Multistage Sampling

Community Blog

Keep up-to-date on postgraduate related issues with our quick reads written by students, postdocs, professors and industry leaders.

Multistage Sampling

Multistage Sampling explained with Multistage Sample

What is Multistage Sampling?

Multistage sampling, also called multistage cluster sampling, is exactly what it sounds like – sampling in stages.

It is a more complex form of cluster sampling, in which smaller groups are successively selected from large populations to form the sample population used in your study. Due to this multi-step nature, the sampling method is sometimes referred to as phase sampling.

In short, multistage sampling works as follows:

First, a random group of one class is selected, for example, US states. Then, within these groups, a random sample of smaller sub-groups is selected, for example, cities or districts; this continues until you reach the smallest level of sub-groups you need, for example, towns. Afterwards, a sample of these smallest sub-groups can be randomly selected to form the population for your study.

Multistage sampling makes data collection more practical for large populations,  especially when a complete list of all elements of a population does not exist or isn’t suitable. It is also often used when costs and implementation time need to be minimised.

Steps for Carrying out Multistage Sampling Technique

When applying multi stage sampling to research studies, it should be implemented in four steps:

  1. Define the first sampling frame, assign a unique number to each group in the population, and select a small sample of discrete clusters to form your primary sampling units.
  2. Define a second sampling frame for the primary sampling units selected in step 1 and then select random samples from these.
  3. If necessary, repeat the previous step.
  4. Select the final sample group from the sub-groups using a form of probability sampling, such as simple random sampling or systematic sampling.

Multistage Sampling Example

A research firm in the UK conducted a survey in which it divided the country into its counties and randomly selected some of these counties as a cluster sample (the first stage of sampling).

Each county was then divided into its towns, and areas were chosen at random from each town (the second stage of sampling).

Finally, within each town, each town was divided into small areas and households were selected at random from each area. These households formed the sample population for the research study (third stage of sampling).

Advantages of Multistage Sampling

  • Practical for primary data collection for large populations that are geographically dispersed.
  • Reduces the costs and time associated with data collection.
  • Provides flexibility, as researchers can break down the population as often as necessary to create the sample population they need.
  • Allows each stage to use its own sampling method, whether it be stratified sampling, cluster sampling or simple random sampling.
  • Typically more accurate than using cluster sampling with the same sample size.

Disadvantages of Multistage Sampling

  • It introduces a considerable degree of subjectivity, based on the sampling design that surrounds the formation of the sub-groups and their selection.
  • The sample will not be 100% representative of the entire population, and there is the potential for biases if there is little variance between members in a sub-group. If, for example, we want to study the vaccination rate in a city, we may divide the city into its towns, and then randomly select households from each town and count the number of vaccinated children within them. Logically, however, it’s likely that when one child in a household is vaccinated, their siblings will also be vaccinated; this can distort the results.
  • To form the sampling frames for multistage random sampling, group-level information is required, sometimes at a national level depending on the target population.
  • Typically not as accurate as using simple random sample with the same sample size.
Share on facebook
Share on twitter
Share on linkedin
Share on pinterest



Join thousands of other students and stay up to date with the latest PhD programmes, funding opportunities and advice.

Browse PhDs Now

Other Posts
DiscoverPhDs procrastination trap
Ending the Procrastination Trap

Are you always finding yourself working on sections of your research tasks right up until your deadlines? Are you still finding yourself distracted the moment you step foot inside your lab? Are you deliberately finding a hundred and one reasons not to get on with your literature review? Everyone faces

Read More »
Amy Tucker

Amy recently entered her third and final year of her PhD at the University of Strathclyde. Her research has focussed on young people’s understanding of mental health stigma in Scotland.

Read More »
Dr Zoë Ayres

Dr Ayres completed her PhD at the University of Warwick in 2017, researching the use of diamond to make electrochemical sensors. She is now a research scientists in the water industry, developing different analytical techniques and sensors to help keep our water systems safe.

Read More »
Charlene N. Rivera Bonet

Charlene is a 5th year PhD candidate at the University of Wisconsin-Madison. She studies depression and neuroticism in people with Temporal Lobe Epilepsy (TLE) using MR Imaging and behavioural tests.

Read More »

Browse PhDs Now

Join Thousands of Students

Join thousands of other students and stay up to date with the latest PhD programmes, funding opportunities and advice.