Violence in Video Games

Games are growing
APA finds Video Games to be strongly associated with increased aggressive behavior, cognition, affect, and arousal.
Also associated with desensitization and decreased empathy

Why Study MMORPGs?
MMORPG players spend significantly more time playing.
Gamers play for an average of 6.25 hours per week (U.S. Games 360 Report for 2018).
Two studies found that MMORPG players play more than 20 hours per week(Yee, 2006; Griffiths, Davies, & Chappell, 2004).
Beyond self selection, another study assigned players to four groups. Those assigned to play MMORPGs played for an average of 14.4 hours per week (Smyth, 2007).

Why Study MMORPGs?
Social interaction positively influences how enjoyable the games are as well as the amount of time spent playing them (Chen, Phuah, & Duh, 2007; Griffiths, Davies, & Chappell, 2004; Dindar, 2018).

Why Study MMORPGs?
They create an emotional attachment and sense of investment through the creation of highly customizable player controlled characters.
These characters become representative of an online identity, the development and status of which over time, drives the need to continue playing (Chen, Phuah, & Duh, 2007; Dindar, 2018) and results in a sense of loyalty to the game (Yee, 2006).

Cultivation Theory
Meta-analyses of violence in television and movies found four distinct cultivation effects: the aggressor effect, the victim effect, the bystander effect, and the appetite effect (Donnerstein, Slaby, & Eron, 1994)
Cultivation effects related to risky driving, fear, and racial stereotypes have been found in video games (Beullens, 2006; Williams, 2006; Behm-Morawitz & Ta, 2014).
The Aggressor, Victim, and Bystander Effects, while not explicitly described as such, have been found in Video Games.

Content Analyses
According to Schmierbach (2009), content analyses of violence in video games tend to use a variety of different methods and, when selecting samples within the games, often do so in a way that compromises the design.
Popular Video Games: Quantifying the Presentation of Violence and Its Content, conducted by Smith, Lachlan, & Tamborini in 2003, sampled segments that were just the first 10 minutes of each game in the sample.
The type of characters, settings, plot-points, and forms of violence all tend to change as a player advances through a video game and even if studying longer and later segments of gameplay those segments may still misrepresent the frequency with which a particular piece of content occurs

The Present Study
The purpose of this study was to content analyze the top five most played Massively Multiplayer Online Roleplaying Games (MMORPGs) for the context of violence. This includes World of Warcraft, The Elder Scrolls Online, Black Desert Online, Guild Wars 2, and Final Fantasy XIV.
Hypotheses based on the results Popular Video Games: Quantifying the Presentation of Violence and Its Content, conducted by Smith, Lachlan, & Tamborini in 2003

Hypotheses
H1: The majority of violence will result in death.
H2: The majority of violence will involve repeated acts.
H3: Humans will be the target of violence more often than other sentient species, droids, or animals/non-sentients.
H4: Males will be the target of violence more often than females.
H5: Humans that are white will be the target of violence more frequently than humans that are not white.
H6: Adults will be the target of violence more often than children.
H7: Violence is more likely to be justified than unjustified.
H8: Violence will be rewarded more than it is punished.
RQ1: Is the violence typically required by the games’ quests or missions?

Method
“any overt depiction of a credible threat of physical force or the actual use of such force intended to physically harm an animate being or group of beings. Violence also includes certain depictions of physically harmful consequences against an animate being or group that occur as a result of unseen violent means” (Smith et al., 1998, p. 30)

Method
The unit of analysis is a PAT which is defined as “an aggressive exchange that occurs between a perpetrator (P) engaging in a particular type of act (A) against a target (T)” (Smith, Lachlan, & Tamborini, 2003)

Coding Categories
The coding categories were developed from categories and variables used in the Smith, Lachlan, & Tamborini study conducted in 2003, either directly borrowed or adapted.
Requirement of Violence
Result of Violence
Extent of Violent Acts
Demographics of the Targets of Violence
Reason for Violence
Reinforcement of Violence
Unused categories.

Procedure
Participant Installed games on his personal laptop.
Played each game for two hours and recorded the footage.
Met with Researcher, given codebook, explanation of codebook, and coded the footage using a shared excel sheet
Pilot study conducted playing SW:TOR to establish reliability and validity.
Coders were in complete agreement for all but two categories.
High reliability even for those two categories (Justified vs not Justified).

Results
Results in death 97.6% of the PATs.
Injury is a result in only 2.8% of the PATs.
Hypothesis 1 Supported
Repeated acts in 85.1% of PATs.
Single acts in 14.9% of PATs.
Hypothesis 2 supported.

Results
Violence is required in 83.4% of PATs
Research Question 1 is answered.
Violence is justified in 52.9% of PATs.
Hypothesis 7 supported.
Violence is rewarded in 97.6% of PATs.
Violence is never punished.
Hypothesis 8 is supported.

Results
Humans not a target at all
Hypothesis 3 is not supported.
Non-human Sentients were the target in 58.1% of PATs.
Animals/Non-Sentient were the target in 45.7% of PATs.

Results
Males are the target in 51.2% of PATs
Females are the target in 4.2% of PATs
Hypothesis 4 is supported.
Adults were the target in 53.4% of PATs.
Children were the target in 4.8% of PATs.
Hypothesis 6 is supported.
Hypothesis 5 could not be tested.

Discussion
Prior content analyses focused on other genres or top selling games, not MMORPGs specifically.
Future content analyses of MMORPGs could attempt to overcome the limitations of this study.
They could also attempt to analyze the portrayal of race or gender in these games as studies have done for other games.
Future studies could determine if MMORPGs have a greater effect on behavioral aggression, cognition, arousal,and desensitization to violence, than other games.
Future studies could examine if repetition of violent acts in particular has an effect.
Future studies could examine if the absence of humans as a target influences perceptions of violence and of the social normativity of violence.

Limitations
The Elder Scrolls Online was dropped from the sample due installation errors and a laptop that failed to function.
Using time (the first two hours) to sample the games artificially divides them in ways that exclude important plot points and content (Schmierbach, 2009). Samples may not accurately reflect the games.
Purposive Sampling was used, meaning that these results may not be representative of ALL MMORPGs
Footage coded on the same device, possibility of artificially high inter-coder reliability.