sciurus Posted February 8, 2016 Share Posted February 8, 2016 Hi everyone. is there anyone that knows if, when I have a subset of a population, I can I standardize the subset?? what I mean is, If I have a unknown number of eggs, I take 30 of that eggs and after I have to use what I know (data obtained from the subgroup) for that 30 eggs on all other eggs, although I do not know how many are the total eggs. How can I do it on R program??? I hope I clearly exposed my question. thank you Link to comment Share on other sites More sharing options...
Prometheus Posted February 8, 2016 Share Posted February 8, 2016 I don't follow. What do you ultimately want to do? Standardisation has several meanings between maths, statistics and science: from the context i'm guessing you mean standard score? If so this is easy in R. What have you tried so far? Link to comment Share on other sites More sharing options...
sciurus Posted February 9, 2016 Author Share Posted February 9, 2016 What I meant was: from an unknown number of eggs (maybe 100, maybe 2000) I take 50 eggs and want to see how long before they hatch. when they hatch I know that 50 eggs hatch in 3 days...can I use this number, 3 days, to refer to all the 100 or maybe 2000 eggs from the original group?? I mean, can I say that the eggs of a species hatch in 3 days, even if I just have a subgroup?? If I should use a t-test, for example, I should know the initial population number to know if the 3 days is significant or not, right?? Link to comment Share on other sites More sharing options...
Prometheus Posted February 9, 2016 Share Posted February 9, 2016 OK, i see the problem. It's difficult. You are correct that you should account for the fact you are sampling from a finite population. However, since you don't know the population number, you can't make the usual corrections. I would just assume an infinite population and proceed with the usual analyses, bearing in mind you will have a biased estimate. The result will be that you have a larger estimate for the standard error than you would otherwise have, meaning you are less likely to detect statistically significant differences in means when comparing two species. Link to comment Share on other sites More sharing options...
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now