Content-Type: text/html

This paper was presented at the Association for Education in Journalism and
Mass Communication in San Antonio, Texas August 2005.
         If you have questions about this paper, please contact the author
directly. If you have questions about the archives, email
rakyat [ at ] eparker.org. For an explanation of the subject line, 
send email to
<a href="/cgi-bin/wa?LOGON=A3%3Dind0602A%26L%3DAEJMC%26E%3D8bit%26P%3D7224159%26B%3D--%26T%3Dtext%252Fplain%3B%2520charset%3Diso-8859-1" >[log in to unmask]</a> with just the four words, "get help info aejmc," in the
body (drop the "").

(Feb 2006)
Thank you.
Elliott Parker
====================================================================

Third-Person Effect and Censorship
of Web Pornography



By Li-jing Arthur Chang

Li-jing Arthur Chang
Assistant Professor
Department of Mass Communications
Jackson State University
P.O. Box 18590
Jackson, MS 39217-0990
Tel: (601)979-1351
Email: <a href="/cgi-bin/wa?LOGON=A3%3Dind0602A%26L%3DAEJMC%26E%3D8bit%26P%3D7224159%26B%3D--%26T%3Dtext%252Fplain%3B%2520charset%3Diso-8859-1" >[log in to unmask]</a>


Submitted to:
Mass Communication and Society Division
Association for Education in Journalism and Mass Communication
Dr. Renita Coleman
MC&S Division Research Chair
Manship School of Mass Communication
Louisiana State University
Baton Rouge, LA 70803-7202
April 1, 2005

Abstract

The study, which surveyed 710 respondents in Singapore, found that 
third-person effect played a role in the support for the censorship 
of Web pornography. Other factors found to predict the support for 
the censorship measure include gender, age and Internet use. In 
addition, the study also confirmed past empirical evidence about the 
link between third-person effect and undesirable media content, and 
the association between third-person effect and the social distance 
between self and others.





Introduction

Censorship of Internet pornography has been an issue receiving high 
attention in many parts of the world, as research has indicated 
sexually explicit media content may affect the behaviors of teenagers 
(Collins, Elliot, Berry, Kanouse, Kunkel, Hunter & Miu, 2004). In 
Asia, which has the highest number of Internet users among the major 
regions in the world (Internet World Stats, 2004a),[1] many countries 
have scrambled to limit public access to Web pornographic materials 
(Fluendy, 1996). Inside Asia, Singapore is an interesting case 
because it has one of the highest Internet penetration rates in the 
continent (Internet World Stats, 2004b). Despite its effort to 
promote Internet usage, the Singapore government is concerned about 
the negative impact from the online media on the public and has tried 
hard to block public access to pornographic Web content (Ang & 
Nadarajan, 1996).
As research showed that censorship of pornography has received 
widespread support in Singapore (Ang et al., 1996; Gunther & Ang, 
1996), it would be interesting to study the reasons behind the 
general public's support for the censorship, especially in the case 
of censorship for Web pornography – an issue of great public concern. 
As no prior study seems to have explored the general public's 
opinions about their reasons behind supporting the censorship of Web 
pornography, findings from the study will be useful for policy makers 
coping with the censorship issue.[2]
Research has shown many in the public favor censorship for 
potentially harmful media content because they believe others may be 
more susceptible to the negative impact of such content than they are 
(Gunther, 1995; Gunther et al., 1996; Hoffner & Buchanan, 2002; 
Hoffner, Buchanan, Anderson, Hubbs, Kamigaki, Kowalczyk, Pastorek, 
Plotkin & Silberg, 1999; McLeod, Eveland & Nathanson, 1997; Rojas, 
Shah & Faber, 1996; Salwen, 1998; Salwen & Dupagne, 1999; Wu & Koo, 
2001). The concern that others, rather than oneself, may be more 
vulnerable to harmful media effects was first coined by Davison 
(1983) as the third-person effect, also known as the perceptual bias 
between self and others regarding the vulnerability to undesirable 
media content.
As third-person effect has been found in the past to predict support 
for censorship of undesirable media messages (Gunther, 1995; Gunther 
et al., 1996; Hoffner et al., 2002; Hoffner et al., 1999; Lo & Wei, 
2002; McLeod et al., 1997; Rojas et al., 1996; Salwen, 1998; Salwen 
et al., 1999; Wu et al., 2001), the present study intends to explore 
how the intriguing concept of third-person effect affects the levels 
of public support for censoring Web pornography in Singapore. In 
addition, the study attempts to probe the influence of other factors 
found in the past to predict censorship of undesirable media content, 
such as gender (Chia, Lu & McLeod, 2004; Gunther, 1995; Gunther et 
al., 1996; Hoffner et al., 1999; Rojas et al., 1996), age (Gunther et 
al, 1996; Hoffner et al., 1999; Rojas et al., 1996; Wu et al., 2001), 
education (Hoffner et al., 1999; Lo et al., 2002; Salwen, 1998; 
Salwen & Driscoll, 1997; Salwen et al., 1999), and media use (Chia et 
al., 2004; Gunther, 1995; Hoffner et al., 1999; Lo et al., 2002; 
Salwen et al., 1997; Wu et al., 2001).

Literature Review

According to a synthesis of past research, factors predicting public 
support of censorship against undesirable media messages (e.g., media 
violence, pornography, etc.) include third-person effect (Gunther, 
1995; Gunther et al., 1996; Hoffner et al., 2002; Hoffner et al., 
1999; McLeod et al., 1997; Rojas et al., 1996; Salwen, 1998; Salwen 
et al., 1999; Wu et al., 2001), gender (Chia et al., 2004; Gunther, 
1995; Gunther et al., 1996; Hoffner et al., 1999; Rojas et al., 
1996), age (Gunther et al, 1996; Hoffner et al., 1999; Rojas et al., 
1996; Wu et al., 2001), education (Hoffner et al., 1999; Lo et al., 
2002; Salwen, 1998; Salwen et al., 1997; Salwen et al., 1999), and 
media use (Chia et al., 2004; Gunther, 1995; Hoffner et al., 1999; Lo 
et al., 2002; Salwen et al., 1997; Wu et al., 2001).
Third-Person Effect
Third-person effect was first coined by Davidson (1983) as the 
perceptual bias about the impact of undesirable media messages on 
others vs. on oneself. Specifically, the third-person effect 
hypothesis postulates that people in general perceived undesirable 
media messages (e.g., propaganda) will have greater effects on other 
people than on themselves (Davidson, 1983; Gunther & Mundy, 1993; 
Mason, 1995; Price, Huang & Tewksbury, 1997; Salwen & Dupagne, 2001).
Since Davison first proposed the third-person effect, at least 45 
published studies have tested the concept and found support for its 
existence (Perloff, 1999). According to the research, whether the 
undesirable messages is media violence (Duck & Mullin, 1995; Innes & 
Zeitz, 1988), negative political advertising (Cohen & Davis, 1991), 
or pornography (Gunther, 1995), people tend to believe others will be 
more susceptible to the impact of such messages (Perloff, 1999).
Applying the above empirical evidence in the present research, it is 
possible that the perceived effect of Web pornography on self (i.e., 
an adult in Singapore) is going to be less than the perceived effect 
of Web pornography on other adults in Singapore. Or, put another way, 
the perceived effect of Web pornography on other adults is likely to 
be greater than the perceived effect of Web pornography on self. 
Based on this rationale, the first hypothesis for the present study 
is generated:
	H1: The perceived effect of the Web pornography on other adults will 
be greater than the perceived effect on self.
	Research has also shown that the greater the social distance between 
oneself and others, the greater the third-person effect (Chia et al., 
2004; Cohen, Mutz, Price & Gunther, 1988; Duck, Hogg & Terry, 1995; 
Gibbon & Durkin, 1995; Gunther, 1991; McLeod et al., 1997; Lo et al., 
2002; White, 1997). For example, Cohen et al. (1988) found the size 
of third-percent perception increased as the social distance between 
self and others became wider. Subjects believed the impact on 
"others" will grow bigger when "others" changed from "other Stanford 
students" to "other Californians" to the public at large (Cohen et 
al, 1988). Similarly, McLeod et al. (1997) discovered from a research 
on a group of mass communication majors in a Delaware university that 
the third-person effect turned larger when the definitions of 
"others" changed from "Delaware students" to "New York/Los Angeles 
youth" to "average person."
	Applying the social distance hypothesis in the current study, it is 
possible that the third-person effect will be greater when the social 
distance becomes wider. For example, according to the social distance 
hypothesis, when the "others" in the current study changed from 
"other adults" to "young people" in Singapore, the third-effect will 
become larger. Following this rationale, the second hypothesis is generated:
	H2: The size of third-person effect will increase as the social 
distance of the comparison group increases (i.e., when "others" 
change from "other adults" to "young people").
	Past research on the impact of undesirable media content also showed 
that third-person effect is closely related to public support for 
censorship of such content (Gunther, 1995; Gunther et al., 1996; 
Hoffner et al., 2002; Hoffner et al., 1999; Lo et al., 2002; McLeod 
et al., 1997; Rojas et al., 1996; Salwen, 1998; Salwen et al., 1999; 
Wu et al., 2001). As the third-person effect is really reflecting the 
phenomenon that the perceived negative media effect on others are 
greater than the perceived negative effect on oneself, it would be 
interesting to see if perceived negative media effect on others would 
be a stronger predictor behind the support for censoring Web 
pornography, compared to the perceived media effect on self. As the 
potentially harmful media content tested in the current study is Web 
pornography, perceived effect of Web pornography on self and 
perceived effect of Web pornography on two groups of others (i.e., 
other adults in Singapore and young people in Singapore) can be 
tested as predictors for public support to censor Web pornographic 
content. Based on the above rationale, the following research 
question is generated:
	RQ1: How do the levels of public support for censoring Web 
pornography affected by perceived negative effects of the Web 
pornography on oneself, other adults and young people?
Research testing the influence of third-person effect on public 
approval for censorship also found other factors, along with 
third-percent perception, are behind the support for censorship of 
undesirable media massages. These factors include gender (Chia et 
al., 2004; Gunther, 1995; Gunther et al., 1996; Hoffner et al., 1999; 
Rojas et al., 1996), age (Gunther et al, 1996; Hoffner et al., 1999; 
Rojas et al., 1996; Wu et al., 2001), education (Hoffner et al., 
1999; Lo et al., 2002; Salwen, 1998; Salwen et al., 1997; Salwen et 
al., 1999), and media use (Chia et al., 2004; Gunther, 1995; Hoffner 
et al., 1999; Lo et al., 2002; Salwen et al., 1997; Wu et al., 2001).
Gender
	Compared with men, women are more likely to support restrictions on 
pornography (Herman & Bordner, 1983; U.S. Commission on Obscenity and 
Pornography, 1970). Several researches testing the influence of 
third-person effect on censorship support also found gender as a 
significant predictor behind public approval for the censorship of 
undesirable media content (Gunther, 1995; Gunther et al., 1996; 
Hoffner et al., 1999; Lo & Wei, 2002; Rojas et al., 1996). 
Specifically, the studies showed women are more likely than men to 
support the censorship of undesirable media content, such as TV 
violence (Gunther et al., 1996; Hoffner et al., 1999) and pornography 
(Gunther, 1995; Lo et al., 2002; Rojas et al., 1996).
Age
	Some past researches testing the impact of third-person effect in 
censorship support also found that age can be another predictor of 
public approval for censorship (Gunther et al., 1996; Hoffner et al., 
1999; Rojas et al., 1996; Wu et al., 2001). For example, Gunther et 
al. (1996) and Hoffner et al. (1999) found out that age is a 
predictor behind the levels of public support for censoring TV 
violence. In addition, two studies conducted based on samples from 
universities students (Rojas et al., 1996; Wu et al., 2001) also 
discovered older students, compared with their younger counterparts, 
are more in favor of censorship measure of undesirable media content.
Education
	There is also evidence showing education could be a good predictor 
behind public support for censoring undesirable media content (Lo et 
al., 2002; Salwen et al., 1997; Salwen, 1998; Salwen et al., 1999). 
Surveys on adults in the United States (Salwen et al., 1997; Salwen, 
1998; Salwen et al., 1999) showed education has a negative 
relationship with support for censorship of media content. However, a 
survey on high school and university students (Lo et al., 2002) found 
out that older students tend to support censorship of Web pornography.
Media Use
	Past research has shown that media use as a predictor affecting the 
levels of public support for censorship measures against undesirable 
media messages (Chia et al., 2004; Gunther, 1995; Hoffner et al., 
1999; Lo et al., 2002; Salwen et al., 1997; Wu et al., 2001).
	The term media in the case of the present study means the Internet. 
Presumably, people who use more Internet content will be less likely 
to support censorship of Web content. This assumption coincides with 
evidence from studies by Lo et al. (2002) and Wu et al. (2001), who 
found that exposure to Internet pornography tends to reduce support 
for restrictions on such Web content. The above assumption and 
empirical evidence suggest that it is possible the Internet usage may 
reduce support for censoring Web pornography.
According to the above review, besides the third-person effect, 
gender, age, education and Internet usage may be other factors 
predicting the levels of public support for restricting Web 
pornographic content. Therefore, a second research question is generated:
RQ2: How do the levels of public support for censoring Web 
pornography affected by factors such as gender, age, education and 
Internet usage?


Method

Sampling
Data for this study was collected in face-to-face interviews with 710 
adults (aged 18 or older) in Singapore from April to May, 2001.[3] 
The 710 subjects with complete responses to the survey questionnaire 
are from the original sample pool of 1447 households. Given the 
procedure to randomly select one adult per household for interview, 
the completed 710 questionnaires represent a response rate of 49.1%. 
The original sample pool of 1447 households was randomly selected 
through stratified sampling, a probability sampling procedure to 
ensure the subjects are randomly selected in Singapore. Specifically, 
the area in the country is divided into clusters, stratified for 
geographical locations, housing types and socio-economic classes. A 
number of clusters were randomly selected and a number of households 
inside the selected clusters are also randomly chosen. Finally, 
inside each chosen household, an interviewee was randomly selected. 
The whole procedure was designed to ensure all the interviewees were 
randomly selected.
Operational Definitions
The interviews went on for about half an hour for each subject 
completing a questionnaire. The questionnaire items covers all 
concepts tested for this research: (1) negative effect of Web 
pornography on self, (2) negative effect of Web pornography on other 
adults, (3) negative effect of Web pornography on young people, (4) 
third-person effect, (5) support for censorship of Web pornography; 
(5) demographics (gender, age and education), and (6) time spent 
weekly on the Web.
Negative effects of Web pornography on self, other adults and young 
people. For the measurements of the three variables (i.e., negative 
effect of Web pornography on self, negative effect of Web pornography 
on other adults, and negative effect of Web pornography on young 
people), the respondents were asked to indicate on a 4-point Likert 
scale (with "not at all" at one end of scale and "a great deal" at 
other end) the amount of influence Internet pornography on self, 
other adults and young people. The responses to the Likert scale was 
coded as follows: "not at all" was coded "1," "a great deal" was 
coded as "4," and the answers in between two polarized responses were 
coded as "2" and "3" accordingly.
Third-person effect. The measurement of the third-person effect is 
derived by comparing the difference between the negative effect of 
Web pornography on self and the negative effect of Web pornography on 
others. As the second hypothesis of the present study is aimed to see 
how third-person effect may change when the social distance between 
self and others widens, there are two types of "others" in this case: 
the first others is "other adults," and the second others is "young 
people," with the latter having wider social distance with "self." 
Therefore, third-person effect between self and others as other 
adults was derived by subtracting negative effect of Web pornography 
on self from negative effect of Web pornography on other adults. 
Similarly, third-person effect between self and others as young 
people was derived by subtracting negative effect of Web pornography 
on self from negative effect of Web pornography on young people.
Demographics (gender, age and education). During the face-to-face 
interviews for the present study, interviewees' genders, age 
categories[4] and education levels were also recorded. The responses 
to the gender question were coded as follows: "0" for "female" and 
"1" for "male." For age questions, respondents' age categories were 
coded from 1 to 10 depending on their answers.[5] In terms of 
education levels, the responses were coded as follows: "1" for "no 
formal education," "2" for "primary school education," "3" for 
"secondary school graduates," "4" for "polytechnic diploma holders," 
and "5" for "university graduates."[6]
Time spent weekly on the Web. For the measurement of Web use, the 
respondents were asked to indicate how many hours or minutes they 
spent on the Internet a week. Their answers, when in the form of 
minutes, were converted to its equivalents in hours by dividing the 
number of minutes by 60. A new variable was created to represent the 
total time of weekly Web usage in terms of hours for all respondents.
Data Analysis
Hypothesis 1. The first hypothesis posits that the perceived effect 
of the Web pornography on other adults will be greater than the 
perceived effect on self. To test the hypothesis, the mean of 
perceived negative effect from Web pornography on self is compared 
with the mean of perceived negative effect from Web pornography on 
other adults to see if the latter is indeed, as third-person effect 
hypothesis posits, larger than the former. Secondly, a t-test is 
performed to see if the perceived effect of Web pornography on self 
is significantly different from the perceived effect of Web 
pornography on other adults.
Hypothesis 2. The second hypothesis posits the size of third-person 
effect will increase as the social distance of the comparison group 
increases (i.e., when "others" change from "other adults" to "young 
people"). To test the hypothesis, the means of perceived negative 
effect from Web pornography on self, other adults and young people 
are compared to see if when social distance between self and others 
become wider, the perceived harmful effect will increase as well. 
Secondly, a t-test was used to see if the perceptual bias about the 
negative effect of Web pornography on self vs. other adults is 
significantly different from the perceptual bias about the effect on 
self vs. young people regarding the same issue. The t-test is 
conducted to ensure if the enlarged third-person effect, if any, is 
statistically significant.
Research Questions 1 and 2. The first research question asks the 
extent to which the levels of public support for censoring Web 
pornography are affected by perceived negative effects of the Web 
pornography on oneself, other adults and young people. The second 
research question inquires the extent to which the levels of public 
support for censoring Web pornography are affected by factors such as 
gender, age, education and Internet usage. To test the two research 
questions, a regression analysis was performed, with the levels of 
support for censoring Web pornography as the dependent variables, and 
gender, age, education, Internet usage, negative effect of Web porn 
on self, negative effect of Web pornography on other adults, and 
negative effect of Web pornography on young people as the independent 
variables.


Results

The sample of 710 adults in Singapore was 51.1% male, 71% married, 
and 26.3% with tertiary education. The median age category was 35-39 
years old, and the median monthly salary was in the range of US$1,801 
to US$2,401. When compared with national census of Singapore 
population (Singapore population, 2001), the demographics for the 
sample used for the current study are very comparable.[7]
Hypothesis 1
	The first hypothesis posits that the perceived effect of the Web 
pornography on other adults will be greater than the perceived effect 
on self. As shown in Table 1, the study found that the perceived 
negative effect of Web pornography on other adults is greater than 
the perceived negative effect of Web pornography on self. In 
addition, according the results of t-test, the perceived negative 
effect on other adults is significantly different from the perceived 
effect on self. Based on this finding, the present study showed ample 
support for the first hypothesis.




Table 1.  	Mean Scores of Perceived Effect of Web Pornography on Self 
and Other Adults
	(Standard Deviations in Parentheses)

N	Self	Other Adults	Difference	t-value


710	2.01 (1.13)	2.63 (1.03)	.61		16.78***


Note: The responses were coded with a 4-point Likert scale, with "1" 
as "not at all" and "4" as "a great deal." Significance level was 
calculated using paired t-test.
*** p < .001




Hypothesis 2
The second hypothesis posits the size of third-person effect will 
increase as the social distance of the comparison group increases 
(i.e., when "others" change from "other adults" to "young people"). 
As shown in Table 2, while the perceived negative effect of Web 
pornography on other adults was larger than the perceived effect of 
Web pornography on self, the perceived effect of Web pornography on 
young people is much greater than the perceived effect of Web 
pornography on self. This finding provided a preliminary support for 
hypothesis 2 because it showed the perceptual difference between self 
and others changed as social distance between self and others widened 
(i.e., the definitions of  "others" changed from "other adults" to 
"young people").



Table 2.  	Mean Scores of Perceived Effect of Web Pornography on 
Self, Other Adults and
	Young People (Standard Deviations in Parentheses)

N	Self	Other Adults	Young People


710	2.01 (1.13)	2.63 (1.03)	3.42 (.85)


Note: The responses were coded with a 4-point Likert scale, with "1" 
as "not at all" and "4" as "a great deal." Significance level was 
calculated using paired t-test.



	A more solid support of hypothesis 2 comes from findings illustrated 
in Table 3. The figures in Table 3 showed that the perceptual bias 
between self and young people regarding the negative effect of Web 
pornography was not only larger than the perceptual bias between self 
and other adults, but was also statistically different from the 
latter. Specifically, the results of paired t-test revealed that the 
difference between the self-young people perceptual bias and the 
self-other adults perceptual bias was statistically significant (see 
Table 3). This evidence provided ample support for hypothesis 2.



Table 3. 	Mean Scores of Perceptual Bias between Self and Other 
Adults, and Perceptual
	Bias between Self and Young People (Standard Deviations in Parentheses)

N	Self & Other Adults	Self & Young People		Difference        t-value


710	.61 (.98)	1.41 (1.22)			.80                   22.66***


Note: The perceptual bias between the perceived negative effect of 
Web pornography on self and the perceived negative effect of Web 
pornography on other adults was derived by subtracting the former 
from the latter.  Similarly, the perceptual bias between the 
perceived negative effect of Web pornography on self and the 
perceived negative effect of Web pornography on young people was 
derived by subtracting the former from the latter. Significance level 
was calculated using paired t-test.
*** p < .001


Research Questions 1 and 2
The first research question asks the extent to which the levels of 
public support for censoring Web pornography are affected by 
perceived negative effects of the Web pornography on oneself, other 
adults and young people. The second research question inquires the 
extent to which the levels of public support for censoring Web 
pornography are affected by factors such as gender, age, education 
and Internet usage. The two research questions were test using a 
regression analysis, with the levels of support for censoring Web 
pornography as the dependent variable, and gender, age, education, 
Internet usage, negative effect of Web porn on self, negative effect 
of Web porn on other adults, and negative effect of Web porn on young 
people as the as the independent variables.
	As shown in Table 4, the results of the regression analysis showed 
gender, age, Internet use, negative effect of Web pornography on 
self, negative effect of Web pornography on other adults, and 
negative effect of Web pornography on young people were found to be 
significant predictors of support for censorship of Web pornography. 
The findings largely confirmed the previous findings about likely 
predictors for censorship of undesirable media content (Chia et al., 
2004; Gunther, 1995; Gunther et al., 1996; Hoffner et al., 2002; 
Hoffner et al., 1999; Lo et al., 2002; McLeod et al., 1997; Rojas et 
al., 1996; Salwen, 1998; Salwen et al., 1997; Salwen et al., 1999; Wu 
et al., 2001).
Examining the positive or negative sign of the significant predictors 
reveals the relationship between the predictors and the dependent 
variable (Schroeder, Sjoquist & Stephan, 1986). Among significant 
predictors found through the test of the research


Table 4.  	Predictors for Support of Censorship of Web Pornography

			Standardized
			Coefficients


Gender			-.13***
Age			-.09*
Education			-.01
Internet Use			-.07+
Negative effect of Web pornography on self	-.14**
Negative effect of Web pornography on other adults	-.22***
Negative effect of Web pornography on young people	-.14***


N = 710
Adjusted R Square = .21
*** p < .001; ** p < .01; * p < .05; + p < .10



questions, age, negative effect of Web pornography on self, negative 
effect of Web porn on other adults, and negative effect of Web 
pornography on young people have positive coefficients, revealing 
their positive relationships with the dependent variable support for 
censorship of Web pornography. For example, such positive 
relationship in the case of age could mean that the older a person 
is, the more likely he or she will support retractions on Internet 
pornography. In contrast to the positive relationship, gender and 
Internet usage have negative coefficients and therefore negative 
relationship with the dependent variable. For example, a negative 
relationship between Internet use and the dependent variable means 
that the more a person uses the Internet, the less likely he or she 
is going to support censorship of Web pornography. In a addition, 
since gender is coded as a dummy variable with "0" representing 
"female" and "1" representing "male," its negative relationship with 
the dependent variable means women are more likely to support 
restrictions on Web pornographic content.
Aside from reviewing the negative/positive relations between the 
predictors and the dependent variable, examining the relative sizes 
of the standardized coefficients also reveals the sizes of effects of 
different significant predictors on the dependent variable 
(Schroeder, Sjoquist & Stephan, 1986). Due to the relative size of 
its coefficient, perceived negative effect of Web pornography on 
other adults has the greatest influence on the dependent variable 
support for censorship of Web pornography (see Table 4). Following 
the same logic, perceived effect of Web pornography on young people 
and perceived effect of Web pornography on self are next most 
important predictors with equal sizes of regression coefficients, 
followed by gender, age, and internet use.
	In addition, as the third-person effect is reflecting the phenomenon 
that the perceived negative media effect on others are greater than 
the perceived negative effect on oneself, the present research was 
aimed to test if perceived negative media effects on both types of 
"others" (i.e., "other adults" and "young people") would be a 
stronger predictors behind the support for censoring Web pornography, 
compared to the perceived media effect on self. However, the findings 
of the present showed that while perceived negative effect of Web 
pornography on "other adults" is a stronger predictor than perceived 
negative effect of Web pornography on "self," perceived negative 
effect of Web pornography "young people" is not different from 
perceived negative effect of Web pornography on "self." Judging from 
the above findings, it seemed that the third-person effect affected 
support for censorship of Web pornography when "others" are "other 
adults" and did not affect support for censorship of Web pornography 
when "others" are "young people."


Discussion

	The present research tests two hypotheses and two research 
questions. The first hypothesis posits that the perceived effect of 
the Web pornography on other adults will be greater than the 
perceived effect on self. The second hypothesis posits the size of 
third-person effect will increase as the social distance of the 
comparison group increases (i.e., when "others" change from "other 
adults" to "young people"). The first research question asks the 
extent to which the levels of public support for censoring Web 
pornography are affected by perceived negative effects of the Web 
pornography on oneself, other adults and young people. The second 
research question inquires the extent to which the levels of public 
support for censoring Web pornography are affected by factors such as 
gender, age, education and Internet usage.
	The present study provided ample support for the hypotheses, 
confirming both the existence of the third-person effect and the 
social distance corollary (i.e., the greater the social distance 
between self and others, the larger the third-person effect) in the 
context of public attitudes about the negative effect of Web 
pornography in Singapore. In addition, the testing of the research 
questions showed gender, age, Internet usage, perceived negative 
effects of the Web pornography on oneself, perceived negative effects 
of the Web pornography on other adults, and perceived negative 
effects of the Web pornography on young people are all significant 
predictors of the public support for censorship of Web pornography in 
Singapore. Findings of the current study are very interesting because 
it largely confirmed previous empirical evidence about the 
third-person effect and the influence of the effect and other factors 
behind the support for censorship of Web pornography. In addition, 
the research also measures the general public's attitudes toward the 
third-person effect and factors behind support for censorship measure 
in Web pornography. The findings apparently have implications for 
policymakers regarding the issues of Web pornography censorship.
	Specifically, the present study confirmed empirical evidence in 
previous research supporting the existence of third-person effect in 
the context of undesirable media content (Perloff, 1999). The 
findings of the current research showed that the public perceived Web 
pornography as having greater negative effect on other adults than on 
themselves. In addition, the present study also supported past 
findings on the social distance corollary (Chia et al., 2004; Cohen, 
Mutz, Price & Gunther, 1988; Duck, Hogg & Terry, 1995; Gibbon & 
Durkin, 1995; Gunther, 1991; McLeod et al., 1997; Lo et al., 2002; 
White, 1997). Similar to previous findings, the present study lends 
further support to the hypothesis that the wider social distance 
between self and others, the larger the third-person effect. Findings 
showed that when the "others" changed from "other adults" to "young 
people," the perceived negative effect of Web pornography on others increased.
	Besides confirming third-person effect and the social distance 
corollary, the present study also support past research findings 
regarding predictors behind the support for censorship of undesirable 
media content. Specifically, past research showed third-person 
effect, or perceptual bias regarding negative effect of undesirable 
media messages on self vs. others, contributed to the support for 
censorship of such media content (Gunther, 1995; Gunther et al., 
1996; Hoffner et al., 2002; Hoffner et al., 1999; McLeod et al., 
1997; Rojas et al., 1996; Salwen, 1998; Salwen et al., 1999; Wu et 
al., 2001). As the third-person effect is really reflecting the 
phenomenon that the perceived negative media effect on others are 
greater than the perceived negative effect on oneself, the present 
research test if perceived negative media effect on "others" would be 
a stronger predictor behind the support for censoring Web 
pornography, compared to the perceived media effect on self. As the 
potentially harmful media content tested in the current study is Web 
pornography, perceived effect of Web pornography on self and 
perceived effect of Web pornography on two groups of "others" (i.e., 
other adults in Singapore and young people in Singapore) was tested 
as predictors for public support to censor Web pornographic content. 
In other words, perceived negative effect of Web pornography on other 
adults, perceived negative effect of Web pornography on young people, 
and perceived negative effect of Web pornography on self were all 
tested as predictors of support for censorship of Web pornography. 
And if third-person effect does influence support for censorship of 
Web pornography, the negative effects of Web pornography on both 
other adults and young people will be stronger predictors of Web 
pornography censorship support than the negative effect of Web 
pornography on self. As predicted, the results of the current study 
showed that the perceived negative effect of Web pornography on other 
adults is a stronger predictor than the perceived negative effect of 
Web pornography on self, judging from their relative sizes of 
regression coefficients on the dependent variable support for 
censorship of Web pornography (see Table 4). However, the findings 
did not show any major difference in the size of regression 
coefficients for both the negative effect of Web pornography on young 
people and the negative effect of Web pornography on self (see Table 
4). Judging from the above findings, it seemed that the third-person 
effect affected support for censorship of Web pornography when 
"others" are "other adults" and did not affect support for censorship 
of Web pornography when "others" are "young people." This finding is 
very intriguing for two reasons: (1) there is an apparent greater 
third-person effect when "others" are "young people" rather than 
"other adults" (see Tables 2 and 3); and (2) the greater third-person 
effect – meaning greater concern for possible harm to "young people" 
– did not translate into support for censorship of Web pornography. 
Put it another way, it seemed that the respondents believed young 
people are more vulnerable to the negative effect of Web pornography 
than other adults, but their support for the censorship of Web 
pornography seemed to be based more on the negative effect of Web 
pornography on other adults than the negative effect Web pornography 
on young people. One possible explanation of this phenomenon may be 
that even though adults believe young people in Singapore are 
vulnerable to negative effect of Web pornography, the youngsters, 
however, may not have as much access to Web pornography as is the 
case of other adults. It is possible that due to strong family ties 
in Singaporean families, parents can effectively prevent young people 
from accessing pornographic Web content.
	Besides the third-person effect, demographic variables (i.e., 
gender, age and education) and Internet usage are also investigated 
in the present study for their possible influences in the levels of 
public support for the censorship of Web pornography. The results 
showed gender, age and Internet usage are significant predictors of 
the public support for censorship of Web pornography (see Table 4). 
Education, on the other hand, was not found to be a significant 
predictor of such censorship measure. The findings mostly confirmed 
past empirical evidence. For example, the findings that women are 
more likely than men to support censorship of Web pornography are 
consistent with previous findings (Gunther, 1995; Gunther et al., 
1996; Hoffner et al., 1999; Lo et al., 2002; Rojas et al., 1996). 
Similarly, the findings that more exposure to the Internet tend to 
lower support for censorship of Web pornography also match the 
researcher's prediction made based on a synthesis of past research on 
the link between media use and support for censorship measure (Chia 
et al., 2004; Gunther, 1995; Hoffner et al., 1999; Lo et al., 2002; 
Salwen et al., 1997; Wu et al., 2001). Although education was not 
found to be a predictor of support for censorship of Web pornography, 
it is not an inconceivable finding. In a society like Singapore that 
prides itself with high Internet penetration (Iritani, 2000), it is 
likely the majority of the public are well aware of the potential 
harms of Web pornography, regardless of their educational backgrounds.
	Besides confirming past research findings, another contribution of 
the current research is the fact it is a large-scale survey seeking 
to use third-person effect and other factors (such as demographics 
and Internet usage) to explain the levels of public support for 
censorship of Web pornography. Previous studies on the link between 
third-person effect and the support for censorship of pornographic 
media content seemed to either focus on non-Internet pornographic 
content (Gunther, 1995; Gunther et al., 1996; Hoffner et al., 2002; 
Hoffner et al., 1999; McLeod et al., 1997; Rojas et al., 1996; 
Salwen, 1998; Salwen et al., 1999), or surveyed student populations 
for the link between third-person effect and support of censoring Web 
pornography (Lo et al., 2002; Wu et al., 2001). The present study 
thus provides more evidence about the general public's attitudes 
toward how third-person effect and other factors (such as gender, 
age, and Internet usage) affect the levels of the support for 
censorship of Web pornography.
	As the present study is a large-scale survey on the public opinions, 
its implications for policy makers may be quite useful. First of all, 
the survey found that women, older people, and infrequent Internet 
users are more likely to support censorship measures against Web 
pornographic content. Given this finding, policymakers will know 
where their support base will come from if they need to establish or 
strengthen restrictions on public access to Web pornography. 
Secondly, as findings from the current study showed that the 
perceived negative impact of Web pornography on other adults, rather 
than the perceived impact on self or young people, is the biggest 
reason why people support censorship of Web pornography in Singapore, 
it may be a sign suggesting access to pornography is more prevalent 
among adults than among young people. This may be vital information 
for policymakers in the country.
	Limitations of the study include the exclusion from the sample 
non-residents in Singapore. Although Singapore citizens and permanent 
residents currently comprise more than 80% of the population[8] and 
the present sample on them should be sufficient to reflect public 
opinion in the country, inclusion of the non-residents, however, will 
provide a more complete picture of the issues investigated. Another 
limitation of the current study is the measurement of media use. 
Although using weekly time spent on the Web should be a good measure 
of media use, a more precise measure of the media use would be the 
respondents' exposure to Web pornographic materials.
	Still, the present research is very interesting and useful. The 
study confirmed past empirical evidence about the existence of 
third-person effect and social distance corollary. In addition, its 
findings on the influence of third-person effect and other factors 
(such as gender, age and Internet usage) on the levels of public 
support for censorship of Web pornography should be useful to 
policymakers. At a time when more and more people are surfing the 
Net, the current study will help to add to the understanding of some 
of the public attitudes toward the use of the young and developing media.
Reference:

Ang, P.H. & Nadarajan, B. (1996). Censorship and the Internet: a 
Singapore perspective. Communications of the ACM, 39, 72-78.

Chia, S. C., Lu, K., & McLeod, D. M. (2004). Sex, lies, and video 
compact disc: a case study on third-person perception and motivations 
for media censorship. Communication Research, 31, 109-130.

Cohen, J., & Davis, R. G. (1991). Third-person effects and the 
differential impact in negative political advertising. Journalism 
Quarterly, 68. 680-688.
Davison, W. P. (1983). The third-person effect in communication. 
Public Opinion Quarterly, 47, 1-15.

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.

Collins, R. L., Elliot, M. N., Berry, S. H., Kanouse, D. E., Kunkel, 
D., Hunter, S. B., & Miu, A. (2004). Watching sex on television 
predicts adolescent initiation of sexual behavior. Pediatrics, 114, 280-289.

Duck, J. M., Hogg, M. A., & Terry, D. J. (1995). Me, us and them: 
political identification and the third-person effect in the 1993 
Australian federal election. European Journal of Social Psychology, 
25, 195-215.

Duck, J. M., & Mullin, B. (1995). The perceived impact of mass media: 
reconsidering the third person effect. European Journal of Social 
Psychology, 25, 77-93.

Fluendy, S. (1996). Pandora's box. Far Eastern Economic Review, 159.

Gibbon, P., & Durkin, K. (1995). The third person effect: social 
distance and perceived media bias. European Journal of Social 
Psychology, 25, 597-602.

Gunther, A. (1991). What we thin others think: cause and consequence 
in the third-person effect. Communication Research, 18, 355-372.

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

Gunther, A. C., & Ang, P. H. (1996). Public perceptions of television 
influence and opinions about censorship in Singapore. International 
Journal of Public Opinion Research, 8, 248-265.

Gunther, A. C., & Mundy, P. (1993). Biased optimism and the 
third-person effect. Journalism Quarterly, 70, 58-67.

Herman, M. S., & Bordner, D. C. (1983). Attitudes toward pornography 
in Southern community. Criminology, 21, 349-374.

Hoffner, C., & Buchanan, M. (2002). Parents' responses to television 
violence: the third-person perception, parental mediation, and 
support for censorship. Mediapsychology, 4, 231-252.

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, 726-742.

Innes, J. M., & Zeitz, H. (1988). The public's view of the impact of 
the mass media: a test of the "third person" effect. European Journal 
of Social Psychology, 18, 457-463.

Internet World Stats (2004a). Internet usage statistics – the big 
picture. http://www.internetworldstats.com/stats.htm.

Internet World Stats (2004b). Top 22 countries with the highest 
Internet penetration rate (percentage of the population using the 
Internet). http://www.internetworldstats.com/top25.htm.

Iritani, E. (2000). Singapore loosens its reins to get Internet 
economy running Asia. Los Angeles Times, July 23, 2000, Section C, 1.

Lo, W., & Wei, R. (2002). Third-person effect, gender, and 
pornography on the Internet. Journal of Broadcasting & Electronic 
Media, 46, 13-33.

Mason, L. (1995). Newspaper as repeater: an experiment on defamation 
and third-person effect. Journalism and Mass Communication Quarterly, 
72, 610-620.

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, 153-174.

Perloff, R. M. (1999). The third-person effect: a critical review and 
synthesis. Media Psychology, 1, 353-378.

Price, V., Huang, L., & Tewksbury, D. (1997). Third-person effects on 
news coverage: orientations toward media. Journalism and Mass 
Communication Quarterly, 74, 525-540.

Rojas, H., Shah, D. V., & Faber, R. J. (1996). For the good of the 
others: censorship and the third-person effect. International Journal 
of Public Opinion Research, 8, 163-186.

Salwen, M. B. (1998). Perceptions of media influence and support for 
censorship: the third-person effect in the 1996 presidential 
election. Communication Research, 25, 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, 47, 60-78.

Salwen, M. B., & Dupagne, M. (1999). The third-person effect: 
perceptions of the media's influence and immoral consequences. 
Communication Research, 26, 523-549.

Salwen, M. B., & Dupagne, M. (2001). Third-person perception of 
television violence: the role of self-perceived knowledge. 
Mediapsychology, 3, 211-236.

Schroeder, L. D., Sjoquist, D. L., & Stephan, P. E. (1986). 
Understanding Regression Analysis: An Introductory Guide (Sage 
Publications Series: Quantitative Applications in the Social 
Sciences, No. 57). Newbury Park, CA: Sage Publications, Inc.

Singapore Population (2001). Census of Population 2000. 
http://www.singstat.gov.sg/keystats/people.html#census.

U.S. Commission on Obscenity & Pornography. (1970). Report. 
Washington, D.C.: U.S. Government Printing Office.

White, H. A. (1997). Considering interacting factors in the 
third-person effect: argument strength and social distance. 
Journalism and Mass Communication Quarterly, 74, 557-564.

Wu, W., & Koo, S. H. (2001). Perceived effects of sexually explicit 
Internet content: the third-person effect in Singapore. Journalism 
and Mass Communications Quarterly, 78, 260-274.

[1] Notes:

  The Website that provides the information, Internet World Stats, 
reported that the Internet usage information comes from data 
published by Nielsen//NetRatings, by International Telecommunications 
Union, by NICs and other reliable sources.
[2]
  Two published studies about reasons behind the support for 
censoring Web pornography based their findings on responses from high 
school and university students (Lo & Wei, 2002; Wu & Koo, 2001), 
rather than from of the large-scale general public. For example, Wu 
and Koo (2001) surveyed a group of students at a university in 
Singapore to explore factors behind the support for censorship of Web 
pornography. Similarly, Lo and Wei (2002) polled high school and 
university students in Taiwan on reasons behind their levels of 
support for censoring Web pornography. The present study targets at a 
different population – the general public – to see their levels of 
support for the censorship of Web pornography.

[3]  The respondents are either Singaporeans or Singaporean permanent 
residents. They do not include non-residents in Singapore. According 
to census figures by the Singapore government, non-residents in 
Singapore comprise of 19.7% of Singapore's total population of 4.13 
million in 2001 (source: Statistics Singapore Newsletter [url: 
http://www.singstat.gov.sg/ssn/feat/nov2003/pg10-13.pdf] released in 
September 2003).

[4]  In an effort to encourage responses to the age question, the 
respondents were asked to indicate their age categories, instead of 
the actual age. The age categories were coded as follows: "1" for 
"18-24 years old," "2" for "25-29 years old," "3" for "30-34 years 
old," "4" for "35-39 years old," "5" for "40-44 years old," "6" for 
"45-49 years old," "7" for "50-54 years old," "8" for "55-59 years 
old," "9" for "60-64 years old," and "10" for "65 years old and older."

[5]  See endnote No. 4.

[6]   The list of coding option for the variable of education (i.e., 
no formal education, primary school education, secondary school 
graduates, polytechnic diploma holders, and university graduates) was 
derived thru conversion from the original response options: no formal 
education, primary 6 or less, some secondary, "O" level, "A" level, 
diploma, and degree & above.  The conversion is designed to translate 
the educational qualifications to easier-to-understand terms to 
facilitate the interpretation of research analysis.

[7]  The demographic statistics of the sample for the current study 
(n  = 710) is very comparable with the population parameters obtained 
during the census in 2000 (Singapore population, 2001):

* The census figures for various demographic categories are for 
Singapore citizens and permanent residents. As in the case of the 
current sample, the demographic figures provided for the census do 
not cover records for non-residents in Singapore.
** The records with a "**" sign are for those aged 15 and over; the 
records without a "**" sign are about people of all ages.
*** Tertiary education includes polytechnic diploma holders and 
university degree holders.
**** The figures are converted from original Singapore dollar figures.

[8]  See endnote 3.