Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: sample size


From   Ronán Conroy <rconroy@rcsi.ie>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: sample size
Date   Thu, 28 Oct 2004 10:35:55 +0100

Anuja wrote:

Hi Statalist,

I have merged two datasets, one with a sample size of
10,696 observations and the other with n<1000.

However when I run a regression, the variables are not
significant coz of the sample size.
Do you mean that there are variables in the large dataset which predict the outcome you are interested in but the same variables do not predict outcome in the smaller set?

There are many things you don't tell us. Are the effect sizes (regression coefficients) similar in both samples, but the confidence intervals wider in the small sample (in which case the lack of statistical significance could well be becoz of the sample size) or are the effect sizes different also.

And the two samples, are they from the same or different populations? Why combine them?


Should I run two seperate regressions? or is there a
way I link both of them, as the dep var is the same.



I think we need more detail before any practical suggestions can emerge.

Many thanks,
Anuja
=====
WE ARE ALL IN THE GUTTER,SOME OF US LOOK AT THE STARS.




__________________________________
Do you Yahoo!?
Yahoo! Mail Address AutoComplete - You start. We finish.
http://promotions.yahoo.com/new_mail *
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/




--

Ronan M Conroy (rconroy@rcsi.ie) Senior Lecturer in Biostatistics Royal College of Surgeons Dublin 2, Ireland +353 1 402 2431 (fax 2764) -------------------- Just say no to drug reps http://www.nofreelunch.org/

*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/




© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index