In the fields of statistics, survey methodology and quality assurance, the term sampling is used to refer to a subset’s selection (a statistical sample) of individuals that are found within a statistical population for purposes of estimating the characteristics of the entire population. Statisticians try for the samples that represent the entire population in question. This practice has two major advantages. They are:
- Sampling has lower cost (saves on money).
- Sampling enables faster collection of data. Unlike measuring the whole population, sampling involves much lesser time.
Every observation measures a single or more properties (like color, location and weight) of bodies that are observable which are distinguished as independent individuals or objects. In survey sampling, weights may be applied to the data for purposes of adjusting for the sample design, especially stratified sampling.
The reliability of the results of a research is dependent on the manner in which the sample was selected. A sample is supposed to be a true representative of the entire population. The sample should include representatives from different spheres and sections of the population so as to become a population’s true representative.
Some of the terminologies that apply in sampling are discussed below. They are:
- Sample. This refers to that part of the population that is selected.
- Sample size. This refers to the number of items that are there in the sample that is selected.
- Sampling frame. This is a list of items or individuals that are included in the sample.
- Sampling technique. This refers to the procedure that is applied in the selection of the sample members.
TYPES OF SAMPLING.
The major types of sampling are two. They are probability sampling and non-probability sampling. They are however sub-divided into sub-types.
PROBABILITY SAMPLING.
This is a sampling type where every population’s member has a probability that is known of being selected. In a highly homogenous population, every member has a chance of being picked in the sample, this chance is known. The types of probability sampling are:
- Simple random sampling. This is where the sample members are picked randomly by chance. Since all the members have an equal chance of selection, random member selection doesn’t affect the sample’s quality.
- Stratified random sampling. In this sampling, the population is first divided into sub-groups which are known as strata. After this, that’s when members are randomly selected from the sub-groups.
- Systematic sampling. This is where a member that occurs after a certain fixed interval is selected. For example: 5, 10, 15, 20………
- Cluster sampling. This is where population segments are taken as clusters then the members from all the clusters are randomly selected.
- Multi-stage sampling. In this sampling method, every sample’s cluster is further sub-divided into smaller clusters then members are randomly selected from the smaller clusters.
NON-PROBABILITY SAMPLING.
This is a type of sampling where all the members of a population do not have a known probability of selection. The types of this sampling are:
- Purposive sampling. This is a sampling type where the sample members are picked with regard to the study’s purpose.
- Convenience sampling. This is a sampling method where the sample members are picked with regard to their convenient accessibility.
- Snow-ball sampling. It is also referred to as the chain sampling. It is a sampling method where a respondent is identified by another respondent. It is applied in situations where there is a difficulty in identification of the sample members.
- Quota sampling. This is the sampling type where the selection of members is done according to specific characteristics that are chosen by the researcher.
