Dear All,
In this query when I said t test, I meant paired
t-test.
Regards,
Indrajit
--- Indrajit Sengupta <[log in to unmask]> wrote:
> Dear All,
> I have a question regarding calculation of sample
> size
> for t-tests.
> Now for doing the above the formula I have is like
> this,
>
> N = 2 * [s(x1-x2)]^2 * (1-r) * [z(1 - a/2) + z(1 -
> b)]^2/(m1-m2)^2
>
> where,
> s(x1-x2) : Estimated std of difference of means;
> r : correlation coefficient between x1 and x2;
> z(t) : Lower t% pt of normal distribution;
> a : Type 1 error;
> b : Type 2 error;
> m1 : Mean of x1;
> m2 : Mean of x2;
>
> I have come across a calculator in website which
> requires only the following parameters :
> 1. Type 1 error;
> 2. Type II error or power;
> 3. Estimated Standard Deviation of differences;
> 4. Minimum detectable difference (m1-m2);
>
> Could anybody provide me with the formula used to
> calculate the sample size in such a situation?
>
> Thanks in advance,
> Regards,
>
>
>
> Indrajit Sen Gupta
> Business Analyst
> Mu Sigma Business Solutions Pvt Ltd
> #21/1-1, Nawab Towers
> Cunningham Road
> Bangalore 560 052, India
> Tel: +91 80 5123 2551
> Fax: +91 80 5123 8853
> Mob: +91 98807 54816
> [log in to unmask]
> www.mu-sigma.com
>
> __________________________________________________
> Do You Yahoo!?
> Tired of spam? Yahoo! Mail has the best spam
> protection around
> http://mail.yahoo.com
>
__________________________________________________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
|