CYBERPSYCHOLOGY, BEHAVIOR, AND SOCIAL NETWORKING Volume 17, Number 4, 2014 ª Mary Ann Liebert, Inc. DOI: 10.1089/cyber.2013.0245
Who Believes Electronic Games Cause Real World Aggression? Andrew K. Przybylski, PhD
Abstract
Electronic games have rapidly become a popular form of human recreation, and the immersive experiences they provide millions have led many to voice concerns that some games, and violent ones in particular, may negatively impact society. Increasingly heated debates make it clear that gaming-related aggression is a topic that elicits strong opinions. Despite a complex and growing literature concerned with violent games, little is known empirically about why some ardently believe, whereas others dismiss, notions that this form of leisure is a source of aggression. The present research recruited three nationally representative samples to investigate this understudied topic. Results showed that belief was normally distributed across the population, prominent among demographic cohorts who did not grow up with games and those who lack concrete gaming experience. Results are discussed in the context of this developing research area, wider social science perspectives, and the place of electronic games in society. Introduction
E
lectronic games are becoming a dominant medium for human recreation.1 The widespread rise of gaming as a mainstream leisure activity has driven many to fear that games might have negative psychosocial effects on players.2 Indeed, it makes intuitive sense that the immersive and interactive nature of games could be more influential than familiar forms of media entertainment such as music, books, or films. Seizing on this intuition, a literature has developed in recent years focusing on concerns that games might drive aggression. Despite growing interest, there is little consensus between researchers surrounding the means and extent to which gaming is linked to real world aggression. Divergent results across a host of correlational,3,4 experimental,5,6 longitudinal,7,8 and meta-analytic studies9,10 serve to underscore the equivocal nature of investigations considering whether violent games are a source of real world aggression. The absence of scientific agreement about the influences of electronic games has not hindered many in the public from forming and acting on beliefs about gaming-related aggression. During the 1990s, electronic games came under scrutiny as concerned parents and media reports expressed fears that games produce real world aggression.2 Many U.S. state and federal lawmakers echoed and amplified these sentiments, and drafted legislation aimed at protecting the public from the perceived danger of electronic games.11 Based on the conviction that games meaningfully contribute to societal violence, laws restricting the sale of mature and objectionable
games made their way into the legal system. However, these policy maker initiatives proved as ambivalent as social science research focused on games. In June 2011, the Supreme Court of the United States (SCOTUS) rebuffed efforts to regulate objectionable electronic games. Crossing a boundary between politics and social science, SCOTUS ruled that games enjoy free speech protection.12 The majority of U.S. Supreme Court justices found the legal and scientific arguments for laws to protect society from the perceived harm of games unpersuasive. Since then, a series of mass shootings have rekindled the question in people’s minds. Heightened public concerns reacting to these events are informing future attempts to study and regulate games. The tendency of many to believe games drive such massacres underlines a more fundamental question concerning the nature of popular beliefs about play: who believes electronic games cause real world aggression? Given that arguments can and have been articulated on both sides of the issue, it is noteworthy that little empirical research has specifically investigated how group and individual-level factors relate to people’s belief in gaming-related aggression. The purpose of the present research was to investigate empirically this understudied aspect of the wider debates surrounding electronic games. Existing research on beliefs in gaming-related aggression Two existing studies—a laboratory-based experiment and a survey-based poll—provide a frame for social science
Oxford Internet Institute, University of Oxford, Oxford, United Kingdom.
228
BELIEF IN GAMING-RELATED AGGRESSION research examining belief in gaming-related aggression. The first of these studies by Ivory and Kalyanaraman investigated how different modes of reflection on gaming experiences related to beliefs about the effects games exert.13 In their experiment, the researchers manipulated concrete versus abstract thinking about violent games, and assessed how these mindsets influenced participants’ belief in gaming-related aggression and support for censoring violent games. Results showed that participants instructed to think about a specific violent game they were familiar with were less likely to believe games promoted aggression. Further, those who thought concretely about a game were also less likely to support steps aimed at limiting access to or censoring objectionable games compared to those who considered games in the abstract. The researchers concluded that thinking about a violent game may increase the availability of thoughts regarding a link, but reflecting on the specific features of any given game undermines a person’s ability to generate compelling rationales to support the banning of objectionable games. Second, recent polling research probed the prevalence of belief in gaming-related aggression across the general population.14 Among a battery of other topics, the study asked a representative sample of American adults to rate the extent to which they believed that there is a link between playing violent games and teenagers showing violent behavior. Findings indicated that the nearly three in five American adults agree or strongly agree with this idea. The report’s descriptive statistics suggested that the views of older Americans potentially undergirded this trend. Nearly three-quarters of older people endorsed a belief in gaming-related aggression, whereas less than half of younger adults reported thinking violent games lead teens to behave aggressively. Taken together, existing research suggests that popular perceptions of gaming-related aggression vary as a function of whether people have had, or can draw on, concrete personal experiences with electronic games. Both lab-based and polling research indicate that depending on abstract knowledge of games, or thinking about games in purely theoretical terms, is associated with believing they contribute to real world violence. Present research The present studies were designed to examine empirically the nature of belief in a causal link between electronic gaming and real world aggression. These studies recruited three nationally representative U.S. samples to address three research questions. The first two studies investigated how demographic variability related to having experience with electronic games (study 1) and belief in gaming-related aggression (study 2). By examining broad trends, these studies establish baselines for both gaming experience and belief in gaming-related aggression across the general population. Study 1 tested the hypotheses that older people, a cohort who grew up in a time before electronic games, may tend to have less concrete experience with games. Study 2 evaluated the predication that this cohort of older adults would be more likely to believe in a causal link between games and violence. Finally, study 3 shifted focus from broad demographic trends to investigate how individuals’ experiences with electronic games relate to believing in gamingrelated aggression. The third study tested the hypothesis that
229 people who have, and can draw on, concrete experiences with games are less likely to believe they cause aggression compared to those who have an abstract understanding of games. Study 1: Concrete Experience with Electronic Games The first study investigated the extent to which recent firsthand experience with electronic games varies across the general population. It tested the hypothesis that older participants, a cohort who grew up in a time before games were a popular recreation activity, would be less likely to have experience with electronic gaming contexts. Method Data collected for the present studies were gathered using Google Consumer Surveys (GCS). Validation work by the Pew Research Center15 and Google16 indicates GCS produces highly accurate results in line with other probability-based panel survey approaches. Importantly, GCS demonstrates substantially higher response rates (15–20%) compared to sampling rates observed industry wide (0.1–2%) across a range of polling topics. This is because the GCS method puts less burden on potential respondents.16 All analyses presented in the present studies were conducted twice: once using population-level weighting based on U.S. Census Current Population Survey data, and once without. Because the direction, magnitude, and significance of relations remained unchanged between these parallel analyses, only the simpler unweighted results are presented in the interests of brevity. Participant demographics. A nationally representative U.S. sample of 1,003 adults was recruited for this study. Of these, 50.2% were identified as female, and participants were coded into one of six age cohorts: 18–24 years (12%), 25–34 years (19.1%), 35–44 years (14.7%), 45–54 years (21.5%), 55–64 years (16.4%), and 65 years or older (15.3%). Recent gaming experience. Concrete experience with electronic games was assessed using a single item measure that asked participants: ‘‘How frequently do you play video/ computer games?’’ This question was paired with a 5-point Likert-style scale participants used to respond: 1 = ‘‘never/ almost never,’’ 2 = ‘‘once a year or less,’’ 3 = ‘‘several times a year,’’ 4 = ‘‘at least once a week,’’ or 5 = ‘‘most days.’’ Fifty-five percent of the sample reported never or almost never playing electronic games, 5.6% reported play once a year or less, 10.2% play several times a year, 11.4% played at least once a week, and 17.8% of participants said they played most days. Analyses A hierarchical regression model evaluated the unique and interacting effects linking age cohort and gender to gaming experience. This model regressed gaming experience on age and gender in the first step of the model, and an interaction term (the product of centered age and gender scores) in the second step. Results indicated the overall model accounted for 2.4% of variability in gaming experience. Age cohort was a significant predictor of experience (b = - 0.21, p < 0.001), whereas gender was not (b = - 0.04, p = 0.18). Results showed
230
PRZYBYLSKI
the interaction term (age by gender) was a significant predictor (b = 0.17, p < 0.001; the pattern of this interaction is presented in Fig. 1). Subsequent simple slopes analyses indicated that the trend was significant and negative for men (b = - 0.20, p < 0.001) and not significant for women (b = 0.02, p = 0.66). In other words, the relation linking age cohorts to gaming depended on participant gender. Younger men tended to have more experience with electronic games compared to their older counterparts, whereas no such pattern was in evidence for women. To contextualize these effects, a logistic regression model focused on men was evaluated, and a discrete gaming experience variable was created. Those who played once a year or less (57.7% of the males in the sample) were judged to be inexperienced with games and coded as 0. Those who reported playing several times a year or more (42.3% of this sample) were considered to have concrete experience with games and coded as 1. The logistic model regressing discrete gaming experience onto age accounted for 3.9% of the observed variability (Nagelkerke R2) in the outcome, showing a negative relation between the factors (B = - 0.23, SE = 0.062, p < 0.001). To put these significant trends in context, the following equation was used to compute the log odds ratio relating age to recent experience of games: Odds Ratio ¼ e(B Score Differences) Compared to their older counterparts (q65 years), younger males (18–24 years) were three times more likely (3.08 · ) to have recent experience playing electronic games.
Study 2: Belief in Gaming-Related Aggression The second study examined the extent to which belief in gaming-related aggression varies across the general population. In particular, this study evaluated the hypothesis that those belonging to older demographic cohorts, as group Study 1 indicated, have a less concrete experiential basis to evaluate games, and would be more likely to believe in a link between electronic games and real world aggression. Method Participant demographics. A second nationally representative U.S. sample of 1,000 adults (49.6% identified as female) was recruited for this study. Participants were coded into one of six age cohorts: 18–24 years (11.6%), 25–34 years (18.3%), 35–44 years (15%), 45–54 years (23.6%), 55–64 years (17.5%), and q65 years (14%). Belief in a gaming–aggression link. Judgments regarding the relation between electronic games and real world aggression were measured using a single item that asked participants to rate their level of agreement with the statement: ‘‘Violent video/computer games cause real world aggression.’’ Participants responded using a 5-point Likert-style scale with anchors labeled: 1 = ‘‘strongly disagree,’’ 2 ‘‘somewhat disagree,’’ 3 ‘‘neither agree or disagree,’’ 4 ‘‘somewhat agree,’’ or 5 ‘‘strongly agree.’’ Overall scores ranged from 1 to 5, M = 3.00, SD = 1.37, skewness = - 0.13, and kurtosis = 1.16. Results from a Kolmogorov–Smirnov test—D(1000) = 5.15, p < 0.001— indicated that belief in a gaming–aggression link was relatively normally distributed across the general population. Analyses
FIG. 1. Moderating effect of participant gender on the relation between age cohorts and experience with electronic games (study 1).
A hierarchical regression model was created to evaluate the unique and interacting effects of age and gender on belief in the existence of gaming-related aggression. This model regressed participants’ belief onto age cohort and gender in the first step of the model and an interaction term in the second step (the product of centered age and gender scores). Results indicated the overall model accounted for a combined 6.8% of the variability (R2) in belief. In particular, participant age (b = 0.23, p < 0.001) and gender (b = 0.09, p = 0.005) showed significant main effects, whereas the interaction term trended toward, yet did not attain, statistical significance (b = 0.08, p = 0.06). To aide in the interpretation of these main effects, a multiple logistic regression model was evaluated. Participants’ belief in a gaming–aggression link was recoded into a new discrete variable. Participants (27%) who neither agreed nor disagreed with a gaming–aggression link were dropped from this analysis. Those who did take a stance on the issue were grouped: people who somewhat or strongly disagreed were coded as 0 (33.7%), and those who somewhat or strongly agreed were coded as 1 (39.3%). The logistic model regressed the discrete belief variable simultaneously on participant age (B = 0.34, SE = 0.05, p < 0.001) and gender (B = 0.37, SE = 0.16, p = 0.022). These factors accounted for 10.3% of the observed variability in evaluations of a gaming–aggression link. To put these significant trends in context, log odds ratios relating gender and age to beliefs about violent games were computed. Compared to younger adults (18–24 years), older
BELIEF IN GAMING-RELATED AGGRESSION Americans (q65 years) were almost five and a half times more likely (5.47 ·) to believe that gaming influenced real world aggression. Results showed a similar yet smaller effect based on gender. Compared to men, women were almost one and a half times more likely (1.45 ·) to believe that electronic games were a cause of real world violence. There were no demographic differences observed for those who neither agreed nor disagreed with a gaming–aggression link. Study 3: Gaming Experience and Belief in Gaming-Related Aggression The final study was conducted to integrate and extend the results derived in studies 1 and 2. Of central interest in this study was the relation between gaming experience and belief in gaming-related aggression on the individual level. This study tested the hypothesis that people who have recent experiences of games are less likely to endorse belief in gamingrelated aggression compared to those who have only abstract knowledge concerning games. Method Participant demographics. A nationally representative U.S. sample of 501 adults was recruited for this study. Of these, 52.3% of participants were identified as female, and participants were coded into one of six age cohorts: 18–24 years (10.5%), 25–34 years (17.6%), 35–44 years (17.2%), 45–54 years (20.7%), 55–64 years (20.7%), and q65 years (13.4%). Recent gaming experience. Following the approach used in study 1, concrete experience with games was assessed using the single-item measure that asked participants: ‘‘How frequently do you play video/computer games?’’ In response to this, 55.7% of the sample reported never or almost never playing, 5.8% reported play once a year or less, 8% played several times a year, 11.6% played at least once a week, and 19% said they played most days.
231 individual-level relation between gaming experience and belief in gaming-related aggression. The first model examining age did not show a moderation effect (b = 0.07, p = 0.118). The second model evaluating gender showed this factor significantly moderated the link between recent experience and belief in gaming-related aggression (b = 0.17, p = 0.006). The shape of this interaction is presented in Figure 2. Simple slopes analyses indicated that the trend was significant and negative for males (b = - 0.27, p < 0.001), and was not significant for females (b = 0.01, p = 0.862). A logistic model examining belief in a gaming–aggression link among male participants (B = - 0.32, SE = 0.092, p < 0.001) indicated that men who never or almost never played electronic games were three and a half times more likely (3.52 ·) to believe that a link exists between gaming and real world aggression exists than their counterparts who reported gaming most days. Testing replications of effects. The data collected for study 3 provided the opportunity to retest the relations uncovered in studies 1 and 2. Overall correlations are presented in Table 1. Two hierarchical regression models evaluated the unique and interacting effects linking participant age and gender to belief in gaming-related aggression and to gaming experience. The model accounted for 7.4% of belief and showed age cohort (b = 0.28, p < 0.001) and gender (b = 0.10, p = 0.022) predicted belief. The interaction term was not significant (b = - 0.07, p = 0.24). The model predicting recent gaming experience accounted for 4.6% of the outcome’s variance and showed participant age (b = - 0.31, p < 0.001) but not gender (b = - 0.01, p = 0.79) were significant predictors. A
Belief in a gaming–aggression link. Just as in study 2, participants rated their agreement with the statement: ‘‘Violent video/computer games cause real world aggression.’’ Overall judgments about the relation between gaming and aggression ranged from 1 to 5, and were relatively normally distributed, D(501) = 4.22, p < 0.001, M = 3.15, SD = 1.40, skewness = - 0.26, and kurtosis = - 1.20. Analysis Experience and belief in gaming-related aggression. To test the main hypothesis that those with experience of games are less likely to think there is a link between gaming and aggression, the latter was regressed on the former (b = - 0.11, p = 0.012). To contextualize this relation, a subsequent logistic model regressed discrete belief onto gaming experience (B = - 0.13, SE = 0.062, p = 0.032), following the approach used in study 1. Overall, those who never or almost never played electronic games were nearly twice as likely (1.7 ·) to believe in a link between gaming and aggression compared to those who played games most days. Expanding on the general demographic trends observed in studies 1 and 2, two hierarchical regression models evaluated the extent to these factors served as moderators of the
FIG. 2. Moderating effect of participant gender on the relation between experience with games and belief in gamingrelated aggression (study 3).
232
PRZYBYLSKI Table 1. Observed Variables in Study 3 1
2
3
1. Age — 2. Gender 0.12** — 3. Belief in gaming-related 0.25*** 0.13** — aggression 4. Recent experience with games - 0.15*** - 0.03 - 0.11* *p < 0.05; **p < 0.01; ***p < 0.001.
significant interaction term qualified the age cohort effect (b = 0.22, p < 0.001). Simple slopes analyses indicated that the trend relating age to gaming was negative and significant for males (b = - 0.30, p < 0.001) and flat for females. The pattern of this interaction closely followed the plot observed for study 2 data. Logistic regression models following those used in studies 1 and 2 indicated that: (a) older people (q65 years) were six times more likely (6.14 ·) than younger adults (18–24 years) to believe games influence aggression; (b) women were almost twice as likely (1.89 ·) as men to believe gaming contributed to real world aggression; and (c) younger men (18–24 years) were almost six times more likely (5.74 ·) than their older counterparts (q65 years) to have concrete gaming experience. These findings fully replicated results derived in studies 1 and 2. There were no demographic differences observed for those who neither agreed nor disagreed with a gaming–aggression link. Discussion At present, lively debates concerning the extent and mechanisms through which electronic games might foment aggression are active in popular, political, and scientific circles. Many are concerned that violent games might constitute a source of real world aggression. Yet, despite these fears, the scientific study of gaming-related aggression is currently at loggerheads, as are legislative efforts to regulate and censor games.17 The present research was not meant to resolve these impasses. Instead, three studies were developed to build a clearer picture of the nature of belief in gaming-related aggression proper. Results derived in the present research indicate that belief in a gaming–aggression link is not uniform across demographic cohorts (see Fig. 3). Results showed that older Americans, a demographic group who reported little experience with games, tended to be between four and six times more likely to believe that electronic games contributed to human aggression compared to younger adults. Analyses showed a similar, yet smaller, demographic effect for American women, a group that tends to be less familiar with violent games.18 These significant demographic trends laid a foundation for investigating belief in gaming-related aggression on the individual level. Evidence showed that people who reported little recent gaming experience were nearly twice as likely to believe in a causal relation between games and violence. This finding was amplified by gender: American men who had an abstract experiential basis for considering games were between three and four times more likely to believe in a link compared those who had concrete experience to draw on.
FIG. 3. Comparison of recent experience with electronic games (study 1) versus belief in gaming-related aggression (study 2) across age cohorts.
Relations with existing literatures Results derived from the present research conceptually replicate and extend findings reported in recent experimental13 and polling research14 concerning beliefs about gamingrelated aggression. In line with this earlier work, evidence suggested the ability to draw on concrete experience with games was associated with skepticism that games influence human aggression. Mirroring figures presented in polling research, findings indicated that generational and personal trends in experience were likewise negatively associated with belief in gaming-related aggression. Research examining experience with specific gaming genres, such as online roleplaying games, may be similarly informative regarding popular concerns about problematic patterns of game engagement.19 The present findings also interface with experimental studies by Kneer et al. that suggest young people are uniform in their automatic associations regarding electronic games.20,21 The authors found nonplayers and regular players do not vary in their reflexive evaluations of violent games; they are equally liable to associate games with social interaction and show less pronounced associations between games and
BELIEF IN GAMING-RELATED AGGRESSION aggression. This pattern of results suggests that the variability observed in the present studies was not due to individual defensiveness or social identity bias. That said, research meant to encourage dynamic, open, and critical processing of thoughts, feelings, and beliefs surrounding games is needed.22 The results also speak to broader social science research framing societies’ tendency to be apprehensive about new forms of entertainment.23 Viewed from sociological and criminological perspectives, the trends uncovered in the present work lend empirical weight to the notion that electronic games are presently the focus of a moral panic—an exaggerated perception that an activity, group, or person is deviant and poses a danger to wider society.24,25 Results from the present studies indicate that those who grew up in a world without electronic games, or lack experience with games, are most likely to endorse the idea that violent games are a material source of society’s ills. Further, they highlight the potential for confirmation bias in how the public construes a link between games and violence. In cases where shooters are young males, evidence of violent gaming is often pursued in news narratives, whereas electronic games are not considered as the cause of mass shootings perpetrated by women or older men.26 These findings may also provide an explanation for the unfortunate tendency of some media reports to misrepresent the nature and content of popular electronic games27 and foolheartedly feed fears about their negative effects.28 This kind of media coverage may be inaccurate and carry an alarmist tone precisely because it is not informed by firsthand experience with games.
233 widespread calls for a paradigm shift in the study of gamingrelated aggression.29 Closing Remarks Historically, the rise of novel forms of entertainment has been met with concerns that the new mediums themselves might negatively impact society as a whole. At present, electronic games are at the nexus of popular, legislative, and scientific debates. Whether electronic games are on balance a good or bad thing remains an open question, but is a topic about which people hold strong opinions. The present research sought to understand a wider aspect of conversation: what factors relate to believing in versus discounting a causal link between electronic games and aggression across the general population? The work presented herein provides a first step to understanding why they are held. How these factors and others yet to be studied relate to popular apprehensions about games will certainly be the focus of future research as this medium for human play becomes evermore ubiquitous. Acknowledgment The author would like to thank Albert H. Robinson for his invaluable contributions to this research project. Author Disclosure Statement No competing financial interests exist. References
Limitations The present studies present a number of limitations that provide rich avenues for future research. First, the approach used to quantify participants’ levels of experience in the first and third studies was not optimal. This measure did not differentiate between participants’ experience of different kinds of games. Future studies comparing how experience with competitive, cooperative, violent, and nonviolent modes of play relate to belief in gaming-related aggression would greatly extend the present work. Second, results from the second and third studies have little to say about those who hold no firm view regarding gaming-related aggression. The focus of the present studies concerned the factors that predicted discounting versus buying into a gaming–aggression link. Future research examining what leads between a quarter and a third of the general population to have no defined opinion on the relationship between games and aggression would be illuminating. Finally, these studies are not informative about shifts in attitudes among subpopulations particularly relevant to the gaming-related aggression debate such as parents, policy makers, and social science researchers. It is possible that the presently heated nature of arguments surrounding games are being stoked by an influx of skeptical young adults whose lived experience of gaming runs directly counter to the commonsense notion that games cause aggression. Future studies may shed empirical light on whether higher levels of firsthand gaming experience among a new generation of social science researchers undergirds the increasingly vocal and
1. Entertainment Software Association. (2012) Essential facts about the computer and video game industry. www.theesa .com/facts/pdfs/ESA_EF_2012.pdf (accessed July 1, 2013). 2. Kent SL. (2001) The ultimate history of video games. New York: Three Rivers Press. 3. Anderson CA, Gentile DA, Buckley KE. (2007) Violent video game effects on children and adolescents: theory, research, and public policy. New York: Oxford University Press. 4. Olson CK, Kutner LA, Baer L, et al. M-rated video games and aggressive or problem behavior among young adolescents. Applied Developmental Science 2009; 13:188–198. 5. Anderson CA, Carnagey NL, Flanagan M, et al. Violent video games: specific effects of violent content on aggressive thoughts and behavior. Advances in Experimental Social Psychology 2004; 36:200–251. 6. Adachi PJC, Willoughby T. The effect of video game competition and violence on aggressive behavior: which characteristic has the greatest influence? Psychology of Violence 2011; 1:259–274. 7. Anderson, CA, Sakamoto A, Gentile DA, et al. Longitudinal effects of violent video games on aggression in Japan and the United States. Pediatrics 2008; 122:1067–1072. 8. Wallenius M, Punama¨ki R-L. Digital game violence and direct aggression in adolescence: a longitudinal study of the roles of sex, age, and parent-child communication. Journal of Applied Developmental Psychology 2008; 29:286–294. 9. Anderson CA, Shibuya A, Ihori N, et al. Violent video game effects on aggression, empathy, and prosocial behavior in eastern and western countries: a meta-analytic review. Psychological Bulletin 2010; 136:151–173. 10. Sherry J. (2007) Violent video games and aggression: why can’t we find links? In Preiss R, Gayle B, Burrell N, Allen M,
234
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
PRZYBYLSKI Bryant J, eds. Mass media effects research: advances through meta-analysis. Mahwah, NJ: Lawrence Erlbaum, pp. 231–248. Denniston L. (2010) Argument preview: kids and video games: SCOTUS blog. www.scotusblog.com/2010/10/argumentpreview-kids-and-video-games/ (accessed July 1, 2013). Brown v. Entertainment Merchants Association. (2011) www.supremecourt.gov/opinions/10pdf/08-1448.pdf (accessed July 1, 2013). Ivory JD, Kalyanaraman S. Video games make people violent—well, maybe not that game: effects of content and person abstraction on perceptions of violent video games’ effects and support of censorship. Communication Reports 2009; 22:1–12. Harris Interactive. (2013) Majority of Americans see connection between video games and violent behavior in teens. www.harrisinteractive.com/vault/Harris%20Poll%2010%20%20Video%20Games_2%2027%2013.pdf (accessed July 1, 2013). Keeter S, Christian L. (2012) A comparison of results from surveys by the Pew Research Center and Google consumer surveys. www.people-press.org/2012/11/07/a-comparisonof-results-from-surveys-by-the-pew-research-center-and-googleconsumer-surveys/ (accessed July 1, 2013). McDonald P, Mohebbi M, Slatkin B. (2012) Comparing Google consumer surveys to existing probability and nonprobability based Internet surveys. www.google.com/insights/consumersurveys/static/consumer_surveys_white paper.pdf (accessed July 1, 2013). Elson M, Ferguson CJ. Twenty-five years of research on violence in digital games and aggression: empirical evidence, perspectives, and a debate gone astray. European Psychologist 2013; in press. Lucas K, Sherry JL. Sex differences in video game play: a communication-based explanation. Communication Research 2004; 31:499–523. Peters CS, Malesky AM. Problematic usage among highlyengaged players of massively multiplayer online role playing games. CyberPsychology & Behavior 2008; 4:481–484. Kneer J, Glock S, Beskes S, et al. Are digital games perceived as fun or danger? Supporting and suppressing different game-related concepts. CyberPsychology, Behavior, & Social Networking 2012; 15:251–256.
21. Kneer J, Munko D, Glock S, et al. Defending the doomed: implicit strategies concerning protection of first person shooter games. CyberPsychology, Behavior, & Social Networking 2012; 15:151–156. 22. Kutner LA, Olson CK, Warner DE, et al. Parents’ and sons’ perspectives on video game play: a qualitative study. Journal of Adolescent Research 2008; 23:76–96. 23. Cohen S. (2002) Folk devils and moral panics: creation of mods and rockers. London: Routledge. 24. Ferguson CJ. The school/violent game link: causal relationship or moral panic? Journal of Investigative & Offender Profiling 2008; 5:25–37. 25. Griffiths DN. (2012) Modern warfare and moral panics—a very old pattern resurfaces. Forbes Magazine. www.for bes.com/sites/danielnyegriffiths/2012/04/20/modern-war fare-and-moral-panics-a-very-old-pattern-resurfaces/ (accessed July 1, 2013). 26. Schiesel S. (2008) Author faults a game, and gamers flame back. The New York Times. www.nytimes.com/2008/01/ 26/arts/television/26mass.html (accessed July 1, 2013). 27. Ferguson CJ. (2013) Don’t blame video games for real-world violence: the chronicle of higher education. http://chronicle .com/blogs/conversation/2013/01/10/dont-blame-videogames-for-real-world-violence/ (accessed July 1, 2013). 28. Person C. (2013) Breaking down the absurd anatomy of a video game scare piece: Kotaku. http://kotaku.com/breakingdown-the-absurd-anatomy-of-a-video-game-scare-487437560 (accessed July, 1, 2013). 29. Ferguson CJ, Dyck D. Paradigm change in aggression research: the time has come to retire the General Aggression Model. Aggression & Violent Behavior 2012; 17:220–228.
Address correspondence to: Dr. Andrew K. Przybylski Oxford Internet Institute University of Oxford 1 St. Giles Oxford, OX1 3JS United Kingdom E-mail: [email protected]
Who Believes Electronic Games Cause Real World Aggression? Andrew K. Przybylski, PhD
Abstract
Electronic games have rapidly become a popular form of human recreation, and the immersive experiences they provide millions have led many to voice concerns that some games, and violent ones in particular, may negatively impact society. Increasingly heated debates make it clear that gaming-related aggression is a topic that elicits strong opinions. Despite a complex and growing literature concerned with violent games, little is known empirically about why some ardently believe, whereas others dismiss, notions that this form of leisure is a source of aggression. The present research recruited three nationally representative samples to investigate this understudied topic. Results showed that belief was normally distributed across the population, prominent among demographic cohorts who did not grow up with games and those who lack concrete gaming experience. Results are discussed in the context of this developing research area, wider social science perspectives, and the place of electronic games in society. Introduction
E
lectronic games are becoming a dominant medium for human recreation.1 The widespread rise of gaming as a mainstream leisure activity has driven many to fear that games might have negative psychosocial effects on players.2 Indeed, it makes intuitive sense that the immersive and interactive nature of games could be more influential than familiar forms of media entertainment such as music, books, or films. Seizing on this intuition, a literature has developed in recent years focusing on concerns that games might drive aggression. Despite growing interest, there is little consensus between researchers surrounding the means and extent to which gaming is linked to real world aggression. Divergent results across a host of correlational,3,4 experimental,5,6 longitudinal,7,8 and meta-analytic studies9,10 serve to underscore the equivocal nature of investigations considering whether violent games are a source of real world aggression. The absence of scientific agreement about the influences of electronic games has not hindered many in the public from forming and acting on beliefs about gaming-related aggression. During the 1990s, electronic games came under scrutiny as concerned parents and media reports expressed fears that games produce real world aggression.2 Many U.S. state and federal lawmakers echoed and amplified these sentiments, and drafted legislation aimed at protecting the public from the perceived danger of electronic games.11 Based on the conviction that games meaningfully contribute to societal violence, laws restricting the sale of mature and objectionable
games made their way into the legal system. However, these policy maker initiatives proved as ambivalent as social science research focused on games. In June 2011, the Supreme Court of the United States (SCOTUS) rebuffed efforts to regulate objectionable electronic games. Crossing a boundary between politics and social science, SCOTUS ruled that games enjoy free speech protection.12 The majority of U.S. Supreme Court justices found the legal and scientific arguments for laws to protect society from the perceived harm of games unpersuasive. Since then, a series of mass shootings have rekindled the question in people’s minds. Heightened public concerns reacting to these events are informing future attempts to study and regulate games. The tendency of many to believe games drive such massacres underlines a more fundamental question concerning the nature of popular beliefs about play: who believes electronic games cause real world aggression? Given that arguments can and have been articulated on both sides of the issue, it is noteworthy that little empirical research has specifically investigated how group and individual-level factors relate to people’s belief in gaming-related aggression. The purpose of the present research was to investigate empirically this understudied aspect of the wider debates surrounding electronic games. Existing research on beliefs in gaming-related aggression Two existing studies—a laboratory-based experiment and a survey-based poll—provide a frame for social science
Oxford Internet Institute, University of Oxford, Oxford, United Kingdom.
228
BELIEF IN GAMING-RELATED AGGRESSION research examining belief in gaming-related aggression. The first of these studies by Ivory and Kalyanaraman investigated how different modes of reflection on gaming experiences related to beliefs about the effects games exert.13 In their experiment, the researchers manipulated concrete versus abstract thinking about violent games, and assessed how these mindsets influenced participants’ belief in gaming-related aggression and support for censoring violent games. Results showed that participants instructed to think about a specific violent game they were familiar with were less likely to believe games promoted aggression. Further, those who thought concretely about a game were also less likely to support steps aimed at limiting access to or censoring objectionable games compared to those who considered games in the abstract. The researchers concluded that thinking about a violent game may increase the availability of thoughts regarding a link, but reflecting on the specific features of any given game undermines a person’s ability to generate compelling rationales to support the banning of objectionable games. Second, recent polling research probed the prevalence of belief in gaming-related aggression across the general population.14 Among a battery of other topics, the study asked a representative sample of American adults to rate the extent to which they believed that there is a link between playing violent games and teenagers showing violent behavior. Findings indicated that the nearly three in five American adults agree or strongly agree with this idea. The report’s descriptive statistics suggested that the views of older Americans potentially undergirded this trend. Nearly three-quarters of older people endorsed a belief in gaming-related aggression, whereas less than half of younger adults reported thinking violent games lead teens to behave aggressively. Taken together, existing research suggests that popular perceptions of gaming-related aggression vary as a function of whether people have had, or can draw on, concrete personal experiences with electronic games. Both lab-based and polling research indicate that depending on abstract knowledge of games, or thinking about games in purely theoretical terms, is associated with believing they contribute to real world violence. Present research The present studies were designed to examine empirically the nature of belief in a causal link between electronic gaming and real world aggression. These studies recruited three nationally representative U.S. samples to address three research questions. The first two studies investigated how demographic variability related to having experience with electronic games (study 1) and belief in gaming-related aggression (study 2). By examining broad trends, these studies establish baselines for both gaming experience and belief in gaming-related aggression across the general population. Study 1 tested the hypotheses that older people, a cohort who grew up in a time before electronic games, may tend to have less concrete experience with games. Study 2 evaluated the predication that this cohort of older adults would be more likely to believe in a causal link between games and violence. Finally, study 3 shifted focus from broad demographic trends to investigate how individuals’ experiences with electronic games relate to believing in gamingrelated aggression. The third study tested the hypothesis that
229 people who have, and can draw on, concrete experiences with games are less likely to believe they cause aggression compared to those who have an abstract understanding of games. Study 1: Concrete Experience with Electronic Games The first study investigated the extent to which recent firsthand experience with electronic games varies across the general population. It tested the hypothesis that older participants, a cohort who grew up in a time before games were a popular recreation activity, would be less likely to have experience with electronic gaming contexts. Method Data collected for the present studies were gathered using Google Consumer Surveys (GCS). Validation work by the Pew Research Center15 and Google16 indicates GCS produces highly accurate results in line with other probability-based panel survey approaches. Importantly, GCS demonstrates substantially higher response rates (15–20%) compared to sampling rates observed industry wide (0.1–2%) across a range of polling topics. This is because the GCS method puts less burden on potential respondents.16 All analyses presented in the present studies were conducted twice: once using population-level weighting based on U.S. Census Current Population Survey data, and once without. Because the direction, magnitude, and significance of relations remained unchanged between these parallel analyses, only the simpler unweighted results are presented in the interests of brevity. Participant demographics. A nationally representative U.S. sample of 1,003 adults was recruited for this study. Of these, 50.2% were identified as female, and participants were coded into one of six age cohorts: 18–24 years (12%), 25–34 years (19.1%), 35–44 years (14.7%), 45–54 years (21.5%), 55–64 years (16.4%), and 65 years or older (15.3%). Recent gaming experience. Concrete experience with electronic games was assessed using a single item measure that asked participants: ‘‘How frequently do you play video/ computer games?’’ This question was paired with a 5-point Likert-style scale participants used to respond: 1 = ‘‘never/ almost never,’’ 2 = ‘‘once a year or less,’’ 3 = ‘‘several times a year,’’ 4 = ‘‘at least once a week,’’ or 5 = ‘‘most days.’’ Fifty-five percent of the sample reported never or almost never playing electronic games, 5.6% reported play once a year or less, 10.2% play several times a year, 11.4% played at least once a week, and 17.8% of participants said they played most days. Analyses A hierarchical regression model evaluated the unique and interacting effects linking age cohort and gender to gaming experience. This model regressed gaming experience on age and gender in the first step of the model, and an interaction term (the product of centered age and gender scores) in the second step. Results indicated the overall model accounted for 2.4% of variability in gaming experience. Age cohort was a significant predictor of experience (b = - 0.21, p < 0.001), whereas gender was not (b = - 0.04, p = 0.18). Results showed
230
PRZYBYLSKI
the interaction term (age by gender) was a significant predictor (b = 0.17, p < 0.001; the pattern of this interaction is presented in Fig. 1). Subsequent simple slopes analyses indicated that the trend was significant and negative for men (b = - 0.20, p < 0.001) and not significant for women (b = 0.02, p = 0.66). In other words, the relation linking age cohorts to gaming depended on participant gender. Younger men tended to have more experience with electronic games compared to their older counterparts, whereas no such pattern was in evidence for women. To contextualize these effects, a logistic regression model focused on men was evaluated, and a discrete gaming experience variable was created. Those who played once a year or less (57.7% of the males in the sample) were judged to be inexperienced with games and coded as 0. Those who reported playing several times a year or more (42.3% of this sample) were considered to have concrete experience with games and coded as 1. The logistic model regressing discrete gaming experience onto age accounted for 3.9% of the observed variability (Nagelkerke R2) in the outcome, showing a negative relation between the factors (B = - 0.23, SE = 0.062, p < 0.001). To put these significant trends in context, the following equation was used to compute the log odds ratio relating age to recent experience of games: Odds Ratio ¼ e(B Score Differences) Compared to their older counterparts (q65 years), younger males (18–24 years) were three times more likely (3.08 · ) to have recent experience playing electronic games.
Study 2: Belief in Gaming-Related Aggression The second study examined the extent to which belief in gaming-related aggression varies across the general population. In particular, this study evaluated the hypothesis that those belonging to older demographic cohorts, as group Study 1 indicated, have a less concrete experiential basis to evaluate games, and would be more likely to believe in a link between electronic games and real world aggression. Method Participant demographics. A second nationally representative U.S. sample of 1,000 adults (49.6% identified as female) was recruited for this study. Participants were coded into one of six age cohorts: 18–24 years (11.6%), 25–34 years (18.3%), 35–44 years (15%), 45–54 years (23.6%), 55–64 years (17.5%), and q65 years (14%). Belief in a gaming–aggression link. Judgments regarding the relation between electronic games and real world aggression were measured using a single item that asked participants to rate their level of agreement with the statement: ‘‘Violent video/computer games cause real world aggression.’’ Participants responded using a 5-point Likert-style scale with anchors labeled: 1 = ‘‘strongly disagree,’’ 2 ‘‘somewhat disagree,’’ 3 ‘‘neither agree or disagree,’’ 4 ‘‘somewhat agree,’’ or 5 ‘‘strongly agree.’’ Overall scores ranged from 1 to 5, M = 3.00, SD = 1.37, skewness = - 0.13, and kurtosis = 1.16. Results from a Kolmogorov–Smirnov test—D(1000) = 5.15, p < 0.001— indicated that belief in a gaming–aggression link was relatively normally distributed across the general population. Analyses
FIG. 1. Moderating effect of participant gender on the relation between age cohorts and experience with electronic games (study 1).
A hierarchical regression model was created to evaluate the unique and interacting effects of age and gender on belief in the existence of gaming-related aggression. This model regressed participants’ belief onto age cohort and gender in the first step of the model and an interaction term in the second step (the product of centered age and gender scores). Results indicated the overall model accounted for a combined 6.8% of the variability (R2) in belief. In particular, participant age (b = 0.23, p < 0.001) and gender (b = 0.09, p = 0.005) showed significant main effects, whereas the interaction term trended toward, yet did not attain, statistical significance (b = 0.08, p = 0.06). To aide in the interpretation of these main effects, a multiple logistic regression model was evaluated. Participants’ belief in a gaming–aggression link was recoded into a new discrete variable. Participants (27%) who neither agreed nor disagreed with a gaming–aggression link were dropped from this analysis. Those who did take a stance on the issue were grouped: people who somewhat or strongly disagreed were coded as 0 (33.7%), and those who somewhat or strongly agreed were coded as 1 (39.3%). The logistic model regressed the discrete belief variable simultaneously on participant age (B = 0.34, SE = 0.05, p < 0.001) and gender (B = 0.37, SE = 0.16, p = 0.022). These factors accounted for 10.3% of the observed variability in evaluations of a gaming–aggression link. To put these significant trends in context, log odds ratios relating gender and age to beliefs about violent games were computed. Compared to younger adults (18–24 years), older
BELIEF IN GAMING-RELATED AGGRESSION Americans (q65 years) were almost five and a half times more likely (5.47 ·) to believe that gaming influenced real world aggression. Results showed a similar yet smaller effect based on gender. Compared to men, women were almost one and a half times more likely (1.45 ·) to believe that electronic games were a cause of real world violence. There were no demographic differences observed for those who neither agreed nor disagreed with a gaming–aggression link. Study 3: Gaming Experience and Belief in Gaming-Related Aggression The final study was conducted to integrate and extend the results derived in studies 1 and 2. Of central interest in this study was the relation between gaming experience and belief in gaming-related aggression on the individual level. This study tested the hypothesis that people who have recent experiences of games are less likely to endorse belief in gamingrelated aggression compared to those who have only abstract knowledge concerning games. Method Participant demographics. A nationally representative U.S. sample of 501 adults was recruited for this study. Of these, 52.3% of participants were identified as female, and participants were coded into one of six age cohorts: 18–24 years (10.5%), 25–34 years (17.6%), 35–44 years (17.2%), 45–54 years (20.7%), 55–64 years (20.7%), and q65 years (13.4%). Recent gaming experience. Following the approach used in study 1, concrete experience with games was assessed using the single-item measure that asked participants: ‘‘How frequently do you play video/computer games?’’ In response to this, 55.7% of the sample reported never or almost never playing, 5.8% reported play once a year or less, 8% played several times a year, 11.6% played at least once a week, and 19% said they played most days.
231 individual-level relation between gaming experience and belief in gaming-related aggression. The first model examining age did not show a moderation effect (b = 0.07, p = 0.118). The second model evaluating gender showed this factor significantly moderated the link between recent experience and belief in gaming-related aggression (b = 0.17, p = 0.006). The shape of this interaction is presented in Figure 2. Simple slopes analyses indicated that the trend was significant and negative for males (b = - 0.27, p < 0.001), and was not significant for females (b = 0.01, p = 0.862). A logistic model examining belief in a gaming–aggression link among male participants (B = - 0.32, SE = 0.092, p < 0.001) indicated that men who never or almost never played electronic games were three and a half times more likely (3.52 ·) to believe that a link exists between gaming and real world aggression exists than their counterparts who reported gaming most days. Testing replications of effects. The data collected for study 3 provided the opportunity to retest the relations uncovered in studies 1 and 2. Overall correlations are presented in Table 1. Two hierarchical regression models evaluated the unique and interacting effects linking participant age and gender to belief in gaming-related aggression and to gaming experience. The model accounted for 7.4% of belief and showed age cohort (b = 0.28, p < 0.001) and gender (b = 0.10, p = 0.022) predicted belief. The interaction term was not significant (b = - 0.07, p = 0.24). The model predicting recent gaming experience accounted for 4.6% of the outcome’s variance and showed participant age (b = - 0.31, p < 0.001) but not gender (b = - 0.01, p = 0.79) were significant predictors. A
Belief in a gaming–aggression link. Just as in study 2, participants rated their agreement with the statement: ‘‘Violent video/computer games cause real world aggression.’’ Overall judgments about the relation between gaming and aggression ranged from 1 to 5, and were relatively normally distributed, D(501) = 4.22, p < 0.001, M = 3.15, SD = 1.40, skewness = - 0.26, and kurtosis = - 1.20. Analysis Experience and belief in gaming-related aggression. To test the main hypothesis that those with experience of games are less likely to think there is a link between gaming and aggression, the latter was regressed on the former (b = - 0.11, p = 0.012). To contextualize this relation, a subsequent logistic model regressed discrete belief onto gaming experience (B = - 0.13, SE = 0.062, p = 0.032), following the approach used in study 1. Overall, those who never or almost never played electronic games were nearly twice as likely (1.7 ·) to believe in a link between gaming and aggression compared to those who played games most days. Expanding on the general demographic trends observed in studies 1 and 2, two hierarchical regression models evaluated the extent to these factors served as moderators of the
FIG. 2. Moderating effect of participant gender on the relation between experience with games and belief in gamingrelated aggression (study 3).
232
PRZYBYLSKI Table 1. Observed Variables in Study 3 1
2
3
1. Age — 2. Gender 0.12** — 3. Belief in gaming-related 0.25*** 0.13** — aggression 4. Recent experience with games - 0.15*** - 0.03 - 0.11* *p < 0.05; **p < 0.01; ***p < 0.001.
significant interaction term qualified the age cohort effect (b = 0.22, p < 0.001). Simple slopes analyses indicated that the trend relating age to gaming was negative and significant for males (b = - 0.30, p < 0.001) and flat for females. The pattern of this interaction closely followed the plot observed for study 2 data. Logistic regression models following those used in studies 1 and 2 indicated that: (a) older people (q65 years) were six times more likely (6.14 ·) than younger adults (18–24 years) to believe games influence aggression; (b) women were almost twice as likely (1.89 ·) as men to believe gaming contributed to real world aggression; and (c) younger men (18–24 years) were almost six times more likely (5.74 ·) than their older counterparts (q65 years) to have concrete gaming experience. These findings fully replicated results derived in studies 1 and 2. There were no demographic differences observed for those who neither agreed nor disagreed with a gaming–aggression link. Discussion At present, lively debates concerning the extent and mechanisms through which electronic games might foment aggression are active in popular, political, and scientific circles. Many are concerned that violent games might constitute a source of real world aggression. Yet, despite these fears, the scientific study of gaming-related aggression is currently at loggerheads, as are legislative efforts to regulate and censor games.17 The present research was not meant to resolve these impasses. Instead, three studies were developed to build a clearer picture of the nature of belief in gaming-related aggression proper. Results derived in the present research indicate that belief in a gaming–aggression link is not uniform across demographic cohorts (see Fig. 3). Results showed that older Americans, a demographic group who reported little experience with games, tended to be between four and six times more likely to believe that electronic games contributed to human aggression compared to younger adults. Analyses showed a similar, yet smaller, demographic effect for American women, a group that tends to be less familiar with violent games.18 These significant demographic trends laid a foundation for investigating belief in gaming-related aggression on the individual level. Evidence showed that people who reported little recent gaming experience were nearly twice as likely to believe in a causal relation between games and violence. This finding was amplified by gender: American men who had an abstract experiential basis for considering games were between three and four times more likely to believe in a link compared those who had concrete experience to draw on.
FIG. 3. Comparison of recent experience with electronic games (study 1) versus belief in gaming-related aggression (study 2) across age cohorts.
Relations with existing literatures Results derived from the present research conceptually replicate and extend findings reported in recent experimental13 and polling research14 concerning beliefs about gamingrelated aggression. In line with this earlier work, evidence suggested the ability to draw on concrete experience with games was associated with skepticism that games influence human aggression. Mirroring figures presented in polling research, findings indicated that generational and personal trends in experience were likewise negatively associated with belief in gaming-related aggression. Research examining experience with specific gaming genres, such as online roleplaying games, may be similarly informative regarding popular concerns about problematic patterns of game engagement.19 The present findings also interface with experimental studies by Kneer et al. that suggest young people are uniform in their automatic associations regarding electronic games.20,21 The authors found nonplayers and regular players do not vary in their reflexive evaluations of violent games; they are equally liable to associate games with social interaction and show less pronounced associations between games and
BELIEF IN GAMING-RELATED AGGRESSION aggression. This pattern of results suggests that the variability observed in the present studies was not due to individual defensiveness or social identity bias. That said, research meant to encourage dynamic, open, and critical processing of thoughts, feelings, and beliefs surrounding games is needed.22 The results also speak to broader social science research framing societies’ tendency to be apprehensive about new forms of entertainment.23 Viewed from sociological and criminological perspectives, the trends uncovered in the present work lend empirical weight to the notion that electronic games are presently the focus of a moral panic—an exaggerated perception that an activity, group, or person is deviant and poses a danger to wider society.24,25 Results from the present studies indicate that those who grew up in a world without electronic games, or lack experience with games, are most likely to endorse the idea that violent games are a material source of society’s ills. Further, they highlight the potential for confirmation bias in how the public construes a link between games and violence. In cases where shooters are young males, evidence of violent gaming is often pursued in news narratives, whereas electronic games are not considered as the cause of mass shootings perpetrated by women or older men.26 These findings may also provide an explanation for the unfortunate tendency of some media reports to misrepresent the nature and content of popular electronic games27 and foolheartedly feed fears about their negative effects.28 This kind of media coverage may be inaccurate and carry an alarmist tone precisely because it is not informed by firsthand experience with games.
233 widespread calls for a paradigm shift in the study of gamingrelated aggression.29 Closing Remarks Historically, the rise of novel forms of entertainment has been met with concerns that the new mediums themselves might negatively impact society as a whole. At present, electronic games are at the nexus of popular, legislative, and scientific debates. Whether electronic games are on balance a good or bad thing remains an open question, but is a topic about which people hold strong opinions. The present research sought to understand a wider aspect of conversation: what factors relate to believing in versus discounting a causal link between electronic games and aggression across the general population? The work presented herein provides a first step to understanding why they are held. How these factors and others yet to be studied relate to popular apprehensions about games will certainly be the focus of future research as this medium for human play becomes evermore ubiquitous. Acknowledgment The author would like to thank Albert H. Robinson for his invaluable contributions to this research project. Author Disclosure Statement No competing financial interests exist. References
Limitations The present studies present a number of limitations that provide rich avenues for future research. First, the approach used to quantify participants’ levels of experience in the first and third studies was not optimal. This measure did not differentiate between participants’ experience of different kinds of games. Future studies comparing how experience with competitive, cooperative, violent, and nonviolent modes of play relate to belief in gaming-related aggression would greatly extend the present work. Second, results from the second and third studies have little to say about those who hold no firm view regarding gaming-related aggression. The focus of the present studies concerned the factors that predicted discounting versus buying into a gaming–aggression link. Future research examining what leads between a quarter and a third of the general population to have no defined opinion on the relationship between games and aggression would be illuminating. Finally, these studies are not informative about shifts in attitudes among subpopulations particularly relevant to the gaming-related aggression debate such as parents, policy makers, and social science researchers. It is possible that the presently heated nature of arguments surrounding games are being stoked by an influx of skeptical young adults whose lived experience of gaming runs directly counter to the commonsense notion that games cause aggression. Future studies may shed empirical light on whether higher levels of firsthand gaming experience among a new generation of social science researchers undergirds the increasingly vocal and
1. Entertainment Software Association. (2012) Essential facts about the computer and video game industry. www.theesa .com/facts/pdfs/ESA_EF_2012.pdf (accessed July 1, 2013). 2. Kent SL. (2001) The ultimate history of video games. New York: Three Rivers Press. 3. Anderson CA, Gentile DA, Buckley KE. (2007) Violent video game effects on children and adolescents: theory, research, and public policy. New York: Oxford University Press. 4. Olson CK, Kutner LA, Baer L, et al. M-rated video games and aggressive or problem behavior among young adolescents. Applied Developmental Science 2009; 13:188–198. 5. Anderson CA, Carnagey NL, Flanagan M, et al. Violent video games: specific effects of violent content on aggressive thoughts and behavior. Advances in Experimental Social Psychology 2004; 36:200–251. 6. Adachi PJC, Willoughby T. The effect of video game competition and violence on aggressive behavior: which characteristic has the greatest influence? Psychology of Violence 2011; 1:259–274. 7. Anderson, CA, Sakamoto A, Gentile DA, et al. Longitudinal effects of violent video games on aggression in Japan and the United States. Pediatrics 2008; 122:1067–1072. 8. Wallenius M, Punama¨ki R-L. Digital game violence and direct aggression in adolescence: a longitudinal study of the roles of sex, age, and parent-child communication. Journal of Applied Developmental Psychology 2008; 29:286–294. 9. Anderson CA, Shibuya A, Ihori N, et al. Violent video game effects on aggression, empathy, and prosocial behavior in eastern and western countries: a meta-analytic review. Psychological Bulletin 2010; 136:151–173. 10. Sherry J. (2007) Violent video games and aggression: why can’t we find links? In Preiss R, Gayle B, Burrell N, Allen M,
234
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
PRZYBYLSKI Bryant J, eds. Mass media effects research: advances through meta-analysis. Mahwah, NJ: Lawrence Erlbaum, pp. 231–248. Denniston L. (2010) Argument preview: kids and video games: SCOTUS blog. www.scotusblog.com/2010/10/argumentpreview-kids-and-video-games/ (accessed July 1, 2013). Brown v. Entertainment Merchants Association. (2011) www.supremecourt.gov/opinions/10pdf/08-1448.pdf (accessed July 1, 2013). Ivory JD, Kalyanaraman S. Video games make people violent—well, maybe not that game: effects of content and person abstraction on perceptions of violent video games’ effects and support of censorship. Communication Reports 2009; 22:1–12. Harris Interactive. (2013) Majority of Americans see connection between video games and violent behavior in teens. www.harrisinteractive.com/vault/Harris%20Poll%2010%20%20Video%20Games_2%2027%2013.pdf (accessed July 1, 2013). Keeter S, Christian L. (2012) A comparison of results from surveys by the Pew Research Center and Google consumer surveys. www.people-press.org/2012/11/07/a-comparisonof-results-from-surveys-by-the-pew-research-center-and-googleconsumer-surveys/ (accessed July 1, 2013). McDonald P, Mohebbi M, Slatkin B. (2012) Comparing Google consumer surveys to existing probability and nonprobability based Internet surveys. www.google.com/insights/consumersurveys/static/consumer_surveys_white paper.pdf (accessed July 1, 2013). Elson M, Ferguson CJ. Twenty-five years of research on violence in digital games and aggression: empirical evidence, perspectives, and a debate gone astray. European Psychologist 2013; in press. Lucas K, Sherry JL. Sex differences in video game play: a communication-based explanation. Communication Research 2004; 31:499–523. Peters CS, Malesky AM. Problematic usage among highlyengaged players of massively multiplayer online role playing games. CyberPsychology & Behavior 2008; 4:481–484. Kneer J, Glock S, Beskes S, et al. Are digital games perceived as fun or danger? Supporting and suppressing different game-related concepts. CyberPsychology, Behavior, & Social Networking 2012; 15:251–256.
21. Kneer J, Munko D, Glock S, et al. Defending the doomed: implicit strategies concerning protection of first person shooter games. CyberPsychology, Behavior, & Social Networking 2012; 15:151–156. 22. Kutner LA, Olson CK, Warner DE, et al. Parents’ and sons’ perspectives on video game play: a qualitative study. Journal of Adolescent Research 2008; 23:76–96. 23. Cohen S. (2002) Folk devils and moral panics: creation of mods and rockers. London: Routledge. 24. Ferguson CJ. The school/violent game link: causal relationship or moral panic? Journal of Investigative & Offender Profiling 2008; 5:25–37. 25. Griffiths DN. (2012) Modern warfare and moral panics—a very old pattern resurfaces. Forbes Magazine. www.for bes.com/sites/danielnyegriffiths/2012/04/20/modern-war fare-and-moral-panics-a-very-old-pattern-resurfaces/ (accessed July 1, 2013). 26. Schiesel S. (2008) Author faults a game, and gamers flame back. The New York Times. www.nytimes.com/2008/01/ 26/arts/television/26mass.html (accessed July 1, 2013). 27. Ferguson CJ. (2013) Don’t blame video games for real-world violence: the chronicle of higher education. http://chronicle .com/blogs/conversation/2013/01/10/dont-blame-videogames-for-real-world-violence/ (accessed July 1, 2013). 28. Person C. (2013) Breaking down the absurd anatomy of a video game scare piece: Kotaku. http://kotaku.com/breakingdown-the-absurd-anatomy-of-a-video-game-scare-487437560 (accessed July, 1, 2013). 29. Ferguson CJ, Dyck D. Paradigm change in aggression research: the time has come to retire the General Aggression Model. Aggression & Violent Behavior 2012; 17:220–228.
Address correspondence to: Dr. Andrew K. Przybylski Oxford Internet Institute University of Oxford 1 St. Giles Oxford, OX1 3JS United Kingdom E-mail: [email protected]
