answersLogoWhite

0

There are at least two situations.

Consider the situation where the population consists of a number of sub-populations (strata) such that units within a sub-population are similar to one another but there are much larger differences between units from different sub-populations. In order to ensure that the sample is representative, it may be sensible to use stratified random sampling. The sampling proportion may be a constant proportion or may even be such that the variance in each stratum is similar.

The situation may also arise if the population is widely scattered geographically. Rather than expend time and money travelling all over the place, you could employ cluster sampling. Select a number of clusters of the population and then, within each cluster, carry out a census.

User Avatar

Wiki User

10y ago

Still curious? Ask our experts.

Chat with our AI personalities

TaigaTaiga
Every great hero faces trials, and you—yes, YOU—are no exception!
Chat with Taiga
RafaRafa
There's no fun in playing it safe. Why not try something a little unhinged?
Chat with Rafa
ProfessorProfessor
I will give you the most educated answer.
Chat with Professor

Add your answer:

Earn +20 pts
Q: When do you need non-probability sampling?
Write your answer...
Submit
Still have questions?
magnify glass
imp