One common error you may encounter in R is:

Error in `contrasts<-`(`*tmp*`, value = contr.funs[1 + isOF[nn]]) : 
  contrasts can be applied only to factors with 2 or more levels

This error occurs when you attempt to fit a regression model using a predictor variable that is either a factor or character and only has one unique value.

This tutorial shares the exact steps you can use to troubleshoot this error.

Example: How to Fix ‘contrasts can be applied only to factors with 2 or more levels’

Suppose we have the following data frame in R:

#create data frame
df <- data.frame(var1=c(1, 3, 3, 4, 5),
                 var2=as.factor(4),
                 var3=c(7, 7, 8, 3, 2),
                 var4=c(1, 1, 2, 8, 9))

#view data frame
df

  var1 var2 var3 var4
1    1    4    7    1
2    3    4    7    1
3    3    4    8    2
4    4    4    3    8
5    5    4    2    9

Notice that the predictor variable var2 is a factor and only has one unique value.

If we attempt to fit a multiple linear regression model using var2 as one of the predictor variables, we’ll get the following error:

#attempt to fit regression model
model <- lm(var4 ~ var1 + var2 + var3, data=df)

Error in `contrasts<-`(`*tmp*`, value = contr.funs[1 + isOF[nn]]) : 
  contrasts can be applied only to factors with 2 or more levels

We get this error because var2 only has one unique value: 4. Since there isn’t any variation at all in this predictor variable, R is unable to effectively fit a regression model.

We can actually use the following syntax to count the number of unique values for each variable in our data frame:

#count unique values for each variable
sapply(lapply(df, unique), length)

var1 var2 var3 var4 
   4    1    4    4 

And we can use the function to display each of the unique values for each variable:

#display unique values for each variable
lapply(df[c('var1', 'var2', 'var3')], unique)

$var1
[1] 1 3 4 5

$var2
[1] 4
Levels: 4

$var3
[1] 7 8 3 2

We can see that var2 is the only variable that has one unique value. Thus, we can fix this error by simply dropping var2 from the regression model:

#fit regression model without using var2 as a predictor variable
model <- lm(var4 ~ var1 + var3, data=df)

#view model summary
summary(model)

Call:
lm(formula = var4 ~ var1 + var3, data = df)

Residuals:
       1        2        3        4        5 
 0.02326 -1.23256  0.91860  0.53488 -0.24419 

Coefficients:
            Estimate Std. Error t value Pr(>|t|)  
(Intercept)   8.4070     3.6317   2.315   0.1466  
var1          0.6279     0.6191   1.014   0.4172  
var3         -1.1512     0.3399  -3.387   0.0772 .
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 1.164 on 2 degrees of freedom
Multiple R-squared:  0.9569,	Adjusted R-squared:  0.9137 
F-statistic: 22.18 on 2 and 2 DF,  p-value: 0.04314