Probability Sampling
Probability sampling is the method that is based on the premise that all the members of the population have an equal possibility of being chosen for study or observation. It usually includes such main sampling types as simple random, systematic, stratified, and cluster samplings (Etikan and Bala, 2017). The former implies that all the people of the population are chosen at random and have an equal possibility of being selected. On the other hand, systematic sampling assumes that the first member of the sample is determined randomly, whereas other members are chosen following the fixed interval. In stratified sampling, the population is divided into subgroups or strata based on certain crucial (for the study) factor(s) which are more homogenous than the overall population. After that, the sample is chosen randomly from each subgroup. Cluster sampling is similar to stratified sampling but usually stratifies people according to their residence, and then cluster(s) that will be included in the study are selected at random. This method is good when the budget and time of the study are restricted.
Non-Probability Sampling
Conversely to probability sampling, non-probability sampling assumes that not all the members of the greater population have an equal chance to be chosen as a sample. This method encompasses quota, accidental, judgmental, and snowball sampling types (Etikan and Bala, 2017). Quota sampling is similar to stratified sampling, but usually, researchers can choose any member with the fitting characteristics at will. Accidental sampling implies that members are selected at will (by convenience) without prior stratification. Judgmental sampling refers to the choice of members based on researchers’ judgment concerning their (members) ability to provide the necessary information. Finally, snowball sampling is based on the determination of members through networks. As such, the researcher finds the first candidate for the study who, in turn, can provide information concerning other members suitable to be included in the sample.
Stratified Sampling
There are two main advantages of stratified sampling, namely high precision and lower cost. As for the former, stratification ensures that all the important groups are going to be adequately represented in the sample. As a consequence, usually, the increase in precision leads to the reduction of the sample population. Thus, this type of probability sampling is better to use when the general population can be divided into non-overlapping groups.
Reference
Etikan, I., & Bala, K. (2017). Sampling and sampling methods. Biometrics & Biostatistics International Journal, 5(6), 00149. Web.