In order to get the most accurate results, researchers must choose and assign their subjects in a random manner. In this lesson, we'll look at random assignment, random selection, and why they are important.
Charlene is a psychologist. She's interested in whether sitting on a jury will increase the level of patriotism that people feel. She goes to her local courthouse and gives jury members from several trials a short questionnaire that tells her how patriotic they feel. After they've served on a jury, Charlene gives them the same questionnaire and then compares their answers to see if they are feeling any more patriotic than before.
Now, there are a lot of things that can affect how patriotic a person feels. They could be hit with a high tax bill and feel less patriotic, or they could talk to a veteran and feel more patriotic. How does Charlene know that the results of her study are because people served on a jury and not because something else happened to influence her results?
The truth is that Charlene can't ever be 100% sure that her results are only the result of serving on a jury. However, one thing that she can do to increase the chances that they are is randomization, or the process of randomly selecting or assigning subjects.
Why does randomization work? Let's look at a specific example. Let's say that Charlene has evidence that tall people feel more patriotic than short people. When she goes to pick out her subjects, she notices a lot of tall people there. If she selects mostly tall people, will that make a difference to her results? If height is related to feelings of patriotism, then it might.
But what if Charlene didn't choose the tall people? What if she just randomly chose people of all different heights? In that case, she's more likely to get a good mix of people that represent the population at large. Because there's a mix of people, height is not as likely to affect her results. Let's look closer at two types of randomization: random assignment and random selection.
Imagine that Charlene goes to her courthouse, and there are 10 trials going on. She only needs to poll the juries from seven of the ten trials. How does she choose?
Charlene knows that three of the trials involve cases that will more than likely make people feel frustrated or disgusted with the government. As a result, they might not feel patriotic after sitting on those trials. So, Charlene decides to ignore those trials and just choose the jurors who are in the other seven trials.
Uh-oh. Charlene hasn't used randomization, and her results might be biased. Instead, she should have used random selection, or the process of randomly choosing which subjects to include in a study. It's easy to remember because of the word 'selection.' You are selecting which people to include in your study, and you are selecting them randomly. In Charlene's case, she should have flipped a coin or drawn names out of a hat in order to choose which trials to include.
Another way that Charlene could randomly select subjects is by randomly selecting actual jurors from each of the 10 trials. Again, she could flip a coin or draw names out of a hat, or she could use an online randomizer tool to help her make sure that she was actually randomly selecting which jurors to poll.
Let's say that Charlene decides that she wants to compare the people who sit on juries to the people who just sit in the juror room but are never selected for an actual trial. She goes to the courthouse and gives everyone her patriotism poll at the beginning. Through a special (and probably illegal) arrangement, Charlene gets to decide which people are put on juries and which people are not selected for a trial. She picks out the people whose polls show that they are patriotic and sends them to serve in a trial.
But wait! She hasn't randomized, so her results might not be accurate. By choosing people who are already patriotic, she might get results that are skewed in one direction or the other. But she could use random assignment instead.
Random assignment is when a researcher randomly decides which subjects go in which groups. The word 'assignment' is key here. You are assigning subjects to groups in a random manner.
Of course, in real life, Charlene can't control who gets chosen for a jury and who sits in a jury room reading. But if she could, she would want to flip a coin, draw names or use another randomizing method to choose who is put on a jury and who isn't.
Randomization is the process of randomly selecting or assigning subjects. There are two major types of randomization: Random selection involves choosing subjects in a random manner, while random assignment involves assigning subjects to treatment groups in a random manner.
Following this lesson, you should have the ability to:
- Define randomization and explain why it is important in psychological research
- Describe two types of randomization