repeated scatterplots

[Doris-18]

Hi,

I read a message this morning which in principle refers also to my problem
(but the answer doesn`t help me because I don`t use Python).
I have lots of potential, but not consecutive predictors (t2pain, t2pfi,
t2vital etc) and I want to create scatterplots of each of these predictors
vs the dependent variable e.g.

GRAPH
  /SCATTERPLOT(BIVAR)=t2pain WITH logt2max
  /MISSING=LISTWISE .

Is there a possibility to avoid to repeat this step 29 times (for the
reminder 29 variables)? and to present the scatterplots of groups
(demograph, economic...) of predictors in one table (like the first column
in the matrix scatterplot)? Please note, I have only very limited
experience with macros.

Many thanks in advance for your help, Doris

[Spousta Jan]

Hi Doris,
The program may look like this one:


GET FILE='C:\Program Files\SPSS\Cars.sav'.

Define myscatte ()
!let !vars = "engine horse weight"
!do !i !in ( !vars ).

GRAPH
  /SCATTERPLOT(BIVAR)=!i WITH accel
  /MISSING=LISTWISE .

!doend
!enddefine.

myscatte .

Just replace the names of variables (here engine horse weight, and
accel).

Regarding the second question, I think you can produce only by editing
the images in a graphic editor or perhaps in Word.

HTH

Jan.

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Doris Gerstner
Sent: Tuesday, May 22, 2007 12:58 PM
To: [hidden email]
Subject: repeated scatterplots

Hi,

I read a message this morning which in principle refers also to my
problem (but the answer doesn`t help me because I don`t use Python).
I have lots of potential, but not consecutive predictors (t2pain, t2pfi,
t2vital etc) and I want to create scatterplots of each of these
predictors vs the dependent variable e.g.

GRAPH
  /SCATTERPLOT(BIVAR)=t2pain WITH logt2max
  /MISSING=LISTWISE .

Is there a possibility to avoid to repeat this step 29 times (for the
reminder 29 variables)? and to present the scatterplots of groups
(demograph, economic...) of predictors in one table (like the first
column in the matrix scatterplot)? Please note, I have only very limited
experience with macros.

Many thanks in advance for your help, Doris

[Mike P-5]

Basic idea

*You variable name.
DEFINE !VANAMES ()
t2pain t2pfi t2vital.. all 29
!ENDDEFINE.

DEFINE !LABPRO (!POSITIONAL !TOKENS(29))
!LET !vin = !1.
!DO !vi !IN (!vin).
GRAPH
  /SCATTERPLOT(BIVAR)=!vi WITH logt2max
  /MISSING=LISTWISE .
!DOEND
!ENDDEFINE.

!LABPRO !VANAMES.

HtH

Mike



-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Doris Gerstner
Sent: 22 May 2007 11:58
To: [hidden email]
Subject: repeated scatterplots

Hi,

I read a message this morning which in principle refers also to my
problem
(but the answer doesn`t help me because I don`t use Python).
I have lots of potential, but not consecutive predictors (t2pain, t2pfi,
t2vital etc) and I want to create scatterplots of each of these
predictors
vs the dependent variable e.g.

GRAPH
  /SCATTERPLOT(BIVAR)=t2pain WITH logt2max
  /MISSING=LISTWISE .

Is there a possibility to avoid to repeat this step 29 times (for the
reminder 29 variables)? and to present the scatterplots of groups
(demograph, economic...) of predictors in one table (like the first
column
in the matrix scatterplot)? Please note, I have only very limited
experience with macros.

Many thanks in advance for your help, Doris

________________________________________________________________________
This e-mail has been scanned for all viruses by Star. The
service is powered by MessageLabs. For more information on a proactive
anti-virus service working around the clock, around the globe, visit:
http://www.star.net.uk
________________________________________________________________________

________________________________________________________________________
This e-mail has been scanned for all viruses by Star. The
service is powered by MessageLabs. For more information on a proactive
anti-virus service working around the clock, around the globe, visit:
http://www.star.net.uk
________________________________________________________________________

[Peck, Jon]

To get the paneling effect, add something like this to your graph command

/panel rowvar = mygroupingvar

If you have a lot of groups, you will probably want to increase the vertical size of the plot, which you can do by just dragging the border in the Viewer.

HTH,
Jon Peck



-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Spousta Jan
Sent: Tuesday, May 22, 2007 6:38 AM
To: [hidden email]
Subject: Re: [SPSSX-L] repeated scatterplots

Hi Doris,
The program may look like this one:


GET FILE='C:\Program Files\SPSS\Cars.sav'.

Define myscatte ()
!let !vars = "engine horse weight"
!do !i !in ( !vars ).

GRAPH
  /SCATTERPLOT(BIVAR)=!i WITH accel
  /MISSING=LISTWISE .

!doend
!enddefine.

myscatte .

Just replace the names of variables (here engine horse weight, and
accel).

Regarding the second question, I think you can produce only by editing
the images in a graphic editor or perhaps in Word.

HTH

Jan.

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Doris Gerstner
Sent: Tuesday, May 22, 2007 12:58 PM
To: [hidden email]
Subject: repeated scatterplots

Hi,

I read a message this morning which in principle refers also to my
problem (but the answer doesn`t help me because I don`t use Python).
I have lots of potential, but not consecutive predictors (t2pain, t2pfi,
t2vital etc) and I want to create scatterplots of each of these
predictors vs the dependent variable e.g.

GRAPH
  /SCATTERPLOT(BIVAR)=t2pain WITH logt2max
  /MISSING=LISTWISE .

Is there a possibility to avoid to repeat this step 29 times (for the
reminder 29 variables)? and to present the scatterplots of groups
(demograph, economic...) of predictors in one table (like the first
column in the matrix scatterplot)? Please note, I have only very limited
experience with macros.

Many thanks in advance for your help, Doris