Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: t test in stata


From   Richard Goldstein <richgold@ix.netcom.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: t test in stata
Date   Fri, 15 Feb 2013 14:09:56 -0500

Hi,

you might also want to read the following:

Diehr, P., et al. (1995), "Breaking the matches in a paired t-test for
community interventions when the the number of pairs is small",
_Statistics in Medicine_, 14: 1491-1504

I have no idea whether your situation matches the conditions in this article

Rich

On 2/15/13 1:59 PM, Daljit Dhadwal wrote:
> Hi Caroline,
> 
> I was recently looking into this issue as well: what to do when you
> have a mix of paired and unpaired data. I came across the following
> paper that you may find helpful. I haven’t implemented the method it
> describes though. You may also find the references helpful.
> 
> Comparing proportions in overlapping samples: An unpublished paper by
> J. Martin Bland and Barbara K. Butland
> 
> Link:  http://www-users.york.ac.uk/~mb55/overlap.pdf
> 
> Abstract:
> 
> Sometimes we want to estimate the difference in proportions between
> two groups where some subjects appear in both. We present an approach
> which does not require the assumption that the proportions in the
> overlapping and non-overlapping samples are the same, and which can be
> extended very easily to comparisons of means, odds, etc. The method
> has the disadvantage that we need each group to contain some subjects
> observed once only. We illustrate the method with an example from the
> UK National Child Development Study, and compare the results with
> other methods.
> 
> Thanks,
> 
> Daljit
> 
> 
> On Fri, Feb 15, 2013 at 9:55 AM, Caroline Wilson <wilson_cj@hotmail.com> wrote:
>> Hi everyone,
>>
>> I'm trying to determine whether there is a statistically significant change in two proportions between two years, 2005 and 2010. I understand I can use a t-test for this, provided the sample size is large enough. However, I'm not sure whether I should conduct a paired t-test or an unpaired t-test because there are some of the same subjects in both years, but for the most part the two groups represent different subjects.  In this case, is it better to conduct paired t-test because at least some of the same people are in both years?
>>
>> Any advice would be much appreciated!
>> Carrie
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/


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