CODEWITHSUNDEEP
rggplot2
mathematical.coffee
mathematical.coffee

Reputation: 56905

Using `geom_line()` with X axis being factors

Suppose I have a dataframe:

hist <- data.frame(date=Sys.Date() + 0:13,
                   counts=1:14)

I want to plot the total count against weekday, using a line to connect the points. The following puts points on each value:

hist <- transform(hist, weekday=factor(weekdays(date),
                                       levels=c('Monday', 'Tuesday', 'Wednesday', 'Thursday', 'Friday', 'Saturday', 'Sunday')))
ggplot(hist, aes(x=weekday, y=counts)) + geom_point(stat='summary', fun.y=sum)

When I try to connect them with a line (geom_line()), ggplot complains about only having one data observation per group and hence is not able to draw a line between the points.

I understand this - it's trying to draw one line for each weekday (factor level).

How can I get ggplot to just pretend (for the purposes of the line only) that the weekdays are numeric? Perhaps I have to have another column day_of_week that is 0 for monday, 1 for tuesday, etc?

Upvotes: 71

Views: 64836

Answers (2)

Fernando_U
Fernando_U

Reputation: 21

A solution that worked for me very easily was:

ggplot(data.frame, aes(X, Y)) + 
       geom_point() +
       geom_path(group=1) +   
       theme(axis.text.x = element_text(angle=90, hjust = 1, vjust = 0.5))

group = 1 lets ggplot know that you want to treat all the observations as part of the same group, and will therefore draw the line. The theme line is just for a cleaner look. You can delete it if you want.

Upvotes: 2

bdemarest
bdemarest

Reputation: 14667

If I understand the issue correctly, specifying group=1 and adding a stat_summary() layer should do the trick:

ggplot(hist, aes(x=weekday, y=counts, group=1)) +
geom_point(stat='summary', fun.y=sum) +
stat_summary(fun.y=sum, geom="line")

enter image description here

Upvotes: 78

Related Questions