[R] Code modification for post-hoc power

Tue Aug 27 18:15:38 CEST 2019

Anne sent me off-line the error message.
============ error message starts here ==========
This error message:
"Beginning Cgee S-function, @(#) geeformula.q 4.13 98/01/27 Error in 
`[.default`(xj, i) : type 'closure' d'indice incorrect"

appears after the following R codes:

library(gee)
attach(geefile) # stored in some spss directory
geereg <- gee(outcome ~ trial + group + trial*group + trial*group*beckce 
+ trial*group*beckref, id = subject, data = geefile, na.omit, tol = 
0.001, maxiter = 25, family = binomial, corstr = "exchangeable", silent 
= TRUE)

The following command would have allowed me to modify the power so as to 
obtain the sample size I have:
lmmpower(geereg, pct.change = 0.10, t = seq(0,6,3), power = 0.80)

================ end ===============

I am not an expert on R programming but in my experience that error has 
meant that I was including as a parameter something which was not of the 
type which the program expected. I would suggest as a first step not 
using attach() at all and instead using data=geefile in the call to gee 
or investigating the with() command if it does not allow a data parameter.

On 27/08/2019 12:25, Michael Dewey wrote:
> Dear Anne
> 
> Can you resend the eror message which you accidentally sent only to me 
> please?
> 
> Michael
> 
> On 27/08/2019 08:02, CHATTON Anne wrote:
>> Dear Michael,
>>
>> Thanks a lot for your suggestion. This is what I am trying to do with 
>> R (longpower and gee packages). But I am getting stuck with a 
>> confusing error message sent earlier I don't understand.
>>
>> Best,
>>
>> Anne
>>
>> -----Message d'origine-----
>> De : Michael Dewey [mailto:lists using dewey.myzen.co.uk]
>> Envoyé : lundi, 26 août 2019 18:29
>> À : Marc Schwartz <marc_schwartz using me.com>; CHATTON Anne 
>> <Anne.Chatton using hcuge.ch>
>> Cc : R-help <r-help using r-project.org>
>> Objet : Re: [R] Code modification for post-hoc power
>>
>> Dear Anne
>>
>> In addition to Marc's comments if you are forced to do this then, 
>> assuming your package computes sample size from power then just feed 
>> it a range of powers and find the one for which it calculates the 
>> sample size you had. There is a more elegant way to do this using 
>> uniroot but brute force should work.
>>
>> Michael
>>
>> On 26/08/2019 13:42, Marc Schwartz via R-help wrote:
>>>
>>>> On Aug 26, 2019, at 6:24 AM, CHATTON Anne via R-help 
>>>> <r-help using r-project.org> wrote:
>>>>
>>>> Hello everybody,
>>>>
>>>> I am trying to accommodate the R codes provided by Donohue for 
>>>> sample size calculation in the package "longpower" with lmmpower 
>>>> function to estimate the post-hoc power (asked by a reviewer) of a 
>>>> binary GEE model with a three-way interaction (time x condition x 
>>>> continuous predictor) given a fixed sample size. In other words 
>>>> instead of the sample size I would like to estimate the power of my 
>>>> study.
>>>>
>>>> Could anyone please help me to modify these codes as to obtain the 
>>>> power I'm looking for.
>>>>
>>>> I would really appreciate receiving any feedback on this subject.
>>>>
>>>> Yours sincerely,
>>>>
>>>> Anne
>>>
>>>
>>> Hi,
>>>
>>> Three comments:
>>>
>>> 1. Don't calculate post hoc power. Do a Google search and you will 
>>> find a plethora of papers and discussions on why not, including these:
>>>
>>>     The Abuse of Power: The Pervasive Fallacy of Power Calculations 
>>> for Data Analysis
>>>     The American Statistician, February 2001, Vol. 55, No. 1
>>>     https://www.vims.edu/people/hoenig_jm/pubs/hoenig2.pdf
>>>
>>>     Post Hoc Power: Tables and Commentary
>>>     https://stat.uiowa.edu/sites/stat.uiowa.edu/files/techrep/tr378.pdf
>>>
>>>     Observed power, and what to do if your editor asks for post-hoc 
>>> power analyses
>>> http://daniellakens.blogspot.com/2014/12/observed-power-and-what-to-do
>>> -if-your.html
>>>
>>>     Retraction Watch:
>>>     Statisticians clamor for retraction of paper by Harvard 
>>> researchers they say uses a “nonsense statistic”
>>> https://retractionwatch.com/2019/06/19/statisticians-clamor-for-retrac
>>> tion-of-paper-by-harvard-researchers-they-say-uses-a-nonsense-statisti
>>> c/
>>>
>>>     PubPeer Comments on the paper cited in the above RW post:
>>>     https://pubpeer.com/publications/4399282A80691D9421B497E8316CF6
>>>
>>>     A discussion on Frank's Data Methods forum also related to the 
>>> same paper cited above:
>>>     "Observed Power" and other "Power" Issues
>>> https://discourse.datamethods.org/t/observed-power-and-other-power-iss
>>> ues/731/30
>>>
>>>
>>> 2. If you are still compelled (voluntarily or involuntarily), you may 
>>> want to review the vignette for the longpower package which may have 
>>> some insights, and/or contact the package maintainer for additional 
>>> guidance on how to structure the code. See the vignette here:
>>>
>>> https://cran.r-project.org/web/packages/longpower/vignettes/longpower.
>>> pdf
>>>
>>>
>>> 3. Don't calculate post hoc power.
>>>
>>>
>>> Regards,
>>>
>>> Marc Schwartz
>>>
>>> ______________________________________________
>>> R-help using r-project.org mailing list -- To UNSUBSCRIBE and more, see
>>> https://stat.ethz.ch/mailman/listinfo/r-help
>>> PLEASE do read the posting guide
>>> http://www.R-project.org/posting-guide.html
>>> and provide commented, minimal, self-contained, reproducible code.
>>>
>>> ---
>>> This email has been checked for viruses by AVG.
>>> https://www.avg.com
>>>
>>>
>>
>> -- 
>> Michael
>> http://www.dewey.myzen.co.uk/home.html
>>
> 

-- 
Michael
http://www.dewey.myzen.co.uk/home.html