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: sampsi and percentages


From   Ricardo Ovaldia <ovaldia@yahoo.com>
To   "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu>
Subject   Re: st: sampsi and percentages
Date   Tue, 30 Aug 2011 11:39:24 -0700 (PDT)

Thank you Ronan. My original question was whether it makes sense to compute power and sample size using means in a study similar to the example I gave with the auto data.
You are correct that  ratio can be greater than 1, while a proportion cannot. However in my case the ratio will always be between 0 and 1. 

I am wondering wheter powering on the difference of means in such comparisons makes sense.
 
Ricardo

Ricardo Ovaldia, MS
Statistician 
Oklahoma City, OK


----- Original Message -----
From: Ronan Conroy <rconroy@rcsi.ie>
To: "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu>
Cc: 
Sent: Tuesday, August 30, 2011 10:10 AM
Subject: Re: st: sampsi and percentages

On 2011 Lún 30, at 15:45, Ricardo Ovaldia wrote:

>  
> Thank you, but these are not proportions. They are intensity measures. You can think of them as ratios of two continous things. 
> For example with the auto data, they could be the ratio of car's length  to weight (length / weight) which is always between 0 and 1.
>  

> Now less say that you want to compare these ratio between between foreign and domestic cars.
> 

I am a little confused. A ratio can be greater than 1, while a proportion cannot. 

In your example, however, you are asking if the difference in weight between foreign and domestic cars is explained by the difference in length between them. 

. regress weight length foreign

      Source |      SS      df      MS              Number of obs =      74
-------------+------------------------------          F(  2,    71) =  316.54
      Model |  39647744.7    2  19823872.3          Prob > F      =  0.0000
    Residual |  4446433.7    71  62625.8268          R-squared    =  0.8992
-------------+------------------------------          Adj R-squared =  0.8963
      Total |  44094178.4    73  604029.841          Root MSE      =  250.25

------------------------------------------------------------------------------
      weight |      Coef.  Std. Err.      t    P>|t|    [95% Conf. Interval]
-------------+----------------------------------------------------------------
      length |  31.44455  1.601234    19.64  0.000    28.25178    34.63732
    foreign |  -133.6775  77.47615    -1.73  0.089    -288.1605    20.80555
      _cons |  -2850.25  315.9691    -9.02  0.000    -3480.274  -2220.225
------------------------------------------------------------------------------


Length is a significant determinant of weight, and adjusted for this there is no significant difference between domestic and foreign cars (the effect size is interesting, but the standard error is very large). 

Is this analogous to what you had in mind?


Ronán Conroy
rconroy@rcsi.ie
Associate Professor
Division of Population Health Sciences
Royal College of Surgeons in Ireland
Beaux Lane House
Dublin 2


*
*  For searches and help try:
*  http://www.stata.com/help.cgi?searchhttp://www.stata.com/support/statalist/faqhttp://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