Quantcast

Getting Regression Output for Effect Size Calculation

classic Classic list List threaded Threaded
6 messages Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Getting Regression Output for Effect Size Calculation

Eric Howard
Statalist,
 
I would like to calculate an effect size (Cohen's d) myself using an Excel file instead of using the sizefx command and was wondering how I can get n sizes, mean outcomes, and standard deviations from OLS (using reg command) regression output for my treated and comparison groups.
 
Thanks in advance!
Eric Howard

Eric S. Howard, M.A.
SERVE Center at the
University of North Carolina at Greensboro
P.O. Box 5367
Greensboro, NC 27435
(800) 755-3277
(336) 315-7431 - Direct
(336) 315-7457 - Fax
[hidden email]
www.serve.org



*
*   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/
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Getting Regression Output for Effect Size Calculation

nshephard
Administrator
On Fri, Oct 29, 2010 at 3:17 PM, Eric Howard <[hidden email]> wrote:
> Statalist,
>
> I would like to calculate an effect size (Cohen's d) myself using an Excel file instead of using the sizefx command and was wondering how I can get n sizes, mean outcomes, and standard deviations from OLS (using reg command) regression output for my treated and comparison groups.

Why use Excel when Stata can do this?

Some of the pitfalls of using M$-Excel/spreadsheets...

http://www.burns-stat.com/pages/Tutor/spreadsheet_addiction.html

Summary of response from Allstat mailing list...

http://biostat.mc.vanderbilt.edu/twiki/bin/view/Main/ExcelProblems

Neil

--
"Our civilization would be pitifully immature without the intellectual
revolution led by Darwin" - Motoo Kimura, The Neutral Theory of
Molecular Evolution

Email - [hidden email]
Website - http://kimura-no-ip.org/
Photos - http://www.flickr.com/photos/slackline/

*
*   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/
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Getting Regression Output for Effect Size Calculation

nshephard
Administrator
Further if you want to do the calculation yourself, why not do that in
Stata too rather than Excel?

You could write your own ado-file and use it time after time.

Neil

--
"Our civilization would be pitifully immature without the intellectual
revolution led by Darwin" - Motoo Kimura, The Neutral Theory of
Molecular Evolution

Email - [hidden email]
Website - http://kimura-no-ip.org/
Photos - http://www.flickr.com/photos/slackline/
*
*   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/
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Getting Regression Output for Effect Size Calculation

Maarten buis
In reply to this post by Eric Howard
--- On Fri, 29/10/10, Eric Howard wrote:
> I would like to calculate an effect size (Cohen's d) myself
> using an Excel file instead of using the sizefx command and
> was wondering how I can get n sizes, mean outcomes, and
> standard deviations from OLS (using reg command) regression
> output for my treated and comparison groups.

I assume you are refering to the user written command -sizefx-
that can be downloaded from SSC. Remember that the Statalist
FAQ asks everybody to report where they got user written
software from. This is in your own interest, as the answer
often depends on that information (i.e. which version of the
user written program you are using).

You can see how the computation is done if you type in Stata
-viewsource sizefx.ado-. This is a very short program and
well documented with comments, so reading it should not pose
a problem.

Hope this helps,
Maarten

--------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
Germany

http://www.maartenbuis.nl
--------------------------


     

*
*   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/
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Getting Regression Output for Effect Size Calculation

Eric Howard
In reply to this post by nshephard
Thanks Neil! Didn't even think about just calculating it all in Stata via ado-file! Definitely the way to go!

Thanks again,
Eric

>>> Neil Shephard <[hidden email]> 10/29/2010 11:28 AM >>>
Further if you want to do the calculation yourself, why not do that in
Stata too rather than Excel?

You could write your own ado-file and use it time after time.

Neil

--
"Our civilization would be pitifully immature without the intellectual
revolution led by Darwin" - Motoo Kimura, The Neutral Theory of
Molecular Evolution

Email - [hidden email]
Website - http://kimura-no-ip.org/ 
Photos - http://www.flickr.com/photos/slackline/ 
*
*   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/
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Getting Regression Output for Effect Size Calculation

Eric Howard
In reply to this post by Maarten buis
Maarten,

Yes, I was referring to the user written command for "sizefx." Thanks for the suggestion on viewsource sizefx.ado!

Best,
Eric

>>> Maarten buis <[hidden email]> 10/29/2010 11:33 AM >>>
--- On Fri, 29/10/10, Eric Howard wrote:
> I would like to calculate an effect size (Cohen's d) myself
> using an Excel file instead of using the sizefx command and
> was wondering how I can get n sizes, mean outcomes, and
> standard deviations from OLS (using reg command) regression
> output for my treated and comparison groups.

I assume you are refering to the user written command -sizefx-
that can be downloaded from SSC. Remember that the Statalist
FAQ asks everybody to report where they got user written
software from. This is in your own interest, as the answer
often depends on that information (i.e. which version of the
user written program you are using).

You can see how the computation is done if you type in Stata
-viewsource sizefx.ado-. This is a very short program and
well documented with comments, so reading it should not pose
a problem.

Hope this helps,
Maarten

--------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
Germany

http://www.maartenbuis.nl 
--------------------------


     

*
*   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/
Loading...