Sampling is the process whereby a section of a given population is selected to represent the whole population for research purposes. There are various methods of sampling and they vary based on the type of population and circumstances under which they are employed. The major categories are the probabilistic and the non probabilistic sampling methods. Probability can be defined as the possibility of an event occurring.
Probabilistic sampling is a method which uses random selection; it involves following a procedure which will ensure that the different units in the population have equal or the same probabilities of selection. This is mostly applied in random numbers generation by computers, picking a team from the basket which one has to face off etc.
Some types of probabilistic sampling methods include simple random sampling, stratified simple random sampling, cluster sampling and systematic sampling. Simple random sample is whereby the representative of the population is chosen based entirely by chance. Each member of the population has an equal chance of being selected and in most cases the element of the sample is avoided to be selected twice.
Stratified simple random sample is a probabilistic sampling method whereby the population is divided into smaller groups which are homogenous. These groups are called strata. Following the division, a simple random sample is the selected from each group and the outcome from these groups is brought together for inference making purposes. Another probabilistic sampling method is the c luster sampling which involves pulling a vast population together and then dividing them into clusters along geographic boundaries. This method is majorly used when one ones to cover a very large geographical area for instance a state. Clusters are then randomly sampled and measured and inference drawn. Finally we have systematic sampling where a random sample is picked systematically. The population is numbered and a decision on the sample size is a made. After this, an interval size is obtained by dividing the total number of the population by the sample size. The sample size is normally represented by the letter k. Select a random integer between one and k and then take every kth unit.
Non probabilistic sampling on the other hand is a sampling method where the population elements do not have equal chances of being selected. Other elements have high chances than others. This method is most suitable for instance when one is doing a qualitative research, where the research results will not be used as a basis for generalization of the whole population, where doing a random selection is impossible for instance in a limit less population.
Under this type of sampling there are other several types of sampling such as convenient sampling, quota sampling etc.
Convenience sampling is whereby a sample is chosen basically because they can be easily accessed by the researcher. This is the least time consuming and the easiest sampling method. Judgmental sampling is also known as purposive sampling. It involves the researcher choosing a given sample based on the purpose and his belief on their appropriateness.