Abstract
This study examined video game–playing aggression among graduate and undergraduate students at Rochester Institute of Technology in upstate New York. The following three research questions were posed: In the context of video game playing, what differences are there in levels of aggression in relation to sex? What differences are there in levels of aggression and type of video games played? Are aggression and length of video game playing related? A nonprobability sample of students (N = 175) was selected and electronically surveyed. The Buss–Perry Aggression Questionnaire, a widely used self-reported measure, was used to assess aggression. No significant difference was found between men and women with regard to levels of aggression, and no statistically significant relationship was found between aggression and length of video game playing (r = −.05, p = .52). However, significant differences were found in levels of aggression between playing and nonplaying respondents of action role-playing (t = −1.48, df = 175, p = .01) and role-playing video games (t = −1.24, df = 175, p = .02). Players of role-playing games had lower levels of aggression than nonplayers. Since role-playing requires one to take the part of another person, empathy might play a role in reduced levels of aggression. Although exposure to violent video games has been related to decreased empathy, role-playing games show just the opposite. It is suggested that controversies regarding violent video games might best be explained by moral panic theory and that competitive games may attract more aggressive players.
Video games have become a mass medium and a big business. The Entertainment Software Association (2011) reports that consumers spent $25.1 billion on video games, hardware, and accessories in 2010 and that 72% of all households in the United States play computer or video games. They are rapidly gaining popularity not only among children but also among young and middle-aged adults. The Entertainment Software Rating Board (2011) reports that the average age of a “gamer” is 34 and that 18- to 49-year-olds make up 49% of video games users. It estimates that video games are played in 67% of U.S. households and that the average gamer spent 8 hours per week playing. The Kaiser Family Foundation (2002) reported that 70% of children between the age of 2 and 18 years had at least one video game player in their homes.
It has been reported (Weber, Ritterfeld, & Mathiak, 2006) that 53% of games contain violence, and this exposure to violence may lead to negative outcomes, including violent behavior in individuals toward the simulated environment. The National Youth Violence Prevention Resource Center (2004) has reported that a 2001 review of the 70 top-selling video games found 49% contained serious violence (Weber et al., 2006). “Mature”-rated games (which contain adult content) are extremely popular with preteen and teenage boys who report no trouble buying the games (Weber et al., 2006).
There is no doubt that violent video games (VVGs) are among the most popular entertainment products for teens and adolescents, especially for boys (Weber et al., 2006). This may be because video game playing has been viewed as an appropriate play for males but not for females (Lucas & Sherry, 2004). Young men are more likely to engage in video game playing with their peers, as it will increase their likelihood for inclusion; consequently, they are likely to have higher levels of aggression as compared with girls (Lucas & Sherry, 2004).
One finding that has been stable throughout several decades of research is that video games are liked more and played more by males than by females. In 2004, Lucas and Sherry’s study of sex differences and video game play showed that there is a difference between type (violent or nonviolent) and time period for video games played by boys and girls. In addition to playing more than girls, boys report a preference for more violent games than girls (Yelland & Lloyd, 2001), while 10- to 13-year-old girls described games they liked as challenging or intellectually stimulating. Boys tended to describe their favorite games as exciting and having good graphics. In a recent survey on family media use, 76% of homes with at least one male child own video games as compared to 58% of homes with at least one female child (Lucas & Sherry, 2004).
Public Concerns
In the wake of school shootings and rampage massacres, VVGs have produced legal and political controversy on an international scale. Thus far, eight states have attempted to enact legislation restricting their sales (Hall, Day, & Hall, 2011) and Europe has considered a Union-wide regulation of VVGs (“Cracking Down,” 2007).
In 2005, the state of California passed a law prohibiting the sale or rental of VVGs to minors and requiring labels on their packaging. Although the law was struck down as violative of the First Amendment, the Supreme Court found that California relied on studies that “have been rejected by every court to consider them” (Brown v. EMA, 2011, p. 12) and that they did not “prove that VVGs cause minors to act aggressively” (pp. 12-13). The Court made direct reference to Dr. Craig Anderson’s claim that violent video games exert some effect on aggression but concluded that the effects were small and no different from the effects of other media.
In response to the question of whether the Australian National Classification Scheme should include an R18+ category for video games, a literature review was prepared (Australian Government, 2010). It concludes that the literature is “divided” and that harmful effects have been neither proven nor disproven. There are, at best, small effects of VVGs on aggressive behavior, but overall, the research is “contested and inconclusive” (p. 42).
The Swedish Media Council (2011) also reviewed the research literature from 2000 to 2011. In examining 161 studies, methodological problems were found in all but 5. Often, methods differed in measuring aggression before and after the gaming sessions. The Council questioned whether laboratory effects could be generalized to real life and concluded that VVGs might attract aggressive people but that there is no proof that VVGs produce aggressive behavior.
Violent Video Games and Violence
In an attempt to isolate the effects of violence and competitiveness on aggressive behavior, Adachi and Willoughby (2011) used both a violent and a nonviolent video game matched in competitiveness, difficulty, and pace of action. Violent content alone did not elevate aggressive behavior, but competitiveness elevated aggressive behavior independent of the level of violent content. The authors conclude that competition may have the greatest influence on aggressive behavior and suggest that future research investigate the degree of difficulty and pace of action on aggressive behavior.
Following government criticism of the VVG effects literature, a longitudinal study was conducted regarding youth aggression and dating violence as consequences of VVG use (Ferguson, San Miguel, Garza, & Jerabeck, 2012). It employed carefully controlled variables related to family, peers, and mental health. Video game violence was not related to any negative outcomes. Depression, antisocial personality traits, exposure to family violence, and peer influences best predicted aggression. In light of the exaggerated fears of video game violence, the authors suggest that the scientific community consider moral panic theory in identifying future panics that may influence scientific study.
To address the issue of causality regarding VVGs and violent behavior, Gunter and Daly (2012) applied propensity score matching. Using a large sample (N = 6,567) of eighth grade students, it was found that playing VVGs did not affect males on any outcomes when samples were matched. Some results indicate that females in matched samples had a slightly elevated likelihood for violent behaviors. Although the link to violence cannot be completely dismissed, they do not support the correlations found in previous research. The authors conclude that “assumptions made by the popular media and by policy-makers may be exaggerated at best and erroneous at worst” (p. 1353).
In what is perhaps the first experimental study to use one video game with both violent and nonviolent content in the game’s story line, Valadez and Ferguson (2012) investigated the effects on aggressive affect (hostile feelings), depression, and visuospatial cognition. No causal relationships were found. The authors conclude that a causal link between strict and aggression is weak or nonexistent when employing the most exacting methods.
Anderson et al. (2010) conducted a meta-analysis of the VVG literature. Likening the effects to television and film effects research, the authors refer to video game violence as the “new kid on the block” (p. 152) without the large number of empirical studies the older media have generated. Citing their own previous reviews, they conclude that the VVG effects may be somewhat larger but add that there has been “recent skepticism.” After a lengthy and detailed analysis, the estimate of VVGs on aggression was r = .19 for all studies. Anderson et al. make this recommendation:
Concerning public policy, we believe that debates can and should finally move beyond the simple question of whether violent video game play is a causal risk factor for aggressive behavior; the scientific literature has effectively and clearly shown the answer to be “yes.” (p. 171)
In response, Ferguson and Kilburn (2010) pointed out that the uncorrected estimate for the effects on aggression were small (r = .15). Methodological disagreements with Anderson et al. (2010) included whether to include unpublished studies, how to correct for publication bias, the value of bivariate correlations, the use of standardized and valid measures of aggression, and how to interpret effect size estimates. The authors call attention to the irony that as VVG sales have increased, youth violence in most industrialized nations has decreased (Ferguson, 2008). The correlation coefficient is reported as r = −.95, “a near-perfect correlation in the wrong direction” (Ferguson & Kilburn, 2010, p. 176). The authors conclude that psychology has exaggerated the focus on VVGs and has distracted attention away from other causes of aggression.
In the same issue of Psychological Bulletin, Bushman, Rothstein, and Anderson (2010) replied to Ferguson and Kilburn (2010) with a rebuttal. They disputed the r = .15 correlation reported by Ferguson and Kilburn and reported the estimate of VVG effect on aggression to be r = .19 for all studies and r = .24 for studies with more rigorous methods. They also claimed that their correlations were between “small” and “medium” but that this most commonly seen in social psychology. They further claimed that effect sizes for VVGs are similar to risk factor in physical health, such as exposure to lead, asbestos, or secondhand smoke (Bushman & Anderson, 2001).
Video game playing has often been compared to and contrasted with television viewing. Past concerns over television violence have been magnified since violence is prevalent in most popular video games (Dietz, 1998; Smith, Lachlan, & Tamborini, 2009). One contrast is that while television viewing has been characterized as active or passive (Sherry, 2007), video game playing has been described as more interactive, offering greater involvement, immersion, and spatial presence (Tamborini & Skalski, 2006). This greater level of involvement has often been taken as a sufficient condition for other outcomes, such as increased aggression (Krcmar & Lachlan, 2009). Moreover, in video games, violent behavior is reinforced since players are frequently rewarded for aggressive play. Television research has shown that reinforced aggression is imitated more than unrewarded behavior (Wilson et al., 1997). A number of scholars have concluded that playing VVGs will result in increases in aggression (Krcmar & Lachlan, 2009).
Theoretical Explanations
Uses and gratifications theory (Berelson, 1949; Blumer, 1933; Herzog, 1944) holds that people are active in their selection of media either to meet needs they have or to fulfill their desires. This approach to studying media reversed the earlier theories of media directly effecting passive audiences. Audiences consciously select the media they use. In speaking of the effects of television on children, Schramm, Lyle, and Parker (1961) said,
In a sense the term “effect” is misleading because it suggests that television “does something” to children . . . Nothing can be further from the fact. It is the children who are most active in this relationship. It is they who use television rather than television that uses them. (p. 1)
It can be theorized that playing video games provides a motivational cycle related to flow and success fueled by confidence. Thus, it may be correct to say that the males are entering into this motivational cycle and the females are not (Hamlen, 2010). If media consumption does not meet female needs in the same way as men’s, this may explain gender differences in video game playing.
Early theories used notions of aggressive instinct, catharsis, and frustration to explain potential origins of human aggression (Weber et al., 2006). However, recent theorizing explains the long-term effects of media violence on aggression as originating from observational learning related to aggressive behavior (Weber et al., 2006). Aggression was typically measured by allowing participants to inflict loud noise through headphones.
The general aggression model (GAM) theory has largely been used in studies of media violence, even though its proponents did not intend its application to be limited to that subject (DeWall, Anderson, & Bushman, 2011). It predicts that repeated simulation of antisocial behavior produces an increase in antisocial behavior (e.g., aggression) and a decrease in prosocial behavior (e.g., helping) outside the simulated environment (i.e., in “real life”; Anderson & Bushman, 2001; Sestir & Bartholow, 2010). The model attempts to explain the effects of VVGs on antisocial behavior (Sestir & Bartholow, 2010). According to this model, the aggressive contents of violent media instigate aggressive behavior through their impact on the person’s internal states, namely, cognitive, affective, and state of arousal, which many times can be misinterpreted as anger when applied to an aggressive situation.
Although the GAM does not specify the internal states that are affected by a particular aggression-enhancing stimulus, prior research (Anderson & Carnagey, 2009) suggests that VVGs can differentially increase both aggressive cognition and aggressive affect even when physiological arousal is controlled (Anderson & Carnagey, 2009; Chory, Goodboy, Hixson, & Baker, 2007). In addition to the input variables, the GAM holds the assumption that the three internal states are interrelated and affected by each other (Chory et al., 2007). Therefore, exposure to VVGs evokes associations with aggressive cognitions, arousal, and affect related to violence and instigates aggressive behavior in video game players (Greitemeyer & Osswald, 2010).
Recently, an argument has been made to “retire” the GAM (Ferguson & Dyck, 2012). The two main reasons cited are the lack of validity in measures to assess aggression and failure to test a wider range of variables predicting aggression (Ferguson, 2010; Freedman, 2002; Gauntlett, 2005; Savage, 2004). It has been reported that when other predictor variables are considered, concepts related to the GAM become nonsignificant (Adachi & Willoughby, 2011; Ferguson et al., 2008; Ferguson et al., 2009; Przybylski, Rigby, & Ryan, 2010).
Research Questions
Research Question 1: What differences are there in levels of aggression in relation to sex? (in the context of video game playing)
Research Question 2: What differences are there in levels of aggression and type of video games played?
Research Question 3: Are aggression and length of video game playing related?
This particular study has employed one variable of the video game–playing experience, aggression, and seeks to investigate its relation to gender, types of video game played (violent vs. nonviolent), and the length of video game playing. Thus, violent games will be compared with nonviolent games to determine what affect content has on feelings of aggression. The current study will also use the Buss–Perry Aggression Questionnaire (Buss & Perry, 1992), along with additional sets of questions (as part of the survey), to seek out answers for questions raised above.
Sherry (2007) found that games containing human or fantasy violence produce more aggressive outcomes than games containing sports violence in the observed behavior of the study participants. Sherry also suggested that playing video games longer may be associated with less aggression. Therefore, the present study seeks to determine the validity of the latter findings along with differences in aggression levels related to sex and types of video game played by males and females.
In the video game literature, it has been suggested that longer game play of an aggressive video game results in greater verbal and physical aggression, since the aggression primes become strengthened during the game play experience. Priming refers to the process by which a mental cue or association serves as a component to trigger related thoughts and behaviors (Krcmar & Lachlan, 2009). Prior research (Krcmar & Lachlan, 2009) produced some important findings regarding the length of play and differences in the levels of aggression. Krcmar and Lachlan’s study used undergraduate students who played the game Mortal Kombat and subsequently took the Buss–Durkee Hostility Inventory (Buss & Durkee, 1957). Taken together, this study suggests that video game play may increase aggression but that this increase in aggression may occur during shorter, not longer, game play (Krcmar & Lachlan, 2009).
Another important insight that contributes toward aggression in relation to video game playing is the competitive situations that stimulate aggressiveness (Anderson & Carnagey, 2009). Links have been found between VVGs and aggression, not because of the violent content but because the games mostly involve competition, whereas nonviolent video games are frequently noncompetitive (Anderson & Carnagey, 2009). The competitive aspect of the game might contribute toward increasing aggression by increasing arousal or by increasing aggressive thoughts or affect (Anderson & Carnagey, 2009). One can argue that the mark of a good video game is the challenge it poses to the player, and today frustration is an inherent quality of most VVGs (Williams, 2011).
Video Games as a Learning Process
Video games enhance learning in the consumers (male or female) and are considered a reasonably fast technique for education or imitation in this era of digital divide. The active involvement and constant attention to playing video games may enhance the learning of aggression (Williams, 2011). Learning occurs in both the mediated and unmediated settings and can be transposed into real life (Eastin, Griffiths, & Lerch, 2005). Additionally, behaviors learned in an unmediated environment can be accessed and used in a mediated context (Eastin et al., 2005). In other words, aggressive ideas found during video game play can combine with other related ideas, increasing the possibility of game players experiencing aggressive thoughts even outside the game environment (Eastin et al., 2005).
Since the average effect of video games on one’s aggression seems to be rather small in size, it is important to consider the amount of video game playing as a factor in increasing the overall levels of aggression present (Weber et al., 2006). Abelson (1985), Prentice and Miller (1992), and Rosenthal (1986) demonstrated that even small effects can result in high societal costs or damage under high-exposure conditions (Weber et al., 2006).
Beside the methodological questions, one explanation of the relationship between playing VVGs and different aggression levels it stimulates is that the players understand and interpret the same games differently (Weber et al., 2006). “Depending on how players internalize a game, and its violent content, playing might have a lesser or greater impact on the player’s attitudes, emotions, and behaviors” (Weber et al., 2006, p. 41).
Method
The study will rely on a nonprobability sample of students enrolled in Rochester Institute of Technology located in upstate New York. The university has both an undergraduate and a graduate degree program in game design and has been rated one of the top 10 schools for video game design (“Top 10 Game Design Programs,” 2011). A survey will be electronically distributed to both undergraduate and graduate students in the College of Liberal Arts and College of Computing and Information Sciences.
Part 1 of the survey begins by eliminating nonplayers from the survey. It then determines the types of video games the subject plays and classifies the participants into categories and calculates the participants’ aggression levels. This also helps in finding out whether there is a behavioral change by playing different video games or not.
Part 2 contains the Buss–Perry Aggression Questionnaire, a widely used self-reported measure of aggression. A number of researchers have investigated the reliability and validity of this instrument (Bernstein & Gesn, 1996; Tremblay & Ewart, 2005). It is a 27-item, 7-point scale instrument ranging from 1 (extremely uncharacteristic of me) to 7 (extremely characteristic of me). It measures four aggression-related dimensions: physical aggression, verbal aggression, hostility, and anger (Bernstein & Gesn, 1996).
Results
A total of 175 completed responses were received and analyzed. Of these, 133 (76%) were males and 42 (24%) were females. Respondents ranged from 18 to 40 years of age with 95% being between 18 and 25 years old. Respondents were from the College of Liberal Arts and the School of Interactive Games and Media at the Rochester Institute of Technology.
The participants were provided with a set of categories of video games: action, action role–playing, action adventure, adventure, fighting, role-playing, shooter, sports, and other. The most common type of video game among with respondents was action role–playing (82.8%) with role-playing (80%) and action adventure (78.8%) not far behind in popularity.
To determine whether there was a significant difference in levels of aggression in relation to sex, the study employed an independent-samples t test. No significant difference was found between men and women who play video games (t = 0.731, degrees of freedom [df] = 173, p = .602). Thus, there was no support for the hypothesis.
Levels of aggression were compared between playing and nonplaying respondents regarding types of video games. Significant differences were found with both the action role–playing (t = −1.48, df = 175, p = .01) and the role-playing video games (t = −1.24, df = 175, p = .02). Players had lower levels of aggression than nonplayers. Other types of video games were tested, but no significant differences were found. Even though action adventure, adventure, fighting, shooter, sports, and other types of video games did not report significant differences, sports sports (t = .595, df = 175, p = 0.59) and shooter (t = −1.420, df = 175, p = .063) video games approached significance. However, sports video games players had higher levels of aggression than nonplayers while shooter video games players had lower levels of aggression. This finding suggests that these differences could become statistically significant if there were a larger sample of respondents. Future research should test these relationships.
And finally, a Pearson correlation test was employed to determine whether aggression and length of video game playing are related. The Pearson correlation was most suitable because the length of video games played by respondents was interval data and they were not confined to a particular range of hours but had the freedom to report their playing hours. No statistically significant relationship was found between aggression and length of video game playing (r = −.05, p = .52).
Discussion
More than 75% of the respondents were males. A large number of them played action role–playing and role-playing video games (82% and 79%, respectively). This study revealed that the respondents spend an average of 7 hours per week playing video games. There were no significant differences in the levels of aggression found in males to those in females, counter to what was previously reported (Lucas & Sherry, 2004). These findings are important because they show that despite the males being heavier users of video games, there are no significant differences in their levels of aggression when compared to females.
The most popular game listed by the respondents was Halo (17.7%) followed by Call of Duty (14.2%) and World of Warcraft (12%). All of these video games fall in the action role–playing and role-playing category, and they involve violent activities in the sense that the player typically spends a considerable amount of time destroying other creatures. Overall, participants reported playing video games progressively less from 9th and 10th grades to their current playing habits in the past month. The respondents reported playing video games on average of 11.82 hours per week while in the 9th and 10th grades, 11.85 hours per week while in 11th and 12th grades, and 7.03 hours per week during the recent month.
With regard to the difference in levels of aggression and type of video games played, the only significant findings concerned action role playing (p = .01) and role-playing (p = .02) video games. In both cases, players had lower levels of aggression than nonplayers, suggesting either a cathartic effect of playing or that these types of games attract less aggressive players.
Sherry (2007) suggested that playing video games longer may be associated with less aggression. Similarly Krcmar and Lachlan (2009) suggested that video game play may increase aggression but that this increase in aggression may occur during shorter game play. The results of the present study do not support either of these claims. There is no relationship between the length of time spent playing video games and aggression.
Ferguson et al. (2012) have suggested moral panic theory as an explanation for the overreaction on the part of certain elements within the public and their legislative representatives. “Moral panic” is a concept used to explain overreaction by society to a perceived threat to social values. Stanley Cohen (1972) has defined it as follows:
A condition, episode, person or group of persons emerges to become defined as a threat to societal values and interests; its nature is presented in a stylized and stereotypical fashion by the mass media; the moral barricades are manned by editors, bishops, politicians and other right-thinking people; socially accredited experts pronounce their diagnoses and solutions; ways of coping are evolved or (more often) resorted to; the condition then disappears, submerges or deteriorates and becomes more visible. (p. 1)
Although first used to explain a disturbance caused by the mods and rockers of the early sixties, a recent update cites more current examples including high school massacres and shooting sprees and computer games and Internet porn (Cohen, 2011).
Cohen (2002) notes that in order to classify an event as a moral panic, the following elements must be included: concern with a potential or imagined threat, hostility toward or disapproval of actions taken, consensus among a majority of people that the actions are inappropriate, disproportional response to the perception of harm than is really called for, and volatility of the fear and panic. It appears the VVG controversy qualifies since it contains every element. Fueled by a volatile mixture of social science research and political righteous indignation, public fears continue.
The uses and gratifications theory holds that audiences choose their media based on needs they want gratified. In this study, it might explain why players choosing to play role playing games had lower levels of aggression. Players who need to take the place of someone else might be less aggressive since they seek video games that (virtually) place them in someone else’s shoes. Although there were no significant differences between sports and shooter games, they approached significance. It may be that sports games are more competitive and hence attract more aggressive players. Aggressive players with a heightened need to compete would likely seek out competitive games. Single-shooter games attracted players with lower levels of aggression. It may be that these games are less competitive since they do not have opponents and attract less competitive and aggressive players. Paraphrasing Schramm et al. (1961), it is the players who are most active. It is they who use video games rather than video games using them.
Footnotes
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
