Research Tips
Probability Sampling Techniques
As discussed in the tip “Probability versus Non-Probability Samples”, for most types of quantitative research it’s best to aim for a probability sample. There are many different ways to draw a probability sample. Some of the most common are:
- Simple random sampling in which every member of the population has an equal chance of inclusion. An example would be using a random number generator to select customers from a company’s data base for inclusion in a survey.
- Systematic sampling in which members of the population are ordered in some way and then those at equal intervals are selected. An example would be selecting every tenth customer in a company’s database after selecting a random number between 1 and 10 as a starting point.
- Stratified sampling in which the population is divided up into groups and then simple random sampling is used to select within each group. An example would be dividing a company’s database into high-, medium- and low-volume customers and then using random sampling to select within each of these groups.
It’s important to remember that the formulas used for estimating population characteristics, determining minimum sample sizes, determining confidence intervals, etc. vary depending on the sampling technique used.