Posted by
Anthony Babinec on
Apr 13, 2007; 3:26pm
URL: http://spssx-discussion.165.s1.nabble.com/individual-graphs-of-change-tp1075013p1075018.html
You can do individual trajectories using the Chart Builder in SPSS 15. NOTE
that I have not pushed this "to the limit" with large datasets, but here's a
small example.
Consider two patients with a number of seizures noted over consecutive
2-week periods. Here are the data.
Id Time Seizures
101 1 11
101 2 14
101 3 9
101 4 8
102 1 8
102 2 7
102 3 9
102 4 4
Make sure that ID has its MEASURE set to NOMINAL, while the others are set
to SCALE.
Then, you do a scatterplot of Seizures by Time, with ID as a GroupColor
variable. After you generate the Chart, edit it and add interpolation lines.
* Chart Builder.
GGRAPH
/GRAPHDATASET NAME="graphdataset" VARIABLES=time seizure id MISSING=
LISTWISE REPORTMISSING=NO
/GRAPHSPEC SOURCE=INLINE.
BEGIN GPL
SOURCE: s=userSource(id("graphdataset"))
DATA: time=col(source(s), name("time"))
DATA: seizure=col(source(s), name("seizure"))
DATA: id=col(source(s), name("id"), unit.category())
GUIDE: axis(dim(1), label("time"))
GUIDE: axis(dim(2), label("seizure"))
GUIDE: legend(aesthetic(aesthetic.color.exterior), label("id"))
SCALE: cat(aesthetic(aesthetic.color.exterior))
ELEMENT: point(position(time*seizure), color.exterior(id))
END GPL.
-----Original Message-----
From: SPSSX(r) Discussion [mailto:
[hidden email]] On Behalf Of
Zdaniuk, Bozena
Sent: Thursday, April 12, 2007 2:00 PM
To:
[hidden email]
Subject: Re: individual graphs of change
Thanks, Gene. I have the book and I found the syntax. It's done using
igraph. Now I have to go search through our list's archives to figure
out the case limit for igraphs. I have 1200 cases (times 4 time points)
and the igraph function I tried to run hung up on me instantly. I
vaguely recall there was a discussion already on the case limits of
igraph functions. Thanks again to all who helped me work through that.
b.
-----Original Message-----
From: SPSSX(r) Discussion [mailto:
[hidden email]] On Behalf Of
Gene Maguin
Sent: Thursday, April 12, 2007 1:48 PM
To:
[hidden email]
Subject: Re: individual graphs of change
Bozena,
Correct me if I'm wrong, but I think that you must be working on a
multilevel analysis and you want to see either the individual points
plotted
by perons and then connected with lines or the individual regression
lines
through those points. Specific examples of both of these are described
in
Judith Singer's book. Unfortunately, mine is at home.
Gene