Skip to content

sleep data

4 messages · Tom Backer Johnsen, Charles C. Berry, Martin Maechler

Original
Tom Backer Johnsen
# 
When reading the documentation for the "sleep" data set in R, the 
impression is clear, this is an "independent groups" kind of design 
(two groups of 10 subjects each).  However, when browsing the original 
article (referred to in the help file), my impression is quite clear, 
this is really a "repeated measures" kind of data (one group of 10 
subjects, two observations).  What is correct?

Tom
Charles C. Berry
# 
Yes, you refer to

 	Cushny, A. R. and Peebles, A. R. The action of optical isomers: II
 	hyoscines. The Journal of Physiology, 1905, 32: 501.510.

which was used by 'Student' to illustrate the paired t-test.

This is indeed a crossover design.
On Wed, 17 Jan 2007, Tom Backer Johnsen wrote:

            

      
      
      
    

        

      
      
    

        
Charles C. Berry                        (858) 534-2098
                                          Dept of Family/Preventive Medicine
E mailto:cberry at tajo.ucsd.edu	         UC San Diego
http://biostat.ucsd.edu/~cberry/         La Jolla, San Diego 92093-0901

  
  


  


      

    

    

      
      

        


  

        

      Martin Maechler
    

    

      
      #
    

  


  

      
Thank you, Chuck and Tom.

I'd gladly improve the help page,

particularly if you can provide a patch against

https://svn.R-project.org/R/trunk/src/library/datasets/man/sleep.Rd

where I've already added the Cushny and Peebles reference,
thanks to Chuck.

Regards,
Martin Maechler, ETH Zurich

            

      
      
      
    

        
ChuckB> Yes, you refer to

    ChuckB>  	Cushny, A. R. and Peebles, A. R. The action of
    ChuckB> optical isomers: II hyoscines. The Journal of
    ChuckB> Physiology, 1905, 32: 501.510.

    ChuckB> which was used by 'Student' to illustrate the paired
    ChuckB> t-test.

    ChuckB> This is indeed a crossover design.

        
ChuckB> On Wed, 17 Jan 2007, Tom Backer Johnsen wrote:

        
>> When reading the documentation for the "sleep" data set
    >> in R, the impression is clear, this is an "independent
    >> groups" kind of design (two groups of 10 subjects each).
    >> However, when browsing the original article (referred to
    >> in the help file), my impression is quite clear, this is
    >> really a "repeated measures" kind of data (one group of
    >> 10 subjects, two observations).  What is correct?
    >> 
    >> Tom
    >> 
    >> ______________________________________________
    >> R-help at stat.math.ethz.ch mailing list
    >> 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.
    >> 

    ChuckB> Charles C. Berry (858) 534-2098 Dept of
    ChuckB> Family/Preventive Medicine E
    ChuckB> mailto:cberry at tajo.ucsd.edu UC San Diego
    ChuckB> http://biostat.ucsd.edu/~cberry/ La Jolla, San Diego
    ChuckB> 92093-0901

    ChuckB> ______________________________________________
    ChuckB> R-help at stat.math.ethz.ch mailing list
    ChuckB> https://stat.ethz.ch/mailman/listinfo/r-help PLEASE
    ChuckB> do read the posting guide
    ChuckB> http://www.R-project.org/posting-guide.html and
    ChuckB> provide commented, minimal, self-contained,
    ChuckB> reproducible code.

  
  


  


      

    

    

      
      

        


  

        

      Charles C. Berry
    

    

      
      #
    

  


  

      
A further note. The Cushny & Peebles article can be viewed here:

 	http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=1465734

and the page with the data is here:

 	http://www.pubmedcentral.nih.gov/pagerender.fcgi?artid=1465734&pageindex=9

A machine readable version of the data is at:

 	https://biostat.ucsd.edu/~cberry/t-test/sleep.dat

The version published in Student's Biometrika article has a typographical 
error, but it is evident that Student's computations were unaffected.

Chuck Berry

        
On Wed, 17 Jan 2007, Charles C. Berry wrote:

        

            

      
      
      
    

        

      
      
    

            

      
      
      
    

        
Charles C. Berry                        (858) 534-2098
                                          Dept of Family/Preventive Medicine
E mailto:cberry at tajo.ucsd.edu	         UC San Diego
http://biostat.ucsd.edu/~cberry/         La Jolla, San Diego 92093-0901