CODEWITHSUNDEEP
r
largecats
largecats

Reputation: 175

compare_means(): "must resolve to integer column positions, not a symbol" when using string/symbol in formula

I'm trying to do ANOVA with compare_means(). Using the sample dataset ToothGrowth as an example, I want to achieve something like this:

#directly calling column name
compare_means(len~supp, data=ToothGrowth, group.by = "dose", method = "anova")

except that my supp in the formula is a string.

Directly passing the string as an argument results in an error in the model formula. So I tried converting the string to name or symbol:

#convert string column name to name/symbol
x <- as.name("supp") #as.symbol("supp")
compare_means(len~x, data=ToothGrowth, group.by = "dose", method = "anova")

But both returned the same error:

Error: x must resolve to integer column positions, not a symbol

I have checked similar error messages like Error All select() inputs must resolve to integer column positions. The following do not:, but none of them seem to address my issue.

Please instruct what I should do. Any help would be much appreciated!

Upvotes: 1

Views: 706

Answers (1)

largecats
largecats

Reputation: 175

So I saw How to use reference variables by character string in a formula? and solved the problem using

f <- "len~supp"
do.call("compare_means", list(as.formula(f), data=as.name("ToothGrowth"),
    group.by = "dose", method = "anova"))

Upvotes: 1

Related Questions