When developing a research study, one of the first things that you need to do is clarify all of the units (also referred to as cases) that you are interested in studying. Units could be people, organizations, or existing documents. In research, these units make up the population of interest. When defining the population, it’s really important to be as specific as possible.
The problem is it’s not always possible or feasible to study every unit in a population. For example, you might be interested in American college students’ attitudes about owning houses. It would obviously be too time-consuming and costly to collect information from every college student in the United States. In cases like these, you can study a portion or subset of the population called a sample. The process of selecting a sample needs to be deliberate, and there are various sampling techniques that you can use depending upon the purpose of the research.
Prior to selecting a sample you need to define a sampling frame, which is a list of all the units of the population of interest. You can only apply your research findings to the population defined by the sampling frame.
Qualities of a Good Sampling Frame
You can’t just use any list you come across! Care must be taken to make sure your sampling frame is adequate for your needs.
- Include all individuals in the target population.
- Exclude all individuals not in the target population.
- Includes accurate information that can be used to contact selected individuals.
Other general factors that you would want to make sure you have:
- A unique identifier for each member. This could be a simple numerical identifier (i.e. from 1 to 1000). Check to make sure there are no duplicates in the frame.
- A logical organization to the list. For example, put them in alphabetical order.
- Up to date information. This may need to be periodically checked (i.e. for address changes).
In some cases, it might be impossible, or very difficult, to get a sampling frame. For example, getting a list of prostitutes in your city isn’t likely (mostly because of the fact that most prostitutes won’t want to be found). Sometimes, techniques like snowball sampling must be used to make up for the lack of sampling frame. Snowball sampling is where you find one person (or a few people) for your survey or experiment. You then ask them to find someone else who would be willing to participate. Then that person finds someone else, and so on, until you have enough people for your needs.