CODEWITHSUNDEEP
rggplot2
moth
moth

Reputation: 2349

ggplot difference in aes()

Consider a plot of a df in ggplot:

ggplot(data = df, aes(x=X,y=Y)) +
geom_point(aes(color=Trial))

The variable Trial in this case is a factor. Could someone explain me what is the difference between plotting the factor inside the geom_point function and outside as in the case below :

ggplot(data = df, aes(x=X,y=Y,color=Trial))

Here is a sample of the df :

structure(list(Sample = c("GID7173723", "GID4878677", "GID88208", 
"GID346403", "GID268825", "GID7399578"), Trial = structure(c(4L, 
6L, 5L, 5L, 5L, 6L), .Label = c("ES1-5", "ES6-13", "ES14-26", 
"ES27-38", "SA1-13", "SA14-25"), class = "factor"), X = c(8.68082802271727, 
-52.0699224029511, -23.9642181294958, 53.3466936371821, -85.064856918598, 
33.4668785456285), Y = c(-35.3997039478218, 46.2365967946906, 
-42.8190962627021, 24.245938561458, 95.9865691321666, 25.6522750117316
), Z = c(12.3326491737533, -24.7722861720316, -11.7262667337085, 
-43.0492006899678, -51.7268052275685, 49.6715770397554)), row.names = c(NA, 
6L), class = "data.frame")

Upvotes: 3

Views: 726

Answers (2)

Su Na
Su Na

Reputation: 344

In this case there is no difference in the "aesthetic mapping" section. The actrual problem is that every ggplot object need at least one "geometric layer" to work functionally. In other words, you need to tell ggplot object to use point to represent the X and Y value by the coordinate of these points, which is exactly what function geom_point() do.

So 

ggplot(data = df, aes(x = X, y = Y)) + geom_point(aes(color = Trial))

p1

And

ggplot(data = df, aes(x = X, y = Y, color = Trial)) + geom_point()

p2

Upvotes: 1

Julius Vainora
Julius Vainora

Reputation: 48191

I suppose you meant

ggplot(data = df, aes(x = X, y = Y)) +
  geom_point(aes(color = Trial))

versus

ggplot(data = df, aes(x = X, y = Y, color = Trial)) +
  geom_point()

rather than just the first line, without geom_point. Not including geom_point clearly would not give any points regardless of specified aesthetics.

Now those two code excerpts, as you can see, give identical results in this case. More generally, the difference would be in that in the first case you set the color aesthetic only for geom_point, while in the second case it is set globally. That is, you could use

ggplot(data = df, aes(x = X, y = Y, color = Trial)) +
  geom_point() + geom_line()

enter image description here

and now both geoms would use the same color aesthetic. On the other hand,

ggplot(data = df, aes(x = X, y = Y)) +
  geom_point(aes(color = Trial)) + geom_line()

enter image description here

would give black lines. Also note that as a side effect lines are also grouped by Trial.

Upvotes: 3

Related Questions