X Views and Counting

Abstract

Academics and activists called to attention decades prior the importance of identifying, analyzing, and tracking the transmission of attitudes, behaviors, and norms correlated with violence against women. A specific call to attention reflected the media as a mode of transmission. This research builds on prior studies of media, with an emphasis on Internet search queries. Using Google search data, for the period 2004 to 2012, this research provides regional analysis of associated interest in rape-oriented pornography and pornographic hubs. Results indicate minor regional variations in interest, including the use of “BDSM” or “bondage/discipline, dominance/submission, and sadomasochism” as a foundational query for use in trend analysis. Interest in rape-oriented pornography by way of pornographic hubs is discussed in the context of microaggression.

Keywords

rape-oriented pornography rape culture Internet pornography microaggression Internet search queries macrolevel assessments

Introduction

For a half century, researchers have studied acquisition, transmission, and reinforcement of intricate markers, reflecting attitudes, behaviors, and norms, correlated with violence against women. This research originally placed emphasis on either personal or sociopolitical explanations, largely grounded in patriarchy (Miller, 1994). Using these theoretical explanations, Heise (1998) introduced an integrated ecological framework conceptualizing “violence as a multifaceted phenomenon grounded in interplay among personal, situational, and sociocultural factors” (Heise, 1998, p. 264). Highlighted within this theoretical frame is the primacy of culturally constructed messages (Heise, 1998).

Conveyance of these culturally constructed messages was at first restrained to the situational level. Interactions with others in the physical space provided an opportunity for transmission, but equally important was the capacity to accept or deny the intended message. However, mass media promoted a new mode for transmission of these culturally framed messages (DeKeseredy, Muzzatti, & Donnermeyer, 2014; Fallows, 1996; Gramsci, 1971; Gunter, 1987; Hayward & Presdee 2010; Slayden, 2010; Surette, 2011; Young, 2009).

Academics and activists have spent decades examining these messages. Flood and Pease (2009) suggest that deconstructing attitudes is a principal way to identify the varying domains of violence within a culture. When specifically assessing violence toward women, the authors suggest that it is important to examine the consistent link between attitudes and beliefs supporting violence and perpetuating violence given following four factors: individual, organizational, community, and societal. For the current analysis, societal-level factors, specifically transmission of violence-supportive attitudes by media, become an important narrowing lens. Flood and Pease (2009) suggest that media representations of violence toward women affect attitudes and behaviors to varying degrees, largely because those who consume media do so in diverse ways and personal attributes can serve to mitigate the effects. However, the authors do note, “the relationship between violent media representations and violent attitudes is reciprocal, in that viewers with violent inclinations and behaviors show greater interest in and enjoyment of violent media representations” (Flood & Pease, 2009, p. 135). Thus, media in various forms (e.g., television, film, music, games, and pornography) serve as a mentor to the masses, reinforcing traditional gender roles and promoting violence-supportive attitudes. Moreover, if the content objectifies women and is sexually violent, it has the ability to reinforce attitudes consistent with violence toward women that are sexually aggressive and sexist (Fisher & Barak, 1991; Flood & Pease, 2009; Foubert, Brosi, & Bannon, 2011; Gossett & Byrne, 2002; Hald, Malamuth, & Yuen, 2010; Harrington & Neilson, 2009).

Despite the focus on media, as a transmission medium for attitudes supportive of violence toward women, the Internet has received limited research (Flood & Pease, 2009). This is disconcerting given the growth of Internet users, representing ~2.4 billion global users and ~275 million users in North America (Miniwatts Marketing Group, 2013). The availability and variety of content, largely unfettered in access, has fundamentally influenced generations of individuals (Castells, 1996). As Greenfield and Yan (2006) offer, although research has been unable to ascertain the relative strength of this influence, researchers must recognize its ability to reproduce subculture norms, behavior, and attitudes.

Recognizing the Internet possesses many avenues of transmission; we use a quantitative analysis of Internet search queries associated with rape-oriented pornography as a proxy measurement for gendered microaggression. The availability of search query data, as made available by Google Trends, enables researchers with the ability to explore popularity and broad utilization of a myriad of search terms. Building upon the methodology of Roussinov and Robles-Flores (2007), this research restrains the analysis to the United States, while introducing factor analysis. With this methodology, our research seeks to understand cultural transmission and gendered microaggression specific to this niche genre of pornography.

Literature Review

Culture is an assemblage of “settings, norms, traditions, values, practices, beliefs, and assumptions” that influences both individual and group behavior while providing a framework for the construction of meaning derived from cultural experiences (Argiero, Dyrdahl, Fernandez, Whitney, & Wooding, 2010, p. 27). Researchers have commonly expressed the importance of culture as a prominent factor, manifesting as an entire subset within criminology and one of the most often attributed factors expressed in comparative studies (Adler, 1997). However, as Adler (1997) contends, culture is merely the lens, to which scholars must narrow the lens of observation to understand the factors associated with developing new perspectives. Culture cannot become merely a “place card” for factors unknown, rather researchers must look holistically, which is to say at a macrolevel to understand and promote research narrowly tailored toward a given relationship.

Specific to violence against women, sociocultural explanations have recognized the importance of identifying and examining attributes exceeding that of the dominant perspective that it is merely patriarchy, responsible for the dominance and abuse experienced by women (Miller, 1994). Although patriarchy initially framed the macrolevel analyses conducted, it was the culturally constructed messages, conveyed, accepted, and reproduced within and between a person’s social environments (Edleson & Tolman, 1992), which became increasingly a focus of inspection.

Researchers examined transmission of attitudes and beliefs reflecting cultural acceptance of violence against women, conveyance and acceptance of rape myths (i.e., false, albeit common, beliefs that rape is deserved and/or enjoyed), and rationalization for violence against women (Argiero et al., 2010; Chapleau, Oswald, & Russell, 2008; Flood & Pease, 2009; Gossett & Byrne, 2002). As Sanday (1981) offered, the cross-cultural explanation for acquiring those specific attitudes and beliefs was a product of hypermasculinity. Deeply engrained and respected was the “macho personality” reflecting traits such as power, dominance, a lack of empathy, and toughness (Mosher & Sirkin, 1984). The transmission and reinforcement within the community promoted hegemonic masculinity and gender-linked crimes. Conditioned were men to aspire to these traits and women, their antithesis. The Media Education Foundation documentary exemplifies this idea in Bro Code: How Contemporary Culture Creates Sexist Men (Keith, 2011). The documentary highlights how culturally transmitted messages of misogyny and hypermasculinity diffuse through popular culture, specifically by television, movies, music, and pornography. Messages promoting the degradation and objectification of women are received en masse, imbedded, and subsequently reinforced by heterosexual male peer groups who “police” other men to ensure adherence to the masculine ideal. Mass media reinforce to men, beginning at a very young age, that sexism is not merely customary, but “cool” (Keith, 2011).

Emerging from the work of Sanday (1981), scholars sought to better understand the sources of transmission by exceeding the traditional social environments (family, peers, and authority figures). This expansion included various sources of media, including print media (Madriz, 1997; Meyers, 1997; Pazzani, 2007), music and music videos (Sommers-Flanagan, Sommers-Flanagan, & Davis, 1993), television (Greenberg, Abelman, & Neuendorf, 1981; Greenberg & Busselle, 1996; Olson, 1994; C. Wilson & Gutierrez, 1995), movies (Adelman, 1989; Bufkin & Eschholz, 2000; Rothman, Powers, & Rothman, 1993; W. Wilson, 1994), and most recently, the Internet (Gossett & Byrne, 2002; Short, Black, Smith, Wetterneck, & Wells, 2012). Two works exemplifying this new focus is the research of Baron and Straus (1989) and later Pazzani (2007). These works introduced proxy measurements analyzing gender inequality and cultural legalization of violence within subscription rates for magazines, including pornographic magazines and magazines about violence. Although a minor point within an impressive work on stranger and acquaintance sexual assaults, the work of Pazzani (2007) empirically tested via a proxy measurement of interest in print media, the sociocultural explanation for violence against women. However, following the work of Pazzani (2007), few studies would look to proxy measurements, which are increasingly important within cross-cultural studies, seeking to identify a common frame of reference.

Problematic within such research has been the subjective nature of the methods most often used within content analysis of popular media (Rothman et al., 1993). Although projects have adhered to the almost dogmatic tenants of their respective qualitative methods, these approaches have had to overcome skepticism of specific coding or explaining relevancy to avoid labeling as merely hyperbole. A good example of such criticism reflects in the research of Gossett and Byrne (2002), undertaking content analysis of rape-oriented websites. Despite a multifaceted analyses of 31 violent pornographic websites, missing was an establishment of the relative interest in such websites. The Internet consists of tens of billions of webpages, and while content analysis of websites is worthy of study, the nature of the Internet affords anyone the ability to create and seek out material, regardless of how odd or depraved. Therefore, the mere identification of these websites, although interesting and theoretically worthy, necessitates an empirical assessment of associated interest and use.

Recognizing that Gossett and Byrne (2002) did not have access to data reflecting interest in rape-oriented pornography, the intent of this study is to provide an empirical foundation for further validating the work. This research also replicates portions of the work of Baron and Straus (1989) and Pazzani (2007) by directly measuring interest in rape-oriented pornography material regionally. Last, this research provides a theoretical and methodological platform for studying interest in rape-oriented pornography using Internet search queries.

Rape Pornography

“Rule #34: If you can imagine it, it exists as Internet porn” (Ogas & Gaddam, 2011, p. 7). This statement, from the caption of a Calvin and Hobbes–inspired comic, highlights the current state of Internet pornography. Pornography easily and anonymously accessed and ripe with variability (Gorman, Monk-Turner, & Fish, 2010; Short et al., 2012). Violence, once relegated to niche genres, is now unexceptional in mainstream pornography as the porn industry competes to satiate the desire for more extreme content (Dines, 2010). In their content analysis of current, best-selling pornographic films, Bridges, Wosnitzer, Scharrer, Sun, and Liberman (2010) found that of the more than 300 scenes analyzed, targets of both physical and verbal aggression were predominantly women (94.4%). Overwhelmingly, 89.8% of the content contained aggression, and when that aggression targeted women, 95.1% responded with expressions of pleasure or neutrally with no discernible opposition. Furthermore, unusual acts potentially harmful and humiliating to women in real life are becoming commonplace in mainstream pornographic films. For example, 41% of scenes analyzed contained what the industry calls ATM or “ass to mouth” (i.e., a performer providing oral sex subsequent to the insertion of a penis into the anal cavity). In addition, 19.1% contained double penetration (i.e., a performer penetrated by two objects orally, vaginally, or anally; Bridges et al., 2010). These findings are consistent with the work of Sun, Bridges, Wosnitzer, Scharrer, and Liberman (2008) and Gorman et al.’s (2010) finding that more than half of the pornographic content analyzed in each of their studies contained verbal or physical aggression performed by men and directed toward women who subsequently responded with signs of pleasure or indifference.

Although not all mainstream pornography depicts aggression, niche genres, specifically rape-oriented pornography, are blatant caricatures of the consumer’s desire for extreme, sexualized content. We define rape-oriented pornography as the depiction of sexual assault or forced sexual intercourse. Content typically accessed via free websites as well as pay-per-view or membership sites (Gorman et al., 2010; Gossett & Byrne, 2002). Most images and videos within this genre depict male rapists assaulting female victims, reflecting much of what is known about rape in the criminal justice community. The following scenario is common: a simulated sexual assault, rape pornography may show a woman forced by a man or several men into nonconsensual sexual acts. The portrayal of the victim within this context is an unwilling participant on screen, forced to perform sexually as a response to verbal or physical demands, threats, or physical manipulation.

Although limited definitive data exist on whether men or women seek out rape pornography with greater frequency, we can begin to surmise the audience from previous research on pornography consumerism. Compared with women, introduction of pornography occurs earlier for men ( Morgan, 2011), they are heavier consumers (Carroll et al., 2008; Gorman et al., 2010; Morgan, 2011), and find it more socially acceptable to do so (Carroll et al., 2008; Morgan, 2011). In fact, Carroll et al. (2008) offers that “pornography use is as common as drinking” among college men (p. 23). Although, the primary consumer demographic of pornography is men, women are not exempt. Carroll et al. (2008) found that nearly half (49%) of young women believe viewing pornography is acceptable, and 31% of these women report actually using pornography. When women view pornography, research suggests a preference for forms that explicitly address the sexual needs and desires of women, as opposed to mainstream forms, which largely address those of men (Sun et al., 2008). Therefore, women prefer to consume pornography within a relationally based context (e.g., with a romantic partner or within the context of an interactive sexual dialogue online ) or when pornography highlights unrestricted and equitable sex roles as opposed to a male-dominating dynamic (Sun et al., 2008). However, other researchers suggest that women may explore pornography alone and in varying forms much like men (Yoder, Virden, & Amin, 2005). In fact, Brosi, Foubert, Bannon, and Yandell (2011) found that within the last 12 months, sorority women surveyed had readily consumed both hardcore (46%) and sadomasochistic (21%) pornography, forms once thought to have an almost exclusively male audience.

Studies suggest that both male and female consumers of pornography, specifically violent pornography, are more likely to accept rape myths (Gossett & Byrne, 2002; Harrington & Neilson, 2009) and are less likely to intervene as a bystander in the event of a sexual assault (Brosi et al., 2011; Foubert et al., 2011). Furthermore, studies suggest that male consumers are more likely to hold views consistent with support for violence toward women after viewing pornography, especially violent pornography (Fisher & Barak, 1991; Foubert et al., 2011; Gossett & Byrne, 2002; Hald et al., 2010; Harrington & Neilson, 2009). On viewing violent pornography, “men’s tendencies to have fantasies about raping a woman” increase as well as accepting the idea that “women secretly wish to be raped” or enjoy the act (Fisher & Barak, 1991, p. 68). In one of the first studies examining the effects of viewing rape-oriented content, Donnerstein, Linz, and Penrod (1987) analyzed exposure to rape images, specifically those involving a female victim. The results from this attitudinal study suggest that nonrapist subjects began to experience arousal based on exposure to such rape-oriented images.

More recently, Foubert et al. (2011) surveyed fraternity members and found similar results accounting for the pornography the men reported watching (i.e., mainstream vs. rape-oriented). The results suggest, “The attitudinal variable of rape myth acceptance is a type of attitude that has a relationship only to the types of pornography where violence is more directly inherent and not necessarily to the broader spectrum of pornography” (p. 224). These results were also consistent in survey research of sorority members. Sorority members reporting viewing violent pornography were less inclined to assist in a bystander situation and more inclined to staunchly adhere to rape myths that justify male aggression (e.g., women deserve to be raped or that no woman can be raped against her will; Brosi et al., 2011).

One of the first studies providing a descriptive account specific to this pornographic niche was the work of Gossett and Byrne (2002). Their research specifically examined pornographic websites marketed as “rape” or “forced sex,” which were easily distinguishable from other niche genres such as BDSM or “bondage/discipline, dominance/submission, and sadomasochism” (Wilkinson, 2011, p. 494) that portray a voluntary (and respectful) exchange of domination and submission between two willing participants (Barker, 2013; Ogas & Gaddam, 2011). Within the 31 websites used in the study, accessed via Internet search engines (e.g., Yahoo!, AltaVista, and Excite), the authors found few male victims or female perpetrators, but rather mostly male perpetrator and female victim scenarios. Male perpetrators within this genre are largely invisible to the viewer, as the aggressor is rarely focused, noticed, or highlighted in any identifiable manner. The authors conclude that this was for ease of viewer assimilation. In other words, anyone viewing the material, regardless of social status, physical presence, or demeanor, could see himself (or herself) as the perpetrator of sexual assault. In addition, the authors acknowledge this relative lack of attention paid to the largely male perpetrator as a reflection of the male gaze. The consumer is essentially viewing the material “through the eyes of the rapist” (Gossett & Byrne, 2002, p. 704). In fact, any descriptive words or images used within these websites to describe the scenario exclusively concerned identifying the victim via age, race, or social class markers.

As an interesting observation, the location of the rape is not included in much of the content analyzed, leading them to believe this was to give the “illusion (or reality) that rape can occur anywhere” (Gossett & Byrne, 2002, p. 687). In addition, many of the sites used imagery that included the use of a weapon to overcome the apparent lack of consent from the victim. In accordance with the use of a weapon, the scenarios do not indicate any semblance of romance or mutual eroticism, rather, the portrayal of the rape in excruciating detail, including amount of physical pain the victim appears to be in throughout the act. Furthermore, the juxtaposition of victims within these scenarios is as follows: either (a) “deserving” of the rape or (b) innocent parties, a more common theme. Finally, the authors note that the unintentional access by unwitting individuals seeking out more mainstream forms of pornography through advertisements and external links to associated websites may encounter rape-oriented pornography. “Thus, actual women may experience increased danger of sexual violence due to the proliferation of violent pornography on the Internet,” even via those consumers who do not actively seek rape pornography out (Gossett & Byrne, 2002, p. 705).

Cultural Explanations

Although not an all-encompassing list, this discourse has produced several explanations for the levels of violence, specifically sexual violence, within a nation. Three macrolevel measurements, gender inequality (Sanday, 1981), social disorganization (Baron & Straus, 1989), and cultural legitimization of violence (Baron & Straus, 1989; Smith & Donnerstein, 1998), offer theoretical explanations for the degree of sexual violence and gender-based crime in a nation. In addition, microlevel explanations emerged, including deviant sexual arousal (Abel, Barlow Blanchard, & Guild, 1977), use of pornography (Brownmiller, 1975; Dworkin, 1989; Zillmann & Bryant, 1982), psychiatric disorders (Hockett et al., 2009), and cultural explanations (Brownmiller, 1975).

Gendered Microaggression

Placing this research into the preexisting theoretical narrative connects deeply to the discussion on microaggression, specifically gendered microaggression. Gendered microaggression reflects choices and actions (or inactions) taken in our daily lives, which promote and reinforce harm, subordination, exploitation, and disrespect of women (Hudson, Ballif-Sanvill, Caprioli, & Emmett, 2013). Initial research in this area explored racial microaggression on the Internet wherein what was initially limited to the physical world became unencumbered by place or time. As researchers highlighted, the Internet created an entirely new opportunity for microaggression. In the research of Tynes, Rose, and Markoe (2013), they discuss how postings of racial jokes and transmission of stereotypes (among other actions) on Facebook created an unhealthy and hostile environment to students of color. Most important within this research was the recognition that microaggression manifests not merely in the approval of postings, through a direct “up vote” but through the tacit approval of such postings, perpetuating future postings.

The relationship between direct approval and tacit approval has become a staple for websites, wherein users are able to promote content they find interesting, agree, or hide content they disagree. The majority of websites allowing commenting have opted for this strategy, in which users are able to promote content, through mere interest (calculated as a view). Content providers are then able to respond by creating content for that interest, interest that appears to have no limitation, as niche pornography has become increasingly available, marketed, and gendered (Gorman et al., 2010; Short et al., 2012).

As it concerns pornography, microaggression has manifested via the increasing interest in pornography, as measured via the proliferation of pornographic hubs. These hubs act as a repository for user-generated and commercially produced content, providing access to, in some instances, over a million pornographic videos. Pornographic hubs are repositories of gendered aggression. Manifestation of gendered microaggression occurs through the consumers’ perceived invisible actions. Representing actions promulgating rape myths and violence against women through the promotion of “most viewed” content, normalizing and providing rationalization that “I am not the only one” interested in this content. In the context of rape-oriented pornography, interest exists as pornography and niche pornography becomes more normalized in society and more readily accessible and categorized.

From a normalized perspective, consider one of the largest pornographic hubs on the Internet. This hub experienced tremendous growth and rising popularity on the Internet, providing access to over one million user-uploaded videos (Ogas & Gaddam, 2011), many of which are increasingly amateur (or portrayed as amateur) in origin. The resulting user base, has coalesced in popularity reflected in the nearly 350 million unique monthly users, consuming over 29 petabytes of bandwidth each month (Anthony, 2012). To place this into context, a single petabyte is ~1,000 terabytes, and the Library of Congress holds roughly 422 terabytes of data, with an addition of 5 terabytes a month (Library of Congress, 2013).

Subsequently, each rape-oriented pornographic video queried, viewed, or commented on acts as a source of gendered microaggression. Specific to the Xvideos website, there are 350 million monthly users receiving a culturally constructed message. This message reaffirms subordination and domination of women promoted through gendered microaggression. Each query and click, further enshrines acceptance of this pornography, as views are tallied, “likes” calculated, and increasingly so, comments added for each video. One comment that remains vivid, when examining websites ranking cinematic portrayals of rape in mainstream movies was that it “would be better if it was real.”

Intra-State Culture and Levels of Sexual Assault

According to Kalra and Bhugra (2013), “higher rates of sexual violence are expected to be more prevalent in cultures that encourage objectification of women, thus making them appear inferior to men (p. 245).”Thus, in the United States where many subcultures exist, it is natural to infer that varying perceptions of women may result in differing levels of sexual violence. Just as the Southern Culture of Violence perspective promoted a narrowing lens for proliferation of homicide across geographic areas (Ousey & Lee, 2010), we believe an extrapolation of this idea is suitable for sexual violence and rape culture on variations of sexual violence across regions. For instance, riddled are college campuses, notorious areas prime for sexual violence, with messages and cultural dynamics consistent with rape culture (Argiero et al., 2010). Thus, other locations and geographic areas could be similarly laden with attitudes, behaviors, and messaging consistent with violence toward women that could prove to be more heavily affected by sexual violence.

As discussed prior, the work of Gossett and Byrne (2002) proved influential in providing a descriptive lens for rape-oriented pornography. The research exposed reproduction of rape culture and the deeply capitalistic orientation of advertisements, profiting on violent pornography. However, missing from that analysis, and subsequently from much of the research, was evidence of associated interest. While a reader would be correct to offer the mere creation and promotion of websites catering to this violent niche is indication of interest, it is important to measure associated interest for analyses and comparison purposes, something largely missing from the discourse. As such, this research is positioned to further validate or potentially refute prior studies. Moreover, this research acts in the ability to provide a first empirical assessment of rape culture as measured by interest regionally.

Method

Pornography research has long suffered criticism and controversy, as it is undoubtedly polarizing and steeped in social politics; starkly contrasted between freedom of expression or sexual liberation and overt representations of violence against women (Attwood, 2005; Diamond, 2009; Dines, 2010; Ogas & Gaddam, 2011; Paasonen, 2011; Sun et al., 2008). Conflicting studies and a diverging discourse have fostered research that seeks to replicate the prior, by expanding and diversifying the population in an attempt to provide improved generalizability. However, rarely have researchers sought out new methods. Recognizing the discourse is strongly polarizing and projects have been limited and often criticized for lacking in generalizability (Paasonen, 2011), we use a rather new data source, Internet search queries.

Although a new data source, validation of Internet search queries exists across multiple disciplines, including political science (Bélanger & Meguid, 2008; Reilly, Richey, & Taylor, 2012; Ripberger, 2011; Scharkow & Vogelgesang, 2011; Weeks & Southwell, 2010), epidemiology (see Althouse, Ng, & Cummings, 2011; Ginsberg et al., 2009), health care (Reis & Brownstein, 2010), consumer behavior (Goel, Hofman, Lahaie, Pennock, & Watts, 2010), suicide studies (Marhan, Săucan, Popa, & Danciu, 2012; McCarthy, 2010), and efficacy-based studies within the social sciences (Scheitle, 2011). The rising popularity of this data source exceeds just its relative “newness” and claimed efficacy. Rather, access to aggregated data, spanning nearly a decade, and reflecting a user population, increasingly generalizable to national demographics, are increasingly important in identifying trends, reflected in associated interest to entered queries.

Data for this project make use of Google Trends, a service provided by Google, enabling access to associated levels of interest, relative to a specific query within a specified period. The period of analysis for this project comprises the years 2004 to 2012, reflecting the data made available for a full calendar year. Google provides a normalized output, reflecting the associated trend, specific to the query, the most commonly searched terms, and rising terms. The data provided do not provide the frequency of searches, expressed as a raw search total. Rather, the normalized output provides the opportunity to evaluate how queries have changed over time (i.e., has interest in the specific search term increased or decreased over the specified period). The scores range from 0 to 100, with 100 representing the highest interest. A score of 0 does not suggest no interest in the specific query. Rather, it may reflect a search volume that does not reach the threshold for a score of 1, relative to the total volume of searches, a known limit in normalized data.

Measures

Dependent variables

There are two dependent variables in this study (Table 1). The first dependent variable is a scale item representing interest in rape-oriented pornography. The scale consists of seven queries, each entered into Google Trends with quotation marks. Query selection reflected those most often associated with searches for rape-oriented pornography. A principal components factor analysis using a varimax rotation method identified five queries displaying internal consistency, removing the queries Rape Porn and Rape Videos. The remaining queries and their factor loadings are as follows: Forced Porn (.91), Free Rape Porn (.88), Rape Porn Videos (.86), Rape Scenes (.50), and Torture Porn (.89). Construction of the rape-oriented pornography scale includes these five queries, and assessment of the internal consistency demonstrates sufficient effects, possessing an alpha coefficient of .87. Removing the Rape Scenes (.50) item provides negligible improvements, with an alpha coefficient of .91.

Table 1.

Scale Items and Factor Loading.

Interest in Rape Pornography (n = 5)	Interest in Pornographic Hubs (n = 7)
Forced porn (.911)	Pornhub (.911)
Free rape porn (.889)	Porntube (.912)
Rape porn videos (.840)	Redtube (.860)
Rape scenes (.518)	Tube 8 (.969)
Torture porn (.868)	Xhamster (.860)
	Xvideos (.955)
	Youporn (.937)
	α = .965

The second dependent variable is interest in pornographic websites. The scale consists of seven queries, representing the top pornographic hubs as ranked by Alexa, entered into Google Trends with quotation marks. A principal components factor analysis using a varimax rotation method confirmed that all seven queries display internal consistency with loadings ranging from .86 to .96, with an alpha coefficient of .96.

As an exploratory study, this study seeks to test whether regional variations exist for interest in two scale items: interest in rape-oriented pornography and interest in pornographic hubs. As prior literature on sexual violence identifies culture and interest in specific types of violence as an appropriate focal point, we include these analyses as initial steps toward confirming whether variations exist. In addition to this regional analysis, we replicate prior research for testing the Southern Culture of Violence thesis. Based on these assumptions, the study tests the following hypotheses:

Hypothesis 1: States categorized as Old South will display higher interest in rape-oriented pornography.

Hypothesis 2: States categorized as Old South will display higher interest in pornographic hubs.

Hypothesis 3: There is a relationship between interest in BDSM and interest in rape-oriented pornography.

Independent variables

There are three independent predictors assumed to have effects on the dependent variables. These include region (categorized by the Uniform Crime Report [UCR]), Old South (Alabama, Arkansas, Florida, Kentucky, Louisiana, Mississippi, Missouri, Tennessee, and Texas), and interest in sadomasochism, measured by the query “BDSM.” As prior findings suggest interest in specific types of sexual violence correlates with higher interests in rape-oriented pornography, we include the query BDSM as a predictor.

Control variables

Control variables include the median age of the state (interval), population density (interval), and broadband ranking (interval). The latter a ranking assessing the population of a state with broadband access and Internet speed. For this research, we use the 2012 report¹ ranking.

Content validity

A methodological concern when using Internet search queries is the issue of content validity. Specific to our research, we must be able to establish the searches, in fact, reflect queries connecting to interest in this content and not searches used to access other content. Google Trends provides an efficient means of assessing content validity by the data output report. The output includes the most commonly used search terms, specific to the original query. The process of determining content validity requires analyzing the ancillary queries with each query—specific to the rape-oriented pornography scale—to determine whether the query reflected intent to procure content. The below list reflects the top searches related to the query “Torture Porn.” We have included the associated interest in parenthesis.

free torture porn (100), torture sex (60), sex torture porn (60), torture porn videos (55), rape torture porn (40), bondage torture porn (30), torture porn movies (30), bdsm torture porn (30), bdsm porn (30), torture porn video (25), gay torture porn (25), gay torture (25), pussy torture porn (25), lesbian torture porn (25), extreme torture porn (20), anal torture porn (20), torture porn tube (20), teen torture porn (20), anal torture (20), tit torture porn (20), torture porn pics (20), teen torture (20), hardcore torture porn (20), cock torture porn (15), horror porn (15), brutal torture porn (10), best torture porn (10), ball torture porn (10)

As displayed in the above, and as was completed for each query, the queries correlate with the intent to find rape-oriented pornography.

Limitations

As a new data source, the use of Internet search queries has several limitations. First, the data measures associated interest in queries. We do not have empirical evidence suggesting that a query made reflects material sought. Second, the demographic characteristics of the users are unavailable. As it concerns this research, trend analysis and macrolevel variations on interest are not concerned with the microlevel individual characteristics of the users. Third, the queries performed include only English phrasing . For example, we cannot account for searches performed in other languages. The decision to use only English terms was not an ethnocentric decision. Rather was a deliberate decision to mirror prior studies. Future research should include variations across language and culturally specific vernacular.

Results

Interest in Rape-Oriented Pornography

To provide context for the results of the multivariate analysis, we first provide the descriptive statistics (means) for the rape-oriented scale item. Figure 1 provides the graphic representation of the regional analysis. Regional analysis produces the following means ranking for interest in queries associated with rape pornography: South (54.43), Midwest (53.95), Northeast (44.84), and West (32.47). Although the South displayed the highest interest in this scale item, the Old South analysis produces the highest interest with a mean of 58.04.

Figure 1.

Regional analysis.

Because of Kurtosis in the rape-oriented scale, we use nonparametric tests. Results of the Kruskal–Wallis test reveal there is no statistically significant difference between the regions. However, results of the Wilcoxon–Mann–Whitney test suggest there is a statistically significant difference between the distributions of the rape-oriented pornography scales of the West and the non-West states (z = −2.135, p = .03). West states have the least interest in this scale item. The Old South is not statistically significant from the other regions for interest in rape-oriented pornography.

Presented in Table 2 are the results of two multiple regression models estimating the effects of the predictor variable BDSM on interest in rape-oriented pornography and pornographic hubs, including the control variables. The first linear regression model tests interest in rape-oriented pornography. The model is significant adjusted R² =.63, F(4, 45) = 22.12, p < .001 and explains 63% of the variance in rape-oriented pornography. With a coefficient of 2.13, a unit increase in BDSM results in an increase of 2.13 in interest in rape-oriented pornography. Subsequently, the more interest in the query “BDSM,” the higher is the interest in queries associated with rape-oriented pornography.

Table 2.

Regression Models Estimating the Effects of Predictors on Interest in Rape-Oriented Pornography and Pornographic Hubs.

	Model 1 (Linear)		Model 2 (Robust)
	Rape-Oriented Pornography		Pornographic Hubs
	B (SE)	t/F	B (SE)	t/F
Constant	−174.00 (54.09)	−3.21**	−19.59 (14.05)	−1.39
BDSM	2.13 (0.24)	8.71***	1.0 (0.06)	15.68***
Broadband ranking	0.67 (0.22)	2.95*	0.07 (0.06)	1.28
State population density	0.009 (0.01)	0.71	.007 (.003)	2.29*
State median age	−0.78 (1.22)	0.64	−0.03 (0.31)	−1.39
R ²		.63		.69

Note. BDSM = bondage/discipline, dominance/submission, and sadomasochism.

p < .05. **p < .01. ***p < .001.

Interest in Pornographic Hubs

Regional analysis produces the following means ranking for queries specific to the top seven pornographic hubs: South (60.79), Midwest (52.78), Northeast (52.26), and West (50.76). As displayed in the prior scale analysis, the Old South also displays the highest interest in the Pornographic Hub scale with a mean of 62.80.

To determine whether these variations are statistically significant, we use ANOVA and t tests. Both were suitable for this scale item. ANOVA results show the variations in interest in this scale item are not statistically different. However, unlike the prior scale, the South and Old South are statically different from the other regions. The Old South had higher interest in Pornographic Hubs (M = 62.81, SD = 13.49) than those non–Old South states (M = 52.95, SD = 13.05), t(48) = −2.03, p < .047. As expected, the UCR classification of South displays higher interest in Pornographic Hubs (M = 60.79, SD = 13.94) than those non–Old South states (M = 51.87, SD = 12.56); t(48) = −2.26, p < .028.

Model 2 tests interest in pornographic hubs. However, because of a violation of Shapiro–Wilk, we use a robust regression equation to control for the outliers and calculate the new R² using the procedure identified by Street, Carroll, and Ruppert (1988) and the script (rregfit) from University of California, Los Angeles (UCLA), Statistical Consulting Group (2014). The model is significant R² = .69, F(4, 45) = 66.12, p < .001 and explains 69% of the variance in queries associated with pornographic hubs. With a coefficient of 1.0, a unit increase in BDSM results in an increase of 1.0 in interest in pornographic hubs. Specific to Model 2, the more interest in the query “BDSM,” the higher the interest in queries associated with pornographic hubs.

BDSM

Recognizing the predictive power of the query BDSM, we further explore this variable item. Correlational analysis performed on the query against the dependent variables and individual items reveals moderate and strong relationships. BDSM is highly correlated with the Rape-Oriented Pornography Scale (.728) and Pornographic Hub Scale (.796) with p < .001. Considering the high alpha for the scale items, all of the individual scale items correlate with the query. However, the individual rape-oriented queries most correlated with the query include Free Rape Videos (.678), Forced Porn (.648), and Torture Porn (.584), with p < .001. Concerning specific pornographic hubs, the hub most correlated with the query is Pornhub (.924) and the least correlated hub is Redtube (.497), with p < .001. Regionally, a moderate correlation exists with the Old South (.434), p < .002.

Discussion

Heise (1998) injected into the discourse an integrated, ecological framework for the study of violence against women, a framework recognizing the interplay between personal, situational, and sociocultural factors. Overarching within these factors was what Heise (1998) referred to as the macrosystem, “a broad set of cultural values and beliefs that permeate and inform the other three layers of the social ecology” (p. 277). Through an analysis of the interplay between personal history, the microsystem, exosystem, and macrosystem, researchers could not merely better understand the causes of violence against women but attempt to intervene and prevent subsequent violence by targeting the precursors to this violence. As Heise (1998) contended, most complicated within those precursors are the macrosystem factors, those sociocultural factors promoting violence against women at a direct and indirect level.

The results of the current analysis, examining search interest in pornographic hubs and rape-oriented pornography, further validate, and expand on, Gossett and Byrne’s (2002) content analysis of rape-oriented pornographic websites. Although Gossett and Byrne (2002) provided a qualitative exploration of the content within these niche websites, our results provide initial empirical evidence measuring interest in the queries associated with rape-oriented pornography. Also, the results provide exploratory data for the macrosystem factors of Heise (1998). The results show support for two of the three main hypotheses. First, states categorized as Old South (Alabama, Arkansas, Florida, Kentucky, Louisiana, Mississippi, Missouri, Tennessee, and Texas) displayed higher interest in pornographic hubs. Therefore, the results from this exploratory study confirm that certain regional variations exist for interest in pornographic hubs. However, no support exists for the hypothesis that the Old South will display higher interest in rape-oriented pornography. Second, there is a relationship between interest in BDSM and interest in rape-oriented pornography.

In further reflecting on the results of the regression models and correlational analysis, it is not our intent to demonize the pornographic hubs utilized in this study. Rather we seek to explore the influence of these hubs on Internet search behavior, content availability, and as a source of microaggression. As it concerns search behavior, correlation analysis performed on the search query “hardcore porn” reveals a negative relationships with the five most prominent hubs: Pornhub (−.550), Redtube (−.633), YouPorn (−.598), Xvidoes (−.521), and Xhamster (−.502) with p < .001. Pornographic hubs hold the capacity to influence search behavior. Therefore, future research should explore the most popular and trending content queries and content tags used for the video material. If pornographic hubs are influencing search queries, their willingness to ban or remove queries associated with rape-oriented pornography could play an important role in making this specific niche pornography more difficult to find. Although many hubs already ban tags for “rape,” hubs do contain tags that would provide a descriptive means of finding content. For example, research should explore tags such as “brutal,” “slave,” “screaming,” “tied up,” “torture,” “pain,” “whipping,” and other tags used synonymously and in combination to examine the qualitative nature of this content. Also, research should explore the relationship between the number of video views and comments associated.

As for the relationship between BDSM and interest in rape-oriented pornography, we gather the nature of the BDSM content found on pornographic hubs is likely the reason for the association. There is a common misconception that BDSM, the acronym for “Bondage/discipline, dominance/submission, and sadomasochism” (Wilkinson, 2011), is solely about inflicting pain on a vulnerable recipient (Ogas & Gaddam, 2011). The BDSM community considers BDSM to be a voluntary and equitable exchange of power between consenting participants. In fact, as typified by the phrases, “safe, sane, and consensual” (SSC) or “risk aware consensual kink (RACK),” real-world BDSM scenarios typically include a lengthy discussion about consent before initiation of sexual activity (Barker, 2013, p. 896). The submissive party also has a safe word that will stop the act instantly when reaching his or her limits (Ogas & Gaddam, 2010).

However, although BDSM culture, erotic literature, and videos portraying all facets of BDSM relations do exist, pornographic hubs often contain only brief depictions of BDSM scenarios, most conceivably lacking the consent briefing that occurs before sexual activity begins and the debrief session following the acts. In fact, even BDSM community bloggers note the definitive lack of dialogue about consent in BDSM pornography (Barker, 2013). Thus, the tags found in hub sites may only lead to small components of BDSM sexual activity, components riddled with blatant representations of domination, submission, and aggression. The characterizations of pain infliction on a submissive party often found in BDSM content on hub sites, may then lead viewers to seek out pornography that freely depicts deliberate nonconsensual physical and sexual domination over a powerless victim, thus rape-oriented pornography. How users identify with either the aggressor or the dominating party introduces a new dimension within psychosexual research. Moreover, comments made by users to these videos could prove instrumental in exploring gendered microaggression and its influence on popular themes emerging within the next iteration of rape-oriented pornography.

Future research should explore BDSM queries and the influence of pornographic hubs on search queries given their association with rape-oriented pornography. Researchers should continue to explore content and user-generated comments on the BDSM tags. In addition, as prior research reveals, it is not simply pornography, but rather the type of pornography consumed. Future research should explore the variation across BDSM genres and tags used to categorize video content.

Although the regional analysis provides a foundation for variations in search interest across regions, future research should narrow the lens to specific contextual differences within city-level data to ascertain whether there are specific relationships between city-level characteristics and search interest. Analysis that is more acute will likely give researchers a better understanding of the specific nuances associated with each area that may elicit varying levels of interest in pornographic hubs and rape-oriented pornography.

Conclusion

As Gerbner and Gross (1976) theorized, media merely reflect the character of society. From this perspective, media are a gendered artifact, and in the context of this research, the Internet is an important litmus for measuring transmission of culturally constructed messages supportive of violence against women. The use of the Internet does not change who we are, rather it reveals who we truly are as an individual and as a society. Internet search queries provide a glimpse into who we are as a society as the social controls that are social desirability fade into the void that is cyberspace.

Albeit mostly a simulated and produced medium, rape-oriented pornography connects deeply with hegemonic masculinity. Each search query, view of a clip, epitomizes the tacit approval and reinforcement of male dominance and subordination, objectification, and commodification of women and other men as objects. As research shows, violent niche pornography dehumanizes and perpetuates commodification of women (and in some search queries other males) as merely an object of physical sexual pleasure. Troubling within this niche pornography and our results is not merely objectification of women, but the commodification that is violence and sex. Content views and comments offered become instruments of microaggression, only further promoting the content and directing others to similar content. However, as displayed in the microaggression research, the capacity for others to restrain behaviors by directly attacking content postings could prove interesting in the context of rape-oriented pornography. Do comments admonishing rape-oriented pornography act to dissuade others from viewing the content.

Perpetuation of rape myths is well documented across traditional media. Specific to the Internet, there are hundreds of millions of Internet users actively consuming pornography and a subset interested in rape-oriented pornography. The purpose of this research was to not merely raise awareness to these searches, but look into the potential relationships that would enable a more systematic analysis. As Rothman et al. (1993) reflected, without a valid frame of comparison, an agreed upon lens of analysis, data can be made to say anything. The high correlation of the query BDSM to the rape-oriented pornography and pornographic hub scales may be the foundation for developing standardized queries for use in tracking associated trends. The use of Internet search queries is merely in its infancy, but as displayed here, it is an important source nonetheless. As Hudson, Ballif-Sanvill, Caprioli, & Emmett (2012) introduced in their discussion on understanding violence against women, “What is there to see, and why aren’t we seeing it?” (p. 229).

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

Author Biographies

David A. Makin is a research fellow at the Washington State Institute for Criminal Justice (WSICJ) Division of Policing and Security. His recent research interests include the relationship between technology, society, and criminal justice.

Amber L. Morczek is a doctoral candidate at Washington State University and works in the Washington State University Center for Health Promotion Violence Prevention Programs. Her research interests include violence against women, sex offending, and corrections.

References

Abel

G.G.

Barlow

D.H.

Blanchard

Guild

(1977). The components of rapists’ sexual arousal. Archives of General Psychiatry, 34, 895-903.

Adelman

(1989). Representations of violence against women in mainstream film. Resources for Feminist Research, 18(2), 21-26.

Adler

(1997). Seizing the middle ground: Constructivism in world politics. European Journal of International Relations, 3(3), 319-363.

Althouse

B. M.

Y. Y.

Cummings

D. A.

(2011). Prediction of dengue incidence using search query surveillance. Public Library of Science, 1-7.

Anthony

(2012, April 4). Just how big are porn sites? Retrieved from ExtremeTech: http://www.extremetech.com/computing/123929-just-how-big-are-porn-sites

Argiero

S. J.

Dyrdahl

J. L.

Fernandez

S. S.

Whitney

L. E.

Woodring

R. J.

(2010). A cultural perspective for understanding how campus environments perpetuate rape-supportive culture. Journal of the Indiana University Student Personnel Association, 43, 26-40.

Attwood

(2005). What do people do with porn? Qualitative research into the comsumption, use, and experience of pornography and other sexually explicit media. Sexuality and culture, 9(2), 65-86.

Barker

(2013). Consent is a grey area? A comparison of understandings of consent in 50 Shades of Grey and on the BDSM blogosphere. Sexualities, 16, 896-914.

Baron

Straus

M. A.

(1989). Four theories of rape in American society: A state-level analysis. New Haven, CT: Yale University Press.

10.

Bélanger

É.

Meguid

B. M.

(2008). Issue salience, issue ownership, and issue-based vote choice. Electoral Studies, 27, 477-491.

11.

Bridges

A. J.

Wosnitzer

Scharrer

Sun

Liberman

(2010). Aggression and sexual behavior in best-selling pornography videos: A content analysis update. Violence Against Women, 16, 1065-1085.

12.

Brosi

M. W.

Foubert

J. D.

Bannon

R. S.

Yandell

(2011). Effects of women’s pornography use on bystander intervention in a sexual assault situation and rape myth acceptance. Oracle: The Research Journal of the Association of Fraternity/Sorority Advisors, 6(2), 26-35.

13.

Brownmiller

(1975). Against our will: Men, women and rape. New York, NY: Fawcett Books.

14.

Bufkin

Eschholz

(2000). Images of sex and rape: A content analysis of popular film. Violence Against Women, 1317-1344.

15.

Carroll

J. S.

Padilla-Walker

L. M.

Nelson

L. J.

Olson

C. D.

Barry

C. M.

Madsen

S. D.

(2008). Generation XXX pornography acceptance and use among emerging adults. Journal of Adolescent Research, 23(1), 6-30.

16.

Castells

(1996). The rise of the network society. Oxford, UK: Blackwell.

17.

Chapleau

K. M.

Oswald

D. L.

Russell

B. L.

(2008). Male rape myths: The role of gender, violence, and sexism. Journal of Interpersonal Violence, 23, 600-615.

18.

DeKeseredy

W. S.

Muzzatti

S. L.

Donnermeyer

J. F.

(2014). Mad men in bib overalls: Media’s horrification and pornification of rural culture. Critical Criminology, 22, 179-197.

19.

Diamond

(2009). Pornography, public acceptance and sex crime: A review. International Journal of Law and Psychiatry, 32, 304-314.

20.

Dines

(2010). Pornland: How porn has hijacked our sexuality. Boston, MA: Beacon Press.

21.

Donnerstein

Linz

Penrod

(1987). The question of pornography: Research findings and policy implications. New York, NY: Free Press.

22.

Dworkin

(1989). Pornography: Men possessing women. New York, NY: Plume.

23.

Edleson

Tolman

R. M.

(1992). Intervention for men who batter: An ecological approach. Newbury Park, CA: Sage.

24.

Fallows

(1996). Breaking the news: How the media undermine American democracy. New York, NY: Pantheon Books.

25.

Fisher

W. A.

Barak

(1991). Pornography, erotica, and behavior: More questions than answers. International Journal of law and Psychiatry, 14(1), 65-83.

26.

Flood

Pease

(2009). Factors influencing to violence against women. Trauma, violence, & abuse, 10, 125-142.

27.

Foubert

J. D.

Brosi

M. W.

Bannon

R. S.

(2011). Pornography viewing among fraternity men: Effects on bystander intervention, rape myth acceptance and behavioral intent to commit sexual assault. Sexual Addiction & Compulsivity, 18, 212-231.

28.

Gerbner

Gross

(1976). Living with television: The violence profile. Journal of Communication, 26, 172-194.

29.

Ginsberg

Mohebb

M. H.

Patel

R. S.

Brammer

Smolinski

M. S.

Brilliant

(2009). Detecting influenza epidemics using search engine query data. Nature, 1012-1014.

30.

Goel

Hofman

J. M.

Lahaie

Pennock

D. M.

Watts

D. J.

(2010). Predicting consumer behavior with web search. Proceedings of the National Academy of Sciences, 107, 17486-17490.

31.

Gorman

Monk-Turner

Fish

J. N.

(2010). Free adult Internet web sites: How prevalent are degrading acts? Gender Issues, 27, 131-145.

32.

Gossett

J. L.

Byrne

(2002). “Click here”: A content analysis of Internet rape sites. Gender & Society, 16, 689-709.

33.

Gramsci

(1971). Selections from the prison notebook. New York, NY: Free Press.

34.

Greenberg

B. S.

Abelman

Neuendorf

(1981). Sex on the soap operas: Afternoon delight. Journal of Communication, 31(3), 83-89.

35.

Greenberg

B. S.

Busselle

R. W.

(1996). Soap operas and sexual activity: A decade later. Journal of Communication, 46, 153-160.

36.

Greenfield

Yan

(2006). Children, adolescents, and the Internet: A new field of inquiry in developmental psychology. Developmental Psychology, 42, 391-394.

37.

Gunter

(1987). Television and the fear of crime. London, England: John Libbey.

38.

Hald

G. M.

Malamuth

N. M.

Yuen

(2010). Pornography and attitudes supporting violence against women: Revisiting the relationship in nonexperimental studies. Aggressive Behavior, 36(1), 14-20.

39.

Harrington

Neilson

(2009). A review of research on sexual violence in audio-visual media. Office of Film & Literature Classification, Wellington, NZ.

40.

Hayward

Presdee

(Eds.). (2010). Framing crime: Cultural criminology and the image. London, England: Routledge.

41.

Heise

L. L.

(1998). Violence against women an integrated, ecological framework. Violence Against Women, 4, 262-290.

42.

Hockett

J. M.

Saucier

D. A.

Hoffman

B. H.

Smith

S. J.

Craig

A. W.

(2009). Oppression through acceptance? Predicting rape myth acceptance and attitudes toward rape victims. Violence Against Women, 15, 877-897.

43.

Hudson

V. M.

Ballif-Sanvill

Caprioli

Emmett

C. F.

(2013). Sex & world peace. New York, NY: Columbia.

44.

Kalra

Bhugra

(2013). Sexual violence against women: Understanding cross-cultural intersections. Indian Journal of Psychiatry, 55, 244-249.

45.

Keith

(Director). (2011). Bro Code: How contemporary culture creates sexist men [Motion picture]. United States: Media Education Foundation.

46.

Library of Congress. (2013, May 1). Web Archiving. Retrieved from Library of Congress: http://www.loc.gov/webarchiving/faq.html#5

47.

Madriz

E. I.

(1997). Images of criminals and victims: A study on women’s fear and social control. Gender & Society, 11, 342-356.

48.

Marhan

A. M.

Săucan

Popa

Danciu

(2012). Searching Internet: A report on accessibility, nature, and quality of suicide-related information. Procedia-Social and Behavioral Sciences, 33, 373-377.

49.

McCarthy

M. J.

(2010). Internet monitoring of suicide risk in the population. Journal of Affective Disorders, 122, 277-279.

50.

Meyers

(1997). News coverage of violence against women: Engendering blame. Newbury Park, CA: Sage.

51.

Miller

S. L.

(1994). Expanding the boundaries: Toward a more inclusive and integrated study of intimate violence. Violence and Victims, 9, 183-194.

52.

Miniwatts Marketing Group. (2013, February 17). Internet world stats. Retrieved from http://www.internetworldstats.com/stats.htm

53.

Morgan

E. M.

(2011). Associations between young adults’ use of sexually explicit materials and their sexual preferences, behaviors, and satisfaction. Journal of Sex Research, 48, 520-530.

54.

Mosher

D. L.

Sirkin

(1984). Measuring a macho personality constellation. Journal of Research in Personality, 18, 150-163.

55.

Ogas

Gaddam

(2010). A billion wicked thoughts: What the Internet tells us about sexual relationships. Strand, UK: Penguin.

56.

Olson

(1994). Sex and the soaps: A comparative content analysis of health issues. Journalism & Mass Communication Quarterly, 71, 840-850.

57.

Ousey

G. C.

Lee

M. R.

(2010). The Southern culture of violence and homicide-type differentiation: An analysis across cities and time points. Homicide Studies, 14, 268-295.

58.

Paasonen

(2011). Online pornography: Ubiquitous and effaced. The Handbook of Internet Studies, 14, 424.

59.

Pazzani

L. M.

(2007). The factors affecting sexual assaults committed by strangers and acquaintances. Violence against women, 13, 717-749.

60.

Reilly

Richey

Taylor

J. B.

(2012). Using Google search data for state politics research: An Empirical Validity Test using roll-off data. State Politics & Policy Quarterly, 12, 146-159.

61.

Reis

B. Y.

Brownstein

J. S.

(2010). Measuring the impact of health policies using Internet search patterns: The case of abortion. BMC Public Health, 10(1), 514.

62.

Ripberger

J. T.

(2011). Capturing curiosity: Using Internet search trends to measure public attentiveness. Policy Studies Journal, 39, 239-259.

63.

Rothman

Powers

Rothman

(1993). Feminism in films. Society, 30, 66-72.

64.

Roussinov

Robles-Flores

J. A.

(2007). Applying question answering technology to locating malevolent online content. Decision Support Systems, 43, 1404-1418.

65.

Sanday

P. R.

(1981). The socio-cultural context of rape: A cross-cultural study. Journal of Social Issues, 37(4), 5-27.

66.

Scharkow

Vogelgesang

(2011). Measuring the public agenda using search engine queries. International Journal of Public Opinion Research, 23, 104-113.

67.

Scheitle

C. P.

(2011). Google’s insights for search: A note evaluating the use of search engine data in social research. Social Science Quarterly, 92(1), 285-295.

68.

Short

M. B.

Black

Smith

A. H.

Wetterneck

C. T.

Wells

D. E.

(2012). A review of Internet pornography use research: Methodology and content from the past ten years. Cyberpsychology, Behavior, and Social Networking, 15(1), 13-23.

69.

Slayden

(2010). Debbie does Dallas again and again: Pornography, technology, and market innovation. In Attwood

(Ed.), Porn.com: Making sense of online pornography (pp. 54-68). New York, NY: Peter Lang.

70.

Smith

S. L.

Donnerstein

(1998). Harmful effects of exposure to media violence: Learning of aggression, desensitization, and fear. In Geen

R. G.

Donnerstein

(Eds.), Human aggression: Theories, research, and implications for social policy (pp. 167-202). San Diego, CA: Academic Press.

71.

Sommers-Flanagan

Davis

(1993). What’s happening on music television? A gender role content analysis. Sex Roles, 28, 745-753.

72.

Street

J. O.

Carroll

R. J.

Ruppert

(1988). Note on computing robust regression estimates via iteratively reweighted least squares. The American Statistician, 42, 152-154.

73.

Sun

Bridges

Wosnitzer

Scharrer

Liberman

(2008). A comparison of male and female directors in popular pornography: What happens when women are at the helm? Psychology of Women Quarterly, 32, 312-325.

74.

Surette

(2011). Media, crime and criminal justice: Images, realities, and policies (4th ed.). Belmont, CA: Thompson.

75.

Tynes

B. M.

Rose

C. A.

Markoe

S. L.

(2013). Extending campus life to the Internet: Social media, discrimination, and perceptions of racial climate. Journal of Diversity in Higher Education, 6, 102-114.

76.

University of California, Los Angeles, Statistical Consulting Group. (2014). Introduction to SAS. Retrieved from http://www.ats.ucla.edu/stat/stata/faq/rregr2.htm

77.

Weeks

Southwell

(2010). The symbiosis of news coverage and aggregate online search behavior: Obama, rumors, and presidential politics. Mass Communication and Society, 13, 341-360.

78.

Wilkinson

(2011). “Extreme pornography” and the contested spaces of virtual citizenship. Social & Cultural Geography, 12, 493-508.

79.

Wilson

Gutierrez

(1995). Minorities and the media: Diversity and the end of mass communication. Beverly Hills, CA: Sage.

80.

Wilson

(1994). Sexuality in the land of Oz: Searching for safer sex at the movies. Lanham, MD: University Press of America.

81.

Yoder

V. C.

Virden

T. B.

III Amin

(2005). Internet pornography and loneliness: An association? Sexual Addiction & Compulsivity, 12(1), 19-44.

82.

Young

(2009). Moral panic its origins in resistance, ressentiment and the translation of fantasy into reality. British Journal of Criminology, 49(1), 4-16.

83.

Zillmann

Bryant

(1982). Pornography and sexual callousness, and the trivialization of rape. Journal of Communication, 32, 10-21.