Imagine that you wanted to conduct a controlled experiment to find out whether playing violent video games causes adults to behave more aggressively in real life. In order to make the causal conclusion, we have to control exposure to violent video games and measure acts of aggression.
The independent variable (IV) is the thing we are going to manipulate. Some adults will play a violent video game and some will play a similar, but non-violent game.
The dependent variable (DV) is the thing we are measuring to see if the IV had an effect
In an experiment like this, there are two groups of participants to compare. The experimental group, in this study, would be the group playing the violent video game. The other group that we are comparing them to is the control group. If the control group did not play any video game at all, it would be possible that playing any video game makes adults more aggressive. So in this example the only thing that is different between the two groups is the level of violence in the game (the IV).
WATCH: The Variables Song
A fun way to remember variables and their purpose in controlled experiments.
To continue with our example, imagine that we want to determine whether eating Cheerios actually causes a change in academic performance. To do that, we have to conduct an experiment where we manipulate what students eat for breakfast (IV) and measure grades (DV). We will study 100 6-year-olds for a semester. We will feed half of them a breakfast of Cheerios every day and give the other half a PopTart. At the end of the semester, we will look to see which group had better grades.
Now, aside from some potential ethical problems with this experiment, how can we be sure that the group who ate Cheerios was not just a smarter group of students to begin with? Maybe they would have earned better grades even if we did not manipulate their breakfast foods. To control for other variables that might influence academic performance (e.g., intelligence, motivation, S.E.S.), we need to break people into the two groups using a random system. That way, every participant has an equal chance of being assigned to the different groups in the study. Sure, some students might be more intelligent than others, but on average, the two groups will have about the same number of especially intelligent students. Therefore, when we compare the two equivalent groups and find a difference in their grades, we can be reasonably sure it is not because of anything other than the manipulated IV.