Question

If i am selecting subjects to be in my study, i necessarily must do sampling without replacement (i can not have the same person in my study twice). What effect does this have on the sample selection process?

Answer 1

I can't understand your question could u write again please

Answer 2

There are major differences between Sampling With Replacement and Sampling Without Replacement

Explanation:

• Sampling with replacement (replacement sampling): Consider a population of card stacks containing 12, 13, 14, 15, 16, 17, or 18 cards, all of which are equally likely. Assume that each integer in this population has exactly one stack. As a result, the population as a whole now possesses seven stacks. If I have two replacement samples, I pick one first (say 14). I only had a one-in-seven probability of choosing that one. Following that, I replace it. After that, I pick a new one. Each of them still has a one-in-seven chance of being chosen. And there are a total of 49 alternatives to choose from (assuming we distinguish between the first and second.)
• Taking samples without replacing them is referred to as "sampling without replacement." Consider a population of card stacks with 12, 13, 14, 15, 16, 17, or 18 cards in each. Assume that each integer in this population has exactly one stack. As a result, the population as a whole now possesses seven stacks. If I had to select between two samples with no substitutes, I'd go with the first (say 14). I only had a one-in-seven probability of choosing that one. After that, I pick a new one. At this point, there are just six options: 12, 13, 15, 16, 17, and 18. As a result, there are only 42 viable outcomes (again assuming that we distinguish between the first and the second.)

• THEREFORE, THE DIFFERENCES ARE:

1. When we sample using replacement, the two-sample outcomes are independent.
2. In practice, this means that the first's outcomes have no influence on the second's outcomes.
3. This means that the mathematical covariance between the two is zero.
4. In sampling without replacement, the two-sample findings are not independent.
5. This implies that what we got on the first has an effect on what we can get on the second.
6. This indicates that the mathematical covariance between the two isn't zero.