ifelse help with logic

[jeremyz]

How do I create a script for the following logic:

If Column A = “Monday” AND Column B >= 17, then convert those instances in Column A to “Tuesday” AND those in Column B to 8, otherwise leave all other instances alone.

My attempt:

df <- ifelse(df$ColumnA == “Monday” 
& df$ColumnB >= 17, 
df$ColumnA == “Tuesday” 
& df$ColumnB == 8, 
df)

[JackDavison]

I think the best way of doing this is to create a "flag" column that identifies the rows that meet your requirements, and then use that flag to change the other two columns one at a time.

# create some data
library(tidyverse)

dat = crossing(day = c("Monday", "Tuesday", "Wednesday"),
               num = seq(0,30,5))

dat
#> # A tibble: 21 x 2
#>    day       num
#>    <chr>   <dbl>
#>  1 Monday      0
#>  2 Monday      5
#>  3 Monday     10
#>  4 Monday     15
#>  5 Monday     20
#>  6 Monday     25
#>  7 Monday     30
#>  8 Tuesday     0
#>  9 Tuesday     5
#> 10 Tuesday    10
#> # ... with 11 more rows

# solution

dat$flag = dat$day == "Monday" & dat$num >= 17
dat$day  = ifelse(dat$flag, "Tuesday", dat$day)
dat$num  = ifelse(dat$flag, 8, dat$num)
dat$flag = NULL

dat
#> # A tibble: 21 x 2
#>    day       num
#>    <chr>   <dbl>
#>  1 Monday      0
#>  2 Monday      5
#>  3 Monday     10
#>  4 Monday     15
#>  5 Tuesday     8
#>  6 Tuesday     8
#>  7 Tuesday     8
#>  8 Tuesday     0
#>  9 Tuesday     5
#> 10 Tuesday    10
#> # ... with 11 more rows

# or do it in the tidyverse
dat |> 
  mutate(flag = day == "Monday" & num >= 17,
         day  = if_else(flag, "Tuesday", day),
         num  = if_else(flag, 8, num)) |> 
  select(-flag)

^{Created on 2021-12-14 by the reprex package (v2.0.1)}

[jeremyz]

Thank you so much! I didn’t think of creating a flag column. Worked great