languages for syntax and documentation

[Art Kendall]

On the Statistics Without Borders list we are starting a discussion
about developing statistical capacity in developing countries.

Does anybody know the link to find out what languages the SPSS syntax
can be in and what languages the documentation is available in?

Is there a link to prices for versions available in education in
developing countries that _allow the use of syntax_?
Soapbox: I consider it very poor pedagogy to rely solely on the GUI.

Art Kendall
Social Research Consultants

=====================
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 K Peck]

Jon Peck
SPSS, an IBM Company
[hidden email]
312-651-3435

From:	Art Kendall <[hidden email]>
To:	[hidden email]
Date:	02/03/2010 05:29 AM
Subject:	[SPSSX-L] languages for syntax and documentation
Sent by:	"SPSSX(r) Discussion" <[hidden email]>

---

On the Statistics Without Borders list we are starting a discussion
about developing statistical capacity in developing countries.

Does anybody know the link to find out what languages the SPSS syntax
can be in and what languages the documentation is available in?
>>>SPSS syntax is not translatable.  Variable names, titles, etc can be in any characters - you should use Unicode for everything, BTW, but the command names, subcommands, and keywords are always the same.  This is important in order that syntax jobs will work regardless of the L10N settings.  

Of course, the output and user interface, including help, are provided in quite a few languages besides English: currently french, italian, spanish, german, polish, russian, japanese, korean, traditional chinese, and simplified chinese.
 There is also a Python module and extension command, SPSSINC TRANSLATE OUTPUT, that can translate output into other languages according to your own specifications.

If someone wanted to make a translated version of syntax, one approach would be to define a language version of the syntax definition files in the syntax_files subdirectory.  That would make coloring and checking of syntax in the editor work with other text, but you would need code to map that back to the standard terms as it is run.  That's doable, but not a small job.

Is there a link to prices for versions available in education in
developing countries that _allow the use of syntax_?
Soapbox: I consider it very poor pedagogy to rely solely on the GUI.

>>>I certainly disagree here at least for purposes of learning statistics.  While syntax is unarguably critical for many purposes, I don't think it has much to do with learning statistics.

Regards,
Jon

Art Kendall
Social Research Consultants

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

[Bruce Weaver]

Jon K Peck wrote

--- snip ---

AK:
Is there a link to prices for versions available in education in
developing countries that _allow the use of syntax_?
Soapbox: I consider it very poor pedagogy to rely solely on the GUI.

JP:
>>>I certainly disagree here at least for purposes of learning statistics.
 While syntax is unarguably critical for many purposes, I don't think it
has much to do with learning statistics.

FWIW, I don't think using the GUI has (or should have) all that much to do with learning statistics either.  Arguably, one reason there is so much bad statistical practice out there is that GUI-based packages hand people a loaded statistical gun before they've learned how to shoot.  This is not to say we should bar students from using stats packages.  However, IMO, before we turn them loose with a stats package, we should first give them a good conceptual understanding of how a particular test or procedure works, what the underlying assumptions are, etc.

Furthermore, I suspect Art was thinking more of pedagogy around data analysis practices than around statistical knowledge per se.

There's my 2 cents.  ;-)

[John F Hall]

See red selection in Jon Peck's reply below.

 

My retort is that I taught data management and analysis with SPSS using only basic syntax.  My students said they learned more statistics from me than they did from their official statistics course.

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

From: [hidden email]

To: [hidden email]

Sent: Wednesday, February 03, 2010 2:59 PM

Subject: Re: languages for syntax and documentation

Jon Peck
SPSS, an IBM Company
[hidden email]
312-651-3435

From:	Art Kendall <[hidden email]>
To:	[hidden email]
Date:	02/03/2010 05:29 AM
Subject:	[SPSSX-L] languages for syntax and documentation
Sent by:	"SPSSX(r) Discussion" <[hidden email]>

---

On the Statistics Without Borders list we are starting a discussion
about developing statistical capacity in developing countries.

Does anybody know the link to find out what languages the SPSS syntax
can be in and what languages the documentation is available in?
>>>SPSS syntax is not translatable.  Variable names, titles, etc can be in any characters - you should use Unicode for everything, BTW, but the command names, subcommands, and keywords are always the same.  This is important in order that syntax jobs will work regardless of the L10N settings.  

Of course, the output and user interface, including help, are provided in quite a few languages besides English: currently french, italian, spanish, german, polish, russian, japanese, korean, traditional chinese, and simplified chinese.
 There is also a Python module and extension command, SPSSINC TRANSLATE OUTPUT, that can translate output into other languages according to your own specifications.

If someone wanted to make a translated version of syntax, one approach would be to define a language version of the syntax definition files in the syntax_files subdirectory.  That would make coloring and checking of syntax in the editor work with other text, but you would need code to map that back to the standard terms as it is run.  That's doable, but not a small job.

Is there a link to prices for versions available in education in
developing countries that _allow the use of syntax_?
Soapbox: I consider it very poor pedagogy to rely solely on the GUI.

>>>I certainly disagree here at least for purposes of learning statistics.  While syntax is unarguably critical for many purposes, I don't think it has much to do with learning statistics.

Regards,
Jon

Art Kendall
Social Research Consultants

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

[Marks, Jim]

Agree—if only because the subcommands highlight assumptions and issues relating to the statistical procedures.

 

For example you can specify listwise or pairwise deletion of cases in a regression procedure.

Why is listwise the default?

When is pairwise a good idea?

What does pairwise deletion do to the parameter estimates and R-sq values?

 

Plus you can run the analysis twice and see the difference each option brings

 

Jim Marks

Director, Market Research

x1616

 

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of John F Hall
Sent: Wednesday, February 03, 2010 8:45 AM
To: [hidden email]
Subject: Re: languages for syntax and documentation

 

See red selection in Jon Peck's reply below.

 

My retort is that I taught data management and analysis with SPSS using only basic syntax.  My students said they learned more statistics from me than they did from their official statistics course.

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

From: [hidden email]

To: [hidden email]

Sent: Wednesday, February 03, 2010 2:59 PM

Subject: Re: languages for syntax and documentation

 

Jon Peck
SPSS, an IBM Company
[hidden email]
312-651-3435

From:	Art Kendall <[hidden email]>
To:	[hidden email]
Date:	02/03/2010 05:29 AM
Subject:	[SPSSX-L] languages for syntax and documentation
Sent by:	"SPSSX(r) Discussion" <[hidden email]>

 

---

On the Statistics Without Borders list we are starting a discussion
about developing statistical capacity in developing countries.

Does anybody know the link to find out what languages the SPSS syntax
can be in and what languages the documentation is available in?
>>>SPSS syntax is not translatable.  Variable names, titles, etc can be in any characters - you should use Unicode for everything, BTW, but the command names, subcommands, and keywords are always the same.  This is important in order that syntax jobs will work regardless of the L10N settings.  

Of course, the output and user interface, including help, are provided in quite a few languages besides English: currently french, italian, spanish, german, polish, russian, japanese, korean, traditional chinese, and simplified chinese.
 There is also a Python module and extension command, SPSSINC TRANSLATE OUTPUT, that can translate output into other languages according to your own specifications.

If someone wanted to make a translated version of syntax, one approach would be to define a language version of the syntax definition files in the syntax_files subdirectory.  That would make coloring and checking of syntax in the editor work with other text, but you would need code to map that back to the standard terms as it is run.  That's doable, but not a small job.

Is there a link to prices for versions available in education in
developing countries that _allow the use of syntax_?
Soapbox: I consider it very poor pedagogy to rely solely on the GUI.

>>>I certainly disagree here at least for purposes of learning statistics.  While syntax is unarguably critical for many purposes, I don't think it has much to do with learning statistics.

Regards,
Jon

Art Kendall
Social Research Consultants

=====================
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 K Peck]

Apart from some options that might be available only in syntax, I don't see why having to express your choices in syntax is helpful in exploring the effects of the choices, and the gui generally exposes the choices more explicitly than syntax would, since the defaults need not be explicit in the syntax, and the gui uses more descriptive text.

My experience years back in teaching with an econometric system (of my own creation) that did not have a gui was that too much time was spent getting the syntax right at the expense of focusing on what the students were supposed to be learning about econometrics.

I'm not arguing at all that syntax isn't important at later stages in the process, but it doesn't help with concepts IMO.


Jon Peck
SPSS, an IBM Company
[hidden email]
312-651-3435

From:	"Marks, Jim" <[hidden email]>
To:	[hidden email]
Date:	02/03/2010 08:11 AM
Subject:	Re: [SPSSX-L] languages for syntax and documentation
Sent by:	"SPSSX(r) Discussion" <[hidden email]>

---

Agree—if only because the subcommands highlight assumptions and issues relating to the statistical procedures.
 
For example you can specify listwise or pairwise deletion of cases in a regression procedure.
Why is listwise the default?
When is pairwise a good idea?
What does pairwise deletion do to the parameter estimates and R-sq values?
 
Plus you can run the analysis twice and see the difference each option brings
 
Jim Marks
Director, Market Research
x1616
 
From: SPSSX(r) Discussion [[hidden email]] On Behalf Of John F Hall
Sent: Wednesday, February 03, 2010 8:45 AM
To: [hidden email]
Subject: Re: languages for syntax and documentation
 
See red selection in Jon Peck's reply below.
 
My retort is that I taught data management and analysis with SPSS using only basic syntax.  My students said they learned more statistics from me than they did from their official statistics course.
----- Original Message -----
From: Jon K Peck
To: [hidden email]
Sent: Wednesday, February 03, 2010 2:59 PM
Subject: Re: languages for syntax and documentation
 

Jon Peck
SPSS, an IBM Company
peck@...
312-651-3435

From:	Art Kendall <Art@...>
To:	[hidden email]
Date:	02/03/2010 05:29 AM
Subject:	[SPSSX-L] languages for syntax and documentation
Sent by:	"SPSSX(r) Discussion" <[hidden email]>

 

---

On the Statistics Without Borders list we are starting a discussion
about developing statistical capacity in developing countries.

Does anybody know the link to find out what languages the SPSS syntax
can be in and what languages the documentation is available in?
>>>SPSS syntax is not translatable.  Variable names, titles, etc can be in any characters - you should use Unicode for everything, BTW, but the command names, subcommands, and keywords are always the same.  This is important in order that syntax jobs will work regardless of the L10N settings.  

Of course, the output and user interface, including help, are provided in quite a few languages besides English: currently french, italian, spanish, german, polish, russian, japanese, korean, traditional chinese, and simplified chinese.
There is also a Python module and extension command, SPSSINC TRANSLATE OUTPUT, that can translate output into other languages according to your own specifications.

If someone wanted to make a translated version of syntax, one approach would be to define a language version of the syntax definition files in the syntax_files subdirectory.  That would make coloring and checking of syntax in the editor work with other text, but you would need code to map that back to the standard terms as it is run.  That's doable, but not a small job.

Is there a link to prices for versions available in education in
developing countries that _allow the use of syntax_?
Soapbox: I consider it very poor pedagogy to rely solely on the GUI.

>>>I certainly disagree here at least for purposes of learning statistics.  While syntax is unarguably critical for many purposes, I don't think it has much to do with learning statistics.

Regards,
Jon

Art Kendall
Social Research Consultants

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

[Art Kendall]

One thing I try to get across at an introductory level is that it is very easy to shoot yourself in the foot by failing to use a team approach to an analysis.

Yes, I was thinking much more of statistics in the context of a complete analysis.

Art

On 2/3/2010 9:33 AM, Bruce Weaver wrote:

Jon K Peck wrote:
  

--- snip ---

AK:
Is there a link to prices for versions available in education in
developing countries that _allow the use of syntax_?
Soapbox: I consider it very poor pedagogy to rely solely on the GUI.

JP:
    

I certainly disagree here at least for purposes of learning statistics.
          

 While syntax is unarguably critical for many purposes, I don't think it
has much to do with learning statistics.


    

FWIW, I don't think using the GUI has (or should have) all that much to do
with learning statistics either.  Arguably, one reason there is so much bad
statistical practice out there is that GUI-based packages hand people a
loaded statistical gun before they've learned how to shoot.  This is not to
say we should bar students from using stats packages.  However, IMO, before
we turn them loose with a stats package, we should first give them a good
conceptual understanding of how a particular test or procedure works, what
the underlying assumptions are, etc.

Furthermore, I suspect Art was thinking more of pedagogy around data
analysis practices than around statistical knowledge per se.

There's my 2 cents.  ;-)



-----
--
Bruce Weaver
[hidden email]
http://sites.google.com/a/lakeheadu.ca/bweaver/
"When all else fails, RTFM."

NOTE:  My Hotmail account is not monitored regularly.
To send me an e-mail, please use the address shown above.
--
View this message in context: http://old.nabble.com/languages-for-syntax-and-documentation-tp27435489p27437219.html
Sent from the SPSSX Discussion mailing list archive at 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

[Art Kendall]

Comments interspersed:
Some context from the other list. (where I am coming from)

We should consider the goals of the whole project. Perhaps we should focus on helping them develop a base of good users of these techniques. Perhaps later, when there is an established user base, they should at that time consider developing a cadre of their own people with advanced graduate degrees.   I usually take the approach that the goals for users of stat/methods include:
-- to give them basic fluency in a package with strong human factors; 
-- help them gain sufficient background so they can work efficiently with consultants;
-- emphasize the importance of good consultation at the design stages, i.e., not shooting from the hip; and
-- develop good quality assurance habits in dealing with data (e.g. fully label their data, incorporate peer review aka referencing, and understand that analysis goes through several drafts so doing things in such a way that the audit trail allows them to go back and refine their approach,  etc.).

In my experience, it would cost some upfront money, but save them a tons of people time (i.e., money) over time to include SPSS.  It comes with very extensive tutorials, "coaches", and documentation.  There is also very strong free support available from the user community.  If what they want to do cannot be done with SPSS, it is likely that they are not doing something that is mainstream (basic stuff).  It would cost some upfront money, but save them a lot of people time (i.e., money) to include SPSS.  It comes with very extensive tutorials, coaches and documentation.  There is also very strong support available from the user community. The availability of clear syntax helps with the review process, posting answerable questions on discussion lists, and going back and doing it better.

I forgot to include mention of my experience that  getting the data ready for stat analysis often exceeds 90% of the staff time for this phase of a project.  This is one area SPSS really shines.

I also forgot to mention that is important to familiarize them with the many sources under <help>.

On 2/3/2010 10:21 AM, Jon K Peck wrote:

Apart from some options that might be available only in syntax, I don't see why having to express your choices in syntax is helpful in exploring the effects of the choices, and the gui generally exposes the choices more explicitly than syntax would, since the defaults need not be explicit in the syntax, and the gui uses more descriptive text.

I do not have a problem with using the GUI.  I believe it it a great way to get a first draft of syntax and to see what some of the options are. I have a problem with using it exclusively. It would require a prodigious short term memory to say this block of output was by checking box a, and this block of output was from checking box b. looking at the two snippets of syntax generated does not require as much cognitive effort.

I also have found that having students act as QA partners/reviewers for each other helps them develop a more flexible understanding.  I believe that having the syntax available helps the review process.

However, I also strongly believe that analysis is an iterative process.  As people progress in their understanding of analysis, stats, and the data at hand, they will inevitably be helped if they can go back and refine the stream of syntax.  The availability of the syntax is also helpful in getting help.  Can you imagine having to deal a post asking for help with "I clicked ... then ... then ..." and received this message '...' "?

Even after doing analyses since 1970, I still go back and refine my approach and clean up the presentation syntax  after the first draft.  This is even so with the small sets of syntax I post on lists.

My experience years back in teaching with an econometric system (of my own creation) that did not have a gui was that too much time was spent getting the syntax right at the expense of focusing on what the students were supposed to be learning about econometrics.

I'm not arguing at all that syntax isn't important at later stages in the process, but it doesn't help with concepts IMO.

On the other hand I think that it can give an additional perspective on underlying processes.

I find that small simulations are helpful in clarifying points.
*THROW 2 DICE.
SET SEED=20090515.
NEW FILE.
INPUT PROGRAM  .
LOOP id=1 to 10000.
COMPUTE die1=RND(RV.UNI(.5,6.5)).
COMPUTE die2=RND(RV.UNI(.5,6.5)).
END CASE.
END LOOP.
END FILE.
END INPUT PROGRAM.
FORMATS DIE1 DIE2 (F1).
CROSSTAB TABLES= DIE1 BY DIE2 /CELLS=ALL.
*.

* finding the number of combinations when presenting the shorthand counting formula.
new file.
* all combinations of 3 variables.
input program.
*generate indices.
numeric id(f4).
numeric index1 index2 index3 (f1).
leave id index1 index2.
*change the "3" on the next 3 lines to see what happens.
loop index1 = 1 to 3.
loop index2 = 1 to 3.
loop index3 = 1 to 3.
compute id = id +1.
end case.
end loop.
end loop.
end loop.
end file.
end input program.

Jon Peck
SPSS, an IBM Company
[hidden email]
312-651-3435

From:	"Marks, Jim" [hidden email]
To:	[hidden email]
Date:	02/03/2010 08:11 AM
Subject:	Re: [SPSSX-L] languages for syntax and documentation
Sent by:	"SPSSX(r) Discussion" [hidden email]

---

Agree—if only because the subcommands highlight assumptions and issues relating to the statistical procedures.
 
For example you can specify listwise or pairwise deletion of cases in a regression procedure.
Why is listwise the default?
When is pairwise a good idea?
What does pairwise deletion do to the parameter estimates and R-sq values?
 
Plus you can run the analysis twice and see the difference each option brings
 
Jim Marks
Director, Market Research
x1616
 
From: SPSSX(r) Discussion [[hidden email]] On Behalf Of John F Hall
Sent: Wednesday, February 03, 2010 8:45 AM
To: [hidden email]
Subject: Re: languages for syntax and documentation
 
See red selection in Jon Peck's reply below.
 
My retort is that I taught data management and analysis with SPSS using only basic syntax.  My students said they learned more statistics from me than they did from their official statistics course.
----- Original Message -----
From: [hidden email]
To: [hidden email]
Sent: Wednesday, February 03, 2010 2:59 PM
Subject: Re: languages for syntax and documentation
 

Jon Peck
SPSS, an IBM Company
[hidden email]
312-651-3435

From:	Art Kendall <[hidden email]>
To:	[hidden email]
Date:	02/03/2010 05:29 AM
Subject:	[SPSSX-L] languages for syntax and documentation
Sent by:	"SPSSX(r) Discussion" <[hidden email]>

 

---

On the Statistics Without Borders list we are starting a discussion
about developing statistical capacity in developing countries.

Does anybody know the link to find out what languages the SPSS syntax
can be in and what languages the documentation is available in?
>>>SPSS syntax is not translatable.  Variable names, titles, etc can be in any characters - you should use Unicode for everything, BTW, but the command names, subcommands, and keywords are always the same.  This is important in order that syntax jobs will work regardless of the L10N settings.  

Of course, the output and user interface, including help, are provided in quite a few languages besides English: currently french, italian, spanish, german, polish, russian, japanese, korean, traditional chinese, and simplified chinese.
There is also a Python module and extension command, SPSSINC TRANSLATE OUTPUT, that can translate output into other languages according to your own specifications.

If someone wanted to make a translated version of syntax, one approach would be to define a language version of the syntax definition files in the syntax_files subdirectory.  That would make coloring and checking of syntax in the editor work with other text, but you would need code to map that back to the standard terms as it is run.  That's doable, but not a small job.

Is there a link to prices for versions available in education in
developing countries that _allow the use of syntax_?
Soapbox: I consider it very poor pedagogy to rely solely on the GUI.

>>>I certainly disagree here at least for purposes of learning statistics.  While syntax is unarguably critical for many purposes, I don't think it has much to do with learning statistics.

Regards,
Jon

Art Kendall
Social Research Consultants

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

[John F Hall]



For a start, syntax is much quicker and easier to use, especially if that's how you've learned it.  I can only speak for analysing data from questiionnaire surveys or similar, but when I was teaching SPSS-X 8 on a Vax cluster with VMS and EDT  (two more languages to learn in addition to SPSS!) the user-friendly on-line interface by Jim Ring was invaluable (error detection, correction and return to last place in SPSS)  The availability of a couple of undergraduate students as demonstrators, who'd done the course the previous year and knew their way around SPSS and the computer system, made everything run smoothly and we got through the syllabus much more quickly and covered additional ground.  It providied a good platform for those needing to go to more advanced statistics.  Three students who did my  course as undergraduates are now full Professors themselves (another one is a Senior Lecturer) and head up departments and/or graduate programs running quantitative courses and turning out Masters and Doctors.

 

A compromise for addictive drop-downers in blissful ignorance of syntax is to follow the advice in Jacquie Collier's new book to save all the syntax generated from the GUI and then use it to wean themselves away from GUI to syntax proper.  My advice is stick to 15, ignore the add-ons and trim all the commands and subcommands in syntax files to an absolute minimum and accept the SPSS defaults.  The GUI automatically generates the full set of syntax in the Output Viewer.  The only advantage of GUI is it generates correct syntax (far too much of it) for people who can't spell or read a manual, many of whom don't know what they're doing anyway (nor in my experience do many of their supervisors).  The only thing I use it for is to open a new syntax file, edit and run all or part of it, and save that and *.sav files when needed.  Inside GUI you can't see what you're doing and it takes forever.  If you don't believe me have a look at my comparisons of syntax versu menus on http://surveyresearch.weebly.com/7-old-dog-old-tricks.html sections 4 and 5 (replicating examples from Julie Pallant's SPSS Survival Manual) and http://surveyresearch.weebly.com/uploads/2/9/9/8/2998485/5.2.4_the_compute_command_1_-_attachment_to_status_quo.doc (my first tutorial on COMPUTE) which uses first syntax, then menus to perform the same simple task.  In the second COMPUTE tutorial I use syntax only and end by saying:

 

The point-and click way

 

If you prefer, you can produce all the above tables, statistics and histogram using the drop-down menus, but it will take you a very long time.  I'm not even going to bother.  You have been warned!

 

Don't get me wrong.  I remember when a single run could take two days to come back by courier.  Mistakes could be expensive on researcher time, so we didn't make too many.  Now that we all have PCs, its sheer speed means forests no longer disappear every time we make a mistake: just delete the output file without saving it, go back to the syntax file, correct it and run it again, and again, and again.   Makes for lazy computing, but it's quick.

 

I just ♥ SPSS: please don't muck about with it.

 

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

From: [hidden email]

To: [hidden email]

Sent: Wednesday, February 03, 2010 4:21 PM

Subject: Re: languages for syntax and documentation

Apart from some options that might be available only in syntax, I don't see why having to express your choices in syntax is helpful in exploring the effects of the choices, and the gui generally exposes the choices more explicitly than syntax would, since the defaults need not be explicit in the syntax, and the gui uses more descriptive text.

My experience years back in teaching with an econometric system (of my own creation) that did not have a gui was that too much time was spent getting the syntax right at the expense of focusing on what the students were supposed to be learning about econometrics.

I'm not arguing at all that syntax isn't important at later stages in the process, but it doesn't help with concepts IMO.


Jon Peck
SPSS, an IBM Company
[hidden email]
312-651-3435

From:	"Marks, Jim" <[hidden email]>
To:	[hidden email]
Date:	02/03/2010 08:11 AM
Subject:	Re: [SPSSX-L] languages for syntax and documentation
Sent by:	"SPSSX(r) Discussion" <[hidden email]>

---

Agree—if only because the subcommands highlight assumptions and issues relating to the statistical procedures.
 
For example you can specify listwise or pairwise deletion of cases in a regression procedure.
Why is listwise the default?
When is pairwise a good idea?
What does pairwise deletion do to the parameter estimates and R-sq values?
 
Plus you can run the analysis twice and see the difference each option brings
 
Jim Marks
Director, Market Research
x1616
 
From: SPSSX(r) Discussion [[hidden email]] On Behalf Of John F Hall
Sent: Wednesday, February 03, 2010 8:45 AM
To: [hidden email]
Subject: Re: languages for syntax and documentation
 
See red selection in Jon Peck's reply below.
 
My retort is that I taught data management and analysis with SPSS using only basic syntax.  My students said they learned more statistics from me than they did from their official statistics course.
----- Original Message -----
From: [hidden email]
To: [hidden email]
Sent: Wednesday, February 03, 2010 2:59 PM
Subject: Re: languages for syntax and documentation
 

Jon Peck
SPSS, an IBM Company
[hidden email]
312-651-3435

From:	Art Kendall <[hidden email]>
To:	[hidden email]
Date:	02/03/2010 05:29 AM
Subject:	[SPSSX-L] languages for syntax and documentation
Sent by:	"SPSSX(r) Discussion" <[hidden email]>

 

---

On the Statistics Without Borders list we are starting a discussion
about developing statistical capacity in developing countries.

Does anybody know the link to find out what languages the SPSS syntax
can be in and what languages the documentation is available in?
>>>SPSS syntax is not translatable.  Variable names, titles, etc can be in any characters - you should use Unicode for everything, BTW, but the command names, subcommands, and keywords are always the same.  This is important in order that syntax jobs will work regardless of the L10N settings.  

Of course, the output and user interface, including help, are provided in quite a few languages besides English: currently french, italian, spanish, german, polish, russian, japanese, korean, traditional chinese, and simplified chinese.
There is also a Python module and extension command, SPSSINC TRANSLATE OUTPUT, that can translate output into other languages according to your own specifications.

If someone wanted to make a translated version of syntax, one approach would be to define a language version of the syntax definition files in the syntax_files subdirectory.  That would make coloring and checking of syntax in the editor work with other text, but you would need code to map that back to the standard terms as it is run.  That's doable, but not a small job.

Is there a link to prices for versions available in education in
developing countries that _allow the use of syntax_?
Soapbox: I consider it very poor pedagogy to rely solely on the GUI.

>>>I certainly disagree here at least for purposes of learning statistics.  While syntax is unarguably critical for many purposes, I don't think it has much to do with learning statistics.

Regards,
Jon

Art Kendall
Social Research Consultants

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

[Ergul, Emel A.]

Dear all:

I need to select 7 specific types of cases
(type1, type2, type3, type4, type5, type6, type7) across the 6 variables
(v1,v2,v3,v4,v5,v6).

Would you be kind enough to direct me to a sample syntax I can look into?

Many thanks!

Emel


The information in this e-mail is intended only for the person to whom it is
addressed. If you believe this e-mail was sent to you in error and the e-mail
contains patient information, please contact the Partners Compliance HelpLine at
http://www.partners.org/complianceline . If the e-mail was sent to you in error
but does not contain patient information, please contact the sender and properly
dispose of the e-mail.

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

[Maguin, Eugene]

Emel,

I think you need to give us some more information so that you can get better
replies.

You're saying that each case has six variable, v1 to v6. Are type1 to type7
possible values of v1 to v6 or are type1 to type7 also variables or does
type1, for instance, refer to a specific combination of values for v1 to v6?


If type1 to type7 are variables, then what values of type1 to type7 are to
be selected?

If type1 to type7 are each separate and specific combinations of values for
v1 to v6, then what is(are) the specific combination(s) of values that
define, for instance, type1?

Gene Maguin


>>I need to select 7 specific types of cases
(type1, type2, type3, type4, type5, type6, type7) across the 6 variables
(v1,v2,v3,v4,v5,v6).

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

[Ergul, Emel A.]

Hi Gene:

hope below is more clear:

Basically I'm trying to pull the specific cases match with my specific criteria
(creating a subset?). I have more than a hundred variables and about 2 thousand
cases. I want to say  "go look into these specific variables (6 types of
incision technique) and give me those 7 specific incision tools used
(toola,toolb,toolc,toold,toole, toolf, toolg) CODED NUMERIC; filter the rest or
create another set. The data at the end has to include only these 7 type of
incision tool used in 6 incision technique along with all the other variables
for each case (when diagnosis happened, medication used, individuals height and
weight etc).

There is no tool combination in any of the technique but the same type of tool
might be used with same patient in different time frame.

I use SPSS mostly through the menus not syntax(I might be annoying many of you
by saying that).

Many thanks for the help!!

Emel



-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Gene
Maguin
Sent: Thursday, February 04, 2010 11:28 AM
To: [hidden email]
Subject: Re: Syntax for case seletion

Emel,

I think you need to give us some more information so that you can get better
replies.

You're saying that each case has six variable, v1 to v6. Are type1 to type7
possible values of v1 to v6 or are type1 to type7 also variables or does
type1, for instance, refer to a specific combination of values for v1 to v6?


If type1 to type7 are variables, then what values of type1 to type7 are to
be selected?

If type1 to type7 are each separate and specific combinations of values for
v1 to v6, then what is(are) the specific combination(s) of values that
define, for instance, type1?

Gene Maguin


>>I need to select 7 specific types of cases
(type1, type2, type3, type4, type5, type6, type7) across the 6 variables
(v1,v2,v3,v4,v5,v6).

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


The information in this e-mail is intended only for the person to whom it is
addressed. If you believe this e-mail was sent to you in error and the e-mail
contains patient information, please contact the Partners Compliance HelpLine at
http://www.partners.org/complianceline . If the e-mail was sent to you in error
but does not contain patient information, please contact the sender and properly
dispose of the e-mail.

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

[Mike]

If I may butt in here:

(1)  You want to select subsets of subjects on the basis of one or
more variables.  What are the names of these variables?  Is
it the case that the variables are:

toola,toolb,toolc,toold,toole, toolf, toolg

The key point is one needs the specific names of the variable(s) that
you want to use as the basis of selecting subjects.

(2)  What are the values of the variables that you want to use to
select subjects?  Do you, say, zero-one coding to indicate absence(=0)
or presence(=1)? Consider the following syntax.

temp.     <-- Temporarily executes the following until the first procedure is done.
select if (toola eq 0).  <--- select only those subjects who have toola=0.
desc var=var1 to var10. <--- provide descriptive stats for selected subjects for var1 to var10.

After desc procedure is done, the temporary selection is turned off and
you have all of the subjects to work with again.

If you wanted to use multiple variables for selecting subjects, consider:

temp.
select if (toola eq 0 and toolb eq 0 and toolc eq 1).
desc var=var1 to var10.

The "select if ( )" allows you to identify the variables and their values
that you want to use for selecting subsets of subjects.  I use the
descriptives procedure but any procedure can be used in its place.

More info can be gotten from the help menu and syntax icon (the "/_" thingie).

Please let us know whether or not this is the issue you are trying
to deal with.

-Mike Palij
New York University
[hidden email]



----- Original Message -----
From: "Ergul, Emel A." <[hidden email]>
To: <[hidden email]>
Sent: Thursday, February 04, 2010 1:08 PM
Subject: Re: Syntax for case seletion-clerification

> Hi Gene:
>
> hope below is more clear:
>
> Basically I'm trying to pull the specific cases match with my specific criteria
> (creating a subset?). I have more than a hundred variables and about 2 thousand
> cases. I want to say  "go look into these specific variables (6 types of
> incision technique) and give me those 7 specific incision tools used
> (toola,toolb,toolc,toold,toole, toolf, toolg) CODED NUMERIC; filter the rest or
> create another set. The data at the end has to include only these 7 type of
> incision tool used in 6 incision technique along with all the other variables
> for each case (when diagnosis happened, medication used, individuals height and
> weight etc).
>
> There is no tool combination in any of the technique but the same type of tool
> might be used with same patient in different time frame.
>
> I use SPSS mostly through the menus not syntax(I might be annoying many of you
> by saying that).
>
> Many thanks for the help!!
>
> Emel
>
>
>
> -----Original Message-----
> From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Gene
> Maguin
> Sent: Thursday, February 04, 2010 11:28 AM
> To: [hidden email]
> Subject: Re: Syntax for case seletion
>
> Emel,
>
> I think you need to give us some more information so that you can get better
> replies.
>
> You're saying that each case has six variable, v1 to v6. Are type1 to type7
> possible values of v1 to v6 or are type1 to type7 also variables or does
> type1, for instance, refer to a specific combination of values for v1 to v6?
>
>
> If type1 to type7 are variables, then what values of type1 to type7 are to
> be selected?
>
> If type1 to type7 are each separate and specific combinations of values for
> v1 to v6, then what is(are) the specific combination(s) of values that
> define, for instance, type1?
>
> Gene Maguin
>
>
>>>I need to select 7 specific types of cases
> (type1, type2, type3, type4, type5, type6, type7) across the 6 variables
> (v1,v2,v3,v4,v5,v6).
>
> =====================
> 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
>
>
> The information in this e-mail is intended only for the person to whom it is
> addressed. If you believe this e-mail was sent to you in error and the e-mail
> contains patient information, please contact the Partners Compliance HelpLine at
> http://www.partners.org/complianceline . If the e-mail was sent to you in error
> but does not contain patient information, please contact the sender and properly
> dispose of the e-mail.
>
> =====================
> 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

[Maguin, Eugene]

Emel,

I'd hope that people who use the menus wouldn't annoy others. I 'always'
work in syntax and while you might be able to do this through the menus, I
simply don't know.

Anyway. First, read up on the Any function. Based on how I understand what
you've said, I'd start by doing this. I don't normally use Any and I'm not
familiar with it. Before you run the command you'll replace
toola,toolb,toolc,toold,toole,toolf,toolg with the value that each of them
refer to. Finally, pick=1 is your subset of cases.

Compute pick=0.
Do repeat x=toola,toolb,toolc,toold,toole,toolf,toolg.
+  if (any(x,v1 to v6)) pick=1.
End repeat.


Gene Maguin


>>Basically I'm trying to pull the specific cases match with my specific
criteria
(creating a subset?). I have more than a hundred variables and about 2
thousand
cases. I want to say  "go look into these specific variables (6 types of
incision technique) and give me those 7 specific incision tools used
(toola,toolb,toolc,toold,toole, toolf, toolg) CODED NUMERIC; filter the rest
or
create another set. The data at the end has to include only these 7 type of
incision tool used in 6 incision technique along with all the other
variables
for each case (when diagnosis happened, medication used, individuals height
and
weight etc).

There is no tool combination in any of the technique but the same type of
tool
might be used with same patient in different time frame.

I use SPSS mostly through the menus not syntax(I might be annoying many of
you
by saying that).

Many thanks for the help!!

Emel



-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Gene
Maguin
Sent: Thursday, February 04, 2010 11:28 AM
To: [hidden email]
Subject: Re: Syntax for case seletion

Emel,

I think you need to give us some more information so that you can get better
replies.

You're saying that each case has six variable, v1 to v6. Are type1 to type7
possible values of v1 to v6 or are type1 to type7 also variables or does
type1, for instance, refer to a specific combination of values for v1 to v6?


If type1 to type7 are variables, then what values of type1 to type7 are to
be selected?

If type1 to type7 are each separate and specific combinations of values for
v1 to v6, then what is(are) the specific combination(s) of values that
define, for instance, type1?

Gene Maguin


>>I need to select 7 specific types of cases
(type1, type2, type3, type4, type5, type6, type7) across the 6 variables
(v1,v2,v3,v4,v5,v6).

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


The information in this e-mail is intended only for the person to whom it is
addressed. If you believe this e-mail was sent to you in error and the
e-mail
contains patient information, please contact the Partners Compliance
HelpLine at
http://www.partners.org/complianceline . If the e-mail was sent to you in
error
but does not contain patient information, please contact the sender and
properly
dispose of the e-mail.

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

[Ergul, Emel A.]

Mike and Gene,

Here is a brief graphic desc. Incisions done with different time intervals. For
each incision,

        Incision1st Incition2nd --incision6th  diagnosis dm  age

Id1   12423     12514    --             78093      CA   NO      34
Id2     12514           78093  --       12423   LD      YES     42
Id3
Id4
--
--
Id2000
***
12423: incision tool silver
12514: incision tool polymer
78093: incision tool cupper
(many more types here)
***
Manually I've tried to get rid off certain tools in series by using
Select cases-->if incision1st <12423 but you can selet one incision per trial.

Let me experiment with both syntax and see.

Thanks so very much!!

emel








-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Mike
Palij
Sent: Thursday, February 04, 2010 1:32 PM
To: [hidden email]
Subject: Re: Syntax for case seletion-clerification

If I may butt in here:

(1)  You want to select subsets of subjects on the basis of one or
more variables.  What are the names of these variables?  Is
it the case that the variables are:

toola,toolb,toolc,toold,toole, toolf, toolg

The key point is one needs the specific names of the variable(s) that
you want to use as the basis of selecting subjects.

(2)  What are the values of the variables that you want to use to
select subjects?  Do you, say, zero-one coding to indicate absence(=0)
or presence(=1)? Consider the following syntax.

temp.     <-- Temporarily executes the following until the first procedure is
done.
select if (toola eq 0).  <--- select only those subjects who have toola=0.
desc var=var1 to var10. <--- provide descriptive stats for selected subjects for
var1 to var10.

After desc procedure is done, the temporary selection is turned off and
you have all of the subjects to work with again.

If you wanted to use multiple variables for selecting subjects, consider:

temp.
select if (toola eq 0 and toolb eq 0 and toolc eq 1).
desc var=var1 to var10.

The "select if ( )" allows you to identify the variables and their values
that you want to use for selecting subsets of subjects.  I use the
descriptives procedure but any procedure can be used in its place.

More info can be gotten from the help menu and syntax icon (the "/_" thingie).

Please let us know whether or not this is the issue you are trying
to deal with.

-Mike Palij
New York University
[hidden email]



The information in this e-mail is intended only for the person to whom it is
addressed. If you believe this e-mail was sent to you in error and the e-mail
contains patient information, please contact the Partners Compliance HelpLine at
http://www.partners.org/complianceline . If the e-mail was sent to you in error
but does not contain patient information, please contact the sender and properly
dispose of the e-mail.

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

[John F Hall]

Without seeing your data, but assuming you want cases selected for each of 6 incision types combined with 7 types of tool:

 

type coded 1-6, tool coded 1-7, try

 

count newvar = type (1 thru 6).

count newvar2  = tool (1 thru 7).

select if newvar1 ge 1 and newvar2 ge 1).

 

 

Just tested two variations of the first part of the syntax on one of my own data sets

 

count x = v248 to v255 (1 thru 5) v256 to v261 (1 thru 5).

freq x.

count y = v248 to v255 v256 to v261 (1 thru 5).

freq y.

 

and it seems to work as both produce the same result, so your select if should work as well.

 

		Frequency	Percent	Valid Percent	Cumulative Percent
Valid	0	6	4.2	4.2	4.2
	3	1	.7	.7	4.9
	7	1	.7	.7	5.6
	9	1	.7	.7	6.3
	10	1	.7	.7	7.0
	11	4	2.8	2.8	9.9
	12	9	6.3	6.3	16.2
	13	14	9.9	9.9	26.1
	14	105	73.9	73.9	100.0
	Total	142	100.0	100.0

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

From: [hidden email]

To: [hidden email]

Sent: Thursday, February 04, 2010 7:08 PM

Subject: Re: Syntax for case seletion-clerification

Hi Gene:

hope below is more clear:

Basically I'm trying to pull the specific cases match with my specific criteria
(creating a subset?). I have more than a hundred variables and about 2 thousand
cases. I want to say  "go look into these specific variables (6 types of
incision technique) and give me those 7 specific incision tools used
(toola,toolb,toolc,toold,toole, toolf, toolg) CODED NUMERIC; filter the rest or
create another set. The data at the end has to include only these 7 type of
incision tool used in 6 incision technique along with all the other variables
for each case (when diagnosis happened, medication used, individuals height and
weight etc).

There is no tool combination in any of the technique but the same type of tool
might be used with same patient in different time frame.

I use SPSS mostly through the menus not syntax(I might be annoying many of you
by saying that).

Many thanks for the help!!

Emel



-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Gene
Maguin
Sent: Thursday, February 04, 2010 11:28 AM
To: [hidden email]
Subject: Re: Syntax for case seletion

Emel,

I think you need to give us some more information so that you can get better
replies.

You're saying that each case has six variable, v1 to v6. Are type1 to type7
possible values of v1 to v6 or are type1 to type7 also variables or does
type1, for instance, refer to a specific combination of values for v1 to v6?


If type1 to type7 are variables, then what values of type1 to type7 are to
be selected?

If type1 to type7 are each separate and specific combinations of values for
v1 to v6, then what is(are) the specific combination(s) of values that
define, for instance, type1?

Gene Maguin


>>I need to select 7 specific types of cases
(type1, type2, type3, type4, type5, type6, type7) across the 6 variables
(v1,v2,v3,v4,v5,v6).

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


The information in this e-mail is intended only for the person to whom it is
addressed. If you believe this e-mail was sent to you in error and the e-mail
contains patient information, please contact the Partners Compliance HelpLine at
http://www.partners.org/complianceline . If the e-mail was sent to you in error
but does not contain patient information, please contact the sender and properly
dispose of the e-mail.

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

[John F Hall]

See my memo just sent and substitute your own varnames and values in the syntax.

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

From: [hidden email]

To: [hidden email]

Sent: Thursday, February 04, 2010 8:21 PM

Subject: Re: Syntax for case seletion-clerification

Mike and Gene,

Here is a brief graphic desc. Incisions done with different time intervals. For
each incision,

        Incision1st Incition2nd --incision6th  diagnosis dm  age

Id1   12423     12514    --             78093      CA   NO      34
Id2     12514           78093  --       12423   LD      YES     42
Id3
Id4
--
--
Id2000
***
12423: incision tool silver
12514: incision tool polymer
78093: incision tool cupper
(many more types here)
***
Manually I've tried to get rid off certain tools in series by using
Select cases-->if incision1st <12423 but you can selet one incision per trial.

Let me experiment with both syntax and see.

Thanks so very much!!

emel








-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Mike
Palij
Sent: Thursday, February 04, 2010 1:32 PM
To: [hidden email]
Subject: Re: Syntax for case seletion-clerification

If I may butt in here:

(1)  You want to select subsets of subjects on the basis of one or
more variables.  What are the names of these variables?  Is
it the case that the variables are:

toola,toolb,toolc,toold,toole, toolf, toolg

The key point is one needs the specific names of the variable(s) that
you want to use as the basis of selecting subjects.

(2)  What are the values of the variables that you want to use to
select subjects?  Do you, say, zero-one coding to indicate absence(=0)
or presence(=1)? Consider the following syntax.

temp.     <-- Temporarily executes the following until the first procedure is
done.
select if (toola eq 0).  <--- select only those subjects who have toola=0.
desc var=var1 to var10. <--- provide descriptive stats for selected subjects for
var1 to var10.

After desc procedure is done, the temporary selection is turned off and
you have all of the subjects to work with again.

If you wanted to use multiple variables for selecting subjects, consider:

temp.
select if (toola eq 0 and toolb eq 0 and toolc eq 1).
desc var=var1 to var10.

The "select if ( )" allows you to identify the variables and their values
that you want to use for selecting subsets of subjects.  I use the
descriptives procedure but any procedure can be used in its place.

More info can be gotten from the help menu and syntax icon (the "/_" thingie).

Please let us know whether or not this is the issue you are trying
to deal with.

-Mike Palij
New York University
[hidden email]



The information in this e-mail is intended only for the person to whom it is
addressed. If you believe this e-mail was sent to you in error and the e-mail
contains patient information, please contact the Partners Compliance HelpLine at
http://www.partners.org/complianceline . If the e-mail was sent to you in error
but does not contain patient information, please contact the sender and properly
dispose of the e-mail.

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

[Richard Ristow]

At 01:42 PM 2/4/2010, Gene Maguin wrote:

First, read up on the Any function. Before you run the command you'll replace toola,toolb,toolc,toold,toole,toolf,toolg with the value that each of them refer to. Finally, pick=1 is your subset of cases.

Compute pick=0.
Do repeat x=toola,toolb,toolc,toold,toole,toolf,toolg.
+  if (any(x,v1 to v6)) pick=1.
End repeat.

Right, though you'd probably have "ToolType=x" rather than "pick=1". And although the poster states that no more than one tool type can occur in any record, I think a check for more than one type occurring would be imperative.
===================== 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

[Maguin, Eugene]

Hi Richard,

Did you see Emel's later (Thursday, 2:24pm) posting (tidied up below). It
seemed possible that multiple, focal tool codes could be used in one
particular case. Thus the pick=1 statement.

However, after reflecting on your message, I wonder if I don't have an error
in my code. Specifically, suppose that when the Any function is evaluated
for x='toola', the result is true and pick is set to 1. Now, and on the same
record, suppose that when Any function is evaluated for x='toolb', the
result is false (i.e., 'toolb' is not a value of v1 to v6 for that record).
Is the current value of pick, which is 1.0, set to 0.0 or left unchanged. If
pick is reset, the code I offered is in error.

Gene Maguin


>>Here is a brief graphic desc. Incisions done with different time
intervals. For
each incision,

      Incision1st Incition2nd -- Incision6th  diagnosis dm   age

Id1   12423       12514       -- 78093        CA        NO   34
Id2   12514       78093       -- 12423        LD        YES  42
Id3
Id4
--
--
Id2000
***
12423: incision tool silver
12514: incision tool polymer
78093: incision tool cupper
(many more types here)







Right, though you'd probably have "ToolType=x" rather than "pick=1". And
although the poster states that no more than one tool type can occur in any
record, I think a check for more than one type occurring would be
imperative.

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

[Ergul, Emel A.]

Richard, Gene and John:

syntax worked perfectly well.
Thanks so much again!

emel


-----Original Message-----
From: SPSSX(r) Discussion on behalf of Richard Ristow
Sent: Fri 2/5/2010 1:44 PM
To: [hidden email]
Subject:      Re: Syntax for case seletion-clerification

At 01:42 PM 2/4/2010, Gene Maguin wrote:



        First, read up on the Any function. Before you run the command you'll
replace toola,toolb,toolc,toold,toole,toolf,toolg with the value that each of
them refer to. Finally, pick=1 is your subset of cases.

        Compute pick=0.
        Do repeat x=toola,toolb,toolc,toold,toole,toolf,toolg.
        +  if (any(x,v1 to v6)) pick=1.
        End repeat.


Right, though you'd probably have "ToolType=x" rather than "pick=1". And
although the poster states that no more than one tool type can occur in any
record, I think a check for more than one type occurring would be imperative.

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



The information in this e-mail is intended only for the person to whom it is
addressed. If you believe this e-mail was sent to you in error and the e-mail
contains patient information, please contact the Partners Compliance HelpLine at
http://www.partners.org/complianceline . If the e-mail was sent to you in error
but does not contain patient information, please contact the sender and properly
dispose of the e-mail.

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