Date:         Wed, 25 Jul 2007 10:31:24 -0600
Reply-To:     ViAnn Beadle <vab88011@gmail.com>
Sender:       "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From:         ViAnn Beadle <vab88011@gmail.com>
Subject:      Re: Old Sunflower Option
Comments: To: "William B. Ware" <wbware@email.unc.edu>
In-Reply-To:  <Pine.LNX.4.63+UNC.0707251147160.28503@login01.its.unc.edu>
Content-Type: text/plain; charset="us-ascii"

The nearest equivalent I can think of to the old sunflower option is the
binhex function in GPL. It groups together nearby points and then you use
the summary.count function to count the number of hits within the bin to set
the size of the point displayed. This produces a plot sometimes referred to
as bubble plot.

Here's some sample syntax which produces a "bubble" plot using the employee
data.sav sample file. The SCALE command constrains the minimum point size to
5 pixels. IMHO, the default sizing creates really, really small 1 pixel
points which look like dust on my monitor--so this gets around that. The
color.interior fills in the points (defaults to circles). I think the
default hollow points are ugly--so this gets around that.


GGRAPH
  /GRAPHDATASET NAME="graphdataset" VARIABLES=prevexp jobtime MISSING=
  LISTWISE REPORTMISSING=NO
  /GRAPHSPEC SOURCE=INLINE.
BEGIN GPL
 SOURCE: s=userSource(id("graphdataset"))
 DATA: prevexp=col(source(s), name("prevexp"))
 DATA: jobtime=col(source(s), name("jobtime"))
 GUIDE: axis(dim(1), label("Previous Experience (months)"))
 GUIDE: axis(dim(2), label("Months since Hire"))
 SCALE: linear(aesthetic(aesthetic.size), aestheticMinimum(size."5px"))
ELEMENT: point(position(bin.hex(prevexp*jobtime, dim(1,2))),
size(summary.count()), color.interior(color.blue))
END GPL.

The IGRAPH procedure also provides a jittering option which nudges the
points slightly apart by adding a small amount of random variation, but I
think the GPL binhex approach works much better.

-----Original Message-----
From: SPSSX(r) Discussion [mailto:SPSSX-L@LISTSERV.UGA.EDU] On Behalf Of
William B. Ware
Sent: Wednesday, July 25, 2007 9:49 AM
To: SPSSX-L@LISTSERV.UGA.EDU
Subject: Old Sunflower Option

In the older versions of SPSS, there was something called the sunflower
option to show multiple cases at the same point in a scatter plot.  Does
anyone know how to do that in Version 14 and up?

Bill

__________________________________________________________________________
William B. Ware, Professor                         Educational Psychology,
CB# 3500                                       Measurement, and Evaluation
University of North Carolina                         PHONE  (919)-962-7848
Chapel Hill, NC      27599-3500                      FAX:   (919)-962-1533
Office:  118 Peabody Hall                            EMAIL: wbware@unc.edu
Adjunct Professor                                    School of Social Work
__________________________________________________________________________