. The consumers' impressions of the

advertisement content would be compared with the actual fact and claim

14

of the advertisement to determine if deception existed. The Expectation
Screening Procedure compares what people expect to see for a given

product class with what they perceive the advertisement to be saying. /
III. THE FISHBEIN ATTITUDE MODEL

Fishbein (1975, p. 4) defined attitude as "the amount of affect

" or in other words, the individual's feel-

for or against some object,
ings of favorableness or unfavorableness towards the object in question."
He views the formation of attitudes as part of the learning process in
which beliefs are associated with various objects. As an object is
associated with various positively or negatively perceived attributes,
characteristics, or qualities, the individual acquires a positive or
negative attitude toward that object. Thus, at any point in time, an
attitude toward an object is a function of beliefs about that object,
and the evaluative aspects of those beliefs (Fishbein, 1975).

An individual may hold many beliefs about an object. However,
the number of beliefs that contribute to an attitude toward an object
at any one time is limited. Research seems to indicate that humans
can deal with only about five to nine pieces of information at a time,
thus there are probably no more than five to nine beliefs that serve
as primary determinants of attitude at any point in time (Fishbein, 1975).

These are the beliefs that are most salient for that individual.

Fishbein's attitude model is represented as follows:

n

B A

A = O O
1 1

0 Z
1
Where:
A0 = attitude toward an object

Bi = the strength of belief 1 about 0

15

A1 = the evaluative aspect of Bi

n = the number of beliefs

Fishbein (1975) further theorized that an attitude toward an
action, like an attitude toward an object, is a function of beliefs

and evaluations of those beliefs. Thus,

Aact = E B.A
1

Where:

Aact = attitude toward an act
B1 = beliefs about the consequences of engaging in that act
A.i = the evaluation of those consequences.

Recently, a great deal of marketing research literature has been
devoted to operational and computational efforts to improve the predic-
tive power of the Fishbein model and other multi-attribute attitude
models (Green, 1964; Bass and Wilkie, 1973; Wilkie and Pessemier, 1973;
Holbrook and Hulbert, 1975; Mazis, Ahtola, and Klippel, 1975; Ryan and
Bonfield, 1975; Wright, 1975; and Holbrook, 1976, 1977). Reviewing
much of this literature, Holbrook concludes:

...one is impressed less by the degree to which any particular

formulation outperforms another than by the consistency with

which many different versions yield good predictive results
as long as some reasonable representation of beliefs is used

to predict affect or choice. (1978, p. 545)

IV. THE FISHBEIN BEHAVIORAL INTENTION MODEL

I

Although evidence seems to support the predictive quality of the
Fishbein attitude model, marketing research is really concerned with

the prediction not of attitudes, but of consumer behavior. Fishbein

16

(1975) theorized that behavioral intentions are influenced not only
by attitudes, but by social or normative considerations as well.
Fishbein has represented this behavioral intention model as:

B s BI = AactW + [(NB) (1%)]Wl

0
Where:
B = overt behavior
B1 = behavioral intention
Aact = attitude toward an act
NB = normative belief
MC = motivation to comply with the normative belief
woandwl = anpirically determined weights

This extended model added two new components, a normative belief
(NB) component and a motivation to comply (MC) component, to the Aact
model to predict behavioral intentions. A relative weighting system
has been added to components because it is theorized that intention to
engage in certain types of behavior is more strongly influenced by social
or normative considerations than intention to engage in other types of
behavior (Fishbein, 1975). For example, social or normative consider-
ations may have a greater part in influencing an individual to buy a
particular brand of a status—related item such as jeans, than rela-
tively status-free items such as milk. Indeed, Witter (1978), found
that the use of normative components increased sensitivity of the Aact
measure of attitude when evaluating three products used in her study:
two brands of footwear and one brand of pantyhose. Ajzen and Fishbein
(1973) reviewed research testing the validity of this model in pre-
dicting a wide variety of behaviors. The prediction of behavioral

intentions for activities such as engaging in premarital sexual intercourse,

l7

cheating on an exam, buying various products, and using contraceptives

has supported the model.

VI. EMPIRICAL RESEARCH FINDINGS

Deceptive Advertising

 

Gurol (1978) identified two main approaches used by researchers
to identify deception. The first approach consisted of asking subjects
if advertisements, claims, or situations are deceptive. Although this
method has been used by several researchers (Haefner, 1972; Haefner and
Permut, 1974; King and Wise, 1974; Ford, Kuehl and Reksten, 1975), it
has been criticized as a poor method for detecting deception. Comment-
ing on this procedure, Armstrong and Russ indicated, "to detect de—
ception, it is insufficient (and logically deficient) to ask exposed
relevant consumers whether an advertisement is deceptive. It appears
self-evident that a consumer who judges an advertisement to be deceptive
is not himself deceived." (1975, p. 26)

Indeed, Haefner (1972) found that a sample of adults' and students'
rating of the deceptiveness of certain television commercials did not
agree with unofficial ratings by FTC lawyers. He found, however, a
strong correlation between advertisements that were seen as deceptive
and the same advertisements being described as annoying, offensive, and
as insulting to the intelligence. The Haefner and Permut study (1974)
used a similar approach, which yielded similar findings.

When asked to rank fourteen previously classified types of deception
from "most" to "least" deceptive, subjects in the Ford, Kuehl, and Reskten
study (1975) were quite consistent in their rankings. Ranking fourteen

bogus print ads designed to correspond to the previously devised classification

18

scheme, however, provided inconsistent rankings. The authors concluded
that, "deception is in the eyes of the beholder."

In a study similar to the Ford, Kuehl, Resktin Study (1975),
respondents were asked to rank-order definitional statements and news-
paper advertisements. King and Wise (1974), found that the subjects
were consistent in their ranking of the statements, but inconsistent in
actual advertisement ranking. After reviewing the above studies, Gurol
concluded, "These results suggest that measuring deception according to
consumers' assessment of deceptiveness is not the right approach . In-

stead, it seems that using consumers' brand attribute beliefs to detect

 

advertising deception is basically the correct approach," (1977, p. 183).
Empirical research designed to measure deception which does make

use of brand attribute beliefs has been completed by Armstrong, Kendall

and Russ (1975), Armstrong, Gurol and Russ (1979), Kuehl and Dyer (1976,

1977).and Witter, (1978). Armstrong, Kendall, and Russ (1975), use a

model based on consumers' perceptions and beliefs of salient claims. They

showed four potentially deceptive television advertisements to church and

PTA groups. The structured instrument they used contained ten claims

for each ad, designed to determine perceived salient and believed claims.

After determining the false claims, the following measures were calculated

for each advertisement: (1) liklihood of an ad to deceive relevant

consumers; (2) what false claims were perceived, believed, and salient,

and (3) the charaCteristics of those deceived. The pilot study had a

number of limitations, among them a small sample size and an after-

only experimental design, which could not measure deceptions held prior

to exposure to the potentially deceptive advertisement.

Armstrong, Gurol and Russ (1979) further tested the Armstrong-

l9

Kendall-Russ model, using a much larger sample size and a pretest-
posttest control group experimental design to eliminate carry-over
effects. Actual Listerine mouthwash ads used in the experimental
treatment were found to be deceptive.

Kuehl and Dyer (1976, 1977) used the original Fishbein attitude
model to measure deception, using junior high school students as subjects
and hypothetical print advertisements about two hypothetical products.
They used a modified version of Gardner's normative belief technique
to empirically test brand attribute measures on Gardner's claim-fact
and claim—belief typology of deception. They found that, in all cases,
the direction of findings indicated that the functional attribute was
more favorably evaluated than the norm after exposure to the hypothetical
deceptive advertisement.

Witter (1978) tested the use of the extended Fishbein behavioral
intention model in detecting deception, using actual advertisements of
mouthwash, pantyhose, blemish cream, and footwear. The results of the
pretest-posttest control group experimental design provided support for
the use of this model in detecting deception. In some cases, the use of
the normative component of the extended model increased the sensitivity
of the model in detecting deception.

Rotfield and Reid (1977, p. 129) noted that, for the most part,
academic researchers have "tended to produce unrelated, unreplicated
studies which are rarely designed for use in specific pending cases."
This research provides further testing of Witter's (1978) proposed
application of Fishbein's behavioral intention model in detecting
deception. It also tests its use in measuring the amount of residual

deception after exposure to a corrective advertisement.

Corrective Advertising

 

Armstrong, Gurol and Russ (1979) classified empirical research
on corrective advertising into two groups, according to the dependent
variables used. Dyer and Kuehl (1974), Hunt (1972), and Kassarjian,
Carlson, and Rossin (1975), studied attitudes toward the brand. Dyer
and Kuehl (1978), Kuehl and Dyer (1976, 1977), Mazis and Adkinson (1976),
and Armstrong, Gurol and Russ (1979) studied brand beliefs on a multi-
attribute basis.

Dyer and Kuehl used radio and print ads for soft drink and suntan
lotion brands as stimuli. Students were used as subjects in this study
of the effects of message strength, message source and number of exposures
of corrective messages upon awareness, attitude, corporate image and
recall. In print ads, a strong corrective statement in the form of an
FTC news release resulted in lower intention to buy. Also, the FTC-
source, high strength message resulted in a less trustworthy firm image,
while company—source high strength messages brought a more trustworthy
image. Neither of these results held true for the broadcast messages,
however. Gurol (1977) criticized the message designs as "artificial"
and "ambiguous."

Hunt (1972) studied the effects of three ad attacks (explicit,
general, and none) and three inoculation ads (explicit, general, and
none) on inhibiting corrective ad effects. Actual Chevron ads or
slight variations were shown to college students and attitudes toward
the brand was measured. The study found that a corrective ad does reduce
favorableness of attitude toward a brand and that by varying the content
of the corrective ad, varied amounts of reduction in favorableness of
attitude can be obtained.

20

21

Kassarjian, Carlson and Rosin (1975) attempted to assess the
impact of a corrective advertisement for an unfamiliar brand of motor-
CyC1e safety device on attitude toward the product and the retailer.
College students viewed the ad in five pages of a New York newspaper
under both natural and forced conditions. It was found that attitudes
toward the product became more negative, but that the negative attitudes
did not generalize toward the retailer.

The three studies reported here consistently found that corrective
advertisements were effective in erradicating the effects of a deceptive
message by changing attitudes in a negative direction. They did not
examine subjects' prior brand usage and attitudes (Gurol, 1977) or try
to advance knowledge by constructing and examining innovative behavioral
science concepts and variables related to remedial messages (Kuehl and
Dyer, 1977).

The Mazis and Adkinson (1976), Kuehl and Dyer (1976, 1977) and
Dyer and Kuehl (1978) studies used the brand attribute beliefs com-
ponent of the Fishbein attitude model. Mazis and Adkinson used four
versions of Listerine messages inserted into an audio tape of a late-
night TV interview show. They were (1) noncorrective copy with germ-
killing appeal, (2) corrective copy with germ-killing appeal (FTC
source), (3) corrective copy with germ-killing appeal (company source),
and (4) noncorrective copy with nonrgerm-killing appeal. After listen—
ing to the program, the college student subjects were asked to evaluate:
(l) the goodness or badness of specific attributes of several brands;
(2) the probability of particular brands being associated with beliefs;
and (3) overall brand attitudes. The study found that: (l) the

corrective messages had an impact on evaluation of the belief about

22

Listerine's cold and sore throat prevention abilities, (2) the

corrective messages caused less favorable perception of Listerine on the
cold and sore throat prevention attribute, (3) perception of germ—killing
properties were also negatively affected, and (4) there was no difference
in the effectiveness of FTC or company source.

In the second stage of the Kuehl and Dyer (1976, 1977) experiment
described earlier (which first determined deception), the researchers
used the original Fishbein model in a modified version of the normative
belief technique to study corrective advertising. After a two-week
time interval following the deception experiment, the same respondents
were asked to read a corrective advertisement and to indicate their brand
belief ratings. Combinations of high or low strength and FTC or company
source messages were used. The study found that the FTC source advertise—
ments resulted in significantly lower brand belief levels than the company
source messages.

Dyer and Kuehl (1978) used the same framework in a longitudinal
experimental design to examine the delayed effects of Listerine corrective
advertising. The study found: (1) the Listerine ad to be potentially
deceptive; (2) that limited exposure to corrective advertisements was
ineffective in bringing brand beliefs into equilibrium with "normative"
levels; and (3) that corrective ads affect other attributes of the product
category even though they are not addressed in a remedial message.

In the second stage of the Armstrong, Gurol and Russ (1979)
experiment, the Armstrong-Kendall-Russ model was used to measure the
effects of a two-sided objective information message in erradicating
the residual effects of deception in consumers' minds. Self-designed

"corrective ads" identified as either FTC-source or company-source were

23

shown to eliminate the deception caused by the actual Listerine ad. The
study found: (1) the two-sided objective information messages to be
highly effective in erradicating the residual effects of the deceptive
advertisement, and (2) a persistent effectiveness of the corrective

message after a six-week period.

CHAPTER III

METHODS AND PROCEDURES

This study further tests the use of the Fishbein behavioral
intention model in determining deception, a procedure proposed by Witter
(1978). Three of the original questionnaires formulated by Witter will
be used in this study. An explanation of the previous evaluative criteria
selection and questionnaire development will be provided for purposes of

clarity.

I. THE PRODUCTS

Three products were selected for use in this study from among
those previously investigated or currently under consideration by the
FTC for possible litigation. These are: mouthwash, pantyhose, and
footwear. Actual advertisements for those products were used in the
initial phase of the study. Actual products and advertisements, rather
than fictitious ones, were chosen because consumers make decisions based
on information and experience. Sawyer (1977) urges the use of real
products in corrective advertising research because only real products
allow the measurement of corrective advertising effects on erradicating
residual beliefs of previous exposure to advertisements of the product.

The specific brands used in this study were determined by focus
group interviews with 31 fnfimnen University of Tennessee, Knoxville

students. Those brands most frequently mentioned were:

24

25

Mouthwash Footwear Pantyhose

Scope Scholl Sheer Energy
Listerine Famolare Hanes--Ultra Sheer
Lavoris Fanfares L'eggs

Listermint Earth Shoes No Nonsense

II. THE EVALUATIVE CRITERIA

After the products were chosen, salient evaluative criteria were
established for each product. Salient evaluative criteria are the
criteria that consumers actually use to compare and evaluate products
and brands. For example, though a customer might consider price to be an
important factor in purchasing goods, if he knows that all brands of a
certain product class he wishes to buy have the same price, price would
not be a salient evaluative criteria for that product class. Focus
groups, any FTC rulings on the product, and company claims made in the
advertisements were used to establish salient evaluative criteria. The
sources from which the evaluative criteria were derived are included in
Table l, (Witter, 1978, p. 29).

The focus groups consisted of six to eight participants who discussed
the attributes they considered important when buying the product. These
discussions were tape recorded and later tabulated for response frequencies.
After discussing the attributes, the students were asked to rank the
evaluative criteria and to evaluate how the various brands differed for
each of the attributes. Each focus group was asked to select evaluative
criteria for one product only. The structure of the focus group interviews

is included in Appendix A.

III. THE QUESTIONNAIRE

After salient evaluative criteria were determined, a questionnaire

Table 1

Final Evaluative Criteria By Source

 

 

ADVERTISING
FOCUS GROUPS FTC CLAIMS
Mouthwash
Taste x x
Smell x
Price x
Odor Control x x
Germ Killing Power x1 x
Cold Prevention x1 x
Brand Name x x
Casual Shoes
Originality x x
Feel (Natural vs. Unnatural) x x
Leg Exercise x
Health Value for Feet x x
Price x x
Effect on Posture x x
Durability x
Pantyhose
Texture x x
Stretch x
Durability x
Cost x
Attractiveness to the
Opposite Sex x
Sizing x x
Feel During Wearing x x

 

lFTC vs. Warner-Lambert Company Docket 8891, June 1972

26

27

was developed. The questionnaire was pretested on 51 freshmen Textiles
and Clothing students at the University of Tennessee, Knoxville and
revised for clarity.

Appendix B contains the questionnaire used for mouthwash.
Questionnaires for footwear and pantyhose were similar in format. Only
the posttest and experimental groups received a questionnaire including
page 6 of the posttest.

Questions seeking demographic data as well as information necessary
to operationalize the Fishbein model components were included. Additional
information was gathered (in parts II and VI of the questionnaire) for
analysis beyond the scope of this study.

Components of the Fishbein model and how they were operationalized
in the questionnaire are found in Table 2, (Witter, 1978, p. 30). Behavioral
intentions were measured by subjects' responses to questions of whether or
not they intended to buy a particular brand. Respondents indicated their
intention on a scale from 0 (I will not buy this brand) to 6 (I will buy
this brand), with 3 being the neutral response. Subjects were asked to
respond to intentions to buy at the brand level, rather than at the pro-
duct class level, since Dulany (1961) has found that the more specific
the intention becomes, the higher the correlation with a specific behavior.

Affect was measured by the subjects' scaling the importance of
the evaluative criteria. The respondent was asked to scale the importance
of the various criteria from 0 (very unimportant) to 6 (very important).
Since the importance of the evaluative criteria would be generalized
across the product level, the response was solicited at that level.

Beliefs were measured by response to the likelihood that specific

brands would contain each attribute. Responses at the brand level were

28

from O (extremely improbable) to 6 (extremely probable).

TABLE 2
FISHBEIN COMPONENT QUESTIONNAIRE
BI (Behavioral Intention) Part I (Questions 8-11)
Aact (Attitude)
n
Aact = z A, B,
1 1 1
A (Affect) Part 111 (Questions 19-25)
B (Beliefs) Part IV (Questions 26—53)
NB (Normative Beliefs) Part V (Questions 54-57)
MC (Motivation to Comply) Part VI (Question 58)

Normative Belief was measured by asking the subjects' estimates
that others would recommend buying each brand. Responses ranged from
0 (extremely improbable) to 6 (extremely probable).

The Motivation to Comply component was measured by asking the
respondents to rate the importance they placed on others' advice when
selecting the product. Responses ranged from 0 (don't care at all) to
6 (care a great deal).

Occupation and education data were weighted and combined according
to Hollingshead's Two Factor Index of Social Position (1965) to create
a social position variable. This factor was used to identify socio-
economic differences between experimental and control groups.

The questions were evaluated on a O to 6 scale, which was con-
verted to a l to 7 scale for statistical analysis. The O to 6 scale
was used in order to allow students to respond on computer scan sheets.
Table 3 (Witter, 1978, p. 36) contains the directional rating of variables

in the study.

29

TABLE 3

VARIABLE RANGE HIGH SCORE LOW SCORE
Behavioral

Intentions l to 7 High Importance Low Importance
Affect 1 to 7 High Importance Low Importance
Belief l to 7 High Importance Low Importance
Normative Belief 1 to 7 High Importance Low Importance
Motivation to

Comply l to 7 High Importance Low Importance
Hollingshead

Social Position ‘ l to 7 Low Position High Position

IV. THE SAMPLE

The sample consisted of public high school students from two
high schools in Grand Rapids, Michigan and one high school in Holt, Mflflngmi.
Most of the students were home economics class members, although some
of the data were gathered in a study hall and an advertising class.
Three hundred sixty—four students participated in the pretest, with
249 students completing valid first posttest questionnaires. The 115
failures to complete the first posttest questionnaires were due to
absenteeism, incorrect completion of the questionnaire, or refusal to
complete the posttest questionnaire. One hundred fifteen students
validly completed the second posttest questionnaire. The final sample
size for each product group was:

mouthwash

posttest l n - 102
posttest 2 n = 54

30

footwear
posttest l n = 78
posttest 2 n = 33
pantyhose
posttest l n = 69
posttest 2 n = 28

Since advertisements for the three products to be tested are
found in magazines designed for high school students, it is assumed
that the sample is made up of potential users of the products selected.
It is essential that the sample be part of the potential market segment,
since, as previously noted, Gardner (1976) defined deception as occuring
within specific market segments. It would not matter, for example,
whether or not a pantyhose commercial was found to deceive men, since
men do not form part of the market segment for pantyhose. For this
reason, the pantyhose and footwear questionnaire were administered to

female students only.
V. THE RESEARCH DESIGN

A two-stage non-equivalent control group experimental design
was used in the study. The pretest-posttest component of the design
allows for measurement of differences in attitude scores, while the
control group provides increased internal validity. Groups constituted
naturally assembled classrooms rather than randomly-assigned assemblages,
since the use of classroom group provided the least interferences
with high school schedules. Campbell and Stanley (1963) consider
this design to be acceptable when random assignment of groups is not
feasible.

An identical design was used in the experimental evaluation for

31

each of the three product classes. Figure 1 contains a diagram

illustrating the research design.

FIGURE 1

TWO—STAGE NON—EQUIVALENT CONTROL GROUP DESIGN

01 O2 02 03
01 X1 02 X2 03
01 = Pretest
02 = First Posttest
03 = Second Posttest
X1 = Experimental Treatment Number 1 (original advertisement)
X2 = Experimental Treatment Number 2 (corrective advertisanent)

Members of the sample were assigned by classroom to one of two
groups-—experimenta1 or control. The pretest questionnaire was
administered to both groups. One week later, the experimental group
was shown an advertisement referred to as Experimental Treatment Number
1 and the first posttest questionnaire was administered to both groups.
Differences in the mean gain scores for the experimental and the control
group were analyzed, using a t-test for differences.

One week after the first posttest, an advertisement modeled after
those developed by the FTC for use as corrective advertisements
(Experimental Treatment Number 2) was shown to the groups. The second
posttest questionnaire was administered and differences in the mean

gain scores were again analyzed, using a t-test for differences.

VI. THE ADVERTISING TREATMENT

The advertisements used in Experimental Treatment Number 1 con-
sisted of actual advertisements for four different brand items of one
product class. The advertisements were obtained from magazines designed
to be sold to the sample group. One advertisement contained an allegedly
deceptive statement. The allegedly deceptive statements, by product
class, were:

Mouthwash: "Two-fisted. A real fighter. Kills germs by millions. 50
strong it keeps breath fresh for hours... and hours. Listerine antiseptic.
Kills Germs by Millions on Contact. For Bad Breath, Colds and Resultant
Sore Throats."

Footwear: "What you don't feel is the toe grip at work. It makes your
toes "hang on." Which exercises your legs to help firm and tone them
while you walk.”

Pantyhose: "Gentlemen prefer Hanes."

Experimental Treatment Number 2 also consisted of four advertise-
ments. Three of the advertisements were identical to the three "non-
deceptive" ads in Experimental Treatment Number 1, while the allegedly
deceptive advertisement was replaced by a "corrective advertisement."
The Listerine corrective advertisement was taken from an advertising
mockrup obtained from Warner-Lambert. The section containing the
corrective message was reproduced. The other "corrective advertisements"
were identical to the original advertisement except that the allegedly
deceptive portion was omitted and a "corrective" portion inserted,
modeled in format after the Listerine corrective advertisement. The
"corrective" messages were:

32

33

Mouthwash: "Listerine will not help prevent colds or sore throats
or lessen their severity."

Footwear: "Scholl Exercise Shoes will not firm or tone legs."
Pantyhose: There is no evidence which indicates that men prefer the
Hanes brand of Pantyhose."

The advertising treatments are contained in Appendix C.

Each participant was provided with a black and white photostatic
copy of the advertisement, while a color slide of the same advertisement
was projected on a screen. They were allowed to refer freely to the
advertisements as they completed the questionnaires. This eliminated

retention of material as a variable in the study.
VII. HYPOTHESES AND STATISTICAL TREATMENT

Hypotheses l and 2 are related to the deceptive advertising issue,
while hypotheses 3 and 4 are related to the erradication of deception by
corrective advertising. Hypotheses are stated in the null form.

H1: There will be no change in attitude following exposure to
Experimental Treatment Number 1.

This Hypothesis deals with the original Fishbein model,

Aact = 2 B1 A1. Mean gain scores comparing attitudes were calculated
and a titest for independent samples was used to compare the two groups.
Mean gain scores were used because it is necessary to compare pretest
and posttest scores on an individual basis rather than as a group.

H2: There will be no change in behavioral intentions following

exposure to Experimental Treatment Number 1.

The second hypothesis deals with the extended Fishbein model

(1967), which measures behavioral intentions, B; BI = Aactw + [(NB)(MC]w
0

1.

34

Mean gain scores comparing behavioral intentions were calculated for
both experimental and control groups, and a t-test for independent
samples was used to compare the two groups.

H3: There will be no change in attitude following exposure to
Experimental Treatment Number 2.

Mean gain scores comparing attitudes were calculated and a t-test
for independent samples was used to compare the two groups.

H4: There will be no change in behavioral intentions following
exposure to Experimental Treatment Number 2.

Mean gain scores comparing behavioral intentions were calculated
for both experimental and control groups, and a t-test for independent

samples was used to compare the two groups.

CHAPTER IV
RESULTS AND DISCUSSION

This chapter summarizes the outcome of the data collection
process and discusses the statistical analysis of the data. The first
major section discusses the validity of the Fishbein models in pre-
dicting attitudes and behavioral intentions, as measured by the question-
naire. The second section describes the sample and compares the control
and experimental groups with respect to certain selected demographic
characteristics. Findings related to the hypotheses are presented

in the final section.
I. ANALYSIS OF THE FISHBEIN MODELS

This study focuses on the use of Fishbein models as a method of
predicting attitudes and behavioral intentions. Therefore, an analysis
was done to determine those prediction capabilities. Multiple regression
analysis was used to analyze both the original and the extended models
and their ability to predict the respondents' stated intention to
buy specific product brands. The results are shown in Tables 18 and
19, found in Apprndix E.

The original attitude model (Aact = 2 B1 A1) was found to
accurately predict behavioral intentions t6 buy 11 of the 12 product
brands. The model was significant at the .0001 level in predicting

intention to buy Scope, Listerine, Lavoris and Listermint brands of

mouthwash and Scholl, Fanfares and Earth Shoes brands of footwear.

35

36

The model was significant at the .001 level for Sheer Energy panty-

hose; at the .01 level for Famalare brand of footwear and Hanes

Ultra Sheer pantyhose; and at the .05 level for No Nonsense pantyhose.
The model for L'eggs brand of pantyhose was not found to be a significant
predictor of behavioral intentions.

The extended Fishbein model was also found to predict behavioral
intention to buy 11 of the 12 products. Behavioral intention to buy Scope,
Listerine, Lavoris, and Listermint mouthwash, and Scholl, Fanfares, and
Earth Shoes footwear were predicted at the .0001 level of significance.
The model was significant in predicting behavioral intentions to buy
Famloare footwear and Sheer Energy pantyhose at the .001 level; Hanes
Ultra Sheer at the .01 level, and No Nonsense at the .05 level. Again,
for only one product brand, L'eggs pantyhose, was the model not found to

be a significant predictor of behavioral intention.

II. THE SAMPLE

Data describing the race, sex, age, parent's occupation, and
parent's education were gathered for both control and experimental
groups. Another demographic variable, socio-economic status, was deter-
mined by the use of Hollingshead's Two Factor Index of Social Position
(1965). The index uses parents' occupation and education as classifi-
cation factors, with each being ranked from 1 to 7. The 1 represents
the highest classification, while the 7 represents the lowest. The
two factors are weighted and combined in the following manner:

SES LEVEL = Occupation x 7 + Education x 4
Final social position scores ranged from a possible 11 to a possible

77. Low scores indicated upper social position, while higher scores

37

indicated lower position.

The control and experimental groups were compared with respect
to certain selected variables: parent's occupation, parent's education,
socioeconomic status level, purchasing frequency, and behavioral in-
tentions. Purchasing frequency was measured by a pretest response to,

"How often do you generally purchase (product class)?" Respondents

 

were asked to circle a response ranging from "several times a month" to
"never." Behavioral intentions were measured by a pretest response to
the question, "Would you estimate how likely it would be that you would

purchase the following brands of (product class)?" Responses were

 

made on a scale of 0 (I will not buy this brand) to 6 (I will buy
this brand), with a neutral response of 3. Demographic characteristics
of the control and experimental groups will be discussed by product

groups.

Mouthwash Group

 

Fifty-four control subjects and forty-nine experimental subjects
completed the first posttest questionnaire. Demographic characteristics
of the two groups are summarized in Table 4. Ninety-six percent of the
control group were males, while males made up only 67 percent of the
experimental group. The high proportion of males completing the
mouthwash questionnaires was due to data collection conditions.

Since the other two products used required female consumers as
respondents, the male class members were often asked to complete
mouthwash questionnaires. The majority of both control and experi—
mental respondents were 17 years old. All of the respondents were

18 years of age or younger. The control group contained a lower

Variable

Race
Caucasian
Black
Other
No Response

Sex
Male
Female
No Response

Age

Under 15

15

16

17

18

19

Over 19

No Response

SES Level
11—17
18-27
28-43
44-60
61-77
No Response

Table 4

Mouthwash Group
Summary of Demographic Information
for Control and Experimental Groups

Control
N Z
29 54
19 35
5 9
l 2
52 96
l 2
l 2
3 6
8 15
6 11
32 59
5 9
0 0
0 0
0 0
0 0
4 7
10 19
20 37
19 35
1 2
N = 54

38

Experimental
N Z

 

67
12
16

4

ch‘w

33 67
15 31

10

67
18

HOOKDWI—‘MO

35

16

39

proportion of subjects in the higher social position than did the
Experimental group. Only 26 percent of the control subjects were
in the three higher social positions, while 43 percent of the experi-
mental subjects were classified into those groups.

Experimental and control groups were compared with respect
to level of parent's occupation, parent's education, social position,
purchasing frequency, and behavioral intentions by means of a t-test
for differences. The results are presented in Table 5. Three variables,
parent's occupation, socio-economic status, and intention to purchase two
brands of mouthwash were found to be significant at the .05 level. For
both the occupational and the social position variables, the experimental
group yielded lower means, indicating a higher social position and a
higher occupational level. The means for the behavioral intention vari-
able also differed significantly for two brands of mouthwash: Lavoris
and Listermint. In both cases, before exposure to the advertising treat—
ments, the experimental group estimated that they were less likely to
buy the brands than did the control group. Both the control and
experimental groups indicated that they purchase mouthwash, on the
average, "several times" to "once" a year. The mean response of
both groups indicated that the average educational attainment of the
household head was high school graduate or some college, business or

vocational training.

Footwear Group

 

Forty-five control subjects and thirty-five experimental sub-
jects completed the first posttest questionnaire. Demographic character-

istics of the footwear group are summarized in Table 6. All respondents

Table 5

Mouthwash Group
Comparison of Control and Experimental Groups
for Selected Demographic Variables

 

N = 102
Variable N Mean SD ‘22
Parent's Occupation
Control 53 5.2642 1.571 100
Experimental 49 4.5102 1.570
Parent's Education
Control 53 3.8868 1.368 100
Experimental 49 3.4898 1.583
SES Level
Control 53 4.0189 .930 100
Experimental 49 3.6327 .906
Purchasing Frequency
Control 53 3.6415 1.922 100
Experimental 49 3.4490 1.757
Behavioral Intentions
Scope-Control 53 4.5849 2.107 100
Scope—Experimental 49 3.9796 2.136
Listerine-Control 53 4.2642 2.237 100
Listerine- 49 3.5306 2.199
Experimental
Lavoris-Control 53 3.4906 1.783 100
Lavoris—Experimental 49 2.7143 1.720
Listermint-Control 53 4.7547 1.860 100
Listermint- 49 3.6531 2.127
Experimental
* p <.05

40

.42*

.36

.12*

.53

.44

.67

.23*

.79*

41

were female. Only 38 percent of the control group were Caucasian, while
Caucasians made up 77 percent of the experimental group. A greater per-
centage of the control group were black (56 percent vs. 14 percent of

the experimental group). Nine percent of the experimental group responded
to the "other" racial category, while none of the control group made

that response. The majority of both control and experimental groups

were 17 years old or younger, but 45 percent of the control group and

only 20 percent of the experimental group were 15 or younger. The
majority of both groups came from the two lowest social position
categories.

Results of t—tests used to compare experimental and control
groups with respect to levels of parents' occupation, parents' educa-
tion, social position, purchasing frequency, and behavioral intention
are contained in Table 7. The purchasing frequency variable was found
to be significant at the .001 level. This indicates that members of
the control group purchase casual shoes, on the average, more fre-
quently than do members of the experimental group. The mean response
indicates a purchase of "several times a year" to "monthly" for the
control group, while only "once a year" to "several times a year"
is indicated for the experimental group. Behavioral intentions to
purchase Fanfare brand of footwear was also significant. 0n the pre-
test questionnaire, the control group indicated a greater behavioral
intention to buy the Fanfare brand than did the experimental group.

In both groups, the average level of educational attainment for the
household head was high school graduate or some college, vocational

or business school training.

Table 6

Footwear Group
Summary of Demographic Information
for Control and Experimental Groups

 

Control Experimental
Variable N z N %
Race
Caucasian 17 38 27 77
Black 25 56 5 14
Other 0 0 3 9
No Response 3 7 0 0
Sex
Male 0 0 0 0
Female 42 93 35 100
No Response 3 7 0 0
Age
Under 15 3 7 0 0
15 17 38 7 20
16 13 29 6 17
17 8 18 18 51
18 3 7 2 6
l9 0 0 2 6
Over 19 0 0 0 0
No Response 1 2 0 0
SES Level
11-17 Higher 1 2 l 3
18-27 4 9 2 6
28-43 5 11 3 9
44-60 14 31 18 51
61-77 21 47 10 29
No Response 0 0 l 3
N = 45 N = 35

42

Table 7

Footwear Group
Comparison of Control and Experimental Groups
for Selected Demographic Variables

N = 79
Variable N Mean SD .DF t
Parent's Occupation
Control 45 5.5333 1.854 77 1 04
Experimental 34 5.1176 1.610 °
Parent's Education
Control 45 3.4889 1.362 77 -l 39
Experimental 34 4.1176 1.343 '
SES Level
Control 45 4.1111 1.071 77 0 48
Experimental 34 4.0000 .953 '
Purchasing Frequency
Control 45 2.8889 1.049 77 _3.95**
Experimental 34 3.8235 1.029
Behavioral Intentions
Scholl-Control 45 4.2444 2.278 77 0 58
Scholl-Experimental 34 3.9412 2.322 '
Famolare-Control 45 3.5778 1.815 77 0 95
Famolare- 34 3.1765 1.914 '
Experimental
Fanfare-Control 45 4.3778 1.969 77 2 24*
Fanfare-Experimental 34 3.4412 1.655 '
Earth Shoes-Control 45 5.1333 2.040 77 1 07
Earth Shoes- 34 4.6176 2.202 '
Experimental
* p <.05
** p <.01

43

44

Pantyhose Group

 

Thirty-eight control subjects and thirty-one experimental sub-
jects completed the first posttest. Demographic characteristics are
summarized in Table 8. In both the control and experimental groups,
about 50 percent of the respondents were Caucasian. Thirty-six percent
of each group were black, with the remaining respondents of other
racial heritage. All respondents were female. In both groups, the
majority of the respondents were in the two lowest socio-economic
groups. Twenty—seven percent of the control subjects were 15 years
old or younger, while the experimental group contained no subjects
under 16 years of age. In both groups, however, the majority of
subjects were 16 or 17 years old.

Results of t-tests used to compare experimental and control
groups with respect to levels of parents' occupation, parents' edu-
cation, social position, purchasing frequency, and behavioral intention
are contained in Table 9. Only the purchasing frequency variable was
found to be significant at the .05 level. The mean responses (3.1579
for the control group, 2.3226 for the experimental), would indicate
that members of the control group, on the average, purchase pantyhose

only "several times a year,”

while the experimental group purchases the
Product "monthly." The 3.9737 mean for the control group and 3.9355
mean for the experimental would indicate that the highest educational

attainment for the head of the household was, on the average, graduation

from high school.

Summary of Demographic Information

 

Control and experimental groups were described with respect to

Variable

Race
Caucasian
Black
Other
No Response

Sex
Male
Female
No Response

Age
Under 15
15
16
17
18
19
Over 19

SES Level
11-17
18-27
28-43
44-60
61-77

Table 8

Pantyhose Group
Summary of Demographic Information
for Control and Experimental Groups

Control
N Z
19 50
14 36
4 ll
1 3
0 0
37 97
1 3
1 3
9 24
13 34
14 37
0 O
l 3
0 0
0 0
1 3
10 26
13 34
14 37
N = 38

45

 

Experimental
N Z
16 52
ll 36
4 l3
0 0
0 0
31 100
0 0
0 0
0 0
18 58
13 42
0 0
0 0
0 0
0 0
l 3
6 19
ll 36
13 43
N = 31

Table 9

Pantyhose Group
Comparison of Control and Experimental
for Selected Demographic Variables and Behavioral Intentions

 

Variable N Mean SD DF t

Parent's Occupation

Control 38 5.3947 1.443 67 O 02
Experimental 31 5.3871 1.647 '
Parent's Education
Control 38 3.9737 1.619 67 -0 10
Experimental 31 3.9355 1.482 °
SES Level
Control 38 4.0526 .868 67 -0 52
Experimental 31 4.1613 .860 °
Purchasing Frequency
Control 38 3.1579 1.551 67 2.46*
Experimental 31 2.3226 1.194
Behavioral Intentions
Sheer Energy-Control 38 4.2895 2.039 67 -0 21
Sheer Energy- 31 4.3871 1.745 °
Experimental
Hanes-Control 38 3.6053 1.868 67 0 29
Hanes-Experimental 31 3.4839 1.503 '
L'eggs-Control 38 4.0789 2.173 67 -1 65
L'eggs-Experimental 31 4.9032 1.921 '
No Nonsense-Control 38 4.9474 2.323 67 l 19
No Nonsense- 31 4.2903 2.209 °
Experimental
* p <.05

46

race, sex, age, and social position. T-tests were used to compare the
groups with respect to parent's occupation, parent's education, social
position, product group purchase frequency, and behavioral intention
to buy specific brands within the product group. The groups were not
always comparable. The mouthwash groups differed in distribution by
social position, parent's occupation, and behavioral intention. The
footwear groups differed in purchasing frequency and behavioral in-
tention distribution. The pantyhose group differed in purchasing fre—

quency distribution.

III. TESTING THE HYPOTHESES

Stage I: Detecting Deception

 

H1: There will be no change in attitude following exposure to
Experimental Treatment Number 1.

This hypothesis is designed to determine if attitude change resulted
after exposure to the potentially deceptive advertisement. Pretest and
Posttest scores were computed according to the original Fishbein model,

n
(Aact = 2 Ai Bi)’ for subjects in the control and experimental groups.

Attitudelscores were computed for each subject and mean gain scores were
calculated for each group. The independent samples t-test was used to
compare the groups.

Tables 10-12 summarize attitude change. No significant difference
was found between the control and experimental groups. Null hypothesis
1 therefore cannot be rejected.

H2: There will be no change in behavioral intention following

exposure to Experimental Treatment Number 1.

This hypothesis deals with change in behavioral intention which

Table 10

Stage 1: Posttest l - Pretest
Attitude Change and Behavioral Intention Change
Gain Scores for Mouthwash Brands

N = 102

Variables N Mean SD ‘DF t
Aact Attitude

Gain Score

Scope-Control 53 - 4.7358 69.912 100 -0 08
Scope-Experimental 49 - 3.6735 70.256 °
Listerine-—Control 53 - 2.7545 70.212 100 -1 l7
Listerine-Experimental 49 12.4694 59.760 '
Lavoris-Control 53 .3585 65.359 100 _0 40
Lavoris-Experimental 49 5.7143 68.246 '
Listermint-Control 53 - 7.7170 59.189 100 0 47
Listermint - 49 -14.1429 78.988 '

Experimental

Fishbein Extended

Model Gain Score

Scope-Control 53 - 4.4717 68.764 100 -0 15
Scope-Experimental 49 - 2.2449 80.105 °
Listerine-Control 53 - 1.0189 67.331 100 -l 03
Listerine-Experimental 49 12.7757 67.210 °
Lavoris-Control 53 2.0755 64.003 100 _0 26
Lavoris-Experimental 49 5.6327 74.261 °
Listermint-Control 53 - 6.1509 58.681 82 271 0 53
Listermint- 49 ~14.1837 88.702 ° °

Experimental '

1Since it cannot be assumed that the 2 groups had the same variance, an
F test of sample variances was performed. F was less than or equal to
.05, so the degrees of freedom were adjusted and a t based on the separate
variance estimate was used.

48

Stage 1:

Gain Scores for Footwear Brands
N:

Variables

Aact Attitude
Gain Score

Scholl-Control
Scholl-Experimental

Famolare-Control
Famolare-Experimental

Fanfares-Control
Fanfares-Experimental

Earth Shoes-Control
Earth Shoes-
Experimental

Fishbein Extended
Model Scores

Scholl-Control
Scholl-Experimental

Famolare-Control
Famolare-Experimental

Fanfares-Control
Fanfares-Experimental

Earth Shoes-Control
Earth Shoes-
Experimental

1Since it cannot be assumed that the 2 groups had the same variance,

44
34

44
34

44
34

44
34

44
34

44
34
44
34

44
34

Table 11

Posttest l - Pretest
Attitude Change and Behavioral Intention Change

78

Mean

.0909
.8529

.2273
.8824

.4318
.2059

.3182
.8529

.5000
.1471

.9091
.0588
.8182
.7059

.9545
.8824

60.
37.

62.
56.

58.
48.

61.
54.

63.
44.

65.
60.
59.
51.

64.
58.

an F test of sample variances was performed.

to .05, so the degrees of freedom were adjusted and t based on the

separate variance estimate was used.

49

SD

651
823

127
110

205
504

491
626

055
076

060
353
052
956

677
676

73.06

76

76

76

76

76

76

76

.16

.71

.71

.48

.03

.50

.54

.29

F was less than or equal

Table 12

Stage 1: Posttest - Pretest
Attitude Change and Behavioral Intention Change
Gain Scores for Pantyhose Brands

N = 69
Variables N Mean SD 2F
Aact Attitude
Gain Score
Sheer Energy-Control 38 - 6.0000 78.163 67
Sheer Energy- 31 -13.2581 77.017
Experimental
Hanes-Control 38 - 4.5789 71.602 67
Hanes-Experimental 31 - 1.1935 76.659
L'eggs-Control 38 -21.0263 82.567 67
L'eggs-Experimental 31 - 4.0645 79.542
No Nonsense-Control 38 —21.3158 83.824 67
No Nonsense- 31 1.9355 90.618
Experimental
Fishbein Extended
Model Gain Scores
Sheer Energy-Control 38 - 2.5000 83.720 67
Sheer Energy- 31 —11.l935 85.030
Experimental
Hanes-Control 38 - .8947 74.844 67
Hanes-Experimental 31 - .8065 84.884
L'eggs-Control 38 -l9.8421 86.026 67
L'eggs-Experimental 31 - 2.0000 85.778
No Nonsense-Control 38 -17.8158 90.442 67
No Nonsense- 31 1.5806 98.872

Experimental

50

.39

.19

.86

.11

.43

.00

.86

.85

51

results from exposure to a potentially deceptive advertisement as
measured by the Fishbein extended model, B a BI = Aactw()+ [(NB)(MC)]wl.
Behavioral intention scores were computed for each subject and mean
gain scores were calculated for each group. The independent samples
t-test was used to compare the groups.

Changes in behavioral intentions are summarized in Tables 10-12.
No significant difference was found in behavioral intention change

between the control and experimental group. Null hypothesis 2 there-

fore cannot be rejected.
Stage 11: Measuring the Effectiveness of Corrective Efforts

The second stage of the experiment dealt with the issue of correc-
tive advertising.

H3: There will be no change in attitude following exposure to
Experimental Treatment Number 2.

This hypothesis is designed to determine if attitude change
resulted after exposure to the corrective advertisement. Pretest and
posttest scores were computed according to the original Fishbein model
for subjects in the control and experimental groups. Attitude scores
were computed for each subject and mean gain scores were calculated for
each group. The independent samples t-test was used to compare the
groups.

Tables 13-15 summarize attitude change. There was no significant
difference between control and experimental groups, therefore null
hypothesis 3 cannot be rejected.

H4: There will be no change in behavioral intention following

exposure to Experimental Treatment Number 2.

52

This hypothesis deals with change in behavioral intention which
results from exposure to the corrective advertisement, as measured by
the Fishbein extended model, B sBI = Aactw0 + [(NB)(MC)]W1. Behavioral
intention scores were computed for each subject and mean gain scores
were calculated for each group. The independent samples t-test was
used to compare the groups.

Tables 13-15 summarize change in behavioral intention. No
significant difference was found in behavioral intention change between

the control and experimental group. The null hypothesis therefore cannot

be rejected.
IV. NORMATIVE BELIEF TECHNIQUE

The data were further analyzed using the conceptual framework of
Gardner's (1976) normative belief technique and Wilkie's (1973)
suggestion of measurement at the brand belief level. For each product
category, two beliefs were selected, one being the potentially deceptive
belief under study. Another belief closely related to the potentially
deceptive belief was selected to examine whether deceptive or corrective
statements about one attribute may affect beliefs about other attributes.
Mean brand belief scores on the pretest, first posttest, and second
posttest were computed for each brand for control and experimental groups.

Mean scores for the specific brand advertising the potentially
deceptive claim were compared with a normative belief score consisting
of an average mean score of the other 3 brands studied. Results are
presented in Table 16.

Figure 2 illustrates the shifts in cold prevention belief scores

that occurred during the experiment's three phases. In the first stage

Stage 2:
Attitude Change and Behavioral Intention Change
Gain Scores for Mouthwash Brands
N:

Variables

Aact Attitude
Gain Score

Scope-Control
Scope-Experimental

Listerine-Control
Listerine-Experimental

Lavoris-Control
Lavoris-Experimental

Listermint-Control
Listermint-
Experimental

Fishbein Extended
Model Gain Score

Scope-Control
Scope-Experimental

Listerine-Control
Listerine-Experimental

Lavoris-Control
Lavoris-Experimental

Listermint-Control
Listermint-
Experimental

21
32

21
32

21
32

21
32

21
32

21
32

21
32

21
32

Table 13

Posttest 2 - Pretest

54

Mean

-12.
-16.

3.
-10.

2.
3.

-15.
-23.

53

9048
7500

8095
1875

4286
8125

6190
0000

.8095
.1563

.9524
.9375

.2381
.0938

.9048
.5000

53.
63.

52.
75.

51.
64.

45.
76.

60

49.

64.
47.

66.
47.

50.
36.

SD

652
930

793
532

215
670

983
665

.355
175

490
200

023
435

605
097

51

51

51

51

51

51

51

51

.23

.74

.08

.40

.44

1.69

.73

.45

Stage 2:

Gain Scores for Footwear Brands
N:

Variables

Aact Attitude
Gain Scores

Scholl-Control
Scholl-Experimental

Famolare-Control
Famolare-Experimental

Fanfare-Control
Fanfare-Experimental

Earth Shoes-Control
Earth Shoes-
Experimental

Fishbein Extended
Model Gain Scores

Scholl-Control
Scholl-Experimental

Famolare-Control
Famolare-Experimental

Fanfare-Control
Fanfare-Experimental

Earth Shoes-Control
Earth Shoes-
Experimental

10
23

10
23

10
23

10
23

10
23

10
23

10
23

10
23

Table 14

Posttest 2 - Pretest
Attitude Change and Behavioral Intention Change

Mean

11.

33

3000

.7826

38
16.

55.
28

— 4.
—12.

24.
14.

-15.

.1000

4783

4000

.4348

9000
3043

.1000
.0879

.0000
.7391

2000
7824

2000

.9565

SD

83.
51.

77.
74.

72.
53.

93.
46.

41.
42.

39.
.609

41

69.
47.

82.
29.

145
949

202
785

864
229

860
528

278
785

282

140

299

436
230

31

31

31

10.

31

31

31

31

97

1.

1.

.44

.76

19

.24

.31

37

.46

.69

l .
Since it cannot be assumed that the 2 groups had the same variance, an

F test of sample variance was performed.

F was less than or equal to

.15, so the degrees of freedom were adjusted and t based on the separate

variance estimate was used.

54

Attitude Change and Behavioral Intention Change
Gain Scores for Pantyhose Brands

Variables

Aact Attitude
Gain Scores

Sheer Energy—Control
Sheer Energy-
Experimental

Hanes-Control
Hanes-Experimental

L'eggs-Control
L'eggs-Experimental

No Nonsense-Control
No Nonsense-
Experimental

Fishbein Extended
Model Gain Scores

Sheer Energy-Control
Sheer Energy-
Experimental

Hanes-Control
Hanes-Experimental

L'eggs-Control
L'eggs-Experimental

No Nonsense-Control
No Nonsense-
Experimental

State 2:

Table 15

Posttest 2 - Pretest

N = 28
Mean SD
- 3.5263 66.204
-29.5556 34.413
13.2105 68.464
-19.5556 47.242
-10.7895 71.303
- 8.3333 38.613
-18.2105 70.025
-20.1111 49.936
8.4211 57.529
13.1111 43.386
12.6842 52.071
14.1111 69.033
15.7895 49.484
14.2222 51.533
12.8947 57.683
14.8889 53.069

55

26

26

26

26

26

26

26

26

.10

.29

.10

.07

.22

.06

.08

.09

Table 16

Mean Belief Scores for Selected Attributes

 

 

 

 

 

 

Belief Pretest Posttest l Posttest 2
Mouthwash Listerine Others Listerine Others Listerine Others
Kills Germs

Control 5.113 4.755 5.130 4.759 5.0476 4.889

Experimental 4.653 4.442 4.653 4.265 4.1250 4.166
Prevents Colds

Control 4.434 3.742 4.056 3.983 4.619 4.222

Experimental 3.429 3.184 3.735 3.395 3.375 3.260
Footwear Scholl Others Scholl Others Scholl Others
Leg Exercise

Control 5.000 4.741 5.068 4.674 6.100 5.033

Experimental 5.118 4.294 5.343 4.505 5.087 4.406
Health Value

for feet

Control 5.156 4.741 5.500 4.750 6.300 5.033

Experimental 4.824 4.225 5.000 4.800 4.870 4.924
Pantyhose Hanes Others Hanes Others Hanes Others
Texture

Control 4.105 4.728 4.447 4.473 4.579 4.351

Experimental 4.065 4.642 3.387 4.484 4.667 4.778
Sex Appeal

Control 4.211 4.430 4.395 4.281 4.895 4.807

Experimental 4.968 4.903 4.645 4.656 5.111 4.407

56

Figure 2

Normative Belief Technique

Belief that Mouthwash Prevents Colds

 

 

7.0 v
6.5 «1
6.0 -
5.5 -
5.0 -

J
4'5 Listerine o
4.0 -

Norm§"""""-
3.5 4 Listerine a—i
a

3.0 ‘ Norm
2.5 4
2.0 w
1.5 w
1.0 -+
0.5 -

 

 

Pretest

Posttest 1

Posttest 2

aNorm means represent an average of beliefs in the other three brands

Experimental

57

Control

58

of the study, the mean brand belief of the group viewing the Listerine
experimental treatment increased slightly, as did the norm. In both
experimental and control cases, the belief about cold prevention power
was higher for Listerine than for the other brands of mouthwash. Thus,
using Gardner's terminology, the Listerine ad would be considered
potentially deceptive. In the second stage, the mean brand belief
scores dropped closer to the normative belief score, indicating that
the experimental group now perceived Listerine as about as likely to
prevent colds as the other brands of mouthwash studied.

The belief that Listerine kills germs may be related closely by
some to the cold prevention claim. The shift in germ-killing beliefs
is illustrated in Figure 3. The brand belief for the Listerine control
group remained higher than the normative belief in all phases of the
experiment, while the experimental group score increased during the
first stage of the experiment and dropped to about the same as the
normative belief during the second stage. This pattern follows closely
the pattern of the cold prevention belief.

The shifts in leg exercise belief scores are illustrated in Figure 4.
In all cases, the belief that Scholl exercise sandals exercise the legs
was greater than the normative belief, although the belief score for
the group which had previously seen the potentially deceptive treatment
decreased slightly in the second stage of the experiment. Gardner would
term the Scholl advertisement potentially deceptive.

The belief that brands have health value for the feet is illustrated
in Figure 5. The belief for Scholl brand of footwear was higher than
the normative belief in the first phase of the study. This could

indicate potential deception. The belief of the Scholl experimental

Figure 3
Normative Belief Technique

Belief That Mouthwash Kills Germs

— Listerine

a. b-——‘—-——

Norm ._a

— Listerine .__ “-_.
a C
Norm

 

 

 

I 1 0
Pretest Posttest l Posttest 2

a ' .
Norm means represent an average of bel1efs in the other three brands

Experimental ---------- Control

 

59

Figure 4

Normative Belief Technique

Belief that Footwear Gives Leg Exercise

 

 

 

 

.
.. /
I
”a’
A 1’
_ Scholl —- / /
0—____ ...p—II— nun-un-
_ Norma '—_ --O—-"" "I. ’-
a ;’7 __'
_ Norm
I 1 v
Pretest Posttest 1 Posttest 2

a .
Norm means represent an average of beliefs in the other three brands

Experimental ---------- Control
60

 

7.0

6.5

6.0
5.5

5.0

4.5

4.0

3.5
3.0

2.5

2.0

1.5

1.0

0.5

Figure 5

Normative Belief Technique

Belief That Footwear Has Health Value For Feet

 

 

’
’
5””
’
- I
’
o
..- —— “...-
Scholl __,_ .. -—-
- 0“"
A d
SChoél W

 

 

_ Norm ._._,_.....-—-— --
Norma /

aNorm means represent an average of beliefs in the other three brands

Experimental Control

 

 

61

62

group decreased in the second phase, to a mean slightly below that of
the norm, indicating a "correcting" effect, even though this particular
belief had not been mentioned in the corrective treatment.

Figure 6 illustrates the shifts in sex appeal belief scores that
occurred. The Hanes experimental group brand belief scores were only
slightly higher than the normative belief at the time of the first
posttest. An even greater difference existed between the Hanes experi-
mental group scores and the normative group scores after Experimental
Treatment Number 2.

The shifts in texture belief scores are illustrated in Figure 7.
The Hanes scores for the experimental group were considerably lower than
the normative belief in the first stage of the study, but increased to

about the same as the normative belief in the second stage.

V. THE AFFECT COMPONENT OF THE FISHBEIN MODEL

The Affect component of the Fishbein model was also analyzed.
Affect was measured by asking the respondents to estimate the importance
0f various attributes when making a purchasing decision. A 1 response
indicated a very low importance, while a 7 response indicated the attribute
was considered very important in the purchase decision. A response of
4 would indicate a neutral evaluation of importance. The response was
solicited at the product class level rather than at the brand level.
Mean affect scores were computed for the same attributes studied using
the normative belief approach. Table 17 contains mean affect scores for
control and experimental groups during the three stages of the study.

The experimental group perceived the germ-killing capability of

mouthwash as less important after viewing the allegedly deceptive and

Figure 6

Normative Belief Technique

Belief that Pantyhose has Sex Appeal

 

 

 

- Hanes
a "" —' a
.. Norm .___ _ _ _,_..”.. -
Hanes ______ ...a-u— c-..”-
1 I !
Pretest Posttest l Posttest 2

3Norm means represent an average of beliefs in the other three brands

Experimental ---------- Control

63

 

7.0

6.5

6.0

5.5

5.0

4.5

4.0

3.5

3.0

2.5

2.0

1.5

1.0

0.5

Hanes
a
Norm
a
Norm
Hanes
1

Figure 7

Normative Belief Technique

Belief that Pantyhose has Texture One Likes

 

 

 

 

Pretest

Posttest 1 Posttest 2

8Norm means represent an average of beliefs in the other three brands

Experimental

 

64

............ Control

Table 17

Mean Affect Scores for Selected Attributes

Affect Pretest Posttest 1 Posttest 2
Importance of ...

 

Mouthwash

Germ-Killing Power

Control 5.717 5.259 4.905
Experimental 5.551 5.286 5.219
Cold Prevention Power
Control 4.642 4.426 5.048
Experimental 4.245 4.531 3.844
Footwear
Leg Exercise
Control 5.244 5.409 5.400
Experimental 4.677 4.714 5.478
Health Value for Feet
Control 5.644 5.659 5.600
Experimental 4.970 5.000 5.478
Pantyhose
Texture
Control 5.211 4.763 5.421
Experimental 5.065 4.774 5.556
Sex Appeal
Control 5.105 4.895 5.368
Experimental 4.936 4.710 5.000

65

66

the "corrective" ads than in their original pretest evaluation. In
all cases, the scores were above the neutral response. The same group
evaluated the cold prevention power as more important after the allegedly
deceptive treatment, but the evaluation was below the neutral response
after viewing the "corrective" treatment.

For the footwear category, the importance of leg exercise
received an increasingly higher rating by the experimental group after
both experimental treatments. The importance of health value for the

feel also increased after each experimental treatment.

VI. DISCUSSION

Multiple regression was used to analyze the effectiveness of the
Fishbein models in predicting attitudes and behavioral intentions, as
measured by the questionnaire. The original Fishbein attitude model
was found to be a significant predictor of attitudes toward ll of the
12 product brands tested. These results agree with Witter (1978), who
found the model to be significant in predicting attitudes toward all
16 of the products tested. The Fishbein extended model was found to
be a significant predictor of behavioral intentions to purchase 11 of
the 12 product brands. These results also agree with Witter's study
which found the model to be a significant predictor of behavioral
intentions to purchase all products studied.

Attitude change and change in behavioral intentions were measured
by mean gain scores. Control and experimental groups were compared
by the use of t-tests. In no case was attitude or behavioral intention
found to change after exposure to either of the advertising treatments.

These results disagree with the results of Witter (1978), who found significant

PLEASE NOTE:

Page 67 is missing in numbering only.

UNIVERSTIY MICROFILMS INTERNATIONAL

68

differences in attitude as measured by the original Fishbein attitude
model for two of the four brands of footwear and three of the four

brands of pantyhose. Witter also found significant differences in be-
havioral intention as measured by the Fishbein extended model for two
brands of footwear and one brand of pantyhose. The Witter study did not,
however, find the Fishbein models to measure any change in attitude or
behavioral intention after exposure to the Listerine advertisement. Mazis
and Adkinson (1976), also did not find the original Fishbein attitude
model to measure any attitude change after exposure to the Listerine
advertisement.

The results of this study do not necessarily indicate that deception
was not present in the advertisements studied, but instead suggest that
the Fishbein models do not measure deception, as defined by Gardner and
others. These results are consistent with Wilkie's (1973) suggestion
that attitudes and behavioral intentions are unsatisfactory levels of
response at which to measure deception. He states that these measures
are "too general to be attributed to exposures to specific messages."

(p. 198) However, it should be acknowledged that the FTC has the right
to prohibit "false or deceptive acts or practices," regardless of whether
they actually change consumers' purchasing habits. (Wheeler-Lea Act of
1938)

Wilkie (1973) suggests that measurement of advertising should
occur at the brand belief level. In order to further investigate the
deceptive advertising issue, specific brand beliefs were investigated,
using Gardner's normative belief technique. Two beliefs from each of
the three product classes were selected for further investigation. Mean

scores of the brand under investigation were compared with an average of

69

the mean brand belief scores of the other three brands. This technique
identified potential deception in beliefs that Listerine mouthwash
"prevents colds" and "kills germs." These results agree with findings
of Witter (1978), who also identified deception in these two beliefs by
use of the normative belief technique. In order to confirm deception,
scientific evidence that Listerine does not, in fact, prevent colds or
sore throats, would need to be provided. As was stated earlier, evidence
was presented in the Warner-Lambert Case which denied the cold prevention
capability of Listerine. In the second stage of the experiment, in
which all subjects saw the corrective advertisement disclaimer for cold-
prevention capabilities, a "corrective" effect was noted for both the
cold prevention belief and the germ-killing belief. The beliefs that
Listerine prevents colds and kills germs more closely approximated the
beliefs for the other brands studied in the control and experimental
groups for germ-killing belief and in the experimental group for the
cold prevention belief. Three other studies (Mazis and Adkinson, 1976;
Kuehl and Dyer, 1976; Dyer and Kuehl, 1978), have noted this "reverse
halo effect" in which a corrective statement for the cold prevention
belief also affected the germ-killing belief. This may be due to a
strong association in the minds of consumers between germ-killing
power and cold prevention. It may also illustrate the inability of a
corrective advertisement to "surgically" remove one deceptive belief
without negatively affecting other beliefs about the brand.

The belief that Scholl exercise sandals provide leg exercise was
found to be much higher than the normative belief for both the control
and experimental groups during the first stage of the study. Using

Gardner's normative belief technique, this would indicate potential

70

deception. In order to prove deception, scientific evidence would

need to be provided which disproved any leg exercise benefit of the
sandals. The belief level dropped only slightly for the experimental
groups after administration of the corrective statement on erradicating
this belief. This may indicate a strongly-held belief by the respondents
from past exposure to the advertisements. The substantial difference
between belief for Scholl and the normative belief on the pretest would
lend further evidence to this explanation.

The mean brand belief that Scholl has health value for feet was
greater than the normative belief for both control and experimental
groups in the first stage of the study. This would indicate potential
deception. Scholl would need to substantiate this claim. The mean
brand belief of the Scholl experimental group fell slightly below the
normative belief during the second stage of the study. As in the

"reverse halo"

Listerine germ-killing belief, this would indicate a
effect, since the belief of health value for feet was not addressed in
the "corrective" portion of Experimental Treatment Number 2.

The belief that Hanes pantyhose has sex appeal was only slightly
greater than the normative belief for both control and experimental
groups in the first stage of the study. In fact, both the normative
belief and the Hanes belief actually dropped from the pretest scores.
No potential deception would be indicated. While the normative belief
for the group who had not seen the original experimental treatment
declined after viewing the "corrective" advertisement, the Hanes and
normative beliefs for the control group actually increased. This

would seem to indicate that the "There is no scientific evidence

which indicates that men prefer the Hanes brand of pantyhose" disclaimer

71

actually increased belief in the sex appeal of Hanes as well as other
brands of pantyhose. This was the first time a disclaimer for this "non-
functional" attribute has been used in an empirical study. It may
indicate that consumers do not trust "scientific evidence to be an
accurate measure of a non-functional attribute such as sex appeal, but
that the disclaimer actually draws attention to the attribute and may
even increase belief in the attribute. More evidence would need to be
gathered before any firm conclusions could be drawn.

While the control groups yielded approximately the same mean
brand belief in texture during the first stage of the experiment,
the Hanes belief level was actually substantially lower than the
normative belief level for the experimental group. Thus, potential
deception was indicated. The Hanes belief substantially increased after
Experimental Treatment Number 2. Since pantyhose texture was not
directly mentioned in the disclaimer, it may be that this change in
belief has no relationship to the experimental treatment. It may also
give further evidence to a "reverse-belief" effect suggested for the
sex appeal belief.

Mean affect scores were also studied, noting the change in the
subjects' perceptions of importance of selected attributes during the
three phases of the study. In general, only slight shifts in affect
were observed. This would indicate that importance of beliefs are not
greatly affected by change in beliefs.

In summary, the results of this study lend evidence to the ability
of_the Fishbein models to measure attitudes and behavioral intentions.
The results also support Wilkie's (1973) theory that deception can

best be detected at the brand belief level. No attitude change or

72

behavioral intention change was detected after either Experimental
Treatment Number 1 or Experimental Treatment Number 2. Gardner's (1976)
Normative Belief Technique showed promise as a useful approach to
detecting deception and measuring the effects of efforts to erradicate

residual deception.

CHAPTER v
SUMMARY, LIMITATIONS AND RECOMMENDATIONS
I. SUMMARY

The objectives of this study were: (1) to empirically test an
attitude change approach to detecting deception in advertising, and (2)
to measure the effectiveness of corrective advertisements on eliminating
the effects Of previous deception. The Fishbein attitude model and the
extended Fisbein behavioral intention models were used to measure con—
sumer attitude and behavioral intention change. The study used a two-
stage non-equivalent control group experimental design. The first stage
dealt with the deceptive advertising issue. The second stage dealt with
the corrective advertising issue. A pretest and two posttests used a
pretested questionnaire which operationalized the Fishbein models.

The questionnaire was administered by classroom to high school
students in three Michigan public schools. Two-hundred forty-nine
students made up the total sample.

The sample experimental and control groups were compared by the
use of t-tests for selected demographic characteristics. The groups
were not always comparable. The mouthwash groups differed in distribution
by social position, parent's occupation, and behavioral intention. The
footwear groups differed in purchasing frequency and behavioral intention
distribution. The pantyhose group differed in purchasing frequency
distribution. It is likely these differences were present because the
groups were composed by classroom rather than by true random assignment.

Twelve brands in three product classes were investigated. Actual

73

74

advertisements for Listerine, Scope, Listermint and Lavoris mouthwash,
Scholl, Fanfare, Famolare, and Earth Shoes footwear, and Hanes, L'eggs,
No Nonsense, and Sheer Energy pantyhose were used. 'Dnee potentially
deceptive statements were contained in the first experimental treatment
and duee "corrective" statements in the experimental treatment of the
final stage.

Multiple regression was used to analyze the effectiveness of
the Fishbein models in measuring attitudes and behavioral intentions,
as measured by the questionnaire. The original Fishbein attitude model
was found to be a significant predictor of attitudes toward ll of the
12 product brands tested. The Fishbein extended model was found to be
a significant predictor of behavioral intentions to purchase 11 of the
12 product brands.

Attitude change and change in behavioral intentions were measured
by mean gain scores. Control and experimental groups were compared by
the use of t-tests. In no case was attitude or behavioral intention
found to change after exposure to either advertising treatments. It was
concluded that the results of this study do not necessarily indicate
that deception was not present in the advertisements studied, but
instead suggest that the Fishbein models do not measure deception, as
defined by Gardner (1976) and others.

In order to further investigate the deceptive advertising issue,
specific brand beliefs were investigated, using Gardner's (1976)
Normative Belief Technique. Two beliefs from each of the three product
classes were selected for further investigation. Mean scores of the
brand under investigation were compared with an average of the mean

brand belief scores of the other three brands. This technique identified

75

deception in the beliefs that Listerine mouthwash "prevents colds"

and "kills germs," the beliefs that Scholl exercise sandals provide

leg exercise and have health value for feet. A "corrective" effect was

noted after administration of the corrective experimental treatment for

the beliefs that Listerine kills germs and prevents colds and the belief
that Scholl exercise sandals have health value for feet.

Mean affect scores were also studied, noting the change in the
subjects' perceptions of importance of selected attributes during the
three phases of the study. In general, only slight shifts in affect were
observed. This would indicate that importance of evaluation criteria are

not greatly affected by change in beliefs.

II. LIMITATIONS AND RECOMMENDATIONS FOR FURTHER STUDY

The study was limited by the use of a focus group and a pretest
sample with characteristics different from the study sample group. The
former groups consisted of college freshmen majoring in clothing and
textiles at University of Tennessee, Knoxville, while the sample consisted
of Michigan high school students. The evaluative criteria of the two
groups may differ greatly, especially since the consumer knowledge of the
clothing and textile majors about the two clothing product groups used
in this study may be much more extensive than that of the high school
students. For example, the researcher noted that some high school
students in the sample complained that they "never wear" pantyhose or
use mouthwash and were unfamiliar with some of the attributes on the
questionnaire. This may point to a need to develop evaluative criteria
with a very specific group in mind.

The design of this study could have been improved by the use of

76
random assignment of the sample. Comparison of the control and
experimental groups detected differences in distribution by social
position, parents' occupation, purchasing frequency and behavioral
intention. Comparison of pretest belief means for attributes studied
also indicated substantial differences in experimental and control groups.
A true random assignment design would have minimized these differences
and yielded more accurate results.

Uneven group size, another major limitation of this study, was
caused partially by the use of fixed classroom groups, absenteeism and
subjects' refusal to complete the questionnaires. It was informally
noted that refusal to complete the questionnaires was more evident on
the second posttest; answering essentially the same quesitons three
weeks in a row was tedious to some students. It may be that use of
some inducement, such as payment for completion of the questionnaires,
would have eliminated this problem.

This study assumes that the respondents actually read the advertise-
ments and that changes in attitude were caused by the advertisements. In
reality, there was no way to guarantee that the students actually read
the ads they were given. Many students paid only brief attention
to the ads before completing the questionnaires. This issue must be
considered in future measurements of deception.

The experimental design, while allowing for control of many vari-
ables, may have certain inherent limitations. Conditions used for viewing
the advertisements were not realistic. Single exposure to an advertisement
may not have an effect on deeply ingrained attitudes. Allowing the sub-
jects to read through magazines containing typical articles as well as
the ads may provide more natural viewing conditions.

The study was limited in the number of products studied. Further

LIST OF REFERENCES

LIST OF REFERENCES

Ahtola, O. T. "The Vector Model of Preferences: An Alternative to the
Fishbein Model," Journal of Marketinngesearch, Vol. 12, February
1975, 52-9.

 

Ajzen, I. and Fishbin, M. "Attitudinal and Normal Variables as Predictors
of Specific Behaviors," Journal of Personality and Social
Psychology, Vol. 27, July 1973, 41-57.

 

 

Armstrong, G. M., Gurol, M. N., and Russ, F. A. "Detecting and Correcting
Deceptive Advertising," Journal of Consumer Research, Vol. 6,
December 1979, 237-246.

 

Armstrong, G. M., Kendall, C. L., and Russ, F. A. "Applications of
Consumer Information Processing Research to Public Policy Issues,"
Communications Research, Vol. 2, No. 3, July 1975, 232—245.

 

Armstrong, G. M. and Russ, F. A. "Detecting Deception in Advertising,"
MSU Business Topics, Spring, 1975, 21-31.

 

Bass, F. M. and Wilkie, W. L. "A Comparative Analysis of Attitudinal
Predictions of Brand Preference," Journal of Marketing Research,
Vol. 10, August 1973, 262-9.

 

Bernacchi, M. D., "Advertising and Its Discretionary Control by the
FTC: A Need for Empirically Based Criteria," Journal of Urban
Law, Vol. 2, 1974, 223-66.

 

Brandt, M. T. and Preston, I. L. "The Federal Trade Commission's Use
of Evidence to Determine Deception," Journal of Marketing, Vol. 41,
January 1977, 54-62.

 

Campbell, D. T. and Stanley, J. C. Experimental and Quasi-Experimental
Designs for Research. Chicago: Rand McNally College Publishing
Company, 1963.

 

 

Dillon, T. "What is Deceptive Advertising?," Journal of Advertising
Research, Vol. 13, October 1973, 9-12.

 

Dulany, D. E. "Hypotheses and Habits in Verbal 'Operant Conditioning,
Journal of Abnormal and Social Psycholggy, Vol. 63, 1961, 251-263.

 

Dyer, R. F. and Kuehl, P. G. "The 'Corrective Advertising' Remedy of
the FTC: An Experimental Evaluation," Journal of Marketing,
Vol. 38, 1974, 48-54.

 

78

79

Dyer, R. F. and Kuehl, F. G. "A Longitudinal Study of Corrective Adver-
tising," Journal of MarketinggResearch, Vol. 15, February, 1978, 3948.

Fishbein, M. "Attitude and Prediction of Behavior," in M. Fishbein, ed.,
Readings in Attitude Theory and Measurement. New York: John
Wiley and Sons, 1967, 477-492.

Fishbein, M. "Attitude, Attitude Change, and Behavior: A Theoretical
Overview," in Philip, Levine, Ogilvy and Mather, ads. Attitude
Research Bridges the Atlantic. Chicago: American Marketing
Association, 1975, 3-16.

Ford, C. T., Kuehl, P. G., and Reksten, 0. "Classifying and Measuring
Deceptive Advertising: An Experimental Approach," in E. M. Mazze,
ed., American Marketing Association 1975 Combined Proceedings,
1975, 493-497.

Gardner, D. M. "Deception in Advertising: A Receiver Oriented Approach
to Understanding," Journal of Advertising, Vol. 5, No. 19, 1976, 5-11.

Gardner, D. M. "Deception in Advertising: A Conceptual Approach,"
Journal of Marketing, 39, January 1975, 40-46.

Gellhorn, E. "Proof of Consumer Deception Before the Federal Trade
Commission." Kansas Law Review, Vol. 17, 1969, 559.

Green, P. E. "Uncertainty, Information, and Marketing Decisions," in
Cox, R., Alderson, W., and Shapiro, S. J., eds., Theory in
Marketing. Homewood, Illinois: Richard D. Irwin, Inc., 1964, 31%54.

Gurol, M. N. "Deception in Advertising: A Review of Past and Current
FTC Practices and an Experimental Evaluation of a New Approach for
Detecting Deception and a New Approach for Eradicating its Effects."
Unpublished PhD. Dissertation. University of North Carolina at
Chapel Hill, 1977.

Haefner, J. E. "The Perception of Deception in Television Advertising,
an Exploratory Investigation." Unpublished PhD. Dissertation.
University of Minnesota, 1972.

Haefner, J. E. and Permut, S. E. "An approach to the Evaluation of
Deception in Television Advertising," Journal of Advertising,
V01. 3, NO. 4’ 1974’ 40-45.

 

Holbrook, M. B. "Attitude Structure and Communication Effects: The
Neglected Interface," working paper, Columbia University, 1976.

Holbrook, M. B. Beyond Attitude Structure: Toward the Informational
Determinants of Attitude." Journal of Marketinngesearch, Vol. 15,
November 1978, 545-56.

Holbrook, M. B. "Comparing Multiattribute Attitude Models by Optimal

Scaling," Journal of Consumer Research, Vol. 4, December 1977,
165-71.

80

Holbrook, M. B. and Hulbert, J. M. "Multi-attribute Attitude Mbdels:
A Comparative Analysis," Proceedings, Association for Consumer
Research, Fifth Annual Conference, 1975, 375-88.

 

Hollingshead, A. Two Factor Index of Social Position. New Haven,
Connecticut: Yale Station, 1965.

Hunt, H. K. "Effects of Corrective Advertising," Journal of Advertising
Research, Vol. 13, No. 5, October, 1973, 15-22.

 

Jones, M. G. "The FTC's Need for Social Science Research," address before
the Second Annual Conference of the Association for Consumer
Research, University of Maryland, College Park, Maryland,

September 1, 1971, 3-4.

Kassarjian, H. H., Carlson, C. J., and Rosin, P. B., "A Corrective
Advertising Study," in M. J. Schlinger, ed., Advances in Consumer

Research, Vol. 2, Chicago: Association for Consumer Research,
631-42, 1975.

 

King, A. L. and Wise, G. L. "A Cross-Tabular Ranking of Advertisements
Based on Extent of Deceptiveness," in B. A. Greenberg, ed.,

Proceedings: Southern Marketing Association 1974 Conference,
1974.

Kuehl, P. G. and Dyer, R. F. "Brand Belief Measures in Deceptive-
Corrective Advertising: An Experimental Assessment," in
Marketing: 1776-1976 and Beyond. 1976 Educators' Proceedings,
Bernhardt, K., ed. Chicago: American Marketing Association,
1976, 373-379.

 

Kuehl, P. G. and Dyer, R. F. "Applications of the'Normative Belief'
Technique for Measuring the Effectiveness of Deceptive and
Corrective Advertisements" in Perrault, W., Jr., ed., Advances
in Consumer Research, 1977, 204-212.

 

LaRue, P. H. "FTC Expertise: A Legend Examined." The Antitrust Bulletin.
Vol. 16, Spring 1971, l-31.

 

Mann, R. A. and Gurol, M. "An Objective Approach to Detecting and
Correcting Deceptive Advertising." Notre Dame Lawyer, Vol. 54,
October 1978, 73-101.

 

Mazis, M. B. and Adkinson, J. E. "An Experimental Evaluation of a
Proposed Corrective Advertising Remedy," Journal of Marketing
Research, Vol. 13, 1976, 178-83.

 

Mazis, M. B., Ahtola, O. T., and Klippel, R. E. "A Comparison of Four
Multi-Attribute Models in the Prediction of Consumer Attitudes."
Journal of Consumer Research, Vol. 2, June 1975, 38-52.

 

Pelgro, J. J. "The Authority of the Federal Trade Commission to Order
Corrective Advertising," Boston College Law Review, Vol. 19,
July 1978, 899-938.

 

81

Pollay, "Deceptive Advertising and Consumer Behavior: A Case for
Legislative and Judicial Reform," Kansas Law Review, Vol. 17,
1969, 625.

Preston, I. L. and Johnson, R. H. "Puffery--A Problem the FTC Didn't
Want (And May Try to Eliminate)," Journalism Quarterly, Autumn
1972, 558-68.

Rotfeld, H. J. and Reid, L. N. "Advertiser Supplied Message Research:
Extending the Advertising Substantiation Program," The Journal
of Consumer Affairs, Vol. 11, Summer 1977, 128-134.

 

Ryan, M. J. and Bonfield, E. H. "The Fishbein Extended Model and Consumer
Behavior," Journal of Consumer Research, Vol. 2, September 1975,
118-36.

 

Sawyer, A. G. "Measuring Communication Effectiveness of Deceptive and
Corrective Advertisingf'flnY. Wind, and M. G. Greenberg, Moving
A Head with Attitude Research. Chicago: American Marketing
Association, 1977, 190-94.

Weaver, Warren. "The Mathematics of Communication." Scientific American,
Vol. 181, 1949, 11-15.

Wilkie, W. L. "Research on Counter and Corrective Advertising," in
S. F. Divita, ed., Advertising and the Public Interest: Selected
Papers from the Conference on Advertising and the Public Interest
Held in Washington, D. C. May 1973, sponsored by the American
Marketing Association, 1973, 189-202.

 

Wilkie, W. L. and Gardner, D. M. "The Role of Marketing Research in
Public Policy Decision Making," Journal of Marketing, Vol. 38,
1974, 38-47.

Wilkie, W. L. and Pessemier, E. A. "Issues in Marketing's Use of Multi-
Attribute Models," Journal of Marketing Research, Vol. 10,
November 1973, 428-41.

Witter, B. S. "Deceptive Advertising and Advertising Substantiation:
An Attitude Change Approach." Unpublished PhD. Dissertation,
University of Tennessee, 1978.

Wright, P. L. "Consumer Choice Strategies: Simplifying vs. Optimizing,"
Journal of Marketing Research, Vol. 12, February 1975, 60—7.

APPENDIX A

STRUCTURE OF FOCUS GROUP

What characteristics do you look for when you buy ?
(the product class)

 

How would you rate these characteristics in their order of
importance to you as a consumer?

What brands of (product class) are you familiar
with as a consumer?

 

Look again at the characteristics you've listed as important.
Are any of these characteristics the same for different brands?
Do you think that (evaluative criteria) is
different for various brands?

(Continue with each listed evaluative criteria.)

 

82

APPENDIX B

83

uw>mz
ummh m muse cmcu haucmzcmum mmmd

umm% m mono

A.mmCOdmmu mach mauufiov

mummh m can“ mmmd

mums» mum
mumm% HHTOH
Hoonom swan scum nmumnvmuo

A.mm:oammu wash mauuwov masocom

.o

.m

.q

.m

umm> m mmafiu Hmum>mm .m
358: .N
cocoa m mmewu Hmum>wm .H

wzmmznusoe ommzousm zaamumamm no» on amumo Bo:

Hoonom Hmcowumoo>
no moonwmnn gov wmeHoo meow .m

mwuwmv meHHoo .N

Hoosom mumavmuu .H

aw ow m>wa 50% sown: aw maosmmsoz mnu mo mmm: mfiu mam Maw 30:

.m

.o

 

A.owmwoomm mum> ma Ommmfimv wm>HH Dom nowza aw naocmmsos one we was: w£u mo cowumaaooo ecu ma umzz .m

A.>Hao mmmoawsa wcwmoo HOW mw uw .xcman wwsu aw Hawm no: onv

.Hmwucmvwmcoo hauowuum ma Haw3 mSOHuwwnv one Ou mmwcomwmu use»

 

 

 

mHmEom mam:

3nd

.hmnum menu cw upon wawxmu wow 30% xcmge

wmnasz xuwuaomm Hmauom no mswz

HmMHmmmlllmm<3mHDOE

mesa 30% cu mw megawapwuomumno mmmcu m0 some unmuHOQEw 30: mumOfivafl mmmmam
050m new 30% 00 ucmuuoaefi %um> on uswwe mowumwumuomumso mmmnu m0 080m

mamz mcmum umH30fiuumm

mmfioo mucm>mum

mshwo mHHHM

muomo mum>oo

moaum 304

HHmEm manusumz

OumMH ummam

84

.moampmmmua ~30% 00 ummmoao uwna33 ecu Odouwo mmmmam

o

o

o

m

m

m

q

q

a

3

m

m

m

m

N

N

N

N

L

o

o

o

o

.ucmuu0a8« %um> on 003 uswfia

smmazu3oa_wcwmmsuu3m am£3IIIHHH Hm<m

msmz mcmum uma30auumm oz
mvHOU ucm>mum uoz moon
warms 333M uoz mmon
muomo u0>oo uoz moon
moans swam

HHmEm uOOBm

mumme vauammfiun<

.ma

.mH

.oH

.ma

.qa

.MH

.NH

.0 m OHOHHO Ofiuammwucm %um> mummu ou 5mm3£u3oa

u30% OxHH 30% «H .m nonesa m OHOMHO va3oz 30% ammsumalaw uswfiu u3n .ummam no: OHuawmflucm “nausea mmuwmu

sufifis me3£u3oe m OxHH 30% we Nae :0H0m033 as mfiaamxm pom

.OOfioso u30% onwuommm ou wanna wcwsoaaom 0:0

wo 080w mm3 unwfla 30% Smm35u3oe m cw causes 30% wmauwam3v m0 mmd%u owns %Huumxm Odomaom Hamu 0a whoa 30%
MH .mcwamma 3H muwmomao mum :0fi53 mauve m0 awed m 00 wwumamu m30fiumm3v wafisoaaow ecu m0 nomMIIIHH Hm¢m

\DOOO

 

mz<mm mHmH wbm AAH3 H

mmmm

.mamun 0:0
OHOHHO .mcmun umnu %3n 00: Haw3

m>wufim0m mum 30% we .maaemxm pom

%meHH 30: Opossumm 30% va3o3

<f<f<f¢

mmmm

A<MBDMZ

NNNN

HF—i—dfi

30% umzu 0>aufimoa mum 30% HH

COCO

unweuwuqu
mwu0>mq
mawuwumfiq
mmoum

 

Qz<mm mHmH Nam Hoz AAH3 H

u3onm meHaomw u30% mandamus On e on o mumnafia 050 m0 %:m OHOHHU
.e m mauuwo .mcwun mfiu %3n Haw3 30% uMfiu
mamm3nu3oa m0 mmamwn mafiaoaflom msu wmmsuu3n ma303 30% umsu on UH303 uw
.me3nu308 m0 wucmun u30w Ou sowuommu u30% 3H mwumwumuufi Em HIIIH Hm<m

.HH
.OH
.3
.m

.o m

H

H.30Hmn mcowumm3c 050 now mmmcoammu u30% wHouHO wmmmHmv .OxHH
30% noHns mummy m0 ma%u ecu O>ms uos meow ucHBMOOMHH mega man Hmuu3w: mum mHu0>mH u3onm wwaHHmmw
u30% umnu .OHHH 30% HOHHB mumme m0 03%3 mxu mvfi>0um %HHmanow waHuoumHH mam maoom umsu Hmmw 30% HH

 

 

o m e m N H 50\ unHauOumHH
o m s «m N H o 3823
0 ..mx e m N H o mowumumfiq
o ..wx c m N H o mmoom
m4m<momm meEMMme ; MHm<mommzH MHmzmmem
“MHm24xm

"mwsu OHHH economy ucwwa 30%

%u0wmumo mmmmm.mcu pom33 mHaamxm pom .waHHOOH one 30% Hafina you MOHuMHumuumumso m0 Oa%u ego 30% O>Hw
mH303 uGHEuOumHH new msuo>wH .mawwmumHH .waoom umnu on mH303 0H mHanOHQ 30: wum3Hm>m 30% CH303 .nmms
n£u3oe :H now wconoH mum 30% 50H£3 MOHumHumuomumso m0 mmd%u msu wwH>0um %ms momma: ucmummealul>H Hm<m

 

 

% o m s m N 3 o mamz 33m .2
o n 3 m N H o GONucw>mHm mHoo .qN
o m e m N H o meson mcHHHHM sumo .mm
o n 3 m N H o Houucoo 9060 .NN
o m 3 m N H O . mowum .HN
o n 3 m N H o HHmEm .ON
0 n d m N H O mummh .mH
BZ<BMO®ZH wmm> HZ<HMOQZHZD me>

 

.m wHouHo HHHa 30% oomDDOMEH %uo> no: unmuuomaHa3 %wm> umsuwma mH mummu um:u Hwom 30%
HH .0 OHOHHO HHHB 30% ucmupdmch3 %um> mH momma umnu xcfinu 30% NH OHaamxw mom .nwmzzu3oa wafiuflflmm

 

86

@000
WWW“)
\‘TQQQ‘
NNNN
I—Iv—lF-lv—l
COCO

 

NOHHH 30% pmzom wCHHHHM Show m0 vawx Ono m>ms vH303 mvcmpn mmmnu H0

@000
mmmm
<T<f<f<f
NNNN
v—4o—cc—4r—I
COCO

 

NOHHH 30% Houuaoo homo m0 mcax mzu w>mc mH303 mvamun mmmcu H0

@000
mmmm
<r<r<r<r
NNNN
H—a—I—d
OOOO

meHH 30% OOHHm mo mcwx mnu m>m£ EH30: mmcmun mmmfiu m0

@000
mmmm
\‘f\'f\'f\'l'
NNNN
o—lv—iv—Ov—d
OOOO

NoxHH 30% HHOEm m0 mcax OHu m>m£ cH3oB mmcmun mmm£u m0

@000
mmmm
<T<T<T<T
mmmm
NNNN
v—Iv—dv—I—a
COCO

 

 

MHm<momm WHmzmmexm MHm<mOMmZH mezmmem

NoxHH 30% mumme mo mcHx 0:0 m>m£ mH303 mmamun mmmsu m0

unwaumumfiq
mHu0>mH
maHpmumHH
maoow

some Own“ an EH HHOEHH

unwaumumfiq
mHu0>mH
OcaumumHH
maoom

Eumm Oman on ma samxHH

unwauOuMHH
mau0>mq
msHumumHH
Odoom

50mm Oman DH ma samxaa

uaHsuOumHH
mHu0>wH
OcfiuODMHH
Odoum

comm Esau “3 m3 samxna
ucHeumumHH
mHu0>MH

ocwumumHH
maoum

comm Oman an EA samxaa

.me
.33
.m3
.N3

30:

.H3
.03
.mm

30:

.Nm
.mm
.mm

30:

.mm
.Nm
.Hm
.om

30m

.aN
.wN
.NN
.oN

3cm

87

o m c m N H o .wm
H<mo H<mmo < mm<o HH< H4 mm<u H.209

 

.e m wHouwo UH303 30% cmzu .mHaomn mmmsu m0 00w>0m
Osu exam 0u mmmmmmmmw mH NH xcHnu 30% HH .0 OHOHHO oH303 30% .0uw .MNOuoom .mucmuma .mmcmHuw n30% mo
00H>vm may mme Ou ucmupomEH33 MH 0H umnu xcHnu 30% MN .OHasmxm pom chmEEOOOu cH303 30% Ou ucwuuomeH
mum 0:3 OHQOOQ umsuo Hons 0o umHHEHm mH smmsnu3oe m0 moHono u30% NH sumo 30% on .Hmuwcmw GHIIIH> Hadm

 

 

 

o m e m N H o uaHaumuMHH .mm

o m e m N H o mHu0>mH .om

o m q n N H o mcHumuwHH .mm

o n e m N H o Odoum .qm
mHm<momm MHmzmmexm MHm<mommzH mezmmexm

Namm35u3oa m0 mucosa mwmnu %3H 30% umnu
vcmeeoomu cH303 .0um MHOHOOU .muamuma .mcaoHuw mm £03m 30% OH unmuuoaaw mum 0:3 OHaoma umnuo umnu ow
uH HcHnu 30% on %HOHHH 30: .00H>vm 30% 0>Hw uste OHQOOQ umauo .GOHMHOOC mmmcuu3m m wcwxma 30:3III> em<m

unwanmumHH .mm
mHu0>mH .Nm
ocHumumHH .Hm
wmoom .om

\D\O\D\D
mmmm
QQQ’Q
mmmm
NNNN
v—iu—iHv—i
OOOO

 

NoxHH 30% memz mcmum m0 mm%e ecu on cH303 mvcmun mmmsu m0 Home umcu UH mH %meHH 30:

 

 

o m .H m N H o 358833 .3

e m .H m N H o 2.823 .3

o m s m N H o 2.2323 .3

o m .H m N H o 38m .3
H.583: szmmsxm mHHH<HH9HHzH szmmexm

 

NOHHH 30% umSom 30Hucm>mum mHoo H0 mcHx ecu m>m£ mH303 mmcmua mmmnu m0 Home uMfiu ufi mw %H0&HH 30m

88

 

 

o m c m N H o udwfihmumHH .No

0 m 3 m N H o mfiuo>mH .H@

c m 3 m N H o mcfiumumHH .00

m m c m N H o maoum .mm
m>HHHmOm %Hmzmmexm H<mebmz m>HH<omz wHMZMMme

 

Nnmm35u3oe mo mmcmun u30w mmmnu 0x08 Howcz mchmmaoo mnu mum3Hw>0

 

30% vH303 3o: .me3su3oe ecu mmxma £0H£3 %3m3500 OHu mo 00Hu03Hm>m 3m mm3H03H %Hnmn0um 30% nonHown
mmm£0u3a m Oxme 30% 00:3 u3n .5m035u3oe H0 mucosa N30H u3onm mcowumm33 moxmm coon w>m£ 30%IIIHH>

 

o
o
o
o
H

Qz<mm me9 %3m HHHB

mumnE3: mau H0 %:m OHOHHU

on mH303 uH %H0HHH 30: mumEHumm 30% mH303

89

uaHauoumHH
mHu0>mH
maHumumHH

maoum

 

s m N H o

s m N H .o

s m N H o

s m N H o
3<mssmz

nz<Mm mHmH %Dm Roz HHH3 H

.vcmun mau u3onm wwaHHOOH u30% mumOHvaH Ou o 0u o

.o w OHOHHO .mcmun mnu %3p uoc HHH3 30% umnu O>HuHmom mum 30% HH
ecu %3n HHHB 30% umsu O>HuHm0a mum 30% HH .mHamem pom

Cmauo .m
xumHm .N

fidHWNUfiNU .H

.o m wHouHu .mamun
Nnmmszu3oa H0 mwcmun w3H30HH0H may monsou3a mH3o3 30% umnu

.Smm3fiu305 HO wUCmHD H30“ Ou fiOHUUme “30% GH fimummhwufiH EM H IIH Hm<m

mmomu p30% MH um53 .N
mHmamm .N
me2 .H

Nxmm u30% mH umfiz .0
NH .q
mH wm>o .N 0H .m
mH .o mH .N
wH .m mH nova: .H

wam u30% mH ans: .m

 

.uwwcm u0u3maoo u30%.30 mumamcm waHsoHHow m£u vaoomu mmmmHm

A.%H00 memoau33 NCHUOO pom mH uH .xcmHn mHHu cH HHHH uoc 00v

AQIHV
nona3z %uHu3omm HmHoom no mamz

 

HmMH Hmomllmm<3mHDOZ

o
o
o
o
o
c
c

 

HZ<HMOHZH %mm>

%um> nonuna wH mummu umnu HOOH 30% HH

.5m03fiu3oa wcHuomHom 30:3 30% 00 MH moHuMHumuumumzo omega H0 some unnuuoasw 30: oumoncH mmmmHm

no
0,
new: mumps umH3oHupmm

mmHou mu00>0pm

mahmw MHHHM

muovo mu0>0o

mowua 30H

HHOEm OSHOHQOE

mummH ummzm

.o m OHOHHO ofiuammHuam %um> mummH 00 smm350308 u30% OHHH 30% HH
.uOOSm ~03 oHudmmHunm HOHHHO: mmummu HOHHB smm3£u3oe mxHH 30% HH NH :0H0m030 3H mHaemxm pom

\quqN-rqq

H F‘ H H H H H

ooooOOO

 

HZ<HMOHZHZD wmm>

mam: HEM um
30H030>0ua mHoo
uoaoa waHHHHH sumo
Houunoo homo

OOHNH

HHoam

mume

.mN
.QN
.NN
.NN
.HN
.ON
.mH

 

.m OHOHHO HHHB 30% unnuwomeH %Hm> soc ucmuu0m8H33

.o OHOHHO HHHa 30% ucmuuoaaHa3 %u0> mH mummu umnu Hawnu 30% HH mHmamxm pom

0000000

mmmmmmm

qqqqqqq

mmmmmmm

NNNNNNN
Hv-dv—Ou—do—lp—d—q
OOOOOO

o

0503 manna N0H30Huumm oz
mvHoo ucm>wua 00: moon
wapmw HHHH 000 moon
muoco u0>00 no: wooa
OOHHQ :me

HHOEm uoo3m

mummu OHuammHuc<

.uamuuoaaH %um> on
003 unwwe 050m was 30% 00 ucmuHOQEH %H0> on unwwa MOHHMHumuumumcu omenu H0 080m nmmsfiu3oe wcwmmnou3a GOHBIIHHH Hm<m

.mH
.NH
.oH
.mH
.qH
.MH
.NH

.wocmumwmua H30% 00 ummmoHo non533 ecu OHONHO mwmmHa
.m ~0Ha33 m OHOHHO mH303_30% 0003umnlcw unwuu 03H
.OOHOHO u30% mnHuommv

ou meumu wcHaoHHOH 050 H0 0503 003 ustE 30% :mm350305 m 0H cwucm3 30% mmHuHHm3v H0 m00%u umna %Huomx0 mcooaom HHmu

0H 0u03 30% HH

.wcHsmOE :H muwmoaao mum HOHH3 msumu H0 uHma m 00 mmumeu wGOHum03U waHSOHHom ecu H0 nummllHH Hm<m

 

.N .a mm<3ms=oz

o m
e m
o m
o m

q
q
c
c

wmxfifi :o»

m
m
m

@000

n

e
c
q
q

wmxHH so»

\0 \o \O \D
In In In In

91

uno> mHouHo mmmmHmv

\TQQ‘

q

m
m
m

m

NNN

N

H
H
H
H

o unfiaumumfig .mm
o mfiuo>mq .om
o mafiumumfiq .mm
o maoom .qm

monm mo uaflx ma“ m>mn vases mucmun mmmgu mo 30mm umau u“ ma hamxfifi 30m

m
m
m
m

NNN

N

v—dr-QHF"

ucfiaumumfig .mm
mfiuo>mH .Nm
mafiumumfig .Hm
o macaw .om

oOO

HHmzm mo vcHx mzu m>mn stoz mvcmun mmmsu Ho numw umnu uw ma hamxfia 30:

m
m
m

m

NNN

N

HHv—lv-l

uaHaumumflq .am
mHuo>MH .wN
mcwumumfia .NN

c) C) C)

o macaw .oN

meflH 50» mym<e mo ccflx mg“ m>mn oases mummy; mmmcu mo comm umnu “H m“ hflmxfia 3cm

A.30Hmn mcowumwsv mnu how mmmconmwu

.mxHH 30> cowna mpm<e Ho mazu wzu m>mn uoc mmow unwahwumHH umSu cam Hmuuswc mum wHuo>mH

uaoam wwcHHmmw pack umsu .mxHH 30% £0H£3 mem<e Ho mama may mvH>oua mHHmuwcmw mcfiumumHH cam maoum umau mew :0» HH

0 m
o m
o H,m
o m

/

 

m4m<momm mezmmem

umwzu mxflH caoammu uanE so» huowmumu m9m<e wsu “was: memem uom

mo waxy mnu Dom m>Hw cHao3 ucHEpmumHH wcw mwuo>mH .mcHuwumHH .maoom umnu ma vHao3 uH mHnmnoua Bo: mumaHm>m Dom
.nmmszusoa CH pom mewxooa mum 30h nows3 mowumHumuompmSU mo mmmmu msu mvH>oua mmE mvamun ucmumHMHa II>H Hm<m

UHDOB

\‘TQQQ

m

Q

mm

N

N
N
N

r—‘o—‘Ho—i

ucwsumumwa
mwuo>mH

mcHumuwHH

mmoum

 

mHm¢mommZH quzmmem
"mum24xm

.wafixOOH mum 30% nuwc3 pom mUHumHumuomumno

 

.m .m,mm<3mepoz

92

 

 

 

 

w w c m N H o acwaamawHa .mw

w w c m N H o wHac>mH .Nw

c w c m N H o mcHamawHa .Hw

c w c m N H o macaw .ow

meHH cca wz<z az<am ac mawe mac mp chca wccmac mwmna ac namm amca aH wH aHmHHH 3cm

c w c m N H o acHaamawHH .wc

w w c m N H o mHac>ma .wc

c w c m N H o chamawHa .Nc

c w c m N H o macaw .cc

meHH cca amzoa onezm>waa cha No ccHH mna mcms chca wcccac mwmca wc namm amca aH ma ameHH 3cm
c w c m N H o acHaamaqu .wc

c w c m N H o wHac>mH .cc

c w c N N H o mcHamawHH .mc

c w c m N H o macaw .Nc

meHH cca amzoa azHHHHx zawa Hc ccHH mac mcmn chcz wccmac mwmca wc nacm amca aH wN ameHH 3cm
c w a m N H o acweamaaHH .Hc

c w a m N H o wHac>mH .oc

w w c m N H o mcHamawHH .am

c w c m N H o macaw .ww

NmHHH cca aoaazoa acao wc ccHx mac acm; chcs accmac mwmna wc namm amna aw ma ameHH ac:

 

 

mHm<momm wqmzmmfixm MHm<mommzH wqmzmmexm

 

. c é HHw<3mSoz

o n q n N H o u:HEhwumHH .No

 

 

w w c m N H c wHaccmH .Hw

c w c m N H o mcHamawHH .oc

w w c m N H o macaw .aw
w>Hanoa wamzmaaxw H<aeawz m>He<awz wwwzmaaxw

 

szmacusoe mo mvcmup know mmmsu mxma nuwnz
mmwsmmaou mzu mumDHm>m 20% szo3 30m .nmmznuaoa may mmxma gowns hammaou wzu mo :oHuMSHm>m cm musHoafi mHnmnoua
so» comeomw mmmzouaa m mxma so» cmsa wan .nmmsnusoe Ho mvamun “new usonm maowummsv cwxwm ammo m>ms nowllHH> Hm<m

 

o m c m N H o .wm
H<mn H<mmu < mm<o 44¢ H< mm<o H.209

 

 

3
9
.m mHuuHo vHaoa

sou cmnu .mHaowa mwmzu mo moH>cm mnu mxmu Cu ucmuuommw mH uH xcHsu 30% HH .0 mHouHu vHaozcaoa ..uuw .mHOuuov
.wuamuma .mwamwuw usom mo moH>vm mnu mxmu cu ucmuuomEHas mH uH umnu Mafinu no» «H .waaamxm pom chmanomu szos so»
Cu unmuuoasfl mum 0:3 mHaomQ umsuo umna cu umHHEHm mH smmazuaoa mo mUHono Mao» ma mpmu 30% cu .Hmumcmm :HIIMMIMMflM

 

 

 

o m c m N H o unwahmumwg .mm

o m w m N H o mwuo>MH .mm

o m c m N H o wcwumuqu .mm

o m q m N H o macaw .qm
wwm<momm wamzmmaxm MHm<mommZH meEMMme

Nsmmssusoe we mvcwun
mmmnu xan so» umnu vamssouwu uHaoa ..uuw .mu0uuov .muamuma .mucwHum mm Loam no» Ou uamuuoaaH mum 0:3 mHaomm HwSuo
umcu mH uH xcwzu so» ow hmeHH Bo: .mUH>cm 30% m>Hm unwwe mauomn pmcuo .conHomv mmmcuuzm m wameE cm53||> Hm<m

 

.w .a mw<3mysoz

94

o

 

mmmu< MHUZOMHm

m

a

.maoom
cmzu umuuma cw>w mxuo3 ucwshmume .oN

.wamehn Hmuo voow you
zmv>pm>m nmm: mp vHsonm ucHEumumHH .oo

.meuH>mo mcwucm>mua aw
Hmwuwwmcmn mw mHuo>mH zuwa wcHkumu .wc

.mwuo>ma mm: Cu ucmuuoqsfi
mH uH umcu m>mHHmn mumHucmn .Nc

.muaoc pow meumw mHHHx mcHumumfiH .oo

.mUHou wcHucm>wum
:H mHou “ohms m xmaa emu mcaumumHH .mo

.uH wGHHHmEm
up mahww mHHHx macaw Hamu :mu :0» .eo

.muwuawmwuam mm
cmmHo mm gummun mw>mmH mmoum .mo

 

mmmu<mHQ »HUZOMHm

Nmucmawumuw wsHSOHHow

mzu Luw3 wmummmwv no mwuwm so» ow .wuHmccoHuwmsc mHnu Cu vwzumuum muamEmmHuum>vm wcu wawvmmu umum¢llHHH> Hm<m

 

.c .m :wSHHHSc:

APPENDIX C

DECEPTIVE ADVERTISING: A CONSUMER RESEARCH APPROACH

CONSENT FORM

 

Advertisements have been criticized as being highly misleading.

This project is concerned with how students use advertising
information. The Federal Trade Commission regulates advertising
practices, but little or no input from consumers is included in
their decisions. This study proposes to assess advertisements for
misleading claims based on how the consumer comprehends and believes
the advertisement. As part of the study, the students will be
informed about the claims the producers have made which were found
by the Federal Trade Commission to be misleading.

Very little risk is involved with participation in this study.

Some students may change their beliefs about the product because of
the advertisements, but every attempt will be made to leave the
student with an informed view of the actual product attributes.

The advertisements which will be used are taken from magazines
frequently read by the students and do not carry any immoral or
subliminal messages.

Any questions regarding procedures involved or results obtained

will be answered by the Project Director. If you would like
additional information about the study before agreeing to participate,
please feel free to call Dr. Brenda Witter, Department of Human
Environment and Design, College of Human Ecology, Michigan State
University (355-3378).

Any participant is free to withdraw consent and to discontinue
participation in the project at any time. No coercive practices
will be used to guarantee participation. The study will require
approximately 20 minutes and will be conducted during one of the
regular class periods of the school day.

Student Signature

 

Date

 

Parent or Legal
Representative
Signature

 

Date

 

95

APPENDIX D

PLEASE NOTE:

Copyrighted materials in this document
have not been filmed at the request of
the author. They are available for
consultation, however, in the author's
university library.

These consist of pages:
96—110

 

 

 

 

 

 

 

Uni °
Micr nlms
lntemational

330 N. ZEEB 80.. ANN ARBOR, MI 48106 (313) 761 ~47OO

 

96

~ - <--~--.~ -\'\- w .. ,- .,
, .-.”..H. ..._. H-,... .’ ......HH .-. —-..

Il- FISTED

A real fighter. Kills germs by millions.
So stron it keeps
breath esh for
hours... and
rs.

 

 

 

98

u ' ' ~
~
. u .

_ TyNwtmt :

y.)

           
 

E~HHmiéis’L-J'7’ xv ' ~ ..

 
 

Tastes as goal
as Scope...

{c

a fresh
minty taste!

 

She figured dentists give Lavoris’ to their patients...
dentists give it to themselves. So instead of using some
mild minty stuff. . .or something that smells like a bottle
of medicine...l used Lavoris. And you know what?
Mother was right. Take it from a dentist's wife: for really
clean, really fresh mouth and breath...

...more dentists (including my husband)
use refreshing Lavoris than any other mouthwash.

Your macaw-Ii tell you
‘ ' bad breath;

 

 

IXFNK Ki ., “Eta

 

 

  
  

  

- Feelm Fa
. W“ ‘

 

 

 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

— ‘ Lookin’ Fury.
laslic.

Kick on a pair of Scholl

, Exercise Sandals

The looks right. In Light or
new dark-toned, natural wood.
Both in so many strap colors,
you can wear them with ' ‘
everything. ‘ g - ‘ ‘

And. the feelin's fine. Pure

.. comfort all day long from the ,
' soft; foam-padded strap. And _

cool, carved European

oodsoles. ' '
. y Whatyou don't feel is the

toe grip at work. It makes your.
toes "hang onl' '

exercises your legs to help firm. .-
and tone them while you walk. .

< So, c'rnon FOxy, feel‘
fantastic. With Scholl Exercise

.. v.“.-;<v:‘r .'l“);:'r ..-. .NN-...

.

in. Fiji! nil-a EIEEEEKCI; F. st.

 

 
 

It’s far. the sake of making feet;
legs and backs; as comfortableg
“ " , as passible: ' ’

. $31“ ”if?"

‘ fdurth pushes

initial shock to the
heel and ankle;

Andi'thefiaggg

 

,Please be patient.
We re making our shoes
as fast as we can.

 

 

y.

So she worked
for 10 years re-
fining every deli-
cate adjustment.
Until finally they
were perfected.

The shoe
that works with
your body.

And the result
was the Earth
shoe.The shoe

 

We’re amazed. Really
amazed. At first people
called our EARTl-l‘ brand
negative heel shoes strange
and ugly. And now they're
standing in line to get them.

And while the ends of
the lines are waiting to get
into our stores, the begin-
nings of the lines are buy-
ing up all of our shoes.

Of course we always
knew Earth8 shoes were
a great invention. And
we knew people would
love them. But we had
no idea the word would
spread so fast.

It all started with
lens liaise.

It started in Denmark
17 years ago, when Anne
Kalso had the idea forthe
negative heel shoe. A
shoe with the heel lower
than the toe.

The concept was that
these shoes would allow

 
 
  

)
i

  

body IS desrgzned to walk.
And that this shoe would

 

Der shoes are sold at stores that sell only the BART!

Who eter heard ofstanding .

in Iinefor n prur ofshoes‘.’

 

   
 
     
     
     
      
 

ii..-
g; :
5a

shoes take time to make. . g.

    
   
   
   
  

this is the natural way the g

that's not just for
your feet.

Not only is the heel
lower than the toe. but the
entire sole is molded in a
very special way. This
allows you to walk in a
gentle rolling motion. And
to walk easily and com-
fortably on the hard,jar-
ring cement of our cities.

Even the arch of the
Earth shoe is different.

to keep your toes
from being
cramped or
squashed.
low everybody .

wants them. ,

buying them. You
told your friends

told their friends.
Until finally it’s
happened. Now you
want them faster
than we can make

them.

It takes time to
make a good shoe.
Earth negative heel

Of course we could .
knock them out fast, by .'

EARTH shoe works. stand

.- on u k. Feel what
begins to happen

104

and the toes are wide

  
   
  
  
   
 
 
  
  
   
 
    
    
  

V about themAnd they

—— "'1;
To get an idea oflzcw the '

 
 
 
  
 
     
  

leaving
out a lot
of important
features. Or by
not paying attention to
quality. But then it
wouldn't be the Earth
brand shoe.

Lowering the heel
isn’t enough.
We knew we had a good
idea. And we knew others
would try to imitate
us by making negative
heel shoes too.
But just because a
shoe looks like ours
doesn‘t mean it works
like ours.
The 10 years that
went into perfecting
the Earth shoe are very
important. We have
many, many features
built into our shoes to
make them work. And
that is why they are
patented.

Soto be sure you're
getting the Earth neg-
ative heel shoe. look on

_ the sole for our Earth
- trademark. and US. pat-
ent number 3305947.

They're worth
waiting for.

Please be patient.We're
sending out more and
more shoes to our stores

 

barefoot with your toes up i!

-

    

 

The EARTHs shoe comes
in stylesfor men and women,
from open sandals to high
3.50 to
$42.50. Prices slightly
higher in
t e west.

   

  
   
   
   

every month. And

if they've run out of

your size or style,
they‘ll have it soon.

And when you do try
them, you‘ll see. perhaps
for the first time in your
life. what it's like to walk
more gracefully, naturally
and comfortably.

And. believe us. that's
worth waiting for.

'EARTH is the registered
trademark of Kalso Systemet,
Incfor its negative heel shoes
and other products.

 

Anne Isles.
[mentor ofthé EARTH‘
negutne lieel slice.

 

hoe. for a list of these stores please see the laclng page.

OW". Kola. 51ml. inc

I"
t
\
t
e
s
e
e
e
e
.
e
O
a
t
|
Q
\
t
|

S. I
fm“\,

 

Sole excitement
in a racer, a
ridge and athick
crepe bottom. In
suede and
leather ties and
mode? The
latest sport
shoes to wear
underneath
it all!

-v"

i

 

...;Jo-11Q’

4"!* U .
Irv-firs .. «

‘qugg

MOST STYLES
. ‘19'70 $27
Wohl Shoe Company, Saint Louis, Missouri - a member of Brown Group. Inc.

LAZARUS. COLUMBUS/RIKE’S, DAYTON/SHILLITO'S. CINCINNATI/PALMS ROYAL. HOUSTON/FA
SPRINGFIELD/CARQWN, PlRlE. SCO

        

TTB.

SHION BAR. DENVER/ALBERTS, DETROITlSTEIGER'S OF
(30.. CHICAGO/STEINBACH'S. NEW JERSEY/MACY'S. CALIFORNIA. KANSAS, MISSOURI, NEW YOP'fi/DONA'.» I . <1
. n . n E ‘

 

*.15
J.

r

 

 

. 1.13.». r_. , . , . .. .. . , . , , . . . .-..
u _ . . . . . _, . . .. .. ., . .... .. .. ,. . .. . .....l.y4v9.3_.4& . In (A

u. ... 1., .

.1;!.....x..f... . ..

......i..:..mwbe , E as
......irtLK) .x .
.5: ...... ...:..,.._.:.._ 25%

0
\f-

3.
. . 5:.
VI i

iettjmcoi , _

3 no a... at.

\/.‘. ..

naval

;

last .

 

 

There is no evidence which
indicates that men prefer
the Hanes brand of pantyhos

/‘~ " .an.

tmerdepanmentanc apparel stores

 

__ v
m-‘-‘*

-..L. J, . fl.

 

 

No nonsense

 

news.

Introducing the cotton-lined ventilated crotch panel.

The best fashion news today IS one
of the oldest fabrics in the world
Cotton. Because it‘s natural. Because
its soft. Because it's cool on your skin.
That‘s why the ventilated crotch
panel in every pair of panty hose
No nonsenseQmakes is now
lined with cotton. Because it's
naturalsoft and cool.You still get
No nonsense fit. with new
No nonsense comfort.
You‘ll look just as good And
you ‘II feel even better
No nonsense fit _a_nd
No nonsense comfort—That‘s
No‘nonsense fashion. o mammal“...

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
  

 

No nonsense fit. No nonsense comfort. No nonsense fashion.

Another hoe p'OOuC' cl Koysor Roth

108

 

    
  
 

  

What makes L'eggs‘E

pantyhose special? Memory

Yarn.” Memory Yarn hugs your

ankles. Memory Yarn holds your

knees. Memory Yarn stretches
out and springs back '_

to fit thin legs, plump legs,

at the L'eggs Boutique, in - at?

 

L

4 .
“ Average or Queensuze. - as

 

tall legs, all legs. Beautifully.

Without bagging or sagging.
Remember to buy L’eggs© pant)

hose with Memory Yarn

 

I

answer:

“a... ...w"

.‘
E
.
‘5 n

‘m s ..w

 

 

 

A. -. i.- -

‘eneda.

 

 
 

 

  
  

  

.. givegthem‘a little helsz'vé thei
. - the pantyhose ’ . .

" . hath all-day massage. . ' .

2*" =I=-f“»*‘*5f“5[ifihtlzern on; andyoullfeel ~~ . ' “as

 
 

     

  

     
    
  
  
   
 

'4 3‘2“ ,-
’ -' ‘ ~_' . . . ,7 .. . '1.
“l “I . .‘ -'r V ‘._

A . _ 'u . . 1 .

V ‘ . ‘,.v_~_ ' ".-

' “we. -

    
 

    
  
    

 
 

Elthéir massage light away. Its wild! . 35$
. . :1: Now,- flex your leg and feel Sheer 4. Efi'figfi
‘ ‘E. g . .- ., N.-

 
 
  
 

  

Energys springy yam massage even more.

" _ And there's something even wilder! The is M”
3 ' . ‘ ,Jnore you move, the more .,
‘ a...“ \ they massage. ' ’ ‘

  
 

  

' j '. r’- ,‘

     

\\ \ 723-5!" . C!

 

APPENDIX E

Table 18

Multiple Regression Analysis of Fishbein Attitude Model

n
Aact = 2 B A,
1

 

i 1
DP SS M2 R2 F
Mouthwash
Scope
Aact 1 116.497 116.497 0.254 34.000****
Error 100 342.679 3.427
Listerine
Aact 1 78.808 78.808 0.156 18.439****
Error 100 427.398 4.274
Lavoris l
Aact 1 75.238 75.238 .233 30.418****
Error 100 247.350 2.474
Listermint l
Aact 1 127.681 127.681 .298 42.541****
Error 100 300.133 3.001
Footwear
Scholl
Aact 1 74.540 74.540 0.183 17.214****
Error 77 333.435 4.330
Famolare
Aact 1 33.043 33.043 0.123 10.781**
Error 77 235.995 3.065
Fanfare
Aact 1 75.923 75.923 0.273 38.937****
Error 77 202.027 2.624
Earth Shoes
Aact 1 100.065 100.065 .287 31.029****
Error 77 248.314 3.225
Pantzhose
Sheer Energy
Aact 1 36.346 36.346 .148 11.652***
Error 67 208.987 3.119
Hanes
Aact 1 21.529 21.529 .109 8.217**
Error 67 175.544 2.620
L'eggs
Aact 1 2.823 2.823 .010 .643
Error 67 294.250 4.392

-CONTINUED-

111

Table 18 - Continued

DF SS M R F

 

No Nonsense
Aact 1 29.538 29.538 .084 6.106*
Error 67 324.114 4.838

.05
.01
.001
.0001

* p

** p
*** p
***‘«'~' p

AAAA

Table 19

Multiple Regression Analysis of
Fishbein Behavioral Intention Model

BI 3' [Aact]w0 + [(NB)(MC)]wl

 

 

 

DF 88 M2 R2 F
Mouthwash
Scope
BI 1 126.362 126.362 0.275 37.968****
Error 100 332.815 3.328
Listerine
BI 1 80.247 80.247 0.159 18.839****
Error 100 425.959 4.260
Lavoris
B1 1 74.867 74.867 0.232 30.222****
Error 100 247.721 2.477
Listermint
BI 1 138.787 138.787 0.324 48.019****
Error 100 289.027 2.890
Footwear
Scholl
B1 1 84.415 84.415 0.207 20.089****
Error 77 323.560 4.202
Famolare
BI 1 36.197 36.197 0.135 11.970***
Error 77 232.841 3.024
Fanfares
B1 1 77.806 77.806 0.280 29.934****
Error 77 200.144 2.599
Earth Shoes
B1 1 113.639 113.639 0.326 37.276****
Error 77 234.741 3.049
Pantyhose
Sheer Energy
B1 1 38.027 38.027 0.155 12.290***
Error 67 207.306 3.094
Hanes
BI 1 20.560 20.560 0.104 7.804**
Error 67 176.513 2.635
L'eggs
BI 1 3.131 3.131 0.011 0.714
Error 67 293.941 4.387
-CONTINUED-

113

No Nonsense
BI
Error

.05
.01
.001
.0001

x.
a.
'U
AAAA

Table 19 — Continued

 

DF SS
1 27.523
67 326.129

114

M2

27.523
4.868

R2

0.078

F

5.654*

   

"Illfillllllllllflll