Imagine that we want to choose a sample of size $ n $ from a population of size $ N $. Let $ j $ be a unit contained within the population of size $ N $.

What is the inclusion probability of unit $ j $ for simple random sampling without replacement? In other words, what is the probability that our sample of size $ n $ will contain $ j $?

According to Thompson M.E. (1997):

"Since there are $ {N-1}\choose{n-1} $ samples of size $ n $ which contain $ j $ and each has probability $ 1/{{N}\choose{n}} $, then $ \pi_j = {{N-1}\choose{n-1}} \big/{{N}\choose{n}} = \frac{n}{N} $."

I understand that the probability of obtaining any given sample is $ 1/{{N}\choose{n}} $. However, I don't understand the other claim, which states that there are $ {N-1}\choose{n-1} $ samples of size $ n $ that contain unit $ j $... Any help or guidance would be appreciated. Thanks!


Side note: there is a far easier way to compute $\pi_j$. A simple random sample can be obtained by randomly permuting the $N$ units and choosing the first $n$. It should be clear that the probability that unit $j$ ends up in the first $n$ after permuting is $n/N$.

The number of samples of size $n$ which contain $j$ is simply counting the number of ways you can choose the remaining $n-1$ elements of your sample. There are $N-1$ other units to choose from, so there are $\binom{N-1}{n-1}$ ways.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.