Back To CoursePsychology 102: Educational Psychology
10 chapters | 123 lessons | 9 flashcard sets
As a member, you'll also get unlimited access to over 70,000 lessons in math, English, science, history, and more. Plus, get practice tests, quizzes, and personalized coaching to help you succeed.Free 5-day trial
Scientists in almost every field of study use experiments to answer research questions. Imagine you are a psychologist, and you want to investigate whether caffeine has an effect on student behaviors and performance in the classroom. How would you go about finding out the answer to this question? The answer is that you would do an experiment. This lesson covers all of the different aspects of an experiment you would want to consider.
The first thing any experimenter needs to decide is what variables you are studying. Let's imagine your hypothesis is that when students in school consume caffeine, their performance on tests is affected. You might hypothesize that caffeine increases test performance because it causes the students to be less sleepy and more focused, or you might hypothesize that caffeine decreases test performance because it makes the students jumpy and hyper. Either way, you have two variables involved in this study.
The independent variable in an experiment is the variable that you control as the experimenter and the one that creates two or more groups in the study. In order to study caffeine, you might give half of the students a caffeinated drink and the other half of the students simply get water. The difference between the two groups is whether they have caffeine or not. So, the independent variable is the variable that you, as the experimenter, have manipulated.
The dependent variable in an experiment is the outcome variable or the one you are simply measuring. Here, you guessed that caffeine might affect test performance. So, in this example, your dependent variable is test performance.
Another way to think about independent variables and dependent variables is in terms of cause and effect. This study is testing whether caffeine (the cause) has an effect on test performance. All experiments are testing if whatever makes the groups different has an effect on some outcome variable. The independent variable is always the cause. Here, that's the caffeine. The dependent variable is always the effect. Here, that's test performance. So, the independent variable always happens first, and the dependent variable always happens second.
Now, let's talk about why we need more than one group in an experiment. Imagine you went into a classroom, gave every student caffeine and then tested them on some kind of performance measure, such as the number of times they can jump a rope. You can see how these students performed after having caffeine. But how can you know if their performance was increased or decreased compared to what they would have done without caffeine? With only one group in your study, you can't be sure what the effects of caffeine were.
So, in an experiment we always need at least two groups to compare. Let's go back to the example of giving half of the students caffeine and half of the students water to drink. When we're testing for the effect of the independent variable, we want to make sure that one of the groups in our study can serve as the natural, or baseline, group. That natural or baseline group is called a control group. In our example, the control group would be all of the children who only drank water.
We then compare the control group to the group of children who received caffeine. In an experiment, the group that receives some kind of change to their natural environment is called the experimental group. In our example, the experimental group would be all of the children who drank caffeine.
We need at least two groups so that we can compare the experimental group to the control group. Control groups are especially necessary when testing for the effect of drugs, like caffeine, because we want to make sure the group doesn't change simply because they think they are supposed to. When you change your behavior just because you expect a change, that's called the placebo effect. For example, let's say we give all of the children soda, but half of the sodas have caffeine and half are caffeine-free. We wouldn't want to tell the children which kind of soda they got, because they might change their behavior simply due to expectations. This type of problem can be avoided with a good control group.
When we divide the class up into the control group versus the experimental group, it's important to make sure that this division occurs at random. When each person in the study has an equal chance of being in either the control group or the experimental group, that's called random assignment. You might decide which group each person is in by flipping a coin, as an example.
Why is random assignment important? We want to make sure that the groups are as identical as possible in every way except for the independent variable. Let's go through an example of why this matters.
Imagine that you decided that all the boys in the class would get the caffeine drink, while all the girls in the class got the no-caffeine drink. Then, you tested for the effects of caffeine using the dependent variable of jumping rope. Now, imagine that you see a difference! The caffeine group is better at jumping rope. But, you can't actually conclude that caffeine was the cause. It could be that boys are better at jumping rope. Unless the groups are identical in every possible way except for the independent variable, you can't be sure what caused any difference in the dependent variable. But if the independent variable really is the only difference between groups, then you can be sure, because there's no other explanation. This is why random assignment is so important: only with random assignment can you be sure of a cause-effect relationship within an experiment. Without random assignment, there might be other reasons why you see a difference between the two groups.
The final decision you have to make when you're designing an experiment is who will participate. When you're studying any research question, the entire group of people you're interested in is called the population. The population is always a relatively large group. If you're interested in the effects of caffeine in children, then your population is actually every single child in the entire world! A more specific population might be every child in a certain school. But, of course, you probably can't include every child in your study. So, you will use a sample. A sample is the group of people actually participating in your study. The sample is always smaller than the population. Next, you need to decide how you will narrow down your population and choose your sample for the study. You have several different options.
A common way to choose people for a study is to simply use people who are readily available and willing to participate, such as your friends or family. You might advertise for the study on your personal Facebook page and wait for people to volunteer. When you use an easy sample like this, it's called a convenience sample because it was so convenient and easy.
However, a convenience sample is not a very good way to get people in the study. Why not? Because when you only use a convenience sample you're probably only getting a certain type of people. For example, you might only use people who live in your local area. A specific type of convenience sample that's based on a particular geographic area is called a clustered sample. This means that your sample probably isn't a good representation of the population. If you lived in California and you haven't asked any people from Iowa to be in your study, you can't really be sure that people in California act the same way as people in Iowa.
One way to attempt to get a good sample from the population is to make a list of every person in the population, then choose randomly from that list. This type of sample is, as you might guess, called a random sample. The definition of a random sample is one in which every person in the larger population has an equal chance of being in the sample. The idea behind a random sample is that it's probably representative of the population, meaning any important variables that are in the population are equally present in the sample. Look at this graph for an example. The colors could represent any important variable, such as different ages of children, different grades in school, people living in different states, or different ethnicities. When you use a convenience sample, you might not get the same breakdown in the sample as you had in the population. But if you choose randomly from the population, you've got a better chance of getting a representative sample.
A more complicated version of a simple random sample is called a stratified random sample. Here you identify important variables in advance, divide up the population based on this variable, then randomly choose an equal number of people from each group for the sample. For instance, you might think that caffeine affects boys differently from girls, and you want to compare. So, you would want to make sure you randomly choose an equal number of boys and girls. When you identify variables like this in advance that you want to make sure are covered by your sample, you used a stratified random sample. It's partially random but also partially structured.
No matter what kind of sample you use, remember that the main concern is that it is representative of the larger population of interest.
In summary, in an experiment, the variable you control that creates groups within your sample is called the independent variable. Experimenters hope that differences in the independent variable cause changes in the outcome variable, which is called the dependent variable. The baseline group that doesn't receive any change from their normal environment is called a control group, whereas the group that receives some change is called the experimental group. You want to make sure that you decide which participant goes into which group using random assignment so that the groups are as identical as possible. There are many different ways you can choose a sample from the larger population. In order to make sure your sample is representative of the population in terms of important variables, the best way to choose people is by using a random sample. These issues can be complicated, but if you do them correctly, experiments can be a very useful way to answer questions from psychology. If you consider all of the issues from this lesson, you can create a true experiment.
After watching this lesson, you should be able to:
To unlock this lesson you must be a Study.com Member.
Create your account
Did you know… We have over 160 college courses that prepare you to earn credit by exam that is accepted by over 1,500 colleges and universities. You can test out of the first two years of college and save thousands off your degree. Anyone can earn credit-by-exam regardless of age or education level.
To learn more, visit our Earning Credit Page
Not sure what college you want to attend yet? Study.com has thousands of articles about every imaginable degree, area of study and career path that can help you find the school that's right for you.
Back To CoursePsychology 102: Educational Psychology
10 chapters | 123 lessons | 9 flashcard sets