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 on April 23, and its replacement, statalist.org is already up and running.


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

Re: st: Re: R: Minimal Detectable Difference


From   Michael McCulloch <mm@pinestreetfoundation.org>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Re: R: Minimal Detectable Difference
Date   Tue, 5 Jun 2012 08:16:44 -0700

Lars,
See the user-written SAMPLESIZE:
"samplesize does everything that sampsi does but in addition allows multiple values for every numeric parameter e.g. the number of follow-up measurements. The resulting values for either power or sample size are plotted in a twoway graph."


Sincerely,

Michael McCulloch, LAc MPH PhD
Pine Street Foundation
124 Pine St,
San Anselmo, CA 94960-2674
tel:		415-407-1357
fax:		206-338-2391
email: 	mm@pinestreetfoundation.org



On Jun 5, 2012, at 6:33 AM, William Buchanan wrote:

> Hi Lars,
> 
> It may not be Stata, but you could also take a look at G*Power.  It's free and can be used to calculate various aspects of statistical power (e.g., required sample size to detect a given effect size, minimum detectable effect size, etc...).  It doesn't calculate these sizes for SEM or MLM, but for most other models you can get quick and easy answers.
> 
> - Billy
> 
> Sent from my iPhone
> 
> On Jun 5, 2012, at 5:41, Lars Folkestad <lfolkestad@health.sdu.dk> wrote:
> 
>> That is what i ended up doing. But i hoped there was a nicer/smerter/more
>> standardized way to do it.
>> lars
>> 
>> Den 05/06/12 14.06 skrev "carlo.lazzaro@tiscalinet.it"
>> <carlo.lazzaro@tiscalinet.it>:
>> 
>>> 
>>> 
>>> 
>>> Lars can probably simulate some scenarios with -sampsi-. I find this
>>> approach particularly useful when the limited number of participants
>>> available is due to budget constraints (as it is often the case, sadly).
>>> Kindest Regards,
>>> Carlo
>>> 
>>> 
>>> -----Messaggio originale-----
>>> Da: owner-statalist@hsphsun2.harvard.edu
>>> [mailto:owner-statalist@hsphsun2.harvard.edu] Per conto di Lars Folkestad
>>> Inviato: martedì 5 giugno 2012 13:39
>>> A: statalist@hsphsun2.harvard.edu
>>> Oggetto: st: Minimal Detectable Difference
>>> 
>>> Dear List
>>> 
>>> This might be a bit more of a stats question than a stata question.
>>> 
>>> I am about to do a cross sectional study comparing two groups regarding a
>>> measured continous variable.
>>> I know the the variance of the measurement. No one has ever investigated
>>> these groups before using this or similar machines. For practical reasons
>>> i
>>> only have a limited number of participants available for this project. I
>>> am
>>> now wondering what the minimal detectable difference between the two
>>> groups
>>> are with the number of participants i have available.
>>> 
>>> I will use a t-test (paired, seeing the two groups are matched)
>>> 
>>> How do i calculate this using stata?
>>> 
>>> lars
>>> 
>>> *
>>> *   For searches and help try:
>>> *   http://www.stata.com/help.cgi?search
>>> *   http://www.stata.com/support/statalist/faq
>>> *   http://www.ats.ucla.edu/stat/stata/
>>> 
>> 
>> 
>> *
>> *   For searches and help try:
>> *   http://www.stata.com/help.cgi?search
>> *   http://www.stata.com/support/statalist/faq
>> *   http://www.ats.ucla.edu/stat/stata/
> 
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   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   |   Site index