Simple random and stratified Sampling is surely equally easy for a large populationWatch
I am currently just beginning the statistics course in A level Maths, and we have been told 'Simple Random sampling is not suitable when the population size is too large due to time consumption, disruption and expense' and a benefit of Stratified sampling has been said to be that it's 'Suitable for large populations' I cannot see how this is true, as it takes literally 5 seconds to generate a list of say 5000 random numbers. In the same way it would take 5 seconds to have a computer take (for example) every fifth person from a population of 25 000 in a sampling frame. These both give a sample size of 5000 in their respective sampling methods. I simply do not understand how its possible that stratified is better for large populations.
Simple random sampling of a large population would not only require you to generate 5000 random numbers out of 25,000, but you would also need to assign a unique number to each of the 25,000 members of the population. If you do not have a list of every member of the population, you would also need to spend time collating this for yourself before beginning the process of assigning a random number to each. This is why it can be time consuming and expensive.
But if the data is in software such as excel you will already have everyone numbered 1-25000, and i would assume you would have them in some kind of software that does that if you have a sample size that large, then you either pick every nth member (stratified) or use your random number generator (random) this appears to take the same amount of time? What you say about collecting a list of every member of the population has to be done for both according to the book as it says that both require a sampling table?