Media Influences: effects of computer games Flashcards
What do experimental studies find about aggression?
These are lab studies of short-term effects of computer games that can demonstrate cause-and-effect. Bartholow and Anderson’s participants played a violent or non-violent computer game for ten minutes - then carried out the Taylor Competitive Reaction Time Task, a standard lab measure of aggression (choosing volume of noise blasts). Those who played the violent game selected significantly higher noise levels compared to non-violent players.
What do correlational studies show about aggression?
They can investigate real-life variables and are usually short-term. DeLisi et al studied 227 juvenile offenders with histories of serious aggressive behaviour. Structured interviews used to gather data on aggression and violent computer-game-playing. Aggressive behaviour was positively correlated with how often they played violent computer games and how much they enjoyed them. Researchers claim the link is so well- established that computer game violence should be considered a significant risk factor.
What do longitudinal studies show about aggressive behaviour?
These are correlational studies conducted over a period of time looking at long-term effects. Robertson et al studied 1037 people born in New Zealand, measuring their TV viewing hours (all types of programme) at regular intervals up to the age of 26 years. Time spent watching TV was a reliable predictor of aggressive behaviour in adulthood (convictions for aggressive and violent crimes) and diagnosis of antisocial personality disorder. The most important media-related factor in influencing aggressive behaviour may be the amount of TV watched rather than whether it has violent content or not.
What do meta-analysis studies show about aggression?
They bring together a variety of studies to give a judgement of size of the effect of violent media on aggression. Anderson et al’s meta analysis of 136 studies included experimental, correlational and longitudinal research. Exposure to violent computer games was associated with increases in aggressive behaviours, thoughts and feelings. This applied to both males and females across collectivist and individualist cultures.
What do Anderson et al claim?
The effect of violent game-playing on aggressive behaviour is greater than the effect of second-hand smoke on cancer. The analysis also showed no indication that publication bias influenced the results.
Limitation: measures of aggression are artificial.
The Taylor Competitive Reaction Time Task measures aggression as the volume of noise selected by participants as punishment - an unrealistic measure. Aggression measures in lab situations are also unrealistic because they do not involve retaliation - the experimenter gives the participant permission to be ‘safely’ aggressive. This casts doubt on the validity of experimental studies and on the link between media violence and aggression.
Limitation of experimental studies: non-equivalence problem.
Experimental studies compare participants who play a violent computer game with those who play a non-violent one - but it is difficult to be sure that the two games are equivalent apart from the presence or absence of violence. For example, the violent game may be much more complex then the non-violent game, which could act as a confounding variable. Participants may behave more aggressively after playing the violent game - but this could be due to frustration caused by the greater complexity of the game rather than the violence the players are exposed to.
Limitation of correlations: we cannot draw cause-and-effect conclusions.
No variables are manipulated or controlled, and there is no random allocation of participants to violent or non-violent media conditions. A positive correlation between violent computer games and aggression can be explained by socialisation (violent games cause people to be aggressive) or selection (people who are already aggressive choose video games). The direction of causality cannot be settled by correlational studies so the finding does not help us choose between two hypotheses, leaving open the question of how computer games influence aggression.
Limitation of longitudinal studies: confounding variables.
Longitudinal studies are conducted over quite lengthy periods of time (from several months to several years). Many sources of aggression interact with media influences over this period (e.g. Role models). It is difficult to separate them and assess contributions to aggressive behaviour. Therefore it is impossible to conclude that it is violent media such as computer games (rather than confounding variables) that have affected aggression.
Limitation of all methodologies: publication bias.
There is a tendency for only statistically significant findings to be published. This is a particular problem for meta-analyses because they generally only include published studies, which mostly show a significant effect. So publication bias creates the false impression that the effects of violent media on aggression are greater then they actually are.
