Graph for Two Multicategory Variables

[Brian Dates]

I wonder if it's possible to use GPL to place two separate multiple-category
variables which share the same categories and and scale into one graph. I
already have graphs that combine multiple single categories for separate
variables, but am stymied about how to put multiple category variables
together. I'm doing this for a matched t-test, so splitting one variable
isn't possible unless I restructure the data, which I'm trying to avoid
because I have a number of other graphs which depend on the present
structure. I've attached a word document with the two variables I'm trying
to combine.

As always, thanks for any help you can provide.

Brian

Sample_Graph.docx
<http://spssx-discussion.1045642.n5.nabble.com/file/t341458/Sample_Graph.docx>  



--
Sent from: http://spssx-discussion.1045642.n5.nabble.com/

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[Jon Peck]

Yes, you can do this with GPL.  However, you might want to try the STATS COMPARE SUBGROUPS extension command which does something like that although with a number of other features.  You can install it from the Extensions > Extension Hub menu if you don’t already have it, and it will appear on the Graphs menu.

On Mon, Nov 26, 2018 at 6:51 PM Brian Dates <[hidden email]> wrote:

I wonder if it's possible to use GPL to place two separate multiple-category
variables which share the same categories and and scale into one graph. I
already have graphs that combine multiple single categories for separate
variables, but am stymied about how to put multiple category variables
together. I'm doing this for a matched t-test, so splitting one variable
isn't possible unless I restructure the data, which I'm trying to avoid
because I have a number of other graphs which depend on the present
structure. I've attached a word document with the two variables I'm trying
to combine.

As always, thanks for any help you can provide.

Brian

Sample_Graph.docx
<http://spssx-discussion.1045642.n5.nabble.com/file/t341458/Sample_Graph.docx> 



--
Sent from: http://spssx-discussion.1045642.n5.nabble.com/

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

--

===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD

[Andy W]

If you just want to do a clustered bar chart this is actually easy via the
legacy graph commands. Below I show that and some other ways to do it via
GGRAPH.

*********************************************************************************************.
DATA LIST FREE / Cat (A1) V1 V2 (2F3.0).
BEGIN DATA
A 40 30
B 20 60
C 10 40
END DATA.
DATASET NAME Bars.


GRAPH /BAR(GROUPED)=MEAN(V1) MEAN(V2) BY Cat.

*GGRAPH code examples.

*Superimposed points.
GGRAPH
  /GRAPHDATASET NAME="graphdataset" VARIABLES=Cat V1 V2
  /GRAPHSPEC SOURCE=INLINE.
BEGIN GPL
  SOURCE: s=userSource(id("graphdataset"))
  DATA: Cat=col(source(s), name("Cat"), unit.category())
  DATA: V1=col(source(s), name("V1"))
  DATA: V2=col(source(s), name("V2"))
  GUIDE: axis(dim(1), label("Cat"))
  GUIDE: axis(dim(2), label("Y Value"))
  SCALE: linear(dim(2), include(0))
  ELEMENT: point(position(Cat*V2), size(size."12"), shape("V2"),
color.interior("V2"))
  ELEMENT: point(position(Cat*V1), size(size."12"), shape("V1"),
color.interior("V1"))
END GPL.
*Might want to also dodge if many points are overlapping, see.
*https://andrewpwheeler.wordpress.com/2013/03/06/some-random-spss-graph-tips-shading-areas-under-curves-and-using-dodging-in-binned-dot-plots/.
*Would need to reshape though wide to long to do that.

*Dodged dot plot with connecting edge.
GGRAPH
  /GRAPHDATASET NAME="graphdataset" VARIABLES=Cat MEAN(V1) MEAN(V2)
   TRANSFORM=VARSTOCASES(SUMMARY="MaxBar" INDEX="V")
  /GRAPHSPEC SOURCE=INLINE
  /FRAME INNER=YES.
BEGIN GPL
  SOURCE: s=userSource(id("graphdataset"))
  DATA: Cat=col(source(s), name("Cat"), unit.category())
  DATA: V=col(source(s), name("V"), unit.category())
  DATA: MaxBar=col(source(s), name("MaxBar"))
  TRANS: base = eval(0)
  COORD: rect(dim(1,2), cluster(3))
  GUIDE: axis(dim(1), label("Cat"))
  GUIDE: axis(dim(2), label("Y Value"))
  SCALE: linear(dim(2), include(0))
  ELEMENT: edge(position(V*(base + MaxBar)*Cat), color(V))
  ELEMENT: point(position(V*MaxBar*Cat), size(size."12"), shape(V),
color.interior(V))
END GPL.
*********************************************************************************************.



-----
Andy W
[hidden email]
http://andrewpwheeler.wordpress.com/
--
Sent from: http://spssx-discussion.1045642.n5.nabble.com/

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD