Passing a string as argument for data column (whilst having another column identifier)

I'm writing a function that finds the mean at 4 different categorical variables and returns a size 4 vector with all of these values. In doing so, I want to pass different columns of a dataframe such that I can find these 4 values for any column of the dataframe. (I would like for the dataframe to be variable as well but thats not important)

makeMean <- function(df, colName){
  x <- c()
  levels <- levels(beetleData$Herbicide.dosage)
  for(i in levels){
    tempdata <- summarize(df, mean.temp = mean(Species.richness[Herbicide.dosage == i]))
    x <- append(x, tempdata)
  }
  x <- unlist(x) 
  return(x)
}

This is the function. Species.richness is the column name I'd like to replace with "colName" but I can't find a way to do that. I can pass it by doing df[colName] (I think) but I'd like to have the other filter too (Herbicide.dosage == i).

R doesn't let me do mean(colName[Herbicide.dosage == i]). I've tried a couple other ways as well and none work as R doesn't convert the string to the dataframe column I want, it just reads it as a string

Passing column names to a function is tricky. But dplyr programming concepts will help. Link

Cannot work on ur data but passing a column name which is not read a sring is possible this way .

library(dplyr)
makeMeans <- function(df,colName){
  
 df%>%
    summarise(Mean=    mean({{colName}},na.rm=T))
  
}
makeMeans(mtcars,mpg)

returns

      Mean
1 20.09062

This way you can just pass columns and manage them in your function with double curly brackets {{

there might be other ways but passing columns this way is easy. You can try approaches in the link

Like @anuanand's answer, but passing a string.

library(tidyverse)
makeMeans <- function(df, colName){
  
  df %>%
    summarise(Mean = mean(.data[[colName]],na.rm=T))
}

See: https://www.tidyverse.org/blog/2019/06/rlang-0-4-0/ regarding passing column names as part of non-standard evaulation (NSE) in R.