R ggplot facet wrap misclassifying observation

Asked Oct 16 '18 at 20:00

Active Oct 16 '18 at 20:03

Viewed 35 times

My graph is functioning properly without facets and runs with facets, but classifies/wraps at least one individual incorrectly. I only noticed it because it is such an outlier. Everything else seems to be functioning properly. Without the facet_wrap, the top left-most point is "Typical" (this is confirmed through my data frame by just looking at it). With the facet_wrap, this point is now faceted with "Dyslexia." The same happens for facet_grid. All I'm doing to change between the two figures is deleting the last line of code here. I don't typically question R, but this is extra bizarre. Any ideas?

ggplot(x, aes(x$split1_FC, x$split2_FC, color = dys)) +
    geom_point(aes(colour = dys), size = 2) +
    geom_smooth(method = lm, se = F, color = 'black') +
    labs(x = 'Split 1', y = 'Split 2', title = paste('Reliability 
         of', component, 'split by', split)) +
    scale_color_discrete(name="Diagnosis") +
    theme_linedraw() +
    facet_wrap(x$dys)

No Facets

Facets

edited Oct 16 '18 at 20:03

bdemarest

14,397
3
53
56

asked Oct 16 '18 at 20:00

Sean McWeeny

2

As a wild guess, try removing `aes(colour = dys)` from inside of `geom_point()`. You definitely don't want both `color=dys` and `colour=dys`, which you have at the moment. Also change last line to `facet_wrap(~ dys)`, the way it was designed to be used. – bdemarest Oct 16 '18 at 20:06
2

Please `dput(x) ` – DJV Oct 16 '18 at 20:07
3

don't put $ in the aes – Richard Telford Oct 16 '18 at 20:11
2

Using `$` is certainly the problem. Correcting to `facet_wrap(~ dys)` and `aes(split1_FC, split2_FC, color = dys)` will fix. – Gregor Thomas Oct 16 '18 at 20:12
Check out the R-faq post on [never using `$` inside `aes`](https://stackoverflow.com/q/32543340/5325862) – camille Oct 16 '18 at 20:15
It was using $ for split1 and split 2, nothing else fixed it. Thanks all! – Sean McWeeny Oct 16 '18 at 20:20

R ggplot facet wrap misclassifying observation

0 Answers0