In this video I explain the experimental method and how the combination of manipulation and random assignment can help to eliminate the third variable problem. I also explain why random assignment helps avoid self-selection and bias so that we can assess the role of the independent variable on the dependent variable.
Don’t forget to subscribe to the channel to see future videos! Have questions or topics you’d like to see covered in a future video? Let me know by commenting or sending me an email!
Need more explanation? Check out my full psychology guide: Master Introductory Psychology: http://amzn.to/2eTqm5s
Video transcript:
Hi, I’m Michael Corayer and this is Psych Exam Review and in this video I’m going to explain the experimental method. So in the last video I talked about the third variable problem and this is the problem that we have with correlational research. If we measure two variables and we find a relationship between these variables we don’t know what’s causing the relationship. And it could always be a third variable that we didn’t measure that’s influencing this relationship and the list of possible third variables that it could be is infinite. We have no way of ruling something out until we actually measure it.
So the experimental method tries to solve this problem by using manipulation and what manipulation refers to is that the researchers take control of one of the variables and they manipulate it for some participants.
What this manipulation accomplishes is it answers the question of “why do people have some score for this variable?” Well the reason is that the researchers made them have that level. So for example, in the previous example I talked about study time. We said that some students maybe study 3 hours some study 1 hour, some study not at all, and we don’t know what’s causing this. We say it could be parental pressure could be how much they enjoy the class, or it could be on how much coffee they’re drinking, it could be anything. We don’t know.
But if we manipulated that variable then we would know the cause. So if I forced some students to study for 3 hours and I didn’t let other students study at all, now I know the answer to that question why does this student study for 3 hours, why does this student not study at all, because I made them do that. So manipulation allows us to know the cause of variation. We don’t have that question anymore.
Now we know the researcher is the reason for this particular variable. Now if we’re going to manipulate the variable amongst participants this means we’re going to need to split our participants up into groups. Generally we’re going to have two groups, we’re going to have an experimental group and we’re going to have a control group.
So the experimental group refers to the group that receives some sort of treatment. So they take a medication or they do some sort of studying or they have some sort of treatment and the control group that they’re compared to does not receive that treatment.
Now it can be important that participants don’t know which group they’re in because their expectations and their beliefs might influence our results so the control group often gets a false treatment so they think they’re getting a treatment but it’s not real. That’s a placebo treatment. So in the case of studying a medication, one group is going to get a pill that actually has medication in it that’s the experimental group and the control group is also going to get a pill but what they don’t know is that that pill has no medicine in it, it’s empty and that would be a placebo pill, that would be the “placebo” group.
OK so how do we decide how we split our participants up into these groups? So we have a bunch of participants here and we’re going to split them up into an experimental group and a control group.
The question is how do we decide? And the answer is, we don’t decide. It’s not really a decision, we have to use random chance. So we assign our participants using random assignment or you might see this called random allocation. This is how we allocate or assign participants to the experimental group or the control group.
And there’s a couple reasons why we want this to be random. There’s certain things we want to avoid. The first thing we want to avoid is self selection. This is where something about the participants causes them to choose one group or another. So for instance if I told my students, if I asked them, “who wants to be in the experimental group and study for 3 hours and who wants to study for 0 hours?” well now the students who are under more pressure might choose the 3 hour group or the students who are drinking more coffee or something and now that third variable problem that I had before is back. Now it’s that that caused them to choose the group that they’re in. So I don’t want any sort of self-selection.
I also want to avoid the potential for bias from the researcher. So I can’t choose who’s in which group because I might decide to put the more hardworking students in a particular group or something I might even not be aware of why I’m putting somebody in a particular group and this introduces the possibility of bias and that could influence our results. So we don’t want to have self-selection and we don’t want to have bias and the best way to avoid those is to just use random chance.
Now if we’ve used random chance that we can say we have an independent variable. The independent variable refers to the variable that’s manipulated. And the reason that it’s called the independent variable is it’s meant to be independent of anything else. In other words, the reason for participants’ score for this variable is not due to anything else. It’s completely independent. It’s not related to how much coffee they drank, it’s not related to their parental pressure, it’s not related to how much they enjoy the class. It’s not related to any of those things and if it’s been randomly assigned then it should be an independent variable. So now the reason why one person is studying for 3 hours and one person is studying for 0 hours is no reason. It’s random chance, they were just assigned to do that. So that’s why it’s called the independent variable.
What we then do this after we manipulate the independent variable we’re going to look at a dependent variable. And the dependent variable refers to what we measure after the treatment.
And what we want to find out is “does this independent variable have an influence on the scores on the dependent variable?”. Does this variable depend on how much somebody studied or does this depend on whether they took the medicine or the placebo pill? So that’s why this is called the dependent variable. We want to find out if it depends on the independent variable.
That’s the basic idea behind the experimental method and the two most important things to remember, first, this idea of manipulation. This is fundamental, we need to control a variable if we only measure things then we can never be sure about the cause. But if we manipulate them we can see “does manipulating this actually have an influence on this later” and that gives us evidence for causation.
And the second important point is this idea of random assignment. We need to avoid self-selection, we need to avoid bias, because if we don’t have random assignment the the third variable problem comes back and we don’t have evidence for causation.
I hope you found this helpful. If so, please like the video and subscribe to the channel for more. Thanks for watching!
