Term
| Define population for purposes of research |
|
Definition
| The entire population you wanted information from. Ex: if you are interested in people who are depressed in the Western Hemisphere, all of your research participants will be people who are despressed in the Western Hemisphere. |
|
|
Term
|
Definition
| A listing of units (people, objects, or events) in a population from which a sample is drawn. For example, if our population is to be all the peole who are depressed in the area swerved by city transit, then we must create a list, of sampline frame, of all the people who are depressed in the area served by city transit. Often one of the hardest parts of a research study. |
|
|
Term
|
Definition
| Once we have our list of depressed people, we have our sampling frame. Next, we need to select our samples. |
|
|
Term
| What are the two main ways of selecting samples? |
|
Definition
1. Probablity Sampling
2. Nonprobability Sampling |
|
|
Term
| List and explain the types of Probability Sampling |
|
Definition
1. Simple Random Sampling (put all names in a bowl and pull out the # you need)
2. Systematic Random Sampling (100/50: Interval - take every second name from the list) (1,000/100 - choose 10th, 20th, 30th, etc.) The larger the sample the better.
3. Stratified Random Sampling - this is the most accurate. Break pop. up into smaller groups. Ex: If depression and sleeping patterns are also concerned with religious affiliation, we count how many are Christians, Jews, etc., and then sample from each category - can be proportionate or disproportionate, depending on the situation.
4. Cluster Random Sampling - Useful if there is difficulty creating our sampling frame. Ex: we want to survey all the people in the area served by city transit to see if they are satisfied w/the transit system, but we don't have a list of all of these people. We do have a list of all the communities in the city served by the transit system and we can use this list as an alternative sampling fame. This type of sampling is more prone to error. |
|
|
Term
| List and explain the types of Nonprobability Sampling |
|
Definition
*In nonprobability sampling, not all the people in the population have the same probability of being included in the sample, and the probabiliity of inclusion is unknown. Used often in exploratory studies where the purpose is just to collect as much data as possible.
1. Availability Sampling - convenience sampling. Literally use who's available as sample for study.
2. Purposive Sampling - Frequently used in qualitative research. Choose the sample you want (Ex: choose not-for-profit agencies who survived the economic down-turn well)
3. Quota Sampling - Set out how much of something you want
4. Snowball Sampling - Seen a lot in qualitative research. Useful when working w/hard-to-find populations (ex: runaways). Find one, interview, then ask them for the name of another, etc.
*Cannot generalize from these.
|
|
|
Term
|
Definition
| Leave out important people (blue M&Ms) that are a big part of your sample |
|
|
Term
|
Definition
| Chance of error: one bag could be just brown, blue, etc. |
|
|
