Population sampling is a method through which a group of individuals are selected from a population for statistical analysis. The sample should have enough size to warrant statistical analysis. Conducting population sampling is very important as errors can lead to misleading data. The ideal approach would be to test every single individual to get the most accurate results. Sampling is done because it is practically impossible to test every single individual in the population and sampling is a reasonable "proxy" that also saves time, money and effort during the research. There are many techniques available for population sampling. The 2 main types of population sampling are probability sampling and Non-probability sampling.
What is Probability sampling?
In probability sampling, every single individual will have equal chance of being selected as the subject for the research. This method assures that the selection procedure is random. Probability sampling is divided into 5 primary sampling strategies. They are
Random sampling: This type of sampling is used in population sampling when analyzing historical or batch data. Random sampling is done by passing on a number to each unit in the population and using a random number table to create the sample list. Using random sampling defends against favoritism or bias being created in the sampling process.
Stratified random sampling: This type of sampling is also for analyzing historical or batch data. Stratified random sampling is used when there are different population groups (strata) and the analyst makes sure that all the groups are represented in the sample. In this sampling, independent samples are drawn from each group. The size of each sample is proportional to the comparative size of the group.
Systematic sampling: This sampling is used in process sampling circumstances when real time data is collected during process operation. Unlike population sampling, a frequency for sampling should be selected. Systemic sampling involves collecting samples according to some systemic rule. For instance, every 4th unit, every hour, the first 5 units etc. One disadvantage in this type of sampling is that the systemic rule may also match some fundamental structure which would bias the sample.
Cluster sampling: In cluster sampling, the researcher selects groups or clusters and then from each cluster he selects the individual subjects by random or systematic sampling. The most familiar cluster used in research is a geographical cluster.
Disproportional sampling: This is a probability sampling method which is used to tackle the difficulty researchers came across with the stratified samples of unequal sizes. This sampling method split the population into subgroups or strata but utilizes a sampling fraction which is not identical for all strata.
Advantages of Probability sampling:
The advantage of probability sampling is the precision of the statistical methods after the research. It is also used to find the population parameters. This sampling is considered to be a trustworthy method to eradicate the bias
What is Non-probability sampling?
In this type of population sampling, the population members will not have equal chance of being selected and so it is not safe to think that the sample completely represents the target population. Also the researcher would have intentionally chosen the individuals who will participate in the study. This sampling is carried out when the parameters of the entire population arenot required. This type of sampling is easy, cheap and quick but not accurate. Non-probability sampling can be classified into 5 types.
Convenience sampling: In this sampling subjects are selected on the convenience of the researcher.
Consecutive sampling: This is also called sequential sampling in which researcher picks a single or group of subjects, conducts his test, examine the results and move on to another group of subjects if needed.
Quota sampling: This sampling is identical to stratified samplings which includes dividing the population into classes (males and females) and then gettinga sample within each class.
Judgmental sampling: In this sampling researcher selects the subjects to be sampled based on the professional judgment.
Snowball sampling: Snowball sampling is the method in which the researcher begins sampling one person, then asks that person to refer some other people and goes on like that. This may be called chain referral sampling.
Advantages of Non-probability sampling:
Non-probability sampling is used in pilot studies, qualitative research, hypothesis development, case studies.