warning message in mean.default argument is not numeric or logical returning na but my vector is numeric

Question

i am running this R code to import a lot of csv file and calculate one mean of a numerico vector (nitrate) for all the datasets filtering by some values of an integer vector, ID.

setwd("C:/Users/americo/Documents/Documenti_2_0/Education/Data science/Coursera/R programming/rworkingdir/specdata")
lista_tot1 = list.files(pattern="*.csv");
lista_tot2 = lapply(lista_tot1, read.csv, sep = ",")#stampa tutti i csv                                  
lista_tot3<-do.call("rbind", lista_tot2)
lista_tot<-lista_tot3[is.element(lista_tot3$ID, 3:4),]
media<-mean(lista_tot$nitrate, na.rm=TRUE)
media

and it works.

But if i try to write this function, which seems similar to me:

pollutantmean <- function(directory, pollutant, id = 1:332) {
  setwd(directory)
  lista_tot1 = list.files(pattern="*.csv");
  lista_tot2 = lapply(lista_tot1, read.csv, sep = ",")  
  lista_tot3<-do.call("rbind", lista_tot2)
  lista_tot<-lista_tot3[is.element(lista_tot3$ID, id),]
  media<-mean(lista_tot$pollutant), na.rm=TRUE)
  return(media)
}

and then

media1<-pollutantmean("C:/Users/americo/Documents/Documenti_2_0/Education/Data science/Coursera/R programming/rworkingdir/specdata","nitrate", 3:4)

it then returns to me the message:

warning message in mean.default argument is not numeric or logical returning na 

...but my vector is numeric

that really doesn't make sense to me. if i ask to return of the structure of the last dataset, lista_tot, nitrate is still numeric.

I really don't know what to do, if some very experienced user could help me i would be extremely grateful.

Answer 1

Change this:

media<-mean(lista_tot$pollutant), na.rm=TRUE)

to this:

media<-mean(lista_tot[[pollutant]], na.rm=TRUE)

because the "$" does not evaluate it's argument. In this case "pollutant" is not a column name so you need an extraction function that will evaluate what the named object, pollutant, is "carrying" (namely the column name, "nitrate")to get the proper extraction from the dataframe.