Which Statistical methods should i use? - (Nov/20/2005 )

Dear all,

I have two independent and non normalized samples where one sample has n=30 and the other has n= 1. Due to both samples don't have a normality distribution so i have to use non-parametric methods. I plan using Mann-Whitney Test, but then it needs that both samples have to be larger than n=5 or both are lesser than n=5. In my case then i have n=30 and the other is n=1. What kind of statistical methods should i use to get a reliable answer?


I thank you alot!

Hope for a quick answer!

---

hi
i'm afraid i'm not too sure about getting your exp.
You mean in first try you had 30 samples and in the second try you only have one sample?...
You can't do a comarison based on only one sample...

please clear my understanding...

---

If Fred 33 is correct, and you are trying to compare to a sample with n=1, I dont think there is anything that would be reliable. You have no way to even estimate the variability in the second sample, so I think you are stuck.

---