CODEWITHSUNDEEP
r
Henry Cann
Henry Cann

Reputation: 101

Error Standardising variables in R : 'only defined on a data frame with all numeric variables'

I am simply looking to standardise my set of data frame variables to a 100 point scale. The original variables were on a 10 point scale with 4 decimal points.

I can see that my error is not unheard of e.g

Why am I getting a function error in seemingly similar R code?

Error: only defined on a data frame with all numeric variables with ddply on large dataset

but I have verified that all variables are numeric using

library(foreign)
library(scales)
ches <- read.csv("chesshort15.csv", header = TRUE)
ches2 <- ches[1:244, 3:10]
rescale(ches2, to = c(0,100), from = range(ches2, na.rm = TRUE, finite = TRUE))

This gives the error: Error in FUN(X[[i]], ...) : only defined on a data frame with all numeric variables

I have verified that all variables are of type numeric using str(ches2) - see below:

'data.frame':   244 obs. of  8 variables:
 $ galtan            : num  8.8 9 9.65 8.62 8 ...
 $ civlib_laworder   : num  8.5 8.6 9.56 8.79 8.56 ...
 $ sociallifestyle   : num  8.89 7.2 9.65 9.21 8.25 ...
 $ immigrate_policy  : num  9.89 9.6 9.38 9.43 9.13 ...
 $ multiculturalism  : num  9.9 9.6 9.57 8.77 9.07 ...
 $ ethnic_minorities : num  8.8 9.6 9.87 9 8.93 ...
 $ nationalism       : num  9.4 10 9.82 9 8.81 ...
 $ antielite_salience: num  8 9 9.47 8.88 8.38

In short, I'm stumped as to why it refuses to carry out the code.

For info, Head(bb) gives :

  galtan civlib_laworder sociallifestyle immigrate_policy multiculturalism ethnic_minorities
1  8.800           8.500           8.889            9.889            9.900             8.800
2  9.000           8.600           7.200            9.600            9.600             9.600
3  9.647           9.563           9.647            9.375            9.571             9.867
4  8.625           8.786           9.214            9.429            8.769             9.000
5  8.000           8.563           8.250            9.133            9.071             8.929
6  7.455           8.357           7.923            8.800            7.800             8.455
  nationalism antielite_salience
1       9.400              8.000
2      10.000              9.000
3       9.824              9.471
4       9.000              8.882
5       8.813              8.375
6       8.000              8.824

Upvotes: 2

Views: 2514

Answers (1)

HAVB
HAVB

Reputation: 1888

The rescale function is throwing that error because it expects a numeric vector, and you are feeding it a dataframe instead. You need to iterate; go through every column on your dataframe and scale them individually.

Try this:

sapply(ches2, rescale, to = c(0,100))

You don't need the range(ches2, na.rm = TRUE, finite = TRUE) portion of your code because rescale is smart enough to remove NA values on its own

Upvotes: 3

Related Questions