RECRUITMENT AND SAMPLING
The people who have participated in your program or received your service would be those you approach for recruitment. You might also want to include a comparison group to strengthen your evaluation - a comparison group is a group of people who share key characteristics relevant to the outcomes that you want to measure (e.g., they are from the same age group, show similar behavioural and psychological symptoms of dementia, and/or live in the same local area etc.) yet have NOT participated in your program or received your service. This inclusion of a comparison group allows you to determine whether or not your program could be making a difference in the outcomes.
If you are working with a very large group, it might be impractical to collect data from everyone in the group. In this case, an option will be to collect data from a portion of the group. However, when you do this, it is important that the group you include is representative of the larger group, this way it will be able to provide you with similar information at a much lower cost.
You may wonder, how do I decide on the number of people to collect data from? There is no standard answer to this question, as the appropriate sample size will depend on your evaluation questions and the resources available to you. Below are four key questions that may help you decide on how large your sample size should be (Krathwohl, 1998):
Other things to keep in mind:
If you are working with a very large group, it might be impractical to collect data from everyone in the group. In this case, an option will be to collect data from a portion of the group. However, when you do this, it is important that the group you include is representative of the larger group, this way it will be able to provide you with similar information at a much lower cost.
You may wonder, how do I decide on the number of people to collect data from? There is no standard answer to this question, as the appropriate sample size will depend on your evaluation questions and the resources available to you. Below are four key questions that may help you decide on how large your sample size should be (Krathwohl, 1998):
- How precise do you want to be?
- How sure do you want to be of your answer?
- How much variation is there in the population you are studying?
- How small of an effect do you want to be able to detect?
Other things to keep in mind:
- Most of the time there will be a number of people you approach who do not respond or refuse to answer your questions. So estimate the rate of non-response and increase your sample size accordingly.
- If you are running a long-term project, there will be people who drop out of the program for various reasons (e.g., no longer require your services, move out of the area, decease etc.) Take these into account and increase your sample size.