In statistics, two of the most common methods used to obtain samples from a population are **cluster sampling** and **stratified sampling**.

This tutorial provides a brief explanation of both sampling methods along with the similarities and differences between them.

**Cluster Sampling**

**Cluster sampling** is a type of sampling method in which we split a population into clusters, then randomly select some of the clusters and include all members from those clusters in the sample.

For example, suppose a company that gives whale-watching tours wants to survey its customers. Out of ten tours they give one day, they randomly select four tours and ask every customer about their experience.

This is an example of **cluster sampling**.

**Stratified Sampling**

**Stratified sampling** is a type of sampling method in which we split a population into groups, then randomly select some members from each group to be in the sample.

For example, suppose a high school principal wants to conduct a survey to collect the opinions of students. He first splits the students into four stratums based on their grade – Freshman, Sophomore, Junior, and Senior – then selects a simple random sample of 50 students from each grade to be included in the survey.

This is an example of **stratified sampling**.

**Similarities & Differences**

Cluster sampling and stratified sampling share the following **similarities:**

- Both methods are examples of
*probability sampling methods*– every member in the population has an equal probability of being selected to be in the sample. - Both methods divide a population into distinct groups (either clusters or stratums).
- Both methods tend to be quicker and more cost-effective ways of obtaining a sample from a population compared to a simple random sample.

Cluster sampling and stratified sampling share the following **differences:**

- Cluster sampling divides a population into groups, then includes
*all*members of*some*randomly chosen groups. - Stratified sampling divides a population into groups, then includes
*some*members of*all*of the groups.

**When to Use Each Sampling Method**

There is a simple rule of thumb we can use to decide whether to use cluster sampling or stratified sampling:

If a population is **heterogeneous** (i.e. there are natural differences between individuals) then it’s best to use **stratified sampling** to obtain a random sample.

- In our previous example with high school students, the students could naturally be divided into four groups based on grade. Thus, it made sense to include some students from each grade in the sample to get a representative sample of all students in the school.

If a population is **homogeneous** (i.e. there are no noticeable differences between individuals) then it’s best to use **cluster sampling** to obtain a sample.

- In our previous example with whale-watching tours, there were no clear differences between one group of customers and the next. Thus, it made sense to just randomly choose some groups and include all customers from those chosen groups to be included in the sample.

Keep this rule of thumb in mind when deciding whether to use stratified sampling or clustering sampling.

**Additional Resources**

Introduction to Sampling Methods

What is Multistage Sampling?

What is Two-Stage Cluster Sampling?

What is Sampling Variability?