Strategies for using functions to store within the global env

Question

Let's say we have the following data frame:

# Data
Id <- c(1,2,3,4,5,6,7,8,9,10)
Type <- c("Beginner", "Expert", "Intermediate", "Beginner", 
  "Professional", "Expert", "Intermediate", "Professional", 
  "Professional", "Expert")
Response<- c(1,1,2,2,1,2,1,2,1,1)
Successful <- data.frame(Id, Type, Response)
Successful

# Dataframe
#   Successful
Id  Type             Response    
1   Beginner         1
2   Expert           1
3   Intermediate     2
4   Beginner         2
5   Professional     1
6   Expert           2
7   Intermediate     1
8   Professional     2
9   Professional     1
10  Expert           1

I know that I could store it as an object (DFRespType) in the global environment by doing the following:

 DFRespType <- 
  as.data.frame(round(100*prop.table(table(Successful$Response, 
                                   Successful$Type),2), 1))

But instead, I would like to create a function for storing the object to make doing this a lot more efficient. Below I tried to make the StoreDF function:

StoreDF <- function(DFName, dataset, variable1, variable2){
  DFName <- as.data.frame(round(100*prop.table(table(dataset$variable1, 
                                              dataset$variable2),2), 1))
}

But when I try and use it in the following way, nothing is stored:

StoreDF(DFRespType, Successful, Response, Type)

Any help with this would be gratefully appreciated.

Try the `<<-` operator instead of `<-` in your `StoreDF()` function definition. Therefore, `DFName <- as.data.frame(some_table)` becomes `DFName <<- as.data.frame(some_table)` — Argon, Jun 27 '19 at 00:19
This is a misuse of functions. You are trying to produce an object through side effect. Don't do it. You might want to look into storing multiple data.frames in a list. See this post: [make list of data.frames](https://stackoverflow.com/questions/17499013/how-do-i-make-a-list-of-data-frames). — lmo, Jun 27 '19 at 00:30

score 2 · Accepted Answer · answered Jun 27 '19 at 00:35

2

Don't store objects in global environment from inside the function. Instead return the dataframe back from the function. Also using quoted variables to subset the dataframe.

StoreDF <- function(dataset, variable1, variable2){
    as.data.frame(round(100* prop.table(table(dataset[[variable1]], 
                        dataset[[variable2]]),2), 1))
}

DFRespType <- StoreDF(Successful, "Response", "Type")
DFRespType

#  Var1         Var2 Freq
#1    1     Beginner 50.0
#2    2     Beginner 50.0
#3    1       Expert 66.7
#4    2       Expert 33.3
#5    1 Intermediate 50.0
#6    2 Intermediate 50.0
#7    1 Professional 66.7
#8    2 Professional 33.3

answered Jun 27 '19 at 00:35

Ronak Shah

377,200
20
156
213

Thanks Ronak, this works well. Could you explain why objects shouldn't be stored in the global environment from inside functions? And why have you used double instead of single square brackets? – interestedindata Jun 27 '19 at 18:53
@interestedindata If you want to store objects from inside the function in global environment you have to use `assign` or `<<-` but using `assign` or `<<-` is not usually a good practice. Read [Why is using assign bad](https://stackoverflow.com/questions/17559390/why-is-using-assign-bad) and [when to use <<-](https://stackoverflow.com/questions/2628621/how-do-you-use-scoping-assignment-in-r). Apart from that double bracket gives you a vector whereas single bracket gives a dataframe back. Check `mtcars[["cyl"]]` and `mtcars["cyl"]`. Maybe in this case it doesn't matter though. – Ronak Shah Jun 28 '19 at 01:35

Strategies for using functions to store within the global env

1 Answers1