Pornography and Attitudes Supporting Violence Against Women: Revisiting the Relationship in Nonexperimental Studies

Abstract

A meta-analysis was conducted to determine whether nonexperimental studies revealed an association between men's pornography consumption and their attitudes supporting violence against women. The meta-analysis corrected problems with a previously published meta-analysis and added more recent findings. In contrast to the earlier meta-analysis, the current results showed an overall significant positive association between pornography use and attitudes supporting violence against women in nonexperimental studies. In addition, such attitudes were found to correlate significantly higher with the use of sexually violent pornography than with the use of nonviolent pornography, although the latter relationship was also found to be significant. The study resolves what appeared to be a troubling discordance in the literature on pornography and aggressive attitudes by showing that the conclusions from nonexperimental studies in the area are in fact fully consistent with those of their counterpart experimental studies. This finding has important implications for the overall literature on pornography and aggression.

Full text

AGGRESSIVE BEHAVIOR
Volume 35, pages 1–7 (2009)
Pornography and Attitudes Supporting Violence
Against Women: Revisiting the Relationship
in Nonexperimental Studies
Gert Martin Hald
1,2
, Neil M. Malamuth
2
, and Carlin Yuen
3
1
University of California, Los Angeles, California
2
University of Copenhagen, Copenhagen, Denmark
3
New York University Law School, New York City, New York
:::::::::::::::::::::::::::::::::::::::::
A meta-analysis was conducted to determine whether nonexperimental studies revealed an association between men’s pornography
consumption and their attitudes supporting violence against women. The meta-analysis corrected problems with a previously
published meta-analysis and added more recent findings. In contrast to the earlier meta-analysis, the current results showed an
overall significant positive association between pornography use and attitudes supporting violence against women in
nonexperimental studies. In addition, such attitudes were found to correlate significantly higher with the use of sexually violent
pornography than with the use of nonviolent pornography, although the latter relationship was also found to be significant. The
study resolves what appeared to be a troubling discordance in the literature on pornography and aggressive attitudes by showing
that the conclusions from nonexperimental studies in the area are in fact fully consistent with those of their counterpart
experimental studies. This finding has important implications for the overall literature on pornography and aggression. Aggr.
Behav. 35:1–7, 2009.
r
2009 Wiley-Liss, Inc.
:::::::::::::::::::::::::::::::::::::::::
INTRODUCTION
In a meta-analysis conducted by Allen et al.
[1995b] the investigators failed to find a significant
association between attitudes supporting violence
against women and pornography consumption in
nonexperimental studies. This result is both at odds
with results emerging from experimental studies and
the overall literature in the area including other
meta-analyses by Allen and associates. Here a
consistent significant association between pornogra-
phy and various dependent measures including both
attitudes supporting violence against women and
actual aggressive behavior has been found [Allen
et al., 1995a,b, 2000].
For many researchers, the incongruity between
the results emerging from experimental vs. non-
experimental studies concerning the association
between attitudes supporting violence against
women and pornography has understandably raised
doubts about the ability of generalizing the conclu-
sions emanating from experimental studies to
‘‘real world’’ settings [e.g., Lo and Ran, 2005;
Seto et al., 2001]. In addition, if such doubts
are well founded, they also constitute a major
challenge to models positing that attitudes
supporting violence against women are one of the
interacting pathways mediating and moderating
behavioral effects of pornography, e.g., The Con-
fluence Model of Sexual Aggression [Malamuth
et al., 1995].
However, as elaborated upon below, serious
doubts may be raised about the accuracy of the
conclusions reached by Allen et al. [1995b] in their
meta-analysis of the relationship between attitudes
supporting violence against women and pornogra-
phy in nonexperimental studies. On this basis, we
conduct a new and up-to-date meta-analysis that
corrects for problems and questionable decisions in
the Allen et al. [1995b] meta-analysis. In addition,
we examine the potential relationship between
attitudes supporting violence against women and
content of pornography across included studies as
violent forms of pornography have been reported to
Published online in Wiley InterScience (www.interscience.wiley.
com). DOI: 10.1002/ab.20328
Received 12 March 2008; Revised 25 August 2009; Accepted 29
September 2009

Correspondence to: Gert Martin Hald, Axel Heides Gade 12, 5 t.v.,
2300 Copenhagen S., Denmark. E-mail: gertmartinhald@gmail.com
r
2009 Wiley-Liss, Inc.

be more clearly associated with risk factors pertain-
ing to sexual aggression than nonviolent forms [e.g.
Boeringer, 1994].
Definitions
The widely accepted conceptualization of ‘‘atti-
tudes’’ usually incorporates three components in-
cluding affective responses, cognitive evaluations,
and behavioral predispositions toward an entity
[e.g., Breckler, 1984]. When applying this approach
to defining ‘‘attitudes supporting violence against
women’’ and deciding which studies should be
included within a meta-analysis of this area, we
included various scales assessing (a) affective res-
ponses to acts such as rape, other types of sexual
aggression, and partner violence, (b) evaluative
cognitions, and (c) behavioral predispositions or
attractions toward such aggressive acts [e.g.,
Malamuth, 1981, 1989a,b]. Thus, we follow the lead
of Allen et al. [1995b] although these investigators
used the term ‘‘rape myth acceptance’’ rather than
‘‘attitudes supporting violence against women.’’ We
believe that this latter term better describes the
conceptual territory encompassed by the various
scales included.
The term ‘‘pornography’’ refers to sexually explicit
materials intended to create sexual arousal in the
consumer. Nonviolent pornography is defined as
sexually explicit materials without any overt
coercive content, but which may sometimes imply
acts of submission and/or coercion by the position-
ing of the models, use of props or display of unequal
power relationships. Violent pornography is defined
as sexually explicit materials in which nonconsen-
sual, coercive, and/or violent sexual relations
are explicitly portrayed [see also Senn and Radtke,
1990].
The Basis for Predicting Associations
The basis for predicting associations between
exposure to violent pornography and aggressive
tendencies, including attitudes supporting violence
against women, may be viewed as in keeping
with more general models of the impact of violent
media on aggressive tendencies [e.g., Anderson and
Carnagey, 2004; Huesmann and Kirwil, 2007],
although additional mechanisms may also be at
play when images of sex and aggression are
intermingled [e.g., Anderson and Anderson, 2008;
Donnerstein and Hallam, 1978].
The proposed processes responsible for predicting
an association between nonviolent pornography and
aggressive responses, including attitudes supporting
violence against women, rely on the fact that
nonviolent pornography often portrays women as
highly sexually promiscuous and frequently as being
dominated and ‘‘used’’ by males. These images may
prime and reinforce various sexually aggressive
schemata and ‘‘rape myth’’ attitudes, e.g., that some
women deserve to or enjoy being harassed, mal-
treated sexually, or raped [Berkel et al., 2004;
Lonsway and Fitzgerald, 1995; Milburn et al.,
2000]. The proposed associations may not occur
for most men, but be particularly likely for men who
hold hostile/power schemas associated with women
and sexuality and/or adhere to attitudes that
dichotomize women into ‘‘whores’’ vs. ‘‘madonnas’’
[see also Bargh et al., 1995; Edelman, 2009;
Kingston et al., 2008; McKenzie-Mohr and Zanna,
1990; Vega and Malamuth, 2007; Zurbriggen, 2000].
METHOD
Problems in the Allen et al. [1995b]
Meta-analysis
First, in our opinion, half of the eight studies
included in the meta-analysis of Allen et al. [1995b]
should not have been included due to lack of fit in
concept definitions, sampling procedures, subject
samples, and/or the assessment instruments used.
These four studies include: Burt [1980], Mosher
[1988], and Padgett et al. [1989, two studies]. For
illustration purposes we will discuss only one
example here namely Burt [1980]. However, a more
detailed description of the reasons why the above
studies were excluded may be obtained from the first
author. In the study by Burt [1980], there is a clear
error in what type of media was classified as
‘‘sexually explicit media’’ or ‘‘pornography.’’ The
media assessed by Burt actually consisted of
‘‘exposure to media treatments of sexual assault,’’
defined as ‘‘television, motion picture, dramatic,
and newspaper treatments of rape or sexual assault’’
(p 221). Such media typically document the
horrors of rape, rather than show sexually explicit
images designed to sexually arouse the consumer
(i.e. pornography). Importantly, the same theoreti-
cal models (e.g., social learning theory) that
would predict a positive association between porno-
graphy use and attitudes supporting violence
against women would in fact predict the opposite,
i.e. a negative association, for this type of docu-
mentary media. For this reason we believe that
Burt [1980] should not have been included in the
meta-analysis.
2 Hald et al.
Aggr. Behav.

Second, in the Allen et al. [1995b] study, we found
a mistake in the statistical analyses concerning the
likely presence of a moderating variable.
1
This error
was graciously acknowledged by Dr. M. Allen
(personal communication, November 25, 2005).
Meta-analyses commonly present a statistical test
of heterogeneity in an attempt to establish whether
all studies are evaluating the same effect [Higgins
et al., 2003; Hunter et al., 1982; Leandro, 2005]. A
failed test of heterogeneity as given by a significant
w
2
indicates the likely presence of a moderating
variable. A nonsignificant w
2
indicates the likely
absence of a moderating variable and hence homo-
geneity across included studies. Allen, Emmers
et al. erroneously reported that ‘‘after deleting the
Check [1985, Experiment 2] and Malamuth and
Check [1985, Nonexperimental] studies, the new
average correlation was homogeneous and that
the sample probably did not contain a mode-
rating variable’’ (p 18). However, our reanalysis
showed that the new average correlation in fact
was heterogeneous, indicating the likely presence
of a moderating variable (w2
ð5Þ¼14:23, P5.0142,
I
2
565% using Cochran’s Qand Higgin’s I
2
). This
calls for a more cautious or even different inter-
pretation of the results and following conclusions
of this particular part of the Allen, Emmers et al.
meta-analysis.
The Present Meta-Analysis
Procedure. We used two methods for collect-
ing studies. First, we examined previous meta-
analyses and reviews on pornography for relevant
studies [in particular Allen et al, 1995a,b; Bauserman,
1996; Fisher and Grenier, 1994; Malamuth et al.,
2000; Oddone-Paolucci et al., 2000]. Second, we
conducted a thorough literature search of the
following databases: PsychInfo, PsycArticles,
PubMed, and Sociological Abstracts using erotica

,
porn

, sexual media

, rape

, and violence

as key
words searching the databases from inception to
February 2009. This resulted in a large number
of references. We then reviewed each reference
carefully according to the following four inclusion
criteria:
1. The definition of pornography matched or approxi-
mated our own. That is, ‘‘sexually explicit materials
intended to create sexual arousal in the receiver.’’
2. The study included a measure of attitudes
supporting violence against women.
3. The study included enough statistical informa-
tion on male participants to estimate the associa-
tion between pornography consumption and
attitudes supporting violence against women.
4. The study used nonoffender samples.
The first three criteria match closely those used by
Allen et al. [1995b] in their meta-analysis. However,
Allen, Emmers et al. included in some studies the data
for both female and male participants. As research
has consistently shown gender to be a strong
differentiating variable in this area of research [e.g.,
Bryant, 2009; Hald, 2006; Hald and Malamuth, 2008]
we elected not to do so, with one exception. In the
Emmers-Sommer and Burns [2005] study ten women
(2.4%) was included in the calculation of results. We
thought it unlikely that such a small percentage
would have much overall impact and decided to
include the study. The fourth criterion does not
explicitly replicate Allen, Emmers et al., although
Allen, Emmers et al. also did not include studies using
offender populations. Our rationale for excluding
studies using offender samples pertain to the fact that
various researchers have raised questions about the
veridicality and validity of self-reports of convicted
offenders as compared with nonoffender samples
[e.g., Hanson and Bussiere, 1998; Hare, 1985].
A total of nine studies and 2,309 participants were
included in the final meta-analysis (Table I) [Barak
et al., 1999; Demare
´et al., 1993]. We acknowledge
that the inclusion of only nine studies in the final
meta-analysis may call for a more cautious inter-
pretation of results.
Measures. The following measures of attitudes
supporting violence against women were used in the
studies included in the meta-analysis:
The acceptance of interpersonal violence scale
(AIV—6 items): The AIV assess attitudes condoning
the use of force and violence in relationships. The
internal reliability of the AIV is .59 as measured by
Cronbach’s a[Burt, 1980].
The adversarial sexual beliefs scale (ASB—9
items): The ASB investigates the degree to which
participants perceive male and female relations as
‘‘fundamentally exploitative’’ [Burt, 1980]. The
1
Depending on the particular focus of the study, individual
differences such as attraction to sexual aggression or attitudes
supporting violence against women may be treated as a mediator, a
moderator or an outcome variable. Mediators reflect the generative
mechanisms or processes through which the identified variable
influences the outcome. That is, how an effect came about. In
contrast, a moderator is a third variable that affects the direction
and/or the strength of a relationship between two variables. In
statistical analyses this is revealed as an interaction effect and in
meta-analyses as a failed test of heterogeneity.
3Pornography Use and Attitudes
Aggr. Behav.

internal reliability of the ASB is .80 as measured by
Cronbach’s a[Burt, 1980].
The rape myth acceptance scale (RMA—11 items):
The RMA measures the degree to which participants
believe in stereotypical rape myths. The internal
reliability of the RMAS is .88 as measured by
Cronbach’s alpha [Burt, 1980].
The attitudes toward rape scale (ATR—15, 25, or 55
items): The ATR includes eight factors. High scores
on these factors reflect various aspects contributing
to a general belief in rape myths, e.g., that women
cause rape through their appearance and/or behavior
[Field, 1978; Garcia, 1986]. The reliability of the ATR
ranges between .81 and .93 as measured by Cron-
bach’s a[Daugherty and Dambrot, 1986].
The likelihood of rape scale (LR), the likelihood of
sexual force (LSF), and the likelihood of sexual
harassment (LSH) scale: The LR, LSF, and LSH are
single item scales used to measure the hypothetical
potential of a man to rape or commit similar sexual
aggressive acts given the assurance that he would
face no punishment [Malamuth, 1981]. Scores on
these scales have been shown to have considerable
construct and predictive validity and to correlate
highly with a much more elaborate measure of
attraction to sexual aggression [e.g., Malamuth,
1989a,b; Malamuth and Dean, 1991].
The perception of sexual harassment scale (PSH–9
items): The PSH examines perceptions of sexual
harassment [Biber et al., 2002]. The reliability of the
PSH is .72 as reported by Lam and Chan [2007] and
measured by Cronbach’s a.
The Sexual Harassment Proclivities Scale (SHP–10
items): The SHP assess participants’ proclivity to
engage in sexual harassment [Pryor, 1987]. The
reliability of the PSH (5 items) is .83 as reported by
Lam and Chan [2007] and measured by Cronbach’s a.
All included measures used Likert scales where
higher scores indicate a higher degree of attitudes
supporting violence against women.
RESULTS
Owing to the findings of heterogeneity in the
analyses reported below all analyses were conducted
using both a fixed effect model and a random effect
model and then compared. As the results of all
TABLE I. Studies Included in (1) The Meta-Analysis and (2) The Sensitivity Analyses
Study/Author Year
Attitudes supporting
violence against
women scales used NCorrelation
a
1. Meta-analysis
Barak, Fisher, Belfry, and
Lashambe
1999 LSH, RMA 31 .310
Boeringer 1994 LF, LR 477 .283
Check 1985 AIV, ASB, LF, LR, RMA 434 .111
Demare, Briere, and Lips 1993 AIV, LF, LR, RMA 383 .142
Emmers-Sommer and Burns 2005 AIV, ASB, RMA 419 .090
Garcia 1986 ATR 115 .045
Lam and Chan 2007 PSH, SHP 227 .208
Malamuth and Check 1985 RMA 121 .290
Vega and Malamuth 2007 ASB, AIV, HTW, RMA 102 .312
2A. Sensitivity analysis I:Nonviolent pornography
Boeringer 1994 LF, LR 477 .205
Demare, Briere, and Lips 1993 AIV, LF, LR, RMA 383 .084
Emmers-Sommer and Burns 2005 AIV, ASB, RMA 419 .030
Garcia 1986 ATR 115 .012
Malamuth and Check 1985 RMA 121 .290
Vega and Malamuth 2007 ASB, AIV, HTW, RMA 102 .312
2B. Sensitivity analysis II:Violent pornography
Boeringer 1994 LF, LR 477 .361
Demare, Briere, and Lips 1993 AIV, LF, LR, RMA 383 .171
Emmers-Sommer and Burns 2005 AIV, ASB, RMA 419 .210
Garcia 1986 ATR 115 .070
Note: Attitudes supporting violence against women scales used: AIV 5Acceptance of Interpersonal Violence, ASB 5Adversarial Sexual Beliefs,
ATR 5Attitudes Toward Rape, LF 5Likelihood of Force, LR 5Likelihood of Rape, LSH 5Likelihood of Sexual Harassment, PSH 5The
Perception of Sexual Harassment Scale, RMA 5Rape Myth Acceptance, SHP 5The Sexual Harassment Proclivities Scale.
a
If an overall rwas not provided, an overall average rwas calculated on the basis of the r-values of each relevant scale included as suggested by
Lipsey and Wilson [2001].
4 Hald et al.
Aggr. Behav.

analyses using either model were very similar, only
the result using the fixed effect model is reported here
with the result using the random effect model being
available from the first author [see also Higgins and
Thompson, 2002; Leandro, 2005; Song et al., 2001].
The overall meta-analysis included nine studies and
2,309 participants. The average correlation between
pornography consumption and attitudes supporting
violence against women using a fixed effect model was
significant (r5.18, Po.001; CI 95% (.14; .22)).
However, a failed test of heterogeneity and incon-
sistency across studies was found indicating the likely
presence of a moderating variable (w2
ð8Þ¼18:21,
Po.001, I
2
556%, using Cochran’s Qand Higgin’s I
2
).
Both theory and the experimental research litera-
ture suggest that violent pornography is more likely
to have association with attitudes supporting violence
against women than nonviolent pornography [e.g.,
Allen et al., 1995b]. Consequently, two sensitivity
analyses based on type of pornography were con-
ducted. Only studies providing the necessary differ-
entiation of statistical information were included.
Across six studies and 1,617 participants, the
average correlation between nonviolent pornogra-
phy and attitudes supporting violence against
women using a fixed effect model was found to be
significant (r5.13, Po.001). However, a failed test
of heterogeneity and inconsistency across studies
was found (w2
ð5Þ¼16:42, P5.006, I
2
570%, using
Cochran’s Qand Higgin’s I
2
) indicating the likely
presence of a moderating variable.
Across four studies and 1,394 participants the
average correlation between violent pornography
and attitudes supporting violence against women
using a fixed effect model was found to be significant
(r5.24, Po.001). However, a failed test of hetero-
geneity and inconsistency across studies was found
(w2
ð3Þ¼14:22, P5.003, I
2
579%, using Cochran’s
Qand Higgin’s I
2
) indicating the likely presence of a
moderating variable.
Using Fisher’s Z transformation to compare the
within-group correlations between violent and non-
violent pornography and attitudes supporting
violence against women, it was found that the
correlation between violent pornography and atti-
tudes supporting violence against women (r5.24)
was significantly higher (Po.001) than the correla-
tion between nonviolent pornography and attitudes
supporting violence against women (r5.13).
DISCUSSION
The result of the present meta-analysis shows a
significant overall relationship between pornography
consumption and attitudes supporting violence
against women in nonexperimental studies. This
relationship was found to be significantly stronger
for violent pornography than for nonviolent porno-
graphy, although both types of pornography showed
significant positive associations with attitudes sup-
porting violence against women. The finding of
heterogeneity in the meta-analysis underlines the
importance of targeting moderators in pornography
research [see also Kingston et al., 2009].
The results are in contrast to earlier conclusions
reported by Allen et al. [1995b] both concerning the
existence of an overall significant relationship
between pornography consumption and attitudes
supporting violence against women in nonexperi-
mental studies and the finding of heterogeneity
indicative of moderators in this relationship.
Further, our reanalysis of the meta-analysis as
originally reported by Allen, Emmers et al. showed
that even in their originally reported meta-analysis
heterogeneity indicative of moderators was found
despite their reporting of the contrary.
Two important implications may be drawn from
this study. First, the results correct a glaring
discrepancy in the research literature by showing
that the relationship between men’s pornography
consumption and their attitudes supporting vio-
lence against women in nonexperimental studies
are in fact fully consistent with those previously
found in experimental studies focusing on the same
association.
Second, the results highlight the role of individual
differences as strong moderators of the association
between pornography and attitudes supporting
violence against women. Such moderation has now
also been well documented in this research area
with other dependent measures [e.g., Bryant, 2009;
Kingston et al., 2008, 2009; Malamuth and Huppin,
2005; Vega and Malamuth, 2007]. More specifically,
it has been consistently found that an association
between pornography consumption and aggression
is particularly likely for men who score high on
other risk factors for sexual aggression.
Does a consistent significant, but relatively small
association between pornography consumption and
attitudes supporting violence against women in
nonexperimental studies have practical significance?
We believe it does. As shown by e.g., Rosenthal
[1986] even small significant associations may
translate into considerable social and practical
significance across larger population samples. In
addition, the type of attitudes studied here have
been found to consistently predict ‘‘real world’’
sexually aggressive proclivities and behaviors in
5Pornography Use and Attitudes
Aggr. Behav.

both cross-sectional and longitudinal research [e.g.,
Hall et al., 2006; Malamuth et al., 1995; Meyer,
2000; Voller et al., 2009]. Finally, as has been well
documented in the area of sexual aggression
research virtually all risk factors have only relatively
small associations with the dependent variables of
interest. However, it is the confluence or interactive
combination of these variables that can have strong
predictive utility and thus social and practical
significance [e.g., Malamuth, 1986; Malamuth
et al., 1995, 2000; Vega and Malamuth, 2007].
REFERENCES
Allen M, D’Alessio D, Brezgel K. 1995a. A meta-analysis
summarizing the effects of pornography II: Aggression after
exposure. Hum Commun Res 22:258–283.
Allen M, Emmers T, Gebhardt L, Giery MA. 1995b. Exposure to
pornography and acceptance of rape myth. J Commun 45:5–25.
Allen M, D’Alessio D, Emmers-Sommer TM. 2000. Reactions of
criminal sexual offenders to pornography: A meta-analytic
summary. In: Roloff M (ed.). Communication Yearbook 22.
Thousand Oaks, CA: Sage, pp 139–169.
Anderson CA, Anderson KB. 2008. Men who target women:
Specificity of target, generality of aggressive behaviour. Aggr
Behav 34:605–622.
Anderson CA, Carnagey NL. 2004. Violent evil and the general
aggression model. In: Miller A (ed.). The Social Psychology of
Good and Evil. New York: Guilford Publications, pp 168–192.
*Barak A, Fisher WA, Belfry S, Lashambe DR. 1999. Sex, guys, and
cyberspace: Effects of Internet pornography and individual
differences on men’s attitudes toward women. J Psychol Hum
Sex 11:63–91.
Bargh JA, Raymond P, Pryor JB, Strack F. 1995. Attractiveness of
the underling: An automatic power sex association and its
consequences for sexual harassment and aggression. J Pers Soc
Psychol 68:768–781.
Bauserman R. 1996. Sexual aggression and pornography: A review
of correlational research. Basic Appl Soc Psychol 18:405–427.
Berkel LA, Vandiver BJ, Bahner AD. 2004. Gender role attitudes,
religion, and spirituality as predictors of domestic violence
attitudes in white college students. J Coll Stud Dev 45:119–131.
Biber JK, Doverspike D, Baznik D, Cober A, Ritter B. 2002. Sexual
harassment in online communications: Effects of gender and
disclosure medium. CyberPsychol Behav 5:33–42.
*Boeringer S. 1994. Pornography and sexual aggression: Associa-
tions of violent and nonviolent depictions with rape and rape
proclivity. Deviant Behav: Interdisciplinary J 15:289–304.
Breckler SJ. 1984. Empirical validation of affect, behavior, and
cognition as distinct components of attitude. J Pers Soc Psychol
47:1191–1205.
Bryant P. 2009. Predicting Internet pornography use and arousal:
The role of individual difference variables. J Sex Res 46:344–357.
Burt M. 1980. Cultural myths and supports for rape. J Pers Soc
Psychol 38:217–230.
*Check J. 1985. The effects of violent and nonviolent pornography.
Ottawa, Canada: Department of Justice (Department of Supply
and Services Contract No. 05SV 19200-3-0899).
Daugherty CG, Dambrot FH. 1986. Reliability of the attitudes
toward women scale. Educ Psychol Meas 46:449–453.
*Demare
´D, Lips HM, Briere J. 1993. Sexually violent pornography,
anti-women attitudes and sexual aggression: A structural
equation model. J Res Pers 27:285–300.
Donnerstein E, Hallam J. 1978. Facilitating effects of erotica on
aggression against women. J Pers Soc Psychol 36:1270–1277.
Edelman B. 2009. Red light states: Who buys online adult
entertainment. J Econ Perspect 23:209–220.
*Emmers-Sommer TM, Burns RJ. 2005. The relationship between
exposure to Internet pornography and sexual attitudes toward
women. Journal of Online Behavior, 1. Retrieved January 22,
from the World Wide Web. http://www.behavior.net/JOB/v1n4/
emmers-sommer.html.
Field HS. 1978. Attitudes toward rape: A comparative analysis of
police, crisis counselors, and citizens. J Pers Soc Psychol 36:156–179.
Fisher WA, Grenier G. 1994. Violent pornography, antiwoman
thoughts and antiwoman acts: in search of reliable effects. J Sex
Res 31:23–38.
*Garcia L. 1986. Exposure to pornography and attitudes about
women and rape: A correlation study. J Sex Res 22:378–385.
Hald GM. 2006. Gender differences in pornography consumption
among young heterosexual Danish adults. Arch Sex Behav 35:
577–585.
Hald GM, Malamuth NM. 2008. Self-perceived effects of porno-
graphy in a random sample of the Danish population. Arch Sex
Behav 37:614–625.
Hall GCN, DeGarmo DS, Eap S, Teten AL, Sue S. 2006. Initiation,
resistance, and persistence of men’s sexual coercion. J Consult
Clin Psychol 74:732–742.
Hanson RK, Bussiere MT. 1998. Predicting relapse: A meta-analysis
of sexual offender recidivism studies. J Consult Clin Psychol
66:348–362.
Hare RD. 1985. A checklist for the assessment of psychopathy in
criminal populations. In: Ben-Aron MH, Hucker SJ, Webster CD
(eds.). Clinical Criminology: The Assessment and Treatment of
Criminal Behavior. Toronto, Canada: M & M Graphics,
pp 157–168.
Higgins PT, Thompson SG. 2002. Quantifying hetereogeneity in a
meta-analysis. Stat Med 21:1539–1558.
Higgins PT, Thompson SG, Deeks JJ, Douglas GA. 2003. Measuring
inconsistency in meta-analyses. Br Med J 327:557–560.
Huesmann LR, Kirwil L. 2007. Why observing violence increases the
risk of violent behavior in the observer. In: Flannery DJ,
Vazsonyi AT, Waldman ID (eds.). The Cambridge Handbook
of Violent Behavior and Aggression. Cambridge, England:
Cambridge University Press, pp 545–570.
Hunter JE, Schmidt FL, Jackson GB. 1982. Meta-analysis:
Cumulating Research Findings Across Studies. New Delhi, India:
Sage.
Kingston DA, Fedoroff P, Firestone P, Curry S, Bradford JM. 2008.
Pornography use and sexual aggression: The impact of frequency
and type of pornography use on recidivism among sexual
offenders. Aggr Behav 34:341–351.
Kingston DA, Malamuth NM, Fedoroff P, Marshall WL. 2009. The
importance of individual differences in pornography use:
Theoretical perspectives and implications for treating sexual
offenders. J Sex Res 46:216–232.
*Lam CB, Chan DKS. 2007. The use of cyberpornography by young
men in Hong Kong: Some psychosocial correlates. Arch Sex
Behav 36:588–598.
Leandro G. 2005. Meta-analysis in Medical Research: A Handbook
for the Understanding and Practice of Meta-analysis. Oxford,
England: Blackwell Publishing, Ltd.
Lipsey MW, Wilson DB. 2001. Practical Meta-analysis.Thousand
Oaks, CA: Sage Publications.
6 Hald et al.
Aggr. Behav.

Lo V, Ran W. 2005. Exposure to Internet pornography and
Taiwanese adolescents’ sexist attitudes and behaviors and.
J Broadcasting Electronic Media 49:221–237.
Lonsway KA, Fitzgerald LF. 1995. Attitudinal antecedents of rape
myth acceptance: A theoretical and empirical reexamination.
J Pers Soc Psychol 68:704–711.
Malamuth NM. 1981. Rape proclivity among males. Journal of
Social Issues 37:138–157.
Malamuth NM. 1986. Predictors of naturalistic sexual aggression.
J Pers Soc Psychol 50:953–962.
Malamuth NM. 1989a. The attraction to sexual aggression scale:
Part I. J Sex Res 6:26–49.
Malamuth NM. 1989b. The attraction to sexual aggression scale:
Part II. J Sex Res 26:324–354.
*Malamuth NM, Check J. 1985. The effects of aggressive
pornography on beliefs in rape myths: Individual differences.
J Res Pers 19:299–320.
Malamuth NM, Dean K. 1991. Attraction to sexual aggression. In:
Parrot A (ed.). Acquaintance Rape: The Hidden Crime. NY:
Wiley, pp 229–247.
Malamuth NM, Huppin M. 2005. Pornography and teenagers:
The importance of individual differences. Adolesc Med 16:
315–326.
Malamuth NM, Linz D, Heavey CL, Barnes G, Acker M. 1995.
Using the confluence model of sexual aggression to predict men’s
conflict with women: A 10-year follow-up study. J Pers Soc
Psychol 69:353–369.
Malamuth NM, Addison T, Koss M. 2000. Pornography and sexual
aggression: Are there reliable effects and can we understand
them? Ann Rev Sex Res 11:26–91.
McKenzie-Mohr D, Zanna M. 1990. Treating women as sexual
objects: Look at the (Gender Schematic) male who viewed
pornography. Pers Soc Psychol Bull 16:296–308.
Meyer SL. 2000. Prevalence and characteristics of sexual aggression
in court-mandated batterers.Unpublished doctoral dissertation,
State University of New York at Stony Brook, NY.
Milburn MA, Mather R, Conrad SD. 2000. The effects of viewing
R-rated movie scenes that objectify women on perceptions of
date rape. Sex Roles 43:645–664.
Mosher DL. 1988. Pornography defined: Sexual involvement theory,
narrative context, and ‘‘goodness-of-fit.’’ J Psychol Hum Sex 1:
67–85.
Oddone-Paolucci E, Genius M, Violato C. 2000. A meta-analysis
of the published research on the effects of pornography. In:
Violato C, Oddone-Paolucci E (eds.). The Changing Family and
Child Development. Aldershot, England: Ashgate Publishing,
Ltd, pp 48–59.
Padgett V, Brislin-Slutz J, Neal J. 1989. Pornography, erotica, and
attitudes toward women: The effects of repeated exposure. J Sex
Res 26:479–491.
Pryor JB. 1987. Sexual harassment proclivities in men. Sex Roles
17:269–290.
Rosenthal R. 1986. Media violence, antisocial behavior, and the
social consequences of small effects. J Soc Issues 3:141–154.
Senn CY, Radtke HL. 1990. Women’s evaluations of and affective
reactions to mainstream violent pornography, nonviolent erotica,
and erotica. Violence Vict 5:143–155.
Seto MC, Maric A, Barbaree HE. 2001. The role of pornography in
the etiology of sexual aggression. Aggr Violent Behav 6:35–53.
Song F, Sheldon TA, Sutton AJ, Abrams KR, Jones DR. 2001.
Methods for exploring hetereogeneity in meta-analysis. Eval
Health Prof 24:126–151.
*Vega V, Malamuth NM. 2007. Predicting sexual aggression: The
role of pornography in the context of general and specific risk
factors. Aggr Behav 33:104–107.
Voller EK, Long PJ, Aosved AC. 2009. Attraction to sexual violence
towards women, sexual abuse of children, and non-sexual
criminal behavior: Testing the specialist vs. generalist models in
male college students. Arch Sex Behav 38:235–243.
Zurbriggen EL. 2000. Social motives and cognitive power–sex
associations: Predictors of aggressive sexual behavior. J Pers
Soc Psychol 78:559–581.
7Pornography Use and Attitudes
Aggr. Behav.