How to do your own experiments in psychology and education
5.2a Define and find your participants
Who will participate in your experiment? What characteristics do they need to have? Why did you choose these people?
For this course, you have little choice with respect to the participants: you need to study normal, consenting adults, usually university students. You can’t study members of at-risk populations: children, the elderly, criminals, etc. This is essentially to avoid the protocol of getting official permission to study them.
On the other hand, one of your factors may be a difference between different kinds (or characteristics of university students: Gender, First Language, Major, etc. This is an interesting option. Remember to use a background questionnaire to be able to document that your participants had different (or similar) characteristics.
Think of reasons why you chose the participants that you chose.
How many participants do you need?
Calculate at least 15 participants per condition of your experimental design. Note that you don’t need the same exact number of participants in each condition. The statistics program will correct for small differences automatically. On the other hand, the numbers shouldn’t be very different either.
How many groups of participants do you need?
It’s likely that you’ll have to run your experiment more than once to get enough participants in each of your experimental conditions. Each time you run your experiment, it’s called a session. Depending on your design, you may need to run each condition separately, or you may be able to run two or more conditions during the same session. You need to know how many sessions you’ll need so that you can figure out how many groups of participants you have to find.
Note that it’s very important to run the experiment the same exact way each time, so that the data are comparable. More on that later.
Where will you find these participants?
You can approach professors to “borrow” their students in class, recruit your friends, or ask passers-by on campus to participate.
If your methods allow it, try to collect data from a whole group of people, for example a whole class, at the same time. Sometimes a former professor will give you access to his or her class for this. Always offer to go back and explain what you did and why!
Contact your potential sources of participants NOW.
Plan as far ahead as possible. It’s normal to have refusals, cancellations, bureaucracy, and no-shows. Be ready with additional sources of participants. You only have a window of two to three weeks for data collection.
For this course, you have little choice with respect to the participants: you need to study normal, consenting adults, usually university students. You can’t study members of at-risk populations: children, the elderly, criminals, etc. This is essentially to avoid the protocol of getting official permission to study them.
On the other hand, one of your factors may be a difference between different kinds (or characteristics of university students: Gender, First Language, Major, etc. This is an interesting option. Remember to use a background questionnaire to be able to document that your participants had different (or similar) characteristics.
Think of reasons why you chose the participants that you chose.
How many participants do you need?
Calculate at least 15 participants per condition of your experimental design. Note that you don’t need the same exact number of participants in each condition. The statistics program will correct for small differences automatically. On the other hand, the numbers shouldn’t be very different either.
How many groups of participants do you need?
It’s likely that you’ll have to run your experiment more than once to get enough participants in each of your experimental conditions. Each time you run your experiment, it’s called a session. Depending on your design, you may need to run each condition separately, or you may be able to run two or more conditions during the same session. You need to know how many sessions you’ll need so that you can figure out how many groups of participants you have to find.
Note that it’s very important to run the experiment the same exact way each time, so that the data are comparable. More on that later.
Where will you find these participants?
You can approach professors to “borrow” their students in class, recruit your friends, or ask passers-by on campus to participate.
If your methods allow it, try to collect data from a whole group of people, for example a whole class, at the same time. Sometimes a former professor will give you access to his or her class for this. Always offer to go back and explain what you did and why!
Contact your potential sources of participants NOW.
Plan as far ahead as possible. It’s normal to have refusals, cancellations, bureaucracy, and no-shows. Be ready with additional sources of participants. You only have a window of two to three weeks for data collection.
| Read this topic next: | Choose a sample of the possible participants |
About
© 2009 Created by Mike Dillinger, PhD on Ning. Create a Ning Network!
