Source: Research; Public Domain http://commons.wikimedia.org/wiki/File:Knox-ut-research-1942.jpg hands over face: public domain; http://morguefile.com/archive/display/87881
Hello, class. So the most widely recognized method of psychological research as well as scientific research in general is the experimental method. And we're going to be talking about that today.
So an experiment is a study that's specifically done or designed to confirm or disconfirm a hypothesis about cause and effects between two different sorts of things or more than two. So generally, an experiment is done within a professional setting like a laboratory, and it's specifically adapted to the needs of the study itself. So whatever the professional setting is, it has certain elements that are needed for the experiment. And this also allows the psychologist to discover the causes for specific events or behavior. So experimental methods are very good at explaining causation and saying this thing makes another thing happen. So let's go over the different elements of the experimental method.
So to begin with in an experiment, we start by selecting experimental subjects. And subjects are the group of people that are going to participate in an experiment. Now, the subjects are chosen as a sample to represent a larger group or population of people. So we're not necessarily testing everybody. We're testing just a select number of people that are going to make a general conclusion about what happens with the larger population or about-- with people as a whole.
These subjects are further divided into control and experimental groups. And there might be more than two, or certain ones might use different aspects. But this is generally what happens.
A control group is a group of people that receive all of the conditions within an experiment except for the experimental variables or the conditions that are being tested. So they get everything else. This is, in other words, kind of the normal or the regular group. And it shows that normally when the events within the experiment happen, that there's no effect. There isn't anything that happens.
And then we have the experimental group, which are the people that receive all of the conditions of the experiment, including the ones from the control group, except they also get whatever the different experimental condition is. So, for example, if I were to do an experiment with a new drug, I would take everybody, all of the different subjects, and I put them into two different rooms. In one of the rooms, I would give the exact same room as the other one, so same chairs, same walls, everything that's the same. I would give each person within that room a glass of water, and I'd give them something small and benign, maybe what we call a sugar pill that would have no effect on the people. This would be the control group.
And then in the other room, which again is exactly the same, because we don't want anything different or anything else affecting our experiment, then I would give those people the glass of water, and I would give them the new drug that I'm trying to test. And then I would look to see what the difference is between these two different groups of people. So I want to control all of the conditions of the experiment to make sure that there's nothing else that's causing the effects that I measure as best you can-- sometimes you can't necessarily do that-- in the most even possible way.
Sometimes people are chosen within an experiment for specific aspects of behavior, specific things there are special to them. Most of the time, we're just trying to choose people that are generally true for the population. So most often, we want to choose groups or subjects for an experiment that are large. We want lots of people, and we want them chose them randomly to prevent any kind of errors within sampling. What's true is that usually bigger is better. The more people we can have in the experiment statistically, the better or more accurate our results will be.
There are also other techniques that we use to keep the results accurate. The first one is random assignment, which is to say when we're putting people within the different groups, we choose them at random and assign them to the experimental or control groups. In this way, we don't want to consciously or unconsciously affect the results by choosing certain people to go in certain types of groups. So random assignment is a helpful way.
The other way we might do it is through a double-blind experiment, which is when the subjects are assigned to their different groups and neither they nor the experimenter know who's in the experimental or control group. So we would take all of the subjects, and without the experimenter even knowing it, we would put them into one room or the other. One of the rooms has the pills that they're studying and one of them has the sugar pills.
And in that way, it prevents either the experimenter or the subject from influencing the results. So if the experimenter knows that the people are taking a certain type of drug, then they might look for more results, and they might emphasize them more, either consciously or unconsciously. Alternatively, if the subject knows that they're taking a new pill, they might feel like they're going to have some kind of result, and they might report it more easily.