x oil

a Q.“

, x

 

Ar‘ . V874 ‘
5 1.351251)?
[.15 in -13!
. a: a...

 

 

..‘(|zlo\.:..lu¢a
. I! 1h)...i>;ufl1$ Q"!

1....13 .2
1; ||

at” . .. , masfié 5mm»

éfipfifi

.2‘1

 

THESIS

u STATE UNIVERSITY LIBRARIES
l

lllllllllllllllllll lllll 1 \' \\\\\\| r

 

 

 

 

 

 

 

 

\

3 1293 01564 4705

This is to certify that the

dissertation entitled

PERSUADING WOMEN TO TAKE PROTECTIVE
ACTION AGAINST RAPE: A COMMUNICATION MODEL
SPECIFYING THE ROLE OF SIGNIFICANT OTHERS

presented by

KELLY MORRISON

has been accepted towards fulfillment
of the requirements for

Ph.D. degree in Communication

 

 

Major professor

Datemmwfi/ 7 ’6

MSU i: an Affirmative Action/Equal Opportunity Institution O~ 12771

 

 

LIBRARY

Mi f aigan State
University

 

 

 

PLACE IN RETURN BOX to remove thie checkwt from your record.
TO AVOID FINES return on or betore date due.

DATE DUE DATE DUE DATE DUE

MN _,>, :
it I GJ‘_ if, 0 9 fyflffi
l e‘ {Ta ? ’
f .-, - ’r .

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU leAan'inndive Action/Ewe! Opportunity Institution
Wanna-9.1

PERSUADING WOMEN TO TAKE PROTECTIVE
ACTION AGAINST RAPE: A COMMUNICATION MODEL
SPECIFYING THE ROLE OF SIGNIFICANT OTHERS
BY

Kelly Morrison

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Communication

1996

ABSTRACT
PERSUADING WOMEN TO TAKE PROTECTIVE
ACTION AGAINST RAPE: A COMMUNICATION MODEL
SPECIFYING THE ROLE OF SIGNIFICANT OTHERS
BY
Kelly Morrison

This paper presents a general model of indirect
persuasion which targets significant others in close
heterosexual relationships. It is suggested that when one
partner in a relationship may be susceptible to a potential
health threat, she or he may experience feelings of perceived
invulnerability and may not take protective action against
this potential health threat. Consequently, it may be more
effective for health campaigns to target messages toward the
significant others of these people and recommend that the
significant others persuade their partners to take protective
action. The general health model is delineated by several
propositions, and the model is specifically exemplified by
hypotheses regarding the issue of heterosexual rape against
women. A pretest-posttest-follow-up design was employed to
assess attitudes, intentions and behaviors. While the data
were not consistent with many of the hypotheses, several
interesting results emerged" The implications are discussed,

as well as directions for future research.

D-ICATION

This dissertation is dedicated to strong women
everywhere, and three in particular who have influenced my
life: Laura E. Grant, who taught me the strength that comes
from discipline and the beauty of dance; Marion Simpson Lyman,
who taught me the strength that comes from a close, loving
family; and Carol Lyman Morrison, who taught me the strength

that comes from determination, optimism, and never quitting.

iii

ACKNOWLEDGMENTS

This dissertation could not have been completed without
the guidance, encouragement and.assistance of several people.
Thanks to Galen Bodenhausen, whose teaching has guided much of
my thinking; Frank Boster, for his continuing instruction in
the craft of scholarship; Kim.Witte, for sparking my interest
in fear appeal research; and Sandi Smith, whose guidance,
teaching and friendship have gotten.me to this point. Thanks
to Carol and Ross Morrison and Connie and Bruce McCornack, for
their interest, support, and encouragement. Thanks to Neera
Singhvi for her help in coding and entering an ungodly amount
of data. Thanks to Jen Kopfman, Steve Robbins and Marge
Barkman, for keeping me sane and making me laugh. A special
thanks to my three redheads; Colin and Kyle for teaching me
the meaning of the word JOY, and especially to Steve, for
giving :me the encouragement to try, the confidence to

continue, and the support to finish.

iv

TABLE OF CONTENTS

LIST OF TABLES . . . . . . . . . . . . . . . . . . . . . .viii
INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . 1
Overview of the Problem . . . . . . . . . . . . . . . 1

The Prevalence of Rape . . . . . . . . . . . . . 1

Women’s Perceived Invulnerability to Rape . . . . 3
Dissertation Overview . . . . . . . . . . . . . . . . 5
LITERATURE REVIEW . . . . . . . . . . . . . . . . . . . . 7
The Just World Hypothesis . . . . . . . . . . . . . . 7
Perceived Invulnerability . . . . . . . . . . . . . . 8

Fear Control Processes . . . . . . . . . . . . . . . 9

El . . . . . . . . . . . . . . . . . . . . . . . . . 10

Hz . . . . . . . . . . . . . . . . . . . . . . . . . 10

H3 . . . . . . . . . . . . . . . . . . . . . . . . . 11
Significant Others and Preventive Health . . . . . . 11

The Theory of Reasoned Action . . . . . . . . . . . . 12
Social Exchange Theories . . . . . . . . . . . . . . 14

H4 . . . . . . . . . . . . . . . . . . . . . . . . . 15
Assessing Relational Closeness . . . . . . .... . . 15
Proposition 1 . . . . . . . . . . . . . . . . . . . . 16

ES . . . . . . . . . . . . . . . . . . . . . . . . . 16

Fear Appeals and Perceived Pear . . . . . . . . . . . 17

Positive relationships .

Negative relationships .

Curvilinear relationships . . .

The Extended Parallel Process Model

Proposition 2

E6 . . . . . . . . . . .
Proposition 3 .

H7 . . . .

Beliefs about Rape, Anger,
Proposition 4 . . . . .
Proposition 5

EB . . . . . .

HS . . .

Proposition 6

H10

METHOD . . .

Procedure . .
Stimulus messages
Instrumentation

Dependent measures

Independent measures .

RESULTS . .

Pretest .
Main Project .

Induction Checks . . . .

Results for Female Respondents

vi

. 18
. 18
. 19
. 20
. 21
. 22
. 22
. 22
. 23
. 25
. 26
. 26
. 26
. 27

. 27

28

. 28

. 29

30

. 35

. 35

36

36

36

37

. 38

Results for Male Respondents

DISCUSSION . . . .
Overview . . . .
Female Respondents

Male Respondents

Parameters and Limitations of

Summary . . . . .
NOTES . . . . . . .
APPENDIX A .
APPENDIX B . . . . . .
APPENDIX C . . . .

LIST OF REFERENCES . .

vii

the

45
55
55
55
57
62
63
65
66
77
78

81

LIST OF TABLES

Table 1: Correlations for the Female Respondents .

Table 2: Means and Standard Deviations for Repeated
Measures ANOVAs . . . . . . . . . . . . . .

Table 3: Differences Between Men and Women for
Attitude and Intention Change . . . . . .

Table 4: Means and Standard Deviations for the Effect
of Communication and Beliefs about Rape on Worry

Table 5: Means and Standard Deviations for the Effect
of Communication and Message Threat Level
on Perceived Fear . . . . . . . . . . . .

Table 6: Means and Standard Deviations for the Effect
of Communication and Message Threat Level on
Sadness . . . . . . . . . . . . . . . . . .

Table 7: Means and Standard Deviations for the Effect
of Communication and Beliefs about Rape on
Attitude Change at Time 1 . .... . . . . . . .

Table 8: Means and Standard Deviations for the Effect
of Communication and Beliefs about Rape on
Intention Change . . . . . . . . . . . . . .
at Time 2

Table 9: Means and Standard Deviations for the Effect

of Message Type and Beliefs about Rape on
Frequency of Talk with Other People . . . .

viii

INTRODUCTION

it will do us large to recall
when the animal in us rises
that all women are someone’s
mother, sister, wife, or daughter
and are not fruit to be stolen when hungry
(Madhubuti, 1993, p. 169).
Overview of the Problem
The Prevalgggg 9f Rape

The crime of heterosexual rape against women has been
described as a "pervasive fact of American life, and its
incidence is growing dramatically" (Buchwald, Fletcher, &
Roth, 1993, p. 9). Suggesting that rape is a Ifact of life"
is alarming, and unfortunately, also accurate. Rape generally
can be defined as forced sexual intercourse against a victim's
will and without the victim/s consent‘.

The National Crime Victimization Survey (NCVS) reported
171,420 rapes in 1991. This indicates that one rape occurred
every 3.5 minutes. The FBI's Uniform Crime Report (UCR, 1991)
suggests that there was a 128 percent increase in the number
of reported rapes between 1972 to 1991. Rape against women
has become a problem of such magnitude that it has begun to

1

2

receive international as well as national attention.
Internationally, the Worldwatch Institute declared the most
common crime worldwide was violence against women (Wolf,
1991). Furthermore, "women's rights to be free from male
violence are now recognized by the united Nations as
fundamental human rights" (Koss, Beise, a Russo, 1994, p.
509). Nationally, it has been included as a health status
objective in the recent.Eealthy People 2000 Report?. Violence
against women, "an. area that historically' has been the
responsibility' of the fields of law' enforcement, social
services, and mental health, has become a national public
health priority“ (p. 226).

Colleges and universities are not immune to this
epidemic. Forty-five percent of all college women surveyed
reported that they had had some form of forced sexual contact,
with 25% of them reporting an experience of rape (Gidycz &
Koss, 1991; Ross, Gidycz, & Wisniewski, 1987). In.a survey of
2,400 colleges, Michigan State university reported the 2;;
highest number of sexual assaults to police on any campus in
1991. While it should be noted that Michigan State University
has one of the largest student populations in the country and
.is one of the few colleges with.a program.that encourages the
reporting of rape (i.e. the Department of Public Safety's
Sexual Assault Guarantee Program) , research suggests that rape
is drastically underreported to police. Typically, ten times

the number of rapes are reported to counseling offices than

3

are reported to police (Shaw, 1993) . For example, the phone
records of Michigan State university's Sexual Assault Crisis
Lines indicate that in 1991, 2 independent rape calls occurred
every day. Thus, the problem of rape against women is
prevalent in general and particularly on college campuses.
This problem warrants attention from communication researchers
so that effective communication campaigns can be created to
help address and diminish this problem.
WWW

Many women are familiar with the statistic that one of
four women may be raped at some time in their lives (Warshaw,
1988) . Despite this knowledge, it seems that some women may
refuse to acknowledge the prevalence of rape and also may
perceive that they are invulnerable to this threat. For
example, a recent study by Morrison (1994) found that while
many women rated sexual assault as a serious and severe
problem, they were not sufficiently motivated by the
persuasive fear appeal utilized to take protective action
against rape. Furthermore, in a recent focus group conducted
with female undergraduate students (Morrison, 1995) , one woman
stated, "I deny, cuz' . . . you say one in four, but, like, I
could name twenty of my friends and none of us have got . .
you know what I mean." This sentiment was echoed by many of
the women participating in the focus group, and also pervades
much of today's culture. As Harvard graduate and author Katie

Roiphe stated, "if I was really standing in the middle of an

4
epidemic, a crisis, if 25 percent of my female friends were
really being raped, wouldn't I know it?" (Roiphe, 1993, p.
52) .

Although many explanations exist for why women may not be
motivated to take protective action against the threat of
rape, perceiving invulnerability to the threat of rape in
order to cope with it seems to be a plausible alternative.
The process of coping with fear has been termed “fear control
processes" (Leventhal, 1970, 1971; Witte, 1992), and can
include defensive avoidance, minimization and denial. These
types of coping processes typically lead to less positive
attitude, intention and behavioral change. Morrison (1994)
did not measure fear control processes, thus the possibility
that these processes affected the outcomes cannot be ruled
out. Because many women may perceive themselves to be
invulnerable to the threat of rape, persuasive fear appeals
targeted toward women and recommending that they take
protective action against rape may be ineffective. A more
effective alternative may be to target messages toward men who
have female significant others and recommend that men talk to
the women they care about regarding the issues of rape and
self-defense. Women may be less likely to feel invulnerable to
a threat if they sense that their partners are concerned about
their safety. Furthermore, women may be more likely to take
protective action against rape if a man they care about

discusses the issue with them and suggests that they enroll in

5
a self defense course together.
Wig

This dissertation. will present a general 'model of
persuasion which targets the significant others of individuals
who perceive invulnerability to health threats. In certain
contexts where perceived invulnerability and fear control
responses (i.e., defensive avoidance, minimization) are likely
to be elicited in the target, more effective behavioral
outcomes may result from persuading a significant other who is
not directly vulnerable to a health threat. (The significant
other is indirectly vulnerable to the health threat because it
may impact the relationship, thus indirectly impacting the
significant other). This general model will be specified by
delineating several hypotheses about the issues of
heterosexual rape against women and self defense.
Additionally, this paper will examine the role of emotion in
the proposed indirect persuasive process. Specifically, this
dissertation will propose that relational closeness, beliefs
about rape and message threat level will interact to influence
emotional arousal and persuasive outcomes (attitude, intention
and behavioral change).

To begin, the possibility that women perceive
invulnerability to the threat of rape in order to cope with it
will be established.by reviewing literature on the Just World
Hypothesis, perceived invulnerability and fear control

processes. Next, current research on significant others and

6
preventive health will be addressed. The Theory of Reasoned
Action, Social Exchange Theories, and assessing relational
closeness then will be discussed in terms of recommending that
persuasive fear appeals be targeted toward significant others
of potential victims (men), rather than potential victims
themselves (women). Several fear appeal theories will be
reviewed and the idea that fear can be elicited in significant
others by an indirect threat will be proposed. Finally, the
influence of stereotyped beliefs and their relationship with
an emotional continuum, ranging from sadness to anger will be

discussed. Hypotheses are offered about each of these tapics.

LITERATURE REVIEW

In this section, the idea that women may cope with a
potential health threat by perceiving invulnerability to it
will be suggested. Research on Lerner's Just World Hypothesis
(1965) , perceived invulnerability, and fear control processes
will be reviewed to support this claim.
MW

Lerner's Just World Hypothesis (1965) suggested that
individuals have a "need to believe that they live in a world
where people generally get what they deserve. The belief that
the world is just enables the individual to confront his
physical and social environment as though they were stable and
orderly“ (p. 1030) . Belief in the Just World Hypothesis plays
the important adaptive role of allowing individuals to
function in an uncertain world. Because of this belief,
victims of crime typically are derogated and/or blamed for
their state, i.e. “if you got raped you must have done
something to deserve it“. This phenomenon of blaming and/or
derogating the victim has been demonstrated by researchers who
applied the Just World Hypothesis to situations involving
victims of rape and victims of AIDS (Anderson, 1992; Connors

5 Heaven, 1989; Furnham a Proctor, 1988; Kleinke & Meyer,

8
1990; Wagstaff, 1982; Wyer, Bodenhausen, & Gorman, 1985).

For example, Kleinke and Meyer (1990) found that after
viewing a six minute videotaped interview'with a female "rape
victimF, individuals with high. belief in a just world
recommended a shorter prison sentence for the rapist than did
individuals with low belief in a just world. Furthermore,
Wyer et al. (1985) found that priming individuals about the
negative consequences of aggression produced increased beliefs
that a rape victim was responsible for the crime.

Thus, just world.beliefs essentially are a belief system
that allows individuals to cope with potential threats. By
adhering to these beliefs, individuals can separate themselves
from. potential victims. Walster (1966) suggested that
individuals blame victims so that they can dismiss the
possibility of their own victimization by subsequently
separating themselves from. the victim. through
characterological and/or behavioral inferences. A similar
method of separating oneself from negative outcomes has been
labelled I'unrealistic optimism" or “perceived invulnerability“
(Perloff & Fetzer, 1986; Weinstein, 1984, 1987).

Pergeivgd Ingulngrapility

Perceived invulnerability is the tendency to believe that
one is more likely to experience positive health outcomes, and
that negative health outcomes are more likely to be
experienced by others. Recent research by Salovey, O’Leary,

Stretton, Fishkin, and Drake (1991) demonstrated that

9

respondents always believed that negative health outcomes were
more likely to occur to their peers than to themselves, and
that positive outcomes were more likely to befall them.
Additionally, research by Hansen, Raynor and Wolkenstein
(1991) indicated that long-term health consequences of
drinking were perceived as least likely to occur among
respondents who were heavy drinkers.

These two different lines of research both suggest that
one way in which people cape with health threats is to adhere
to belief systems which allow people to separate themselves
from potential victims. Similar conclusions have been
suggested by fear appeal researchers in their discussion of
fear control processes.

W

The idea that individuals use defensive reactions to cope
with threatening information has been suggested by several
fear appeal researchers (Janis & Feshbach, 1953, 1954;
Leventhal, 1970, 1971; Witte, 1992). Janis and Feshbach
described these reactions as ones which "tend to be of an
interfering nature - inattentiveness, perceptual distortions,
defensive efforts to deny or minimize the threat, etc.“ (1954,
p. 162). Specifically, when a message is perceived to be
highly threatening, yet the recomendations are not perceived
to be effective in deterring the threat (Witte, 1992) people
will respond to persuasive fear appeals by trying not to think

about the message information (defensive avoidance),

10

downplaying the nature of the threat (minimization), and/or
reacting against the communicator (perceived manipulation).
Additionally, Witte has described fear control as "primarily
emotional processes where people respond to and cope with
their fear, not the danger" (1992, p.116) . While she
hypothesized positive relationships between fear and defensive
avoidance and between fear and minimization (see Witte, 1994),
the results indicated negative relationships. Intuitively, if
fear control processes function as coping mechanisms to deal
with fear, then negative relationships should occur. When the
coping mechanisms emerge they should function to reduce the
fear.

It follows that if perceived invulnerability (the
tendency to believe that one is more likely to experience
positive health outcomes while negative health outcomes are
more likely to be experienced by others) functions as a coping
mechanism to deal with negative emotions, we should see a
negative relationship between fear and perceived
invulnerability as well as between worry and perceived
invulnerability and a positive relationship between perceived
invulnerability and other coping mechanisms (i.e., defensive
avoidance, minimization and perceived manipulation) .

H1: There will be a significant negative relationship between
fear and perceived invulnerability.
H2: There will be a significant negative relationship between

worry and perceived invulnerability.

11

H3: There will be a significant positive relationship between

‘perceived invulnerability and fear control processes.

Perceived.invulnerability should ultimately lead to less
desirable persuasive outcomes (i.e. negative or no attitude
change, lack of intention and behavioral changes). Thus,
persuasive fear appeals which are targeted toward women who
are vulnerable to a potential threat, yet feel invulnerable to
the threat may be ineffective. A potentially more effective
persuasive strategy would.be to persuade an individual who is
not directly vulnerable to a particular health threat to
intercede and.persuade someone they care about (i.e., someone
who is directly vulnerable to this health threat) to take
protective action. Literature on significant others and
preventive health will be discussed next to support this
claim”
Significant Others and Preventive Health

While the area of social support and health is a growing
literature, "only a handful of studies have looked at the
relationship between social support and/or social networks and
preventive health behavior" (Zimmerman & Connor, 1989, p. 58) .
Most of the research has focused on getting social networks
involved in supporting a target's behavior change process
(Gottlieb & Green, 1979; Lauer & Newlin, 1983; Wingard t:
Berkman, 1985), rather than recommending a significant other
(or key member of the social network) instigate the behavior

change. Given that social network support has proven to be

12

effective in sustaining behavior change, it seems likely that
social network support also may provide the impetus needed for
action in many individuals who deny that potential health
threats exist. Literature on the theory of reasoned action
and social exchange theory will be discussed next to support
this claim"

The Theogy of Reasoned Action

Fishbein and Ajzen's Theory of Reasoned Action (1975)
suggested that an individual's intention to behave is
influenced by the individual’s attitude toward the behavior
and her/his subjective norm (beliefs about whether significant
others think s/he should engage in the behavior and motivation
to comply with others' beliefs). This theory applies to
instances where the behavior is voluntary (i.e. the behavior
is performed because a person chooses to perform it). While
both the attitude and subjective norm.components ultimately
influence an individual's behavioral intention, an important
construct in this research is the subjective norm.component
and its relationship to attitude change.

The subjective norm component is comprised of normative
beliefs (perceptions of whether or not significant others
think the individual should engage in the behavior), and the
individual's motivation to comply with the preferences of the
significant others. For example, a woman who perceives her
boyfriend's strong desire for her to enroll in a self-defense

course, and who is motivated to comply with his desire, should

13

yield stronger behavioral intention to enroll in a self-
defense course than a woman who does not perceive these
desires in her boyfriend or is not motivated to comply with
them. Therefore, one method of persuading women to enroll in
self-defense courses in order to protect themselves against
rape (changing’ behavioral intention about a volitionary
behavior) entails making this subjective norm. component
salient to women. Increasing motivation to comply with the
values and.expectations of others has, in fact" been suggested
as a possible intervention approach for alcohol abuse (Hansen,
Raynor, & Wolkenstein; 1991) . Hence, by making the subjective
norm component salient to women and by persuading men to talk
with women whom they care about regarding the issues of rape
and self-defense, women who initially denied the threat of
rape may ultimately be persuaded to take protective actions.

Thus far, it has been.proposed that targeting persuasive
fear appeals toward the significant others of potential
victims may be an effective way to circumvent perceived
invulnerability and fear control processes (defensive
avoidance, minimization) in potential victims (women). Next,
social exchange theory' and. relational closeness will ibe
discussed to establish why significant others should be
motivated to persuade the women they care about to take

protective action.

14
W

Social exchange theories suggest that men in satisfied
relationships should be motivated to influence their
significant others to protect themselves. Kelley and Thibaut
(1978) posited that interpersonal relationships are evaluated
in terms of their rewards and their costs. Rewards include
pleasures and gratifications that result from the relationship
(such as value placed on friendship or companionship) and
costs include negative consequences that result from the
relationship (such as physical effort, mental effort, or
anxiety). The extent to which a relationship is valued, and
will continue, depends upon the relative outcomes (costs and
rewards) associated with that relationship.

A key component of their theory is the idea of mutual
interdependence, or the notion that one person's behavior
ultimately affects the other person. Individuals within a
relationship are outcome interdependent, and should therefore
be concerned about the well-being of their partners. For
example, if men value their relationships with women, and see
their relationships as sources of positive outcomes (the
rewards outweigh the costs), they should be motivated to
maintain their relationships. Since women are sources of

positive outcomes for men, hin i ll hrea s

the well being of the womee aleo coeld indirectly threagen the

W. However, because these men are
only indirectly susceptible to the threat (through their

15

relationships with women), they should.not experience as much

fear, perceived invulnerability or fear control processes as

do the women who are directly susceptible to the threat.

Therefore, men should be motivated to protect their

significant others as this ultimately helps sustain their

relationships and their continued positive outcomes.

If this reasoning is correct, what we should see is men
reporting more positive attitude and intention changes than
women. Thus,

H4: .Men will report significantly'more positive attitude
and intention changes than will women regarding enrolling
in self defense.

Not all men, however, may be motivated to persuade their
female partners. Their degree of motivation will be
influenced by the degree of closeness they feel toward their
female partners. Defining and measuring 'close" relationships
has been approached from several different perspectives (see
Berscheid, 1983). One promising approach is to examine the
activities that partners share, and this approach will be
reviewed next.

Aegeeeing Releeienel Cleeeeese
One method of describing and measuring the concept of

relational closeness is to assess the amount of

interdependence experienced by the partners (Berscheid,

Snyder, & Omoto, 1989). This “relational closeness inventory‘

consists of three subscales which measure how frequently the

16
partners interact (frequency), the strength of the partner’s
influence (i.e., estimates of the impact partner has on
decisions, plans, activities) (strength) , and the diversity of
their interactions (i.e., how many different activities they
do together) (diversity). Berscheid (1983) has suggested that
partners who are more “meshed“ (those with many
interconnections or shared activities) have the potential to
experience the most emotion. However, unless an
interconnection (shared activity) is broken, emotion probably
won't be experienced. Emotion is elicited when an
interconnection is broken (i.e., when a shared activity is no
longer shared). For example, many rape victims subsequently
have intimacy problems in their relationships. Thus, the
thought of not being able to share in the same relational and
intimate activities that they currently participate in may
arouse emotion in some men.
Therefore, in general
Proposition 1: Men in close relationships will experience
significantly greater emotional arousal in response to a
persuasive fear appeal than will men in relationships
that are less close.
Specific to the issue of heterosexual rape against women,
H5: Men in close relationships will experience significantly
greater emotional arousal (perceived fear, anger, worry,
and sadness) in response to a persuasive fear appeal

regarding rape against women than will men in

17

relationships that are less close.

To summarize, thus far it has been suggested that
persuasive fear appeals regarding rape against women may be
ineffective when messages are targeted toward women because
they may perceive themselves to be invulnerable to this
threat. Utilizing the theory of reasoned action and social
exchange theories, it has been suggested that persuasive
campaigns be targeted toward men in close heterosexual
relationships. Persuasive messages should recommend that men
talk to their significant others about the issue of rape and
persuade their significant others to enroll in self-defense
courses. Next, the role of emotion in persuasion will be
addressed. Specifically, the role of perceived fear and fear
appeal theories will be reviewed, then the relationship
between anger, sadness and beliefs about rape will be
discussed.

F r A 1 nd Per iv F r

Central to persuasive fear appeals is the idea that by
evoking fear of a potential threat, and then providing an
effective recommended action to avert this threat, individuals
will be motivated to comply with the recommended action.
Thus, fear is elicited by the persuasive message and
ultimately functions to motivate attitude, intention, and
behavioral change. As defined by Witte (1994) , a fear appeal
is 'a persuasive message that attempts to arouse the emotion

fear by depicting a personally relevant and significant threat

18
and then follows this description of the threat by outlining
recommendations presented as feasible and effective in
deterring the threat“ (p. 114). However, the way in which
fear functions is of considerable debate among fear appeal
researchers.

Poeitive relationships. Several researchers have found
a positive relationship between fear-arousing content and
persuasion such that as fear-arousing content increases,
degree of persuasion increases as well (Beck & Davis, 1978;
Burnett, 1981; Hewgill & Miller, 1965; Janis & Mbnn, 1965;
Leventhal, Singer 5 Jones, 1965; Miller a Hewgill, 1966;
Sutton & Eiser, 1984). These results have been demonstrated
regarding topics which range from smoking to nuclear fallout
shelters. Despite the empirical support for this model of
fear appeals, Boster and Mongeau (1984) described it as I'an
inadequate explanation of the effect of fear-arousing
messagesI (p. 365) due to the amount of support for opposing
models, including negative relationships and the existence of
moderating variables.

Negeeive relagionships. Researchers also have found
support for a negative relationship model, which suggests that
as fear-arousing content increases, degree of persuasion will
decrease (Goldstein, 1959; Janis & Feshbach, 1953; Janis &
Terwilliger, 1962; Leventhal & Watts, 1966). High levels of
fear arousing content may lead to resistance to persuasion for

several reasons, including feelings of invulnerability

19

(Leventhal, 1970) or defensive avoidance (Janis, 1967; Miller,
1963). Again, Boster and Mongeau (1984) suggested that this
model of fear appeals provided an inadequate explanation due
to the empirical support for positive relationships.

Qurvilineer relegionehipe. Proponents of curvilinear
relationships (Janis, 1967; Janis & Leventhal, 1968; McGuire,
1968; 1969) suggest that moderate levels of fear-arousing
content are the optimal level to employ in order to produce
desired persuasive effects. With low levels of fear-arousing
content, people are not motivated to attend to the persuasive
message. With high levels of fear-arousing content, defensive
processes (such as message minimization and defensive
avoidance) predominate and people are not motivated to adhere
to the message recommendations. However, with moderate levels
of fear-arousing content people are aroused enough to pay
close attention to the message recommendations, but not
overwhelmed such that defensive processes are initiated.
Therefore, desired persuasive outcomes are achieved. Although
support for optimal message processing occurring at moderate
levels of arousal has been articulated in the cognitive
processing literature (Darke, 1988; Gur, Gur, Skolnick,
Resnick, Silver, Chawluk, Muenz, Christ, 5. Reivich, 1988),
results from Boster and Mongeau's meta-analysis (1984) suggest
that the curvilinear hypothesis is inconsistent with the fear
appeal data (p. 365).

20

The Expenged Parallel Precese Medel. A recent fear
appeal model (Witte, 1992) combines Roger's (1975) conception
of protection motivation with Leventhal's (1970, 1971)
parallel processing model. The Extended Parallel Process
Model (EPPM) suggests that desired.persuasive outcomes can.be
achieved by eliciting the combination of high levels of both
perceived threat (fear-arousing content) and perceived
efficacy. This combination is proposed to be effective
because it initiates danger control responses. Leventhal
originally described danger control responses as a problem
solving control process where an individual responds to danger
by attempting to act upon and control the outer world while
ignoring or paying less attention to her/his fear. Thus,
danger control is guided by external cues, or information
primarily culled from the external environment.

Alternatively, less desirablejpersuasive outcomes (i.e.,
less attitude, intention, or behavior change, message
minimization, or defensive avoidance) will occur when high
levels of perceived threat combine with low levels of
perceived efficacy. This combination of perceptions is
proposed. to initiate fear control responses. ILeventhal
originally described fear control as the interpretation of
emotional behavior or internal cues, and as a process that
"may be' independent from the danger control processes,
"although the two processes should interact“ (p. 126). Thus,

in fear control people respond to, and attempt to control,

21

internal emotions rather then external (outer world) cues.

While considerable debate exists regarding which model
sufficiently explains how and why fear appeals succeed or
fail, general agreement exists regarding the arousal of
perceived fear. Typically, l'high fear messages produce more
perceived fear on the average than do moderate fear messages,
and moderate fear messages produce more fear on the average
than do low fear messages“ (Boster & Mongeau, 1984, p. 331).
One question explored by this dissertation is whether or not
perceived fear can be elicited when message recipients are not
direct targets of a threat. Literature from relational
closeness and Social Exchange Theories seem to support this
reasoning. In. particular, the concept of mutual
interdependence from social exchange theories suggests that a
threat that impacts one partner could ultimately impact the
other partner as well. Furthermore, Berscheid suggested that
partners in close relationships have the potential to
experience more emotion than those who are less close. Thus,
it seems reasonable that significant others could experience
perceived fear in response to threats to which they are
indirectly vulnerable. This suggests that, in general:
Proposition 2: a high threat persuasive.message will elicit

significantly more perceived fear in individuals in close

relationships than will a low threat message.

Specific to the issue of rape against women,

22

.H6: a high threat persuasive.message regarding rape
against women will elicit significantly more perceived
fear in men in close relationships than will a low threat
message.

Additionally, in general,

Proposition 3: A high threat persuasive message will yield
significantly'.more ‘positive attitude, intention and
behavioral change in individuals in close relationships
than will a low threat message.

Specific to the issue of rape against women:

H7: A high threat persuasive message will yield significantly
.more‘positive attitude, intention, and.behavioral change
in men in close relationships than will a low threat
message.

It also has been suggested that individual difference
variables (such as prior experiences, culture, and personality
characteristics) will influence perceptions. “Thus, the same
fear appeal may produce different perceptions in different
people“ (Witte, 1992, pp. 338-339). One personality
characteristic that may influence emotional arousal as well as
persuasive outcomes is the degree to which an individual
demonstrates stereotyped beliefs about a health issue.
Specific to the issue of rape, this is an individual's beliefs

about rape.

23

W

A growing literature on rape against women has examined
the significance of beliefs about rape and the impact which
accepting false beliefs about rape, rapists, and rape victims
has upon individuals and their social perceptions and
judgments (Brownmiller, 1975; Burt, 1980; Burt &.Albin, 1981;
Feild, 1978; Ward, 1988). Burt (1991) described “rape myths"
as an element of the general culture. “People learn them in
the same way they acquire other attitudes and beliefs--from
their families, their friends, newspapers, movies, books,
dirty jokes, and, lately, rock videos“ (p. 28). Burt’s
argument that these beliefs stem from culture has been
strengthened.by other feminist researchers who suggested.that
the culture we live in supports and condones rape against
women by perpetuating attitudes and beliefs which encourage
victimization and control over them.(Brownmiller, 1975; Clark
a Lewis, 1977; Griffin, 1971; Weis & Borges, 1973). An
example of how cultural messages perpetuate these beliefs
about rape appeared in Neweweek magazine in response to
Antioch College's new rules regarding consensual sex,

“And the rules, by postulating a culture of female

victimization and by creating many permutations of sexual

offenses, delight those feminists who consider America a

predatory ‘rape culture'“ (Will, 1993, p. 92).

Because accepting false beliefs about rape, rapists and
rape victims narrows the definition of repe (Burt & Albin,
1981), the outcome of supporting these stereotyped beliefs is

to deny that many instances involving coercive sex actually

24
constitute “rape“. Thus, the population can be divided
according to beliefs about rape; those individuals who support
stereotyped beliefs and those who don't.

Burt’s (1980) rape myth acceptance scale was designed to
assess these “prejudicial, stereotyped, or false beliefs about
rape, rape victims, and rapists“ (p. 217). This scale has
been demonstrated to correlate positively with sex role
stereotyping, adversarial sexual beliefs, and acceptance of
interpersonal violence. The scale taps five general
classifications of beliefs (Burt, 1991) . Four of these groups
of beliefs focus on the victim/woman and redefine sexual
assault as: nothing happened, no harm was done, she wanted it
or liked it, and she asked for it or deserved it. The fifth
group of beliefs focuses upon the man/assailant and
essentially condones his action.

Beliefs about rape may affect the emotions that are
aroused by the persuasive fear appeal. Specifically, as
beliefs about rape vary according to how strongly individuals
endorse stereotypes, the emotions elicited may vary from
empathy to anger. Specifically, men who are in close
relationships, yet endorse stereotyped beliefs about rape
should experience both anger and perceived fear in response to
a persuasive fear appeal regarding rape against women.

Three explanations exist for why anger may be elicited.
First, men may experience anger as a defensive response.

Because men are the perpetrators of heterosexual rape against

25

women, they may feel defensive over the issue that “not all

men are rapists“. Additionally, this anger may be directed

toward the hypothetical rape victim in the message, (i.e.

“anyone who does that deserves what she gets“). Finally,

anger may be elicited by the desire to separate his

significant other from the victim (i.e. “my girlfriend is not
like that“).

If men experience the mixed emotions of anger and
perceived fear in response to the persuasive fear appeal, they
should be more likely to stereotype the victim in the message
according to the rape myths of “deserving“ or “provoking“ the
rape. Bodenhausen (1993) recently demonstrated that
individuals are more likely to use stereotypes when they are
angry. Since stereotyping the victim functions to separate
the man's significant other from the victim in the message, he
should be less likely to attempt to persuade his significant
other regarding the issues of rape and self-defense. Thoughts
that may occur include, “since my girlfriend is not like that
I don't need to worry her about this“.

Therefore, in general:

Proposition 4: Stereotyped beliefs regarding the persuasive
issue will significantly influence emotional arousal such
that the more individuals endorse stereotyped beliefs,
the more anger will be elicited in individuals in close

relationships .

26

Proposition 5: Stereotyped beliefs regarding the persuasive
issue will significantly influence persuasive outcomes
such that the .more individuals endorse stereotyped
beliefs, the less attitude, intention and behavioral
change will be elicited in individuals in close
relationships.

Specific to the issue of rape against women,

H8: Beliefs about rape will significantly influence emotional
arousal such that the more individuals endorse
stereotyped beliefs about rape the more anger will be
elicited in men in close relationships.

H9: Beliefs about rape will significantly influence
.persuasive outcomes such that the .more individuals
endorse stereotyped beliefs about rape the less
attitude, intention and behavioral change will be
elicited in men in close heterosexual relationships.
Alternatively, men who are in close heterosexual

relationships yet do not endorse stereotyped beliefs about

rape should experience empathy or sadness in response to a

persuasive fear appeal regarding rape against women. Although

many forms of conceptualizing empathy exist, a basic
definition suggests that it is the reaction of one person to
the observed experience of another (Davis, 1983). Recently,
attention has been devoted to the possible multidimensionality
of this construct (Stiff, Dillard, Somera, Kim a Sleight,

1988; Tamborini, Salomonson, & Bahk, 1993), and its separate

27

affective and cognitive dimensions. Of particular interest is
the affective dimension of empathic concern, in which an
individual experiences emotional reactions due to another's
distress, and which is linked with altruistic helping
behaviors (Houston, 1990) . Since the persuasive message
depicts another person’s distress (an actual rape), men who do
not endorse stereotyped beliefs about rape should experience
empathic concern regarding the rape victim in the message, and
specifically their significant others, such that:

Proposition 6: Stereotyped beliefs regarding the persuasive
issue will significantly influence emotional arousal such
that the more individuals endorse stereotyped beliefs,
the less sadness will be elicited in individuals in close
relationships.

H10: Beliefs about rape will significantly influence emotional
arousal such that the more individuals endorse
stereotyped beliefs about rape the less sadness will be
elicited in men in close heterosexual relationships.

In summary, a general model of indirect persuasion
regarding health threats has been offered" Additionally, the
general model has been exemplified by several hypotheses
regarding heterosexual rape against women. Interactions
between relational closeness, :message threat level, and
beliefs about rape on emotional arousal and. persuasive
outcomes have been predicted. The methodology of this study

will be discussed next.

METHOD

Pro r

A pretest-posttest-follow-up design was emloyed with
separate samples of college male (N - 115) and female (N -
144) undergraduates from a large midwestern university.
Respondents were recruited from large introductory
interpersonal communication courses and given extra credit for
their participation in this research. The respondents
completed pretest measures assessing beliefs and attitudes
about rape, rape victims, and self defense as a protective
action, along with measures assessing their intent to
participate in self defense. Berscheid, Snyder and Omoto’s
relational closeness inventory (1989) was completed by the
male participants to determine the nature of their
relationships .

The respondents then read either a high threat or a low
threat stimulus message, and completed mood measures assessing
levels of fear, worry, anger and sadness. Measures assessing
the quality of the message arguments and plausibility of the
message were then completed, as well as ratings of severity,
susceptibility, response efficacy and self efficacy, defensive

avoidance, perceived manipulation, message derogation and

28

29
perceived invulnerability. Finally, the respondents completed
post-test measures assessing attitudes and intentions about
enrolling in a self defense course.

One week later, a follow-up survey was completed, again
assessing attitudes, intentions, and whether or not
respondents had talked with.a significant other/friend and/or
enrolled in self defense. At this time, respondents were
given postcards to return via campus mail to indicate whether
or not any subsequent action was taken (talking with a
significant other/friend and/or enrolling in self defense).
Stimulus Messages

Separate high threat and low threat messages were created
for both the female and male samples, thus, there were four
different messages. All of the messages included information
about the efficacy of self defense and information regarding
specific self defense classes. Eachmmessage recommended that
respondents enroll in self-defense courses and/or persuade
their friends (for female respondents) or significant others
(for male respondents) to enroll in self defense courses. The
female low threat message reported about an advertising
canpaign a man had created to help stop rape, while the female
high threat message reported about an attempted attack at
Michigan State University and statistics about the prevalence
of rape. The male low threat message reported about a
professor who suggested that rape statistics are exaggerated,

while the high threat message reported an acquaintance rape

30

and statistics about the prevalence of rape. Thus, threat was
manipulated by either including a vivid description of a rape
or reporting general information about rape. (See Appendix A
for all messages, pretest results are reported in the results
section).
We

Male participants completed a slightly different survey
from the female participants. Male participants were
instructed to think of a female whom they considered to be
close, and keep her in mind during the completion of the
survey; Additionally, several measures were completed.by the
men that were not completed by the women. All measures were
completed by both men and women unless noted otherwise.

Aptitudee goward self defense. Attitudes toward self-
defense was assessed with one seven-item semantic differential
scale. Female participants considered the statement “taking
a self defense or martial arts class to learn how to protect
myself from. rape would be:“ while male participants
considered the statement “taking a self defense or martial
arts class with a woman I care about so that she could learn
how to protect herself from rape would be:“. Response items
included bad/good, undesirable/desirable,
pleasant/unpleasant, (ranging from 1 - 7).

Inteetiene. The intentions scale was composed of eight
items with Likert-type response options. Example items

include, “I'm.going to sign up for a self defense or martial

31
arts course with a friend (female)/with a woman I care about
(male)“, “I am thinking about enrolling in a self defense or
martial arts course“ (both female and.male), and “I intend to
talk with a friend (female)/with the woman I care about (male)
regarding the two of us signing up for a self defense or
martial arts course. Response options ranged from 1 (strongly
disagree) to 7 (strongly agree).

Beliefe abegt rape. Both.men and women completed Ward's
Attitudes Toward Rape Victims Scale, which consists of twenty-
eight Likert-type items. Response options ranged from 1
(strongly disagree) to 7 (strongly agree). Example items
included, “a raped woman is usually an innocent victim}, “the
extent of a woman's resistance should be the major factor in
determining if a rape has occurred“, and “intoxicated women
are usually willing to have sexual relations“.

Additionally, men completed the rape empathy scale, which
consists of nineteen forced choice items. Participants were
instructed to choose which statement they preferred, ranging
from 1 (most prefer “A“) to 7 (most prefer “3“). Example
statements include; A) “In.general, I feel that rape is an act
that is provoked by the rape victimF, B) “In general, I feel
that rape is an act that is not provoked.by the rape victimP,
and A) “In a court of law, I feel that the rapist must be held
accountable for his behavior during the rape“, B) “In a court
of law, I feel that the rape victim.must be held accountable

for her behavior during the rape“.

32

Releeienal eleegese. Only the male participants
completed Berscheid, Snyder and Omoto's relational closeness
inventory. This inventory includes measures of shared
activities (i.e., in the past week I talked about personal
things/ate a meal/engaged in sexual relations. . .with “X“) and
the strength of the partner's influence (i.e, “X“ influences
inportant things in my life, influences the way I feel about
myself) .

Mood meaeure. Immediately after reading the persuasive
message, both male and female participants completed a mood
measure, consisting of twenty Likert-type items (1 to 7 range)
and assessing four moods. Sadness was assessed by five items,
including “not at all“/“very“ sad, sorrowful and depressed.
Anger was assessed by five items, such as “not at all“/“very“
angry, furious, and enraged. Fear was assessed by five items,
including “not at all“/“very“ afraid, frightened and scared.
Worry was measured by five items, such as “not at all“ / “very“
worried, apprehensive and concerned.

Severity. The severity scale was comprised of four items
with Likert-type responses. Participants responded to items
such as, “I believe that rape is a severe problem“ and “Rape
is a critical problem“. Options ranged from 1 (strongly
disagree) to 7 (strongly agree).

Sgeeepe'ipiligy. Four Likert-type response items
comprised the susceptibility scale. Respondents completed

items ranging from 1 (strongly disagree) to 7 (strongly

33
agree), such as “it is likely that I (the woman I care about)
could get raped“ and “I am.at risk (the woman I care about is
at risk) for the threat of rape“.

Reepenee effieeey. The response efficacy scale consisted
of four items with. Likert-type responses. Participants
responded to items such.as, “taking self defense/martial arts
classes are an effective way to protect women against rape“
and “self defense/martial arts classes work in protecting
women against rape“. Options ranged from. 1 (strongly
disagree) to 7 (strongly agree).

Self effieacy. Four Likert-type response items made up
the self efficacy scale. Respondents completed items ranging
from 1 (strongly disagree) to 7 (strongly agree), such as “I
(the woman I care about) could learn how to protect myself
(herself) from rape if I (she) took a self defense/martial
arts class“ and “Self defense/martial arts classes would not
teach me (the woman I care about) effective ways to protect
myself (herself) from rape“.

Pereeived genipeletion. The perceived manipulation scale
was composed.of four items with.Likert-type response options.
Example items include, “reports of rape are exaggerated“ and
“reports of rapetare overblownfi. Response options ranged.from
1 (strongly disagree) to 7 (strongly agree).

Minimizeeiee. Four Likert-type response items comprised
the minimization scale. Respondents completed items ranging

from 1 (strongly disagree) to 7 (strongly agree), such as “I

34
feel that prevention messages concerning rape are distorted“
and “I feel that prevention messages concerning rape are
deceptive“.

Defgeive aveignce. Defensive avoidance was assessed
with one six-item semantic differential scale. Participants
considered the statement “when I first read the message about
rape, my first instinct was to:“, and responded to items such
as “want to/not want to think about rape“, “want to/not want
to try to remember the message“. Response alternatives ranged
from 1 (want to) to 7 (not want to).

Meeeege quality. One five-item semantic differential
scale assessed message quality. Participants considered the
statement “I believe that the quality of the arguments
presented in this message is:“, and responded to items such as
“high/low“, “poor/outstanding“, and “superior/inferior“.

Response alternatives ranged from 1 (poor) to 7 (outstanding) .

Message plausibility . One five- item semantic
differential scale measured message plausibility .

Participants considered the statement “I believe that this
message was:“, and responded to items such as “plausible/not
plausible“, “believable/not believable“, and “possible/not
possible“. Response alternatives ranged from 1 (believable)
to 7 (not believable).

Per iv n ili . Five Likert-type response
items composed the perceived invulnerability scale.

Respondents completed items ranging from 1 (strongly disagree)

35
to 7 (strongly agree), such as “other women have more of a
chance of getting raped.than I (the woman I care about) really
do (does)“ and “rape is more likely to happen to other women
than it is to me (the woman I really care about)“.

Bel:_1a_vior meaeureg. Several one-item behavioral measures
were utilized to assess a range of behavioral action.
Respondents reported whether or not they talked with a
friend/woman they care about regarding the issues of rape and
self defense, if they'spoke‘with.others regarding these issues
(men only), and if they signed up for a self-defense course.
De M a re

The dependent measures included mood measures in response
to the stimulus messages, attitude change regarding self
defense as a protective measure against rape, intentions to
enroll and/or persuade friends/significant others to enroll in
self defense courses, and the behavioral measures of whether
or not a conversation occurred, and whether or not anyone
enrolled in self defense.

Ind M r

The independent measures included the threat level of the

stimulus messages, relational closeness, and beliefs about

rape.

RESULTS

Preteet

The messages were extensively pre-tested with a different
sample from a large midwestern university (N s 27 men, N = 30
women). T-test results indicated the messages to be
significantly different in terms of perceived fear for both
the male respondents (_t;(1,25) - 2.41, p < .05, M :- 5.69 high
threat, M - 4.64 low threat) and the female respondents
(5(1,28) - 3.50, p < .05, M - 5.08 high threat, M a 3.51 low
threat).

Mein Prejee;

Scale Reliabilitiee. Confirmatory factor analyses
(Hunter & Gerbing, 1982) were employed to test the
dimensionality of the measures. The analyses resulted in the
deletion of several items from several measures. Cronbach's
coefficient alpha and the number of items for each measure are
reported for all scales in Appendix B.

The relational closeness inventory was particularly
problematic in terms of the lack of unidimensionality of the
behavioral activities used to assess closeness. Because this
research presents a communication model, the decision was made

to focus solely on items measuring talk between partners.

36

37
This was measured by three items responding to the question,
“please check all those activities that you have engaged in
alone with (“X“/the woman you care about) in the past week“:
“talked.about nonpersonal things“, “talked on the phone“, and
“talked about personal things“. Thus, this research uses
communication as an indicator of relational closeness.

To measure beliefs about rape, items from Ward's
Attitudes Toward Rape Victims Scale were combined with items
from the rape empathy scale to create a unidimensional
measure. These scales were combined.because of the degree of
item overlap and similarity between the two measures. The
final scale consisted of seventeen items (see Appendix C for
Beliefs About Rape Scale).

Inducgion Checke

Considering the male respondents first, T-test results
indicated thezmessages t0>be significantly different in terms
of perceived fear (t(l,109) - -2.17, p < .05, M n 4.68 high
threat, M - 4.09 low threat) and severity (e(1,109) - -2.40,
p < .05, M . 6.39, high threat, M a 5.96, low threat).
Additionally, the messages were not perceived to differ
according to response efficacy (t(l,109) - .41, p > .05, ns,
M :- 5.40, high threat, M :- 5.49, low threat), self-efficacy (_t
(1,109) - -.13, p > .05, ns, M - 5.63, high threat, M . 5.61,
low threat), message plausibility (t(1,108) - .12, p > .05,
ns, M - 5.67, high threat, M a 5.69, low threat), or message

quality (e(1,109) - -.52, p > .05, ns, M - 5.22, high threat,

38
M = 5.13, low threat).

Similarly, the female respondents reported the messages
to be significantly different in terms of perceived fear
(;(1,146) a 2.80, p < .05, M . 4.95 high threat, M = 4.28 low
threat). However, the messages also differed according to
message plausibility (t(l,l43) a 3.46, p_< .05, M a 6.35, high
threat, M - 5.90, low threat) and.message quality (5(1,146) =
1.97, p . .05, M = 5.43, high threat, M . 5.10, low threat).
Ironically, the women seemed to perceive it to be more
plausible for a man to rape a woman than to create an
advertising campaign to reduce rape against women. Finally,
the women did not perceive the messages to differ according to
severity (g (1,144) = .33, p > .05, ns, M = 6.42, high threat,
M - 6.38, low threat), response efficacy (5(l,146) - -.13, p
> .05, ns, M - 5.75, high threat, M - 5.77, low threat), or
self-efficacy (_t;(1,l46) - .71, p > .05, ns, M - 5.99, high
threat, M = 5.89, low threat).

Re ul for F 1 Re n

Hypethesis 1. This hypothesis suggested that perceived
fear would be negatively related to perceived invulnerability
for women. An examination of the correlation matrix (see
Table 1) revealed a small and insignificant relationship
between perceived invulnerability and perceived fear (; = -

.06, p_> .05, ns), thus the data were not consistent with this

hypothesis.

39
Table 1

gerreletione fer Female Reepondents

 

 

FEAR WORRY PI MIN PM DA
FEAR
WORRY . 86"
PI -.06 -.06
MIN -.06 -.12 -.00
PM -.06 -.14 -.01 .52“
DA -.07 -.02 .13 .09 .22'

 

“indicates significance at the .001 level, 'indicates
significance at the .01 level, PI - perceived invulnerability,
MIN - minimization, PM . perceived manipulation, DA ..
defensive avoidance.

Hypotheeig 2. This hypothesis suggested that worry would
be negatively related.to perceived.invulnerability for women.
An examination of the correlation matrix revealed a small
negative relationship between perceived invulnerability and
worry (; - -.06, p > .05, ns). Again, the relationship was
not significant so the data were not consistent with this
hypothesis.

Hypoehesis 3. This hypothesis suggested that perceived
invulnerability would be positively related to fear control
processes for women. The correlation matrix revealed a
positive correlation between perceived invulnerability and
defensive avoidance, however it was not significant (p a .13,
p > .05, ns). Additionally, there were no relationships
between perceived invulnerability and minimization (; s - .00)

or between perceived invulnerability and perceived

manipulation (; - -.01) . Hence, the data were not consistent

40
with this hypothesis.

To summarize thus far, the data were not consistent with
the first link in the model. This link suggested that women
would experience perceived invulnerability in response to a
potential health threat (in this case, rape against women).
In general, women felt vulnerable to the threat of rape (M
=2.03, ed a 1.3, range a 1-7, low scores indicate low
perceived invulnerability). Additionally, the data were not
consistent with the posited relationships between perceived
invulnerability, emotions and fear control processes for
women.

Hypotheeis 4. This hypothesis suggested that men would
report more positive attitude and intention change toward self
defense than would women. To test this hypothesis, both
repeated measures ANOVAs and T-tests were employed. First, a
series of six repeated measures ANOVAs were performed with
time as the within groups factor (pretest, posttest and
follow-up) for both men and women. The first ANOVA assessed
the attitudes of the male respondents. This analysis
indicated a significant effect for the within groups factor of
time on attitudes (2(2,168) a 4.52, p - .01, e3 - .05). The
means and standard deviations for all ANOVAs are listed in
Table 2. The second ANOVA assessed the male respondents'
intentions to take action with their female partners (i.e.,
“intentions for both“) . This analysis indicated a significant

effect for the within groups factor of time on intentions

41
(2(2,166) - 7.32, p - .001, e’ . .08) . The last ANOVA for the
male respondents assessed their intentions to talk with their
female partners so that the women would take action by
themselves. This analysis also indicated a significant effect
for the within groups factor of time on intentions (F(2,168)
= 10.40, p < .0001. n3 - .11).

Similarly, the ANOVAs for the female respondents also
indicated significant effects for the within groups factor on
intentions. The analysis assessing intentions for women to
enroll by themselves in self defense revealed a significant
and.substantial effect for the within.groups factor of time on
intentions (E(2,252) . 33.48, p . .000, p: a .21), as did the
analysis assessing intentions to enroll in self defense with
a friend (E(2,252) - 33.85, p,- .000, p3 - .21). The effect
of the within groups factor of time on attitudes toward self
defense was not significant (§(2,252) - 1.90, p > .05, ns).
However, it should.be noted that women had.very high.positive
attitudes about self defense at all three points in time.
Furthermore, an examination of the means revealed that women
initially reported higher scores than men for each of the
measures at the pretest, and continued to display this trend

for both the posttest and the follow-up surveys.

42

 

 

 

 

 

 

 

 

Table 2
Meane egg Standare Deviatione for Repeeted Measuree ANOVAs
Atti§eee Pregest, Postteet Follog up
Men M a 5.91 M - 5.64 M = 5.62
eg . .98 eg a 1.02 eg a 1.10
Women M I 6.21 M = 6.16 M = 6.04
fig . .86 eg - .90 eg a 1.02
In T Pr e t Poeeeeet Follow up
Men M I 3.06 M I 3.58 M I 3.41
fig . 1.47 eg . 1.46 pg 3 1.60
Women M I 3.35 M = 4.24 M = 4.18
g = 1.62 _S_d I: 1.62 _8_g = 1.68
Intent/A Pretegt Posttegt Follow up
Men M I 3.58 M = 4.42 M I 4.07
E, I 1.68 'Ld I 1.63 eel I 1.73
Women M a 3.72 M - 4.49 M = 4.39
eg - 1.57 eg - 1.55 eg = 1.65

 

N a 84 (men), N - 127 (women), Intent/T - intentions to enroll
in self-defense together (men with their partners and women
with another female friend), Intent/A - intentions to enroll
in self-defense alone (men to persuade their partners to
enroll alone and women to enroll alone).

The differences between.men and.women also were analyzed
by T-tests. Three sets of change scores were created for the
attitude and intention measures; the difference between the
posttest and the pretest, between the follow-up and the
posttest, and between the follow-up and the pretest. The data
indicated a significant difference between men and women for

intention change regarding enrolling in self defense alone

(women by themselves, men's partners by themselves) between

43
the posttest and the pretest (t(1,252) a 2.22, p < .05, M =
.81, men, M - .51, women), and for intention change regarding
enrolling in self defense alone (women by themselves, men’s
partners by themselves) between the follow-up and the posttest
(£(1,83) . -2.00, p < .05, M . -.36, men, M - -.08, women).
No other significant differences emerged between men and women

for any of the other change scores (see Table 3).

44
Table 3

Differences Between Men and Women for Attitude end Intention
Cheese

 

 

t-value df p M sd

ATTl -l.67 211 >.05 -.26 .97
-.05 .82

ATT2 -l.25 211 >.05 -.34 1.03
-.17 .89

ATT3 .25 212 >.05 -.09 1.13
-.13 .92

INTENTl/T —1.29 128 >.05 .79 1.03
.89 1.09

INTENTl/A’ 2.22 85 <.05 .81 1.64
.51 1.30

INTENTZ/T -1.26 126 >.05 .71 1.30
.83 1.49

INTENT2/A .41 83 >.05 .50 1.60
.44 1.44

INTENT3/T - .29 125 >.05 -.09 .99
-.06 1.14

INTENT3/A' -2.00 83 <.05 -.36 1.27
-.08 1.17

 

ludifference between posttest and pretest, 2=difference
between follow-up and pretest, 3=difference between follow-up
and posttest, Teenrolling in self defense together,
A-enrolling in self defense alone. First mean and standard
deviation scores listed are for men, second are for women.
Therefore, the data were consistent with.hypothesis four
regarding intentions to enroll in self defense alone (men's
partners by themselves/women by themselves). Men revealed

more positive intention change than did women between the

posttest and pretest in terms of enrolling in self defense.

45

However, the data were not consistent with this hypothesis for
any of the other change scores (attitudes or intentions to
enroll in self defense together). In terms of the general
model, the data are consistent with the notion that men will
demonstrate more positive intention change than women, but not
the notion that they will demonstrate more positive attitude
change than women.

Results_fgr_hale_82§222dents

It should again be noted that communication was employed
as an indicator of relational closeness. Thus, while the
hypotheses will be reviewed in terms of their original wording
(i.e., closeness), the results will be reported in terms of
communication.

Median splits were used to dichotomize each predictor
variable (communication.M - .67, beliefs about rape M - 5.71)
and ANOVA procedures were employed to test the hypotheses.
Since main effects should not be interpreted when interactions
exist, three-way ANOVAs (between message threat level,
communication.and.beliefs about rape on emotional arousal and
persuasive outcomes) were conducted to test for interactions.
While the data did not reveal any three-way interactions,
several two-way interactions and main effects were discovered.

Hypetheeie Five. This hypothesis suggested that men in
close relationships would experience significantly greater
emotional arousal than wouldumen in less close relationships.

The data were consistent with this hypothesis for arousal of

46
anger. Men who communicated. more with their partners
experienced significantly more anger (M = 4.90) than did those
who communicated less (M - 4.17), (g(1,84) a 4.33, p < .05, e2
= .05).

Several unpredicted results also emerged regarding
emotional arousal as a dependent measure. The data revealed
a main effect for message type on the amount of anger
experienced (g(1,84) - 4.42, p < .05, e3 a .05). Men who read
the high threat message experienced significantly more anger
(M a 4.96) than did those who read the low threat message (M
= 4.23). Finally, a significant.two-way interaction.was found
for communication and beliefs about rape on the amount of
worry experienced (E(1,84) - 4.60, p - .05, e? - .04). Men
who communicated more with their partners and did.not endorse
stereotyped beliefs about rape experienced the greatest amount
of worry, followed by those who communicated more with their
partners and endorsed stereotyped beliefs about rape more
strongly. The least amount of worry was experienced.by those

who communicated less and did not endorse stereotyped beliefs

about rape (see Table 4).

47

 

Table 4
Meeee end Steedard.Deviation§ for The Effect ef Communication
and Belief R e on Wor

Low Comm High Comm

 

Low Stereotyped M a 4.81 M = 5.98
Beliefs §_d_ - 1.58 pg - .85
n a 16 n I 22
High Stereotyped M . 5.04 M a 5.27
Beliefs 5A - 1.19 M1 = .89
n I 19 n I 28

 

Therefore, the data were only partially consistent with
hypothesis five. The predicted main effect for closeness on
emotional arousal only emerged for anger, but not for fear,
worry or sadness. Additionally, emotional arousal was
affected by message type (those who read the high threat
message experienced more anger) and by the combination of
closeness and beliefs about rape (for the amount of worry
experienced). In general, the data appear to be consistent
with the contention that closeness (as measured by
communication with partner) influences emotional arousal (both
by itself and in combination with stereotyped beliefs).

Hypeeheeig Six, This hypothesis suggested that men in
close relationships would experience significantly more
perceived fear in response to the high threat message compared
to the low threat message. .A significant two-way interaction

between communication and message threat level emerged for

48
perceived fear, (g(1,84) . 8.47, p :- .005, p2 - .08), however,
it was not in the predicted direction (see Table 5 below).
The most perceived fear was experienced by men who
communicatedmmore with their partners and read the Meg threat
message, followed by those who communicated more with their
partners and read the high threat. message. The least
perceived fear was experienced by men who communicated less

with their partners and read the low threat message.

Table 5

Means and Standard.Deviations for The Effect of Communication
and Meseage Threat Level on Perceived Fear

 

 

Low Comm High Comm
Low Threat M a 3.34 M a 5.10
fig 3 1.46 eg = 1.24
n = 18 n a 24
High Threat M = 4.61 M a 4.70
gg a 1.45 eg - 1.19
n 17 n 26

 

While the hypothesized two-way interaction occurred, it
was not in the predicted direction. Therefore, the data were
not consistent with hypothesis six.

Additionally, an 'uopredicted significant two-way
interaction between communication and message threat level
emerged for sadness (2(1,84) - 5.07, p < .05, p3 - .06). Men

who communicated less and read the high threat message

49
experienced the most sadness, followed by men who communicated
more and read the high threat message. The least sadness was
experienced by men who communicated less and read the low

threat message (see Table 6 below).

Table 6
Means and §tendard Deviations for The Effect of Communicetion

and Mes a Thr Level on a

 

 

 

Low Comm High Comm
Low Threat M I 3.15 M I 4.04
£1 a 1.37 Ld a 1.51
n I 18 n I 24
High Threat M I 4.63 M I 4.09
fig . 1.49 pg 8 1.53
n I 17 n I 26

 

H he i even. This hypothesis suggested that the
high threat message would evoke more attitude, intention and
behavioral change than would the low threat message for men in
close relationships. A significant main effect was revealed
for message type on attitude change between the pretest and
follow-up surveys (g(l,83) I 3.10, p < .05, e’ I .05). Men
who read the low threat message reported a lower mean for
attitude change (M I -.48) than did those who read the high
threat message (M I -.09), consistent with the hypothesized

direction. However, since a main effect emerged when a two-

way interaction was predicted, the data were not consistent

50
with this hypothesis. No other significant main effects or
interactions were detected for message type on intentions.
(A significant interaction between message type and beliefs
about rape on behaviors is described later in Table 9).

Hypotheeig Eight. This hypothesis suggested that beliefs
about rape would influence anger arousal for men in close
relationships. The effect for beliefs about rape on anger was
not significant (_F_(l,84) I .132, p > .05), thus, the data were
not consistent with this hypothesis.

Hypeehegie Nine, This hypothesis suggested that beliefs
about rape would influence persuasive outcomes for men in
close relationships. While several interesting effects
emerged that are discussed below, in general, the data were
not consistent with this hypothesis.

AteiEugeg. A significant two-way interaction between
communication and beliefs about rape was found on attitude
change between the pretest and posttest (E(1,83) I 4.10, p <
.05, p’ I .05) . Men who communicated less and did not endorse
stereotyped beliefs about rape reported the lowest mean (M I -
.66) for attitude change. However, results from a one-sample
T-test revealed that the mean was not significantly different
from zero (t(1,15)I -1.71, p > .05, ns), essentially
indicating no change. The only positive mean for attitude
change was small and insignificant (M I .02) , and was reported
by men who communicated more and did not endorse stereotyped

beliefs about rape. Overall, though the data revealed

51
significant differences between the groups (as displayed in
Table 7), none of the changes in.attitudes toward self defense

were significantly different from zero.

Table 7

Meeng ang Stangerg.Devia§ione fer The Effect of Commpeication
ang Beliefg epeg; Rape en Ateigude Change a; Time 1

 

Low Comm High Comm

 

 

Low Stereotyped M I -.66 M I .02
Beliefs eg I 1.54 fig I .56

n I 16 n I 22
High Stereotyped M I -.14 M I -.36
Beliefs Mg I .88 g_ I .83

n I 18 n = 28
Ineeetione. A significant two-way interaction between

communication and beliefs about rape emerged for intention
change to enroll in self defense together between the pretest
and follow-up surveys (2(1,82) I 9.88, p < .05, p2 I .06).
Men who communicated less and more strongly endorsed
stereotyped beliefs about rape reported the highest mean for
intention change. Results from a one-sample T-test revealed
the meon to be significantly different from zero (e(1,17).
2.23, p_ < .05, M I .89) . None of the other reported means for
intention change were significantly different from zero (see

Table 8).

52

Table 8
Meens egg §§eeeerg Deviagione for The Effect of Communicetion

and Beliefe about Repe on Ingention” Change at Time 2

 

 

Low Comm High Comm
Low Stereotyped M I -.10 M I .55
Beliefs eg I 1.06 eg I 1.63
n I 16 n I 22
High Stereotyped M I .89' M I .11
Beliefs eg I 1.69 ed I 1.25
n I 18 n I 27

 

'denotes significantly different from.zero at .05 level
"denotes intentions to enroll in self-defense together
BeMeviore. A significant two-way interaction between
message type and.beliefs about rape emerged for the frequency
with which men spoke with people besides their significant
other regarding the issue of rape (E(l,84) I 8.93, p,< .05, g3
I .05) . Men who read the low threat message and endorsed
stereotyped beliefs about rape more strongly reported the
highest frequency, followed by men who read the high threat
message who did not endorse stereotyped beliefs about rape.
The lowest frequency was reported by men who read the high
threat message and endorsed stereotyped beliefs about rape

more strongly (see Table 9).

53

Table 9
Meane egg Spandard Deviations for The Effect of Meesage Me

and Beliefe epoet Repe on Frequency of Talk with cher Peeple

 

 

Low Threat High Threat
Low Stereotyped M I 2 . 06 M I 2 . 50
Beliefs eg I 1.55 gg,= 1.54
n I 18 n I 20
High Stereotyped M I 2.63 M :- 1.78
Beliefs eg I 1.76 gg.= .94
n I 24 n I 23

 

Lastly, a significant main effect emerged for beliefs
about rape on whether or not men spoke with their significant
others regarding rape and self defense (E(1,84) I .90, p <
.05, pf I .06) . Men who did not endorse stereotyped beliefs
about rape were more likely to speak with their significant
others (M I .89) than were men who endorsed stereotyped
beliefs about rape more strongly (M I .68) .

Hypoehesie Tee. This hypothesis suggested that beliefs
about rape would influence sadness for men in close
relationships. The data revealed no effect for beliefs about
rape on sadness (2(1,84) I .656, p > .05) thus the data were
not consistent with this hypothesis.

To summarize the importance of stereotyped beliefs about
rape, while the data were not consistent with many of the

hypotheses, the results are consistent with the notion that

54
stereotyped beliefs should be measured. Beliefs about rape
did.not influence anger or sadness (contrary to predictions),
but did influence whether or not men talked with their
significant others. Additionally, beliefs about rape combined
with closeness to influence intention change and attitude
change, and combined with message type to influence the
frequency with which men spoke with other individuals about

the topics of rape and self-defense.

DISCUSSION

Overview

The purpose of this research was to suggest a general
model of indirect persuasion and to test it utilizing the
specific issues of rape and self-defense. It was suggested
that women.may perceive invulnerability to the threat of rape
which would result in less desirable persuasive outcomes
(i.e., less positive attitude, intention and. behavioral
change). To circumvent this process it was proposed that l)
persuasive messages be targeted toward men in close
relationships, and 2) these messages recommend that men talk
with the women they care about regarding the issues of rape
and self defense.

Female Reependeege

The data were only partially consistent with the proposed
model. The relationships between perceived invulnerability
and emotions (fear and worry) as well as the relationships
between perceived invulnerability and fear control processes
(perceived manipulation, defensive avoidance and minimization)
for female respondents were small and insignificant. An
examination of the means for these constructs revealed a

basement effect for perceived invulnerability. Put another

55

56

way, the female respondents in this study gel feel vulnerable
to the threat of rape. However, additional analyses revealed
a significant positive relationship between perceived
invulnerability and defensive avoidance (; I .24, p < .01) in
the follow-up survey one week after the message induction.
This finding suggests that perceived invulnerability and fear
control processes may both function as coping mechanisms that
are best measured well after the message induction. As Witte
suggested, “to truly assess avoidance patterns, one must delay
measurement of the construct“ (1994, p. 122) . Intuitively, it
seems reasonable to suggest that a coping mechanism will
emerge well after, as opposed to immediately after, the
message processing. During the initial message processing,
respondents are in a data collection environment and may be
cognitively taxed and inwardly focused. Alternatively, when
an individual is outside of a data collection environment and
not as cognitively taxed, unwanted thoughts and feelings from
the initial message processing may emerge. This is when a
coping mechanism should become more active.

To adequately conceptualize and measure perceived
invulnerability and fear control variables, researchers need
to decide how and when to measure these constructs. These
findings support Witte's suggestion that delayed measurement
is necessary. Possible approaches to assessing these
constructs include diary methods or delayed thought-listing

tasks. One option is to utilize delayed thought-listing

57

measures and code the thoughts according to whether or not
they reveal threatening, efficacious or neutral information.
This approach may allow researchers to better distinguish fear
control from danger control processes.
file Reepengengg

gloeepeee and Epotionel Arous_a_l. Perhaps the most
interesting results from these data were the relationships
between relational closeness (as measured by communication
with partner) and emotional arousal. Even though the male
respondents were not directly vulnerable to the threat
conveyed in the message, they still reported fear arousal.
Specifically, the most fear was reported by those who
communicated more with their partners and read the low threat
message. The notion that it is possible to evoke fear in
indirect targets has several practical implications for fear
appeal researchers. First, these findings broaden the concept
of susceptibility. Several fear appeal theorists have
suggested that eliciting a feeling of vulnerability is a
central component when crafting a fear appeal. These data
suggest the need for expanding our notions of susceptibility
to include indirect targets of a threat. Also, these findings
provide another persuasive route for fear appeals. As Boster
and Mongeau noted (1984), fear manipulations typically are
small, likely due to ethical and human subjects restrictions.
This research suggests that by emphasizing the threat to the

relationship, perceived fear can be elicited in indirect

58

targets. Finally, this research has emphasized the importance
of examining the range of emotions experienced by message
recipients. Varying degrees of fear, sadness, anger and worry
were elicited by the persuasive messages used in this
research. To adequately parse out effects, fear appeal
researchers should measure the range of negative emotions
elicited by fear appeals and test causal models.

Beliefe epout Rape. This research suggests that
stereotyped beliefs regarding the persuasive issue of interest
(in this case, beliefs about rape) are an important construct
to examine. Beliefs about rape influenced emotional arousal
(worry), attitude change, intention change, and behaviors.
Specifically, beliefs about rape combined with closeness (as
measured by communication with partner) to influence worry,
attitude change and intention change. Also, beliefs about
rape combined with message threat level to influence how often
men spoke with someone besides their significant others
regarding rape and self-defense. Although no consistent
pattern emerged in the data in terms of the two-way
interactions, the behavioral outcome measure of whether or not
men spoke to their significant others about the issues of rape
and self defense was determined solely by beliefs about rape.
Men with more empathy toward rape victims (i.e. , those who did
not endorse stereotyped beliefs) were more likely to speak
with their significant others. While inconsistent with the

predicted interaction, this finding highlights the importance

59

of assessing stereotyped beliefs regarding the persuasive
issue of interest and. assessing their influence on the
critical behavioral outcomes. For example, if people believe
the ideas that only gay men can acquire HIV, they may be less
likely to take protective action against the threat of HIV
(i.e., always using a condom during sexual intercourse).
Consequently, rather than solely focusing on evoking fear,
these data indicate the need for communication campaigns which
address stereotyped beliefs as well.

Pereueeive Qgtegee. The most intention change was
reported by men who more strongly endorsed stereotyped beliefs
about rape and communicated less with their partners.
However, the results from the behavioral measures indicate
that this same group of men ultimately did.not follow through
with their intentions, since the men who were less likely to
endorse stereotyped beliefs about rape were more likely to
talk.with their significant others. These findings emphasize
a common problem in persuasion research, namely, motivating
people to act on their intentions. Future research should
attempt to uncover the barriers that are inhibiting action,
such. as the question of whether or' not. perceptions of
invulnerability influence men's intentions and behaviors as
well as reasons men may give for not talking with their
significant others.

The results comparing men's attitude and intention change

to women's were only partially consistent with predictions.

60

Men reported more positive intention change than did women (in
terms of enrolling in self defense alone, i.e., men persuading
their partners to enroll in self defense alone). However,
they did not report more positive attitude change than women.
Furthermore, the means from the repeated measures data
revealed that women had more positive attitudes and intentions
than did men at all three time points.

These results can be interpreted several ways. First,
the basement effect for women's perceived invulnerability must
be considered. The model presented in this research targets
women who perceive themselves invulnerable to a health threat
and this sample was not captured. This limitation reiterates
the notion that some barriers to action (i.e., perceived
invulnerability) need to be measured well after the initial
survey date and may require multiple methods in order to
better understand them (such as the combination of diary and
thought-listing techniques).

Additionally, the issue of efficacy must be addressed.
As Hale and Dillard suggested (1995) , a recommended response
must be easy to enact. Unfortunately, presenting efficacious
recommendations is not always possible for many health issues
(such as rape prevention or organ donation) . For example, the
recommended response in this research was to enroll in a self
defense class which meets more than once. Perhaps more
intention change or behavioral outcomes would have been

obtained by recommending a brief seminar as opposed to an

61

extended course. Because attitudes about the efficacy of self
defense as a protective measure against rape were high in
general (as mentioned above), some other barriers must have
influenced people's decisions not to take action. Future
research should seek to uncover these barriers, as well as
deal with the question of how to create the perception that a
recommendation may be more efficacious than it actually is.

Subsidiag analyses. Post hoc analyses revealed two
important findings. First, 58% of the men in this research
talked about the issues of rape and self defense to people
other than the women they cared about. Second, 22% of the men
in this research talked with their significant others about
the issues of rape and self defense. It is important that in
our search for statistical significance and predicting
differences between groups that we not lose sight of important
behavioral outcomes. Although this data was not consistent
with the hypothesized differences between:men and women, over
half of the men surveyed discussed the issues of rape and self
defense and.almost one quarter of the:men surveyed.spoke with
their significant others about the issues of rape and self
defense.

These findings have important theoretical implications
for the literature on social support and preventive health.
Specifically, these data suggest that persuading relational
partners to instigate behavior change in their significant

others may be a promising persuasive route to explore. The

62

findings indicate that emotional arousal, intention change and
behavioral action all can be achieved in people who are not
directly vulnerable to a health threat. An implication of
these findings is that significant others can play a pivotal
role in promoting healthy lifestyles, not just supporting
them. The next question to be answered is, how effective were
significant others in promoting behavioral changes in their
jpartners? To adequately assess this question, a couples study
must be conducted.
Par ters d Limi a ions of the Model

The goal of the general model presented in this research
is to circumvent the coping' mechanisms (perceived
invulnerability, fear control processes) that individuals may
use to deal with health threats. A limitation of this model
is that it excludes those individuals not in close
relationships (as indicated by communication with partner).
While this model has been specified in terms of close
heterosexual relationships, it seems reasonable to assume that
it could be extended to include close same-sex friends, and
other types of intimate relationships. .As long‘ as an
individual is involved in a close relationship that displays
mutual interdependence, and only one partner is susceptible to
a health threat, this model could apply.

.Additionally, this model has been explicated.by applying
fear appeals to significant others. However, this model

conceivably could have utilized a general persuasion theory

63
rather than a fear appeal theory. The central issue is the
interaction between individuals who deny a health threat and
their significant others. Future research could test the
usefulness of this model by utilizing a variety of health
issues and a variety of general theories.

While this model does not reconcile previous research, it
provides a useful framework to employ in order to persuade
those who have not been persuaded by previous research
efforts. Typically, researchers are happy to change the
attitudes and actions of a small group of people. However,
when current research efforts are ineffective with an entire
group of individuals, new and creative solutions must be
attemptedm The model presented in this dissertation utilizes
social exchange theories and the theory of reasoned action to
suggest an alternative approach to persuade individuals who
perceive invulnerability to health threats.

SREEAEI

Powell (1965) stated that “anxiety appeals will change
attitudes significantly only if explicitly directed at the
listener or at those with whom he is personally and closely
involved“ (p. 106, emphasis added). Indeed, the findings from
this research suggest that relational closeness (as indicated
by communication with partner) is an important component in
emotional arousal, as well as intention change. Welch Cline
and McKenzie (1994) also have suggested that indirect

persuasion:may'be a viable and effective persuasive route for

64
HIV prevention because women tend to perceive more
vulnerability to HIV than do men.

Thus, although the idea of indirect persuasion is not
new, the need for it has begun to be recognized by researchers
so that perceived invulnerability can be circumvented and
protective health actions can be pursued. This research
represents a first attempt at examining this indirect
persuasive process, and the results appear promising. These
data indicated that indirect emotional arousal can occur, and
that peOple can be motivated to take behavioral action when
they are not directly vulnerable to a health threat. The
question to be explored by future research is how best to take

advantage of this indirect process.

NOTES

1The legal definition of rape varies by state, and many terms
are used interchangeably. See Bohmer and Parrot, p. 4, 1993,
for discussion.

2Health status objective number 7.7: “Reduce rape and
attempted rape of women aged 12 and older to no more than 108
per 100,000 women. (Baseline: 120 per 100,000 in 1986).
These numbers are likely underreported as “victims of
acquaintance and date rape and other rapes that do not meet
common stereotypes often fail to define their experience as
rape“ (pp. 234-235, Healthy People 2000).

65

APPENDICES

APPENDIX A

STIMULUS MESSAGES

Female Low Threat Message

FIGHTQQ RAPE I§ QM Q's CRUSADE

Rape
she met him.at a party she asked him.on a date
she took him.to dinner
she flirted she drove him home she kissed him
she went inside she kissed
him again she said no he said yes you will

is he guilty?
(Example ad from Charles Hall)

“Charles Hall has spent $5,000 on a series of 15 stickers and
13 posters that challenge men to reconsider any cavalier
attitudes toward rape, a subject that ‘usually is just talked
about in the locker room in terms of - do you think Mike Tyson
is guilty?-'

Why this costly leap from thinking to doing the right thing?
‘I want to make people understand there's a line. And no
woman, no matter how turned on she may be, is asking for
someone to do that to her.’

The provocative messages so far have not been picked up by
anyone in the mood to donate free ad space. But they have
garnered write-ups in.Elle,'Vibe, The Paper, and.Playboyu ‘We
get a lot of flak for objectifying women, and we thought
(Hall's) images were sexually charged but showed that we don't
condone (rape),' says Terry Glover, Playboy assistant editor.
She says many women have written letters in support of the
January spread, which included posters of women in provocative
poses. The line, ‘This is not an invitation to rape me,’
appears in places where the eye is naturally drawn. All the
phrases came from Hall, ‘just things that I either read or am
feeling.’

But feminist activist Holland Utley told Elle that Hall's
campaign is questionable, since it ‘still serves women up as
the delicious object.’ Hall counters that the images he uses

66

67

are ‘lifted from ads and fashion layouts in women's magazines
. . It’s what’s out there.’

Others think Hall is on the right track.

‘I think his campaign can reach men and boys who never would
think of themselves as rapists,’ said Karel Amaranth of the
Women’s Action Alliance. ‘I’m surprised he hasn’t gotten
anyone to run the ads. If I could raise the money, I’d pay to
run them.’

‘This is a very innovative way to get the message out.
Remember, most men don’t listen to what women say, they listen
to other men’, says Cassandra Thomas, director of the Rape
Crisis Program at the Houston Area Women’s Center and last
year’s president of the National Coalition Against Sexual
Assault.

‘The myth out there is that if you behave like x or dress like
Y, then you deserve it,’ she says. ‘What we do here is Band-
Aid work, but it’s really changing the message about rape that
is important, and that’s what a campaign like this can do.’

Hall hasn’t given up on free ad space, and in May starts
production with donated funds and time on a series of TV
spots. He also is hoping to get colleges interested in
printing and distributing the stickers and posters on campus.

Thomas thinks junior high school would be just as appropriate.
‘We need to alert kids young, get them to think in terms of
checking their level of entitlement’ to a girl when on a date,
she says. As for whether one man can make a difference:
‘This is the way we handled racism in the ’60s. It wasn’t
corporations that got together, it was one person getting fed
up, stepping on a bus and going to demonstrate.’

Although Hall considers himself someone ‘who stands up for
something (I) feel’, he admits that his personal crusade has
landed him in some situations unfamiliar to the average ad
executive, like getting stopped by police while stickering a
New York subway car. (LISA TQM, 3/28/95)

It is important that you do not become paranoid about the idea
of being sexually assaulted - you are not under siege - though
at the same time it is necessary for you to exercise a
reasonable degree of awareness and forethought with respect to
your environment and with respect to your actions. While
sexual assault may not be a problem on your campus (and
indeed, the sexual assault numbers have declined compared to
previous years), one thing that you may want to consider is
enrolling yourself in a self defense or martial arts course.
Not just a one time 2 hour workshop, but a continuing course.

68
Consider the facts:

1. Statistics show that people who are trained in self
defense are less likely to be chosen as victims because they
act differently and don’t put themselves in potentially
dangerous situations.

2. The National Center for Prevention and Control of Rape
reports that women who fight, yell and attelpt to flee are
more successful in avoiding rape than those who talked, cried,
tried to appeal to the assailant, or tried to make themselves
appear less desirable.

3. Completing a course (rather than a short workshop) trains
you to react automatically and instinctively to defend
yourself in attack situations.

4. Many self defense and martial arts classes are available
on or around campus, including:
PES 106L - Self Defense through the physical education
department
MSU Karate Club - 355-0822
Okinawan Karate Club at the Michigan Athletic Club -
337- 0002
Self Defense Workshop sponsored by intramural and
recreative services 355-9991 ext. 373 or 355-5250

Please enroll in a self defense course today, either by
yourself, with a female friend, or your boyfriend. It will
protect you and give you confidence. Do it today and do it
for yourself!

69

Female High Threat Message

FI RAPE I ' E

“Two recent sexual assaults at MSU have illustrated a
traditional college fear turned into reality. The victim of
an attempted rape early Sunday morning never saw her
attacker’s face, but police continue to search for him. While
the 19-year old West Wilson Hall resident was walking home
from an East Lansing party, she was tackled and groped by a
tall, heavy set man near Landon field, said Detective Alicia
Spalding of the MSU Department of Police and Public Safety.

‘The victim said she was fighting with him the whole time,’
Spalding said. The woman managed to kick the man in the back
several times, but it was voices nearby that scared him away,
Spalding said.

When the assailant fled toward the IM Sports-Circle, the woman
ran to the Sparty statue, where two men passing by escorted
her to Case Hall, Spalding said. A resident assistant from
Wilson Hall reported the attack to police.

Spalding said this particular incident strays from other cases
of sexual assault this year in that the perpetrator was a
stranger. In a majority of sexually related attacks, the
victim knows or has a relationship with the assailant.

Just this month, a woman was allegedly assaulted by an MSU
student . . . who knew her attacker and was locked in his dorm
room. He made sexual advances toward her and covered her
mouth when she tried to scream.

Regardless of the relationship between the victim and the
attacker, sexual assault is damaging to a victim’s sexual and
emotional growth, said Bonnie, a volunteer staffer for the MSU
Crisis Line. The call-in line provides counseling and legal
and medical advocacy for victims of sexual assault and
harassment.

‘Psychologically, their emotional development gets fixed at
the age when (the attack) occurs, ’ said Bonnie. ‘Their sexual
development is sort of impeded.’

Although sexual assault is widespread, it is difficult if not
impossible to approximate the total number of incidents at
MSU.

Last year, there were seven reported incidents of criminal
sexual conduct on campus. In 1993, the records show 14
reported cases. However, not all cases are reported to the
police.

70

‘I’ve had a lot of people calling in but not wanting to report
it - a lot of guilt comes out, and they don’t feel they have
the right to report it,’ Bonnie said.’ (The State Mews,
3/22/95)

Realize that somebody else won’t always be available to
protect you. There may come a day when the only person you
have to rely on is yourself. Would you be ready?

Nobody likes to think that situations like the ones described
above can happen, but these are true stories and they do
happen. IT CAN HAPPEN TO YOU. College campuses are not
immune from violence, and as much as you might like to believe
otherwise, MSU is not immune from violence and neither are
you. College women represent the age group of women most
likely to be victimized by rape. consider the facts:

1. One out of four women will be raped at some time during
her life. That means that if you know four women, one of them
may be sexually assaulted at some time in her life. This
includes girlfriends, sisters, and yourself.

2. Women are more likely to be sexually assaulted by
acquaintances than by strangers. This is especially
problematic, because women tend to “let down their guard“ when
they are around people they know, and this is exactly what
they should pee do.

3. In a 1991 nationwide survey of 2400 colleges, Michigan
State University reported the 2nd highest number of sexual
assaults on campus.

4. Statistics from the sexual assault crisis line at MSU

indicate that there were ewo ingepgeene cells per day

regarding sexual assault in 1992.

5. In a recent study, 6595 of college educated males said they
would rape a woman if they thought they could get away with
it.

It is important that you do not become paranoid about the idea
of being sexually assaulted - you are not under siege - though
at the same time it is necessary to exercise a reasonable
degree of awareness and forethought with respect to your
environment and actions. While there are many answers to the
problem of rape, one answer is for you to take action for
yourself. Don’t wait until it’s too late. One of the best
things you can do for yourself is to get enrolled in a self
defense or martial arts course now! Statistics show that
people who are trained in self defense are leee likely to be
choegp ag vic§i_m§ because they act differently and don’t put
themselves in potentially dangerous situations. The National

71

Center for Prevention and Control of Rape reports that women
who fight, yell and atte-pt to flee are more successful in
avoiding'rope than those who talked, cried, tried to appeal to
the assailant, or tried. to :make themselves appear less
desirable. Additionally, completing a course (rather than a
short workshop) trains you to react automatically and
instinctively to defend yourself in attack situations. Many
self defense and martial arts classes are available on or
around campus, including:

PES 106L - Self Defense through the physical education
department

MSU Karate Club - 355-0822

Okinawan Karate Club at the Michigan Athletic Club -
337-0002

Self Defense Workshop sponsored by intramural and
recreative services 355-9991 ext. 373 or 355-5250

Please enroll in a self defense course today, either by
yourself, with a female friend, or with your boyfriend. It
will protect you and give you confidence. Don’t count on
other people to be there for you, do it for yourself!

72

Male Low Threat Message

FIGHTQQ RAPE IS ONE MAN'S CRUflE

“Radical feminists aren’t the only ones talking about the rape
crisis anymore. Since the mid eighties the media have been
kindling public interest in rape with a series of
revelations.

“According to the widely quoted Me. survey, one in four
college women is the victim of rape or attempted rape. One in
four. I remember standing outside the dining hall at Harvard
looking at a purple poster with this statistic written in bold
letters. It didn’t seem right. If sexual assault was really
so pervasive , it seemed strange that the intricate gossip
networks hadn’ t picked up more than one or two shadowy
instances of it. If I was really standing in the middle of an
epidemic, a crisis, if 25 percent of my female friends were
really being raped, wouldn’t I know it?

Neil Gilbert, professor of social welfare at the University of
California at Berkeley, has written several articles attacking
the two sociological studies that are cornerstones of the
rape-crisis movement, the Me. magazine study and one done in
the early eighties by Diana Russell. Having taken a closer
look at the numbers he questions the validity of the one-in-
four statistic. He points out that in the Me. study, which is
the one most frequently quoted, 73 percent of the women
categorized as rape victims did not define their experience as
“rape“ . It was Dr. Mary Ross, the psychologist conducting the
study, who did. These are not self-proclaimed victims, then -

these are victims according to someone else. From Koss’s
point of view, these women were suffering from what they used
to call false consciousness. The way it is usually and
tactfully phrased these days is that they don’ t recognize what
has really happened to them.

Gilbert also points out that 42 percent of the women
identified in this study as rape victims later had sex with
the man who supposedly raped them after the supposed rape .

. According to Gilbert, in the Me. study one of the questions
used to define rape was ‘Have you ever had sexual intercourse
when you didn’t want to because a man gave you alcohol or
drugs?’ The strange phrasing of this question itself raises
the issue of agency. Why aren’t college women responsible for
their own intake of alcohol or drugs? A man may give a woman
drugs, but she herself decides to take them . . . Many of
these instances, as Gilbert points out, are simply too vague
for statistical certainty. Classifying a positive answer to
Koss’s ambiguous question as rape further explains how she
could have reached the conclusion that one in four women on
college campuses has been raped.

73

As Gilbert points out, at Berkeley, a campus with 14,000
female students, only 2 rapes were reported to the police in
1990, and between 40 and 80 students sought assistance from
the campus rape-counseling service. Even if we assume that
many students don’t report rapes, even to the sympathetic
rape-crisis center, the one-in-four statistic would still
leave thousands of rapes unaccounted for.

At George Washington University a few years ago, a student was
caught inventing a rape story. ‘Mariam’ , a sophomore who
worked in a rape crisis center, told a story about ‘two
muscular-looking males in torn dirty clothing’ raping a female
student. She later admitted to fabricating the story and
wrote in a letter of apology that ‘my goal from the beginning
was to call attention to what I perceived to be a serious
safety concern for women’ . The fabricated rape was not just
a lie, but a lie promoting stereotypes.“ (Katie Roiphe,
1993) .

It is important that women do not become paranoid about the
idea of being sexually assaulted - they are not under siege -
though at the same time it is necessary for them to exercise
a reasonable degree of awareness and forethought with respect
to their environment and actions. While sexual assault may
not be a problem on your campus (and indeed, the numbers have
declined compared to previous years) , one thing that women may
want to consider is enrolling in a self defense or martial
arts course. Consider the facts:

1. Statistics show that people who are trained in self
defense are less likely to be chosen as victims because they
act differently and don’t put themselves in potentially
dangerous situations.

2. The National Center for Prevention and Control of Rape
reports that women who fight, yell and attupt to flee are
more successful in avoiding rape than those who talked, cried,
tried to appeal to the assailant, or tried to make themselves
appear less desirable.

3. Completing a course (rather than a short workshop) trains
you to react automatically and instinctively to defend
yourself in attack situations.

4. Many self defense and martial arts classes are available
on or around campus, including:

PES 106L - Self Defense through the physical education
department

MSU Karate Club - 355-0822

Okinawan Karate Club at Michigan Athletic Club - 337-0002
Self Defense Workshop sponsored by intramural and

74

recreative Services - 355-9991 ext. 373 or 355-5250

Get a woman you care about enrolled in a self defense or
martial arts course now! .At least talk with a woman you care
about regarding getting her enrolled in a self defense course,
or better yet, enroll in the course Segether. You may have to
bring up the topic more than once in order to convince her,
but it will be worth it. It will protect her, and protect

your relationship with her.

75
Male High Threat Message

“Sometimes an event gets under your skin. And other times it
creeps deep into your brain. ‘Charles’ (a student) threw a
party a year ago at his apartment at MSU. Three women friends
of his left with men in tow, men they had met at his party.
‘Charles’ wished them well, wondering if he had the match-
making touch. Instead, one woman was raped by her new date.
A complaint was filed, but ‘Charles’ was overcome by a
crushing sense of responsibility. ‘Beating the (rapist’s)
brains out was tempting, but I knew it wouldn’t solve the
problem’, says Hall . . . I just kept thinking of my mother,
my girlfriend, my cousins, thinking-I don’t want them to
suffer from this, so what can I do? . . .I want to make people
understand there’s a line. And no woman, no matter how turned
on she may be, is asking for someone to rape her.’“

Realize that you won’t always be available or able to protect
someone you care about, as this woman’s experience
illustrates:

“On my 21st birthday at MSU, my roommate had a party for me.
One guest was a friend of my boyfriend, a man who I made
’small talk’ with. At 2:30 a.m., he and a few other guests-
including my boyfriend - were out front talking, but I was
tired so I said goodnight and went to bed. I woke up with
someone on top of me, shoving something up inside me. The
panic I felt was incredible: I knew everybody had still been
up when I went to sleep, and all I could think was, ’this man
has killed everybody, and now he’s going to kill me!’ After
a struggle that seemed like an eternity, the man lost his
balance on my waterbed. Only when I opened the door to escape
and saw him in the light did I recognize who it was - he had
broken back into the house while my boyfriend slept
downstairs, it was his friend. Since then, my relationship
with my boyfriend has been different and we go to counseling.
I am different now, and our relationship will never be the
same. It’s hard for me to even let my boyfriend hold me now.
The only thing worse than the attack was that my boyfriend was
sleeping downstairs while it happened. He feels guilty about
not being there to defend and protect me. You always think
that you’ll be able to protect yourself, but until you are
faced with a situation like the one I was in, you really don’t
know. That’s why my boyfriend and I have enrolled in a self
defense course together. We both are learning how to defend
ourselves, and I think its helping us recover from that awful
night.“

Nobody likes to think that situations like the ones described
above can happen, but these are true stories and they do
happen. IT CAN HAPPEN TO SOMEONE YOU KNOW. College campuses
are not immune from violence, and as much as you might like to

76

believe otherwise, MSU is not immune from violence and neither
are you. Consider the facts:

1. College women represent the age group of women most likely
to be victimized by rape.

2. One out of four women will be raped at some time during
her life. That means that if you know four women, one of them
may be sexually assaulted at some time in her life. This
includes girlfriends, sisters, relatives and

close female friends.

3. Women are more likely to be sexually assaulted by
acquaintances than by strangers. This is especially
problematic, because women tend to “let down their guard“ when
they are around people they know.

4. In a 1991 nationwide survey of 2400 colleges, Michigan
State University reported the 2nd highest number of sexual
assaults on campus.

5. Statistics from the sexual assault crisis line at MSU
indicate that there were ewo indepenth calle per day

regarding sexual assault in 1992.

6. In a recent study, 6595 of college educated males said they
would rape a woman if they thought they could get away with
it.

7. This month, a woman was sexually assaulted by an MSU
student.

While there are many answers to the problem of rape, one
answer is for you to persuade a woman you care about to enroll
in a self defense or martial arts course now! Statistics show
that people trained in self defense are leee likely Se Me
chosen ae viceims because they act differently and don’t put
themselves in potentially dangerous situations. The National
Center for Prevention and Control of Rape reports that women
who fight, yell and attempt to flee are more successful in
avoiding rape than those who talked, cried, tried to appeal to
the assailant, or tried to make themselves appear less
desirable. Additionally, completing a course (rather than a
short workshop) trains you to react automatically and
instinctively to defend yourself in attack situations. Many
self defense and martial arts classes are available on or
around campus, including:

PES 106L - Self Defense through the physical education
department

MSU Karate Club - 355-0822

Okinawan Karate Club at Michigan Athletic Club - 337-0002

APPENDIX B

SCALE RELIABILITIES

mas Loni—tens

attitudes

post-attitude

follow up attitude

attitude change

intentions (male/both)
intentions (male/her)
post-intentions (male/both)
post-intentions (male/her)
f/up intentions (male/both)
f/up intentions (male/her)
intentions (female/both)
intentions (female/self)
post-intentions (female/both)
post-intentions (female/self)
f/up intentions (female/both)
f/up intentions (female/self)

beliefs about rape

misogyny
closeness (talk)

sadness
anger
fear
worry

severity

susceptibility

response efficacy

self efficacy
minimization

perceived manipulation
defensive avoidance
quality

plausibility

perceived invulnerability
f/u perc. invulnerability

77

wwwwwwwwwwwwamme

thH
4

common

hibihUllblhIhibwihw

Cronbach’ e alpha

.81
.88
.87
.74
.88
.95
.92
.98
.95
.98
.96
.94
.96
.96
.95
.94

.83
.68
.74

.90
.95
.97
.86

.70
.83
.85
.79
.92
.93
.71
.92
.89
.89
.91

APPENDIX C

BELIEFS ABOUT RAPE SCALE

The following items are from the Attitudes Toward Rape
Victhms Scale. Responses ranged from 1 - strongly disagree

to 7

1.

2.

The following

- strongly agree.
Women often claim rape to protect their reputations.
Most women secretly desire to be raped.

Claims of rape by strippers and prostitutes should be
viewed with suspicion.

Women put themselves in situations in which they are
likely to be sexually assaulted because they have an
unconscious wish to be raped.

If a woman leads a man on, he should not be blamed if a
rape occurs.

Women often really mean “yes“ or “maybe“ when they say
“no“.

In most rape against women, the women are promiscuous or
have bad reputations.

items were from. the Rape Empathy Scale.

Responses ranged from 1 - strongly prefer A) to 7 - strongly
prefer B).

A) I feel that the situation in which a man compels a
woman to submit to sexual intercourse against her will
is an unjustifiable act under any circumstances.

B) I feel that the situation in which a man compels a
woman to submit to sexual intercourse against her will
is a justifiable act under certain circumstances.

A) I would find it easier to imagine how a rapist might
feel during an actual rape than how a rape victim
might feel.

B) I would find it easier to imagine how a rape victhm
might feel during an actual rape than how a rapist
might feel.

78

10.

11.

12.

13.

14.

15.

16.

A)

B)

A)

B)

A)

B)

A)

B)

A)

B)

A)

B)

A)

B)

79

Under certain circumstances, I can understand why a
man would use force to obtain sexual relations with
a woman.

I cannot understand why a man would use force to
obtain sexual relations with a woman under any
circumstances.

In a court of law, I feel that the rapist must be held
accountable for his behavior during the rape.

In a court of law, I feel that the rape victim must be
held accountable for her behavior during the rape.

When.a woman dresses in a sexually attractive way, she
must be willing to accept the consequences of her
behavior, whatever they are, since she is signalling
her interest in having sexual relations.

A woman has the right to dress in a sexually
attractive way whether she is really interested in
having sexual relations or not.

I believe that it is impossible for a rape victim.to
enjoy being raped.

I believe that it is possible for a rape victim to
enjoy the experience of being raped, whether she
admits it or not.

I feel it is impossible for a man to rape a woman
unless she is willing.

I feel it is possible for a man to rape a woman
against her will.

Once a couple has had sexual intercourse, then that
issue is resolved and it is no longer possible for
that man to rape that woman.

Even if a couple has had sexual intercourse before,
if the man forces the woman to have sexual intercourse
with him.against her will, this should be considered
rape.

I can understand a wife’s humiliation and anger if her
husband.forced.her to have sexual relations with.himu

A husband has every right to determine when sexual
relations with his wife occur, even if it means
forcing her to have sex with him.

80

17. A) If I were a member of a jury in a rape trial, I would
probably be more likely to believe the woman’s
testimony than the man’s, since it takes a lot of
courage on the woman’s part to accuse the man of rape.

B) If I were a member of a jury in a rape trial, I would
probably be more likely to believe the man’s testimony
than the woman’s, since rape is a charge that is
difficult to defend against, even if the man is
innocent.

Item #s 1,2,3,4,5,6,7,8,ll,13,16,17 were recoded, thus high
scores indicate high empathy for rape victims and less
endorsement of stereotyped beliefs about rape.

LIST OF REFERENCES

LIST OF REFERENCES

Anderson, V.N. (1992). For whom is this world just?:
Sexual orientation and AIDS. Seurnal of Applied Social
Psyehelogy, 22. 248-259.

Beck, K.H., & Frankel, A. (1981). A conceptualization of
threat communications and protective health behavior. Social

Psychology Qparterly, 11, 204-217.

Berscheid, E. (1983). Emotion. In H.H. Kelley, E.
Berscheid, A. Christensen, J.H. Harvey, T.L. Huston, G.
Levinger, E. McClintock, L.A. Peplau, & D.R. Peterson, Slose
Relapionehips (pp. 110-168). New York: Freeman.

Berscheid, E., Snyder, M., a Omoto, A.M. (1989). The
relationship closeness inventorye .Assessing the closeness of

interpersonal relationships. Sournel of Personeligy and
Sociel Peycholegy, S1, 792-807.

Bodenhausen, G.V. (1993) . Emotions, arousal, and
stereotypic judgments: A heuristic model of affect and
stereotyping. In.D.M. Mackie & D.L. Hamilton (Eds.), Affect,

eogpipien, epd gpereepyping. San.Diego, CA: .Academic Press.

Boster, F.J., & Mongeau, P. (1984). Fear-arousing
persuasive messages. In R.N. Bostrom, & B.H. Westley (Eds.),

Communic ion.Y r ook (Vol. 8, pp. 330-375). Beverly Hills:
Sage.

Brownmiller, S. (1975). Ageing; epr will; Men, womep and
pepe. New York: Simon 5 Schuster.

Buchwald, E., Fletcher, P.R., & Roth, M. (1993). Are we
really living in a rape culture? In E. Buchwald, P.R.

Fletcher, & Roth, M (Eds.), Trepefgrming e rape culture (pp.
7-10). Minneapolis, MN: Milkweed Editions.

Burnett, J.J. (1981). Internal-external locus of control
as a moderator of fear appeals. Seprnel of Applied

25191191491. 5.6. 303-393 -

81

82

Burt, M.R. (1978). Appippdee supportive of rape in
american eplpere. House Subcommittee on Science and

Technology, Subcommittee on Domestic and International
Scientific Planning Analysis, and Cooperation, Research into
Violent Behavior: Sexual Assaults (Hearing, 95th Congress, 2nd
Session, January 10-12, 1978). Washington, D.C.: Government
Printing Office, 277-322.

Burt, M.R. (1980). Cultural myths and supports for rape.
Journel of Pereonality end Sociel ngchology, SS, 217-230.

Burt, M.R. (1991). Rape myths and acquaintance rape. In
A. Parrot & L. Bechhofer (Eds.) . Acgeaintance rape: The

hieeep erime (pp. 26-40). New York: Wiley and Sons.
Burt, M.R., r. Albin, R.S. (1981). Rape myths, rape

definitions, and probability of conviction. gournel of
Applied SocielcPsychology, 11, 212-230.

 

Clark, L., a Lewis, D. (1977). Repe-= The price of
eoereive peppelipy. Toronto: The Women’s Press.

Connors, J., 8 Heaven, P.C.L. (1989). Belief in a just
world and.attitudes toward AIDS sufferers. Journal of Social

Psychelegy, 139, 559-560.

Darke, S. (1988) . Anxiety and working memory capacity.
Sogpitien epd Emopion, 3, 145-154.

Davis, M.H. (1983). Measuring individual differences in
empathy: Evidence for a multidimensional approach. Journal

of Pereenalipy Peyeholegy and Seeiel Peyehelogy, 11, 113-126.

Dillard, J.P. (1993). Reehinking the etudv of fee;
appeals; An epepional perepective. Paper presented at the

annual meeting of the International Communication Association,
Washington, D.C.

Federal Bureau of Investigation, United States Department

of Justice. W. 1 91. August 30, 1992.
pp. 14, 17, 24, 58, Table 1. -Index of Crimes, United States,

1972-1991.

Feild, H.S. (1978). .Attitudes toward rape: .A comparative
analysis of police, rapists, crisis counselors, and citizens.

Soernel ef Pereenelipy epe Seeiel Peycholegy, 36, 156-179.

Fishbein, M., a Ajzen, I. (1975). Belief, eppippee,

inteppien, EM peeevier: An ingroepcpien to Sheen and
reseerch. Reading, MA: Addison-Wesley.

 

83

Furnham, D., a Proctor, E. (1988). Supt world.beliefe and
AID- Avli i lsu ofanwmulidimeninal

werle pelief eeele. Unpublished paper.

Gidycz, C.Ar, a Koss, M.P. (1991). The effects of
acquaintance rape on the female victim. In A” Parrot a L.

Bechhofer (Eds.), Acgeeintepce rape: The hidden crime (pp.
270-283). New York: Wiley and Sons.

Goldstein, M.J. (1959). The relationship between coping
and avoiding behavior and response to fear-arousing

propaganda. Seurnel ef Spneppel and Soeiel Ppyeholegy, SS,
247-252.

Gottlieb, N.H., a Green, L.W. (1979). Life events, social
networks, life-style, and health: An analysis of the 1979
National Survey of Personal Health Practices and Consequences .

Health Educapion Qearterly, 13, 91-105.

Griffin, S. (1971). Rape: The all-american crime.
Seppeppe, September, 26-35.

Gur, R.C., Gur, R.E., Skolnick, B.E., Resnick, S.M.,
Silber, F. I., Chawluk, J., Muenz, L., Obrist, W.D., &
Reivich, M. (1988). Effects of task difficulty on regional
cerebral blood flow: Relationships with anxiety and
performance. Peychephysiology, 24, 392-399.

 

Hale, J.L., a Dillard, J.P. (1995). Fear appeals in.health
promotion campaigns: Too much, too little, or just right? In
E. Maibach a R.L. Parrott (Eds.), Deeigping heelth meseages
(pp. 65-80). Thousand Oaks, CA: Sage Publications.

Hansen, WuB., Raynor, A.E., & Wolkenstein, B.H. (1991).
Perceived personal immunity to the consequences of drinking
alcohol: The relationship between behavior and perception.

gournal of Beeevieral Medieine. 11. 205-224.

Hewgill, M.A., & Miller, G.R. (1965). Source credibility
and response to fear-arousing communications. Speech
Moneseanhs. 32. 95-101.

Houston, D.A. (1990). Empathy and.the self: Cognitive and
emotional influences on the evaluation of negative affect in

others. Seprnel ef Pereeneliey end Soeiel ngchology, 52.
859-868.

Hunter, J._E. , & Gerbing, D .W. (1982) . Unidimensional
measurement, second order factor analysis and causal models.
In B.M. Straw and L.L. Cummings (Eds.). Reeeereh in

orgepizepiepel peeevier (Vol. 4), Greenwich, CT: JAI Press.

84

Janis, I.L. (1967). Effects of fear arousal on attitude
change: Recent developments in theory and experimental

research. In L. Berkowitz (Ed.), Aevaneee i e_xperimeptal
eeeiel peyehology (Vol.3, pp. 166-225). New York: Academic
Press.

Janis, I.L., & Feshbach, S. (1953). Effects of fear-

arousing communications. The Journal of Abnormal and Social
Psychology, 4_8_, 78-92.

Janis, I.L., & Feshbach, S. (1954). Personality
differences associated with responsiveness to fear-arousing
communications. Journal of Personality, 23, 154-166.

 

Janis, I.L., a Leventhal, H. (1968). Human reactions to
stress. In E. Borgatta a W. Lambert (Eds.), H ook of

pereenelity pheog and reeeerch. Chicago: Rand McNally.

Janis, I.L., 6: Mann, L. (1965) . Effectiveness of emotional
role playing in modifying smoking habits and attitudes.

Jeurnel ef E_xperimgpel Reeeereh in Persenelipy, _:l_., 84-90.

Janis, I.L., & Terwilliger, R.F. (1962). An experimental
study of psychological resistances to fear-arousing
communications. Sournal of Abnormal and Social Psychology,
5.5. 403-410.

Kelley, H.H. , a Thibaut, J.W. (1978) . Inpeppereonal
R l i ns- A h f in r dence. New York: John
Wiley.

Kleinke, C.L., & Meyer, C. (1990). Evaluation of rape
victims by men and women with high and low belief in a just

world. PSyehelegy ef Wmen wereerly, 14, 343-353.

Koss, M.P. (1985). The hidden rape victim; Personality,
attitudinal, and situational characteristics. Psychology of

Womg mergerly, 2, 193-212.

Koss, M.P., Gidycz, C.A., R Wisniewski, N. (1987). The
scope of rape: Incidence and prevalence of sexual aggression
and victimization in a national sample of higher education

students. Jeernel ef Ceneplping and Qlinicel Peychology, _5_5,
162-170.

 

Koss, M.P., Heise, L. & Russo, N.R. (1994). The global

health burden of rape. W. 1;.
509-537.

Lauer, F., & Newlin, G. (1983). Pe r or in
in f w i h . Paper presented at the American

Public Health Association Conference, Dallas, Texas.

85

Lerner,‘M.J. (1965). Observer’s evaluation of performance
as a function of performer’s reward and attractiveness.

Journal of Pereenalipy and Soeial Peychology. 1. 355-360.

Leventhal, H. (1970). Findings and theory in the study of
fear communications. In L. Berkowitz (Ed.), Advgces in

eMperimental eocial peychology. vol. 5 (pp. 119-186). New
York: Academic Press.

 

Leventhal, H. (1971). Fear appeals and persuasion: The
differentiation of a motivational construct. American Sournal

of gpplic Healph. 51. 1205-1224.

Leventhal, H., Singer, R., E Jones, S. (1965). Effects of
fear and specificity of recommendation upon attitudes and

behavior. Journal of Personalipy and Social Psychology, ;,
20-29.

Leventhal, H., 5 Watts, J. (1966). Sources of resistance
to fear-arousing communications on smoking and lung cancer.

Jeurnel of Pergonelipy, 31. 155-175.

Madhubuti, H.R. (1993). On becoming anti-rapist. In E.
Buchwald, P.R. Fletcher, & Roth, M. (Eds.) Treneforming e rape
culpure, (pp. 165-178) . Minneapolis, MN: Milkweed Editions.

McGuire, W.J. (1968). Personality and susceptibility to

social influence. In E. Borgatta & W. Lambert (Eds.).
Hepepeek ef pereonelipy pheopy ane reeeereh (pp. 1130-1187).

Chicago: Rand McNally.

McGuire, W.J. (1969). The nature of attitude and.attitude
change. In G. Lindzey a E. Aronson (Eds.), The handboek of
sociel peychelogy (Vol. 3, pp. 136-314). Reading; MA:
Addison-Wesley.

Miller, G.R. (1963). Studies on the use of fear appeals:

A summary and analysis. Central Stetee Speech JOurnel, 12.
117-124.

Miller, G.Rr a Hewgill, M.A. (1966). Some recent research
on fear-arousing message appeals. Speeeh Monegraphe, 33, 377-
391.

 

Morrison, K. (1994) . The infleence of e_xperigce QM prior
knowledge on the reception of persuasive fear appeals: A test

ef the extended perellel preceeg model. Paper presented at
the annual meeting of the Speech Communication Association,
New Orleans, LA.

86

Morrison, K. (1995) . Telking wiph male ge fflle college
Students abeu; preventing rape: Insighte frem a focue group
eiecpesion. Unpublished manuscript. Department of

Communication, Michigan State University, East Lansing.

Perloff, L.S., a Fetzer, B.K. (1986) . Self-other judgments
and perceived vulnerability to victimization. Journal of

Persenelipy and Social Peychology, 0, 502-510.

Powell, F. A. (1965). The effect of anxiety-arousing
messages when related to personal, familial, and impersonal

referents. Speech Menegrephe, 1;, 102-106.

Roiphe, K. (1993). The morning after: Soggy fe_ar. and
feminism on ceppee. New York: Little Brown.

 

Rogers, R.W. (1975). A protection motivation theory of

fear appeals and attitude change. Seemel ef Peyehelegy, 3,
93-114.

Salovey, P., O’Leary, A., Stretton, M.S., Fishkin, S.A., a
Drake, C.A. (1991). Influence of mood on judgments about
health and illness. In J. Forgas (Ed.), Emotion and Social
MM: (pp. 241-262). New York: Pergamon Press.

Shaw, K. (1993, January 25) . ’U’ sexual assaults up, crime
down. The Spepe New, pp. 1, 11.

Stiff, J.B., Dillard, J.P., Somera, L., Kim, H., Sleight,
C. (1988). Empathy, communication, and prosocial behavior.

Sommgieeeien Menegraphe, _5_5, 198-213.

Sutton, S.R., & Eiser, J.R. (1984). The effects of fear-
arousing communications on cigarette smoking: An expectancy-

value approach. Seurnel ef Behavioral Meeicine, 1, 13-33.

Tamborini, R., Salomonson, K., a Bahk, C. (1993) . The
relationship of empathy to comforting behavior following film

exposure. Communicatigm_gesearch. 20. 723-738.

United States Department of Health and Human Services a

Public Health Service. (1990) . Healthy peeple 2000:
Nepional health premepien gs; eieeeee prevention oejecpives

(DHHS Publication No. PHS 91-50212). Washington, DC: U.S.
Government Printing Office.

Wagstaff, G.F. (1982). Attitudes to rape: The “just

world“ strikes again? Sellepin ef phe Bripieh Peyehelogicel
5.9.9.1551. 3.5. 277-279-

87

Walster, E. (1966) . Assignment of responsibility for an

accident. Soeppel ef Pergenaligy gs; Seciel Peychology, _3_,
73-79.

Ward, C. (1988) . The attitudes toward rape victims scale.
Psycholegy ef Women Qperperly, 12, 127-146.

 

Warshaw, R. (1988). I n ver calle i ra - The Ms.
report; on reeogizing, fighping, ane Surviving gee ane
W- New York: Harper 8. Row.

Weinstein, N.D. (1984). Why it won’t happen to me:
Perceptions of risk factors and susceptibility. Health
Peyeholegy. _3_, 431-437.

Weinstein, N.D. (1987). Unrealistic optimism about
susceptibility to health problems: Conclusions from a

community-wide sample. Seurnal of Behaviorel Meeicine, 1S,
481-500.

Weis, K., & Borges, S. (1973). Victimology and rape: The
case of the legitimate victim. Issues in Cripinology, S, 71-
114.

Welch Cline, R.J., a McKenzie, N.J. (1994). Sex
differences in communication and the construction of HIV/AIDS.

Journel ef Appliee gemmunicegion Reeeareh, 22, 322-337.

 

Will, G. (1993, October). Sex amidst semicolons.
Negsxeek. p- 92-
Wingard, D., & Berkman, L.F. (1985). A multivariate

analysis of health practices and social networks. In S. Cohen

and L. Syme (Eds.), Seeial Sepport and Health (pp. 161-175) .
New York: Academic Press.

Witte, K. (1992) . Putting the fear back into fear
appeals: The extended parallel process model. Sommunication
Monographs. 52. 329-349.

Witte, K. (1994) . Fear control and danger control: A test
of the extended parallel process model (EPPM) . Cflunieepion

Monegrephe, SS, 113-134.

Wolf, N. (1991). The Beeuey flth. New York: William
Morrow.

Wyer, R.S., Bodenhausen, G.V., a Gorman, T.F. (1985).
Cognitive mediators of reactions to rape. Joprnal of

Pereonelipy See Seeiel Peyehelegy. 48, 324-338.

 

88

Zimmerman, R.S., 5. Connor, C., (1989) . Health promotion in
context: The effects of significant others on health behavior

change. Heelph Eepeepien Qperperly, SS, 57-75.