Add by= variable to tbl_svysummary(include=) by default

[coreysparks]

I may have found a bug, below is reproducible code to generate the error. I expect to get a simple summary of variable v151 for the levels of v139, but I get an error stating

Error: Problem with `mutate()` column `df_stats`.
i `df_stats = pmap(...)`.
x Can't subset columns that don't exist.
x Column `v139` doesn't exist.
Run `rlang::last_error()` to see where the error occurred

---

Brief description of the problem.

dhs<-haven::read_dta("https://github.com/coreysparks/data/blob/master/ZZIR62FL.DTA?raw=true")
dhs<-haven::zap_labels(dhs)
dhs$v005<- dhs$v005/1000000
dhs2<- select(dhs, v021, v022, v005, v139, v151)

names(dhs2)
library(dplyr); library(gtsummary)
options(survey.lonely.psu = "adjust")
t1<- survey::svydesign(ids = ~v021,
               strata=~v022,
               weights = ~v005,
               data=dhs2)%>%
  gtsummary::tbl_svysummary(by = "v139", include = c(v151))

[ddsjoberg]

hi @coreysparks ! thanks for the post! you'll need to add the by= variable to the inlcude= argument to get your code working. it's added by default in tbl_summary() but looks like i need to add it for tbl_svysummary() as well

library(gtsummary)

dhs<-haven::read_dta("https://github.com/coreysparks/data/blob/master/ZZIR62FL.DTA?raw=true")
dhs<-haven::zap_labels(dhs)
dhs$v005<- dhs$v005/1000000
dhs2<- select(dhs, v021, v022, v005, v139, v151)

names(dhs2)
#> [1] "v021" "v022" "v005" "v139" "v151"
library(dplyr); library(gtsummary)
#> 
#> Attaching package: 'dplyr'
#> The following objects are masked from 'package:stats':
#> 
#>     filter, lag
#> The following objects are masked from 'package:base':
#> 
#>     intersect, setdiff, setequal, union
options(survey.lonely.psu = "adjust")
survey::svydesign(ids = ~v021,
                  strata=~v022,
                  weights = ~v005,
                  data=dhs2)%>%
  tbl_svysummary(by = "v139", include = c(v151, v139)) %>%
  as_kable() # convert to kable to display on github

Characteristic	1, N = 2,850	2, N = 1,599	3, N = 2,279	4, N = 1,562	97, N = 58
sex of household head
1	1,975 (69%)	1,180 (74%)	1,494 (66%)	1,162 (74%)	41 (72%)
2	876 (31%)	418 (26%)	786 (34%)	400 (26%)	16 (28%)

^{Created on 2021-06-22 by the reprex package (v2.0.0)}

[coreysparks]

Gotcha! Thanks for the swift reply!