When you want to study a process, you can’t test every output (All the population). Sampling helps to get a representative subset of the population.
There are two main kinds of sampling, that are:
- Non-probability sampling;
- Probability sampling.
In non-probability sampling, the probability of getting one element of a population is not the same for all the elements. One non-probability sample is sequential sampling. In this method, we ordered the item (for example, the newest first), and we got the first N elements. In this way, the newest element has more chances to be selected.
In probability sampling, all the population elements have the same probability of being selected. For this reason, they are preferred. Some of the more used probability sampling methods are:
- Simple random: we assign at each output a number, and then we generate, by an algorithm, some random number;
- Stratified sampling: Similar of the simple random, but in this case, we can divide the population into some non-overlapping groups (every group are homogeneous) and then we made a “simple random sample” for each of one;
- Clustered sampling: You divide the population into cluster, then you randomly sample the entire cluster.
For the exam you need to remember that:
- Sampling helps to extract a representative subset of the entire population;
- In non-probability sampling, the element of the population hasn’t the same probability to be selected;
- In probability sampling, each element of the population has the same probability to be selected.