$35....
5.5

:
.. rs

1:? 1.;

2.1.1. 5 I

in... 9.1“.

tﬂst .
x .37
1; .3315
0 .5.

.rv. at

‘ r . .
x . (10.19
V 4...}: J.
i. ca .57..
3.: .25.
u .. 6.2.
2 a

‘ .Lm...‘
a

 

 

.2:
ﬁvﬁ :
. .vtn....:« : ‘-

 

THESIS

tool

 

LIBRARY
Michigan State
University

 

 

 

This is to certify that the

thesis entitled

PERCEPTION AND SUPPORT FOR CENSORSHIP OF INTERNET

PORNOGRAPHY: THE INFLUENCE OF INDIVIDUALISM/COLLECTIVISM
AND INTERNET SELF-EFFICACY ON THE THIRD-PERSON EFFECT

presented by

BYOUNGKWAN LEE

has been accepted towards fulﬁllment
of the requirements for

MA . d . COMMUNICATION
egree 1n

 

 

Major professor

Date ﬁ'IB'Ol

0-7639 MS U is an Afﬁrmative Action/Equal Opportunity Institution

 

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

 

DATE DUE

DATE DUE

DATE DUE

 

‘07:? 1) 6720023

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 c1CIRC/Datepr65-pt15

——

PERCEPTION AND SUPPORT FOR CENSORSHIP OF INTERNET PORNOGRAPHY:
THE INFLUENCE OF lNDIVIDUALISM/COLLECTIVISM AND INTERNET SELF-
EFFICACY ON THE THIRD-PERSON EFFECT
By

Byoungkwan Lee

AN ABSTRACT OF A THESIS
Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of
MASTER OF ARTS
Department of Communication

2001

Professor Ronald C. Tamborini

ABSTRACT
PERCEPTION AND SUPPORT FOR CENSORSHIP OF INTERNET PORNOGRAPHY:
THE INFLUENCE OF INDIVIDUALISM/COLLECTIVISM AND INTERNET SELF-
EFFICACY ON THIRD-PERSON EFFECT
By

Byoungkwan Lee

This study examined the inﬂuence of individualism/collectivism and Internet self-
efﬁcacy on third-person perception in the context of Internet pornography and the
relationship between this third-person perception and support for censorship as its
behavioral consequence. Results of a survey of both US. and South Korean college
students (N = 232) supported the existence of the third-person effect in the context of
Internet pornography. Participants in this study perceived Internet pornography to have a
greater negative effect on others than on themselves (third-person perception), and the
difference was signiﬁcantly associated with support for censorship. The ﬁndings also
confirmed the inﬂuence of individualism/collectivism and Internet self-efﬁcacy on the
third-person perception. Finally, the results of path analysis suggested that the data were
consistent with a causal model in which both individuals’ individualism/collectivism and
Internet self-efﬁcacy inﬂuence third-person perception, which then leads to support for

censorship of Internet pornography.

ACKNOWELEDGEMENT

This thesis would not have been completed without assistance and encouragement
of many people. I would like to be deeply thankful to Dr. Ron Tarnborini, my advisor, for
his support, direction, patience, and encouragement throughout my master studies. I wish
to express my deep gratitude to all my committee members, Dr. Hairong Li and Dr. Stacy
Smith, for their valuable comments. I really appreciate Dr. Yong-sung Lee, Dr. Hyung-il
Kim, and Mr. Ki-jung Lee for their help in collecting data.

Finally, I thank all my family. Specially, I thank my wife, Joung-a, and my son,

Min-young, who have been encouraging me.

iii

TABLE OF CONTENTS

LIST OF TABLES .................................................................................... v

LIST OF FIGURES .................................................................................. vi

INTRODUCTION .................................................................................... 1

CHAPTER 1

LITERATURE REVIEW AND HYPOTHESES ................................................. 5
Internet Pornography ....................................................................... 5
Third-Person Perception and Support for Censorship of Media Content ............ 8
Hypotheses ................................................................................ 15
A Model of Culture and Internet Self-Efﬁcacy’s Impact
on the Third—Person Effect ............................................................. 20

CHAPTER 2

METHOD .................................................................................................................. . 22
Sample........ ......................................................................................................... 22
Procedures ............................................................................................................ 22
Measures .................................................................................................................. 23

CHAPTER 3

RESULTS ............................................................................................. 28
Third-Person Perception and Internet Pornography .................................. 28
Individualism/Collectivism and the Third-person Perception ....................... 29
Internet Self-Efﬁcacy and Third-person Perception ................................. 29
Third-Person Perception and Behavior ................................................. 33
Culture and Internet Self-Efﬁcacy’s Impact on the Third-Person Effect ............ 35

CHAPTER 4

DISCUSSION ....................................................................................... 37

APPENDIX .......................................................................................... 41

REFERENCES ....................................................................................... 59

iv

LIST OF TABLES

Table 1 — Modiﬁed Individualism/Collectivism (INDCOL) Items ......................... 24
Table 2 — Modiﬁed Internet Self-Efﬁcacy Items ............................................. 25
Table 3 — Modiﬁed Support for Censorship on Internet Pornography ...................... 27
Table 4 — Self and Others Effects of Internet Pornography Comparison ................... 28
Table 5 — Pearson Correlation Matrix for Independent Variables and

Dependent variables ................................................................... 30
Table 6 — Hierarchical Multiple Regression Predicting Third-Person Perception . . . . .32

Table 7 — Hierarchical Multiple Regression Predicting Support for Censorship
of Internet Pornography ............................................................... 35

LIST OF FIGURES
Figure l — Initial Path Model ...................................................................... 21

Figure 2 — Path Coefﬁcients among Individualism/Collectivism, Internet Self- Efﬁcacy,
Third-person Perception, and Support for Censorship ............................ 36

vi

INTRODUCTION

The Internet is a double-edged sword that must be handled carefully. While useful
information and a variety of entertainment are widely provided and consumed on the
Internet, there has been a marked increase in sexually explicit material. Alluring text, chat
rooms, bulletin boards, and newsgroups on the Internet have intensiﬁed parental anxiety
about their vulnerable children. In fact, a simple search using the Yahoo or Alta Vista
search engines reveals that there are hundreds of thousands of pornographic sites on the
Internet. Moreover, “the absence of control, censorship, limits, or grading," as Weimann
(2000) expressed, makes the Internet look like “an ideal site for pornography. "

Censorship is seated at the heart of the controversy in the Internet pornography
issue. A point of contention in this debate has focused on the First Amendment right to
Internet content against the protection of minors from sexually explicit material on the
Internet. In Reno v. ACLU (American Civil Liberties Union) (1997), there was a legal
battle between the government and a coalition of business and ﬂee speech advocates. The
constitutionality of the Communication Decency Act of 1996 (CDA), which was passed by
Congress to protect minors from sexually explicit material on the Internet, was taken to the
Supreme Court. The Supreme Court ruled that the First Amendment right to Internet
content must be upheld identifying “two Provisions of the CDA to be unconstitutional due
to their vagueness and chilling effect their application would have on Internet
communications” (Simon, 1998, p.1015). The Court’s decision meant that “any ﬁiture

attempt to regulate sexually explicit Internet transmissions must be drafted with sufﬁcient

speciﬁcity such that no ambiguity exists as to the scope of its enforcement" (Simon, 1998,
p. 101 5).

In the fall of 1999, the Child Online Protection Act was before Congress. The gist
of the act was to impose criminal penalties against commercial Web sites that display
sexually explicit material on the Internet without appropriate measures to exclude minors
(Munro, 1999). The constitutionality of this act is still pending a judicial decision, due to an
appeal by the ACLU and the Electronic Privacy Information Center.

The very core of this debate is the inﬂuence of sexually explicit material or
pornography on individuals. The majority of research on pornography has empirically
supported its harmful effects, at least on women and children, although the scientiﬁc
community has not come to an absolute consensus about its attitudinal and behavioral
effects (Beaver, 2000). Regardless of the lack of consensus about the harmful effects of
pornography, public concern about Internet pornography has been increasing. Perhaps,
people may think that sexually explicit material or pornography on the Internet is more
harmful and dangerous than in traditional media, such as print, video, television, and ﬁlm.
This thinking might be attributed to individuals’ perceptual differences rather than the
actual difference between the effects of Internet pornography and traditional pornography.
The absence of regulation, the ease of access, and the plethora of pornographic images on
the Internet may magnify people’s anxiety about the harmﬁrl inﬂuence of Internet
pornography on minors.

Consequently, this anxiety gives people the perception that Internet pornography
is more harmful and dangerous than pornography in its traditional forms. This collective

anxiety may affect public opinion about Internet pornography, inﬂuence decisions of

policy makers, (Dearing & Rogers, 1996; Page & Shapiro, 1983) and shape judicial
decisions related to the First Amendment right of free speech (Marshall, 1988; Mishler &
Sheehan, 1993). The third-person effect hypothesis can provide a theoretical base to
explain the relationship between public opinion about socially undesirable media content
and their support of censorship for it.

The third-person effect, as Greenberg and Salwen (1996) indicated, is rising "as a
popular model generating a paradigmatic body of empirical research" (p. 76). Since the
initial study of Davison (1983), which suggested that people will tend to overestimate the
inﬂuence of mass communications on the attitudes and behavior of others rather than on
themselves, a number of studies have examined the third-person effect hypothesis. Support
for the third-person effect has been found in research examining the relationship between
people's biased perception of socially undesirable media content’s inﬂuence and their
overwhelming support of censorship in a variety of contexts including pornography and
sexually explicit material (Gunther, 1995; Lee & Yang, 1996), television violence
(Hoffner et. al., 1999; Rojas, Shah, & Faber, 1997), hate speech (Lambe & Shah, 1999),
press coverage (Rucinski & Salmon, 1990; Salwen & Driscoll, 1997), misogynic rap
lyrics (McLeod, Eveland, & Nathanson, 1997), advertising (Youn, Faber, & Shah, 2000),
and general inﬂuence (Gunther & Hwa, 1996). Most research results in this area show
that the individuals’ tendency to perceive greater media inﬂuence on others than on
themselves (the perceptual component of third—person effect) leads to support for
censorship of the media content (the behavioral component of third-person effect).

The main purpose of this study is to extend research on the third-person effect to

the context of Internet pornography. In addition, this study examines the inﬂuences of both

individualism/collectivism and Internet use on third-person perception. Although the likely
inﬂuence of culture might seem a critical factor in explaining individual differences in the
third-person effect, the relationship between individualism/collectivism and third-person
perception has not been empirically tested‘. Beginning from a cross-cultural perspective,
this study examines whether cultural differences, speciﬁcally those distinguishing
individualistic culture and collectivistic culture, affect third-person perception in the
context of Internet pomography and, subsequently, leads to support for its censorship.

This study also considers Internet use as a signiﬁcant recipient factor that may
affect third-person perception. Although literature in this area generally considers media
use as an important recipient factor magnifying the third-person effect (Brosius & Engel,
1996; Chapin, 2000; Paxton, 1997; Price, Huang, & Tewksbury, 1997; Salwen, 1998;
Salwen & Driscoll, 1997; Salwen & Dupagne, 1999), with the exception of only one
study (Price et al., 1997) most research on traditional media surprisingly'fails to support
the assumption that media use is a signiﬁcant predictor of the third-person effect.
However, we might expect the outcome to change when considering differences between
Internet use and traditional media use. For example, while no special skill and knowledge
are needed to watch television or read newspapers, these are required to access the
Internet. As such, we might expect Internet use to inﬂuence the third—person perception

more than is found with traditional media.

 

1 The research on other recipient factors has been well documented: age (Boynton & Wu, 1999; Tiedge et al.,
1991), education (Driscoll & Salwen, 1997; Gunther, 1995; Salwen, 1998), gender (Boynton & Wu, 1999;
Lee & Yang, 1996), and media use (Brosius & Engel, 1996; Chapin, 2000; Paxton, 1997; Price, Huang, &
Tewksbury, 1997; Salwen & Dupagne, 1999). There has been only study (Park, 1998) on the relationship
between cultural factor and third-person effect.

Chapter 1

LITERATURE REVIEW AND HYPOTESES

Internet Pornography

By this time, numerous empirical studies have been conducted regarding exposure
to sexually explicit material in terms of sexual arousal (e. g., Bancroft & Mathews, 1971),
cognitive effects (e. g., Zillrnann & Bryant, 1982; Zillrnann, Bryant, & Huston, 1994;
Zillrnann & Weaver, 1989) and behavioral effects (e.g., Donnerstein & Berkowitz, 1981;
Zillrnann, Bryant, Comisky, & Medoff, 1981)2. However, very few of these studies focuses
on Internet pornography. In addition, most of those studies focusing on Internet
pornography have focused on the legal aspect of regulation (Cate, 1996; Simon, 1998;
Vandiver, 2000) or content of Internet pornography (Mehta & Plaza, 1997; Rimm, 1995;
Sanders, Deal, & Myers-Bowman, 1999). According to content analysis, pomographic
images on the Intemet are not fundamentally different from those existing in magazines or
videos OVIehta & Plaza, 1997; Sanders, Deal, & Myers-Bowman, 1999). As such, beyond
anything resulting from the unique attributes of the Internet, we might expect strong third-
person effects simply due to content.

Nevertheless, attributes of Internet pornography could be a crucial deterrrrining
factor in third-person effects. The popularity of Internet user and access, the incredible
growth of the on-line sex industry, the global scale of dissemination, and the limits of
control and regulation amplify public anxiety about the social impacts of Internet

pornography. Some Internet use reports indicate this concern may be justiﬁed. According

 

2 Bryant and Zillmann (1996) broadly classiﬁed the effects of the consumption of sexually explicit
materials or pornography into these three categories.

to a recent report (Stanford Institute, 2000), over one third of the respondents3 regularly use
the Internet more than ﬁve hours per day, with users who spend more than ten hours per
day comprising a surprising 14 percent. A quarter of regular Internet users (more than 5
hours daily) feel that it has reduced their time with friends and family or attending events
outside the home. In addition, many respondents indicated that their Internet usage has
reduced both their television viewing and newspaper reading time.

The economics of Internet pornography clearly demonstrate that the Internet has
ﬁrmly entrenched itself in our society as a medium that transmits pornographic material.
According to Slade (2000), about one third of the current commercial activity on the
Internet is pornographic. He remarked:

" Time says that adult Web sites collectively bring in nearly $1 billion a year, or

about a tenth of entire sex industry, and predicts that the share will increase. The

chief industry source also reports that pornography (loosely deﬁned) accounts for

69 percent (or roughly $1 billion worth) of total Internet content sales. The same

source maintains that about 84 percent of that 69 percent of content comes from

American sources and sites. Large, hot sites can generate revenues of $1 million a

month. Porn stars Asia Carrera, Jenteal, Mimi Miyagi, and Vanessa Del Rio have

opened Web sites that can bring in up to $5 million a year, in some cases ﬁve times
what they made from porn videos. Obviously these are rough ﬁgures, since if every
site pulled in $5 million consistently then the aggregate numbers would be

astronomical” (pp.238 - 239).

From a practical point, it may be impossible to completely isolate minors from
access to Internet pornography. Although all commercial Internet pomo-sites reqmre
conﬁrmation that the user is an adult, the conﬁrmation screen itself displays pornographic
images provided to attract and titillate new customers. In addition, non-commercial sites for
Internet pornography are a crucial source for pornographic material, although there was a

ﬁnding (Mehta & Plaza, 1997) that commercial vendors generally post more pornographic

images than non-commercial distributors. Adolescents who cannot use credit cards for

 

3 The sample was of 4,1 13 adults in 2,689 households in the US.

accessing commercial pomo—sites on the Internet share pornographic material with each
other on the Usenet, chat rooms, personal sites, and so on.4 Although net ﬁlter software
programs such as SurfWatch and Swaanny are employed to block youth access to adult
pomo-sites, such net ﬁlters are not practical when they are used on public computers in
schools and libraries because they often ﬁlter out more than they were intended to block
(Melillo, 1999).

Several Asian countries are prohibiting the production and dissemination of
pomography or sexually explicit materials by law. South Korea, China, and Singapore
consider Internet pornography as a serious threat to their social order and morals. In South
Korea, for instance, there is no legal basis for punishing the dissemination of pornography
through the Internet if the dissemination occurs outside the country. A recent instance in
South Korea illustrates how Internet pornography becomes a serious social issue in
countries that are prohibiting the production and dissemination of pornography by law. A
video clip showing a top female pop star and a former producer having sex was uploaded
on a commercial Internet pomo-site without her consent, and the video was posted on the
personal web site of a college student who downloaded it by hacking. The video was
shared with countless other users by using FTP and was disseminated at an amazing speed
throughout the country.

"For a short time, there was a cyber-war. Fans of singer Baek Ji- Young tried
to paralyze Internet sites offering the video showing Baek and a former producer
having sex. The Internet site built new ﬁrewalls in defense. In the end, the

voracious nature of the Web could not be denied. The video clip multiplied and
raced across the Internet, reportedly at a rate of 200,000 copies in one day. Back, 24,

 

’ Though at present there is no empirical evidence of this, we can easily observe this fact on any Usenet,
chatting room, or personal site. See Gutfeld (1999) about pornography on chatting room and Mehta and
Plaza (1997) about pornography on the Usenet.

ﬁrst tried to deny it was she. She hid on an island, then capitulated. She held a
tearﬁrl news conference to confess and apologize... .. Since an estimated one-third
of South Korea's 47 million people use the Internet - one of the highest rates in the
world - 'it was like branding her with a scarlet letter,' quipped one observer” (The
Washington Post, 2000).

According to a report (Digital Chosun, 2000), the request for this video clip, throughout
several search engines, placed it among the most frequent search requests for the year 2000.
Internet pornography has proven itself sufﬁcient to attract the concerns of public
opinion and policy makers regardless of the debate of its actual effects. For them, public
anxiety about Internet pornography may be more important than its actual impacts. In other
words, people's perceptions that Internet pornography has harmful effects can inﬂuence
policy makers and judicial rulings regulating pomography on the Internet. The third-
person effect, as noted above, is a theoretical model to explain how people's perceptions

about the effects of Internet pornography impact public opinion about regulating it.

Third-Person Perception and Support for Censorship of Media Content

The third-person effect hypothesis consists of two components, the perceptual
component and the behavioral component. Gunther (1995) noted: "the ﬁrst suggests that
people are prone to a perceptual bias, leading them to estimate that a communication will
have more inﬂuence on others than on themselves. The second component proposes that
people may react in some way according to this estimate of larger effects on others” (p.27).
In this study, the perceptual component of the third-person effect hypothesis refers to
individuals’ tendency to perceive greater Internet pornography inﬂuence on others than

themselves, and the behavioral component refers to individuals’ tendency to support

censorship of Internet pornography on the basis of their perceptual discrepancy between

perceived inﬂuence on self and others.

The perceptual component. The ﬁrst component of the third-person effect
hypothesis is the perceptual discrepancy by which people are more likely to perceive
greater media inﬂuence on others than on themselves. The perceptual discrepancy is the
difference between the degree to which people estimate the effect of mass media on others
and the degree to which people estimate that effect on themselves (Gunther, 1991, 1995).
The hypothesis suggests that most individuals give higher estimates of media’s impact on
the perceptions of others.

The perceptual discrepancy can be explained in terms of biased optimism. Biased
optimism is a tendency in which people believe that negative experiences occur to
themselves less often than others (Weinstein, 1989). Exarrrining why the third-person
phenomenon occurs, Gunther and Mundy (1993) suggested biased optimism as a cause of
perceptual discrepancy’ . They insisted: "One explanation for this bias is ego—enhancement,
that people are motivated to reinforce their self-esteem by seeing themselves as smarter or
better than their peers. Such a self-serving bias offers a simple and persuasive explanation
for the third—person phenomenon” (p.60). The optimistic bias phenomenon, however,

depends on whether the type of message is harmful or beneﬁcial (Gunther and Mundy,

 

5 Recently, on the contrary, a study (Chapin, 2000) did not support this assurrrption. This study empirically
tested a relationship between optimistic bias and third-person perception from survey research with urban
minority at-risk youth. The ﬁndings showed that although a weak relationship between optimistic bias and
the third-person perception was found, the ﬁndings did not support their assumption that optimistic bias
maybe an underlying mechanism of third-person perception. This result may be caused from the use of a
unique sample presented as a limitation of the study. For more discussion, see Chapin (2000).

1993). In other words, while a harmful message should magnify the discrepancy between
the estimated effect on the self and on others, a beneﬁcial message should make no
difference. This reversed third-person effect—that when people judge a message in
communication as a positive or beneﬁcial one, they perceive themselves to be more
inﬂuenced than others would—has been veriﬁed also by many studies (Duck & Mullin,
1995; Duck, Hogg, & Terry, 1999; Duck, Terry, & Hogg, 1995; Eichholz, 1999; Gunther
& Thorson, 1992; Henriksen & Flora, 1999).

Findings consistent with the perceptual discrepancy hypothesis have been observed
in several studies. Cohen and his colleagues (1988) supported claims that negative or
undesirable messages in communication magnify perceptual discrepancy in the context of
libel law. Conducting an experiment, they found that when the source of the defamatory
communication is perceived as negatively biased, the discrepancy between perceived
media inﬂuence on self and on others is greater. In the context of violent and misogynic rap
lyrics, McLeod and colleagues (1997) also strongly supported the third-person perception
hypothesis that the perceived effect of harmful messages on others would be greater than
the perceived effect on self. Through a nationwide telephone survey about the O. J.
Simpson trial, Salwen and Driscoll (1997) supported their third-person perception
hypothesis that "people will perceive press coverage of the O. J. Simpson trial to exert a
greater inﬂuence on other people's opinions than their own” (p.64). Boynton and Wu
(1999) also supported the third-person perception hypothesis in research on TV violence.
Respondents from a telephone survey showed a conspicuous perceptual discrepancy

between the estimates of TV violence's inﬂuence on themselves and others.

10

In the context of pornography and sexually explicit material, two studies show
strong support for the third-person perception hypothesis. Gunther (1995) conducted a
telephone survey on respondents from the Continental US. age 18 or over (N = 648). He
demonstrated clearly that people perceive pornography to wield greater inﬂuence on others
than on themselves. The results showed that "a majority of respondents, 61%, perceived
others to be more negatively inﬂuenced by X-rated material than themselves” (p.33). A
study of Rojas and colleagues (1996) produced the same results. In a survey on college
students (N = 133) at a midwestem university, they found that while only 7.1% of
respondents perceived pornography to have a greater inﬂuence on themselves than others,
71.7% believed that pornography affects others more than themselves.

Several studies attempted to examine the third-person effect hypothesis in a
different cultural context. Lee and Yang (1996) examined the third-person perception
hypothesis of sexually explicit television program in a Korean cultural setting. Since the
dissemination of pornography is technically illegal in South Korea, the term "sexually
explicit" used in this research is different from the term in the US. Thus, what is termed
sexually explicit in their research referred to simple nudity, obscenity, and indecent
sexual content. "Mild nudity" and "sexual violence" were used as the sexually explicit
content in television programs in South Korea. Their ﬁndings in the Korean context
replicated Gunther's observations (1995). Also, they found a gender difference in third-
person perception. While males did not show a perceptual discrepancy between estimates
of mediated sexual content's inﬂuence on themselves and same gender others, females
tended to overestimate the inﬂuence of the mediated sexual content on same-gender others

as compared to on themselves. Although this research has an implication in terms of

11

testing the third-person effect outside the US, differences in the effect across the two
countries, US. and Korea, was not directly compared.

Gunther and Hwa (1996) examined the third-person effect hypothesis in another
Asian cultural context, Singapore. Data were collected from 506 randomly selected
Singaporeans who evaluated ten categories of sensitive television content (violence,
premarital sex, extramarital sex, foul language, illegitimate children, homosexuality,
nudity, men with long hair, foreign programs, and religious programs). They found a
perceptual discrepancy in all content categories. However, while respondents tended to
perceive more negative inﬂuence on others than themselves when they evaluated the
harmful inﬂuence of media content (c. g. sex-related categories, violence, and foul
language), when they evaluated positive inﬂuence (e. g. religious programs), they
estimated the content to wield more inﬂuence on themselves than others. In addition, they
indicated the importance of cultural factors in the third-person effect research. They
maintained:

"People in this study exhibited the perceptual'bias just as strongly and
consistently as have their counterparts in Western cultures where most third-person
effect research has been conducted. At least in the non-Westem context of
Singapore, the majority of people were very much inclined to separate their
conceptions of self from that of others in questions of media inﬂuence” (Gunther
and Hwa, 1996, p.262).

Once again, however, this study did not provide empirical evidence on whether or not

third-person effects differs between the two cultures.

The belgvioral component. The behavioral component, as Salwen (1998) suggests,
' refers to "any behavioral outcome attributable to third-person perception" (p.262). As such,

he suggests that people tend to act on the basis of these perceptual discrepancies. A direct

12

behavioral consequence of perceptual discrepancy appears as support for censorship of
socially undesirable media content.6 Davison (1983) describes the phenomenon of
censorship as one of the most interesting behavioral consequences of third-person
perception of media content dealing with faith and morals.
"Insofar as faith and morals are concerned, at least, it is difﬁcult to ﬁnd a
censor who will admit to having been adversely affected by the information
whose dissemination is to be prohibited. Even the censor's friends are usually safe

from pollution. It is the general public that must be protected. Or else, it is
youthful members of the general public, or those with impressionable minds”

(p.14).

Gunther ﬁrst (1995) tested the relationship between third-person perception and
support for the censorship of pomography as a behavioral consequence of the third-
person effect. He raised a question, "is it the perception of greater negative inﬂuence on
the morals of others that motivates people to support censorship of pornographic media
content” (p.28). His research ﬁndings supported the behavioral hypothesis that the
magnitude of perceptual discrepancy is positively related to support for censorship of
pornography. Since then, research has left little doubt that a concomitant relationship
exists between perceptual discrepancy and support for the censorship. Rojas and
colleagues (1996) showed that as the perceptual discrepancy between media effects on
others and self increases, people were more likely to exhibit pro-censorship attitudes

toward three media topics, the media in general, violence on television, and pornography.

 

6 However, behavioral outcomes, according to Salwen and Driscoll (1997), may be affected by issue
dynamics. Their study that examines the relationship between third-person perception and support for press
restrictions on the O. J. Simpson trial coverage did not support the behavioral hypothesis. Rather, they
found that people's predispositions about the trial interacted with perceptual bias as a predictor of support
for press restrictions on trial coverage. They argued that issue legitimacy perceived as legitimate or
illegitimate might inﬂuence people's willingness to support press restrictions. For more discussion, see
Salwen and Driscoll (1997).

13

They also found a positive correlation between the magnitude of perceptual discrepancy
and an intention to act in support of censorship for pornography (r = .44, p < .001). In the
area of television ﬁlm, Paxton (1996) showed that as the discrepancy between media
effect on self and others increases, people are more likely to want to restrict the freedom
of expression. Salwen (1998) and Salwen and Dupagne (1999) conﬁrmed that third-person
perception is a positive predictor of support for censorship of media content using
hierarchical regressions.

In the context of violent and misogynic rap lyrics McLeod and colleagues (1997)
also tested the behavioral hypothesis and found a strong association between third-person
perception and support for censorship of harmﬁrl content of rap music. Hoffner and
colleagues (1999) examined the link between third-person effects and support for
censorship of television violence. They measured and tested three different effects: mean-
world perception, aggressive attitude, and aggressive behavior. The ﬁndings showed that
while there was a strong association between the third-person effect for aggression and
support for censorship, the third-person effect for a mean-world perception did not predict
support for censorship.

The behavioral hypothesis of the third-person effect has been supported also in
Asian cultural contexts (Gunther & Hwa, 1996; Lee & Yang, 1996). Of particular
importance to present concerns, Gunther and Hwa (1996) found that when people regard
the media content as very sensitive, as with sex-related content, their support for
censorship appeared to be stronger. However, of interest here is the fact that although two
studies (Gunther & Hwa, 1996; Lee & Yang, 1996) have examined the third-person effect

in an Asian cultural context and emphasized cultural factors as an important criteria for

14

explaining cultural differences in the third-person effect’s impact on censorship of socially
undesirable media content, no true cross-cultural study on the third-person effect has been

conducted.

Hypotheses

The purpose of this study is to extend research on the third-person effect to the
context of Internet pornography and to consider the inﬂuence of two factors, cultural
individualism/collectivism and Internet self-efﬁcacy, on the third-person effect. For the

purpose of this study, the following hypotheses are suggested.

Third-person perception and Internet pomographv. The ﬁrst hypothesis attempts
to replicate previous ﬁndings on third-person perception in the context of Internet
pornography. This study predicts the following:

Hypothesis 1: Individuals will perceive Internet pornography to have a

greater negative inﬂuence on others than on themselves.

Individualism/collectivism and the third-person perception. As the studies of
Gunther and Hwa (1996) and Lee and Yang (1996) intimated, it is strongly assumed that
there are cultural differences in the third-person effect between Western culture and non-
Westem culture. Gunther and Hwa (1996) postulated that the magnitude of the perceptual
discrepancy in Asian cultures would be smaller than that in Western culture. In spite of
their assumption, however, cultural differences in the third-person effect across Western

culture and non-Westem cultures have not empirically tested.

15

In cross-cultural research, cultural variability is often considered along four
dimensions: power distance, uncertainty avoidance, individualism/collectivism, and
masculinity-femininity (Hofstede, 1991; Ting-Toomey, 1999). The present study focuses
on individualism/collectivism. Individualism/collectivism, the most studied among the
four dimensions of cultural variability, provides an important theoretical base for
explaining "the basic differences and similarities concerning communication behavior
between clusters of cultures" (Ting-Toorney, 1999, pp. 66—67).

Individualism is deﬁned as "the broad value tendencies of a culture in
emphasizing the importance of individual identity over group identity, individual rights
over group rights, and individual needs over group needs,” while collectivism refers to
"the broad value tendencies of a culture in emphasizing the importance of the ‘we’
identity over the ‘1’ identity, group rights over individual rights, and in-group-oriented
needs over individual wants and desires” (Ting-Toomey, 1999, p. 67).

How individuals deﬁne themselves, according to Ting-Toomey (1999), is a
fundamental base for making differences between individualism and collectivism across a
wide variety of cultures. For instance, while individuals in individualistic cultures view
themselves as independent ﬁom others and groups, individuals in collectivistic cultures
view themselves as interdependent with others and groups. The degree to which
individuals deﬁne themselves as independent or interdependent to others is termed as
independent self-construal and interdependent self-construal (Markus & Kitayama, 1991).
In terms of self-conception, according to Markus and Kitayama (1991), while individuals
who have high independent self-construal tend to view themselves as distinct and unique

from others, individuals who have high interdependent self-construal tend to ﬁt in with

16

others. By comparing two cultural groups (Japanese and the US. college students), they
found differences between two nations in their estimations of their own uniqueness. Thus,
while the US. students estimated that only 30% of people on average would be better
than themselves, Japanese students did not show this high self-perceived uniqueness
(Markus & Kitayama, 1991).

In general, while independent self-construal prevails in individualistic cultures,
interdependent self-construal prevails in collectivistic cultures (Gudykunst et al., 1996;
Markus & Kitayama, 1991; Singelis & Brown, 1995). Thus, according to- the ﬁndings of
Singelis and Brown (1995), the more people tend to show collectivism, the more they
show interdependent self-construal rather than independent self-construal. In this manner,
it might be assumed that differences between the self-conception of individualistic
cultures and collectivistic cultures create aggregate differences in cross-cultural optimistic
bias. In other words, people in Asian cultures (collectivistic culture) are less likely to view
themselves as distinct or unique from others and as smarter or better than others because
they tend to show interdependent self-construal more than do people in Western cultures
(individualistic culture). Research by Park (1998) partly supports this assumption. By
comparing two cultural groups (Korean and the US. college students), she examined the
relationship between self-construal and false uniqueness tendency. Her ﬁnding showed
that the more individuals hold independent self-construal, the more they showed false
uniqueness tendency.

Since optimistic bias is thought to determine the magnitude of third-person
perception, presumably, the expected cultural differences in optimistic bias should lead to

difference between the two cultures in third-person perception. Therefore, this study

17

predicts that people who are high on collectivism are less likely to show third-person
perception in the context of Internet pornography than are those who are high on
individualism.

Hypothesis 2: Collectivism will be negatively associated with third-person
perception in the context of Internet pornography. Thus, the more individuals show
collectivism, the more the perceptual discrepancy between the negative effects of

Internet pornography on others and on themselves will decrease.

Internet usL and third-person perception. As noted above, media use has been
considered as an important recipient factor even though most studies have failed to
support this assumption. However, this study assumes that the unique attributes of
Internet use distinguish it from other media and magnify its impact on third-person
perception. Central here are technological differences between Internet use and
traditional media use. In order to use the Internet, special skills and knowledge are
required. Eastin and LaRose (2000) argue:

"The Internet requires development of a further set of skills that, to the novice

user, at least, may be daunting. These include establishing and maintaining a

stable Internet connection, learning how to navigate on the Internet, and searching

it for relevant information”.
Self-evaluation of one's capabilities to use the Internet differs. Some people may highly
evaluate their capabilities to use the Internet, while others do not. This individuals' belief
"in one's capabilities to organize and execute courses of Internet actions required to
produce given attainments" is termed Internet self-eﬂicacy (Eastin & LaRose, 2000).

They maintained:

"Within social cognitive theory (Bandura, 1982; 1997) self-efﬁcacy is a form of
self-evaluation that inﬂuences decisions about what behaviors to undertake, the

18

amount of effort and persistence put forth when faced with obstacles, and ﬁnally,

the mastery of the behavior. Self-efﬁcacy is not a measure of skill; rather, it

reﬂects what individuals believe they can do with the skills they possess” (Eastin

and LaRose, 2000).

It is important to note that Internet self-efﬁcacy does not refer to the individuals'
skills for using the Internet but rather to the beliefs of what he or she can do with the
skills. This belief, namely, individuals' self-evaluation of capability to use the Internet,
can also create optimistic bias. As noted above, optimistic bias is a tendency in which
people believe that negative experiences occur to themselves less often than others.
Optimistic bias can be inﬂuenced by perceived controllability (Brosius & Engel, 1996).
Weinstein (1980) found that if people believed that they could control a negative event by
their own effort, their estimate that likelihood that the negative event could happen to
others is bigger than the likelihood it could happen to self. In a similar manner, this study
assumes that Internet self-efﬁcacy will inﬂuence third-person perception. Individuals
whose self-evaluation of their own Internet use capability is high may view themselves
more capable to control its inﬂuence. As such, they are likely to consider themselves less
vulnerable to Internet pornography than others.

Based on this logic, this study predicts that people who have high Intemet self-
efﬁcacy are more likely to show the third-person perception of Internet pornography than

people who have low Internet self-efﬁcacy.

Hypothesis 3: There will be a positive relationship between Internet self-
efficacy and third-person perception in the context of Internet pornography. Thus,
the more individuals show Internet self-efﬁcacy, the more the perceptual
discrepancy between the negative effects of Internet pornography on others and on

themselves will increase.

19

The behavioral component of the third-person effect. The ﬁnal hypothesis is to
replicate the behavioral consequence of third-person perception in the context of Internet
pornography. This study predicts the following:

Hypothesis 4: The magnitude of perceptual discrepancy will be positively

associated with support for censorship of Internet pornography.

A Model of Culture and Internet Self-Efﬁcacy’s Impact on the Third-Person Effect
Based on literature and the logic supporting hypotheses discussed above, this study
constructed an initial path model to investigate the causal links relating individualism/
collectivism, Internet self-efﬁcacy, third-person perception and its behavioral consequence
(Figure 1). The ﬂow of the model demonstrates sets of theoretical propositions regarding
the interrelatedness of the factors. For instance, both individualism/collectivism and
hitemet self-efﬁcacy are shown as antecedents to third—person perception. Collectivism is a
negative predictor of perceptual discrepancy, while self-efficacy is a positive predictor.
Subsequent to this, third-person perception is seen as a positive predictor of its behavioral

consequence, support for censorship of Internet pornography.

20

Figure 1: Initial Path Model

 

 

Collectivism

 

 

 

 

Support for Censorship

Third- ers __,
p on Internet pornography

Perception

 

 

 

 

 

 

 

 

mmmm
Self-Efﬁcacy

 

 

21

Chapter 2

METHOD

Sample

During May through June 2001 , a self-administered survey method was conducted
using a convenience sample of college undergraduate students in communication courses
from two nations, the United States and South Korea. These groups were selected for
previously identiﬁed differences in cultural individualism/collectivism. The US. is the
most representative individualistic country, whereas South Korea represents the
collectivistic cultural area. According to Hofstede (1991) individualism index (IDV) values
for 50 countries and 3 regions, the US. showed the highest IDV score (91), and South
Korea ranked at 43rd (IDV score: 18) among 50 countries.

The ﬁnal sample size was 232 with 118 undergraduate students at Michigan State
University in the US. and 114 undergraduate students at Hanyang University and Hanseo
University in South Korea. Of participants, 47 percent were male, 53 percent were female,
and the mean age was 21.44 (SD = 2.19, maximum = 34, minimum = 18). The mean age
observed in the two groups did not signiﬁcantly differ, t(230) = -.67, p = .95 (M age =

21.45 and 21.43 for the US. and South Korean students, respectively).

Procedures
Translation procedures in this study followed Hui's translation method. Two
versions of each questionnaire (English and Korean) were created simultaneously. This

method enables two versions of the questionnaire to have "equal status as being original"

22

(Hui, 1988, p.21). Three bilinguals checked the two versions of the questionnaire to assure
the equivalence of the wording in different languages (Hui, 1988). The respondents
answered questions about their cultural orientation, their Internet use and Internet self-
efﬁcacy, their opinion about the effects of Internet pornography and its censorship, and

Socio—demographic information.

Measures

Individualism/Collectivism. Cultural individualism/collectivism is the construct
measured by the individual's level of collectivism. In order to measure this construct, a
modiﬁed version of Hui's (1988) individualism/collectivism (INDCOL) scale was
employed. Hui reports a consistent pattern of construct validation. The INDCOL consists
of six subscales: spouse, parent, kin, neighbor, ﬁiends, and co-workers (or classmates).
The scale was modiﬁed through three steps. First, all items with an item-total correlation
in subscales below .i20 were omitted. Also, three items (1 and 16 in the parent subscale
and 11 in the co-worker subscale) were eliminated because they were redundant or do not
translate well across the two cultures. For example, the proverb cited in item 11 of the co-
worker subscale, "Too many cooks spoil the broth", has a different meaning for the
collectivistic culture in South Korea. Second, a total 45 of items ﬁom Hui's 63 items were
assessed for internal consistency and factor loadings. In order to perform these tests,
Conﬁrmatory Factor Analysis (CFA) was conducted using program PACKAGE (Hunter
& Gerbing, 1982). Based on these tests, a total of 12 items that did not violate the internal

consistency theorem were selected (see Table 1).

23

This 12-item modiﬁed scale showed unidimensionality with a low, but acceptable,
internal reliability (or = .50). As such, the 12 items were summed into a single score for
the analysis. Each item was measured on a 5-point Likert-type scale afﬁxed by "strongly
agree" and "strongly disagree." Higher scores on the summed score represented greater
collectivism (M = 38.97, SD = 4.56, max = 51.00, rrrin =26.00). The mean for the US.
participants was 37.58 (SD = 4.27), and the mean for Korean participants was 40.33 (SD

= 4.42). The two groups signiﬁcantly differed, t (223) = 4.75, p < .001.

TABLE 1: Modiﬁed IndividualismIColIectivism (INDCOL) Items

 

1. If a husband is a sports fan, a wife should also cultivate an interest in sports.

If the husband is a stock broker, the wife should also be aware of the current

market situation.*

When making important decisions, I seldom consider the positive and

negative effects my decisions have on my parents.

Teenagers should listen to their parents' advice on dating.*

Whether one spends an income extravagantly or stingily is of no concern to

one's relatives (cousins, uncles).

When deciding what kind of education to have, I would pay absolutely no

attention to my uncles' or cousins' advice.

Each family has its own problems unique to itself. It does not help to tell

relatives about one's problems.

7. I am often inﬂuenced by the moods of my neighbors.*

8. I enjoy meeting and talking to my neighbors everydayf

9. My good friends and I agree on the best places to shop.*

10. I would pay absolutely no attention to my close friends' views when deciding
what kind of work to do.

11.There is everything to gain and nothing to lose for classmates to group
themselves for study and discussion.*

12.I would help if a classmate at school told me that he/she needed money to
pay something.*

N

5".“

s» .01

 

* represents reverse scored items.

24

Internet Use and Internet Self-Efﬁcag3y. A single item was used to measure
Internet use. Respondents were asked an open-ended question on how many hours a week
they use the Internet regardless of personal or professional use (open-ended questions).
The mean on Internet use was 10.52 hours a week (SD = 11.47, max = 70.00, min =1.00).
The construct of individuals' lntemet self-efﬁcacy was measured with the 8 item Internet
Self-efﬁcacy scale developed by Eastin and LaRose (2000). After an internal consistency
test using CFA seven of the eight items (see Table 2) showed good reliability (or = .87)
and were summed into single Internet self-efﬁcacy score for the analysis. Each item was
measured on a 5-point Likert-type scale from 1 (strongly agree) to 5 (strongly disagree).
All items were reverse scored. Higher scores on the summed score indicated greater

Internet Self-efﬁcacy (M = 21.48, SD = 5.34, max = 35.00, min = 7.00).

TABLE 2: Modiﬁed Internet Self-Efﬁcacy Items

 

I feel conﬁdent understanding terms/words relating to lntemet hardware.

I feel conﬁdent describing functions of I ntemet hardware.

I feel conﬁdent trouble shooting Internet problems.

I feel conﬁdent explaining why a task will not run on the lntemet.

I feel conﬁdent using the lntemet to gather data.

I feel conﬁdent Ieaming advanced skills within a speciﬁc lntemet program.
I feel conﬁdent turning to on-line discussion group when help is needed.

NP’P‘PP’NT‘

Third-Person Perception. Third-person perception was measured with four items
adapted from Gunther (1995). The differences between perceived ﬁrst-person and third-
person effects of lntemet pornography both on moral values concerning sex and on

attitudes toward the opposite sex were calculated to measure third-person perception. The

25

ﬁrst two items measured the ﬁrst-person effect, the estimated effects of lntemet
pornography on self. The second two items assessed the perceived effects of lntemet
pornography on others. Respondents were asked twice to indicate how they strongly agreed
or disagreed with the same two statements asked in regard to themselves or others: 1) The
pornographic content on the lntemet has a negative effect on your (other people's) moral
values concerning sex and 2) The pornographic content on the Internet has a negative
effect on your (other people's) attitudes toward the opposite sex. These questions were
measured on a 7-point Likert-type scale from 1 (strongly agree) to 7 (strongly disagree).
In this study the question order of self and others was not considered. Previous
research has found that self-other question order does not inﬂuence the perceptual
hypothesis of the third-person effect (Dupagne et al., 1999; Gunther, 1995; Salwen, 1998;
Salwen & Driscoll, 1997) even though placement of self-other questions prior to the
restrictions question might affect the behavioral hypothesis in some conditions (Dupagne

et al., 1999).

Support for Censors_hip_. In order to measure support for censorship on lntemet
pornography, a modiﬁed version of Rojas et al.'s (1997) four pornography items (a = .81)
extracted from 16 censorship items was employed. Respondents were asked to respond
how they strongly agreed or disagreed with four statements (see Table 3). The result of
CF A on the four-items showed unidimensionality (or = .72), and they were summed into
single support-for—censorship score for analysis. These questions were also measured on

7-point Likert-type scales from 1 (strongly agree) to 7 (strongly disagree). Higher score

26

on the summed scores indicated greater support for censorship (M = 16.97, SD = 5.27,

max = 4.00, min = 28.00).

TABLE 3: Modiﬁed Support for Censorship on lntemet Pornography Items

 

Pornography sites on the lntemet should be censored.*

Today's standards for regulating pornography site on the lntemet are too strict.
For lntemet pomographers freedom of expression should be restricted.*

If pornography site on the lntemet has an appropriate device to block the
access of children, it should not be restricted.

swwe

 

* represents reverse scored items.

27

Chapter 3

RESULTS

Third-Person Perception and Internet Pornography

Hypothesis I attempted to replicate previous ﬁndings on third-person perception

in the context of lntemet pornography. The results supported the hypothesis 1.

Individuals are likely to perceive lntemet pornography to have a greater negative effect

on others than on themselves. The data show that 45.7 percent of participants estimated a

greater negative effect of Internet pornography on others than on themselves, while 38.4

percent responded no difference in inﬂuence, and 15.9 percent evaluated more effect on

themselves than on others. Paired t-test ﬁndings show that there are signiﬁcant

differences between the estimated effects of lntemet pornography on self and other both

for moral values and for attitudes toward the opposite sex (Table 4). When responses on

moral value and attitudes were combined, the perceptual discrepancy for scores combined

was also signiﬁcant.

TABLE 4: Self and Others Effects of lntemet Pornography Comparison

 

 

 

Self Others Perceptual
N M SD M SD Discrepancy T-Value
Effect on Moral Value 232 4.02 1.81 4.54 1.54 0.53 512*“
Effect on Attitudes 232 3.66 1.81 4.47 1.57 0.81 8.85***
Combined Effect 232 3.84 1.67 4.51 1.49 0.67 7.76***

 

**" p < .001.

28

Individualism/Collectivism and the Third-Person Perception

Hypothesis 2 proposed that cultural individualism/collectivism inﬂuences third-
person perception. Thus, people who are high on collectivism are less likely to show
third-person perception in the context of lntemet pornography than those are low on
collectivism. In order to test the association between individualism/collectivism and third-
person perception, this study assessed the correlation between the summed INDCOL
scores and combined perceptual discrepancy scores (i.e. effect on self versus others when
combined for both moral values and attitudes: M = .67, SD = 1.31, max = 6.00, min = -
2.50). As hypothesized, Table 5 indicates that the summed INDCOL score was associated
signiﬁcantly with the magnitude of perceptual discrepancy (r = -.15, p < .05, one-tailed, n
= 225). This ﬁnding supported the prediction of hypothesis 2 that individual’s collectivism
is negatively associated with third-person perception. Thus, as an individual’s
collectivism increases, the perceived difference between the negative effects of lntemet

pornography on others and on themselves decreases.

lntemet Self-Efﬁcacy and Third-Person Perception

Hypothesis 3 predicted that people who have high lntemet self-efﬁcacy are more
likely to show third-person perception of lntemet pornography than people who have low
lntemet self—efﬁcacy. To test this prediction, the summed lntemet self—efﬁcacy scores
were correlated with the combined perceptual discrepancy scores. The results support
hypothesis 3. As Table 5 shows, there was a positive association between Internet self-
efﬁcacy and the magnitude of perceptual discrepancy (r = .13, p < .05, one-tailed, n =

227). Thus, as an individuals’ lntemet self-efﬁcacy increases, the perceived difference

29

.5. Va... .8. v u.

.toqaam .285 3629: 908 8:91 a
Soto .985 pozmoﬁg 98m 5:91 a
.95: n F .oEom n c we cocaooom a
.29: u _. SEES u c an penaooom m

 

 

MO. ecNV. aim”. NOV 00. tsQNf $6... «5.3... «aw—ﬁr UQ_£W._OWC@0 50% tOQQDW O.—.
ram. ram..- .2. .9: 8. 8.- no... 8.- 388985 .5829. .m
too. No. mo. «LN: vor tom... «of- 029:0 co Sootw .w
met :3. tmmr No. .3... KC. 1.0m co mﬂootm .x.
nor tum. :3. mo. mo. Ocean—meg ﬁEoE. .o
nor 3. L r. no... .5002. .m
«or tow. :FN. EEQEQOEOA “9:95 9 22890 .w
8. L F. om: “059:. .m
tom. .xmm .N
mm< .F
m m n m m w m N F

 

point: «copcaon new moi—3...; Eon—Boon:— Lo. x522 scam—230 coma—wed ”m m._m<._.

30

between the negative effects of lntemet pornography on others and on themselves
increases.

Though not a formal hypothesis here, the results of this study are also consistent
with previous ﬁndings (Eastin & LaRose, 2000) in which Internet self-efﬁcacy is
positively correlated with lntemet use (r = .16, p < .01, one-tailed, n = 226). In addition, a
positive relationship between lntemet self-efﬁcacy and exposure to lntemet pornography

was found (r = .24, p < .01, one-tailed, n = 226).

Tests for modifying variables. For exploratory purposes, this study also conducted
a hierarchical multiple regression to see if the observed associations hold when other
independent variables are controlled, and to see if there are additional variables that could
explain third-person perception. Six variables grouped into four blocks were entered into
a multiple regression model (see Table 6) used to predict combined perceptual
discrepancy scores. The overall model accounted for 5 percent of the variance (p = .06).
The ﬁrst block of the regression equation consisted of demographic variables (age and
sex). The results show no signiﬁcant effect for block 1. Although the two variables in this
block were not signiﬁcant, their patterns are consistent with previous ﬁndings. Past work
has shown that age is negatively correlated with third-person perception (e. g. Boynton &
Wu, 1999; Peiser, W. & Peter, J ., 2001; Salwen, 1998), and females tend to show more
third-person perception than males (e. g. Boynton & Wu, 1999; Lee & Yang, 1996; Salwen,
1998). The second block consisted of media use variables (lntemet use and exposure to

lntemet pornography). Neither was signiﬁcantly related to third-person perception. The

31

third block consisted of lntemet self-efﬁcacy. This block, although not signiﬁcant,
explained additional 2 percent of variance in the model. The standardized B of .13 for the
ﬁnal equation is relatively high within the model, and leaves open the possibility that
Internet self-efﬁcacy can be a notable predictor for third-person perception. The fourth
block was INDCOL. As Table 6 demonstrates, INDCOL had a signiﬁcant negative
association with third-person perception (B = -.15, p < .05) and explained 2 percent of
variance in the model. This suggests that collectivism remained a signiﬁcant negative
predictor for third-person perception even when controlling for other independent

variables.

TABLE 6: Hierarchical MuItIple Regression Predicting Third-Person
Perception "‘

 

Final [3 R2 Change

 

Block 1 .00
Age -.02
Sex b -.08

Block 2 .01
Internet Use -.07

Exposure to lntemet Pornography d .05

Block 3 .02
lntemet Self-Efﬁcacy ° .13

Block 4 .02*
INDCOL" -.15*

Total R2 .05

 

a High score indicates greater third-person perceptual bias.
b Recorded as 0 = female, 1 = male.

° High score indicates greater lntemet self-efﬁcacy.

d Recorded as 0 = some, 1 = none.

° High score indicates greater collectivism.

* p < .05.

32

Third-Person Perception and Behavior

Hypothesis 4 attempted to test the behavioral consequence of third-person
perception in the context of lntemet pornography. This study predicted that the
magnitude of perceptual discrepancy would be positively associated with support for
censorship of Internet pornography. In most studies on the behavioral consequence of
third-person perception, the independent inﬂuence of effect-on-self on the association
between perceptual discrepancy and behavioral outcome has been controlled as a possible
suppressor variable (see Gunther, 1995 or Salwen, 1998). Agresti and Finlay (1997)
suggest that when no bivariate association is evident, if a third variable is related to both
explanatory and response variables, it should be statistically controlled. Inspection of
these data shows that effect-on-self is related both perceptual discrepancy (r = -.52, p
< .01, one-tailed, n = 232) and support for censorship (r = .35, p < .01, one-tailed, n =
231). Thus, the suppressor effect of perceptual discrepancy on support for censorship was
controlled in this study.

First-order partial correlation between perceptual discrepancy and support for
censorship controlling for effect on self shows that they are signiﬁcantly and positively
associated with each other (r = .27, p < .001, one-tailed, n = 228). This result supports
hypothesis 4 that the more individuals perceive lntemet pornography to have a greater
negative effect on others than on themselves, the more they are likely to support for
censorship of lntemet pornography. This study afﬁrmed the previous ﬁndings about the
behavioral consequence of third-person perception (Gunther, 1995; Gunther & Hwa,
1996; Hofﬁrer et. al., 1999; Lee & Yang, 1996; McLeod et. al., 1997; Rojas et. Al., 1996;

Salwen, 1998; Salwen & Dupagne, 1999).

33

Tests for modifyinuariables. In order to assess this ﬁrst-order association and
other factors in predicting support for censorship, a hierarchical multiple regression
analysis was conducted. Eight variables grouped into three blocks were entered into the
multiple regression model. The ﬁrst block of the regression equation consisted of
demographic and psychographic variables (age, sex, and INDCOL). Table 7 indicates
that the ﬁrst block explained 19 percent of the variance in the model (p < .001). Sex was
a substantial and signiﬁcant negative predictor of support for censorship (B = -.33, p
< .001). The direction of age, although not signiﬁcant, was negative ([3 = -.01, n.s.); again
consistent with previous ﬁndings (e. g. Boynton & Wu, 1999; Gunther, 1995; Lee & Yang,
1996; McLeod, Eveland, & Nathanson, 1997; Rojas, Shah, & Faber, 1997; Salwen, 1998;
Salwen & Driscoll, 1997).

The second block consisted of media use and efﬁcacy variables (Internet use,
lntemet self—efﬁcacy, and exposure to lntemet pornography). This block was not
signiﬁcantly related to the support for censorship.

The third block consisted of perceptual discrepancy and perceived effect on self.
This block explained 13 percent of variance in the model (p < .001). As Table 7 indicates,
both perceptual bias and perceived effect on self were signiﬁcantly associated with
support for censorship (B = .25, p < .001 and B = .44, p < .001, respectively). These
results strongly support hypothesis 4 that the magnitude of perceptual discrepancy would

be signiﬁcantly associate with support for censorship.

34

TABLE 7: Hierarchical Multiple Regression Predicting Support for
Censorship of lntemet Pornography ’

 

Final [5 R2 Change

 

Block 1 .19***
Age -.01
Sex b -.33***
INDCOL ° .00

Block 2 .02
Internet Use -.1 1
lntemet Self-Efﬁcacy d .07
Exposure to lntemet Pornography ‘3 -.05

Block 3 .13***
Perceptual Discrepancy ' .25***
Self Effects ‘ .44***

Total R2 .33“:9

 

a High score indicates greater support for censorship.

b Recorded as 0 = female, 1 = male.

° High score indicates greater collectivism.

6 High score indicates greater lntemet self-efﬁcacy.

° Recorded as 0 = some, 1 = none.

' High score indicates greater effect.

9 The total R2 does not equal the sum of the R2 increment because of rounding error
*” p < .001.

Culture and Internet Self-Efficacy’s Impact on the Third-Person Effect

Following these initial analyses, the proposed path model was tested to investigate
the causal links among third-person perception, support for censorship, and two recipient
factors, cultural individualism/collectivism and lntemet self-efﬁcacy.

To test the model, path analysis was conducted using PACKAGE (Hunter and
Gerbing, 1982). The results of analysis for the missing path links between collectivism and
support for censorship (error = .02, z = .22, p > .05) and between lntemet self-efﬁcacy and

support for censorship (error = .02, z = .26, p > .05) indicated that the data were consistent

35

with the model suggested (see Figure 2). The overall ﬁt test also supported that the data
were consistent with the model as a whole (x2 (2) = .11, p > .05). These results suggest that
individuals’ collectivism and lntemet self-efﬁcacy inﬂuence third-person perception,

which then leads to support for censorship of lntemet pornography.

Figure 2: Path Coefﬁcients among IndividualismlCollectivism, Internet
Self- Efﬁcacy, Third-person Perception, and Support for

 

 

 

 

 

 

 

 

 

 

 

 

 

Censorship
Collectivism
-.15*
.27!"A
. * Support for Censorship
Third-person __,
Perception lntemet pornography
.13*
lntemet
Self-Efﬁcacy

 

 

 

*p<.05,**p< .01 Overall ﬁt,12(2)= .11,p> .05

Missing path link between collectivism and support for censorship: Error = .02, z = .22, p = .829
Missing path link between lntemet self-efﬁcacy and support for censorship: Error = .02, z = .26,
p = .794

36

Chapter 4

Discussion

This study examined the inﬂuence of cultural individualism/collectivism and
lntemet self-efﬁcacy on third-person perception in the context of lntemet pornography, and
the relationship between this third-person perception and support for censorship as its
behavioral consequence. Consistent with past ﬁndings on the third-person effect,
participants in this study perceive lntemet pornography to have a greater negative effect
on others than on themselves. The ﬁndings also supported the behavioral component
hypothesis showing that third-person perception was signiﬁcantly associated with support
for censorship. As hypothesized, the inﬂuence of individualism/collectivism and lntemet
self-efﬁcacy on the third-person perception was also observed. Finally, the results of path
analysis suggest that the data were consistent with a causal model in which both
individuals’ collectivism and lntemet self-efﬁcacy inﬂuence third-person perception,
which then leads to support for censorship of lntemet pornography.

Tests of hypothesis 1 supported the previous ﬁndings on the third-person
perceptual hypothesis showing that individuals are likely to perceive lntemet
pornography to have a greater negative inﬂuence on others than on themselves.
Irnportantly, the results replicate previous ﬁndings from traditional media and extend
them to include lntemet exposure.

Tests of hypothesis 2 examined cultural individualism/collectivism as a causal
factor of the third-person perception. The negative association found between these two

variables demonstrates that as collectivism increases, the perceptual discrepancy between

37

the negative effects of lntemet pornography on others and on self will decrease. For the
ﬁrst time, the present study empirically supported the assumptions of others (Gunther &
Hwa, 1996; Lee & Yang, 1996), claiming that people who are high on collectivism are
less likely to show third-person perception than people who are low on individualism.
Additionally, these data are consistent with arguments that the difference between two
cultures associated with self-concept impacts optimistic bias and subsequent perceptual
discrepancy.

Tests of hypothesis 3 examined the effect of lntemet self-efﬁcacy on the third-
person perception. Even though most prior research ﬁndings failed to support the
assumption that media use is related to the third-person perception, this study found a
signiﬁcant positive association between lntemet self-efﬁcacy and third-person perception.
This ﬁnding might stem from the fact that unlike traditional media, special skills and
knowledge required to use the lntemet magnify its impact on third-person perception.
This set of circumstances may enhance the ability of Internet self—efﬁcacy to cause an
optimistic bias and lead to this third-person perception.

The ﬁnal hypothesis was to test the behavioral component of the third-person
effect. A signiﬁcant association between third-person perception and support for
censorship of lntemet pornography was found, even after controlling for other
independent variables. This result strongly supports previous ﬁndings showing that as the
perceived dependency of media’s negative impact on self and others increases, so too
does support for censorship of the media content.

This study has several implications. First, it extends previous ﬁndings on the

third- person effect to the context of lntemet pornography. There is now little doubt that

38

the lntemet ﬁrnctions as a crucial medium in our society. As such, there are likely to be
several important areas in which the Intemet’s role in the third-person effect phenomena
should be studied. This study’s demonstration that media use impacts those third-person
effects associated with lntemet content suggests that the factors associated with this
phenomenon may be qualitatively different for Internet exposure. In light of the fact that
third-person effects can inﬂuence public opinion and resulting policy-making processes,
studying the third person effects of the lntemet seems critical to understanding those
opinions likely to shape future regulation of the most inﬂuential medium of our time.

This study examined also the inﬂuence of people’s collectivism on the third-
person effect. Although cultural individualism/collectivism has been considered as an
important factor in prior research, this is the ﬁrst study to demonstrate empirically that
individualism/collectivism and third-person perception are related. Though third-person
effect studies have focused primarily on Western culture, this study suggests that cultural
differences are an important consideration in explaining this phenomenon.

Finally, there are several limitations in this study. First, as suggested in most third-
person effect research, this study used measures of attitude, not actual response, as the
behavioral component of the third-person effect. Clearly, this is a concern. Nevertheless, as
stated earlier, since public policy and judicial decisions are based on public opinion, the
study of these public attitudes or opinions toward the regulation of socially undesirable
media content tends to emerge eventually as a social consequence (Gunther, 1995; McLeod,
Eveland, & Nathanson, 1997). Second, the generalization of results is limited since this
study used a convenience sample of college students from two countries. In addition to

other potential bias, this purposive sampling suppressed the possible role of some

39

 

 

 

 

lama
liter:
using
cons
no
\‘31’
IIEI

p01

demographic variables (e. g. age, income, education, etc.) in the third—person effect.
Therefore, ﬁrture study should attempt to replicate the ﬁndings suggested in this study
using a more representative sample of the general population. Finally, this study did not
consider other variables that might inﬂuence and explain the third-person effect. The
exogenous variables included in this study did not explain a substantial amount of the total
variance of third-person effect. For example, other attitudinal variables, such as support for
free speech, political afﬁliation, or religiosity, could be examined in relationship to lntemet

pornography and the third-person effect.

40

APPENDIX

Questionnaire (English Version)

I. In this ﬁrst set of questions, you will read statement about different aspects
of daily life and social relationships. Read each statement carefully and
please circle your level of agreement / disagreement how much you agree
with that statement.

1. If a husband is a sports fan, a wife should also cultivate an interest in sports. If the
husband is a stock broker, the wife should also be aware of the current market

situation.
Strongly Agree Agree Neutral Disagree Strongly Disagree
1 2 3 4 5

2. If one is interested in a job about which the spouse is not very enthusiastic, one
should apply for it anyway.

Strongly Agree Agree Neutral Disagree Strongly Disagree
1 2 3 4 5

3. It is better for a husband and wife to have their own bank accounts rather than to have
a joint account.

Strongly Agree Agree Neutral Disagree Strongly Disagree
1 2 3 4 5

4. The decision of where one is to work should be jointly made with one's spouse, if one
is married.
Strongly Agree Agree Neutral Disagree Strongly Disagree
1 2 3 4 5
5. In these days parents are too stringent with their kids, stunting the development of
initiative.
Strongly Agree Agree Neutral Disagree Strongly Disagree
1 2 3 4 5

6. When making important decisions, I seldom consider the positive and negative effects
my decisions have on my parents.

Strongly Agree Agree Neutral Disagree Strongly Disagree
1 2 3 4 5

41

7. Teenagers should listen to their parents' advice on dating.

Strongly Agree Agree Neutral Disagree Strongly Disagree
1 2 3 4 5

8. Even if the child won the Nobel prize, the parents should not feel honored in any way.

Strongly Agree Agree Neutral Disagree Strongly Disagree
1 2 3 4 5

9. It is reasonable for a son to continue his father's business.

Strongly Agree Agree Neutral Disagree Strongly Disagree
1 2 3 4 5

10. I would not share my ideas and newly acquired knowledge with my parents.

Strongly Agree Agree Neutral Disagree Strongly Disagree
1 2 3 4 5

11. I practice the religion of my parents.

Strongly Agree Agree Neutral Disagree Strongly Disagree
1 2 3 4 5

12. I would not let my parents use my car (if I have one), whether they are good drivers
or not.

Strongly Agree Agree Neutral Disagree Strongly Disagree
1 2 3 4 5

13. Children should not feel honored even if the father were highly praised and given an
award by a government ofﬁcial for his contribution and service to the community.

Strongly Agree Agree Neutral Disagree Strongly Disagree
1 2 3 4 5

14. Young people should take into consideration their parents' advice when making
education or career plan.

Strongly Agree Agree Neutral Disagree Strongly Disagree
1 2 3 4 5

15. The bigger a family, the more family problems there are.

Strongly Agree Agree Neutral Disagree Strongly Disagree
1 2 3 4 5

42

16. I have never told my parents the number of sons I want to have.

Strongly Agree Agree Neutral Disagree Strongly Disagree
1 2 3 4 5

17. I would help, within my means, if a relative told me that he/she is in ﬁnancial
difﬁculty.

Strongly Agree Agree Neutral Disagree Strongly Disagree
1 2 3 4 5

18. If I met a person whose last name was the same as mine, I would start wondering
whether we were, at lest remotely, related by blood.

Strongly Agree Agree Neutral Disagree Strongly Disagree
1 2 3 4 5

19. Whether one spends an income extravagantly or stingily is of no concern to one's
relatives (cousins, uncles).

Strongly Agree Agree Neutral Disagree Strongly Disagree
1 2 3 4 5

20. I would not let my cousin use my car (If I have one).

Strongly Agree Agree Neutral Disagree Strongly Disagree
1 2 3 4 5

21. When deciding what kind of work to do, I would deﬁnitely pay attention to the views
of relatives of my generation.

Strongly Agree Agree Neutral Disagree Strongly Disagree
1 2 3 4 5

22. When deciding what kind of education to have, I would pay absolutely no attention to
my uncles' or cousins' advice.

Strongly Agree Agree Neutral Disagree Strongly Disagree
1 2 3 4 5

23. Each family has its own problems unique to itself. It does not help to tell relatives
about one's problems.

Strongly Agree Agree Neutral Disagree Strongly Disagree
1 2 3 4 5

43

24. I can count on my relatives for help if I ﬁnd myself in any kind of trouble.

Strongly Agree Agree Neutral Disagree Strongly Disagree
1 2 3 4 5

25. I have never chatted with my neighbors about the political future of this state

Strongly Agree Agree Neutral Disagree Strongly Disagree
1 2 3 4 5

26. I am often inﬂuenced by the moods of my neighbors.

Strongly Agree Agree Neutral Disagree Strongly Disagree
1 2 3 4 5

27. My neighbors always tell me interesting stories that have happened around them.

Strongly Agree Agree Neutral Disagree Strongly Disagree
1 2 3 4 5

28. I am not interested in knowing what my neighbors are really like.

Strongly Agree Agree Neutral Disagree Strongly Disagree
1 2 3 4 5

29. One need not worry about what the neighbors say about whom one should marry.

Strongly Agree Agree Neutral Disagree Strongly Disagree
1 2 3 4 5

30. I enjoy meeting and talking to my neighbors everyday.

Strongly Agree Agree Neutral Disagree Strongly Disagree
1 2 3 4 5

31. In the past, my neighbors have never borrowed anything from me or my family.

Strongly Agree Agree Neutral Disagree Strongly Disagree
1 2 3 4 5

32. One needs to be cautious in talking with neighbors, otherwise others might think you
are noisy.

Strongly Agree Agree Neutral Disagree Strongly Disagree
1 2 3 4 5

44

33. I don't really know how to beﬁiend my neighbors.

Strongly Agree Agree Neutral Disagree Strongly Disagree
1 2 3 4 5

34. I feel uneasy when my neighbors do not greet me when we come across each other.

Strongly Agree Agree Neutral Disagree Strongly Disagree
1 2 3 4 5

35. I would rather struggle through a personal problem by myself than discuss it with my
ﬁiends.

Strongly Agree Agree Neutral Disagree Strongly Disagree
1 2 3 4 5

36. I like to live close to my good ﬁiends.

Strongly Agree Agree Neutral Disagree Strongly Disagree
1 2 3 4 5

37. My good friends and I agree on the best places to shop.

Strongly Agree Agree Neutral Disagree Strongly Disagree
1 2 3 4 5

38. I would pay absolutely no attention to my close friends' views when deciding what
kind of work to do.

Strongly Agree Agree Neutral Disagree Strongly Disagree
1 2 3 4 5

39. It is a personal matter whether I worship money or not. Therefore, it is not necessary
for my ﬁiends to give any counsel.

Strongly Agree Agree Neutral Disagree Strongly Disagree
1 2 3 4 5

40. I have never loaned my car/lap top to any classmates.

Strongly Agree Agree Neutral Disagree Strongly Disagree
1 2 3 4 5

45

41. A group of people at their workplace was discussing where to eat. A popular choice
was a restaurant which had recently opened. However, someone in the group had
discovered that the food there was unpalatable. Yet the group disregarded this person's
objection, and insisted on trying it out. There were only two alternatives for the person
who objected: either to go or not to go with others. In this situation, not going with the
others is a better choice.

Strongly Agree Agree Neutral Disagree Strongly Disagree
1 2 3 4 5

42. There is everything to gain and nothing to lose for classmates to group themselves for
study and discussion.

Strongly Agree Agree Neutral Disagree Strongly Disagree
1 2 3 4 5

43. Classmates' assistance is indispensable to getting a good grade at school.

Strongly Agree Agree Neutral Disagree Strongly Disagree
1 2 3 4 5

44. I would help if a classmate at school told me that he/she needed money to pay
something.

Strongly Agree Agree Neutral Disagree Strongly Disagree
1 2 3 4 5

45. In most cases, to cooperate with someone whose ability is lower than one's own is not
as desirable as doing the thing alone.

Strongly Agree Agree Neutral Disagree Strongly Disagree
1 2 3 4 5
II. The next questions ask you some questions about your perception or behavior

regarding Internet use.

46. On average, how many hours a week do you use lntemet regardless of personal use or
professional use?

hours

 

47. I feel conﬁdent understanding terms/words relating to Internet hardware.

Strongly Agree Agree Neutral Disagree Strongly Disagree
1 2 3 4 5

46

48. I feel conﬁdent understanding terms/words relating to lntemet software.

Strongly Agree Agree Neutral Disagree Strongly Disagree
1 2 3 4 5

49. I feel conﬁdent describing functions of lntemet hardware.

Strongly Agree Agree Neutral Disagree Strongly Disagree
1 2 3 4 5

50. I feel conﬁdent trouble shooting lntemet problems.

Strongly Agree Agree Neutral Disagree Strongly Disagree
1 2 3 4 5

51. I feel conﬁdent explaining why a task will not run on the lntemet.

Strongly Agree Agree Neutral Disagree Strongly Disagree
1 2 3 4 5

52. I feel conﬁdent using the lntemet to gather data.

Strongly Agree Agree Neutral Disagree Strongly Disagree
1 2 3 4 5

53. I feel conﬁdent learning advanced skills within a speciﬁc Internet program.

Strongly Agree Agree Neutral Disagree Strongly Disagree
1 2 3 4 5

54. I feel conﬁdent turning to on-line discussion group when help is needed.

Strongly Agree Agree Neutral Disagree Strongly Disagree
1 2 3 4 5

III. The next two questions ask you some questions about your experience on

Internet pornography.

55. Have you seen or come across pornographic content on the lntemet?

Yes
No ----- > Skip Q57.

47

56. If yes, how many times have you seen or come across pornographic content on the
lntemet since January, 2001?

times

 

IV. The next set of questions asks about your general opinions or perception about
Internet pornography. Please answer the following questions even if you have never
seen or come across Internet pornography. Please indicate your level of agreement/
disagreement.

57. The pornographic content on the lntemet has a negative effect on my moral values
concerning sex.

Strongly Agree Neutral Strongly Disagree
1 2 3 4 5 6 7

58. The pornographic content on the lntemet has a negative effect on my attitudes toward
the opposite sex.

Strongly Agree Neutral Strongly Disagree
1 2 3 4 5 6 7

59. The pornographic content on the lntemet has a negative effect on other people's moral
values concerning sex.

Strongly Agree Neutral Strongly Disagree
1 2 3 4 5 6 7

60. The pornographic content on the lntemet has a negative effect on other people's
attitudes toward the opposite sex.

Strongly Agree Neutral Strongly Disagree
1 2 3 4 5 6 7

61. Pornography sites on the lntemet should be censored.

Strongly Agree Neutral Strongly Disagree
1 2 3 4 5 6 7

62. Today's standards for regulating pornography site on the lntemet are too strict.

Strongly Agree Neutral Strongly Disagree
1 2 3 4 5 6 7

48

63. The freedom of expression of people who post pornographic materials on the lntemet
should be restricted.

Strongly Agree Neutral Strongly Disagree
1 2 3 4 5 6 7

64. promography site on the lntemet has an appropriate device to block the access of
children, it should not be restricted.

Strongly Agree Neutral Strongly Disagree
1 2 3 4 5 6 7
V. Finally, We would like to know a little bit about you. Please answer the following
questions.
65. What is your gender? (Check one)
Male Female

66. What is your age?

years

 

67. With which of the following groups or group do you most identify?

Cl African-American Cl Asian-American or Paciﬁc Islander
Cl Latino Cl Native American

Cl White Cl Prefer not to answer

Cl Other

 

Thank you for your valuable participation in this survey.

49

Questionnaire (Korean Version)

QlEl‘ﬂ 01-8-34 913% £§i°ll 3&1} ”SE-25.4}

r. creel IIWEIE- eeei ale—se- Aiee—aa aw area 14317-1 anion :Ie 319.11%}. are
an age at 2 751201 man were OI}:— 78E are are emaiaezls Marci #2471 via
bit}.

1. ‘it? 01% EEO] ii}: 1801 EVE :1 ONIE SE 12$qu Ell-(slot 123% 7IX‘IOI (5113}. 3.11.11
{‘14. E3591 31' 0]%ﬂ Eiﬂai‘fl :1 0N1 93*] P1111194 *l’éi %-§i°ll PISS: 7l%°ﬁl°l= {Elli}.
DH?- %9|P=* %9—|’.§I §El t‘JEII El 0H5?- ‘EJEH Pet

1 2 3 4 5

2. Olifi *ltiol ilﬁlE 318.10] 915131., 43./“I :1 HIWZPI H81 ‘BziEIEIE :1 3121-8: Q7] 94311-841311

of EDI.

one egg: sale ea ”Jill er one ELHlH a
1 2 3 4 5

3 exam owl—:- %%9I SSIHEIE Mamie 7424 new eerae are 7.101%EI

one esale gala ea ear :2: one trials
1 2 3 4 5

4. viewers. xlaa 347801] area aee ans—w are new eat

one sale sew ea aura one Eml a
l 2 3 4 5

5 gee ease xILI-s—el ewe new 71 7lEEIE— LII-r 217—16I7l‘i'. 6ch

uls— ERIE sale ea ear 22 oil-r- FLIEH a
1 2 3 4 5

6. eat} aes— LIe rrI, LR:— :1 E7310] esteem uni-axle as— we ase- ezgz—I eaten a
sIoI a: maxi ere—cl.

UH

Eglai E
0 C3 0

....-IO

50

seeO'
4 Par—non ﬁ%7l%01<>t EDI

91 til OI £5.01] 3&3“ *1

:15—
E

20
at:

7.10:1

In!

.6D
mu 5

2.

1.!

8a
W4

UL

€21

an
o. 2

so

an
cl.

:3
2. 1

E8151.

$3513} 311313
lineanzamaea
0 33319-1? 5.1-2.8.

.Q.
a

at!A
IOIREI 0l017llrsé‘el

Ir

8a

Ir

ES

2.

Ir

.am
W 4

5..

LI

8a

0.2

So

If

.65

5o
2. l

or...

9. Kit-101 OIBIXI 91 ”+21

Ir

an
m: 5

2.

In!

an
W4

U.—

Ir

Sn
0. 2

go

as

0.
so
2. 1

LII7IAH§ t
«AH *7
Ila SWOILIZIéIS-LII—PE‘EUI
—1 %L%5}X] 0142.7}
L'O'E'. A015}.

10. LIE-

Ir

to
ma 5

2.

LI

8c
W4

5..

Ir

6a
0. 2

go

mm

o.
50
gr I

'73. "
JI—a ELITE 72301:}

Ma
an

a.
E

In!

to
ma 5

2.

In!

ﬁg
W4

U.—

LI

.GD
0. 2

5o

l—r

8a

50
2. 1

EEI ZI

saaaasuwweeEHAJ
D ea

1:.
O

12.03%! ' 04 T 4—
LH‘dIHIZVI 214..)"E ‘ l
lac $140210]
\_.L.]

ﬁx
oﬁa

1r
0

Is!

8s
m... 5

2.

Ir

8a
W4

H;

Egg

0..-

Ir

8a

:3
2. I

ﬁx

:1.

or

Edam :L 2i“
_,

oi

El Sit}

E.
E,

110

13. BIS 0
M2171 21914331011 tll '1
as. SE9} SHIP.

Oil Ell‘<"SIl 0333521 $1 61W

CHE}.

._‘-:_
L.

LI

6a
.Hu 5

2.

LI

6n
W4

H.—

ID!

as
o. 2

_._._o

as

oz

50
2. 1

Ir

Sn
.3 5

0T

Ll

6a
W4

H._

Ir

85
o. 2

5o

ma

5o
0T 1

If

6n
.3 5

0T

as
W4

Ht

Ewe

C)"

an
.9.

go
2. 1

51

are -a
TluaTeaamaaeamewI

reeuvwraes

16. LIE 0

I»!

so
ma 5

2.

Ir

6a
W4

U.—

I»!

an
0. 2

2o

.2.

0.

so
2. 1

17.91%} LII ilio
Alt-121131 0181822 Lila] F201
22.-Elsi: LIME

Eeaeq

Ir

2°C

022

so

as

20
2. 1

BIB] Emil
._ JIM 20M 7l77l—3— 2]? £1 a}

wenee

“11.1%:-

3
L

1015}.

.‘1‘.
L.

212

LI

an
.3 5

2.

If

an
W4

B.—

If

._0D

0.2

So

an

0.
20
2. 1

l9 i131} :LO 3 3: .1
m. a t .1 . u , a 4525).)“ 75123143 014136171] 5).: a 0 $512
L- L. I -._..] :LA
|___._-o]

OI] ‘A
aaauneaeanquq

Ir

.32
me 5

2.

lb!

.02
W4

H1

§EI
0117“ '23-?“
120L274
Lea A01EI'.

mg

022

2°

Ir

3m
0.
so
2. 1

32;
L.

*I

E
._.
E

2 DO
o<raqan2wIaqmyIi

Ir

3m
W4

Hi

we

0.2

50

mg

0.
2o
2. 1

Z
T91%7I%

seem

£2.
a

La% r... '1 ' '
_1 01 :1 a ”I‘é HILL}- XjEHZTO 14.9,]. 7&0 o
t g .1] — -
L. .2 , A CH ,1 21x31

°é 33011:}.

If

an
we 5

2.

Ir

3a
W4

H..—

[—1

6o

0.2

50

non

2o
2. 1

- 7 00
12 IX] Pat—.53: 31°15}

2a2

reeeeeeaena

._‘:.)~
L.

L112 HILL}

It!

to
ma 5

2.

If

.62
W4

H.—

ZS-El

Ll

.02
0. 2

Eo

.18

2°
2. 1

0

HA
.5

201171] '33}

X

23. 2} 7I7S—8— :L 7]-

seaaao

LI

.62
Ma 5

2.

mm
H.
E 4.
H.—

as
o. 2

50

we

5o
2. 1

52

222

wmi%%3%2

24342

244%m°1*
$222421
aéqﬁw+

[pl

.02
w. 5

2.

L-

20G
w 4

H..—

[r

.02
0. 2

5o

.2.
9.

20
2. 1

ﬂWﬂﬂ§ﬂﬂﬂﬁaﬁr

E.
a

2522°
Taqﬂﬂﬁﬂﬁumwh+
1&442

l—l

._0a
m. 5

2.

Ir

.2.
W4

5..

542

0"

.2.
0|.

2o
2. 1

LH0]00 B
T47HWWW4%%QmﬂE
an 2%

o.

0
ti)?”

2&4ﬂ2

Ir

.UD
m... 5

2.

IL.-

..on
W 4

B2

.2.
0. 2

:8

2mm
0|.

go
2. l

2
11:5}

2202 ‘
aaggﬁumwwwn

22

42:28
0.22ngn24

I—l

8a
W4

H._

In!

.02
W 2
2..

IL!

.02

so
2. 1

Ir

.02
W4

H...

2.
0. 2

50

E.
W

2.0
2.. l

223

ﬁEﬁﬂéﬁi

gl-
cl

N444

1'1.
\_.

2446

E
a

NZMPHWEEESWN-

E.
E1

2 4 ‘
9.22%q%

ﬁQQ.

Ir

.XVD
mu 5

2.

LI

.02
W4

U._

EB

Ir

.02
0. 2

£0

.2.
ma

2o
2. 1

Ir

Sm
m... 5

2.

Ir

6a
W4

U._

LI

..0u

0.2

50

2mm
9.

2o
2. l

3.2me
M%%%
qﬂWﬂééﬂmﬁﬁﬂiﬁﬂ
A ﬂﬂﬁq

IL:

.02
.3 5

2.

If

.02
W4

H...

as
0. 2

50

.2.

20
2. 1

N%ﬂmﬂ'
§”‘* 6}
)TW~EWE%4§%WW¥

0t
2...
Eu

1

3zqﬂﬂ
4&6}

‘_;‘.
L.

'5'

Sn
m. 5

2.

.2.
W4

5..

Ir

.02

0.2

5c

Ir

20D

2
$1

2...

#2

0.2
mm
12.0.0

0..

W

"

.XL

2...
or

E

If

20D
w. 5

0T

Ir

6a
W4

H...

Ir

82
o. 2

so

an
0|.
:6
2. l

53

20

3
22

1.7x
2...

2...
\Ir

‘l

O

0.2

Ir

.60
.00.. 5

2.

IV

.00
W4

5..

Ir

.00
0. 2

2.0

.2.

2o
2. l

ﬂﬂaqhtﬂiéﬂﬂﬂ

\—

Swat?“

5:
L.

ma

E.
E

a?

L.

1...

410.13% EH10“ BREE/‘1

}

U
1..

35. Uroll 7%] ‘33 *3

qunq.

22

L!

.00
9.2

20

.2.

2.0
2. l

- w :2
I: .129} 2131 79.2.2 7.74012}
31.23}.

lb!

.00
W4

U.

I»!

.00
9.2

2.0

In!

.00

20
2. 1

#22

IF

.6D
.3 5

2.

lbl

.00
W4

H:

[F

.BD
W 2

20

$0
0|.

2..
2. 1

1

38.2}é191

I?

.00
we 5

2.

Isl

.00
W4

5..

SE]

Ema

0"

I5!

.00

20
2. l

133% 7110124 D
0H101E}.E}3}*1LH21517}:L740114
A 13“ zit-31*?!

\—

44C

L.

9174‘:

«In:
J02

391414—32

322%.

Ir

.00
m. 5

2.

Ir

.00
W4

5.

If

.00

0.2

2.0

If

.62

0.0
2. 1

a2222ama2+

E

£7.22 ‘
JLAaawwihnn

lb!

.00
w. 5

2.

IL!

.00
W4

0..

.2.
o. 2

2°

Ll

.00

2.0
2. 1

0

SHAE‘-° c
In W.Q¢mqa°annm H
_ ._ 7%}?!

amq;1

iw+1424ﬂ%ﬁﬂﬂawll
.m:14%Q%4
4234 @011 7.9121, ohm ix}

E:
L.

E

\_.

.‘2‘.
L.

%°Il 7}
12:0 0 .9.
lAaﬂqqéanqgmq

4L2

M
.1

Ir

.00
m... 5

2.

Ir

.00
W4

0..

an

0.2

_...0

mm

2.0
2. l

54

2:93: 94311122 ‘3}21‘: 723.9: 2. :ILA3%J.% 35.201171123112711 31%.23134 133

DH?- %9l?=* %9l§ €24 ”JEN; 0H?- ELHZH 3:!
l 2 3 4 5
43 0.20... 2% 0.42221271222: 2:: 0. 2122.0. 22 2 23.-3.0.2..

Elsa! §El 'EJEH .3; 0H5?- UJCH E!
2 3 4 5

44. E202 {26.3.9.2 $221232} 9% 14 73517} 9.13.95, '42 7177101E2 330W}.

0..

29's. 29
o O

C!

E! éil

2 3

.— 40
I’EI

HZH E! 0H?- ”JEH 3:!
4

45.2.1229. 7.0.0., 2.2122. 22.0. 2.2 1.2, .3. 2.». 24.20.27,. 2 2.2... 22. 8.2 7,3 2c. 2-
0.2..

DH

29's.
0 D

0.11
N \0
DJ

._.-10

11. ‘4-3-9] 5.3- 25'.- Jé‘ih‘ﬁ- QlEi‘i °I%°ﬂ .734} 443.9] 9143”. 335112 22111} {MB}.

46. #16}:- 02224011338 2 *IGOIL} (215.2% 6}”..M7I}?(7H°J?4 01%014 751-331 01%011 2139.1
0])

All}

 

47.1.2 012... 0.29.20.22.21 030.22 0. 3.0.20. M10. 912..

8}

0 t”

L...

”H

0.”

9| is.

D

0.“

NP.

is. 0”

D

‘03’...“
A;

-10
u. FE
Q
02*

4sq:ma22222221222222m222242422.

0.2 20.2. 2.0.2. 22. 2.2. 2. 0.2 2.1.2.
I12 3 4 5

49. 2.: 01.2... 6.20.0.2 7.22-2- .42. 62:. 2.2.10. 0.2..

0.2 20.2. 2.0.2. 22 0.1a. 2. 0.2- .1:. 2.
1 2 3 4 5

55

so. LF—r: 2.61 Lyn—2.: olgsm sAgar—t: 21w. sag—g saga—13:41 mm 21:}.

D

0H? %9—|3* % at a: raw 8.; DH-‘?— m a:
l 4 5

3

N10

51. ONE {“5191 733210] 1521335111 925% “H, L)r‘f: QH 1710145383511] ‘QE—Xl’I—é it’éﬁH‘ﬁil XHJOI 9,1

14.
UH-?- %-9| 6.; gel at 321 gun 2: DH? inn 2:
l 2 3 4 5

52. 4% 9.112153% 01%8W1 43% $25 230112on 9M.

0H5?- %9|3* % IE“:F ’zf-E 'E'JCH 33 0H9?- PJEH ‘3
l 4

2
2 3 5
53. Uri-3 515} {131561 1323 01‘31 L11 35;;1‘339] 71%OIL} 7l%%- 9} 31-i— 75301144101915}.

0H

...-iO

§9l§=f %-9|§=.* éﬂ 'E'JCH E! 0H?- “.JEHE!
3 4 5

N

54. (215191-8- 0] %6}Ei %, E—S—OI 353%} “H 19:11}?! E"?- 1%(on-line discussion group)% 01%5}-‘E
33 01] 1H] 0] 9&5}.

UH-‘?i%91‘°3 %9l§ 23E! 'E’JEH 3:! UH?— ‘E‘JCH E!
4

III. 41°79] 25 Psi—Ii— ~ESJ%% (LE1 Ll‘ Egioﬂ 't’rii} #1 “M 73%% %—‘3 3304143}.

55. #1 8%- 9.161% QM] £211: 3:,

(‘10
U10
(2“.
M0
Mo
WI
.12.
.9.
.5;
mid
f0!

.12“.
.2.
59.
III)-
L
$1,
° \3

Oil
omg ----> 57% EJJEE 7mg.

 

56. 2124 Jag-WE, $1545 %2001€1%¥E1 9? 2% 2310114 3.753% y.*lﬂb} ﬁﬁW-ﬁ-HW}?

3'11

 

56

IV. 11173-43.- °1EI‘51’2}91 £91: 1%: 1%?315-41 ‘11?! 4181-4 §3J31°J91213194é11%= QE- {vi-ii}
54211451 “1E 413171°J51ﬂ ’89] iii-'9‘- 4% %ﬂ%% 75?} 78§°1 8111131131315- %‘&31} EH]
7] 31%}‘4‘4 1114511". 2J€E°1I £11121 %«1 iii-€- ‘i‘lEH-g] 785% ﬁﬁlﬁdl -’r‘-’~l 7] ”1'3‘451.

57. 9151 L11 ”8191 Sigh-.110: %%1%
08130195 “1’2 7530”}.

r10

L194 *JEEEL, 5% 1801] 336% 5W1/zral 1* 7121011 233:!

5.
2

[101'
01)!
[IE
FE

EH33

OH?-

0H?- %9|
1 7

b)
.5}.
Ln
ON

58°JE151’8191EME— §%%?}%~8—LH7177<131°1—‘501*é°11583113511qu1:11-2831‘1813619513123101
51.

DH

._..qo

%9|3.=1 §El ”HS?- ‘E‘JCH E1
2 4

b.)

59. 01H L1 ”91 £8 1:—_, 9: %%}%€: 94%ﬂii 51% ”Fa/1&9] 183%}, % 18011112318 $53.34/$1—
a123, 7}x101]—?—131—1 086018: ”1%; 33015}.
0H5? %9I§ €52! UH? “JEN?!

1 2 3 4 5 6 7
60 (JEN ”81-91 £51: 3% 3.1%:- (QEMEE the: A1%%°1 7x131 alv‘: 01801] 12:1 51150111?
78711. °3601% 131%! 330151.
UHFT’ %9I J §El 0H5? tJJLHEI

1 2 3 4 5 6 7

61. 9113191 891 £221: 401.12% 615.81 73%5101018} 6m.

UH

%9l E1 §a=l OHS?- E‘JEH E1
2 3 4 7

M
O‘

62 3: 31.13111 *0191 EL?- E: *MEE ﬁﬂﬂa— 71—i—g. Lil-,1-

119.

2% am.

E1H

$943+
0

D

a DH?
7

[E

1EH 21

5....‘10
A01)!

2

9.)
Ln
0‘

63.911511}! ”81°11iéi s—g— gal-:2 111%1-3‘91 Eﬁgl 1}%E- 2111121510101: 12.1 w.

0H?- %9I§ éﬂ 0H5?- BJEH E1
m 2 3 4 5 6 7

57

64.13%1 2112M 6121 £21: 40127} wax—494 set-a- 24 @6171] 911% 4: 551—:- 621% 851 9m

:5 :1 4015-1:— ﬁzﬂ ENE <21 8:}.

1"

DH

59's
0

...-10

§E 0H5?- E‘JF—H 3.21
2 3 4

v. 332$ 418194 41801! {1%} Q 712] QE—dl %"a1311-’1‘-*1‘d #6181386145}.
65. 4'1 6191 182%}?
‘38 04 ’8

663113191 (251.98%?

‘3} *1]

 

@1554! €451] 215114 11111331 34211—1131.

58

REFERENCES

Agresti, A. & Finlay, B. (1997). Statistical methods for the social sciences (3rd
ed. 1. Upper Saddle River, NJ: Prentice-Hall Inc.

Bancroft, L. & Mathews, A. (1971). Autonomic correlates of penile erection.
Journal of Psychosomatic Research, 15, 159 — 167.

Beaver, W. (2000). The dilemma of lntemet pornography. Business and Society
Review. 105(3). 373 - 382.

Boynton, L. A. & Wu, H. D. (1999, August). Don't loolgat me! Third-person
effects and TV violence. Paper presented at the annual meeting of the Association for
Education in Journalism and Mass Communication, New Orleans, LA.

Brosius, H. -B. & Engel, D. (1996). The causes of third-person effects: Unrealistic
optimism, impersonal impact, or generalized negative attitudes toward media inﬂuence?
International Journal of Public opinion Research, 8(2), 142 ~162.

Bryant, J. & Zillmann, D. (1996). Violence and sex in the media. In M. B. Salwen,
and D. W. Stacks (Eds.), An integrated Approach to communication theory and research
(pp. 195 — 209), Mahwah, NJ: Lawrence Erlbaum Association, Inc., Publishers.

Cate, F. H. (1996). Cybersex: regulating sexually explicit expression on the Internet.
Behavioral Sciences and the Law, 14(2), 145 - 166.

Chapin, J. R. (2000). Third-person perception and optimistic bias among urban
minority at-risk youth. Communication Research, 27(1). 51 - 81.

Digital Chosun. (2000). Top 10 News Stories of the Information and Technology
Indusﬂ. December 22.

Cohen, J ., Mutz, D., Price, V., & Gunther, A. (1988). Perceived impact of
defamation: An experiment on third-person effects. Public Opinion Quarterly, 52, 161 -
173.

Davison, W. P. (1983). The third-person effect in communication. Public Opinion
QuarterLY.47(1L 1 - 15.

Dearing, J. W. & Rogers, E. M. (1996). Agenda-setting. Thousand Oaks, CA: Sage.
Donnerstein, E. & Berkowitz, L. (1981). Victim reactions in aggressive erotic ﬁlms

as a factor in violence against women. Journal of Personality and Social Psychology, 41,
710 — 724.

59

Duck, J. M., Hogg, M. A., & Terry, D. J. (1999). Social identity and perceptions
of media persuasion: Are we always less inﬂuenced than others? J oumalgof Applied
Social Psychology. 29(9). 1879 - 1899.

 

Duck, J. M. and Mullin, B. A. (1995). The perceived impact of the mass media:
Reconsidering the third-person effect. European J om] of Social Psychology, 25(1), 77 -
93.

Duck, J. M., Terry, D. J ., & Hogg, M. A. (1995). The perceived inﬂuence of
AIDS advertising: Third-person effects in the context of positive media content. Basic
ad Applied SociQPsychology, 17(3), 305 - 325.

Dupagne, M., Salwen, M., & Paul, B. (1999). Impact of question order on the third-
person effect. International Journal of Public Opinion ReseaLch, 11(4). 334 - 345.

Eichholz, M. (1999, August). The third-person effect and the Hiegrchy of
communication effects: The perceivedgersuasive power of public relations. Paper
presented at the annual meeting of the Association for Education in Journalism and Mass
Communication, New Orleans, LA.

Eastin, M. S. and LaRose, R. (2000). lntemet self-efﬁcacy and the psychology of
the digital divide. J ouml of Computer-Mediated Communication [On-line], 6(1).
Available: http://www.ascusc.org/jcmc/vol6/issue1/eastin.html.

Greenberg, B. S. and Salwen, M. B. (1996). Mass communication
theory and research: Concepts and models. In M. B. Salwen, and D. W. Stacks (Eds.), An__
integrated Approach to communication theory and research (pp. 63 — 78), Mahwah, NJ:
Lawrence Erlbaum Association, Inc., Publishers.

Gudykunst, W., Matsumoto, Y., Ting-toomey, S., Nishida, T., Kim, K. S., &
Heyman, S. (1996). The inﬂuence of cultural individulism-collectivism, self-construals,

and individual values on communication styles across cultures. Human Communication
Research, 22(4), 510 - 543.

Gunther, A. C. (1991). What we think others think: Cause and consequence in the
third-person effect. Communication ResearchJI 8(3), 355 - 372.

Gunther, A. C. (1995). Overrating the X-rating: The third-person perception and
support for censorship of pornography. Journal of Communication 45(1), 27 - 38.

Gunther, A. C. & Hwa, A. P. (1996). Public perceptions of television inﬂuence and
opinions about censorship in Singapore. Interaational Journal of Public Opinion Research,
813 ), 248 — 265.

Gunther and Mundy (1993). Biased optimism and the third-person effect.
Jomlism Quarterly. 70(1), 58 - 67.

6O

Gunther, A. C. & Thorson, E. (1992). Perceived persuasive effects of product
commercials and public service announcements: The third-person effects in new domains.
Communication Research. 19(5), 574 - 596.

Gutfeld, G. (1999). The sex drive. Men's Health, 14(8), 116 -120.

 

Henriksen, L. & Flora, J. A. (1999). Third-person perception and children:
Perceived impact of pro- and anti-smoking ads. Communication Research, 26(6), 643 -
665.

Hoffner, C., Buchanan, M., Anderson, J. D., Hubbs, L. A., Kamigaki, S. K.,
Kowalczyk, L., Pastorek, A., Plotkin, R. S., & Silberg, K. J. (1999). Support for censorship
of television violence: The role of the third-person effect and news exposure.
Communication Research, 26(6), 726 — 742.

Hofstede, G. (1991). Cultures and gganizations: Soﬂware of the mind. London:
McGraw-Hill.

Hui, C. H. (1988). Measurement of individualism-collectivism. Journal of
Research in Personality, 22, 17-36.

Hunter, J. E. & Gerbing, D. W. (1982). Machiavellian beliefs and personality:
Construct invalidity of the Machiavellianism dimension. J om] of Personality and Social
Psychology. 43. 1293 -— 1305.

Lambe, J. L. & Shah, D. V. (1999, August). Me speech and the third-person
effect: Suscgptibility, severig, and the willingaess to censor. Paper presented at the annual

meeting of the Association for Education in Journalism and Mass Communication, New
Orleans, LA.

Lee, C. & Yang, S. (1996, August). Third-person perception and support for
censorship of sexually explicit visual content: A Korean case. Paper presented at the annual

meeting of the Association for Education in Journalism and Mass Communication,
Anaheim, CA.

Markus, H. R. and Kitayama, S. (1991). Culture and the self: Implications for
cognition, emotion, and motivation. Psychological Review, 98(2), 224 - 253.

Marshall, T. R. (1988). Public opinion, representation, and the modern Supreme
Court. American Politics Quarterly, 16, 293 — 316.

McLeod, D. M., Eveland, W. P. & Nathanson, A. I. (1997). Support for censorship

of violent and misogynic rap lyrics: An analysis of the third-person effect. Communication
Research, 24(2), 153 — 174.

61

Mehta, M. D. & Plaza, D. (1997). Content analysis of pornographic images
available on the lntemet. The Infonﬂtion Society. 13, 153 -161.

Melillo, W. (1999). Policing the net. Adweek, January 4.

Mishler, W. & Sheehan, R. S. (1993). The Supreme Court as a countermajoritarian
institution? The impact of public opinion on Supreme Court decisions. American Political
Science Review, 87, 87 — 101.

Munro, N. (1999, September). The Web’s pomucopia. National Journal, 38 — 45.

Page, B. I. & Shapiro, R. Y. (1983). Effects of public opinion on policy. American
Political Science Review, 77, 175 — 190.

 

Park, E. (1998). IndivicMism/collectivism, self-concepL_ and social behavior:
F alse-uniqueness and the spiral of silence hypotheses. Unpublished doctoral dissertation,
Cornell University.

Paxton, M. (1996, August). The third-person effect and attitudes toward
expression. Paper presented at the annual meeting of the Association for Education in
Journalism and Mass Communication, Anaheim, CA.

Paxton, M. (1997, March). Mega use, "experts" and the third-person effect. Paper
presented at the annual meeting of the Association for Education in Journalism and Mass
Communication, Southeast Colloquium, Knoxville, TN.

Peiser, W. & Peter, J. (2001). Explaining individual differences in third-person
perception: A limits/possibilities perspective. Communication Research. 28(2). 156 — 180.

Price, V., Huang, L., & Tewksbury, D. (1997). Third-person effects of news
coverage: Orientations toward media. J ournalismﬁand Mass Communication Quarterly,
74(3), 525 - 540.

 

Rimm, M. (1995). Marketing pornography on the Information Superhighway: A
survey of 917,410 images, descriptions, short stories, and animations downloaded 8.5
million times by consumers in over 2000 cities in forty countries, provinces, and territories.
Georgetown Law iournaflj3(5), 1849 - 1934.

Reno v. ACLU, 117 S. Ct. 2329 (1997).
Rojas, H., Shah, D. V., & Faber, R. J. (1997). For the good of others: Censorship

and the third-person effect. International J om] of Public Opinion Research, 8(2), 163 —
186.

62

Rucinski, D. & Salmon, C. T. (1990). The ‘Other’ as the vulnerable voter: A study
of the third-person effect in the 1988 US. presidential campaign. International J oumal of
Public Opinion Research. 2(4). 345 — 368.

Salwen, M. B. (1998). Perceptions of media inﬂuence and support for censorship:
The third-person effect in the 1996 presidential election. Communication Research, 25(3),
259 — 285.

Salwen, M. B. & Driscoll, P. D. (1997). Consequences of third-person perception in
support of press restrictions in the O. J. Simpson trial. Journal of Communication, 432), 60
—78.

Salwen, M. B. & Dupagne, M. (1999). The third-person effect. Communication
Research, 26(5), 523 — 549.

Sanders, G., Deal, J ., & Myers-Bowman, K. (1999). Sexually explicit material on
the lntemet: Implications for family life educators. Journal of Family aad Consumer
Sciences, 91(3), 112 - 115.

Simon, G. E. (1998). Cyberpom and censorship: Constitutional barriers to
preventing access to lntemet pornography by minors. The Journal of Criminal Law &

Criminology, 88(3), 1015 — 1048.

Singelis, T. M., & Brown, W. J. (1995). Culture, self, and collectivist
communication: Linking culture to individual behavior. Human Communication
Research, 21, 354 - 389.

Slade, J. W. (2000). Pornograpm' in America: A reference handbook. Santa Barbara,
CA: ABC-CLIO, Inc.

Stanford Institute for the Quantitative Study of Society (SIQSS). (2000, February).
Internet and socieg: A preliminm rgport. Stanford University.

The Washington Post. (2000). Koreans clash on lntemet sex video: Traditional
defend old values in 'wiredﬁrtion. December 25.

Tiedge, J. T., Silverblatt, A., Havice, M. J ., & Rosenﬁeld, R. (1991). Discrepancy
between perceived ﬁrst-person and perceived third-person mass media effects. Journalism

Quarterly, 68, 141 - 154.

Ting-Toomey, S. (1999). Communicating across cultures. New York, NY:
Guilford Press.

Vandiver, D. M. (2000). Regulating lntemet pornography. Crime & Justice
htemtional. 16(47). 11 - 14.

63

Weimann, G. (2000). Communicatig unreality: Modern media aral the
reconstruction of reality. Thousand Oaks, CA: Sage.

Weinstein, N. D. (1980). Unrealistic optimism about future life events. Journal of
Personality and Social Psychology, 39, 802 - 820.

Youn, S., Faber, R. J ., & Shah, D. V. (2000). Restricting gambling advertising and
the third-person effect. Psychologyaml Marketing, 17(7). 633 — 649.

Zillrnann, D. & Bryant, J. (1982). Pornography, sexual callousness, and the
trivialization of rape. Journal of Communication, 32, 10 — 21.

Zillrnann, D., Bryant, J ., Comisky, P. W., & Medoff, N. J. (1981). Excitation and
hedonic valence in the effect of erotica on motivated intennale aggression. European
Journal of Social Psychology. 11, 233 — 252.

Zillrnann, D., Bryant, J. & Huston, A. C. (1994). Mediaachildrengand the family:
Social scientiﬁc. psychodynamic, and clinicai perspectives. Hillsdale, NJ: Lawrence
Erlbaum Association, Inc., Publishers.

Zillrnann, D. & Weaver, J. B. (1989). Pornography and men’s sexual callouseness
toward women. In D. Zillrnann and J. Bryant (Eds), Pomography: Research advances and
policy considerations (pp. 95 — 125), Hillsdale, NJ: Lawrence Erlbaum Association, Inc.,
Publishers.

 

   

G

1111111111111111111151111