Skip to content

I don't understand the result of `svyboxplot` function from the Survey package

3 messages · Iulia Dumitru, Anthony Damico, Thomas Lumley

Iulia Dumitru
# 
After following the example given here: https://www.rdocumentation.org/packages/survey/versions/4.1-1/topics/svyhist
for `svyboxplot` I get the result in the attached image. This is a box plot of the `enroll` variable from the stratified dataset `apistrat`, grouped by `stype`: E (elementary school), M (middle school) and H (high school). If I use the `svyby` function to group the data by `stype` and find the mean for each group, I get this result:
stype  enroll       se
E     E  416.78 16.41740
H     H 1320.70 91.70781
M     M  832.48 54.52157

Clearly the means are very different from each other. Then why don?t the box plots show this? I don?t know how to interpret the plot. Could someone please offer some insight on this? Thank you!
Anthony Damico
# 
hi, nice catch!  i'm ccing the author of the survey package because this
might be a issue.  when i run ?svyboxplot, i also see all three boxes in
the exact same place..  seems like the svyby() call inside of svyboxplot
does something unexpected when svyquantile gets passed using keep.var=FALSE
and ci=FALSE


library(survey)

data(api)

dstrat <- svydesign(id = ~1, strata = ~stype, weights = ~pw, data =
apistrat, fpc = ~fpc)

# looks OK
svyby(~enroll, ~stype, dstrat, svyquantile, quantiles = c(0, 0.25, 0.5,
0.75, 1), na.rm = TRUE)

# returns each result three times in an unexpected configuration..
svyboxplot then grabs the repeated information from the first six columns
svyby(~enroll, ~stype, dstrat, svyquantile, ci = FALSE, keep.var = FALSE,
quantiles = c(0, 0.25, 0.5, 0.75, 1), na.rm = TRUE)
On Tue, Aug 2, 2022 at 7:30 AM Iulia Dumitru <iuliadmtru at gmail.com> wrote:

            

      
      
      
    

        

      
      
    

        

  
  


  


      

    

    

      
      

        


  

        

      Thomas Lumley
    

    

      
      #
    

  


  

      
It's a bug resulting from the new svyquantile() implementation.  It's fixed in the development version, which you can get from r-forge here: https://r-forge.r-project.org/R/?group_id=1788

  -thomas

Thomas Lumley
Professor of Biostatistics