- Graeme, you could do a scatterplot first (perhaps in a light color) with points labeled if desired, then do the boxplots next, which would effectively superimpose them.
I do this (although with error bars rather than boxplots) in some of my production stuff.
I realize this isn't ideal compared with the old functionality. I can't think of any way to communicate to scatterplot which data points qualify as outliers.
I'll add a note to the requested features list on this.
From: email@example.com [firstname.lastname@example.org] On Behalf Of Mackay [mackay_30@...]
Sent: Wednesday, September 24, 2008 8:57 AM
To: Stephen Grubb
Sorry for not replying in the forum, just a quickie.. Boxplots with
outliers are very key in my industry.
Do you know of a work around with something like scatter plot or the