Contrasts can be applied only to factor

Question

I have a question about R.

I am using a test called levene.test to test a homogeneity of variance.

I know that you need a factor variable with at least two l

Answer 1

You need to actually convert your variable to a factor. Just having three (or a finite) number of values does not necessarily make it a factor.

use x <- factor(x) to convert

---

When you look at the output of str(), it shows you the type of each variable:

<..cropped..>
$ SIF1         : num  19.6 17 NA 23.8 24.1 ...
$ sex          : Factor w/ 2 levels "0","1": 1 1 2 2 2 2 1 1 1 1 ...
$ k            : Factor w/ 3 levels "0","1","2": 1 1 2 3 1 3 3 3 1 2 ...

notice that $k is a factor but SIF1 is not
Thus, use

 geno1rs11809462$SIF1 <- factor(geno1rs11809462$SIF1)

Answer 2

If your factor has only one level, you will get this error. To check to see the levels of your factor variables, use lapply(df, levels). It will return nothing for non-factor variables, but will easily let you identify which variable is the offender. This is especially helpful if, like me, you have hundreds of variables.

Answer 3

I think I may have solved the problem. I believe it is due to NA value in the data. Because after I removed the na using say

x<-na.omit(original_data)

then apply the levene test on x, the warning message disappears.

Hopefully this is the cause of the problem.