How to Fix in R: invalid model formula in ExtractVars

In this article, we will discuss how we can fix the “invalid model formula in ExtractVars” error in the R programming language.

The error that one may face in R is:

Error in terms.formula(formula, data = data) : 
  invalid model formula in ExtractVars

The R compiler produces such an error we try to fit a decision tree and use at least one variable incorrectly in the formula. 

When this error might occur in R

Let’s firstly create a data frame:

R

# Make a data frame
dataframe <- data.frame(marks=c(86, 74, 99, 92, 77, 88, 82, 89),
                 score=c(11, 17, 22, 24, 27, 12, 29, 32),
                 total=c(17, 27, 16, 18, 16, 15, 27, 4))
dataframe

Output:

Here, we need to use the rpart function from the rpart package to fit a decision tree model and further evaluate it.

Syntax to install the rpart package in the R console:

install.package(‘rpart’)

Suppose that we want to use the rpart() function to fit a decision tree model in the data and then the R compiler produces the error of “invalid model formula in ExtractVars” because we have given quotations at the ends of the predictor variables.

R

# Importing the library
library(rpart)

# Make a data frame
dataframe <- data.frame(marks=c(86, 74, 99, 92, 77, 88, 82, 89),
                 score=c(11, 17, 22, 24, 27, 12, 29, 32),
                 total=c(17, 27, 16, 18, 16, 15, 27, 4))

# Try to fit decision tree model to data
model <- rpart(total ~ "marks" + "score", data = dataframe)

Output:

Output

How to Fix the Error:

We can fix this error easily by simply removing quotations from the predictor variables and writing the formula as given in the code below:

R

# Importing the library
library(rpart)

# Make a data frame
dataframe <- data.frame(marks=c(86, 74, 99, 92, 77, 88, 82, 89),
                 score=c(11, 17, 22, 24, 27, 12, 29, 32),
                 total=c(17, 27, 16, 18, 16, 15, 27, 4))

# Try to fit decision tree model to data
model <- rpart(total ~ marks + score, data = dataframe)

# Print the summary of the model
summary(model)

Output: