There are two branches in statistics, descriptive and inferential statistics. Of these two main branches, statistical sampling concerns itself primarily with inferential statistics. The basic idea behind this type of statistics is to start with a statistical sample. After we have this sample, we then try to say something about the population. We very quickly realize the importance of our sampling method.

There are a variety of different types of samples in statistics. Each of these samples is named based upon how its members are obtained from the population. It is important to be able to distinguish between these different types of samples. Below is a list with a brief description of some of the most common statistical samples.

### List of Sample Types

- Random sample – Here every member of the population is equally likely to be a member of the sample. Members are chosen via a random process.
- Simple random sample – This type of sample is easy to confuse with a random sample as the differences between them are quite subtle. In this type of sample individuals are randomly obtained, and so every individual is equally likely to be chosen. It is also necessary that every group of
*n*individuals is equally likely of being chosen. - Voluntary response sample – Here subjects from the population determine whether they will be members of the sample or not. This type of sample is not reliable to do meaningful statistical work.
- Convenience sample - This type of sample is characterized by the selection of easy to obtain members from the population. Again, this is typically not a worthwhile style for a sampling technique.
- Systematic sample - A systematic sample is chosen on the basis of an ordered system.
- Cluster sample – A cluster sample involves using a simple random sample of evident groups that the population contains.

- Stratified sample - A stratified sample results when a population is split into at least two non-overlapping sub-populations.

It is important to know the distinctions between the different types of samples. For example, a simple random sample and a systematic random sample can be quite different from one another. Some of these samples are more useful than others in statistics. A convenience sample and voluntary response sample can be easy to perform, but these types of samples are not randomized to reduce or eliminate bias. Typically these types of samples are popular on websites for opinion polls.

It is also good to have a working knowledge of all of these kinds of samples. Some situations call for something other than a simple random sample. We must be prepared to recognize these situations and to know what is available to use.

### Resampling

It is also good to know when we are resampling. This means that we are sampling with replacement, and the same individual can contribute more than once in our sample. Some advanced techniques, such as bootstrapping, requires that resampling be performed.