 

 

"Z" LIBRARY
2001, Michigan State
University

 

 

 

This is to certify that the
dissertation entitled

EXAMINING ATTITUDE DECAY IN PREVIOUSLY FORMED
AND NEWLY FORMED ATTITUDES FOLLOWING
PERSUASION

presented by

Sun Young Lee

has been accepted towards fulﬁllment
of the requirements for the

PhD. degree in Department of Communication

 

 

9%]. at

Major Professor’s Signature
A Ndrlarbtr 9-005.

Date

 

MSU is an Afﬁrmative Action/Equal Opportunity institution

—v--‘-:-v- --t-.— -»--.-.- -t-.-.-.-.—u-'vsio

 

PLACE IN RETURN BOX to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2/05 p:IClRC/Date0ue.indd-p.1

 

EXAMINING ATTITUDE DECAY IN PREVIOUSLY FORMED AND NEWLY
FORMED ATTITUDES FOLLOWING PERSUASION

By

Sun Young Lee

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Communication

2005

ABSTRACT

EXAMINING ATTITUDE DECAY IN PREVIOUSLY FORMED AND NEWLY
FORMED ATTITUDES FOLLOWING PERSUASION

By

Sun Young Lee

This study aimed at gaining understanding of attitude decay in persuasion.
It examined whether previously formed attitudes versus newly formed attitudes are
subject to decay, and if reinforcement affects attitude decay. This study was
composed of three single factor experiments, with participants randomly assigned to
one of three studies. For the three studies, Study 1 used the topic of increasing
university tuition on which people have previously formed attitudes in order to
investigate the decay of attitudes modiﬁed in the positive direction after exposure to a
persuasive message. Study 2 employed the topic of binge drinking on which people
have also previously formed attitudes in order to investigate the decay of attitudes
modiﬁed in the negative direction after exposure to a persuasive message. Study 3
used the topic of apheresis donation to examine the decay of newly formed attitudes
in the positive direction aﬁer exposure to a persuasive message. The results showed
that previously formed attitudes decayed to their initial positions without
reinforcement. With reinforcement, however, previously formed attitudes did not

decay. Regardless of reinforcement, newly formed attitudes did not decay to their

initial attitudinal positions. Interestingly, however, newly formed attitudes were not

consistent, and instead decreased toward the direction of initial attitudinal positions.

ACKNOWLEDGEMENTS

I thank my advisor Dr. Frank Boster, for his advice, challenge, and always
guiding me in the right direction. I’ve learned a lot from him, and without his help I
could not have ﬁnished my dissertation successfully. I also thank the rest of my
dissertation committee: Dr. Sandi Smith, Dr. Tim Levine, and Dr. Stan Kaplowitz,
who gave insightful comments and reviewed my work. Besides my advisor, I would
like to thank Dr. Mary Bresnahan for a productive collaboration and as co-authors. I
appreciate all her friendship and encouragement throughout my graduate program.

I thank my parents, Minwoo Lee and Soonlee Kim. Their understanding and
love encouraged me to work hard and to continue pursuing a Ph. D. program abroad.
A special thanks goes to my devoted husband, Hansung Kim for listening to my
complaints and frustrations, and for believing in me. Without his love, support and
encouragement I could never have completed this work.

Last, I would like to thank professors at my home university, Ewha Women’s
University for getting me interested in Communication and coming to the United
States. There are so many friends, both in South Korea and in East Lansing, who

offered me so much kindly help. I give my sincere thanks to all these people.

TABLE OF CONTENTS

LIST OF TABLES .............................................................. vii
LIST OF FIGURES ............................................................ viii
INTRODUCTION ................................................................. 1
CHAPTER I ....................................................................... 3
Persuasion Studies in Communication
and Attitude Decay ...................................................... 3
Attitude Decay in Previously Formed Attitudes
versus Newly Formed Attitudes ....................................... 5
The Double Comparison Model of
Attitude Change and Attitude Decay ................................. 7
Information Integration Model and
Attitude Decay ........................................................... 9
Reinforcement of Attitudes Following
Persuasive Communication ........................................... 13
CHAPTER 2 ....................................................................... 19
Method of study I ...................................................... 20
Results of study 1 ...................................................... 24
Method of study 2 ...................................................... 26
Results of study 2 ...................................................... 29
Method of study 3 ...................................................... 31
Results of study 3 ....................................................... 34

CHAPTER 3 ..................................................................... 36

Previously Formed Attitudes,
Newly Formed Attitudes, Reinforcement,

and Attitude Decay .................................................... 37
Attitude Fluctuation in Newly Formed Attitudes:
Not Decay but Variation .............................................. 40
Implication to Persuasive Message Design ........................ 41
Limitations .............................................................. 42
Future Research ........................................................ 42
Conclusion ................................................................ 43
ENDNOTES ..................................................................... 45

APPENDIX A: PERSUASIVE MESSAGES......................................46

APPENDIX B: BINGE DRINKING ADVERTISEMENTS... . ...50

APPENDIX C: MEASUREMENT RETAINED AFTER CFA ............ 53

BIBLIOGRAPHY ............................................................... 55

vi

LIST OF TABLES

Table I. Attributes of Attitude Strength ..................................... 64

Table 2. Means and Standard Deviations for
Attitudes toward Increasing University Tuition
at the Three Tests by Essay and No Essay Condition .................... 65

Table 3. ANOVA Summary Table for Attitudes
toward Increasing University Tuition ....................................... 66

Table 4. Means and Standard Deviations for
Attitudes toward Binge Drinking at the Three Tests
by Essay and No Essay Condition ........................................... 67

Table 5. ANOVA Summary Table for Attitudes
toward Binge Drinking ........................................................ 68

Table 6. Means and Standard Deviations for
Attitudes toward Apheresis Donation at the
Three Tests by Essay and No Essay Condition ............................ 69

Table 7. ANOVA Summary Table for Attitudes
toward Apheresis Donation .................................................. 7O

vii

LIST OF FIGURES

Figure 1. ANOVA Summary Table for Attitudes

toward Apheresis Donation ...................................................... 7]
Figure 2. Factors Affecting Attitude Accessibility ........................... 72
Figure 3. Design of Study ........................................................ 73

Figure 4. Attitudes Change toward Increasing University
Tuition as a Function of Essay/No Essay and the Three Tests .............. 74

Figure 5. Attitudes Change toward Binge Drinking
as a Function of Essay/No Essay and the Three Tests ....................... 75

Figure 6. Attitudes Change toward Apheresis Donation
as a Function of Essay/No Essay and the Three Tests ....................... 76

viii

INTRODUCTION

During the past four decades, considerable persuasion research has been
conducted to illustrate how to modify people’s attitudes or form new ones, with
researchers expressing increased interest in the various factors affecting attitudes. They
have investigated message characteristics, such as argument strength (Hosman, Huebner,
& Siltanen, 2002), argument sidedness (O'Keefe, 1990), message order effects
(Haygtvedt & Wegener, 1994), framing effects (Maheswaran, 1990), and fear appeals
(Witte, 1992). They also investigated receiver and context characteristics, such as
receivers’ personality traits, level of involvement (Leippe & Elkin, I987), cognitive
ability to process messages (Petty & Cacioppo, 1986), and mood (Petty, Schumann,
Richman, & Strathman, 1993). Other factors researchers were concerned about include
source characteristics such as source credibility (Burgoon, Birk, & Pfau,1990), source
expertise (Homer & Kahle, 1990), source likeability (Roskos-Ewoldsen & Fazio, 1992),
and source’s speech rate (Smith & Shaffer, 1995).

Almost all persuasion research thus far, however, investigates attitude
modiﬁcation or new attitude formation immediately after exposure to a persuasive

message, and fails to examine attitude decay. In general, persuasion scholars

acknowledge that persuasive effects decay over time. O’Keefe (1990) stated, “Old
habits and attitudes can return, competing persuasive messages can be received, and
hence the impact of a given persuasive effort is likely to diminish over time” (p. 186).
Cacioppo, Petty, and Green (1989) also argue that when attitudes are changed by a
persuasive message, the attitude system works to return the modiﬁed attitudes to their
initial positions. However, it is not clear if all types of attitude are subject to decay,
since there is a lack of research on the attitude decay phenomenon. The attitude decay
issue is of signiﬁcance because of the practical importance of studying long-term effects
of persuasion.

The current study seeks to identify conditions under which attitude decay is likely
to occur. In particular, it examines whether previously formed attitudes versus newly
formed attitudes produce attitude decay. In addition, it investigates if reinforcement of
those previously formed or newly formed attitudes affects attitude decay. Given the
lack of empirical ﬁndings, the following review will serve as the rationale for the

hypotheses developed for the study.

CHAPTER I

Persuasion Studies in Communication and Attitude Decay

Persuasion is deﬁned as “a successful intentional effort at inﬂuencing another’s
mental state through communication in a circumstance in which the persuadee has some
measure of freedom” (O’Keefe, 1990, p.17). As for the mental state mentioned in the
deﬁnition, the concept of attitude has been seen as the principal representative of it.

Persuasion, in general, has been studied in three ways in communication.
Researchers investigate (a) the method and (b) the process of producing attitude
modiﬁcation or formation, and (c) the way of inducing attitudinal resistance to persuasive
effort. The two variables involved are message as the independent variable and attitude
as the dependent variable (Dillard, 1993). Message characteristics that have been
studied include argument strength, argument sidedness, message order, message framing,
and the type of appeal (see O’Keefe, 1990). In addition to message characteristics,
researchers have expanded this simple equation of persuasion with additional
independent characteristics of sources, receivers, and context. Source characteristics
studied include source expertise, credibility, likeability, and speech rate (see Burgoon et

al., 1990; Homer & Kahle, 1990; Roskos-Ewoldsen & Fazio, 1992; Smith & Shaffer,

1995). Receiver and context characteristics consist of receivers’ personality traits,
cognitive abilities, involvement, and moods (see Leippe & Elkin, 1987; Petty &
Cacioppo, 1986; Petty et al., 1993). Many studies have investigated the method of
forming or modifying attitudes by using those message, source, receiver, and context
characteristics. For instance, researchers ﬁnd that more credible sources generate more
attitude change than less credible sources (Burgoon, 1990).

Despite the abundance of persuasion research, little attention has been paid to the
attitude decay problem. Attitude decay occurs when attitudes return to their initial
positions following persuasive communication (see O’Keefe, 1990). Attitude decay is
expected if attitudes are part of a larger belief structure. Poole and Hunter (1979) show
attitudes that are embedded in a larger belief structure tend to return to their initial
positions following persuasive communication. Given that attitudes that are embedded
in a belief structure are relatively strong attitudes, decay is likely to occur with these
strong attitudes when they are modiﬁed by persuasive communication. Attitude decay
may be of importance when the goal of a persuasive message is not immediate decision
making or action. Many studies ﬁnd that persuasive messages have an effect on
immediate decision making, but also that modiﬁed attitudes decay as early as two days

after persuasive communication (Haugtvedt & Strathman, 1990). Therefore, if a time

gap exists between persuasive effort and decision making, or if attitude persistence over

time is desired, reinforcement is needed.

Attitude Decay in Previously Formed Attitudes versus Newly Formed Attitudes

As attitude decay implies lack of attitude persistence, the feature of persistent

attitudes needs to be examined. Although no reliable empirical information is available

regarding factors affecting attitude persistence, some researchers speculate about them

(O’Keefe, 1990). One factor might be the number of arguments presented in a

persuasive message that supports the attitude. If the modiﬁed attitude is affected by

many arguments, it appears to be persistent (compare Leventhal & Niles, 1965 with

Calder, Insko, & Yandell, 1974). Communicator credibility also might affect attitude

persistence. Cook and Flay (1978) suggest that a highly credible source is associated

with attitude persistence. In addition, the receiver’s high involvement with the issue

might be associated with attitude persistence (see Cook & Flay, 1978). The

communication modality also seems to be related to attitude persistence, as Chaiken and

Eagly (1983) found that written messages generate slightly stronger attitude persistence

than audiotaped or videotaped messages. Another factor might be the amount of

cognitive elaboration employed in processing a persuasive message. This factor is

inferred from the Elaboration Likelihood Model, which suggests two routes of attitude

modiﬁcation. That is, when people process a persuasive message through a central

route, which involves careful attention to issue relevant information, modiﬁed attitudes

are strong and persistent. On the other hand, when people’s attitudes are modiﬁed

through a peripheral route, which relies on heuristics, their modiﬁed attitudes are

relatively weak and not persistent over time.

Another potential factor might be the type of attitudes people acquire after

receiving persuasive communication, namely previously formed attitudes or newly

formed attitudes. Previously formed attitudes refer to attitudes that are formed positive

or negative way toward a topic before receiving a persuasive message, and may be

modiﬁed in either the negative or the positive direction toward the topic after receiving a

persuasive message. In particular, previously formed positive or negative attitudes

toward a topic are strong attitudes because they have been fully formed and embedded in

a larger belief structure before exposure to persuasive communication. Those

previously formed attitudes are, therefore, likely to decay to their initial positions

following exposure to a persuasive message because the initial attitudes are strong.

Newly formed attitudes refer to attitudes that are initially neutral toward a topic

before receiving a persuasive message because of unfamiliarity toward the topic.

People form their new attitudes toward the topic in either the positive or the negative

direction after exposure to a persuasive message. In other words, people ﬁrst form their

attitudes toward an unfamiliar topic through information shown in a persuasive message.

Because these newly formed attitudes become an individual’s fully formed initial

attitudes, they would not likely decay to their initial neutral positions. Figure 1

summarizes this speculation.

The Double Comparison Model of Attitude Change and Attitude Decay

The argument that attitude decay pertains only to previously formed attitudes is

supported by Hunter, Levine, and Sayers (1976). According to their research, most

attitudes are embedded in a hierarchy of beliefs about concepts, which are arranged on

levels from the most abstract to the least abstract. They also suggest that downward

inﬂuence of beliefs is stronger than upward inﬂuence (if it exists), and that horizontal

inﬂuence does not occur. Attitude modiﬁcation can be acquired by having messages at

any or all levels of abstraction. Hunter et al. (1976) propose a double comparison model

for attitude change with a single message. The model suggests that, (a) people compare

their attitudes toward a speciﬁc object with the position advocated in the external

message, and (b) they compare their attitudes toward the object with their attitudes

toward objects (or concepts) directly above in the hierarchy. They suggest that this

comparison is made between two adjacent levels. In other words, people compare their

feelings for an object with both external and internal messages aﬁer exposure to a
persuasive message. Based on previous persuasion research, it seems that people
mainly engage in external message comparison during exposure to a persuasive message,
as numerous research reports a positive persuasive message effect. As time passes,
people seem to engage mostly in internal message comparison, which explains the
attitude decay phenomenon. The internal message comparison affects attitude decay
because people’s fully established initial attitudes are already well-organized and
associated with superordinate concepts in their belief systems, and the downward
inﬂuence of the beliefs associated with this initial attitude is strong. The model also
implies that attitudes embedded in a belief system are strong attitudes because weak
attitudes are not well associated with other beliefs (Krosnick & Petty, 1995). In the
same fashion, initial attitudes that are well organized in a belief system before exposure
to a persuasive message are strong attitudes, such that they cause previously formed
attitudes to decay.

However, newly formed attitudes after exposure to a persuasive communication
are not well-formed because people create their new attitudes toward a novel topic
through a persuasive message. Their not well-formed initial attitudes are, therefore, not

embedded in a belief structure and not inﬂuential on internal message comparison. As a

result, newly formed attitudes are not likely to decay.

Information Integration Model and Attitude Decay

In addition to the double comparison model of attitude change, an information

integration model suggested by Kaplowitz, Fink, Armstrong, and Bauer (1986) also

provides relevant information for predicting attitude decay in certain kinds of previously

formed attitudes, and no attitude decay in certain kinds of newly formed attitudes. Their

model is based on the information integration theory suggested by Anderson (1981)

describing the process by which different pieces of information are combined to form

evaluations. According to Anderson’s information integration theory, evaluations such

as attitudes or beliefs are newly formed or modiﬁed as people receive, interpret, evaluate,

and integrate information with their existing beliefs or attitudes (Simonin, & Ruth, 1998).

The information integration model suggested by Kaplowitz et al. (1986), however, further

explains the process for long-term attitude change.

From their model, a factor relating to attitude decay is the number of delayed

messages people receive (messages people receive aﬁer persuasive effort). They

suggest that “the greater is the total weight of the delayed messages, the more the ﬁnal

attitude will approach the average scale value of the delayed messages” (p. 514). The

total weight means the total weight of messages people receive after persuasive

communication, and it depends on length of and number of arguments in the message,
and source credibility (Kaplowitz et al., 1986). The average scale value of the delayed
messages refers to “the position that the subject interprets the message as advocating” (p.
510).

For the previously formed attitudes toward familiar issues, it is likely that people
have a social environment containing others who I) care about these familiar issues and
2) have an average view that is similar to their own initial view. Consequently, after
exposure to a persuasive communication, people are likely to receive many messages
advocating their own initial view. Therefore, attitude decay toward this initial view is
expected, unless someone (e.g., the experimenter) provides messages that reinforce the
new opinion.

Conversely, newly formed attitudes toward novel issues are not just newly formed.

In addition, they are probably something that most others the subject encounters know
almost nothing about. Consequently, these others in the social environment are not
likely to try to inﬂuence the subjects to return to their initial neutral opinion toward the
issues. Therefore, there will probably be no messages from their peers that encourage
decay of the newly formed attitudes. In short, attitude decay is not expected regardless

of whether the subject is provided with messages reinforcing their newly formed attitudes.

lO

Reinforcement of Attitudes Following Persuasive Communication

If attitude decay phenomena exist in previously formed attitudes, then
reinforcement to those previously formed attitudes is needed to have a long-term
persuasive effect. The reinforcement would make the previously formed attitudes
strong enough not to decay in a given time following persuasive communication. In
other words, the reinforcement assists previously formed attitudes to have characteristics
of strong attitudes: durability and impactfulness (Krosnick & Petty, 1995). Durability is
persistence over time and resistance to counterpersuasion, and impactfulness causes
consequential inﬂuence on thought and behavior.

Researchers report numerous attributes of strong attitudes that can be applied to
reinforcement. As shown in Table 1, the attributes can be grouped into four categories
(see Krosnick & Petty, 1995). One category consists of features of the attitude itself,
such as extremity (Judd & Johnson, 1981; Osgood & Tennenbaum, 1955) or intensity
(Cantril, 1946; Stouffer, Guttman, Suchman, Lazarsfeld, Star, Clausen, 1950). The
second category consists of cognitive structure associated with the attitude object, such as
attitude accessibility (Bassili & Fletcher, 1991; Bizer & Krosnick, 2001; Fazio, 1995;
Krosnick, 1989; Roese & Olson, 1994), knowledge (Kanwar, Grund, & Olson, 1990), or

affect-cognitive consistency (or matching one’s feelings about an object with one’s belief

ll

about its attributes) (Chaiken & Baldwin, 1981; Norman, 1975). The third category

consists of cognitive processes by which an attitude is formed, such as the amount of

elaboration (Bradburn & Caplovitz, 1965; Kendall, 1954) or direct experience (Regan &

Fazio, 1977; Schurnan & Presser, 1981). The fourth category consists of the subjective

beliefs about an attitude object and an attitude, such as personal importance of the

attitude (Bizer & Krosnick, 2001; Boninger, Krosnick, Berent & Fabrigar, 1995;

Krosnick, 1989), centrality of the attitude to the self-concept (Bizer & Krosnick, 2001),

attitude certainty (Holland et al., 2003; Krosnick & Schuman, 1988), or latitudes of

rejection and noncommitrnent (Sherif, Sherif, & Nebergall, 1965; Sherif, Kelly, Rogers,

Sarup, & Tittler, 1973).

Among the attributes of strong attitudes, attitude accessibility has been a frequent

focus of study, and is also exploitable as a target of reinforcement. Attitude accessibility

is deﬁned as the strength of the link between an object and its evaluation in memory

(Fazio, 1995), such that a highly accessible attitude has a stronger link between the object

and its evaluation. A network model of memory presents an explanatory framework

regarding attitude accessibility in memory.

A network model of memory (see Anderson, 1983; Green, 1984; Smith, 1994)

posits that any concept obtained is represented in memory as a node, and certain nodes

12

are linked by associative pathways. A node is activated by exposure to related
information or thinking about such information, and then activates other nodes through
associative pathways. In this model, attitudes are seen as associations in memory
between attitude objects and their evaluations (Fazio, 1986). Nodes that are strongly
connected have strong associative pathways and vice versa (Pfau et al., 2003). When
the connection between an attitude object and its evaluation is strong, the accessibility of
the attitude toward that object is high. Consequently, highly accessible attitudes are
easily and quickly retrievable from memory. Therefore, reinforcement attempting to
increase attitude accessibility would produce strong attitudes, and thus less likelihood of
attitude decay.
The Method of Writing an Essay as Reinforcement of A ttitudes Following Persuasion

To affect attitude accessibility, the factors affecting attitude accessibility need to
be examined. As shown in Figure 2, one factor is frequency of expression of attitudes.
The more often people express their attitudes, the more accessible they are. Direct
behavioral or sensory experience in forming attitudes also affects attitude accessibility
because physical inputs are memorable. Another factor is the strength of the emotional
reaction to an attitude object. For example, if someone has a strong emotional reaction

toward an object, the attitude toward that object becomes highly accessible (Fazio, 1995).

I3

In addition, the hierarchical arrangement of attitudes affects accessibility such that higher

ranking attitudes are more accessible than lower ranking attitudes (Dillard, 1993).

Higgins and King (1981) also suggest that attitude Lmiqueness and an attitude’s link to

other constructs affect attitude accessibility. Thus, if an attitude is more distinctive, it is

more accessible. Furthermore, the more strongly an attitude is linked to other constructs

(e.g., values) through associative pathways in memory, the more accessible it is. Others

suggest that the expectation of evaluating an attitude object in the future, the amount of

cognitive elaboration used in evaluation, and how recently the attitude has been activated

from memory, inﬂuence attitude accessibility (see Roskos-Ewoldsen et al., 2002).

The current study, in particular, focuses on strengthening the link of an attitude to

other constructs to increase attitude accessibility, and suggests that writing an essay about

the attitude- associated construct would strengthen the link. The method of writing an

essay about the construct would make the construct salient. Information is salient if it is

noticeable, important, and memorable (Entman, 1993). If a persuasive message focuses

on a certain value in its context, previously formed or newly formed attitudes after

exposure to this persuasive message are linked to this certain value. That is, people

modify their attitudes when they think about that value. For example, if a persuasive

message supports a certain political candidate, and this candidate emphasizes equal

14

opportunity and national defense in a message, the message promotes the values of
equality and security. When people have strong negative attitudes toward this candidate,
but embrace those values, then they will change their attitudes in the direction of
supporting the candidate. Their attitude change toward the candidate in the positive
direction is associated with equality and security values. The attitude decay
phenomenon, however, suggests that their changed attitudes will return to their initial
negative attitudes toward that candidate if the persuasive message does not require
immediate decision making, or if listeners do not receive additional persuasive messages.
That is, the link between their modiﬁed attitudes toward the candidate and valuing
equality and security becomes weak, other beliefs associated with their initial attitudes
toward the candidate come to mind, and they therefore refer to their initial negative
attitudes.

If the audience, however, writes an essay about the importance of “being equal”
or “being secure,” then they will view the value of equality or security as more salient
than other beliefs in their belief structure. Moreover, people will recognize that the
equality or security value is associated with their modiﬁed attitudes toward the candidate.
The writing, consequently, plays a role in strengthening the link between their modiﬁed

attitudes toward the candidate and valuing equality and security, and increases

15

accessibility toward the modiﬁed attitudes.

As noted before, because reinforcement is necessary only for previously formed

attitudes, not for newly formed attitudes, after exposure to a persuasive message, the

method of writing an essay would be effective only with previously formed attitudes.

That is, when the persuasive outcome is attitude modiﬁcation from previously formed

attitudes, previously formed attitudes are likely to decay without the reinforcement of

writing an essay, but are not likely to decay with reinforcement of writing an essay.

However, if the persuasive outcome is new attitude formation, newly formed attitudes are

unlikely to decay, regardless of the reinforcement of writing an essay. Thus, the

following predictions are asserted.

H1: If there is no reinforcement, previously formed attitudes, after exposure to a

persuasive message, are likely to decay. However, if there is reinforcement,

previously formed attitudes are not likely to decay.

H2: Despite the presence or absence of reinforcement, newly formed attitudes,

after exposure to a persuasive message, are not likely to decay.

CHAPTER 2

METHOD

Overview

This study was composed of three single factor experiments, with participants

randomly assigned to one of three studies. All studies consisted of a pretest, initial

posttest, and delayed posttest design. Across the three studies, reinforcement was

manipulated by assigning people to either an essay condition or a no essay condition.

Therefore, essay or no essay condition was the independent variable. Attitudes at

pretest, initial posttest, and delayed posttest were the dependent variables.

At pretest, participants’ initial attitudes toward an issue before exposure to a

persuasive message and issue involvement were assessed. Then, they received a

persuasive message designed to induce positive, more positive, negative, or more

negative attitudes toward the issue. The context of the message focused on a certain

value. At the initial posttest, participants were asked to rate their attitudes toward the

issue after exposure to a persuasive message, issue involvement, and source credibility.

After 21 days, participants were randomly assigned to either the essay or no essay

condition. In the essay condition, participants were asked to compose an essay about

17

importance of the value presented in the persuasive message. At the delayed posttest,
participants in both the essay and no essay conditions were asked about their attitudes
toward the issue (see Figure 3 for the design). Attitude decay would be detected by
comparing attitude scores at initial posttest with those at the delayed posttest.

To examine attitude decay in common persuasive efforts (positive attitude change,
negative attitude change, and positive attitude formation), this study employed three
topics. Study 1 used the topic of increasing university tuition to investigate the decay in
attitudes which had previously formed and modiﬁed in the positive direction after
receiving a persuasive message. Study 2 used the topic of binge drinking to investigate
the decay in attitudes previously formed and modiﬁed in the negative direction after
receiving a persuasive message. Study 3 used the topic of apheresis donation to
examine the decay in attitudes newly formed in the positive direction after receiving a
persuasive message. For organization purposes, this section presents the methods and
results of Study 1, Study 2, and Study 3.

Method of Study 1
Topic Selection and the Persuasive Message
This study focused on decay in previously formed attitudes and modiﬁed in the

positive direction after exposure to a persuasive message. It employed the topic of

increasing university tuition to produce fully formed initial attitudes, speciﬁcally negative
ones, and to allow positive attitude change. Participants’ initial attitudes toward the
topic would be well-organized in their belief structures and be strong because this topic
has been familiar and controversial among college students. The persuasive message

was designed to support increasing university tuition and focused on the success value,

 

such that increasing university tuition would help students obtain better jobs (see

Appendix A for all messages).

Design and Data Analysis

This experiment used a pretest, initial posttest, and delayed posttest design. The

attitudes toward increasing university tuition were measured on three tests and analyzed

using 3 (three tests) x 2 (essay/no essay) ANOVA. At pretest, attitudes toward

increasing university tuition before exposure to a persuasive message and issue

involvement were measured. At posttest, attitudes toward increasing university tuition

after exposure to a persuasive message, issue involvement, and source credibility were

measured. After 21 days, participants were randomly assigned to either the essay or no

essay condition to manipulate the independent variable. The essay condition asked

participants to write an essay about the importance of “being successful.” At delayed

posttest, their attitudes toward increasing university tuition were measured.

19

Participants

Fifty-three students at a major midwestem university participated in the study.
Three students were unavailable at follow-up. Therefore, ﬁfty undergraduate students’
responses were retained for analyses. Participation was voluntary, and the investigator
assured conﬁdentiality.

Of these participants, 27 students participated in the essay condition (E) and 23
students participated in the no essay condition (C). The sample was composed of 32
females (64%) and 18 males (36%). They ranged in age from 18 to 24 years, with the
average being 20.30 years (SD=1.30). Among them, 33 were Caucasian (67.3%), 5
were African American (10.2%), 5 were Asian ( 10.2%), 3 were Hispanic (6.1%), one
was Native American (2.0%), and two people were unidentiﬁable (4.1%). The
majority of participants were juniors (44%), followed by seniors (26%), sophomores
(16%), freshmen (12%), and other (2%). Most participants were familiar with (M =
4.78, SD = 1.36) or had heard of increasing university tuition (M = 6.22, SD = 1.00)
before the study.1

Statistics show no evidence that participants’ levels of knowledge, t (48) = 0.43,

p>.05, familiarity with increasing university tuition, t (48) = -0.02, p>.05, and age, t (48)

20

= 0.02, p>.05, differed between conditions. In addition, there was no evidence that
gender signiﬁcantly differed across conditions ()8 = .03, p>.05).
Instrumentation

Conﬁrmatory factor analyses (Hunter & Gerbing, 1982) were conducted for all
scales to test their content validity. Items passing internal consistency and parallelism
tests were retained.

Attitudes toward the issue. Attitudes toward increasing university tuition were
measured three times: at pretest, initial posttest, and delayed posttest. The items were
bad/good, unfavorable/favorable, negative/positive, disagreeable/agreeable, foolish/wise,
unacceptable/acceptable, and wrong/right. The reliability of the attitude scale at pretest
was a = 0.90 (M = 2.34, SD = .86), a = 0.93 at initial posttest (M = 3.52, SD = 1.10), and
a = 0.96 at delayed posttest (M = 3.35, SD = 1.15).

Issue involvement. Issue involvement scale was measured twice: at pretest and
initial posttest (pretest: a = 0.89, M = 6.11, SD = 0.92; initial posttest: a = 0.92, M = 5.99,
SD = 1.07). This variable was measured as a control variable.

Source credibility. Source credibility has two dimensions, expertise and
trustworthiness. The reliability of expertise and trustworthiness scales were 0.77 (M =

4.55, SD = 0.96) and 0.72 (M= 4.32, SD = 0.96), respectively. This variable served as a

21

control variable.
Results of Study 1
T est of the Hypothesis

The control variables (issue involvement and source credibility), gender, race, and
year in college were not signiﬁcantly associated with the dependent variable (attitudes).
Therefore, they were not considered in the analyses.

H1: If there is no reinforcement. previously formed attitudes after exposure to 4a
persuasive messag§_a_.re likely to decay. However, if there is reinforcement. previouslv
formed attitudes are not likely to decay.

Table 2 presents the means and standard deviations for attitudes toward increasing
university tuition at the three tests by essay/no essay conditions. The table shows that,
in general, participants initially had negative attitudes toward increasing university tuition
(M = 2.34, SD = 0.86 at pretest) but they changed their attitudes in the positive direction
after exposure to a persuasive message (M = 3.52, SD = 1.10 at initial posttest, t(49) =
8.55, p<.001).

To test the hypothesis, ANOVA was conducted and results are summarized in
Table 3. Results show a signiﬁcant interaction effect of essay/no essay in the three tests,

F (2, 96) = 17.96, p<.001, partial eta-squared = 0.27. Figure 4 indicates that this

22

interaction may be attributed to the differing pattern of decay graph (the line between

attitude at initial posttest and delayed posttest) of the essay condition compared with the

delay graph of the no essay condition. Here, attitudes toward increasing university

tuition in the essay condition changed more favorable from initial posttest to delayed

posttest. For the no essay condition, attitudes changed less favorable toward increasing

university tuition from initial posttest to delayed posttest.

The interaction effect was further explored by examining the pattern of attitude

scores between the essay and no essay conditions, and it was found that the no essay

condition’s attitudes were signiﬁcantly less favorable than the essay condition, but only at

the delayed posttest, t(48) = -7.88, p<.001. With the no essay condition, attitudes

changed less favorable from initial posttest to delayed posttest, t(22) = 4.55, p<.001, and

the attitude score at delayed posttest did not differ from the attitude score at pretest, t(22)

= -0.94, p>.05, suggesting that previously formed attitudes decayed to their initial

positions without an essay. However, with an essay, attitudes changed more favorable

from initial posttest to delayed posttest, t(26) = -2.52, p<.05. These ﬁndings were

evident for the positive reinforcement effect on avoiding attitude decay in previously

formed attitudes.

23

Method of Study 2

Topic Selection and the Persuasive Message

This study focused on the decay in attitudes modiﬁed in the negative direction
after exposure to a persuasive message. It employed the topic of binge drinking to
produce fully formed initial attitudes, and to allow negative attitude change.
Participants’ initial attitudes toward the topic would be well organized in their belief
structure and be strong because this topic has been familiar and controversial among
college students. The persuasive message was designed to oppose binge drinking. It
illustrated the negative effects of binge drinking, supported the development of campus-
wide programs to decrease binge drinking, and focused on the Mtg value, such that the
programs would help students to be physically and psychologically healthy.
Design and Data Analysis

This experiment used a pretest, initial posttest, and delayed posttest design.
Attitudes toward binge drinking were measured during the three tests and analyzed using
a 3 (three tests) x 2 (essay/no essay) AN OVA. At pretest, attitudes toward binge
drinking before exposure to a persuasive message and issue involvement were measured.
At posttest, attitudes toward binge drinking after exposure to a persuasive message, issue

involvement, and source credibility were measured. After 21 days, participants were

24

randomly assigned to either an essay or no essay conditions to manipulate the

independent variable. The essay condition asked people to write an essay about the

importance of “being healthy.” At delayed posttest, their attitudes toward binge

drinking were measured.

Participants

Forty-two students at a major midwestem university participated in the study.

Two students were unavailable at follow-up. Therefore, forty undergraduate students’

responses were retained for analyses. Participation was voluntary and the investigator

assured conﬁdentiality.

Of these participants, 21 students participated in the essay condition (E), and 19

students participated in the control condition (C). The sample was composed of 28

females (70%) and 12 males (30%). They ranged in age from 18 to 28 years, with an

average age of 20.65 years (SD=1.72). Among them, 35 were Caucasian (89.7%), 2

were Asian (5.1%), 1 was African American (2.6%), and 1 was Hispanic (2.6%). The

majority of participants were juniors (67.5%), followed by sophomores (15%), seniors

(12.5%), and freshmen (5%). Most participants were familiar with (M = 6.38, SD =

0.87) or had heard of binge drinking (M = 6.78, SD = 0.62) before the study.

25

Statistics show no evidence that participants’ age, t (38) = -0.43, p>.05, level of
knowledge, t (38) = 0.32, p>.05, or familiarity with binge drinking, t (38) = 1.21, p>.05
differed between conditions. In addition, there was no evidence that gender
signiﬁcantly differed across conditions ()8 = .04, p>.05).

Instrumentation

Conﬁrmatory factor analyses (Hunter & Gerbing, 1982) were conducted for all
scales to test their content validity. Items passing internal consistency and parallelism
tests were retained.

Attitudes toward the issue. Attitudes toward binge drinking were measured three
times: at pretest, initial posttest, and delayed posttest. The items were bad/good,
unfavorable/favorable, negative/positive, disagreeable/agreeable, foolish/wise,
unacceptable/acceptable, and wrong/right. The reliability of the attitude scale at pretest
was a = 0.94 (M = 3.50, SD = 1.20), a = 0.95 at initial posttest (M = 3.11, SD = 1.17),
and a = 0.95 at delayed posttest (M = 2.99, SD = 1.16).

Issue involvement. The issue involvement scale was measured twice: at pretest
and initial posttest (pretest: a = 0.88, M = 4.49, SD = 1.37; initial posttest: a = 0.96, M =
4.95, SD = 1.30). This variable was measured as a control variable.

Source credibility. Source credibility has two dimensions, expertise and

26

trustworthiness. The reliability of the expertise and trustworthiness scales were 0.85 (M

= 4.72, SD = 1.33) and 0.81 (M = 4.92, SD = 1.23), respectively. This variable served

as a control variable.

Results of Study 2

Test of the Hypothesis

The control variables (issue involvement and source credibility), gender, race, and

year in college were not signiﬁcantly associated with the dependent variable (attitudes).

Therefore, they were not considered in the analyses.

H1: If there is no reinforcement, previously formed:attitudes jitter exposure to g

persuasive message are likely to decay. However, if there is reinforcement, previously

formed attitudes are not likely to decaL

Table 4 presents the means and standard deviations for attitudes toward binge

drinking at the three tests by essay/no essay conditions. Results show that, in general,

participants initially had negative attitudes toward binge drinking (M = 3.50, SD = 1.20 at

pretest), but they changed their attitudes in a further negative direction after exposure to a

persuasive message (M = 3.11, SD = 1.17 at initial posttest, t(38) = -4.52, p<.001).

ANOVA was conducted to examine the reinforcement effect on attitude decay,

and results are summarized in Table 5. Results show a signiﬁcant interaction effect of

27

the essay/no essay conditions among the three tests, F (2, 74) = 9.62, p<.001, partial eta-
squared = 0.21. Figure 5 indicates that this interaction may be attributed to the differing
pattern of decay graph (the line between attitude at initial posttest and delayed posttest) of
the essay condition compared with the delay graph of the no essay condition. Here,
attitudes in the essay condition changed less favorable toward binge drinking from the
initial posttest to delayed posttest. For the no essay condition, attitudes changed more
favorable toward binge drinking from the initial posttest to delayed posttest.

The interaction effect was further explored by examining the pattern of attitude
scores between the essay and no essay conditions. Analyses showed that the essay
condition’s attitudes were signiﬁcantly less favorable than the no essay condition’s, both
at initial posttest, t(38) = 2.45, p<.05, and at delayed posttest, t(38) = 4.29, p<.001. In
addition, in the no essay condition, attitudes changed signiﬁcantly more favorable from
initial posttest to delayed posttest, t(18) = -3.13, p<.05, and the attitude score at delayed
posttest did not differ from the attitude score at pretest, t(17) = -1.88, p>.05, suggesting
that previously formed attitudes toward binge drinking decayed to initial positions with
no essay. However, in the essay condition, the attitude score at initial posttest did not
differ from the attitude score at delayed posttest, t(20) = 0.95, p>.05, and attitudes at

delayed posttest were signiﬁcantly less favorable than attitudes at pretest, t(20) = 3.14,

28

p<.05, showing that previously formed attitudes toward binge drinking did not decay to
initial positions with the use of an essay.
Method of Study 3
Topic Selection and the Persuasive Message
This study focused on the decay in attitudes newly formed in the positive

direction after exposure to a persuasive message. It employed the topic of apheresis
donation2 to produce not well-formed initial attitudes, and to allow positive attitude
formation. Lee (2004) showed that aphaeresis donation is such a new topic that most
people are not familiar with it and, as a result, do not have well-formed attitudes toward
the issue. The persuasive message was designed to promote apheresis donation and
emphasized the help others value, such that aphaeresis donation would help patients and
their families.
Design and Data Analysis

This experiment used a pretest, initial posttest, and delayed posttest design.
Attitudes toward apheresis donation were measured in the three tests and analyzed using
3 (three tests) x 2 (essay/ no essay) ANOVA. At pretest, attitudes toward apheresis
donation before exposure to a persuasive message and issue involvement were measured.

At posttest, attitudes toward apheresis donation after exposure to a persuasive message,

29

issue involvement, and source credibility were measured. After 21 days, participants
were randomly assigned to either the essay or no essay condition to manipulate the
independent variable. The essay condition asked people to write an essay about the
importance of “helping others.” At delayed posttest, their attitudes toward apheresis
donation were measured.

Participants

Forty-six students at a major midwestem university participated in the study.

Five students were unavailable at follow-up. Therefore, forty-one undergraduate
students’ responses were retained for analyses. Participation was voluntary and the
investigator assured conﬁdentiality.

Of these participants, 20 students participated in the essay condition (E), and 21
students participated in the control condition (C). The sample was composed of 22
females (53.7%) and 19 males (46.3%). They ranged in age from 18 to 35 years, with
an average age of 20.80 years (SD=2.53). Among them, 37 were Caucasian (90.2%), 3
were African American (7.3%), and one person was unidentiﬁable (2.4%). The
majority of participants were juniors (53.7%), followed by seniors (24.4%), sophomores

(9.8%), freshmen (7.3%), and others (4.9%). Most participants were not familiar with

30

(M = 1.63, SD = 1.26) or had not heard of apheresis donation (M = 1.59, SD = 1.22)
before the study.

Statistics show no evidence that participants’ level of knowledge, t (1, 39) = -1.84,
p>.05, familiarity with apheresis donation, t (1, 39) = -1.62, p>.05, or age, t (1, 39) = 0.62,
p>.05, differed between conditions. In addition, there was no evidence that gender
signiﬁcantly differed across conditions (x2 = 2.02, p>.05).

Instrumentation

Conﬁrmatory factor analyses (Hunter & Gerbing, 1982) were conducted for all
scales to test their content validity. Items passing internal consistency and parallelism
tests were retained.

Attitudes toward the issue. Attitudes toward apheresis donation were measured
three times: at pretest, initial posttest, and delayed posttest. The items were bad/ good,
unfavorable/favorable, negative/positive, disagreeable/agreeable, foolish/wise,
unacceptable/acceptable, and wrong/right. The reliability of the attitude scale at pretest
was a = 0.92 (M = 4.13, SD = 0.36), a = 0.94 at initial posttest (M = 6.16, SD = 0.83),
and a = 0.99 at delayed posttest (M = 5.97, SD = 1.08).

Issue involvement. Issue involvement scale was measured twice: at pretest and

initial posttest (pretest: a = 0.90, M = 4.09, SD = 0.79; initial posttest: or = 0.90, M = 5.98,

31

SD = 0.88). This variable was measured as a control variable.

Source credibility. Source credibility has two dimensions, expertise and
trustworthiness. The reliability of the expertise and trustworthiness scales are 0.77 (M =
5.21, SD = 1.09) and 0.80 (M = 5.26, SD = 0.98) respectively. This variable served as a
control variable.

Results of Study 3
Test of the Hypothesis

The control variables (issue involvement and source credibility), gender, race, and
year in college were not signiﬁcantly associated with the dependent variable (attitudes).
Therefore, they were not considered in the analyses.

H2: Despite the presence or @sence of reinforcement. newly formed ammdes. after
exposure to apersuasive message, are not likely to decay.

Table 6 presents the means and standard deviations for attitudes toward apheresis
donation in the three tests by essay/no essay conditions. Results show that, in general,
participants initially did not have well-formed attitudes toward apheresis donation (M =
4.13, SD = 0.36 at pretest),3 but they had newly formed attitudes in the positive direction
after exposure to a persuasive message (M = 6.16, SD = 0.83 at initial posttest, t(40) =

14.52, p<.001).

32

ANOVA was conducted to examine the reinforcement effect on attitude decay and
results are summarized in Table 7. Results show no signiﬁcant interaction effect of
essay/no essay conditions in the three tests, F (2, 78) = 0.16, p>.05, partial eta-squared =
0.004, and no signiﬁcant main effect of essay/no essay, F (2, 39) = 1.79, p>.05, partial
eta-squared = 0.04. There was, however, a main effect of the three tests, F (2, 78) =
124.76, p<.001, partial eta-squared = 0.76. Figure 6 indicates that this main effect may
be attributed to an attitude score increase from pretest to initial posttest. The main effect
of the three tests was further explored by examining the pattern of attitude scores among
the three tests. Analyses showed that people had signiﬁcantly more positive attitudes
toward apheresis donation at initial posttest than at pretest, t(40) = -14.52, p<.001, and
attitudes signiﬁcantly changed less favorably from the initial posttest to delayed posttest,
t(40) = 2.04, p = 0.045. Furthermore, people showed signiﬁcantly more favorable
attitudes at delayed posttest than at pretest, t(40) = -10.36, p<.001. These ﬁndings
suggest that newly formed attitudes toward apheresis donation after exposure to a
persuasive message did not decay to initial attitudinal positions, regardless of essay/no

essay conditions, but they did become less favorable over twenty-three days.

33

CHAPTER 3

DISCUSSION

Persuasion scholars acknowledge that persuasive effects decay over time, but no
research has been conducted to examine what kinds of attitude are subject to decay
following persuasion. This study suggested that persuasive effects produce two types of
attitudes, previously formed attitudes and newly formed attitudes. The goal of this
study was to examine attitude decay in previously formed attitudes and newly formed
attitudes after exposure to a persuasive message. It also examined if reinforcement
affects attitude decay.

This study was composed of three single factor experiments and used a pretest,
initial posttest, and delayed posttest design. Study 1 employed the topic of increasing
university tuition to examine decay of previously formed attitudes, which were modiﬁed
in the positive direction after exposure to a persuasive message. Study 2 utilized the
topic of binge drinking to investigate attitude decay of previously formed attitudes, which
were changed in the negative direction after exposure to a persuasive message. Study 3
used the topic of apheresis donation to examine attitude decay of newly formed attitudes,

which were newly formed in the positive direction after exposure to a persuasive message.

34

At pretest, participants’ initial attitudes toward the given issue before exposure to a

persuasive message and issue involvement were assessed, and then, they received a

persuasive message designed to induce previously formed or newly formed attitudes in

the desired direction. The context of the message focused on a certain value. At initial

posttest, participants were asked to rate their attitudes toward the issue, issue involvement,

and source credibility after exposure to a persuasive message. After 21 days,

participants were randomly assigned to either the essay or no essay condition. In the

essay condition, participants were asked to compose an essay about the importance of the

value presented in the persuasive message, to strengthen the link of value to their

previously formed or newly formed attitudes. At delayed posttest, participants in both

essay and no essay conditions were asked about their attitudes toward the issue. In the

remainder of this chapter, discussion of the key ﬁndings and implications for persuasive

message design are presented.

Previously F ormed Attitudes, Newly F armed Attitudes, Reinforcement, and Attitude

Decay

Hypothesis I predicted that previously formed attitudes would decay if there was

no reinforcement, but would not decay if there was reinforcement. This hypothesis was

consistent with the ﬁndings from Studies 1 and 2. In Study 1, people whose attitudes

35

were modiﬁed in the positive direction toward increasing university tuition, showed that
their modiﬁed attitudes decayed to initial negative positions toward the topic without
reinforcement. With reinforcement, however, their attitudes did not decay. In Study 2,
people whose attitudes were modiﬁed further in the negative direction toward binge
drinking after exposure to a persuasive message, showed that their modiﬁed attitudes
decayed to their initial positions toward the topic without reinforcement. Similar to
Study 1, in Study 2, participants’ attitudes did not decay with reinforcement.

Hypothesis 2 predicted that newly formed attitudes would not decay, regardless of
reinforcement because newly formed attitudes are true initial attitudes. The data in
Study 3 showed evidence consistent with the prediction. With or without reinforcement,
newly formed attitudes toward apheresis donation did not decay to their initial attitudinal
positions.

The ﬁndings indicate ﬁrst, not all attitudes are subject to decay. Only previously
formed attitudes are subject to decay, but not newly formed attitudes. In terms of the
double comparison model of attitude change (Hunter et al., 1976), people who have
existing attitudes before receiving persuasive communication are likely to have attitude
decay. Their motivation to return to their initial attitudinal position toward the topic

might be to reduce discomfort in their belief system. If they see unbalance between held

36

beliefs and inﬂuenced beliefs (by persuasive communication), they are going to endorse

their held beliefs. The results also conﬁrm the information integration model suggested

by Kaplowitz et al. (1986), showing that information used to form initial attitudes has

more inﬂuence on the ﬁnal attitude judgments than persuasive message in the previously

formed attitudes.

This is an important ﬁnding in light of the fact that many persuasive advertising

messages are designed to form new attitudes toward new products. Attitude decay, then,

may be considered more importantly when persuasion topics are familiar and social

issues such as organ donation, smoking, drinking, and voting. People are likely to be

exposed to those issues and might establish attitudes that are hard to maintain change, if

successfully changed.

Second, the results show that reinforcement affects attitude decay. The absence

of attitude decay in the presence of reinforcement in previously formed attitudes conﬁrms

the contention suggested by Higgings and King (1981). They argued that a strong link

of attitude to another construct affects attitude accessibility, and consequently strengthens

attitudes. The method of writing an essay about the proposed values as reinforcement,

may make the values salient among other beliefs associated with the topic; hence,

strengthen the link of those values to the formed attitudes.

37

The reinforcement effect also might be explained by match/mismatch between
attitude-relevant cues and information used in attitude formation. Sepupta and
Fitzsimons (2004) showed that attitude-relevant cues on delayed judgments affected the
stability of brand attitude. They suggested that, when there are attitude-relevant cues on
delayed judgments, people compare the inputs used to form initial attitudes with the
available attitude-relevant cues at delay. They found that people had stable attitudes
toward a brand when there was a match in the comparison. In this study, people also
might compare values they think of when they write an essay with those they think of
when forming their attitudes toward increasing university tuition and binge drinking.
They showed attitude stability because they would ﬁnd that the information they
compared was matched.

Attitude Fluctuation in Newly Formed Attitudes: Not Decay but Variation

Interestingly, however, it was found that newly formed attitudes after exposure to
a persuasive message were not consistent, and instead, decreased in the direction of the
initial attitudinal positions. The change was not, obviously, decay to initial positions.

It is possible that people needed more information to maintain their newly formed
attitudes toward apheresis donation. For example, although participants’ newly formed

positive attitudes toward apheresis donation as a result of external message comparison

38

were consistent with their belief systems, they may later consider the pain involved in the
apheresis donation procedure-- a fact they did not consider during the external message
comparison. The lack of sufficient information to support their newly formed attitudes
would likely cause an attitude decrease over time. Another possibility to consider is that
participants might think of blood donation in deciding their attitudes toward apheresis
donation, since these topics are similar in terms of donating blood, and blood donation is
familiar to them. Their attitudes toward blood donation, therefore, would affect
attitudes toward apheresis donation at initial posttest. However, over time, they may
think less of blood donation and think more of other factors associated with apheresis
donation, resulting in an attitude decrease toward apheresis donation. Future research
should examine these two alternative explanations.
Implication to Persuasive Message Design

The results of the study provide an implication to persuasive message designers
who wish to produce a successful persuasive effect for a long time. As was done in the
current study, with a message focusing on prevalent values that people generally consider
important, it will be effective because there will be a high chance of encountering those

values in daily life. Message recipients, then, might have high likelihood of thinking

39

about these values, activate the connection between formed attitude and related values in

their mind, and be likely to keep their modiﬁed/formed attitudes for a long time.

Limitations

The main limitation of this study is the small sample size in each study. Due to

the small sample size, the statistics calculated in this study might be less accurate than

those of a larger sample size, owing to sampling error. Therefore, making

generalizations about other samples might be problematic. Another limitation is

possible environmental effects in regards to the topic of binge drinking. At the time of

data collection, there was a campus-wide campaign on binge drinking prevention.

Appendix C shows the contents of advertisements and posters run in the local newspaper

to prevent binge drinking in the period of data collection for the study of attitude decay in

binge drinking. The campaign might have inﬂuenced subjects’ initial negative attitudes

toward binge drinking, the additive effect on reinforcement, and consequently, attitude

decay. Future research needs to be conducted to examine decay in binge drinking

attitudes controlling environmental effects.

Future Research

Studies that address the aforementioned limitations are areas for future research.

It would be beneﬁcial to replicate this type of study, using similar or different topics and

40

a larger sample size in order to understand further attitude decay following persuasion.

Additionally, studies need to be completed in which there is an examination of the effect

on the need for knowledge toward apheresis donation and attitude change. For example,

an examination of attitude decrease with a comparison of people who are provided

sufficient knowledge toward apheresis donation after attitude formation, versus those

who are not provided any more knowledge could provide additional information. In

addition, a study needs to examine whether attitudes toward blood donation inﬂuenced

attitudes toward apheresis donation, yielding further results on how people express their

attitudes toward unfamiliar topics. Future research also needs investigate other types of

reinforcement that may prevent attitude decay. Finally, studies are needed that examine

the effect of initial attitudinal position (i.e., negative, positive, or neutral) and the

direction of the persuasive effort (i.e., positive or negative attitude modiﬁcation) on

attitude decay. These ﬁndings may give clues to conditions where attitude decay occurs

most frequently.

Conclusion

This study examined decay of previously formed and newly formed attitudes with

reinforcement. Because research on the topic at hand has not previously been

41

completed, this study is somewhat exploratory in that it attempted to begin to look at

attitude decay phenomena in persuasion. However, the ﬁndings may be useful in

understanding this phenomenon, which would inform persuasion scholars as well as

persuasion practitioners that previously formed attitudes tend to decay without

reinforcement. Additionally, because these questions have not been answered in the

persuasion literature, results of this study may help with the development of a conceptual

framework to guide future research in attitude decay and the reinforcement of attitudes.

This study contributes to the persuasion literature by exploring the way that previously

formed and nery formed attitudes decay, showing that strengthening a link of formed/

newly formed attitudes to associated values is a way to reduce attitude decay.

42

ENDNOTES

lLevel of knowledge and familiarity with the given issue were measured by 7-
point Likert scales, anchored by strongly disagree and strongly agree. The smaller the
number, the less knowledge people have, and the less they have heard about the issue.

2 Unlike a “whole blood” donation, in which a unit of whole blood is removed
from the body, the aphaeresis process connects the body to a machine that removes only
one or two components from the blood, and then returns the rest of the blood to the body.

3 Most studies report that an attitude score of 4 on a 1-7 Likert scale, is a neutral
position, indicating that people do not have well-formed positive or negative attitudes.

The current study also interpreted an attitude score of 4 in this way.

43

APPENDICES

44

APPENDIX A: PERSUASIVE MESSAGES

PERSUASIVE MESSAGE FOR INCREASING UNIVERSITY TUITION

The Michigan State University Board of Trustees is considering a proposal to

increase tuition at Michigan State University. The decision will be made based on the

students’ vote for this proposal. The tuition increase would be used to hire more faculty

members for each department and students would be offered many smaller sized classes.

Ultimately, the increased tuition would help students to succeed. Smaller sized

 

classes would allow teachers pay more attention to each student’s performance which

would result in better skills and grades. Better skills and grades will increase the

possibility of ﬁnding a good job after graduation. Consequently, MSU students would

be offered quality education and would be in a better position in the job market compared

to other university students elsewhere. Therefore, MSU students need to vote for this

proposal.

45

PERSUASIVE MESSAGE FOR BINGE DRINKING

The Michigan State University Board of Trustees is considering a proposal to
develop campus-wide programs to decrease Binge Drinking at Michigan State University.
The decision will be made based on the students’ vote for this proposal. Binge Drinking
is consuming ﬁve or more drinks at a single sitting for men and four or more for women.
National studies show that almost 45% of college students engage in binge drinking and
it causes drunk driving, vandalism on campus, sexual assaults, academic failure, and
physical illness. Also, studies show that binge drinkers suffer guilt, regret, and physical
injuries as consequences. According to MSU Police Department, most of crimes and
accidents occurred on and off campus are related to students’ binge drinking and the
offenders and the victims suffer mental and physical illness for a long time.

Ultimately, the campus-wide programs would help students stay physically and
psychologically hm. Studies show that one of the reasons students engage in binge
drinking is lack of refusal skills for situations in which peer pressure is high. The
programs will teach MSU students detailed refusal skills in realistic settings so that they
can stay mentally and physically healthy while maintaining good relationships with

friends. Therefore, MSU students need to vote for this proposal.

46

PERSUASIVE MESSAGE FOR APHERESIS DONATION

The Michigan State University Olin Health Center and Sparrow Hospital are

asking for apheresis donations from Michigan State University students. Apheresis

donation is the donation of one or two components from blood and the rest of the blood

returns to the body. According to Olin Health Center and Sparrow Hospital, 35 pints of

apheresis components are needed for some Michigan State University students. Also,

families of these students are in great grief and long for donations to help their daughters

and sons in need.

Ultimately, apheresis donation is for helping others in need. If some MSU

students do not get the apheresis component they will die because their surgeries have

been delayed while waiting for apheresis components replenishment. Donating

apheresis components will save fellow MSU students’ lives as well as their families’.

Therefore, MSU students needed to donate their apheresis components.

47

APPENDIX B: BINGE DRINKING ADVERTISEMENTS

0 Two sideline advertisements of “Celebrate the Spartan Way” were run in the local

newspaper during data collection.

1) One read:

Most MSU Students:

Watch Spartan Sports

Have 0-5drinks when they party

Keep track of the number of drinks they consume

2) The other one read:

When MSU Students drink, most do one or more of the following:

Pace their drinks

Avoid drinking games

Keep track of the number of drinks they have consumed

48

0 There were also two posters ran as advertisements in the regular edition of local

newspaper twice weekly during data collection

1) One read:

"What Do Spartans Do?”

Most Spartans:

Listen to Music to manage stress (97%)

I Stay with the same group of friends the entire time when they go out drinking

(78%)

I Play recreational sports (57%)

I Study at least 4 evenings per week (77%)

I Consume 5 or fewer drinks when they party (63%)

I Work for wages during the academic year (61%)

I Keep track of the number of drinks they have consumed

2) The other one read:

"What Do Spartans Think?”

Most Spartans think:

I Academic Study is a priority.

49

83% of MSU Students report academics are their ﬁrst priority.

Drinking should be done in Moderation.

94% of MSU Students report they disapprove of drinking to the point

of passing out; 55% disapprove of other students getting "wasted."

Drinking is a personal decision.

96% of MSU students disapprove of pressuring someone to drink more

than they want to.

50

APPENDIX C: MEASUREMENT RETAINED AFTER CFA

O Attitude Items

(1)Bad

Good

 

(2) Unfavorable

Favorable

 

(3) Negative

Positive

 

(4) Disagreeable

Agreeable

 

(5) Foolish

Wise

 

(6) Unacceptable

Acceptable

 

(7) Wrong

Right

 

0 Issue Involvement Items

(1) Unimportant

Important

 

(2) Does not matter to me

Matters to me

 

(3) Insigniﬁcant

Signiﬁcant

 

(4) Of no concern

Of much concern

 

51

O Source Credibility Items

(1) Not knowledgeable

 

(2) Not expert

 

(3) Untrustworthy

 

(4) Unbelievable

 

52

Knowledgeable

Expert

Trustworthy

Believable

BIBLIOGRAPHY

53

BIBLIOGRAPHY

Abelson, R. P. (1988). Conviction. American Psychologist, 43, 267-275.

Anderson, N. H. (1981). FoundLions of information integration theogy. New York:
Academic Press.

Bassili, J. N., & Fletcher, J. F. (1991). Response-time measurement in survey research: A
method for CATI and a new look at nonattitudes. Public Opinion Quarterh, 55,
331-346. '

Bizer, G. Y., & Krosnick, J. A. (2001). Exploring the structure of strength-related attitude
features: The relation between attitude importance and attitude accessibility.
Journal of Personality and Social Psychology, 81, 566-586.

Boninger, D. S., Krosnick, J. A., Berent, M. K., & Fabrigar, L. R. (1995). The causes and
consequences of attitude importance. In R. E. Petty & J. A. Krosnick (Eds),
Attitude stregth: Antecedents and conseqpences (pp. 159-190). Mahwah, NJ:
Erlbaum.

Bradbum, N. M., & Caplovitz, D. (1965). Reports onﬂpniness: A pilot study of behavior
related to merﬂ health Chicago: Aldine.

Burgoon, J. K., Birk, T., & Pfau, M. (1990). Nonverbal behaviors, persuasion, and
credibility. Human Communication Research, 17, 140-169.

54

Cacioppo, T. J ., Petty, R. E., & Green, T. R. (1989). Attitude structure aad function: From
the tripartite to the homeostasis model of attitudes. In A. R. Pratkanis, S. J.
Breckler, & A. G. Greenwald (Eds.), Attitude structure and function (pp. 275-309).
Hillsdale, NJ: Erlbaum.

Calder, B. J ., Insko, C. A., & Yandell, B. (1974). The relation of cognitive and memorial
processes to persuasion in a simulated jury trial. 10ml of Applied Socaal
Psychology, 4, 62-93.

Cantril, H. (1946). The intensity of an attitude. Jourpal of Abnormal and Social
Psychology, 41, 1-12.

Chaiken, 8., & Baldwin, M. W. (1981). Affective-cognitive consistency and the effect of
salient behavioral information on the self-perception of attitudes. Journal of
Personality and Social Psychology, 41, 1-12.

Chaiken, S., & Eagly, A. H. (1983). Communication modality as a determinant of
persuasion: The role of communicator salience. Journal of Personality and Social
Psychology, 45, 241-256.

Cook, T. D., & Flay, B. R. (1978). The persistence of experimentally induced attitude

change. In L. Berkowitz (Ed.), Advances in experimenjtal social psychology (Vol.
11, pp. 1-57). New York: Academic Press.

Dillard, J. P. (1993). Persuasion past and present: Attitudes aren’t what they used to be.
Communication Monoggaphs, 60, 90-97.

Entrnan, R. M. (1993). Framing: Toward clariﬁcation of a fractured paradigm. Journal of
Communication. 43, 51-58.

55

Fazio, R. H. (1986). How do attitudes guide behavior? In R. H. Sorrentino & E. T.
Higgins (Eds.), The handbook of motivation aad cognition: Foundations of social
behavior (pp. 204-243). New York: Guilford Press.

Fazio, R. H. (1995). Attitudes as object-evaluation association: Determinants,
consequences, and correlates of attitude accessibility. In R. E. Petty & J. A.
Krosmck (Eds.), Attitude strergth: Antecedents and consequences (pp. 247-282).
Mahwah, NJ: Erlbaum.

Green, J. O. (1984). A cognitive approach to human communication: An action assembly
theory. Communication Monoggaphs, 51, 289-306.

Haugtvedt, C. P., & Strathman, A. (1990). Situational personal relevance and attitude
persistence, Advances in Consumer Research, 1 7, 766-769.

Haugtvedt, C. P., & Wegener, D. T. (1994). Message order effects in persuasion: An
attitude strength perspective. Journal (1 Consumer Research 21, 205-218.

Higgins, E. T., & King, G A. (1981). Accessibility of social constructs: Information
processing consequences of individual and contextual variability. In N. Cantor &

J. Kihlstrom (Eds.), Personality, cognition, and social interaction (pp. 69-122).
Hillsdale, NJ: Erlbaum.

Holland, R. W., Verplanken, B., & van Knippenberg, A. (2003). From repetition to
conviction: Attitude accessibility as a determinant of attitude certainty. Journal of
Experimental Social Psychology, 39, 594-601.

Homer, P. M., & Kahle, L. R. (1990). Source expertise, time of source identiﬁcation and
involvement. Journal of Advertising. 19, 30-39.

56

Hosman, L. A., Huebner, T. M., & Siltanen, S. A. (2002). The impact of power-of-speech
style, argument strength, and need for cognition on impression formation,

cognitive responses, and persuasion. Journal of Language and Social Psychology,
21, 361-380.

Hunter, J. E., Levine, R. L, & Sayers, S. E. (1976). Attitude change in hierarchical belief
systems and its relationship to persuasibility, dogmatism, and rigidity. Human
Communication Research, 3, 3-28.

Judd, C. M., & Johnson, J. T. (1981). Attitudes, polarization, and diagnosticity: Exploring
the effects of affect. Journg of Personality and Social Psychology, 41, 26-36.

Kanwar, R., Grund, L.,& Olson, J. C. (1990). When do measures of knowledge measure
what we think they are measuring? Advances in Coasumer Research, 17, 603-608.

Kaplowitz, S. A., Fink, E. L., Armstrong, G. B., & Bauer, C. L. (1986). Message
discrepancy and the persistence of attitude change: Implications of an information
integration model. Journal of Experimental Social Psychology, 22, 507-530.

Kendall, P. (1954). Conﬂict and mood: Factors affecting stability of response. Glencoe,
IL: Free Press.

Krosnick, J. A., & Schuman, H. (1988). Attitude intensity, importance, and certainty and
susceptibility to response effects. Journal of Personalig and Social Psychology.
54, 940-952.

Krosnick, J. A. (1989). Attitude importance and attitude accessibility. Personality and
Sociﬂsychology Bulletin, 15, 297-308.

57

Krosnick, J. A., & Petty, R. E. (1995) Attitude strength: An overview. In R. E. Petty & J.
A. Krosnick (Eds.), Attitude strength: Antecedents and coraecyuences (pp. 1-24).
Mahwah, NJ: Erlbaum.

Lee, S. Y. (2004). Value-expressive attitude formation processes as a function of
familiarity and involvement. Unpublished paper.

Leippe, M. R., & Elkin, R. A. (1987). When motives clash: Issue involvement and
response involvement as determinants of persuasion. Journal of Personality aria
SocialﬁychologyL 52, 269-278.

Leventhal, H., & Niles, P. (1965). Persistence of inﬂuence for varying durations of
exposure to threat stimuli. Psychological Reports, 16, 223-233.

Maheswaran, D., & Meyers-Levy, J. (1990). The inﬂuence of message framing and issue
involvement. Jomal of MarketiagResearch, 27, 361-367.

Norman, R. (1975). Affective-cognitive consistency, attitudes, conformity, and behavior.
Joumi of Personality and Social Psychology, 32, 83-91.

O'Keefe, D. J. (1990). Factors inﬂuencing persuasive effects. Persuasion: TheogI and
research. Thousand Oaks, CA: Sage.

Osgood, C. E., & Tennenbaum, P. H. (1955). The principle of congruity in the prediction
of attitude change. Psychological Revievg 62, 42-55.

Petty, R. E., & Cacioppo, J. T. (1986). Communication and persuasion: Central and
ﬁripheral routes to attitude changa New York: Springer-Verlag.

58

Petty, R. E., Schumann, D. W., Richman, S. A., & Strathman, A. J. (1993). Positive mood
and persuasion: Different roles for affect under high-and low-elaboration
conditions. Joummf Personality and Social Psycholcgy 64, 5-20.

Pfau, M., Roskos-Ewoldsen, D., Wood, M., Yin, S., Cho, J., Lu, K., & Shen, L. (2003).
Attitude accessibility as an alternative explanation for how inoculation confers
resistance. Communication Monoggaphs, 70, 39-51.

Poole, M. S., & Hunter, J. E. (1979). Change in hierarchically organized attitudes. In D.
Nimmo (Ed.), Communication Yearbook 3 (pp. 157-176). New Brunswick, NJ:
Transaction.

Regan, D. T., & F azio, R. H. (1977). On the consistency between attitudes and behavior:
Look to the method of attitude formation. Journal of Experimental Social
Psychology, 13, 28-45.

Roese, N. J ., & Olson, J. M. (1994). Attitude importance as a function of repeated attitude
expression. Journal of Experimental Social Psycholagy, 30, 39-51.

Roskos-Ewoldsen, D. R., & Fazio, R. H. (1992). The accessibility of source likability as a
determinant of persuasion. Personality and Social Psychology Bulletin, 18, 19.

Roskos-Ewoldsen, D. R., Arpan-Ralstin, L., & Pierre, J ., ST. (2002). Attitude
accessibility and persuasion: The quick and the strong. In J. P. Dillard, & M.

Pfau (Eds.), The persua_sion handboak: Developments in theory and practice (pp.
39-62). Thousand Oaks, CA: Sage.

59

Schuman, H., & Presser, S. (1981). Questions and answers in attitude surveys:
Experiments on gestion form. wording, and context. San Diego, CA: Academic
Press.

Sepupta, J ., & Fitzsimons, G (2004). The effect of analyzing reasons on the stability of

brand attitudes: A reconciliation of opposing predictions. Journal of Consumer
Research, 31, 705-711.

Sherif, C. W., Sherif, M., & Nebergall, R. E. (1965). Attitude and attitude change.
Philadelphia: Saunders.

Sherif, C. W., Kelly, M., Rogers, H. L., Sarup, G, & Tittler, B. I. (1973). Personal
involvement, social judgment, and action. Journal of Personality and Social
Psychology, 27, 311-328.

Simon, B. L., & Ruth, J. A. (1998). Is a company known by the company it keeps?
Assessing the spillover effects of brand alliances on consumer brand attitudes.
Journal of Marketiag Research, 35, 30-42.

Smith, E. R. (1994). Procedural knowledge and processing strategies in social cognition.
In R. S. Wyer, Jr., & T. K. Srull (Eds.), Handbook of sopial cognition: Vol. 1.
Basic processes (pp. 99-151). Hillsdale, NJ: Lawrence Erlbaum Associates, Inc.

Smith, S. M., & Shaffer, D. R. (1995). Speed of speech and persuasion: Evidence for
multiple effects. Personality and Social Psycholongulletin, 21, 1051-1060.

Stouffer, S. A., Guttman, L., Suchman, E. A., Lazarsfeld, P. F., Star, S. A., & Clausen, J. A.
(1950). Measurement and prediction Princeton, NJ: Princeton University Press.

60

Witte, K. (1992). Putting the fear back into fear appeals: The Extended Parallel Process
Model. Communication Monoggaphs, 59, 329-349.

61

Table 1. Attributes of Attitude Strength

 

 

Categories Attributes

Feature of attitude itself Extremity
Intensity
Cognitive structure associated Attitude accessibility
'th ttitud d tti d
wr. a e an a tu e Knowledge
object
Affect-cognitive consistency
Cognitive process Amount of elaboration
Direct experience

Subjective beliefs about Personal importance

attitude and attitude object ,
Centralrty to the self-concept

Attitude certainty

Latitudes of rejection and noncommitment

 

62

Table 2. Means and Standard Deviations for Attitudes toward Increasing University
Tuition at the Three Tests by Essay and No Essay Condition

 

 

Pretest Initial Posttest Delayed Posttest
No essay M = 2.22, M = 3.39, M = 2.41,
(n = 23) SD = .72 SD = 1.24 SD = .95
Essay M = 2.43, M = 3.62, M = 4.16,
(n = 27) SD = .97 SD = .97 SD = .53
All M = 2.34, M = 3.52, M = 3.35,
(n=50) SD=.86 SD=1.10 SD=1.15

 

63

Table 3. ANO VA Summary Table for Attitudes toward Increasing University Tuition

 

 

. . Type III Sum
Source of Variation DF F p
of Squares
Tests of Within-Subjects Eﬂbcts
Tests 38.972 2 36.23 .000
Tests x Essay/No Essay 19.31 2 17.96 .000
Error (Tests) 51.631 96
Tests of Between-Subject Eﬂects
Intercept 1377.64 1 957.313 .000
Essay/No Essay 19.89 1 13.82 .001
Error 69.08 48

 

64

Table 4. Means and Standard Deviations for Attitudes toward Binge Drinking at the
Three Tests by Essay and No Essay Condition

 

 

Pretest Initial Posttest Delayed Posttest
No Essay M = 3.87, M = 3.56, M = 4.24,
(n=l9) SD=1.17 SD=1.14 SD=1.53
Essay M = 3.19, M = 2.70, M = 2.52,
(n=21) SD=1.17 SD= 1.07 SD=.96
All M= 3.50, M=3.11, M=3.34,
(n=40) SD=1.20 SD=1.17 SD=1.52

 

65

Table 5. ANO VA Summary Table for Attitudes toward Binge Drinking

 

 

. _ Type III Sum
Source of Variation DF F p
of Squares
Tests of Within-Subjects Eﬂects
Tests 3.15 2 4.78 .011
Tests x Essay/No Essay 6.34 2 9.62 .000
Error (Tests) 24.39 74
Tests of Between-Subject Eﬂects
Intercept 1307 1 364 .000
Essay/No Essay 34.83 1 9.7 .004
Error 1 32.86 3 7

 

66

Table 6. Means and Standard Deviations for Attitudes toward A pheresis Donation at the
Three Tests by Essay and No Essay Condition

 

 

Pretest Initial Posttest Delayed Posttest
No Essay M = 4.05, M = 6.01, M = 5.82,
(n=21) SD=.15 SD=.89 SD= 1.14
Essay M=4.21, M=6.3l, M=6.13,
(n=20) SD = .48 SD = .76 SD = 1.02
All M=4.13, M=6.16, M=5.97,
(n=41) SD=.36 SD=.83 SD= 1.08

 

67

Table 7. ANO VA Summary Table for Attitudes toward Apheresis Donation

 

Type [11 Sum of

 

Source of Variation DF F p
Squares
Tests of Within-Subjects Eﬂects
Tests 103.22 2 124.76 .000
Tests x Essay/No Essay .13 2 .16 .851
Error (Tests) 32.27 78
Tests of Between-Subject Eﬂects
Intercept 3614.41 1 3 124.76 .000
Essay/No Essay 2.08 1 1.79 .188
Error 45.1 1 39

 

68

Figure I. Speculation of F actors Aﬂecting Attitude Persistence

 

 

Number of arguments

 

 

 

 

Communicator credibility

 

 

 

 

Involvement with the

 

 

 

 

Communication module

 

 

 

 

Cognitive elaboration

 

 

Type of attitudes resulting
from persuasive
communication

(previously formed
attitudes vs. newly formed

 

69

 

Attitude Persistence

 

 

Figure 2. Factors A ﬂeeting Attitude Accessibility

 

Frequent Expression

 

 

Direct Behavioral or Sensory
Experience

 

 

Emotional reaction to an
attitude object

 

 

Hierarchical Arrangement of
Attitudes

 

 

Attitude Uniqueness

 

 

 

 

Attitude Accessibility

 

 

 

 

Link to Other Construct

 

 

Expectation of Evaluating an
Object in the Future

 

 

 

 

Cognitive Elaboration Used
In Evaluation

 

 

Recency of Attitude

 

 

 

70

Figure 3. Design of Study

 

Randomly
assign people
to essay/no
essay condition
after 21 days

 

 

 

 

 

 

 

 

 

PRETEST INITIAL after initial DELAYED

POSTTEST posttest POSTTEST

0 Measure 0 Measure 0 Measure

attitudes toward attitudes attitudes

an issue before toward the toward the

exposure to same issue same

a persuasive after exposure issue 23

message to a days after
persuasive initial
message posttest

 

 

 

 

 

 

 

 

 

71

Figure 4. Attitudes Change toward Increasing University Tuition as a Function of
Essay/No Essay and the Three Tests

4.5

 

,r’
,

4.0 4 ./ "’

3.5 - /

Attitudes

3.0‘

2.5l
E 3 D No Essay

_———~

 

ll
2.0 ' D Essay
Pretest Initial Posttest Delayed Posttest

 

 

 

Tests

72

Figure 5. Attitudes Change toward Binge Drinking as a Function of Essay/No Essay

and the Three Tests

Attitudes

 

4.5

4.0 '

3.5 -

3.0 r

2.5 .

2.0

.\\

’\ _ ——

D No Essay

8 Essay

 

 

Pretest

Initial Pbsttest

Tests

73

 

Delayed Posttest

Figure 6. Attitudes Change toward Apheresis Donation as a Function of Essay/No
Essay and the Three Tests

 

Attitudes

 

 

 

 

D No Essay
3.5 . D Essay
Pretest Initial Posttest Delayed Posttest

Tests

74

   

I"I'Lllllllﬂjlljljjlllljjjljlll