ggplot2 adding regression line(logic)

Question

I have 2 different lines of code:

ggplot(mpg,aes(displ,hwy,colour = factor(cyl))) + geom_point() + geom_smooth(method = 'lm')

ggplot(mpg,aes(displ,hwy)) + geom_point(aes(colour = factor(cyl))) + geom_smooth(method = 'lm')

The first code produces 3 regression lines for 3 different groups(factor variable) . Second code produces only one line for the whole dataset.

My question is: What is the logic behind this difference? I see, that the output depends on colour = factor(cyl) but can you explain me the logic of ggplo2 in this case?

Well, I'd think of it like the 1st one uses a global aesthetics mapping, which is - by default - inherited by every geom. The 2nd one uses a local mapping for the color aesthetic. — lukeA
– lukeA, Commented Dec 17, 2016 at 17:41

Sandipan Dey · Accepted Answer · 2016-12-17 18:24:17Z

1

Consider the following 2 lines of code which are equivalent (in the first case we are grouping both the geom_point and geom_smooth by globally providing the colour variable in ggplot and in the second case locally with both geom_point and geom_smooth):

ggplot(mpg,aes(displ,hwy,colour = factor(cyl))) + 
geom_point() + 
geom_smooth(method = 'lm')

ggplot(mpg,aes(displ,hwy)) + 
geom_point(aes(colour = factor(cyl))) + 
geom_smooth(aes(colour = factor(cyl)), method = 'lm')

Now the other example code that you provided, there you color the points with different colors by grouping them with cyl variable but you dont group geom_smooth that's why it fits on the entire data instead of fitting on 3 groups separately like the earlier case.

ggplot(mpg,aes(displ,hwy)) + 
geom_point(aes(colour = factor(cyl))) + 
geom_smooth(method = 'lm')

answered Dec 17, 2016 at 18:24

Sandipan Dey

23.4k4 gold badges59 silver badges72 bronze badges

Sign up to request clarification or add additional context in comments.

Collectives™ on Stack Overflow

ggplot2 adding regression line(logic)

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related