CODEWITHSUNDEEP
rdynamic
Joe Thomson
Joe Thomson

Reputation: 1

Functions with dynamic variables

I am analyzing data in R with many different data sets and I want to send in dummy variables to a function which then subsets the main data set and outputs the mean of a variable in the subsets.

For instance, my data set is named "two" and my dummy variable is "over50" and my function is:

getMean <- function(varName) {
   sub1 <- two[two$varName == 1, ]
   sub2 <- two[two$varName == 0, ]
   print(mean(sub1$return)
   print(mean(sub2$return)
}

However, when I call getMean(over50) I do not get the answer expected.

Is there a way to translate the function input into var names so I can do this dynamically? Or do I have to manually do these calculations?

Upvotes: 0

Views: 93

Answers (2)

lmo
lmo

Reputation: 38510

It is easier in this instance to pass a string to your function. Here is a generalized function that takes a data.frame and a variable name (string).

getMean <- function(df, varName) {
   mean1 <- mean(df[df[[varName]] == 1, ]$return)
   mean2 <- mean(df[df[[varName]] == 1, ]$return)

  return(c("mean1"=mean1, "mean2"=mean2))
}

This returns a named vector with the two means. the df argument must be a data.frame name (without quotes) whereas the varname should be a character string.

Upvotes: 0

ChrisW
ChrisW

Reputation: 111

I think the syntax you want is two[two[, varName] == 0, ].

More generally, you can access columns/rows of a data frame by passing in a string as data[c("row1", "row2"), c("col1", "col2")].

Side note: I think you're also missing a couple closing parentheses in your print() statements.

Upvotes: 1

Related Questions