FW: Line Graph of Repeated Measures Data with One Line for Each Case?
Locked
FW: Line Graph of Repeated Measures Data with One Line for Each Case?
 
|
Oops, the d was stripped from
the data list command in the last email, so here’s the corrected version. From: ViAnn Beadle
[mailto:[hidden email]] If you have no categorical
variable (as in your example below), then the mean is the observed value. The
complication arises when you add a categorical variable such as school to the
algebra, because then you get the mean of the school. You can use the split
function to split the lines by student id. As has been discussed here
repeatedly, things get very complicated within GPL when you try to compare
multiple variables. It was really designed to compare rows of data. It is true
that the GGRAPH provides a shortcut for restructuring on the fly via the
TRANSFORM keyword, but this is tied to aggregation and now your mean is an mean
within school. If multiple students are in the same school, you do not get what
you want. The simplest way to approach
this is to restructure the data using VARSTOCASES command first to generate a
dataset in which each row is a student’s score on a single
administration. The following simple example does that and then uses split to
get a line for each student. data
list list/ student test1 test2 test3 test4 (5f1.0)school (A1). begin data 1 5 3 4 2 A 2 3 3 3 4 A 3 4 3 5 2 B 4 2 2 2 4 B end data. VARSTOCASES MAKE testresult from
test1 to test4 / INDEX = time. * Chart Builder. variable level time student
(nominal) testresult(scale). * Chart Builder. GGRAPH /GRAPHDATASET
NAME="graphdataset" VARIABLES=time testresult[LEVEL=SCALE] school
student /GRAPHSPEC
SOURCE=INLINE. BEGIN GPL SOURCE:
s=userSource(id("graphdataset")) DATA: time=col(source(s),
name("time"), unit.category()) DATA:
testresult=col(source(s), name("testresult")) DATA:
school=col(source(s), name("school"), unit.category()) DATA:
student=col(source(s), name("student"), unit.category()) GUIDE: axis(dim(1),
label("time")) GUIDE: axis(dim(2),
label("testresult")) GUIDE:
legend(aesthetic(aesthetic.color.interior), label("school")) SCALE: linear(dim(2),
include(0)) ELEMENT:
line(position(time*testresult), color.interior(school), split(student),
missing.wings()) END GPL. From: SPSSX(r) Discussion
[mailto:[hidden email]] On Behalf Of Justin Meyer Hi
SPSS Listers, I want
to graph data from 4 test administrations on a line graph, with the y axis as
each student’s test score and the x-axis the 4 administrations. My data has
each student as a case with one text variable indicating school and each of the
four test scores as a scale variable. I dragged the four scores into the y axis
box in chart builder but it will only let me summarize the data as a mean, etc
(shown below). How do I get it to show the line for each student? I suppose I
could create a variable with a unique number for each student and then use it
as a grouping variable, but there has to be a cleaner method. Once I have a
line for each student, how can I color code them by school? Thanks! * Chart Builder. GGRAPH /GRAPHDATASET
NAME="graphdataset" VARIABLES=MEAN(SS4) MEAN(sSS4) MEAN( F08SS4) MEAN(S09SS4)
MISSING=LISTWISE REPORTMISSING=NO
TRANSFORM=VARSTOCASES(SUMMARY="#SUMMARY" INDEX="#INDEX") /GRAPHSPEC SOURCE=INLINE. BEGIN GPL SOURCE:
s=userSource(id("graphdataset")) DATA: INDEX=col(source(s),
name("#INDEX"), unit.category()) DATA: SUMMARY=col(source(s),
name("#SUMMARY")) GUIDE: axis(dim(1),
label("INDEX")) GUIDE: axis(dim(2),
label("Mean")) SCALE: cat(dim(1),
include("0", "1", "2", "3")) SCALE: linear(dim(2), include(0)) ELEMENT:
line(position(INDEX*SUMMARY), missing.wings()) END GPL. ____________________________________ Researcher Rowland
Reading Foundation 6120
University Avenue Middleton,
WI 53562 phone:
866-370-7323 fax: 608-204-3846 ____________________________________ |
| Free forum by Nabble | Edit this page |