Select a random sample of units from a sampling frame of a clustered population
(for example, animals with different owners in a village, pigs in pens in a piggery, cattle in pens in a feedlot, etc).
A full sampling frame of all units is generated with units in each cluster
numbered sequentially. Simple random sampling is then done to generate a sample
with units identified by cluster id and then by sequential number for each cluster.
Sampling can be based on either a specified number of units to be sampled or a specified
proportion of the total number of units in the sampling frame. Sampling can also be with or without (default) replacement.
A list of all clusters and the number of units in each cluster is required for
generation of the sampling frame. This can be copied and pasted from excel or other software
and should be formatted as two columns: the first column is the unique cluster id and the second
is the number of units in each cluster.
Sample size should be entered as an integer number of units to be sampled or a proportion
(>0 and <1) of the total population to be sampled.
Download and check out the example data if you are unsure what is required.
Please note: very large sampling frames may take several minutes to process, so please be patient.
Paste columns of data for the sampling frame in the space below and click on submit. The first column
must be a unique id for each cluster (owner, pen number etc), the second the number of units for
each cluster. Include
a header row containing column headings for the data.
Download example data
About this site
| 87 recent calculations
This site was created by Ausvet
with funding from a range of sources.It provides a range of epidemiological tools for the use of researchers and epidemiologists, particularly in animal health. Please send
any comments, questions or suggestions to Evan Sergeant
© 2018 Ausvet