[R] lm() with spearman corr option ?

[Hoji, Akihiko]

Hi,

A following function was  kindly provided by GGally’s maintainer, Barret Schloerke.

function(data, mapping, ...) {
    p <- ggplot(data = data, mapping = mapping) +
        geom_point(color = I("blue")) +
        geom_smooth(method = "lm", color = I("black"), ...) +
        theme_blank() +
        theme(panel.border=element_rect(fill=NA, linetype = "solid", color="black"))

    lmModel <- eval(substitute(lm(y ~ x, data = data), mapping))
    fs <- summary(lmModel)$fstatistic
    pValue <- pf(fs[1], fs[2], fs[3], lower.tail = FALSE)

    if (pValue < 0.05) {
        p <- p + theme(
            panel.border = element_rect(
                color = "red",
                size = 3,
                linetype = "solid",
                fill = "transparent"
            )
        )
    }

    p
}

Basically, this function draws red squares  over pairwise corr plots with p<0.05.  Now, since I need to use the spearman rank corr, I tried to modify the lm function by adding “method=spearman”  but this did not work at al.  Could anybody suggest the way to add the spearman rank corr function in this particular function ?

Thanks.




-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 842 bytes
Desc: Message signed with OpenPGP using GPGMail
URL: <https://stat.ethz.ch/pipermail/r-help/attachments/20160429/ac0f5a6a/attachment.bin>