I was trying to make a correlation matrix with the scatterplotMatrix() command in R and get this error:
scatterplotMatrix (a_heterophylla_NAY_SIN, spread = FALSE, lty.smooth = 2, main="Matrix correlations") Error in pairs.default (x, labels = var.labels, cex.axis = cex.axis, cex.main = cex.main,:
non-numeric argument to 'pairs'
What does it mean and how do I resolve it? Please help !!
I will try to answer your question based on the information you gave. Fundamentally the problem is data, particularly in a_heterophylla_NAY_SIN , the error triggers when scatterplotMatrix() invokes internally pairs() assigning the x parameter to your a_heterophylla_NAY_SIN object, if you see the documentation:
x: the coordinates of points given as numeric columns of a matrix or data frame Logical and factor columns are converted to numeric in the same way that data.matrix does.
That is, a matrix or a dataframe is expected and fundamentally all values must be "mapped" to numerical values. Reviewing the code of pairs() , we see:
if (!is.matrix(x)) {
x <- as.data.frame(x)
for(i in seq_along(names(x))) {
if(is.factor(x[[i]]) || is.logical(x[[i]]))
x[[i]] <- as.numeric(x[[i]])
if(!is.numeric(unclass(x[[i]])))
stop("non-numeric argument to 'pairs'")
}
} else if (!is.numeric(x)) stop("non-numeric argument to 'pairs'")
That is, o a_heterophylla_NAY_SIN is a non-numeric matrix or an object that is still mapped as a dataframe, some of its columns are not numeric. The solution would then go through "normalize" a_heterophylla_NAY_SIN , and eventually convert any non-numeric value into a Factor, so that it can be processed correctly.