R and ggplot- Changing x axis to Date removes position dodge

Question

I've been using ggplot to create plots, and I always like to offset my data points horizontally so the error bars don't overlap. I've found that when I used Date data for my x axis, I loose the ability to offset my data points.

DF=data.frame(Date=c("2006-09-01", "2007-09-01", "2008-09-01" ,"2009-09-01", "2010-01-01" ,"2011-01-01"),Count=c(1,2,3,4,5,6,2,4,6,8,10,12),Type=c('A','A','A','A','A','A','B','B','B','B','B','B'))

If I plot the above data frame using ggplot like this, the points are offset

dodge=position_dodge(.5)
ggplot(data=DF,aes(x=Date,y=Count,group=Type))+geom_point(position=dodge)

But if I designate the Date data as actual dates

DF$Date=as.Date(DF$Date)
ggplot(data=DF,aes(x=Date,y=Count,group=Type))+geom_point(position=dodge)

The points are no longer offset, even though I specify them to be

I've avoided this problem in the past by keeping my date data limited to only "year" and treating it as numeric. But now I need to include month information (I include the day in my date since "as.Date gets a little funky if I don't include it).

So can anyone tell me if there is a way I can plot a time series using ggplot that will allow me to offset my data points and error bars.

Answer 1

Actually the points are dodged slightly.

You need to keep in mind that the interval between one year is 365 days. position_dodge(0.5) moves the point by (a half day)/2. So, the position moves 0.5/2/365 of one year interval.

If you choose `position_dodge(0.5 * 365), then you will get the plot as expected:

DF <- data.frame(
  expand.grid(Date=c("2006-09-01", "2007-09-01", "2008-09-01" ,"2009-09-01"),
     Type = c("A", "B")),
  Count = 1:4)
DF$Date2 <- as.Date(DF$Date)

ggplot(data=DF,aes(x=Date2,y=Count,group=Type))+ geom_point(position = position_dodge(365*0.5))