Re: two-sample t-test question.
From: Paige Miller (paige.miller_at_kodak.com)
Date: 07/20/04
- Next message: Paige Miller: "Re: Weighting regression points?"
- Previous message: Jonathan Greenberg: "Weighting regression points?"
- In reply to: Jonathan L Jacobs: "two-sample t-test question."
- Messages sorted by: [ date ] [ thread ]
Date: Tue, 20 Jul 2004 15:58:45 -0400
Jonathan L Jacobs wrote:
> sci.stat.math:
>
> I want to determine if two sample datasets are statistically different
> (don't come from the same population) at some predefined confidence level
> (e.g. alpha=0.05) and get a p-value for this determination.
Are you implying that you want to compare the means of the two
datasets? Or do you want to compare some other aspect of the datasets?
> The two datasets in question are small (N<30) and Nx != Ny.
>
> The data is also continuous in nature (i.e. not discrete) with values
> ranging from ~0 to 1.0.
>
> The two datasets are confirmed to be independent and normally distributed.
> This is not a bivariate dataset.
> I want to use an unpaired two-sample t-test. Is this what you would
> recommend?
If your answer above is affirmative that you want to compare means,
then I guess I would have to say the two sample t-test is the proper
test.
> If this is the case, for the t-test t statistic and for determining the
> degrees of freedom, should i use the standard deviation of the sample mean
> (i.e. standard error) or an estimate of the population standard deviation?
I belive the two sample t-test formula is quite specific about what
to use. But, I suppose it also depends on which textbook you have.
You can get there either way, with the standard deviation of the
sample mean or the estimate of the population standard deviation. So
consult your favorite textbook.
-- Paige Miller Eastman Kodak Company paige dot miller at kodak dot com http://www.kodak.com "It's nothing until I call it!" -- Bill Klem, NL Umpire "When you get the choice to sit it out or dance, I hope you dance" -- Lee Ann Womack
- Next message: Paige Miller: "Re: Weighting regression points?"
- Previous message: Jonathan Greenberg: "Weighting regression points?"
- In reply to: Jonathan L Jacobs: "two-sample t-test question."
- Messages sorted by: [ date ] [ thread ]
Relevant Pages
|
