SPSSX Discussion

creating multiple charts with chart builder

Classic

List

Threaded

6 messages Options

LUCINDA M TEAR

creating multiple charts with chart builder

Hello,

I am just venturing into Chart Builder instead of using Interactive Charts because I want to be able to use the "logarithmic" axis transformation on the scale axis when I am creating a set of multiple boxplots.

I have read the syntax help, but I am not sure how to create a "macro" that will allow me create multiple boxplots, one for each of 11 variables, such that each boxplot uses the variable LABEL name as the axis label and does NOT include labels for outliers.

I have the following syntax - I have tried to use the "NAME=" function so that I would have to change only the variable name after the VARIABLES = command, but this doesn't work to change the axis label and I'm not sure how to get it to read a list of variable names so that several boxplots are created.

Any help would be appreciated! Thank you, Lucinda Tear

* Chart Builder.

GGRAPH

/GRAPHDATASET NAME="graphdataset" VARIABLES=chem1 [NAME="chem"] MISSING=VARIABLEWISE

REPORTMISSING=NO

/GRAPHSPEC SOURCE=INLINE

TEMPLATE=["C:\Program Files\SPSS\Looks\air_boxes.sgt"].

BEGIN GPL

SOURCE: s=userSource(id("graphdataset"))

DATA: chem=col(source(s), name("chem")) #I replaced chem1 with "chem" here to try to make it more generic

DATA: id=col(source(s), name("$CASENUM"), unit.category())

COORD: rect(dim(1), transpose())

GUIDE: axis(dim(1), label("WHAT CAN I PUT HERE SO THAT THE AXIS LABEL NAME CHANGES WITH EACH VARIABLE"))

SCALE: log(dim(1), base(10))

ELEMENT: schema(position(bin.quantile.letter(chem)), label(id)) #I replaced chem1 with "chem" here to try to make it more generic

END GPL.

====================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

ViAnn Beadle

Re: creating multiple charts with chart builder

When Chart Builder creates syntax, it knows what the variable label for the
selected variable is and uses that when generating the GUIDE statement.
Unlike Chart Builder, macros are simple substitutions of strings and have no
direct knowledge of dictionary information. I have seen some solutions in
which SPSS itself is coaxed into writing a macro with dictionary information
within it but I haven't a clue as to how to go about this.

It's fairly easy to use python to retrieve a variable label from a variable
name and place that in the middle of the label function on the GUIDE
statement. However, I don't think that you can embed a python program inside
of a macro used to generate 11 separate instances of the GGRAPH command. You
might be able to use a python program which does the looping through the
names and I'm sure that python fans on the list can come up with something
that works.

Tell us more about your sets of 11 variables. Do they have some common name
part to them such as chem1 to chem11?

On a simpler note, to suppress the outlier labels, omit the label(id)
function from your ELEMENT statement and drop the DATA statement defining
the id variable.

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
LUCINDA M TEAR
Sent: Wednesday, August 13, 2008 5:29 PM
To: [hidden email]
Subject: creating multiple charts with chart builder

Hello,

I am just venturing into Chart Builder instead of using Interactive Charts
because I want to be able to use the "logarithmic" axis transformation on
the scale axis when I am creating a set of multiple boxplots.

I have read the syntax help, but I am not sure how to create a "macro" that
will allow me create multiple boxplots, one for each of 11 variables, such
that each boxplot uses the variable LABEL name as the axis label and does
NOT include labels for outliers.

I have the following syntax - I have tried to use the "NAME=" function so
that I would have to change only the variable name after the VARIABLES =
command, but this doesn't work to change the axis label and I'm not sure how
to get it to read a list of variable names so that several boxplots are
created.

Any help would be appreciated! Thank you, Lucinda Tear

* Chart Builder.

GGRAPH

/GRAPHDATASET NAME="graphdataset" VARIABLES=chem1 [NAME="chem"]
MISSING=VARIABLEWISE

REPORTMISSING=NO

/GRAPHSPEC SOURCE=INLINE

TEMPLATE=["C:\Program Files\SPSS\Looks\air_boxes.sgt"].

BEGIN GPL

SOURCE: s=userSource(id("graphdataset"))

DATA: chem=col(source(s), name("chem")) #I replaced chem1 with "chem" here
to try to make it more generic

DATA: id=col(source(s), name("$CASENUM"), unit.category())

COORD: rect(dim(1), transpose())

GUIDE: axis(dim(1), label("WHAT CAN I PUT HERE SO THAT THE AXIS LABEL NAME
CHANGES WITH EACH VARIABLE"))

SCALE: log(dim(1), base(10))

ELEMENT: schema(position(bin.quantile.letter(chem)), label(id)) #I replaced
chem1 with "chem" here to try to make it more generic

END GPL.

=======
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

LUCINDA M TEAR

Re: creating multiple charts with chart builder

Thank you, ViAnn for your help re: outlier labels. My variable names (actually, I have 19 - more than I thought!) do have common names: chem1 to chem19. I tried using the !concat() function and creating a (define - !enddefine) macro, but that didn't work. The syntax guide for ggraph indicates that name can be used to create multiple graphs with the same syntax, but doesn't provide any examples of that....

Thanks again!

Lucinda

----- Original Message -----
From: ViAnn Beadle<mailto:[hidden email]>
To: [hidden email]<mailto:[hidden email]>
Sent: Wednesday, August 13, 2008 4:57 PM
Subject: Re: creating multiple charts with chart builder

When Chart Builder creates syntax, it knows what the variable label for the
selected variable is and uses that when generating the GUIDE statement.
Unlike Chart Builder, macros are simple substitutions of strings and have no
direct knowledge of dictionary information. I have seen some solutions in
which SPSS itself is coaxed into writing a macro with dictionary information
within it but I haven't a clue as to how to go about this.

It's fairly easy to use python to retrieve a variable label from a variable
name and place that in the middle of the label function on the GUIDE
statement. However, I don't think that you can embed a python program inside
of a macro used to generate 11 separate instances of the GGRAPH command. You
might be able to use a python program which does the looping through the
names and I'm sure that python fans on the list can come up with something
that works.

Tell us more about your sets of 11 variables. Do they have some common name
part to them such as chem1 to chem11?

On a simpler note, to suppress the outlier labels, omit the label(id)
function from your ELEMENT statement and drop the DATA statement defining
the id variable.

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
LUCINDA M TEAR
Sent: Wednesday, August 13, 2008 5:29 PM
To: [hidden email]<mailto:[hidden email]>
Subject: creating multiple charts with chart builder

Hello,

I am just venturing into Chart Builder instead of using Interactive Charts
because I want to be able to use the "logarithmic" axis transformation on
the scale axis when I am creating a set of multiple boxplots.

I have read the syntax help, but I am not sure how to create a "macro" that
will allow me create multiple boxplots, one for each of 11 variables, such
that each boxplot uses the variable LABEL name as the axis label and does
NOT include labels for outliers.

I have the following syntax - I have tried to use the "NAME=" function so
that I would have to change only the variable name after the VARIABLES =
command, but this doesn't work to change the axis label and I'm not sure how
to get it to read a list of variable names so that several boxplots are
created.

Any help would be appreciated! Thank you, Lucinda Tear

* Chart Builder.

GGRAPH

/GRAPHDATASET NAME="graphdataset" VARIABLES=chem1 [NAME="chem"]
MISSING=VARIABLEWISE

REPORTMISSING=NO

/GRAPHSPEC SOURCE=INLINE

TEMPLATE=["C:\Program Files\SPSS\Looks\air_boxes.sgt"].

BEGIN GPL

SOURCE: s=userSource(id("graphdataset"))

DATA: chem=col(source(s), name("chem")) #I replaced chem1 with "chem" here
to try to make it more generic

DATA: id=col(source(s), name("$CASENUM"), unit.category())

COORD: rect(dim(1), transpose())

GUIDE: axis(dim(1), label("WHAT CAN I PUT HERE SO THAT THE AXIS LABEL NAME
CHANGES WITH EACH VARIABLE"))

SCALE: log(dim(1), base(10))

ELEMENT: schema(position(bin.quantile.letter(chem)), label(id)) #I replaced
chem1 with "chem" here to try to make it more generic

END GPL.

=======
To manage your subscription to SPSSX-L, send a message to
[hidden email]<mailto:[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]<mailto:[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

Peck, Jon

Re: creating multiple charts with chart builder

Python string substitution works much better than macro with GPL. You can find an example of a chart generated this way in the paretochart.py module downloadable from SPSS Developer Central (www.spss.com/devcentral) or in the pls.py module installed when you install the Partial Least Squares addon for SPSS 16. But these do require some Python knowledge.

As a general approach, I would define a Python function whose arguments are the parameters you want to substitute in GGRAPH/GPL, make a template of the syntax you want, and pass it the parameters. If this is all Greek to you, though, this might not be the shortest route to a solution.

HTH,
Jon Peck

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of LUCINDA M TEAR
Sent: Wednesday, August 13, 2008 6:21 PM
To: [hidden email]
Subject: Re: [SPSSX-L] creating multiple charts with chart builder

Thank you, ViAnn for your help re: outlier labels. My variable names (actually, I have 19 - more than I thought!) do have common names: chem1 to chem19. I tried using the !concat() function and creating a (define - !enddefine) macro, but that didn't work. The syntax guide for ggraph indicates that name can be used to create multiple graphs with the same syntax, but doesn't provide any examples of that....

Thanks again!

Lucinda

----- Original Message -----
From: ViAnn Beadle<mailto:[hidden email]>
To: [hidden email]<mailto:[hidden email]>
Sent: Wednesday, August 13, 2008 4:57 PM
Subject: Re: creating multiple charts with chart builder

When Chart Builder creates syntax, it knows what the variable label for the
selected variable is and uses that when generating the GUIDE statement.
Unlike Chart Builder, macros are simple substitutions of strings and have no
direct knowledge of dictionary information. I have seen some solutions in
which SPSS itself is coaxed into writing a macro with dictionary information
within it but I haven't a clue as to how to go about this.

It's fairly easy to use python to retrieve a variable label from a variable
name and place that in the middle of the label function on the GUIDE
statement. However, I don't think that you can embed a python program inside
of a macro used to generate 11 separate instances of the GGRAPH command. You
might be able to use a python program which does the looping through the
names and I'm sure that python fans on the list can come up with something
that works.

Tell us more about your sets of 11 variables. Do they have some common name
part to them such as chem1 to chem11?

On a simpler note, to suppress the outlier labels, omit the label(id)
function from your ELEMENT statement and drop the DATA statement defining
the id variable.

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
LUCINDA M TEAR
Sent: Wednesday, August 13, 2008 5:29 PM
To: [hidden email]<mailto:[hidden email]>
Subject: creating multiple charts with chart builder

Hello,

I am just venturing into Chart Builder instead of using Interactive Charts
because I want to be able to use the "logarithmic" axis transformation on
the scale axis when I am creating a set of multiple boxplots.

I have read the syntax help, but I am not sure how to create a "macro" that
will allow me create multiple boxplots, one for each of 11 variables, such
that each boxplot uses the variable LABEL name as the axis label and does
NOT include labels for outliers.

I have the following syntax - I have tried to use the "NAME=" function so
that I would have to change only the variable name after the VARIABLES =
command, but this doesn't work to change the axis label and I'm not sure how
to get it to read a list of variable names so that several boxplots are
created.

Any help would be appreciated! Thank you, Lucinda Tear

* Chart Builder.

GGRAPH

/GRAPHDATASET NAME="graphdataset" VARIABLES=chem1 [NAME="chem"]
MISSING=VARIABLEWISE

REPORTMISSING=NO

/GRAPHSPEC SOURCE=INLINE

TEMPLATE=["C:\Program Files\SPSS\Looks\air_boxes.sgt"].

BEGIN GPL

SOURCE: s=userSource(id("graphdataset"))

DATA: chem=col(source(s), name("chem")) #I replaced chem1 with "chem" here
to try to make it more generic

DATA: id=col(source(s), name("$CASENUM"), unit.category())

COORD: rect(dim(1), transpose())

GUIDE: axis(dim(1), label("WHAT CAN I PUT HERE SO THAT THE AXIS LABEL NAME
CHANGES WITH EACH VARIABLE"))

SCALE: log(dim(1), base(10))

ELEMENT: schema(position(bin.quantile.letter(chem)), label(id)) #I replaced
chem1 with "chem" here to try to make it more generic

END GPL.

=======
To manage your subscription to SPSSX-L, send a message to
[hidden email]<mailto:[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]<mailto:[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

=====================
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

LUCINDA M TEAR

Re: creating multiple charts with chart builder

Thank you, Jon! The general terms are not Greek, but I don't know Python, so I'll see if I can decipher the link you sent. Thank you.

Lucinda
----- Original Message -----
From: Peck, Jon<mailto:[hidden email]>
To: [hidden email]<mailto:[hidden email]>
Sent: Wednesday, August 13, 2008 7:27 PM
Subject: Re: creating multiple charts with chart builder

Python string substitution works much better than macro with GPL. You can find an example of a chart generated this way in the paretochart.py module downloadable from SPSS Developer Central (www.spss.com/devcentral<http://www.spss.com/devcentral>) or in the pls.py module installed when you install the Partial Least Squares addon for SPSS 16. But these do require some Python knowledge.

As a general approach, I would define a Python function whose arguments are the parameters you want to substitute in GGRAPH/GPL, make a template of the syntax you want, and pass it the parameters. If this is all Greek to you, though, this might not be the shortest route to a solution.

HTH,
Jon Peck

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of LUCINDA M TEAR
Sent: Wednesday, August 13, 2008 6:21 PM
To: [hidden email]<mailto:[hidden email]>
Subject: Re: [SPSSX-L] creating multiple charts with chart builder

Thank you, ViAnn for your help re: outlier labels. My variable names (actually, I have 19 - more than I thought!) do have common names: chem1 to chem19. I tried using the !concat() function and creating a (define - !enddefine) macro, but that didn't work. The syntax guide for ggraph indicates that name can be used to create multiple graphs with the same syntax, but doesn't provide any examples of that....

Thanks again!

Lucinda

----- Original Message -----
From: ViAnn Beadle<mailto:[hidden email]<mailto:[hidden email]>>
To: [hidden email]<mailto:[hidden email]<mailto:[hidden email]%3Cmailto:[hidden email]>>
Sent: Wednesday, August 13, 2008 4:57 PM
Subject: Re: creating multiple charts with chart builder

When Chart Builder creates syntax, it knows what the variable label for the
selected variable is and uses that when generating the GUIDE statement.
Unlike Chart Builder, macros are simple substitutions of strings and have no
direct knowledge of dictionary information. I have seen some solutions in
which SPSS itself is coaxed into writing a macro with dictionary information
within it but I haven't a clue as to how to go about this.

It's fairly easy to use python to retrieve a variable label from a variable
name and place that in the middle of the label function on the GUIDE
statement. However, I don't think that you can embed a python program inside
of a macro used to generate 11 separate instances of the GGRAPH command. You
might be able to use a python program which does the looping through the
names and I'm sure that python fans on the list can come up with something
that works.

Tell us more about your sets of 11 variables. Do they have some common name
part to them such as chem1 to chem11?

On a simpler note, to suppress the outlier labels, omit the label(id)
function from your ELEMENT statement and drop the DATA statement defining
the id variable.

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
LUCINDA M TEAR
Sent: Wednesday, August 13, 2008 5:29 PM
To: [hidden email]<mailto:[hidden email]<mailto:[hidden email]%3Cmailto:[hidden email]>>
Subject: creating multiple charts with chart builder

Hello,

I am just venturing into Chart Builder instead of using Interactive Charts
because I want to be able to use the "logarithmic" axis transformation on
the scale axis when I am creating a set of multiple boxplots.

I have read the syntax help, but I am not sure how to create a "macro" that
will allow me create multiple boxplots, one for each of 11 variables, such
that each boxplot uses the variable LABEL name as the axis label and does
NOT include labels for outliers.

I have the following syntax - I have tried to use the "NAME=" function so
that I would have to change only the variable name after the VARIABLES =
command, but this doesn't work to change the axis label and I'm not sure how
to get it to read a list of variable names so that several boxplots are
created.

Any help would be appreciated! Thank you, Lucinda Tear

* Chart Builder.

GGRAPH

/GRAPHDATASET NAME="graphdataset" VARIABLES=chem1 [NAME="chem"]
MISSING=VARIABLEWISE

REPORTMISSING=NO

/GRAPHSPEC SOURCE=INLINE

TEMPLATE=["C:\Program Files\SPSS\Looks\air_boxes.sgt"].

BEGIN GPL

SOURCE: s=userSource(id("graphdataset"))

DATA: chem=col(source(s), name("chem")) #I replaced chem1 with "chem" here
to try to make it more generic

DATA: id=col(source(s), name("$CASENUM"), unit.category())

COORD: rect(dim(1), transpose())

GUIDE: axis(dim(1), label("WHAT CAN I PUT HERE SO THAT THE AXIS LABEL NAME
CHANGES WITH EACH VARIABLE"))

SCALE: log(dim(1), base(10))

ELEMENT: schema(position(bin.quantile.letter(chem)), label(id)) #I replaced
chem1 with "chem" here to try to make it more generic

END GPL.

=======
To manage your subscription to SPSSX-L, send a message to
[hidden email]<mailto:[hidden email]<mailto:[hidden email]%3Cmailto:[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]<mailto:[hidden email]<mailto:[hidden email]%3Cmailto:[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]<mailto:[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]<mailto:[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

Peck, Jon

Re: creating multiple charts with chart builder

Here is an example of parameterizing a chart specification.

I created a Python module named chartlib containing a function named shadedscatter. You might build up a whole library of functions like this in your chartlib module.

The code for the function is below. You could save this as chartlib.py in c:/python25/lib/site-packages if you have a standard Python installation.

(This requires that you have installed the Python plugin and, if you have an oldish SPSS version, that you download spssaux.py from Developer Central, www.spss.com/devcentral).

In the function, the parameters x, y, and z are inserted into the GGRAPH and GPL syntax by referring to them as

%(x)s and similar.

The function looks up the variable labels and inserts them where needed.

To use this function, you would write something like this using the cars.sav dataset shipped with SPSS.

begin program.

import chartlib

chartlib.shadedscatter(y="mpg", x="accel", z="engine")

end program.

Note that in Python code, case and indentation matter.

HTH,

Jon Peck

(If the listserv mangles the code below, I can email it to anyone who wants it.)

--------- start of shadedscatter function definition

# graphics function that parameterizes a chart

import spss, spssaux

def shadedscatter(y, x, z):

"""Create a scatterplot of y by x with point color determined by variable z.

Use variable labels for labeling the chart"""

vardict = spssaux.VariableDict([x, y, z])

xlabel = vardict[x].VariableLabel

ylabel = vardict[y].VariableLabel

zlabel = vardict[z].VariableLabel

cmd=r"""GGRAPH

/GRAPHDATASET NAME="graphdataset" VARIABLES=%(y)s %(x)s %(z)s MISSING=LISTWISE

REPORTMISSING=NO

/GRAPHSPEC SOURCE=INLINE.

BEGIN GPL

SOURCE: s=userSource(id("graphdataset"))

DATA: x=col(source(s), name("%(x)s"))

DATA: y=col(source(s), name("%(y)s"))

DATA: z=col(source(s), name("%(z)s"))

GUIDE: axis(dim(1), label("%(xlabel)s"))

GUIDE: axis(dim(2), label("%(ylabel)s"))

GUIDE: legend(aesthetic(aesthetic.color), label("%(zlabel)s"))

ELEMENT: point(position(x*y), color(z))

END GPL.""" % locals()

spss.Submit(cmd)

________________________________

From: LUCINDA M TEAR [mailto:[hidden email]]
Sent: Wednesday, August 13, 2008 10:46 PM
To: Peck, Jon; [hidden email]
Subject: Re: Re: creating multiple charts with chart builder

Thank you, Jon! The general terms are not Greek, but I don't know Python, so I'll see if I can decipher the link you sent. Thank you.

Lucinda

----- Original Message -----

From: Peck, Jon <mailto:[hidden email]>

To: [hidden email]

Sent: Wednesday, August 13, 2008 7:27 PM

Subject: Re: creating multiple charts with chart builder

Python string substitution works much better than macro with GPL. You can find an example of a chart generated this way in the paretochart.py module downloadable from SPSS Developer Central (www.spss.com/devcentral) or in the pls.py module installed when you install the Partial Least Squares addon for SPSS 16. But these do require some Python knowledge.

As a general approach, I would define a Python function whose arguments are the parameters you want to substitute in GGRAPH/GPL, make a template of the syntax you want, and pass it the parameters. If this is all Greek to you, though, this might not be the shortest route to a solution.

HTH,
Jon Peck

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of LUCINDA M TEAR
Sent: Wednesday, August 13, 2008 6:21 PM
To: [hidden email]
Subject: Re: [SPSSX-L] creating multiple charts with chart builder

Thank you, ViAnn for your help re: outlier labels. My variable names (actually, I have 19 - more than I thought!) do have common names: chem1 to chem19. I tried using the !concat() function and creating a (define - !enddefine) macro, but that didn't work. The syntax guide for ggraph indicates that name can be used to create multiple graphs with the same syntax, but doesn't provide any examples of that....

Thanks again!

Lucinda

----- Original Message -----
From: ViAnn Beadle<mailto:[hidden email]>
To: [hidden email]<mailto:[hidden email] <mailto:[hidden email]%3cmailto:[hidden email]> >
Sent: Wednesday, August 13, 2008 4:57 PM
Subject: Re: creating multiple charts with chart builder

When Chart Builder creates syntax, it knows what the variable label for the
selected variable is and uses that when generating the GUIDE statement.
Unlike Chart Builder, macros are simple substitutions of strings and have no
direct knowledge of dictionary information. I have seen some solutions in
which SPSS itself is coaxed into writing a macro with dictionary information
within it but I haven't a clue as to how to go about this.

It's fairly easy to use python to retrieve a variable label from a variable
name and place that in the middle of the label function on the GUIDE
statement. However, I don't think that you can embed a python program inside
of a macro used to generate 11 separate instances of the GGRAPH command. You
might be able to use a python program which does the looping through the
names and I'm sure that python fans on the list can come up with something
that works.

Tell us more about your sets of 11 variables. Do they have some common name
part to them such as chem1 to chem11?

On a simpler note, to suppress the outlier labels, omit the label(id)
function from your ELEMENT statement and drop the DATA statement defining
the id variable.

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
LUCINDA M TEAR
Sent: Wednesday, August 13, 2008 5:29 PM
To: [hidden email]<mailto:[hidden email] <mailto:[hidden email]%3cmailto:[hidden email]> >
Subject: creating multiple charts with chart builder

Hello,

I am just venturing into Chart Builder instead of using Interactive Charts
because I want to be able to use the "logarithmic" axis transformation on
the scale axis when I am creating a set of multiple boxplots.

I have read the syntax help, but I am not sure how to create a "macro" that
will allow me create multiple boxplots, one for each of 11 variables, such
that each boxplot uses the variable LABEL name as the axis label and does
NOT include labels for outliers.

I have the following syntax - I have tried to use the "NAME=" function so
that I would have to change only the variable name after the VARIABLES =
command, but this doesn't work to change the axis label and I'm not sure how
to get it to read a list of variable names so that several boxplots are
created.

Any help would be appreciated! Thank you, Lucinda Tear

* Chart Builder.

GGRAPH

/GRAPHDATASET NAME="graphdataset" VARIABLES=chem1 [NAME="chem"]
MISSING=VARIABLEWISE

REPORTMISSING=NO

/GRAPHSPEC SOURCE=INLINE

TEMPLATE=["C:\Program Files\SPSS\Looks\air_boxes.sgt"].

BEGIN GPL

SOURCE: s=userSource(id("graphdataset"))

DATA: chem=col(source(s), name("chem")) #I replaced chem1 with "chem" here
to try to make it more generic

DATA: id=col(source(s), name("$CASENUM"), unit.category())

COORD: rect(dim(1), transpose())

GUIDE: axis(dim(1), label("WHAT CAN I PUT HERE SO THAT THE AXIS LABEL NAME
CHANGES WITH EACH VARIABLE"))

SCALE: log(dim(1), base(10))

ELEMENT: schema(position(bin.quantile.letter(chem)), label(id)) #I replaced
chem1 with "chem" here to try to make it more generic

END GPL.

=======
To manage your subscription to SPSSX-L, send a message to
[hidden email]<mailto:[hidden email] <mailto:[hidden email]%3cmailto:[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]<mailto:[hidden email] <mailto:[hidden email]%3cmailto:[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

=====================
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