order() using data frame column name containing spaces produces unexpected results - r

Question

I was trying to order a data frame in R according to a column named 'credit card usage'. The name of the data frame is mydata. The following command without a comma gives an error
newdata = mydata[order('credit card usage')]
But the following command with a comma works absolutely fine
newdata = mydata[order('credit card usage'),]
I need to understand why do we need the comma. Please can someone explain in a simple language whats going on behind the scenes?
Also the following command
mydata[order('credit card usage'),]
gives only the first row and not the whole dataframe. Why?

Answer 1

Why mydata[order('credit card usage'),] returns only first row is a tricky one.
The name of the column when used after , in [ refers to value of column otherwise it is just a string.
order('credit card usage') call considers just a string is passed to it and it sorts it out and pass the index(which is 1). Hence:
mydata[order('credit card usage'),] reduces to
mydata[1,]
=> which is 1st row of the mydata.

Answer 2

MKR's answer explains why the OP obtained the results described in the post. Here, we'll explain how to deal with the column named credit card usage to correctly sort the entire data frame.
Generally speaking, it's not advisable to use column names containing spaces in R because it often leads to unexpected results as experienced by the OP.
To use a column in a data frame whose name contains spaces, one must use the [[ form of the extract operator. We'll illustrate with some sample data...
set.seed(95014123)
mydata <- data.frame(matrix(round(runif(100)*1000,0),nrow=50))
names(mydata) <- c("credit card usage","value")
head(mydata)
head(mydata[order(mydata[["credit card usage"]]),])
...and the output:
> set.seed(95014123)
> mydata <- data.frame(matrix(round(runif(100)*1000,0),nrow=50))
> names(mydata) <- c("credit card usage","value")
> head(mydata)
credit card usage value
1 795 217
2 816 613
3 342 323
4 126 751
5 618 780
6 625 529
> head(mydata[order(mydata[["credit card usage"]]),])
credit card usage value
47 25 109
44 81 534
18 91 985
31 99 931
19 109 190
4 126 751
>
One can replace the spaces with underscores via the gsub() function, which will enable one to use the $ form of the extract operator in subsequent functions.
# replace spaces with underscores
names(mydata) <- gsub(" ","_",names(mydata))
head(mydata[order(mydata$credit_card_usage),])
...and the output:
> # replace spaces with underscores
> names(mydata) <- gsub(" ","_",names(mydata))
> head(mydata[order(mydata$credit_card_usage),])
credit_card_usage value
47 25 109
44 81 534
18 91 985
31 99 931
19 109 190
4 126 751
>