Different Sampling Methods
Why sample is required?
Sample as the word itself suggests is a small portion of the population that represents the whole population. The sample is used to conduct the study, draw observations, and make inferences about the whole population. Since a small fraction is representing the whole population, it should be selected sedulously to include maximum cardinal characteristics and attributes of the population. This may involve specifically targeting hard to reach clusters of population.
Get complete CFA Online Course by experts Click Here
For example: if the electoral roll for a town was used to identify participants, some people, such as the homeless, would not be registered and therefore excluded from the study by default.
It would be impractical and a costly as well as time-consuming affair to study a whole population. Sampling is a tool that enables researchers to derive knowledge and draw conclusions about a population based on data from a sample sub-set, without having to analyze the whole population. Reducing the universe of the population in a study reduces the cost and workload, and make it viable to obtain quality information, given population characteristic is carefully attributed to sampling.
Sampling Methods are predominantly are divided into two broad categories namely:
- Probability Sampling – Involves random selection, allowing you to make statistical inferences about the whole group. Probability sampling means that every item in the population has an equal chance of being included in the sample. This method of selection has the greatest freedom from bias but could be costly in terms of time and effort.
- Non-Probability Sampling –includes non-random selection based on convenience or other factors, allowing you to collect initial data easily. This is associated usually with a case study and qualitative research, not intending to represent any population at large. A clear rationale and motif exist behind the inclusion of particular cases as well as individuals in the sample study.
Probability Sampling & Non-Probability Sampling can further be divided as follows:
Probability Sampling Methods: –
1)Simple Random Sampling:
In this case, each individual is chosen entirely by chance and each member of the population has an equal chance, or probability, of being selected. One method to undertake random sampling would be if the researcher was to construct a sampling frame first and then used a random number generation computer program to pick a sample from the sampling frame.
2) Systematic sampling:
Systematic sampling is based upon sequences, say every 10th observation from the population will be selected. The convenience is its simplicity that makes this method easy to apply. Observations are picked at regular intervals to maintain uniformity.
3)Stratified sampling:
In this method, the population is divided into subgroups (or strata) with similar characteristics. This is done to ensure the participation of observations with analogous characteristics. Then the study sample is obtained from each stratum by taking equivalent sample sizes. This method ensures the inclusion of different population attributes of the population thereby reducing bias however clustering based on characteristics could be effortful.
4) Clustered sampling:
Clustering is similar to stratified sampling for that matter population is divided into small clusters however the division is random irrespective of characteristics. This is the preferred method where the population is huge and has similar characteristics.
5)Multi-Stage Sampling:
Multi-Stage sampling is a recursive process whereby starting from a large sample size we move to narrower and well-defined samples. This method divided the whole population/sample into smaller chunks.
Get complete FRM Online Course by experts Click Here
Non-Probability Sampling Methods: –
- Quota sampling:
As the name itself says, the quota sample method gives more weight to observations meeting some set quota i.e. requisition. Samples are selected based on predetermined characteristics thereby reducing the time and effort of studying population as well as sorting samples. This is done to ensure the sample purely has characteristics of the population.
- Snowball sampling:
Snowball sampling aims to increase the sample size by demonstrating the participation of the current small sample size. This is done where the population is small, niche, and closed thereby making it difficult to approach. Due to inaccessibility, it is difficult to create a sample.
- Judgment sampling:
Often known as selective or subjective sampling, this approach is based on the researcher’s judgment when choosing who to ask to take part. This allows researchers to implicitly select a “representative” sample to suit their needs; thereby portraying definitive characteristics.
- Convenience Sampling:
This type of sampling method is favored due to its easiness inapplicability, it simply chooses the data/sample available i.e. choosing just because participants are available. It is relatively easy, less time consuming, and also cheaper.
Get complete CFA Online Course by experts Click Here
Conclusion:
To conclude sampling is an essential statistical technique used for data analysis and interpretation. Today almost every field from media to health to the finance sector uses data analysis and analytics to deal with business problems and work more effectively as well as to make predictions regarding market size and nature.
Author: Varsha Bhambhani
About Author: I’m an FRM professional and CFAL-2 candidate with 3 yrs of work ex in the risk domain. My interest includes mathematics and finance. From a personal life perspective, I love health and fitness, in not studying one can find me inside the gym aka nerd gym rat.
Related:
