To conduct a cluster sample, the researcher first selects groups or clusters and then from each cluster, selects the individual subjects either by simple random sampling or systematic random sampling. Or, if the cluster is small enough, the researcher may choose to include the entire cluster in the final sample rather than a subset from it.
One-Stage Cluster Sample
When a researcher includes all of the subjects from the chosen clusters into the final sample, this is called a one-stage cluster sample. For example, if a researcher is studying the attitudes of Catholic Church members surrounding the recent exposure of sex scandals in the Catholic Church, he or she might first sample a list of Catholic churches across the country. Let’s say that the researcher selected 50 Catholic Churches across the United States. He or she would then survey all church members from those 50 churches. This would be a one-stage cluster sample.
Two-Stage Cluster Sample
A two-stage cluster sample is obtained when the researcher only selects a number of subjects from each cluster – either through simple random sampling or systematic random sampling. Using the same example as above in which the researcher selected 50 Catholic Churches across the United States, he or she would not include all members from those 50 churches in the final sample. Instead, the researcher would use simple or systematic random sampling to select church members from each cluster. This is called two-stage cluster sampling. The first stage is to sample the clusters and the second stage is to sample the respondents from each cluster.
Advantages of Cluster Sampling
One advantage of cluster sampling is that it is cheap, quick, and easy. Instead of sampling the entire country when using simple random sampling, the research can instead allocate resources to the few randomly selected clusters when using cluster sampling.
A second advantage to cluster sampling is that the researcher can have a larger sample size than if he or she was using simple random sampling. Because the researcher will only have to take the sample from a number of clusters, he or she can select more subjects since they are more accessible.
Disadvantages of Cluster Sampling
One main disadvantage of cluster sampling is that is the least representative of the population out of all the types of probability samples. It is common for individuals within a cluster to have similar characteristics, so when a researcher uses cluster sampling, there is a chance that he or she could have an overrepresented or underrepresented cluster in terms of certain characteristics. This can skew the results of the study.
A second disadvantage of cluster sampling is that it can have a high sampling error. This is caused by the limited clusters included in the sample, which leaves a significant proportion of the population unsampled.
Let’s say that a researcher is studying the academic performance of high school students in the United States and wanted to choose a cluster sample based on geography. First, the researcher would divide the entire population of the United States into clusters, or states. Then, the researcher would select either a simple random sample or a systematic random sample of those clusters/states. Let’s say he or she chose a random sample of 15 states and he or she wanted a final sample of 5,000 students. The researcher would then select those 5,000 high school students from those 15 states either through simple or systematic random sampling. This would be an example of a two-stage cluster sample.
Babbie, E. (2001). The Practice of Social Research: 9th Edition. Belmont, CA: Wadsworth Thomson.
Castillo, J.J. (2009). Cluster Sampling. Retrieved March 2012 from: http://www.experiment-resources.com/cluster-sampling.html