University of Alabama at Birmingham

UAB Digital Commons

All ETDs from UAB UAB Theses & Dissertations

2008

The Relationship Between Pornography And Rape Rates: Does

The Distribution Of Pornography Increase The Incidence Of Rape?
Jamie Lee Blalock
University of Alabama at Birmingham

Follow this and additional works at: https://digitalcommons.library.uab.edu/etd-collection

Part of the Arts and Humanities Commons

Recommended Citation
Blalock, Jamie Lee, "The Relationship Between Pornography And Rape Rates: Does The Distribution Of
Pornography Increase The Incidence Of Rape?" (2008). All ETDs from UAB. 3557.
https://digitalcommons.library.uab.edu/etd-collection/3557

This content has been accepted for inclusion by an authorized administrator of the UAB Digital Commons, and is
provided as a free open access item. All inquiries regarding this item or the UAB Digital Commons should be
directed to the UAB Libraries Office of Scholarly Communication.
 THE RELATIONSHIP BETWEEN PORNOGRAPHY AND RAPE RATES:
DOES THE DISTRIBUTION OF PORNOGRAPHY INCREASE THE INCIDENCE OF
RAPE?

by

JAMIE LEE BLALOCK

HEITH COPES, COMMITTEE CHAIR

WENDY GUNTHER-CANADA
JOHN J. SLOAN, III

A THESIS

Submitted to the graduate faculty of The University of Alabama at Birmingham,

in partial fulfillment of the requirements for the degree of
Master of Science

BIRMINGHAM, ALABAMA

2008
 THE RELATIONSHIP BETWEEN PORNOGRAPHY AND RAPE RATES:
DOES THE DISTRIBUTION OF PORNOGRAPHY INCREASE THE INCIDENCE OF
RAPE?

JAMIE LEE BLALOCK

JUSTICE SCIENCES

ABSTRACT

This study was conducted to test consumption of pornographic material, specifi-

cally Playboy, increases the likelihood of rape at the county level. To do this, a cross-

sectional design was used to examine this relationship in 3,133 counties in the United

States. Overall, the findings suggest that when households headed by females with child-

ren, persons 15 and older who are divorced, percent of the population with a high school

education or higher, percent of the population who are unemployed, percent of the popu-

lation living in urban areas, percent of the population ages 18 to 24, percent of the popu-

lation that are black, and ratio of males per 100 females are present in counties, rape rates

tend to be higher. However, these findings do not suggest pornography consumption has

a significant effect on rape rates.

ii
 ACKNOWLEDGMENTS

The successful completion of my graduate education involved more than just my

desire to learn. I was helped by the guidance and assistance of a number of people for

whom I will forever be indebted to. I extend my gratitude to Dr. Heith Copes, for his

support in undertaking a quantitative research project. I appreciated his reliability and

patience, which is unprecedented, and knowing that his door was always open. I extend

my gratitude to Dr. John J. Sloan III for his immeasurable support and encouragement

throughout my graduate school career. I thank Dr. Wendy Gunther-Canada for serving

on my thesis committee and her invaluable insights into feminism. Others who have

helped me include Dr. Tomislav Kovandzic and Lynne Vieraitis. They provided help

with data collection and analysis. I am also appreciative of the help given to me from

Andrea Brown. She is the glue that holds the department together.

Special thanks to Darrell Reid for his words of encouragement and friendship

throughout this endeavor. I also thank Michele Way for being a support system and

someone who helped me learn to write properly. Last but not least, I could not have done

this without the love, support, and sacrifice of my parents, George and Pat, without them

none of this would have been accomplishable.

iii
 TABLE OF CONTENTS

Page

ABSTRACT ........................................................................................................................ ii

ACKNOWLEDGMENTS ................................................................................................. iii

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

CHAPTER

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

2. REVIEW OF THE LITERATURE ..........................................................................5

The Early Studies (1970s).................................................................................5

Experimental Research .....................................................................................7
Survey Research................................................................................................8
Other Works ....................................................................................................10

3. METHOD ...............................................................................................................12

Research Design..............................................................................................12
The Dependent Variable: Rape Rates .............................................................12
Primary Independent Variable: Pornography .................................................14
Control Variables ............................................................................................15
Data Analytic Strategy ....................................................................................17

4. ANALYSIS .............................................................................................................17

Multicollinearity .............................................................................................17
Playboy Rate and Rape Rate ...........................................................................18
Other Predictors of Rape Rate ........................................................................19

5. DISCUSSION AND CONCLUSION ....................................................................26

6. LIST OF REFERENCES ........................................................................................33

iv
 LIST OF TABLES

Table Page

1. Description of Variables for Regression Analysis .................................................21

2. Descriptive Statistics for Variables Used in the Analysis .....................................22

3. Correlations for Variables Used in the Analysis.............................................. 23-24

4. Regression Coefficients Representing the Effects of Playboy Rates on Rape,

Controlling Other Variables and Tolerance and Variance Inflation Factors ..........25

v
 1

“One can consult all the experts he chooses, can write reports, make studies, etc., but

the fact that obscenity corrupts lies within the common sense, the reason, and the logic of

every man.” ~ Charles Keating

INTRODUCTION

In 2005, over 176,000 rapes or sexual assaults of women occurred in the United

States. This translates into a victimization rate of 1.4 victimizations per every 1,000

women ages 12 or older in this country (Bureau of Justice Statistics, 2007). While rape

rates have declined in recent years, the question remains why is such violence perpetrated

against women? One possible answer is a link between viewing pornography and in-

creased sexual aggression among males. While this debate began in the late 1960s, pub-

lic concern over this link continues, as does concern over the psychological and physical

damage suffered by women as a result of such violence (see: Seto, Maric, & Barbaree,

2001). As a result, social scientists continue to examine and debate the possible link be-

tween exposure to pornography among males’ proclivities to rape (Scott & Culvelier,

1987b).

Whether exposure to pornography increases the likelihood that males will engage

in sexually aggressive acts remains controversial. The controversy stems from the fact

that some people believe that exposure to pornography promotes attitudes, practices, and
 2

behaviors detrimental to women, including sexual violence. By depicting women in sub-

ordinate roles, pornography, especially in its most violent forms, is thought to affect male

attitudes toward women in ways that promote a sexist ideology, perpetuates the tradition-

al “double standard,” 1 and sexually objectifies women (Brownmiller, 1975; Garcia,

1986; Johnson & Goodchilds, 1973).

The academic debate over possible links between males’ viewing pornographic

material and subsequent sexually assaultive behavior mounted in 1968, when President

Johnson appointed the President’s Commission on Obscenity and Pornography to ex-

amine the long-term effects of pornography on sexually aggressive behavior (see: Byrne

& Kelley, 1984). After two years of reviewing published research on the topic, the

Commission released its findings in 1970 and concluded there was no evidence to link

pornography with heightened sexually aggressive behavior in males (President's Com-

mission on Obscenity & Pornography, 1970). Once released, the findings were widely

criticized by feminists, specifically those who opposed pornography, and were rejected

by Congress. Some 40 years later, the Commission’s findings continue to be challenged,

and the debate over pornography and its effects on behavior continues (see: Demare',

Briere, & Lips, 1988; Kutchinsky, 1991; Malamuth & Ceniti, 1986; Nemes, 1993; Scott

& Cuvelier, 1987b; Seto, Maric, & Barbaree, 2001).

One challenge to the Commission’s findings was the Reagan administration’s At-

torney General’s Commission on Pornography (1986), which concluded that there was a

causal relationship between males viewing sexually aggressive material and behaving in

1The double standard exists when society approves of males but disapproves of females
engaging in sexual activity outside a marital relationship.
 3

a sexually aggressively manner towards women. The Attorney General’s report also

concluded that pornography had become substantially more violent since the 1970s,

which further contributed to the link between pornography and sexual aggressive beha-

vior by men against women. Importantly, this conclusion was based entirely on anecdotal

evidence and not careful review of published scientific research (Scott & Cuvelier,

1987b).

Anti-pornography feminists also challenged the Commission’s findings regarding

the effects of pornography. This faction of feminist theorists opposes pornography on

three assertions. First, is that pornography is morally wrong for two reasons: 1) men use

pornography (among other things) to subjugate, degrade and dehumanize woman, and

maintain their own position of power and 2) pornography sexually objectifies women,

which entails fragmentation of a woman’s body and eroticization of her anatomy (Baron

& Strauss, 1984; Brownmiller, 1975; Dworkin & Mackinnon, 1988; McElroy, 1995;

Russell, 2000).

Second, anti-pornography feminists argue that pornography leads directly to vi-

olence against women. That is, they believe that there is a cause-and-effect relationship

between men viewing pornography and their committing rape (McElroy, 1995). The de-

piction of physical assaults against women, often occurring in pornographic movies or

photographs, is thought to serve as a behavioral model for offenders. Therefore, as por-

nography depicting sexually aggressive acts becomes more prevalent and images of

women being coerced, brutalized, and raped become more acceptable in society, rape and

sexual assault levels should increase (Baron & Straus, 1987; Dworkin, 1981; Dworkin &

Mackinnon, 1988).
 4

Finally, anti-pornography feminists argue that pornography, in and of itself, is vi-

olence against women (Dworkin & Mackinnon, 1988, p. 36; McElroy, 1995, p. 97).

These feminists contend that women participating in pornography are either physically or

psychologically coerced (either through patriarchy or through capitalism) into doing so,

and therefore are incapable of giving informed or “real” consent. They believe that

women are “so brainwashed by white male culture that they cannot render consent”

(McElroy, 1995, p. 105). They argue that since women make less money than men and

have fewer opportunities, they are forced into unsavory professions, such as pornography,

to make a living. According to Catherine Mackinnon, “the first victims of pornography

are the ones in it…” (cite in de Grazia, 1992, p. 587).

An important issue in this debate is to determine the extent that research has

found a link between viewing pornography and sexual assault. If evidence for such a link

exists, then arguments by feminists and others concerning the need to suppress ready

access to pornography by men is given greater credence. If, on the other hand, no such

evidence exists, then arguments about pornography and its ills become more philosophi-

cal and move outside the realm of social science. As shown below, despite over four

decades of research on the subject, researchers are still unclear as to the exact nature of

the relationship between viewing pornography and sexual assault against women, despite

utilizing different methodologies including experimental and quasi-experimental designs.

The focus of the present study is to determine if the relationship between pornography

and sexual violence truly exists. After reviewing the literature on the topic, the specifics

of the study design will be discussed. Based on the results of the analysis, implications

for theory and policy will be discussed.

 5

REVIEW OF THE LITERATURE

The relationship between pornography and sexually aggressive acts has been stu-

died using a variety of methods. Early studies using non-experimental designs generally

did not find a relationship between the availability of pornography and sexually aggres-

sive behavior. Similarly, later studies using experimental designs and surveys found a

minimal relationship between pornography and sexually aggressive behavior. Other

works, however, found a positive relationship between pornography and rape rates. Be-

low, is a review of the relevant research on the link between pornography and sexual vi-

olence, distinguishing this work based on the method(s) used by the researchers.

The Early Studies (1970s)

After pornography became readily available in Denmark between1965 and 1967,

Kutchinsky (1971) examined the official reporting of sex crimes, specifically rape, exhi-

bitionism, peeping, coitus with minors, verbal indecency, and other indecency in Copen-

hagen from the 1950s to 1970. Kutchinsky found that the increased availability of por-

nography actually coincided with a decrease in the total number of sexually-related

crimes reported to the Copenhagen police. More specifically, during the period studied,

the number of reported sex offenses decreased while rape rates remained steady. He

nonetheless concluded that pornography provided a cathartic effect such that “some earli-

er offenders may have stopped or at least reduced their criminal activity, while potential

new offenders may never engage in committing sexual offenses because they get suffi-

cient sexual satisfaction through the use of pornography” (Kutchinsky, 1971, p. 288). In

a similar study, Ben-Veniste (1971) examined the same sex crimes reported to police in
 6

Copenhagen during the period 1958 – 1969 and reported similar results, including de-

creases in the number of reported sex offenses while little change occurred in the reported

instances of rape.

Kutchinsky’s (1971) and Ben-Veniste’s (1971) studies have been criticized on a

number of grounds. Baron and Straus (1984, p. 188), for example, criticized Kut-

chinsky’s work stating, “a close[r] look at his data shows that, although the total number

of sexual offenses decreased, the number of rapes reported either stabilized or increased.

The change in the composite number of sex offenses hides increases in the number of

rapes.” Similarly, Court (1984) found that all sexual crimes such as peeping, exhibition-

ism, and indecency had been reduced, but rape rates had actually increased. Bachy

(1976), in his study of sex crimes reported from 1965 to 1974 in Copenhagen acknowl-

edges the discrepancies in Kutchinsky’s results, but argues the lifting of censorship on

pornography increased rapes from 1970 to 1972 and then declined to levels similar to

those before the increase.

Kupperstein and Wilson (1971) replicated Kutchinsky’s (1971) study in the Unit-

ed States using the period 1960 to 1969. They investigated a possible link between sex

crime rates (using data found in the Uniform Crime Reports (UCR)) and the availability

of solicited and unsolicited pornography sent to mailboxes. Based on the data and evi-

dence compiled through their field research, Kupperstein and Wilson (1971) concluded

there was no support for the link between pornography and sexually aggressive acts.

They also examined the rates of nonsexual offenses in the study. Kupperstein and Wil-

son (1971) believed that although rape rates, as well as rates for all index crimes, had in-

creased, the. increase in sexually related acts may be attributed to pornography whereas
 7

increases in non-sexually related acts may have been the result of different factors (Bau-

serman, 1996).

Experimental Research

Initial research examining trends in official crime data was inconclusive as to a

definitive link between pornography and sexually aggressive acts. Researchers, there-

fore, turned towards experimental approaches to study the presumed link. Experiments

conducted in laboratory settings allowed the researchers to examine the effects of porno-

graphic stimuli on sexual aggression while controlling for extraneous factors (see: Allen,

D’Alessio, & Brezgel, 1995a; Allen, Emmers, Gebhardt, & Giery, 1995b; Donnerstein,

1984; Malamuth & Ceniti, 1986). Donnerstein (1984), for example, conducted an expe-

riment on males and found increased aggression in the male participants who were ex-

posed to aggressive pornography compared to those shown non-aggressive pornography.

Allen, Emmers, Gebhardt, and Giery (1995b) found that studies that questioned partici-

pants about their previous exposure to sexually aggressive material and their endorsement

of rape myths 2 showed no relationship, while studies that actually exposed participants

to sexually aggressive materials and then asked about their endorsement of rape myths

found a positive link. Allen et al. (1995b), in a meta-analysis of prior research, found ex-

posure to pornography increased the acceptance of rape myths.

Another study utilizing meta-analysis conducted by Allen, D’Alessio, and Brezgel

(1995a) found that exposure to pornography was associated with minimal increases in

2 Rape myths are the belief that the perpetrator has little or no responsibility for their ac-
tion; therefore the responsibility is placed on victims (Allen et al. (1995b).
 8

sexual aggression. Of the pornography that participants were exposed to, photographs

depicting nudity alone decreased aggressive behavior, while both violent and non-violent

pornography increased aggressive behavior. Malamuth and Ceniti (1986) conducted a

laboratory experiment involving male college undergraduates that looked at the long-term

effects of repeated exposure to violent and non-violent pornography and male sexual ag-

gression towards women. Malamuth and Ceniti (1986) found repeated exposure to vio-

lent and non-violent pornography revealed no significant effect on laboratory aggression

towards women.

Survey Research

Surveys have been another tool used by researchers to investigate the possible

link between pornography and sexually aggressive acts. In general, survey research has

found an inverse relationship between pornography and sexual aggression (e.g., Bergen

& Bogle, 2000; Malamuth & Donnerstein, 1984; Russell, 1993). For example, interviews

conducted with sex offenders found that while pornography provided a “safety valve” for

their antisocial impulses, it also played a role in provoking their sexual offenses (Carter,

Prentky, Vanderveer, & Boucher (1987); Langevin, Lang, Wright, Handy, Frenzel, &

Black (1988); Walker, 1971). Carter et al. (1987), in their study of perpetrators of sex-

ually aggressive acts, found that rapists utilized pornography to relieve impulses to com-

mit sexually aggressive acts. Langevin et al. (1988) reported similar findings in their

study of sexually aggressive perpetrators. Cook, Fosen, and Pacht (1971) conducted a

self-report survey and found that sex offenders reported less exposure to pornography and

sexually aggressive material than non-sex offenders. Walker (1971) found similar results
 9

when he compared groups of imprisoned aggressive sex offenders, non-sex offenders,

and non-offender controls from colleges and men’s clubs in Texas. The non-offender

controls who participated in the study reported more frequent exposure to pornography

and exposure at an earlier age than sex offenders. Other aggressive offenders reported

the use of pornography to relieve the impulses of committing sexually aggressive beha-

vior (Bauserman, 1996).

Garcia (1986) and Demare', Briere, and Lips (1988) studied the relationships be-

tween self-reported use of pornography by undergraduate college students and attitudes

towards women. Garcia (1986) found the consumption of non-violent pornography was

not correlated with anti-women attitudes whereas violent pornography had a weak asso-

ciation with these attitudes and “greater ‘pro-rape’ beliefs” (Demare' et al., 1988, p. 142).

Demare’ et al. (1988) also studied undergraduate college students and concluded there

was no correlation between non-violent pornography and self-reported likelihood of sex-

ually aggressive behavior. Moreover, Demare’ et al. (1988) found that exposure to vio-

lent pornography was positively correlated with anti-woman attitudes and the likelihood

of sexual aggression. Boeringer (1994) surveyed enrolled male college undergraduates

regarding their exposure to pornography and their proclivity for sexually aggressive be-

havior, specifically rape. Of those respondents, more than 48% admitted to the likelihood

of using force to obtain sex if they knew that they were not going to get caught. Howev-

er, when the word “rape” was introduced into the survey the majority of the respondents

reported no likelihood of committing rape. Still, more than 15% of all the respondents

reported some likelihood of rape (Boeringer, 1994). Boeringer (1994) found no signifi-

cant correlation between soft-core pornography and the likelihood of rape, but he did find
 10

a strong correlation between hard-core violent pornography exposure and a self-reported

likelihood of using force to obtain sex.

Other Works

Researchers have also utilized aggregate-level analyses in which they examined

adult sex magazine circulation rates to determine the relationship between pornography

and sexual aggressive behavior. These analyses found that circulation rates of adult sex

magazines, such as Playboy, Penthouse, and Hustler, are positively correlated with rape

rates (Baron & Straus, 1984; Scott & Schwalm, 1988; Scott & Culvelier, 1987a; 1987b).

Research conducted using the Audit Bureau of Circulation’s 3 distribution records, found

that of all the adult sex magazines for which distribution records were kept, Playboy

magazine had the highest correlation by state between circulation rates and rape rates

(Scott & Culvelier, 1987a; 1987b; Scott & Schwalm, 1988).

Baron and Straus (1984) and Scott and Schwalm (1988) utilized adult sex maga-

zines to compare adult magazine circulation rates and rape rates for each state. Baron and

Straus (1984) and Scott and Schwalm (1988) found states with the highest rape rates also

have the highest circulation rates for adult sex magazines. Gentry (1991), however, in a

later replication of these studies used standard metropolitan statistical areas (SMSAs) as

the unit of analysis and concluded there was no relationship between the circulation of

adult sex magazines and SMSAs rape rates.

3 The Audit Bureau of Circulation (ABC) maintains an electronic database of audited-

circulation information and an array of verified readership, subscriber demographics, and
online activity data. ABC audits daily newspapers, consumer magazines, and business
and farm publications- consisting of paid, non- paid direct request and free circulation.
 11

Jaffee and Straus (1987) also looked at the circulation of adult sex magazines and

its relationship to sexually liberal 4 attitudes to reported rape. They concluded that sex-

ually liberal attitudes and adult sex magazine circulation were both positively correlated

with reported rape rate. Jaffee and Straus (1987) argue that rape is a function of two sets

of forces, social disorganization and “hypermasculinity 5 .” They determined “hypermas-

culinity” leads to both a higher rape rate and a higher adult sex magazine readership. Jaf-

fee and Straus (1987) conclude that the correlation between adult sex magazine circula-

tion and rape rates is due to the respondent’s association with hypermasculinity, rather

than a causal relationship (Bauserman, 1996).

Building on previous research and what was found to be significant; the present

study reexamines the relationship between pornography and sexually aggressive acts.

Specifically, it examines the relationship between Playboy subscription rates and reported

incidents of rape at the county level. Playboy subscription rates will be used to measure

the distribution rate of pornography in 3,133 counties in the United States. Subscription

rates, along with control variables, will be analyzed using multiple regression analysis to

determine if they have a significant effect on reported incidents of rape.

4 Sexual liberalism is the openness or the permissiveness that exists in a state as measured
by attitudes towards sex-related issues. States that are more sexually liberal rather than
conservative, women are more likely to report rape.
5 Jaffee and Straus synthesized the term “hypermasculinity” from Parsons (1947) and To-

by (1966) term of compulsive masculinity. They defined compulsive masculinity as (a)

adolescent boys fascinated with weight-lifting and body-building; (b) males extensive use
of profanity; (c) males endless heterosexual conquests to reassure themselves of their
own virility; and (d) senseless violence. However, Mosher and Sirkin (1984) conceptua-
lized hyper-masculinity as (a) calloused sex attitudes towards women; (b) a conception of
violence as manly, and (c) a view of danger as exciting. These components reflect a
man’s hyper-masculinity to appear powerful and dominant in their interactions with other
men, women, and their environment.
 12

METHOD

Research Design

The present study examines the relationship between pornography and rape using

county-level, cross-sectional data for the year 2000 for which complete data on all study

variables were available (n=3,143). The county level was chosen because it represents

the smallest and most homogenous unit of analysis for which rape data for a large nation-

al sample of ecological units was available. Further, county-level data exhibit tremend-

ous between-unit variation in both pornography-consumption and rape rates, which can

be masked when larger aggregates such as states or metropolitan areas are used. Because

the purpose of this research is to explain between-unit variation, counties were chosen as

the unit of analysis. The use of counties also lowers the risk of aggregation bias 6 by mi-

nimizing the likelihood of within-unit variation in both pornography and rape rates.

The Dependent Variable: Rape Rates

The dependent variable was the Uniform Crime Report (UCR) reported average

rape 7 rate per 100,000 resident population in each county for the years 1999 to 2001

(Federal Bureau of Investigation, 1999-2001). To obtain the female rape rate (RAPE-

RATE) the number of rapes known to the police from each county was divided by the

female population of the county in 2000 and multiplied by 100,000 to obtain the female

rape rate (RAPERATE). The use of female population as the denominator results in a

6 Aggregation bias means that findings at the aggregate level may be completely different
than findings at the individual level. Using to large an aggregate, you might miss impor-
tant causal processes that are taking place at the individual level.
 13

more realistic measure because UCR figures for rape pertain solely to female victims

(Bailey, 1999).

Rape data for each county 8 was obtained from the National Archive of Criminal

Justice Data (NAJCD), 9 which, unlike the FBI, calculates county level data. The NAJCD

uses an algorithm to calculate incomplete reporting by individual law enforcement juris-

dictions. Within each county, data from agencies reporting 3 to 11 months of information

were weighted to yield 12-month equivalents. Agencies reporting less than 3 months of

data were replaced with data estimated by rates calculated from agencies reporting 12

months of data located in the agency's geographic stratum within its state. Data from

agencies reporting only statewide figures were allocated to the counties in the state in

proportion to each county's share of the state population (U.S. Department of Justice,

2000). Following convention for cross-sectional studies on crime a three-year average

was used to avoid random fluctuations in rape rates between years. This study used the

UCR data because they were consistent, rather than because they were especially accu-

7 The UCR definition of rape is “the carnal knowledge of a female forcibly and against
her will.”
8 The following counties were dropped from the study sample because Playboy subscrip-

tion data were not available: Lake and Peninsula Borough, Alaska, Skagway-Hoonah-
Angoon Census Area, Alaska, Yakutat City and Borough, Alaska, Loving County, Texas,
Yellow State National Park, Montana, Manassas Park City, Virginia, Poquoson, Virginia,
and South Boston, Virginia. Illinois counties were also dropped from the study sample
because rape figures furnished by the Illinois Department of State Police were not in ac-
cordance with national UCR guidelines (U.S. Federal Bureau of Investigation, 2001).
Therefore, 3,133 counties were utilized to conduct the analysis.
9 The NAJCD releases county-level data from the UCR data provided by the FBI. UCR

program staff at the Federal Bureau of Investigation (FBI) was consulted in developing
the new adjustment procedures. However, these UCR county-level files are not official
FBI UCR releases and are being provided for research purposes only. Users with ques-
tions regarding these UCR county-level data files can contact the National Archive of
Criminal Justice Data at ICPSR (U.S. Department of Justice, 2000).
 14

rate. As is well-known, the UCR underreports the actual number of rapes that occur each

year in the United States (Baron & Straus, 1987; Gove, Hughes, & Geerken, 1985; Hin-

delang, 1974; Kimmel & Linders, 1996). Although underreporting is a limitation of the

UCR data, prior research indicates it represents a valid indicator of the most serious

forms of rape occurring each year in the United States (Baron & Straus, 1987; Gove,

Hughes, & Geerken, 1985; Hindelang, 1974).

Primary Independent Variable: Pornography

The principle independent variable of interest was pornography, measured as sub-

scription rates for the adult sex magazine Playboy. Although previous research has uti-

lized a variety of adult sex magazines to measure exposure to pornography (Baron &

Straus, 1984; Baron & Straus, 1987; Scott & Schwalm, 1988), the only pornographic

magazine for which county-level data were available was Playboy. Playboy is the oldest

continually published adult sex magazine, which has the highest total circulation of all

sexual magazines currently published (Scott & Cuvelier, 1987a; 1987b) and has also been

found to have the highest correlation to rape rates among pornographic magazines (Scott

& Schwalm, 1988). The total number of Playboy subscribers in each county in 2002 was

obtained from the Audit Bureau of Circulations. The total number of Playboy subscribers

was then divided by county population and multiplied by 100,000 to obtain a Playboy

subscription rate (PLAYBOY).

 15

Control Variables

Various control variables are introduced into the statistical analyses presented be-

low due to the possibility these variables could also be related to rape rates (see Table 1).

Some of these controls are based in specific theories, such as social disorganization

(Shaw & McKay, 1942), while others have been shown by prior research to be related to

violent crime more generally.

A first set of control variables included in the analyses is taken from social disor-

ganization theory (e.g., Shaw & McKay, 1942; Sampson, 1987). Social disorganization

specifies several variables such as family disruption, residential instability, and popula-

tion density, economic status, proximity to urban areas and ethnic heterogeneity 10 that are

linked to violent crime including, presumably, rape. Sampson (1987) argued that marital

and family disruption decreases informal social controls at the community level. There-

fore family households that have a female householder with children (HOUSEHOLD)

and persons 15 and older who are divorced (DIVORCED) were used to measure family

disruption. Baron and Straus (1987) argued that in areas where there were a disproportio-

nate number of divorced and separated persons, rape rates would be higher.

Shaw and McKay (1942) argued that “transitional neighborhoods” have the high-

est rates of deviance. Transitional neighborhoods are those that suffer from a high rate of

population turnover. This turnover disrupts the community’s network of social relations

and its opportunities for collective efficacy. Population densities in these areas are also

very high. Thus percent population net change (POPCHG) and the proportion of owner

10These control variables were obtained from the 2000 edition of the U.S. Census Bu-
reau’s City and County Data Book.
 16

occupied housing units (OWNUNITS) were used to measure residential instability. Pop-

ulation per square mile (DENSITY) was used to measure the population density.

Sampson & Groves (1989) argued that socioeconomic status has been an ecologi-

cal correlate of crime and delinquency. Studies have indicated that rapists are more like-

ly to come from areas of poverty (Baron & Straus, 1987). Moreover, cities with higher

levels of unemployment are expected to have higher levels of rape because joblessness

may contribute to a climate of resentment and hostility, which is conducive to rape (Ba-

ron & Straus, 1987). Therefore the percent of the county population who possess a high

school diploma or (EDUCATION), percent unemployed (UNEMPLOYED), and percent

poor (POOR) were used to measure economic status.

Shaw and McKay (1942) argue that the social process of urbanization creates dis-

organization in the community, which reduces social control, resulting in violent beha-

vior. Bailey (1999) argues that rape occurs in large areas of urbanization, while Scott and

Schwalm (1988) specifically looked at the incidence of rape and found it to be higher in

more populated urban areas and poorer neighborhoods. Therefore, the percent of popula-

tion living in urban areas (URBAN) was used. The percent of populated urban areas

(URBAN) was calculated by the total number of persons living inside urbanized areas

divided by the total population

Three other measures used as control variables are the percent of the population

ages 18 to 24 (PCT18T24), the percent of the population who are black (BLACK), and

the ratio of males per 100 females (SEXRATIO). Studies have shown that areas with

disproportionate numbers of young, black, and single residents have a higher incidence of

rape (Bailey, 1999; Baron & Straus, 1987; Baron & Straus, 1984). Also, when there are
 17

more men than women in a particular geographic area like a county, rape rates tend to be

higher (Baron & Straus, 1987).

Data Analytic Strategy

The analysis utilizes bivariate correlations and multiple regression (Ordinary

Least Squares) to examine the influence of pornography on rape rates. Multiple regres-

sion allows for multiple independent variables to be tested on the dependent variable si-

multaneously. Multiple regression analysis allows the investigator to determine the

unique contribution of the various independent variables on the dependent variable. The

multiple regression program used in this study is SPSS.

ANALYSIS

Because the tested theory postulates linear relationships between the unit va-

riables of interest, measured continuously, a series of ordinary least-squares (OLS) re-

gression models was used to predict the rape rate. Table 2 presents the descriptive statis-

tics for all variables included in the analyses.

Multicollinearity

Table 3 presents the bivariate correlations for all variables included in the analysis. The

variable of interest, PLAYBOY, is significantly correlated at the .001 level with RAPE-

RATE. However, some of the other predictors are highly correlated with each other,

which raises the specter of multicollinearity—a common problem that may lead to biased

and unstable parameter estimates in cross-county studies of crime (Antonaccio & Tittle,
 18

2007; Messner & Rosenfeld, 1997; Pratt and Godsey, 2003; Savolainen, 2000). In this

analysis, the following correlation seemed to be problematic with multicollinearity:

EDUCATION with POOR (-.700) (see Table 3). Therefore, formal regression diagnostic

tests and variance inflation factors were computed for all of the multivariate models. In

the OLS regression models, the variance inflation factors for EDUCATION was 3.298

and POOR was 3.145 (see Table 4, Model 1). However, the variance inflation never ex-

ceeded 4.00, which is the threshold conventionally accepted as indicative of serious mul-

ticollinearity (Antonaccio & Tittle, 2007; Pratt and Godsey, 2003; Savolainen, 2000).

Playboy Rate and Rape Rate

Table 4 describes the results of the analysis, based on two models. Model 1 is the

bivariate coefficient that demonstrates the relationship between PLAYBOY and RAPE-

RATE. Anti-pornography feminists argue that pornographic magazines lead directly to

violence against women. In the first model (Table 4 Model 1) the affects of PLAYBOY

on RAPERATE explains 1.3% of the cross-county variation. PLAYBOY is related sig-

nificantly to RAPERATE (.008) in the direction that anti-pornography feminists posited;

that is, the higher the rate of Playboy subscriptions, the higher the rape rate. However,

because the rape rates may be linked with other variables, another model (2) was ex-

amined in which twelve control variables were included as predictors. Once these va-

riables were introduced (Model 2), PLAYBOY ceased to be significant (.001), thus

PLAYBOY is not related to RAPERATE when controlling for other factors.

 19

Other Predictors of Rape Rate

Although the primary interest was the effect of pornographic magazine subscrip-

tion rates (PLAYBOY) on rape rates (RAPERATE), the findings about the usual predic-

tors of rape rates are somewhat consistent with previous studies. Counties characterized

by households headed by females with children (HOUSEHOLD), persons 15 and older

who are divorced (DIVORCED), percent of population with a high school education or

higher (EDUCATION), percent of population who are unemployed (UNEMPLOYED),

percent of the population living in urban areas (URBAN), percent of the population ages

18 to 24 (PCT18T24), percent of the population that are black (BLACK), and ratio of

males per 100 females (SEXRATIO), were statistically significant (at the .05 level).

Model 2 explains 17.6% of the cross-county variation in rape rate.

Although not all of the variables that measured social disorganization in the mod-

el were significantly related to rape rates, the majority of the variables were. Similar to

previous research, the findings indicated that the rape rate increases in proportion to the

level of disorganization in society. As seen in model 2, as HOUSEHOLD (.067) and

DIVORCED (.140) increased so did the occurrence of rape rates. This would indicate

that as marital and family disruption increases, informal social controls decrease, which

can contribute to higher incidences of rape.

Two variables, UNEMPLOYED (.059) and EDUCATION (.110), which meas-

ured the socioeconomic status of the county, had significant, positive direct effects on the

rape rate. However, POOR, which also measured socioeconomic status, in contrast to

previous research, did not have a significant direct effect on rape rates. Therefore, a
 20

county with societal characteristics of higher unemployment rates and a high school edu-

cation increased the likelihood of rape.

A similar result was found using URBAN (.241) which also showed a significant

positive direct effect on rape rates. Comparing across the independent variables, UR-

BAN (10.501) was the highest predictor of rape in the model.

Furthermore, as in previous research, the analysis shows that the percent of the

population ages 18 to 24 (PCT18T24), percent of the population who are black

(BLACK), and the ration of males to females (SEXRATIO) are associated with higher

levels of rape. The results indicate that if a larger proportion of the population is black

and the higher the sex ratio (higher percent of males in the population) then the incidence

of rape will be higher.

Alternatively, three of the control variables in the model did not have a significant

relationship with rape rates. The variables that described Shaw and McKay’s (1942)

“transitional neighborhoods,” which should have had a high rate of rape because of the

disruption of the community’s network of social relations were not significant.

POPCHG, OWNUNITS, and DENSITY did not explain the occurrence of rape in coun-

ties.

The results confirm previous research on rape rates, which shows that several va-

riables are persistently significant predictors of rape rates, whereas others are not. How-

ever, the introduction of those predictors into the equation does alter the finding that the

subscription rate to pornographic magazines is not a predictor of rape rates.

 21

Table 1

Description of Variables for Regression Analysis

Variable Name Variable Description

The average number of rapes known to the police from 1999 to 2001
RAPERATE
in each county divided by the female population of the county then
multiplied by 100,000

PLAYBOY The number of Playboy subscriptions in 2002 per 100,000 population

HOUSEHOLD Percent of family households with a female householder

DIVORCED Percent of persons 15 and over who are divorced

POPCHG The percent of population net change from 1990- 2000

OWNUNITS Proportion of owner occupied units

DENSITY Population per square mile

EDUCATION Percent of county population with High School Education

UNEMPLOYED Percent of civilian labor force who is unemployed

POOR Percent of persons below poverty level

Total persons living in urban areas (can be in metropolitan or non-

URBAN
metropolitan areas)

PCT18T24 Percent of the population that is 18 to 24 years

BLACK Percent of population that is Black or African American

SEXRATIO Population by sex: males per 100 females

 22

Table 2

Descriptive Statistics for Variables Used in the Analysis

Variable Range Mean SD

RAPERATE .00-776.98 42.31 40.88

PLAYBOY 25.95-9656.71 1102.53 543.19

HOUSEHOLD 26.70-100.00 60.50 6.37

DIVORCED .00-19.13 9.50 1.95

POPCHG -42.00-191.00 11.13 16.05

OWNUNITS .00-89.90 73.95 7.79

DENSITY .00-66835 243.59 1666.35

EDUCATION 32.00-100.00 69.62 10.39

UNEMPLOYED 1.00-28.00 4.77 2.62

POOR .00-47.00 15.02 6.34

URBAN .00-1.00 .40 .31

PCT18T24 1.00-46.00 8.86 3.54

BLACK .00-87.00 8.94 14.61

SEXRATIO 74.00-205.00 98.62 8.98

 23

Table 3

Correlations for Variables Used in Analysis

RAPE OWN
PLAYBOY HOUSEHOLD DIVORCED POPCHG DENSITY
RATE UNITS

RAPERATE 1.00

PLAYBOY .113** 1.00

HOUSEHOLD .148** .327** 1.00

DIVORCED .195** .058** .059** 1.00

POPCHG .112** -.45** .072** .181** 1.00

OWNUNITS -.245** -.239** -.187** -.098** .092** 1.00

DENSITY .044** .037* -.091** -.016 -.006 -.343** 1.00

EDUCATION .169** .520** .495** -.025 .109** -.223** .058**

UNEMPLOYED .016 -.262** -.075** .134** -.112** .014 -.034*

POOR -.032* -.411** -.242** .089** -.172** -.126** .006

URBAN .353** .251** .147** .169** .138** -.532** .221**

PCT18T24 .190** .133** .091** -.081** .051** -.501** .062**

BLACK .102** -.323** -.269** -.013 -.039* -.157** .097**

SEXRATIO .028 .008 .147** .031* .196** -.003 -.074**

 24

Table 3

Correlations for Variables Used in Analysis (Continued)

SEX
EDUCATION UNEMPLOYED POOR URBAN PCT18T24 BLACK RATIO

EDUCATION 1.00

UNEMPLOYED -.426** 1.00

POOR -.700** .550** 1.00

URBAN .371** -.138** -.193** 1.00

PCT18T24 .071** -.033* .133** .362** 1.00

BLACK -.376** .234** .427** .085** .168** 1.00

SEXRATIO .054** .014 .019 -.124** .046** -.110** 1.00

**. Correlation is significant at the 0.01 level (1-tailed).

*. Correlation is significant at the 0.05 level (1-tailed).
 25

Table 4

Regression Coefficients Representing the Effects of Playboy Rates on Rape, Controlling

Other Variables a and Tolerance and Variance Inflation Factors

Model 1 Model 2

Variables b β t b β t Tolerance VIF

RAPERATE .008 (.001)* .113 6.242 .001 (.002) .017 .795 .641 1.559

PLAYBOY .428 (.131)* .067 3.271 .665 1.505

HOUSEHOLD 2.966 (.386)* .140 7.693 .844 1.185

DIVORCED .091 (.047) .036 .795 .805 1.242

POPCHG -.124 (.141) -.023 -.878 .394 2.541

OWNUNITS -.001 (.000) -.027 -1.440 .808 1.237

EDUCATION .431 (.119)* .110 3.626 .303 3.298

UNEMPLOYED .928 (.323)* .059 2.870 .660 1.516

POOR .097 (.192) .015 .504 .318 3.145

URBAN 31.921 (3.039)* .241 10.501 .533 1.876

PCT18T24 .715 (.242)* .062 2.949 .638 1.568

BLACK .339 (.056)* .122 6.111 .703 1.423

SEXRATIO .211 (.082)* .046 2.578 .882 1.134

R2 .013 .176
a
Standard errors of estimates in parenthesis
*p< .05, one tailed
 26

DISCUSSION AND CONCLUSION

For almost a half century, mainstream criminology has been confronted with fe-

minist arguments about the causes, consequences, and cures for male-on-female crime

(Bailey, 1999). While there are considerable differences among feminist perspectives,

many feminists have argued that the viewing of pornography, by men, influences the pre-

valence of rape. Although there have been numerous studies pertaining to male porno-

graphy exposure and rape, the findings of these studies are far from conclusive.

The purpose of the research presented here was to answer the question posed by

feminists’ theorists: “is there is a direct positive relationship between pornography and

rape.” This study evaluated if the rate of pornographic magazine subscriptions was a

significant predictor of rape rates or if there are some other leading variable(s) that are a

predictor(s) of rape.

Ostensibly, the findings of this study do not support the hypothesis derived from

the anti-pornography feminist explanation for rape. Instead, the results indicate that there

is no relationship between the rate of pornographic magazine subscriptions, specifically

Playboy, and rape rates at the county level, once other variables are added to the model.

Therefore, this finding is inconsistent with the theory that pornography contributes to a

cultural climate favorable toward violence against women, specifically rape. These find-

ings, however, are consistent with previous experimental studies which showed that ex-

posure to extremely graphic but non-violent pornography diminishes aggression against

women (Donnerstein, Linz, & Penrod, 1987). Subsequently, a time series analysis of

several European countries showed either no increase or a decrease in the rape rate fol-

lowing the legalization and subsequent increase in circulation rate of pornographic mate-
 27

rials (Kutchinsky, 1986). Based on these findings the viewing of pornography may not

have the effects on men that anti-pornography feminists have previously believed. The

argument is that by men viewing pornography, it subjugates, degrades and dehumanizes

woman which can cause men to have calloused attitudes towards women. Thus men

maintain their position of power over women through rape. However, based on findings

in this study and previous studies, these attitudes are not generated by sexually explicit

magazines, but arise elsewhere.

The current findings are also similar to that of Gentry (1991), who studied stan-

dard metropolitan statistical areas (SMSAs), but contrasts Baron and Straus (1984) and

Scott and Schwalm (1988) who studied states. A reason for this discrepancy could be the

unit of analysis used in the research. Ellis and Beattie (1983) proposed that, because rape

offenders and/or victims are geographically mobile within a metropolitan area, the SMSA

concept provides a much more realistic unit of analysis for relating demographic features

of a city population to the population’s relative prevalence of rape than does a central

concept. This is because the central city concept is a less valid way of defining a “self-

contained community of interacting individuals” than is the concept of SMSA, at least as

far as accounting for the phenomenon of rape in a modern, highly mobile population such

as the United States is concerned (Ellis & Beattie, 1983). A similar concept could be true

with counties. Offenders and/or victims are also geographically mobile within a county

and therefore could provide similar results as SMSAs. In fact, this study found a similar

relationship utilizing county level data. Counties and SMSAs, which are smaller units of

analysis, show no relationship between pornography and rape whereas larger units of

analysis show a positive relationship between pornography and rape, thus the relationship
 28

between pornography and rape in larger units of analysis, such as states, could be spu-

rious or an artifact. Baron and Straus (1989) and Gentry (1991) both acknowledged that

by using different units of analysis it would represent different “realities” of the commu-

nity. Gentry (1991) proposed that studies using different levels of analysis are needed to

have a complete understanding of why some states, counties, SMSAs, cities, census

tracts, and neighborhoods are more dangerous to women than others. Future studies need

to focus on smaller units of analysis such as the county and SMSAs to determine whether

or not certain areas are more dangerous to women and the underlying reasons behind a

high concentration of female victimization within these areas.

Although not the main focus of the study, results from some of the control va-

riables are of theoretical interest in helping explain why some of these areas are more

dangerous to women than others. Several of the control variables that looked specifically

at social disorganization, population who is divorced, population of households headed

by females with children, population with a high school education or higher, population

who are unemployed, population living in urban areas, were shown to contribute to rape

directly. Thus it seems that in the context of a sexist and violent society, a low level of

social integration and social controls contributes to a high incidence of rape (Baron &

Straus, 1987). Population ages 18 to 24 , population that are black, and ratio of males per

100 females are associated with higher rates of female victimization, specifically rape.

These results provide some support for the predictions derived from macrolevel theories

of victimization and are comparable to those reported in prior studies examining the

structural covariates of rape.

 29

There are, however, several limitations of these findings. In previous research,

investigators have operationalized pornography in a variety of ways. The definition of

pornography is pivotal to its study because there is virtually no agreement on either con-

ceptual or operational definitions of pornography (Baron & Straus, 1987). That is, re-

searchers have failed to conceptualize pornography in such a way as to allow them to de-

velop a standardized operational definition of the concept. This can explain inconsisten-

cies in the research. For example, one study, using a particular operational definition of

pornography (or violence), may find a positive correlation between pornography and

rape, while another study using different operational definitions may find the opposite.

In this study, the effects of pornography were operationalized by measuring the Playboy

subscription rates. Using this operational definition, pornography was not found to be a

significant predictor of rape at the county level.

If pornography was operationalized using a combination of both softcore (lack of

showing aroused genitalia) and hardcore (features explicit sexual acts with an aroused

genitalia) magazines the results may have been different. Other operationalized mea-

surements, or a combination of these measurements of pornography, such as the con-

sumption of pornography by viewing internet pornography and the sales of pornographic

videos, could also have generated different results. Therefore, future research could ex-

pand the knowledge in this area by composing a more detailed definition of pornography

to include the consumption of video pornography, internet pornography, and all sales and

distribution of pornographic magazines. However, even with this limitation, the results

of this study proved important in theorizing about the relationship between pornographic
 30

material and rape rates, as well as for implications for the development of policies con-

cerning the circulation and distribution of pornographic material.

Another limitation is the general debate surrounding experimental versus non-

experimental research methods. The issue of experimental vs. non-experimental methods

has been a problem for social scientists (Allen, Emmers, Gebhardt, & Giery, 1995) and is

not limited to methods used to research the link between pornography and sexual vi-

olence. Scholars have utilized both correlational and laboratory research to study the re-

lationship between pornography and sexually aggressive acts.

Correlational studies examine the relationship between rates of reported sexually

aggressive acts and the circulation of pornography (Bauserman, 1996). Experimental re-

search, on the other hand, while controlling for threats to internal validity, does not ad-

dress very well threats to external validity—the generalizability of the findings involving

“real-world aggression” (Nemes, 1993, p. 474). Childress (1991) details the limits of ex-

perimental designs for studying the link between pornography and sexual violence, in-

cluding: (a) the unreal nature of violence depicted in laboratory settings, (b) the lack of

real punishment or social control for subjects, (c) lack of control for subjects’ inhibitions

while being observed or interviewed, (d) the use of willing college students as the norm,

(e) experimenter demand effects, (f) publication of studies only if they have positive re-

sults (that is, a positive relationship between pornography and sexual violence), (g) the

lack of good definitions of violence and aggression, (h) the ethical inability to produce

real violence (p. 184). Based on these limitations, correlational research, at the individual

and macro-level, is the best way to obtain evidence outside the laboratory setting regard-
 31

ing the potential effects of exposure to sexual materials on sexual violence (Bauserman,

1996).

This study found that percent divorced, percent of the population with a high

school education or higher, percent of population who are unemployed, percent of the

population living in urban areas, percent of the population ages 18 to 24, percent of the

population that are black, and population by sex: males per 100 females, were significant

predictors of rape. A potentially more thorough examination should consider other con-

trol variables that were beyond the scope of the current study. Desired control variables

would include the GINI index, gender in/equality, the breakdown of race-specific rape

and a measurement of sex offenders as to how much pornography they had consumed

before raping.

By looking at gender equality/inequality as variables in the study, it could deter-

mine if there was a backlash effect occurring. The backlash hypothesis suggests that as

women gain more power whether it is occupational, educational, or political, violence

against them increases because men will attempt to control the strides made by women.

Therefore in effect, there is a “backlash” against these women because their advancement

is seen as a threat to male dominance, and males use violence to remove that threat (Vie-

raitis & Williams, 2002).

In conclusion, this study failed to find support for a major deductive element of

the feminist explanation for rape. Instead, the present study could explain other femin-

ist’s perspectives in which the viewing of pornography provides a cathartic effect for

males and therefore it alleviates some of the rape that might occur in society rather than

contributing to it. Therefore the fundamental causes of rape are to be found in violence,
 32

sexism, and social disorganization—not in pornography (Baron & Straus, 1987). Future

research and policy procedures designed to reduce rape should focus on the underlying

causes of violence against women.

 33

References

Allen, M., D' Alessio, D., & Brezgel, K. (1995a). A meta-analysis summarizing the ef-

fects of pornography II: Aggression after exposure. Human Communication Re-

search, 22(2), 258-283.

Allen, M., Emmers, T., Gebhardt, L., & Giery, M. A. (1995b). Exposure to pornography

and acceptance of rape myths. Journal of Communication, 45(1), 5-26.

Antonaccio, O., & Tittle, C. R. (2007). A cross-national test of Bonger's theory of crimi-

nality and economic conditions. Criminology, 45(4), 925-958.

Attorney General's Commission on Pornography. (1986). Final Report, U.S. Department

of Justice. Washington, DC.

Audit Bureau of Circulation. (2000). Supplemental Data Report: Playboy Magazines.

ABC, Schaumburg, IL.

Bachy, V. (1976). Danish "permissiveness" revisited. Journal of Communication, 26(1),

40-43.

Bailey, W. C. (1999). The socioeconomic status of women and patterns of forcible rape

for major U.S. cities. Sociological Focus, 32, 43-63.

Baron, L., & Straus, M. (1984). Sexual stratification, pornography, and rape in the United

States. In N. Malamuth & E. Donnerstein (Eds.), Pornography and sexual aggres-

sion (pp. 185-209). Orlando, FL: Academic Press, Inc.

 34

Baron, L., & Straus, M. A. (1987). Four theories of rape: A macrosociological analysis.

Social Problems, 34(5), 467-489.

Baron L., & Straus, M. A. (1989). Four Theories of Rape in American Society: A State-

Level Analysis. New Haven: Yale University Press.

Bauserman, R. (1996). Sexual aggression and pornography: A review of correlational re-

search. Basic and Applied Social Psychology, 18(4), 405-427.

Ben-Veniste, R. (1971). Pornography and sex crime: The Danish experience. In

Technical report of the Commission on Obscenity and Pornography. (Vol. 7, pp.

91-147). Washington, DC: U.S. Government Printing Office.

Bergen, R. K., & Bogle, K. A. (2000). Exploring the connection between pornography

and sexual violence. Violence and Victims, 15(3), 227-234.

Boeringer, S. (1994). Pornography and sexual aggression: Associations of violent and

nonviolent depictions with rape and rape proclivity. Deviant Behavior, 15, 289-

304.

Brownmiller, S. (1975). Against our will: Men, women, and rape. New York: Simon &

Schuster.

Bureau of Justice Statistics. (2007). Criminal Victimization in the United States, 2000

Statistical Tables. Washington, DC: Bureau of Justice Statistics, US Department

of Justice.

Byrne, D., & Kelley, K. (1984). Introduction: Pornography and sex research. In Porno-

graphy and sexual aggression (pp. 1-15). Orlando, FL: Academic Press.
 35

Carter, D. L., Prentky, R. A., Knight, R. A., Vanderveer, P. L., & Boucher, R. J. (1987).

Use of pornography in the criminal and developmental histories of sexual offend-

ers. Journal of Interpersonal Violence, 2, 196-211.

Childress, S. A. (1991). Reel "rape speech": Violent pornography and the politics of

harm. Law and Society Review, 25, 177-214.

Cook, R. F., Fosen, R. H., & Pacht, A. (1971). Pornography and the sex offender: Pat-

terns of previous exposure and arousal effects of pornographic stimuli. Journal of

Applied Psychology, 55, 503-511.

Court, J. (1984). Sex and violence: A ripple effect. In Pornography and Sexual Aggres-

sion (pp. 143-172). Orlando, FL: Academic Press, Inc.

de Grazia, E. (1992). Girls Lean Back Everywhere: The Law of Obscenity and the Assault

on Genius. New York: Random House.

Demare', D., Briere, J., & Lips, H. M. (1988). Violent pornography and self-reported li-

kelihood of sexual aggression. Journal of Research in Personality, 22, 140-153.

Donnerstein, E. (1984). Pornography: Its effect on violence against women. In N. Mala-

muth & E. Donnerstein (Eds.), Pornography and sexual aggression (pp. 53-81).

Orlando, FL: Academic Press, Inc.

Donnerstien, E., Linz, D. & Penrod, S. (1987). The Question of Pornography. New York:

Free Press.

Dworkin, A. (1981). Pornography: Men Possessing Women. New York: G.P. Putnam's

Sons.

Dworkin, A., & Mackinnon, C. (1988). Pornography and Civil Rights. Minneapolis: Or-

ganizing Against Pornography.

 36

Ellis, L., & Beattie, C. (1983). The feminist explanation for rape: An empirical test.

Journal of Sex Research, 19(1), 72-93.

Garcia, L. T. (1986). Exposure to pornography and attitudes about women and rape: A

correlational study. The Journal of Sex Research, 22, 378-385.

Gentry, C. S. (1991). Pornography and rape: An empirical analysis. Deviant Behavior,

12, 277-288.

Gove, W., Hughes, M., & Geerken, M. (1985). Are uniform crime reports a valid indica-

tor of the index crimes? An affirmative answer with minor qualifications. Crimi-

nology, 23, 451-491.

Hindelang, M. J. (1974). The uniform crime reports revisited. Journal of Criminal Jus-

tice, 2, 1-17.

Jaffee, D., & Straus, M. A. (1987). Sexual climate and reported rape: A state-level analy-

sis. Archives of Sexual Behavior, 16, 107-123.

Johnson, P., & Goodchilds, J. D. (1973). Comment: Pornography, sexuality, and social

psychology. Journal of Social Issues, 29(3), 231-238.

Kimmel, M. S., & Linders, A. (1996). Does censorship make a difference? An aggregate

empirical analysis of pornography and rape. Journal of Psychology and Human

Sexuality, 8(3), 1-20.

Kupperstein, L., & Wilson, W. C. (1971). Erotica and antisocial behavior: An analysis of

selected social indicator statistics. In Technical report of the Commission on Ob-

scenity and Pornography (Vol. 7, pp. 311-324). Washington, DC: U.S. Govern-

ment Printing Office.

 37

Kutchinsky, B. (1971). Towards an explanation of the decrease in registered sex crimes

in Copenhagen. In Technical report of the Commission on Obscenity and Porno-

graphy (Vol. 7, pp. 263-310). Washington, DC: U. S. Government Printing Of-

fice.

Kutchinsky, B. (1991). Pornography and rape: Theory and practice? International Jour-

nal of Law and Psychiatry, 14, 47-64.

Kutchinsky, B. (1986). Paper presented at the International Congress on Rape, Tel Aviv,

Israel, April 7-11.

Langevin, R., Lang, R. A., Wright, P., Handy, L., Frenzel, F. R., & Black, E. L. (1988).

Pornography and sexual offenses. Annals of Sex Research, 1, 335-362.

Malamuth, N., & Ceniti, J. (1986). Repeated exposure to violent and nonviolent porno-

graphy: Likelihood of raping ratings and laboratory aggression against women.

Aggressive Behavior, 12, 129-137.

Malamuth, N., & Donnerstein, E. (1984). Pornography and sexual aggression. Orlando,

FL: Academic Press, Inc.

McElroy, W. (1995). XXX: A Woman's Right to Pornography. New York: St. Martin's

Press.

Messner, S. F., & Rosenfeld, R. (1997). Political restraint of the market and levels of

criminal homicide: A cross national application of institutional anomie theory.

Social Forces, 75, 1393-1416.

Mosher, D. L., & Sirkin, M. (1984). Measuring a macho personality constellation. Jour-

nal of Research on Personality 18, 150-163.

 38

Nemes, I. (1993). The relationship between pornography and sex crimes. The Journal of

Psychiatry & Law, 20, 459-481.

Parsons, T. (1947). Certain primary sources and patterns of aggression in the social struc-

ture of the Western World. Psychiatry 10: 167-181.

Pratt, T. C., & Godsey, T. W. (2003). Social Support, inequality, and homicide: A cross-

national test of an integrated theoretical model. Criminology, 41, 611-643.

President's Commission on Obscenity, & Pornography. (1970). Report of the Commission

on Obscenity and Pornography. Washington D.C.: U.S. Government Printing Of-

fice.

Russell, D. E. H. (1993). Making Violence Sexy: Feminist Views on Pornography. New

York: Teachers College Press.

Russell, D. E. H. (2000). Pornography and rape: A causal model. In Feminism & Porno-

graphy (pp. 48-93). Oxford, NY: Oxford University Press.

Sampson, R. J. (1987). Urban black violence: The effect of male joblessness and family

disruption. American Journal of Sociology, 93, 348- 382.

Sampson, R. J., & Groves, W. B. (1989). Community structure and crime: Testing social

disorganization theory. American Journal of Sociology, 94(4), 774-802.

Savolainen, J. (2000). Inequality, welfare state, and homicide: Further support for the in-

stitutional anomie theory. Criminology, 38, 1021-1042.

Scott, J. E., & Culvelier, S. J. (1987a). Sexual violence in Playboy magazine: A longitu-

dinal content analysis. Journal of Sex Research, 23, 534-540.

Scott, J. E., & Culvelier, S. J. (1987b). Violence in Playboy magazine: A longitudinal

analysis. Archives of Sexual Behavior, 16, 279-288.

 39

Scott, J. E., & Schwalm, L. A. (1988). Rape rates and the circulation rates of adult maga-

zines. Journal of Sex Research, 24, 241-250.

Seto, M. C., Maric, A., & Barbaree, H. E. (2001). The role of pornography in the etiology

of sexual aggression. Aggression and Violent Behavior, 6, 35-53.

Shaw, C., & McKay, H. (1942). Juvenile Delinquency and Urban Areas. Chicago: Uni-

versity of Chicago Press.

Toby, J. (1966). Violence and the masculine ideal: Some qualitative data. Annual Ameri-

can Academic Political Social Science 364, 19-28.

U.S. Bureau of the Census County and City Data Book. Retrieved February 27, 2006.

http://fisher.lib.virginia.edu/collections/stats/ccdb

U.S. Dept. of Justice, Federal Bureau of Investigation. Uniform Crime Reporting Pro-

gram Data [United States]: County-Level Detailed Arrest and Offense Data, 2000

[Computer file]. ICPSR03451-v4. Ann Arbor, MI: Inter-university Consortium

for Political and Social Research [producer and distributor], Retrieved January 16,

2006. http://webapp.icpsr.umich.edu/cocoon/NACJD-STUDY/03451.xml

Vieraitis, L. M. & Williams, M. R. (2002). Assessing the impact of gender inequality of

female homicide victimization across U.S. cities: A racially disaggregated analy-

sis. Violence Against Women, 8 (1), 35-63.

Walker, C. E. (1971). Erotic stimuli and the aggressive sexual offender. In Technical re-

port of the Commission on Obscenity and Pornography (Vol. 7, pp. 91-147).

Washington, DC: U. S. Government Printing Office.