Violence in Teen-Rated Video Games

Kevin Haninger; M. Seamus Ryan; Kimberly M. Thompson, MS, ScD

Disclosures
In This Article

Results

We report the summary results and statistics based on the random sample of 81 T-rated video games, but we tabulate results and discuss insights from our observations of all 90 T-rated video games we played. Table 1 compares the distribution of genres in our random sample of 80 video game titles with that of all 396 T-rated video game titles released by April 1, 2001. Video games titles in the action (25%), fighting (24%), and shooting (14%) genres represented nearly two thirds of T-rated video games titles available at the time we drew our random sample.

Table 2 groups ESRB content descriptors by type and shows the breakdown of the content descriptors by genre for all 396 T-rated games released by April 1, 2001. Although T-rated games receive content descriptors for a wide range of content, almost all (94%) received some type of content descriptor for violence (ie, Animated Violence, Mild Animated Violence, Realistic Violence, Violence, and/or Mild Violence) and 26% received some type of a content descriptor for blood (ie, Animated Blood, Animated Blood and Gore, Blood, or Blood and Gore). We found that 280 game titles (71%) received a content descriptor for violence but not blood, 93 game titles (23%) received content descriptors for violence and blood, 9 game titles (2%) received a content descriptor for blood but not violence, and 14 game titles (4%) did not receive content descriptors for violence or blood.

Table 3 groups the 90 T-rated video games we played by genre and summarizes our observations of violence and blood. The table shows the game title, console, and release year in the first column, and the ESRB content descriptors in the second column. In the random sample of 81 T-rated video games we played, 79 games (98%) involved intentional acts of violence, with 77 games (95%) receiving content descriptors for violence. Only the trivia game (You Don't Know Jack, Mock 2) and one adventure game (Overblood) did not contain acts of violence within our hour of game play, although we observed depictions of injuries from violence in the beginning of Overblood, and we know that acts of violence occur later in the game. We observed blood in 34 of 81 games (42%), with 22 games (27%) receiving content descriptors for blood.

We observed wide variation in the amount of violence in the games, with the 79 games (98%) that involved violence containing an average of 36% violent game play time (range 0.1% to 87%). If we include the additional game play associated with planning violence and depiction of injuries (ie, "All Violence" shown in Table 3 , column 5), then we observed an average of 39% of game play time directly related to violence (range 0.5% to 88%) in these 79 games. Overall, we found that the percentage of game play that involved planning violence and depiction of injuries, while measurable, is smaller than the percentage involved in committing violent acts. Compared with our analysis of E-rated video games, this study suggests that T-rated video games are significantly more likely to contain violence than are E-rated video games (ie, 98% of T-rated video games vs 64% of E-rated video games depicted violence,[13] chi-squared = 27.8, P < .001). However, T-rated and E-rated video games that depicted violence did not contain statistically different amounts of violent game play time, based on a 2-sided Wilcoxon rank-sum text. In coding whether each second of game play contained acts of violence, we found excellent agreement between the author who coded all of the video games and the independent coder who assessed the violence in 10 games (kappa = 0.93).

We observed a total of 11,499 character deaths in approximately 95 hours of game play for the random sample of 81 video games, occurring at an average rate of 122 character deaths per hour of game play (range 0 to 1310). This total included 5689 human deaths, occurring at an average rate of 61 human deaths per hour of game play (range 0 to 1291). Overall, we identified 44 games (54%) that depicted deaths to nonhuman characters and 51 games (63%) that depicted deaths to human characters, including the player.

Analysis of the overall portrayal of violence in the random sample of 81 games indicated that 44 games (54%) depicted injuries to nonhuman characters, while 72 games (89%) depicted injuries to human characters, including the player. In classifying the portrayal of violence by severity (ie, "mild," "medium," or "strong"), we found excellent agreement between the author who coded all of the video games and the author who played them (kappa = 0.92). Examining the portrayal of violence to nonhuman characters, we found that 11 of 81 games (14%) portrayed violence categorized as "mild" to nonhumans, 27 games (33%) portrayed "moderate" violence to nonhumans, and 6 games (7%) portrayed "strong" violence to nonhumans. Examining the portrayal of violence to human characters, we found that 8 of 81 games (10%) portrayed "mild" violence to humans, 45 games (56%) portrayed "moderate" violence to humans, and 19 games (23%) portrayed "strong" violence to humans. These results suggest that T-rated video games containing violence portray violence to human characters more severely than violence to nonhuman characters based on a significant result in a test for trend (chi-squared = 28.4, P < .001). We also found that video games that received content descriptors for Mild Animated Violence or Mild Violence portrayed violence less severely to human (P = .001) and nonhuman characters (P = .0001) than video games that received content descriptors for Animated Violence, Violence, or Realistic Violence, based on 2-sided Wilcoxon rank-sum tests.

We observed 37 of 81 games (46%) that rewarded or required destroying objects. We identified 73 games (90%) that rewarded or required injuring characters, with injuring human characters rewarded or required in 54 games (67%). This included 56 games (69%) that rewarded or required killing characters, with killing human characters rewarded or required in 32 games (40%). Compared with our analysis of E-rated video games, these results suggest that significantly more T-rated video games give players incentives or requirements to commit acts of violence (ie, 73% of T-rated video games vs 60% of E-rated video games,[13] chi-squared = 17.3, P < .001).

Table 3 also shows that 71 of 81 games (88%) depicted weapons other than the body. More than half of games (51%) depicted 5 or more types of weapons, with players able to select weapons in 48 games (59%). A total of 59 games (73%) used the body as a weapon, 56 (69%) used projectiles, 46 (57%) used guns, 44 (54%) used explosives, 36 (44%) used knives or swords, 27 (33%) used fire, 24 (30%) used magic, 11 (14%) used toxic substances, and 44 (54%) used other weapons (eg, automobile, hammer, police baton). This does not represent an exhaustive list of the weapons that players might encounter in the video games because of the limited amount of time we played each game; consequently, the list should be viewed as a subset of the weapons depicted in these games.

Table 4 summarizes our observations of violence and blood by game genre. We measured the highest average percentages of violent game play in the 2 most prevalent genres of T-rated video games: fighting (60%) and action (41%). Shooting and action games depicted the highest average number of character deaths per hour (394 and 221, respectively) and the highest average number of human deaths per hour (270 and 75, respectively). In contrast, fighting games, which contained the highest average percentage of violent game play, showed an average of 7 character deaths per hour, only 3 of these human. Although many fighting games contained the use of lethal weapons, we observed that they often indicated a defeated character was knocked out, but not dead. Role-playing games allowed the player to select the greatest variety of weapons, but they also portrayed injuries less severely ( Table 3 ) and involved the player in more nonviolent game play than did video games from the more prevalent genres. These results offer more evidence that genre provides important information about the violent content of video games.[13]

Our analysis of the 9 additional video games we played on Xbox and GCN suggest similar results for newer games, which generally contain more vivid depictions of content than some of the older games in our random sample. Table 5 summarizes our quantitative observations of sexual themes, profanity, and substances in these additional games. Table 6 provides notable examples of the different types of content we observed.

Finally, we found that the R-rated films The Matrix and The Matrix: Reloaded depicted significantly less violence (10% and 16% of screen time, respectively) and fewer human deaths per hour (18 and 22, respectively) than the T-rated video game Enter the Matrix, which contained 45% violent game play and 117 human deaths per hour (mainly deaths of police officers, security guards, and postal workers). While the films contained blood and the video game did not, we noted that all portrayed graphic violence involving martial arts and guns. We also noted that the game manual contained a $3 rebate toward the purchase of The Matrix DVD, which indicates the continued marketing of R-rated violent entertainment to children.[3,4]

Comments

3090D553-9492-4563-8681-AD288FA52ACE
Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.

processing....