stdev error

Hello,

I cannot reproduce this error with a built-in data set.
Can you post str(my_tbl)?


suppressPackageStartupMessages(library(dplyr))

mtcars %>%
   mutate(hp = round(hp * 2) / 2) %>%
   group_by(cyl, hp) %>%
   summarise(
     count = n(),
     hp = mean(hp),
     stdev = sd(hp)
   )
#> `summarise()` has grouped output by 'cyl'. You can override using the 
`.groups`
#> argument.
#> # A tibble: 23 x 4
#> # Groups:   cyl [3]
#>      cyl    hp count stdev
#>    <dbl> <dbl> <int> <dbl>
#>  1     4    52     1    NA
#>  2     4    62     1    NA
#>  3     4    65     1    NA
#>  4     4    66     2    NA
#>  5     4    91     1    NA
#>  6     4    93     1    NA
#>  7     4    95     1    NA
#>  8     4    97     1    NA
#>  9     4   109     1    NA
#> 10     4   113     1    NA
#> # ... with 13 more rows

Hope this helps,

Rui Barradas


?s 14:14 de 11/03/2022, Jeff Reichman escreveu:

r-help forum

  

When I run the following code

  

my_tbl %>%

   mutate(Bse_bwt = round(Bse_bwt * 2) / 2) %>%

   group_by(Cat, Bse_bwt) %>%

   summarize(count = n(), Bse_ftv = mean(Bse_ftv), stdev = sd(Bse_ftv))

  

I get the following error:

  

Error: `stdev` refers to a variable created earlier in this summarise().

Do you need an extra mutate() step?

  

I suspect it is because the standard deviation of a length-one vector is NA
and R is errorerrors out on the standard deviation  of 1. So then I tried

  

summarize(count = n(), Bse_ftv = mean(Bse_ftv), stdev = if(n()>1)
sd(Bse_ftv) else 0) and this didn't seem to work either. So there has to be
a way to add some sort of error checker to my standard deviation function to
check if n > 1 and then take the standard deviation in dplyr.

  

Jeff

  


	[[alternative HTML version deleted]]

______________________________________________
R-help at 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.