Re: st: Spearman's rank correlation coefficient - power calculation

[Richard Goldstein <richgold@ix.netcom.com>]

Sounds to me as both of the following are true:

1. you did not do a sample size estimation prior to the study

2. the reviewer is now asking for a retrospective power calculation; 
this is very problematic -- for an intro to the problems, and some 
citations, see Russ Lenth's page on power: 
http://www.stat.uiowa.edu/~rlenth/Power/

Rich

Theo Serge wrote:

> Dear Statalist, 
> I would be grateful if you could help me on a problem:
>
> Having submitted an article to a Journal, the reviewers asked me for power calculation on a Spearman’s rank correlation coefficient.
>
> This happened although I had been very careful with the interpretation of negative findings in the manuscript. Moreover, Spearman’s rank correlation coefficient was the appropriate choice for the univariate analysis performed, due to a variety of underlying factors.
>
> To my knowledge, there is no command through which STATA can perform power calculations on Spearman’s rank correlation coefficients.
>
> Please inform me if there is a way to overcome this difficulty. Are there any commands I am not familiar with? Would programming STATA be of value? 
> What would be the optimal approach to this issue? The need for power calculations in Spearman's rank correlation coefficients seems justified.
>
> I thank you in advance.
>
> T.S.
> University of Athens, Greece
*
*   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/