Classical Experiment Design: Violence in Video Games

Introduction and Components

This paper presents the design of a classical experiment examining the relationship between violent video games and aggressive behavior. The experiment makes use of the following apparatus:

A video game console — for example, a Nintendo 64 game system — along with appropriate game cartridges, including one violent video game (wrestling) and one non-violent video game (basketball). In the wrestling game, human violence is prevalent: the object of the game is to punch, kick, and use blunt weapons and other wrestling moves to subdue the opponent. The basketball game (NBA Live '99), by contrast, contains no violence; players are not supposed to harm opponents, and any such action results in a warning and a penalty. Players are also expected to focus on scoring more points than the opponent in the shortest time possible (Uhlmann & Swanson, 2004).

Additional apparatus include one VHS camcorder, a one-meter-tall inflatable Bobo doll, and a television set.

Aggressiveness will be measured using two cognitive measures and one behavioral measure. The behavioral measure involves close observation of each participant's interaction with the Bobo doll. The two cognitive measures consist of one designed to prime aggression and one to assess normative beliefs about aggression. The priming measure is a word-stem completion task in which the participant is given a word with a missing letter; each stem can be completed to form at least one familiar aggressive word and at least three non-aggressive words (Funk, 2005).

The study is guided by the following research questions:

Research Questions and Hypothesis

1. What are the short-term effects of violent video games on a person's behavior?

2. What is the long-term effect of violent video games on a person who has played them over an extended period?

Subject Selection Criteria

3. What impact does watching a violent game being played have on a third party?

The central hypothesis of the study is that playing violent video games has a measurable impact on the behavioral tendencies — specifically, the aggression — of someone who is exposed to them.

Participants will be randomly selected from a single school, chosen for convenience. The randomly selected participants will range from third grade to sixth grade, and their grade level will determine their group assignment. The experiment will involve more boys than girls for two reasons: boys constitute the vast majority of video game players, and boys are more likely than girls to demonstrate overt aggression.

Experimental Procedure

Inclusion criteria are as follows: participants must be at least ten years of age, and they must be in sound mental health.

The experiment employs a 2×2 between-groups factorial design, with grade level (third grade vs. sixth grade) and media content (violent vs. non-violent) as the two factors. Two participants will be randomly assigned to each of the four conditions.

The experiment will take place in a quiet room at the participants' school. Each participant will be individually escorted to and from their classroom by the experimenter. Upon arrival, the participant will sit down, read the provided instructions, and complete an assent form. The participant will then be exposed to a 15-minute video game session, during which they will play either the violent wrestling game or the non-violent basketball game (Funk, 2005).

For both games, participants receive instructions on how to play if they are unfamiliar with the controls. After completing the 15-minute session, the experimenter presents the word-stem completion task and the normative beliefs task in counterbalanced order. The word-stem task is administered first: a series of partial words is read aloud, and the participant must supply the missing letter to complete each word. For example, the experimenter might present the stem _UN. This stem can be completed as an aggressive word (e.g., GUN) or a non-aggressive word (e.g., FUN), depending on the letter the child selects. If a participant does not respond within 15 seconds, the experimenter provides a hint by offering the two possible letters (e.g., "F or G").

In the second stage, the participant is asked whether they would like to play with the Bobo doll. The experimenter then leaves the room, while a hidden camera continues to record the participant's interaction with the doll. Scoring for this stage is based on the intensity of the participant's engagement with the doll rather than the frequency of interactions. Research on media violence and its effects on children has consistently highlighted the value of behavioral observation as a complement to self-report measures.

Several limitations are anticipated. First, some students may fail to provide answers for the missing-letter task even after receiving a hint, which could result in incomplete data. Second, because the word-stem and Bobo doll tasks are counterbalanced, the data obtained in the second stage may be influenced by the participant's performance in the first stage. Third, the Hawthorne Effect may compromise result quality, given that the experimenter will be closely engaged with participants throughout the session, potentially altering their natural behavior.

Permission for each selected participant will be obtained through a formal letter sent to their parents or guardians. This letter will include a full description of the experiment as well as a consent form requiring the signatures of both the parent and the student.