Stratified random sampling differs from Cluster - random sampling

 Stratified random sampling differs from Cluster - random sampling

 Stratified random sampling

Stratified sampling is a type of probability sampling that divides a population into strata and draws members of the sample at random from these strata. In stratified sampling, the strata must be homogeneous, collectively exhaustive, and mutually exclusive. The strata must define a subset of the population.

Furthermore the sample members must be distinct, that is, each element must belong to one and only one stratum in the population. This implies that the samples must include the entire population. In each stratum, simple

Examples of Stratified Sampling

Stratified sampling is more representative of the population. When you anticipate that the subgroup you're studying will have different mean values for variables, stratified sampling is a better sampling method to use.

Eg: You want to know how an MBA degree can affect the wage disparities between different gender identities in your company. Based on the employee list, you estimate that only a small percentage of the company's employees are MBA graduates. So you use stratified sampling to compare men, women, and other gender identities who have an MBA to those who do not.

Cluster - random sampling

Cluster sampling is a technique used in market research tools when homogeneity is external but heterogeneity is internal within clusters/groupings. It is the process of dividing a population into multiple groups/clusters.

Cluster sampling is commonly used to reduce the number of interviews and costs while maintaining the desired accuracy. In a fixed sample size, the chance of random error is reduced when the majority of heterogeneity is internal within the group. The total population is divided into clusters, and a simple random sample of the cluster is chosen, followed by a sample of the elements in each of these clusters.

Cluster sampling methods include single-stage, two-stage, and multi-stage sampling. The number of steps required to create the desired sample determines the method used.

Examples of Cluster Sampling 

Because it is less expensive, cluster sampling is advantageous when a large population requires a survey, as previously stated. As a result, one type of cluster sampling is area sampling. It is also used for estimation in high-mortality cases such as wars, famines, and natural diseases.

Respondents are divided into clusters within a local area during the cluster sampling process. However, precision in the estimate is also required, for which the sample size must be increased.

1. One-stage cluster sampling example: Assume the retail store where you work wants to know how many people in town buy their product.

The shop divides the town into neighborhoods and randomly selects people to form cluster samples. Everyone in the neighborhood can take part in the study.

2. Two-stage cluster sampling example: Assume a restaurant owner wants to know how all of its restaurant branches are performing. The owner groups the branches based on their location and then randomly selects samples from the clusters to study the performance.

All the above information highlights the difference between the two categories of Sampling.

• Cluster sampling is done on a population of clusters, so each cluster/group is considered a sampling unit.
• Elements within each stratum are sampled in Stratified Sampling.
• Only selected clusters are sampled in Cluster Sampling.
• A random sample is drawn from each stratum in Stratified Sampling.
• The goal of Cluster Sampling is to reduce sampling costs while increasing sampling efficiency.
• The goal of Stratified Sampling is to increase precision in order to reduce error.