In the dplyr package can you mutate a column based on the values in a different column-CodePudding

My aim is to create a new df column called audit_cat which is based on two already existing df column called audit_score (discrete numeric variable) and gender ("Male", "Female").

audit_cat will be a dichotomous string variable with ranges 0-2 ~"Not hazardous" and 3-12 ~"Hazardous" for "Females" and for "Males" 0-3 = "Not hazardous" and 4-12.

It would be ideal if audit_cat contained both the female and male scores.

Using filter to separate by gender and then mutate the audit_score with case_when has worked except it filters out the other gender.

df2<- df2%>% 
   filter(gender == "Male") %>%
   mutate(audit_score_cat =
            case_when(audit_score >= 0 & audit_score <= 3 ~ "Not hazardous",
                      audit_score >=4 & audit_score <= 12 ~ "Hazardous"))

Is there a way I can create the new df column audit_cat based on the two different scoring systems which is dependent on gender?

Thank you.

CodePudding user response：

You could list the conditions for "Not hazardous" and assign the rest to "Hazardous".

df2 %>%
  mutate(audit_score_cat = case_when(
    gender ==    "Male" & between(audit_score, 0, 3) ~ "Not hazardous",
    gender == "Females" & between(audit_score, 0, 2) ~ "Not hazardous",
    TRUE                                             ~ "Hazardous"
  ))