在 R 中，当替换列不为空时，如何将一列中的值替换为另一列的值？

Question

I'm working on automating a report in R. One thing we do is look at clients who enter a shop and estimate their age record that as Estimate Age . We are instructed to use the Estimated Age as Age in the report so that all parties have an age. If we are able to record someone's Actual Age that age then becomes the Age . Most records do not have a value for Actual Age . For the records that do have an Actual Age value I need to replace the Estimated Age value with the Actual Age value whenever it exists. The records without Actual Age should remain unchanged.

I'm a newbie and have been stuck on this step for months. Asking the stackoverflow gods for a blessing. See image if it helps.

Replacing Estimate Age with Actual Age

I've already tried: Tried several variations of 2 different methods for replacing Estimated Age with Actual Age, again to no avail:

1) Age <- ifelse(is.null(MyReport$ActualAge), MyReport$ActualAge, MyReport$EstimatedAge)

View(MyReport) 2) Also something like this but I tweaked it so much so not exactly like this I messed it up

select <- is.null(MainReportload$ActualAge) < 0.01
df[select,MyReport$EstimatedAge] <- df[select, MyReport$ActualAge]

3)

if(is.null(MyReport$ActualAge)) {
  MyReport$Age <- MyReport$EstimatedAge
} else {
  MyReport$Age <- MyReport$ActualAge
  }
MyReport$Age
View(MyReport)

8.6.19 Alternative based on brain and minimal SQL knowledge, just do a coalesce, coalesce is available in the dplyr library. Result: same issue as the above attempt will continue with research.

Answer 1

I'm not sure why you say that dplyr::coalesce doesn't work, here's a simplified generic example of it. If you post a reproducible version of your data we can help more. The key to coalesce is that it returns the first non missing value. So coalesce(estimated_age, age),= coalesce(age, estimated_age)

# example data
df <- readr::read_csv("
age, estimated_age
12, 14
NA, 13
NA, NA
15, NA
")

# coalesce
df2 <- dplyr::mutate(df, new_age = dplyr::coalesce(age, estimated_age))

Answer 2

Part of the problem was the age values were stored as a range always leading with zero. So a 24 year old would have age as 0-24. I couldn't use the dplyr::coalesce solution until I fixed that price. Also, it didn't work with the mutate piece so I took that off.

Below is what finally worked for me!

    #USING ACTUAL AGE WHENEVER IT IS PRESENT AND ESTIMATED AGE WHEN THERE ISN'T AN ACTUAL AGE
#Remove hyphens from age fields and store the column as integers

main_df$Actual.Age <- as.integer( gsub("-", "", main_df$Actual.Age))
main_df$EstimatedAge <- as.integer( gsub("-", "", main_df$EstimatedAge))


#Use Coalesce to create a new column that contains the NEW Age values. 
main_df$new_EstimatedAge <- dplyr::coalesce(main_df$Actual.Age, 
main_df$EstimatedAge)
#view(head(main_df$new_EstimatedAge, 30))