Can we "freeze" at SPSS 15 for several years?

[John F Hall]

Mike

 

Nice contribution: keeps the debate going.  Had similar debates in UK in the 70s and 80s with statcomp nerds, but legitimate problems nevertheless.  It's a bit late over here, so I'll check the citations tomorrow.  Fortunately all my surveys were small enough to fit in SPSS.  Did you ever see the manuals for BMDP and Osiris?  Or Fakad?  Yikes!!  At PNL they used Minitab (double yikes!) but some people keep talking about SNAP.  Must investigate.

 

John Hall

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

From: [hidden email]

To: [hidden email] ; [hidden email]

Cc: [hidden email]

Sent: Friday, September 25, 2009 1:53 PM

Subject: [Norton AntiSpam] Re: Can we "freeze" at SPSS 15 for several years?

If one wants "SPSS versus other software" one should look at the Tabachnick & Fidel

"Using Multivariate Statistics" which is presently in the 5th edition.  Available on

Amazon and other booksellers, I suggest  looking at the 4th edition which compares

how SAS, SPSS, and SYSTAT (remember that?) compare in doing the same analyses:

http://www.amazon.com/Using-Multivariate-Statistics-Barbara-Tabachnick/dp/0321056779/ref=sr_1_1?ie=UTF8&s=books&qid=1253876611&sr=8-1

or

http://tinyurl.com/y8kxzxh

 

Back when I was in grad school (late 1970s, early 1980s) it was emphasized that doing

any serious (high stakes) statistical analysis should be in at least two different statistical

packages and to compare the output for both (I typically used BMDP and SPSS).  This

was a policy followed by some physicists who wanted to make sure that their home grown

programs behaved appropriately even if then used different algorithms for calculations. 

Of more direct relevant to experimental psychologists like myself and social science researchers

was the 1977 American Statistician article by Leland Wilkinson and Gerard Dallal which is

available on www.jstor.org. Here is the citation info along with a stable Jstor URL:

 

• Accuracy of Sample Moments Calculations among Widely Used Statistical Programs
• Leland Wilkinson and Gerard E. Dallal
• The American Statistician, Vol. 31, No. 3 (Aug., 1977), pp. 128-131
  (article consists of 4 pages)
• Published by: American Statistical Association
• Stable URL: http://www.jstor.org/stable/2682964

Wilkinson and Dallal showed how several different mainframe software programs

(i.e., SPSS, BMDP, OSIRIS, and DATATEXT) could produce errorneous results because

the programmers failed to take into account the possibility of underflow or overflow

errors because they had not allocated sufficient space in memory to represent very

small or very large numbers.  Only BMDP, because it used a different algorithm was

able to produce correct results.  I wondered how many people who had used SPSS

or the other programs for their dissertations or published researched actually went

back to their data and re-analyzed it in order to find out whether they had made any

errors.  Subsequent versions corrected these problem but given the complexity of

statistical programs, new problems will inevitably crop up.

 

Wilkinsion and Dallal would go on and develop various statistical testing batteries

to see how accurate the calculations are for the newer statistical packages. B.D.

McCullough has also done the same; see;

 

Assessing the Reliability of Statistical Software: Part I

B. D. McCullough

The American Statistician, Vol. 52, No. 4 (Nov., 1998), pp. 358-366

Published by: American Statistical Association

Stable URL: http://www.jstor.org/stable/2685442

 

Assessing the Reliability of Statistical Software: Part II

B. D. McCullough

The American Statistician, Vol. 53, No. 2 (May, 1999), pp. 149-159

Published by: American Statistical Association

Stable URL: http://www.jstor.org/stable/2685736

 

 

McCullough has also evaluated the statistical capabilities of off-the-shelf Excel and

his recommendation is don't use Excel for any serious statistical analysis. You

don't have to take his word

 

It should also be noted that Leland Wilkinson (his webpages is

http://www.cs.uic.edu/~wilkinson/ ) developed the SYSTAT software

package.  It might be remembered that SPSS bought SPSS (along with

a lot of other software packages) but then spun SYSTAT off  (the story

on this is provided by Wilkinson on his website; see:

http://www.cs.uic.edu/~wilkinson/SYSTAT/systat.html )

 

So,  SYSTAT is still available as a statistical package (I own a copy of

version 11) and more information about it can be obtained from the SYSTAT

website:   www.systat.com

I would suggest that people take it into consideration as a vaiable alternative

to SPSS but such a decision would be dependent on how well SYSTAT meets

you immediate and foreseeable needs.

 

-Mike Palij

New York University

[hidden email]

 

 

 

 

 

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

From: [hidden email]

To: [hidden email]

Sent: Friday, September 25, 2009 5:55 AM

Subject: Re: Can we "freeze" at SPSS 15 for several years?

Looks fairly intuitive.  Can it really do everything I want?  If so, does this mean I have to ditch 38 years of SPSS?

 

My tutorials are built round the processing and analysis of actual data from real questionnaire surveys (no artificial data whatsoever) from initial checking of raw data to publishable tables and charts.  I use SPSS because it was there when everyone needed it, was widely available and had a manual which made sense (even though I had to write tutorials to suit the research sequence rather than follow the alphetical sequence of commands). 

 

I'm too busy doing my SPSS versions, but it would be interesting to see parallel STATA syntax alongside that for SPSS.  More than 30 years ago, Steve Tagg and I thought of producing a Clods' Guide to Survey Analysis with SPSS, but we later also had the idea of a loose-leaf edition with side-by-side examples from other software.  I don't know if anyone has actually produced such a manual, but if anyone wants to try it with any of my tutorials, feel free.

 

They're all listed on  Survey Analysis Workshop (July 2009) with hyperlinks for free downloads.  By the end of today there'll be even more uploads.

 

This thread is throwing up some really interesting comments, but I'm not sure the subject is catching everyone who could contribute.  Perhaps something like, "SPSS versus other software"?

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

From: [hidden email]

To: [hidden email]

Cc: [hidden email]

Sent: Friday, September 25, 2009 11:00 AM

Subject: Re: Can we "freeze" at SPSS 15 for several years?

I don't get the complete SPSS-synta, but I try my best.
> recode         V363 (1 3 5 7 =1)(2 4 6 8 =2) into ethnic.
One possibility:
recode  V363 (1 3 5 7 =1)(2 4 6 8 =2)
ren V363 ethnic

> rename variables (v348 = sex).
ren V348 sex

> freq     sex ethnic.
tab sex ethnic (maybe with some suboptions, depending on what you want
in each cell)

In fact, I don't think there's much of a difference between SPSS and
Stata for these simple commands. The main advantage of the
Stata-syntax is that it always follows the structure "command
variables, options" which makes it pretty intuitive. For instance,

Linear regression
reg y x

with robust standard errors
reg y x, robust

with bootstrapped SEs (500 reps)
reg y x, vce(bootstrap, reps(500))

Logit
logit y x

with robust standard errors
logit y x, robust

And so on.

Best,

Nils
On Thu, Sep 24, 2009 at 10:58 PM, John F Hall <[hidden email]> wrote:

> If SPSS syntax is:
>
> recode V363 (1 3 5 7 =1)(2 4 6 8 =2) into ethnic.
> rename variables (v348 = sex).
> count sexism = V248 V252 V253 V256 V261 (3,4)
>
> V251 V255 V259 V260 (1,2).
>
> freq sex ethnic.
> freq sexism /his nor.
> cros sexism by sex ethnic
> /sexism by sex by ethnic /cel per.
> means sexism by sex ethnic
> /sexism by sex by ethnic.
>
> ...what would the syntax be in STATA and would it be as easy to understand?
>
>

[SR Millis-3]

Yves,

As a long-time user of SPSS, Stata, SAS, Systat, and R, I can say, without a doubt, it is time for you to switch to Stata.

I agree with all of your observations.  Yes, the Stata Journal is outstanding.

The Stata listserv is much more active than this listserv.  A much broader range of topics tends to be covered.

Scott Millis
~~~~~~~~~~~
"Kunst ist schön, macht aber viel Arbeit."

Scott R Millis, PhD, ABPP (CN,CL,RP), CStat, CSci
Professor & Director of Research
Dept of Physical Medicine & Rehabilitation
Dept of Emergency Medicine
Wayne State University School of Medicine
261 Mack Blvd
Detroit, MI 48201
Email:  [hidden email]
Tel: [phone]
Fax: [phone]


--- On Fri, 9/25/09, Yves Therriault <[hidden email]> wrote:

> From: Yves Therriault <[hidden email]>
> Subject: Re: Can we "freeze" at SPSS 15 for several years?
> To: [hidden email]
> Date: Friday, September 25, 2009, 4:46 PM
>
> Dear SPSS listers,
>
>
>
> I've been using SPSS for almost 20 years in my work as
> a research analyst in public health. I've been a SPSS
> fan since the old days when I was working with SPSS Pc +.  I
> rarely use the GUI and prefer to work with syntax. I learned
> a lot from this list (many thanks to Ray Levesque, Hector
> Malletta, and many others).
>
>
>
> Yet, I'm seriously considering to switch to Stata
> instead of upgrading to the new SPSS 18 release (from my
> current 12.01 version). I've had the opportunity to test
> a demo version of Stata (Stata intercooled) earlier this
> year and I'm very impressed with the software. I'm
> not a statistician, but it seems to me that Stata is
> generally superior to SPSS in many ways.
>
>
>
> -  First of all, with Stata, we have the
> "complete" thing. No need to buy rather expensive
> add-on modules if one needs and wants to do more advanced
> statistical analysis or has to analyse data from complex
> samples. Even though Stata has a complete set of features,
> the IC version is less expensive than the SPSS Base version
> (sorry, I'm still not used to PASW name).
>
>
>
> - Stata syntax seems to be far more intuitive.
>
>
>
> - Better documentation.
>
>
>
> - Features and procedures very useful in public health
> (like the standardization of rates, direct or indirect).
>
>
>
> - For what I know, Stata is much more internet aware than
> SPSS. Updating Stata is really easy as Garry said.  There
> are hundreds and hundreds of  user-written routines that can
> be installed quicky and directly from the internet with a
> particular Stata command. No need no login to Stata website
> do so.
>
>
>
> -  Stata website has much more useful resources. The
> information we seek is very easy to find.
>
>
>
> -  Licensing policies that allow Stata users to use their
> license code to install the software on 3 computers (office,
> home, and personal laptop) provided of course that only one
> instance of Stata is running at the same time.
>
>
>
> - Stata concentrates its efforts into developing a product
> in a way I agree with. I never really agreed with the
> orientation that SPSS has taken in the last years (focusing
> on "predictive analysis" for big corporations in
> order to help them to make better decisions).
>
>
>
> - Stata publishes its own quarterly peer-reviewed journal
> in Statistics. I can't judge the merits of Stata journal
> nor its quality, but I think its a good indicator that Stata
> is very "statistically oriented". The journal also
> publishes notes, columns, comments and tips written by
> strong Stata users.
>
>
>
> I totally agree with Garry that SPSS output capabilities
> are far better than Stata's. Stata is currently light
> years behind SPSS regarding output management. It's the
> only advantage I see in SPSS over Stata and it's the
> only reason why I didn't switch to Stata yet.
>
>
>
> Nevertheless, is it time to see goodbye to an old friend ?
>
>
>
>
> Kind regards,
>
>
>
> Yves
> Therriault, Ph. D.
>
> Agent de recherche
>
> Surveillance de l'état de santé de la
> population (Santé publique)
>
> Agence de la santé et des services sociaux de la
> Côte-Nord
>
> Baie-Comeau
> (QC), Canada
>
>
>
>
>
> -----Original Message-----
>
>
>
> Date:    Thu, 24 Sep 2009 20:23:04 +0100
>
> From:    Garry Gelade
> <[hidden email]>
>
> Subject: Re: Can we "freeze" at SPSS 15 for
> several years?
>
>
>
> Bruce
>
>
>
> One of the great advantages of Stata is the ease with
> which new statistical
>
> procedures can be added, eg if you want a robust statistics
> module, or a
>
> spatial autocorrelation module, you can just download the
> relevant package
>
> for free, and install it quickly. You don't have to
> 'upgrade'. SPSS on the
>
> other hand comes with a fixed set of procedures. Of course
> you can add on
>
> modules - at a price - but even then the range on offer is
> nowhere near as
>
> comprehensive as in Stata.
>
>
>
> That said, SPSS is a fantastic product in many ways,
> and its output
>
> capabilities are far better than Stata. But instead of
> spending development
>
> effort on annual upgrades, which are expensive and time
> consuming for many
>
> users, I'd like to see SPSS extending the Python and R
> side, and building a
>
> range of downloadable packages.
>
>
>
> Garry Gelade
>
> Business Analytic Ltd

=====================
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]

I still have a couple of editions of the BMDP manuals plus the

small spiral bound booklets (right next my copies of really old

SPSS manuals)  as well as two version of BMDP-PC which I

purchased with my own funds (I stopped using BMDP PC in

the late 1990s after SPSS bought them out and no longer supported

the software as the Windows OS got upgraded).  There are a

number of fans of Minitab in academia who use it in their

stats courses  (I have had arguments with these folks over the

decades, saying that it was better to use SPSS because of its wider

acceptance in research contexts). 

 

The McCullough articles and related materials on software

reliability should be well known to some of the people on this list. 

I assume that David Nichols, whom I recognize from when I was

on this list in the early/mid 1990s, is probably well aware of these

issues.  I assume that whoever is responsible for quality control

of the software at SPSS routinely tests the software for calculation

errors/problems.

 

-Mike Palij

New York University

[hidden email]

 

 

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

From: [hidden email]

To: [hidden email]

Sent: Friday, September 25, 2009 5:05 PM

Subject: Re: Can we "freeze" at SPSS 15 for several years?

Mike

 

Nice contribution: keeps the debate going.  Had similar debates in UK in the 70s and 80s with statcomp nerds, but legitimate problems nevertheless.  It's a bit late over here, so I'll check the citations tomorrow.  Fortunately all my surveys were small enough to fit in SPSS.  Did you ever see the manuals for BMDP and Osiris?  Or Fakad?  Yikes!!  At PNL they used Minitab (double yikes!) but some people keep talking about SNAP.  Must investigate.

 

John Hall

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

From: [hidden email]

To: [hidden email] ; [hidden email]

Cc: [hidden email]

Sent: Friday, September 25, 2009 1:53 PM

Subject: [Norton AntiSpam] Re: Can we "freeze" at SPSS 15 for several years?

If one wants "SPSS versus other software" one should look at the Tabachnick & Fidel

"Using Multivariate Statistics" which is presently in the 5th edition.  Available on

Amazon and other booksellers, I suggest  looking at the 4th edition which compares

how SAS, SPSS, and SYSTAT (remember that?) compare in doing the same analyses:

http://www.amazon.com/Using-Multivariate-Statistics-Barbara-Tabachnick/dp/0321056779/ref=sr_1_1?ie=UTF8&s=books&qid=1253876611&sr=8-1

or

http://tinyurl.com/y8kxzxh

 

Back when I was in grad school (late 1970s, early 1980s) it was emphasized that doing

any serious (high stakes) statistical analysis should be in at least two different statistical

packages and to compare the output for both (I typically used BMDP and SPSS).  This

was a policy followed by some physicists who wanted to make sure that their home grown

programs behaved appropriately even if then used different algorithms for calculations. 

Of more direct relevant to experimental psychologists like myself and social science researchers

was the 1977 American Statistician article by Leland Wilkinson and Gerard Dallal which is

available on www.jstor.org. Here is the citation info along with a stable Jstor URL:

 

• Accuracy of Sample Moments Calculations among Widely Used Statistical Programs
• Leland Wilkinson and Gerard E. Dallal
• The American Statistician, Vol. 31, No. 3 (Aug., 1977), pp. 128-131
  (article consists of 4 pages)
• Published by: American Statistical Association
• Stable URL: http://www.jstor.org/stable/2682964

Wilkinson and Dallal showed how several different mainframe software programs

(i.e., SPSS, BMDP, OSIRIS, and DATATEXT) could produce errorneous results because

the programmers failed to take into account the possibility of underflow or overflow

errors because they had not allocated sufficient space in memory to represent very

small or very large numbers.  Only BMDP, because it used a different algorithm was

able to produce correct results.  I wondered how many people who had used SPSS

or the other programs for their dissertations or published researched actually went

back to their data and re-analyzed it in order to find out whether they had made any

errors.  Subsequent versions corrected these problem but given the complexity of

statistical programs, new problems will inevitably crop up.

 

Wilkinsion and Dallal would go on and develop various statistical testing batteries

to see how accurate the calculations are for the newer statistical packages. B.D.

McCullough has also done the same; see;

 

Assessing the Reliability of Statistical Software: Part I

B. D. McCullough

The American Statistician, Vol. 52, No. 4 (Nov., 1998), pp. 358-366

Published by: American Statistical Association

Stable URL: http://www.jstor.org/stable/2685442

 

Assessing the Reliability of Statistical Software: Part II

B. D. McCullough

The American Statistician, Vol. 53, No. 2 (May, 1999), pp. 149-159

Published by: American Statistical Association

Stable URL: http://www.jstor.org/stable/2685736

 

 

McCullough has also evaluated the statistical capabilities of off-the-shelf Excel and

his recommendation is don't use Excel for any serious statistical analysis. You

don't have to take his word

 

It should also be noted that Leland Wilkinson (his webpages is

http://www.cs.uic.edu/~wilkinson/ ) developed the SYSTAT software

package.  It might be remembered that SPSS bought SPSS (along with

a lot of other software packages) but then spun SYSTAT off  (the story

on this is provided by Wilkinson on his website; see:

http://www.cs.uic.edu/~wilkinson/SYSTAT/systat.html )

 

So,  SYSTAT is still available as a statistical package (I own a copy of

version 11) and more information about it can be obtained from the SYSTAT

website:   www.systat.com

I would suggest that people take it into consideration as a vaiable alternative

to SPSS but such a decision would be dependent on how well SYSTAT meets

you immediate and foreseeable needs.

 

-Mike Palij

New York University

[hidden email]

 

 

 

 

 

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

From: [hidden email]

To: [hidden email]

Sent: Friday, September 25, 2009 5:55 AM

Subject: Re: Can we "freeze" at SPSS 15 for several years?

Looks fairly intuitive.  Can it really do everything I want?  If so, does this mean I have to ditch 38 years of SPSS?

 

My tutorials are built round the processing and analysis of actual data from real questionnaire surveys (no artificial data whatsoever) from initial checking of raw data to publishable tables and charts.  I use SPSS because it was there when everyone needed it, was widely available and had a manual which made sense (even though I had to write tutorials to suit the research sequence rather than follow the alphetical sequence of commands). 

 

I'm too busy doing my SPSS versions, but it would be interesting to see parallel STATA syntax alongside that for SPSS.  More than 30 years ago, Steve Tagg and I thought of producing a Clods' Guide to Survey Analysis with SPSS, but we later also had the idea of a loose-leaf edition with side-by-side examples from other software.  I don't know if anyone has actually produced such a manual, but if anyone wants to try it with any of my tutorials, feel free.

 

They're all listed on  Survey Analysis Workshop (July 2009) with hyperlinks for free downloads.  By the end of today there'll be even more uploads.

 

This thread is throwing up some really interesting comments, but I'm not sure the subject is catching everyone who could contribute.  Perhaps something like, "SPSS versus other software"?

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

From: [hidden email]

To: [hidden email]

Cc: [hidden email]

Sent: Friday, September 25, 2009 11:00 AM

Subject: Re: Can we "freeze" at SPSS 15 for several years?

I don't get the complete SPSS-synta, but I try my best.
> recode         V363 (1 3 5 7 =1)(2 4 6 8 =2) into ethnic.
One possibility:
recode  V363 (1 3 5 7 =1)(2 4 6 8 =2)
ren V363 ethnic

> rename variables (v348 = sex).
ren V348 sex

> freq     sex ethnic.
tab sex ethnic (maybe with some suboptions, depending on what you want
in each cell)

In fact, I don't think there's much of a difference between SPSS and
Stata for these simple commands. The main advantage of the
Stata-syntax is that it always follows the structure "command
variables, options" which makes it pretty intuitive. For instance,

Linear regression
reg y x

with robust standard errors
reg y x, robust

with bootstrapped SEs (500 reps)
reg y x, vce(bootstrap, reps(500))

Logit
logit y x

with robust standard errors
logit y x, robust

And so on.

Best,

Nils
On Thu, Sep 24, 2009 at 10:58 PM, John F Hall <[hidden email]> wrote:

> If SPSS syntax is:
>
> recode V363 (1 3 5 7 =1)(2 4 6 8 =2) into ethnic.
> rename variables (v348 = sex).
> count sexism = V248 V252 V253 V256 V261 (3,4)
>
> V251 V255 V259 V260 (1,2).
>
> freq sex ethnic.
> freq sexism /his nor.
> cros sexism by sex ethnic
> /sexism by sex by ethnic /cel per.
> means sexism by sex ethnic
> /sexism by sex by ethnic.
>
> ...what would the syntax be in STATA and would it be as easy to understand?
>
>

[John F Hall]

When I was teaching Data Management and Analysis (aka Survey Analysis Workshop under another name) to 2nd year undergraduates reading Social Research or Sociology (using SPSS), a colleague was simultaneously teaching them Statistics (using Minitab).  The students told me they learned more statistics from my course. 

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

From: [hidden email]

To: [hidden email] ; [hidden email]

Cc: [hidden email]

Sent: Friday, September 25, 2009 11:22 PM

Subject: Re: Can we "freeze" at SPSS 15 for several years?

I still have a couple of editions of the BMDP manuals plus the

small spiral bound booklets (right next my copies of really old

SPSS manuals)  as well as two version of BMDP-PC which I

purchased with my own funds (I stopped using BMDP PC in

the late 1990s after SPSS bought them out and no longer supported

the software as the Windows OS got upgraded).  There are a

number of fans of Minitab in academia who use it in their

stats courses  (I have had arguments with these folks over the

decades, saying that it was better to use SPSS because of its wider

acceptance in research contexts). 

 

The McCullough articles and related materials on software

reliability should be well known to some of the people on this list. 

I assume that David Nichols, whom I recognize from when I was

on this list in the early/mid 1990s, is probably well aware of these

issues.  I assume that whoever is responsible for quality control

of the software at SPSS routinely tests the software for calculation

errors/problems.

 

-Mike Palij

New York University

[hidden email]

 

 

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

From: [hidden email]

To: [hidden email]

Sent: Friday, September 25, 2009 5:05 PM

Subject: Re: Can we "freeze" at SPSS 15 for several years?

Mike

 

Nice contribution: keeps the debate going.  Had similar debates in UK in the 70s and 80s with statcomp nerds, but legitimate problems nevertheless.  It's a bit late over here, so I'll check the citations tomorrow.  Fortunately all my surveys were small enough to fit in SPSS.  Did you ever see the manuals for BMDP and Osiris?  Or Fakad?  Yikes!!  At PNL they used Minitab (double yikes!) but some people keep talking about SNAP.  Must investigate.

 

John Hall

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

From: [hidden email]

To: [hidden email] ; [hidden email]

Cc: [hidden email]

Sent: Friday, September 25, 2009 1:53 PM

Subject: [Norton AntiSpam] Re: Can we "freeze" at SPSS 15 for several years?

If one wants "SPSS versus other software" one should look at the Tabachnick & Fidel

"Using Multivariate Statistics" which is presently in the 5th edition.  Available on

Amazon and other booksellers, I suggest  looking at the 4th edition which compares

how SAS, SPSS, and SYSTAT (remember that?) compare in doing the same analyses:

http://www.amazon.com/Using-Multivariate-Statistics-Barbara-Tabachnick/dp/0321056779/ref=sr_1_1?ie=UTF8&s=books&qid=1253876611&sr=8-1

or

http://tinyurl.com/y8kxzxh

 

Back when I was in grad school (late 1970s, early 1980s) it was emphasized that doing

any serious (high stakes) statistical analysis should be in at least two different statistical

packages and to compare the output for both (I typically used BMDP and SPSS).  This

was a policy followed by some physicists who wanted to make sure that their home grown

programs behaved appropriately even if then used different algorithms for calculations. 

Of more direct relevant to experimental psychologists like myself and social science researchers

was the 1977 American Statistician article by Leland Wilkinson and Gerard Dallal which is

available on www.jstor.org. Here is the citation info along with a stable Jstor URL:

 

• Accuracy of Sample Moments Calculations among Widely Used Statistical Programs
• Leland Wilkinson and Gerard E. Dallal
• The American Statistician, Vol. 31, No. 3 (Aug., 1977), pp. 128-131
  (article consists of 4 pages)
• Published by: American Statistical Association
• Stable URL: http://www.jstor.org/stable/2682964

Wilkinson and Dallal showed how several different mainframe software programs

(i.e., SPSS, BMDP, OSIRIS, and DATATEXT) could produce errorneous results because

the programmers failed to take into account the possibility of underflow or overflow

errors because they had not allocated sufficient space in memory to represent very

small or very large numbers.  Only BMDP, because it used a different algorithm was

able to produce correct results.  I wondered how many people who had used SPSS

or the other programs for their dissertations or published researched actually went

back to their data and re-analyzed it in order to find out whether they had made any

errors.  Subsequent versions corrected these problem but given the complexity of

statistical programs, new problems will inevitably crop up.

 

Wilkinsion and Dallal would go on and develop various statistical testing batteries

to see how accurate the calculations are for the newer statistical packages. B.D.

McCullough has also done the same; see;

 

Assessing the Reliability of Statistical Software: Part I

B. D. McCullough

The American Statistician, Vol. 52, No. 4 (Nov., 1998), pp. 358-366

Published by: American Statistical Association

Stable URL: http://www.jstor.org/stable/2685442

 

Assessing the Reliability of Statistical Software: Part II

B. D. McCullough

The American Statistician, Vol. 53, No. 2 (May, 1999), pp. 149-159

Published by: American Statistical Association

Stable URL: http://www.jstor.org/stable/2685736

 

 

McCullough has also evaluated the statistical capabilities of off-the-shelf Excel and

his recommendation is don't use Excel for any serious statistical analysis. You

don't have to take his word

 

It should also be noted that Leland Wilkinson (his webpages is

http://www.cs.uic.edu/~wilkinson/ ) developed the SYSTAT software

package.  It might be remembered that SPSS bought SPSS (along with

a lot of other software packages) but then spun SYSTAT off  (the story

on this is provided by Wilkinson on his website; see:

http://www.cs.uic.edu/~wilkinson/SYSTAT/systat.html )

 

So,  SYSTAT is still available as a statistical package (I own a copy of

version 11) and more information about it can be obtained from the SYSTAT

website:   www.systat.com

I would suggest that people take it into consideration as a vaiable alternative

to SPSS but such a decision would be dependent on how well SYSTAT meets

you immediate and foreseeable needs.

 

-Mike Palij

New York University

[hidden email]

 

 

 

 

 

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

From: [hidden email]

To: [hidden email]

Sent: Friday, September 25, 2009 5:55 AM

Subject: Re: Can we "freeze" at SPSS 15 for several years?

Looks fairly intuitive.  Can it really do everything I want?  If so, does this mean I have to ditch 38 years of SPSS?

 

My tutorials are built round the processing and analysis of actual data from real questionnaire surveys (no artificial data whatsoever) from initial checking of raw data to publishable tables and charts.  I use SPSS because it was there when everyone needed it, was widely available and had a manual which made sense (even though I had to write tutorials to suit the research sequence rather than follow the alphetical sequence of commands). 

 

I'm too busy doing my SPSS versions, but it would be interesting to see parallel STATA syntax alongside that for SPSS.  More than 30 years ago, Steve Tagg and I thought of producing a Clods' Guide to Survey Analysis with SPSS, but we later also had the idea of a loose-leaf edition with side-by-side examples from other software.  I don't know if anyone has actually produced such a manual, but if anyone wants to try it with any of my tutorials, feel free.

 

They're all listed on  Survey Analysis Workshop (July 2009) with hyperlinks for free downloads.  By the end of today there'll be even more uploads.

 

This thread is throwing up some really interesting comments, but I'm not sure the subject is catching everyone who could contribute.  Perhaps something like, "SPSS versus other software"?

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

From: [hidden email]

To: [hidden email]

Cc: [hidden email]

Sent: Friday, September 25, 2009 11:00 AM

Subject: Re: Can we "freeze" at SPSS 15 for several years?

I don't get the complete SPSS-synta, but I try my best.
> recode         V363 (1 3 5 7 =1)(2 4 6 8 =2) into ethnic.
One possibility:
recode  V363 (1 3 5 7 =1)(2 4 6 8 =2)
ren V363 ethnic

> rename variables (v348 = sex).
ren V348 sex

> freq     sex ethnic.
tab sex ethnic (maybe with some suboptions, depending on what you want
in each cell)

In fact, I don't think there's much of a difference between SPSS and
Stata for these simple commands. The main advantage of the
Stata-syntax is that it always follows the structure "command
variables, options" which makes it pretty intuitive. For instance,

Linear regression
reg y x

with robust standard errors
reg y x, robust

with bootstrapped SEs (500 reps)
reg y x, vce(bootstrap, reps(500))

Logit
logit y x

with robust standard errors
logit y x, robust

And so on.

Best,

Nils
On Thu, Sep 24, 2009 at 10:58 PM, John F Hall <[hidden email]> wrote:

> If SPSS syntax is:
>
> recode V363 (1 3 5 7 =1)(2 4 6 8 =2) into ethnic.
> rename variables (v348 = sex).
> count sexism = V248 V252 V253 V256 V261 (3,4)
>
> V251 V255 V259 V260 (1,2).
>
> freq sex ethnic.
> freq sexism /his nor.
> cros sexism by sex ethnic
> /sexism by sex by ethnic /cel per.
> means sexism by sex ethnic
> /sexism by sex by ethnic.
>
> ...what would the syntax be in STATA and would it be as easy to understand?
>
>

[Marta Garcia-Granero]

Hi everybody:

To close this active thread, this is the result of our meeting at the
University. We all agreed (more or less) that:

- We wanted SPSS 15, for teaching purposes, for at least 3 years
- If we could not have it, then we would move to Stata
(teaching&research) and R (research). My particular vote was I would go
on using PASW for my work, but I would also learn Stata and R (since I
can benefit from the PASW+R combination)

Curiously, part of the problem was they thought Stata was superior to
SPSS/PASW, only because they did not know how to write macros. I gave
them a copy of my most used macros (multiple comparisons for non
parametrics, confidence intervals for proportions, medians and their
differences, boostrapping ROC AUC...) and they left the meeting think
that, after all, SPSS was not that bad... If we keep on using SPSS/PASW,
they want me to give them training on writing SPSS macros.

Unfortunately, when we contacted the local sales manager, we got no
answer to the question "Can we have SPSS 15 for 3 more years?". Their
reply was something like "yes, no, maybe, I don't know, could you repeat
the question?" (this is a joke of mine, a reference to one of my
favorite TV series, Malcolm in the middle). Now, seriously, their reply
was there would be no reply until January-February, until they got the
chance to talk with the new IBM bosses. I found that a bit surprising,
are we being bamboozled?

Therefore:

- We are going to get someone to give us a good R course.
- We are going to start talking to Stata people, just in case (to get a
couple a licenses to start playing with the program, get some training
from them).
- We will get some StatTransfer licenses as well (to move old SPSS
databases to Stata)
- We are contacting research centers and hospitals in Spain to find out
what they use and what they plan to use in the future (to move with the
stream)

I am afraid that we will get a NO from SPSS Spain in February (it has
simply been put off), and I will have to switch to Stata (at least all
my teaching material will).

Regards,
Marta

--
For miscellaneous SPSS related statistical stuff, visit:
http://gjyp.nl/marta/

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

[Robert Marshall-7]

Thought everyone would be interested, there was a article in yeserday's Chicago Tribune, discussing SPSS's potential sale to IBM, and some background on the name change.  SPSS was started here in Chicago. 

----- Original Message -----
From: "Marta García-Granero" <[hidden email]>
To: [hidden email]
Sent: Monday, September 28, 2009 2:42:24 AM GMT -06:00 US/Canada Central
Subject: Re: Can we "freeze" at SPSS 15 for several years? Epilogue

Hi everybody:

To close this active thread, this is the result of our meeting at the
University. We all agreed (more or less) that:

- We wanted SPSS 15, for teaching purposes, for at least 3 years
- If we could not have it, then we would move to Stata
(teaching&research) and R (research). My particular vote was I would go
on using PASW for my work, but I would also learn Stata and R (since I
can benefit from the PASW+R combination)

Curiously, part of the problem was they thought Stata was superior to
SPSS/PASW, only because they did not know how to write macros. I gave
them a copy of my most used macros (multiple comparisons for non
parametrics, confidence intervals for proportions, medians and their
differences, boostrapping ROC AUC...) and they left the meeting think
that, after all, SPSS was not that bad... If we keep on using SPSS/PASW,
they want me to give them training on writing SPSS macros.

Unfortunately, when we contacted the local sales manager, we got no
answer to the question "Can we have SPSS 15 for 3 more years?". Their
reply was something like "yes, no, maybe, I don't know, could you repeat
the question?" (this is a joke of mine, a reference to one of my
favorite TV series, Malcolm in the middle). Now, seriously, their reply
was there would be no reply until January-February, until they got the
chance to talk with the new IBM bosses. I found that a bit surprising,
are we being bamboozled?

Therefore:

- We are going to get someone to give us a good R course.
- We are going to start talking to Stata people, just in case (to get a
couple a licenses to start playing with the program, get some training
from them).
- We will get some StatTransfer licenses as well (to move old SPSS
databases to Stata)
- We are contacting research centers and hospitals in Spain to find out
what they use and what they plan to use in the future (to move with the
stream)

I am afraid that we will get a NO from SPSS Spain in February (it has
simply been put off), and I will have to switch to Stata (at least all
my teaching material will).

Regards,
Marta

--
For miscellaneous SPSS related statistical stuff, visit:
http://gjyp.nl/marta/

=====================
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 a person refers to a newspaper article, it's useful to be a little

more specific about whether one read it in its paper form or online

in electronic form.  I think that we're familiar here with how to

document citations and what makes for a useful reference.

 

That being said, I am not sure if the author below was referring to

the following article or not:

http://www.chicagotribune.com/business/chi-sun-nie-spss-sep27,0,3297617.story

 

The reason why I am uncertain is because the author below says the SPSS

started in Chicago while the article says:

 

|Nie worked with two friends to create the software in 1960s

|while he was a graduate student at Stanford University. In 1968

|Nie moved to the political science department at the University of Chicago

|and brought the software and one of his friends, C. Hadlai "Tex" Hull,

|with him.

 

The actual software was started at Stanford while the company was started

in Chicago.  Quoting the Tribune article:

 

|Together they refined the software and distributed it to other universities.
|
|Soon the software business became too big for the university, and

|Nie and Hull formed a separate company called SPSS in 1975.

|Nie and Hull registered as owners of the "SPSS" trademark and

|licensed the mark to the company. The agreement provided the

|company with an exclusive, perpetual, worldwide, royalty-free license.

 

In any event, SPSS closed on Friday at $49.91 and IBM closed at $121.08

(according to finance.yahoo.com ).

 

SPSS paid no dividend while IBM pays a 1.80% dividend.  Some might say

that buying SPSS stock now would a cheap way to buy IBM and get a

dividend as a bonus.  Note that the sale of SPSS to IBM will take place

on October 2, 2009.  See:

http://news.yahoo.com/s/ap/20090828/ap_on_bi_ge/us_spss_ibm

http://www.chicagobusiness.com/cgi-bin/news.pl?id=35296&seenIt=1

 

Even if SPSS goes the way of Lotus, you'll still have the IBM stock.

 

-Mike Palij

New York University

[hidden email]

 

 

 

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

From: [hidden email]

To: [hidden email]

Sent: Monday, September 28, 2009 6:23 AM

Subject: Re: Can we "freeze" at SPSS 15 for several years? Epilogue

Thought everyone would be interested, there was a article in yeserday's Chicago Tribune, discussing SPSS's potential sale to IBM, and some background on the name change.  SPSS was started here in Chicago. 

----- Original Message -----
From: "Marta García-Granero" <[hidden email]>
To: [hidden email]
Sent: Monday, September 28, 2009 2:42:24 AM GMT -06:00 US/Canada Central
Subject: Re: Can we "freeze" at SPSS 15 for several years? Epilogue

Hi everybody:

To close this active thread, this is the result of our meeting at the
University. We all agreed (more or less) that:

- We wanted SPSS 15, for teaching purposes, for at least 3 years
- If we could not have it, then we would move to Stata
(teaching&research) and R (research). My particular vote was I would go
on using PASW for my work, but I would also learn Stata and R (since I
can benefit from the PASW+R combination)

Curiously, part of the problem was they thought Stata was superior to
SPSS/PASW, only because they did not know how to write macros. I gave
them a copy of my most used macros (multiple comparisons for non
parametrics, confidence intervals for proportions, medians and their
differences, boostrapping ROC AUC...) and they left the meeting think
that, after all, SPSS was not that bad... If we keep on using SPSS/PASW,
they want me to give them training on writing SPSS macros.

Unfortunately, when we contacted the local sales manager, we got no
answer to the question "Can we have SPSS 15 for 3 more years?". Their
reply was something like "yes, no, maybe, I don't know, could you repeat
the question?" (this is a joke of mine, a reference to one of my
favorite TV series, Malcolm in the middle). Now, seriously, their reply
was there would be no reply until January-February, until they got the
chance to talk with the new IBM bosses. I found that a bit surprising,
are we being bamboozled?

Therefore:

- We are going to get someone to give us a good R course.
- We are going to start talking to Stata people, just in case (to get a
couple a licenses to start playing with the program, get some training
from them).
- We will get some StatTransfer licenses as well (to move old SPSS
databases to Stata)
- We are contacting research centers and hospitals in Spain to find out
what they use and what they plan to use in the future (to move with the
stream)

I am afraid that we will get a NO from SPSS Spain in February (it has
simply been put off), and I will have to switch to Stata (at least all
my teaching material will).

Regards,
Marta

--
For miscellaneous SPSS related statistical stuff, visit:
http://gjyp.nl/marta/

=====================
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]



Bit more on the history of SPSS.  The following is an extract from Appendix 1 of my 2006 paper Old Dog, Old Tricks. 

 

The full paper is on Old Dog, Old Tricks: Using SPSS Syntax to Avoid the Mouse Trap 

[Hall J F, 2006] Paper to ASSESS (SPSS Users) York Univ. (UK) Dec 2006

 

---

John Hall to Tony Coxon 12 June 2006

 

Tony

 

Tracked you down after all these years.  I'm doing a turn for the SPSS
users' group ASSESS in York in November and want to stick a bit in at the
beginning about how SPSS got to UK.  David Muxworthy says you brought it
over and it was first installed at Edinburgh in 1970.

I've been doing a bit of work here restoring old SPSS files from SSRC Survey
Unit days and teaching materials from PNL.  Had to teach myself Windows,
Word and mouse from scratch when I offered to review Julie Pallant's SPSS
Survival Manual for the Social Research Association.  Luckily I got SPSS to
let me have an evaluation copy of the Windows version and then they gave me
a free 5-year licence (expires Sep 2007).

I can send you the review and other materials if you're interested.

Also found some old QSN material from 1974 and 1978.  Interesting reading!

 

2:        SPSS arrives in Edinburgh, 1970

 

Tony Coxon to John Hall 12 June 2006-08-31

 

Excellent to hear from you, even though the route taken by the email was circuitous! In the signature below you will find a selection of addresses, though [hidden email] is the preferred one.

I retired from Essex (with relief) in 2002 and moved here to the Isle of Islay (eight whisky distilleries ...)

…………….

 

Now to the subject matter of the email. As you (and David) rightly say 'twas I that was responsible for bringing SPSS over to the UK ... though I keep quiet about it .

 

The background is in some ways more interesting than the mere fact. You especially will remember that the late '60s were characterised by governmental protection of home computer industries (EELM/ ICT/ICL of course) and it was virtually impossible for Universities to get their hands on an IBM. Indeed, it was only those with big natural-science clout that did -- Imperial, Newcastle, Edinburgh come to mind. Equally, you will remember that in those days all the interesting new social science software emanated from the US and Michigan in particular and that it a tedious expensive job to adapt for bloody non-IBM machines, so that those Universities with an IBM were in the privileged position to be able to implement/ run such software immediately.

 

My involvement was that in 1968 I moved from Leeds and took a year as Visiting lecturer at MIT/Harvard Political Science, returning to take up a new post as Lecturer at Edinburgh (now you see the link?). During that time, the first interactive survey package was being developed at MIT (PI: Ithiel de Solla Pool), called ADMINS and also at Harvard, David Armor was developing a general package for the analysis of BOTH survey AND textual data called DATA-TEXT (see my comments at: http://www.qualitative-research.net/fqs-texte/2-05/05-2-40-e.htm ). Armor was pissed off that the Chicago lot had pinched most of his structure  and evacuated it of the qualitative component. He was not a happy bunny ... and of course "the Chicago lot" were SPSS!. However, I did take a copy of SPSS back with me from the US to Edinburgh.

 

The story continues. Edinburgh University had a standing committee for creating a new Edinburgh all-singing, all-dancing Survey Package. Tom Burns sent the new poilitically-naive Tony Coxon as the departmental representative. I couldn't believe it ... Big departments with entrenched positions (Agriculture, Economics ...) were arguing furiously and were unprepared to give an inch. In desperation, I interjected with something like " In the mean time, would it be an idea to use an existing new package until the subtleties of the Edinburgh package are decided? ... it so happens I happen to have a tape with SPSS" ... and that was it. It was agreed, and PLU (with Marjory Barritt -- remember?!-- and David Mux. ) ran with it. Nothing more was heard of the Edinburgh Package!

 

So there you have it!

 

Incidentally, I'm still persevering with MDSX -- now in a Windows incarnation. Have a look: www.newmdsx.com

 

Tell me more about yourself and your developments.

 

 

John Hall to Tony Coxon and David Muxworthy, 16 August 2006

 

I've got quite a long way with a draft presentation for the November ASSESS meeting in York.  One paragraph reads as follows:

 

SPSS was originally written in Fortran for an IBM by three postgraduate students.  It came from Chicago to Edinburgh in 1970 via Tony Coxon and was implemented at ERCC (one of the few places with an IBM) by David Muxworthy and Marjorie Barritt (thereby scotching university plans to commission a survey processing facility at great expense from scratch) and when first installed was reputedly called more times than the Fortran compiler.  Conversions to ICL followed later, but those with CDC and DEC machines got SPSS sooner.

 

Is this accurate and complete?  Please amend to suit and return to me.

 

 

David Muxworthy to John Hall 16 and 19 Aug 2006:

 

As I understand it SPSS first appeared in 1968.  The first UK installation was indeed at Edinburgh RCC in 1970, brought in on Tony  Coxon's recommendation.  I'm not sure about plans for commissioning  something from scratch.  I'll poke about in my loft to see if I've  anything to jog the memory.

Norman Nie and Dale Bent were political science postgrads at Stanford  in the late 1960s and, fed up with the 'put a 1 in column 72'  type  command language of the programs at the time, they devised a language  that a political scientist would want to write to specify an analysis.  They scraped together some funds and hired Tex Hull to  help with coding the program, which was in Fortran IV for the 360.  I 
think Tex must have been finishing his first degree at Yale or maybe a Masters at Stanford, but  I'm not  sure what he did at Stanford. Norman and Tex both moved to Chicago, Norman to the  National Opinion Research Center, Tex to the Computing Center.   (Norman was originally from St Louis, Tex from Minnesota).  Dale went  back to Alberta and, apart from having his name on some of the  manuals, dropped out of SPSS.

People got to hear about the program, which was superior in user  interface to much that was available at the time, and requested  copies.  This led to Patrick Bova, a librarian at NORC, being hired  25% of his time to act as distribution agent and Karin Steinbrenner  being hired as full time programmer.  When I visited Chicago in the  summer of 1972 this was the total staff.  I thought I was going to a large software house.  It was surprising to find it not much bigger than a one man  and a dog in a bedroom outfit (at that time at least). Tex acted largely as  advisor but was busy as associate director of the computing center.

As I remember it, Jean Jenkins was hired as programmer later in 1972  or in 1973.  She was certainly around at the SCSS planning meetings  in the summer of 1973.  The program was so successful that NORC  became wary of losing their non-profit status and strongly encouraged Norman to form a company and move out.  This happened sometime  between 1974 (when I worked with them at NORC) and 1977 (when they  had moved to an office block in downtown Chicago).

In Edinburgh the program grew to be so popular there were demands to move it to the ICL system 4 and later the ICL 2980, the IBM having  been removed by higher authority.  This led to PLU organising 
conversions to some other platforms in UK universities, notably the  ICL 1900.  SPSS themselves arranged conversions to other series,  notably the CDC 6600 at Northwestern University, just up the road  from Chicago.

 

(JFH: …..thereby scotching university plans to commission a survey  processing facility at great expense from scratch.)

You were quite right.  Attached are the relevant minutes of the  Edinburgh committee looking into this.  These appear to be the only  references to SPSS - much of the committee time was spent on building  up maths and engineering software.  The names won't mean anything to  you but will revive memories (good or bad) for Tony.

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

From: [hidden email]

To: [hidden email]

Sent: Monday, September 28, 2009 1:02 PM

Subject: Re: Can we "freeze" at SPSS 15 for several years? Epilogue

If a person refers to a newspaper article, it's useful to be a little

more specific about whether one read it in its paper form or online

in electronic form.  I think that we're familiar here with how to

document citations and what makes for a useful reference.

 

That being said, I am not sure if the author below was referring to

the following article or not:

http://www.chicagotribune.com/business/chi-sun-nie-spss-sep27,0,3297617.story

 

The reason why I am uncertain is because the author below says the SPSS

started in Chicago while the article says:

 

|Nie worked with two friends to create the software in 1960s

|while he was a graduate student at Stanford University. In 1968

|Nie moved to the political science department at the University of Chicago

|and brought the software and one of his friends, C. Hadlai "Tex" Hull,

|with him.

 

The actual software was started at Stanford while the company was started

in Chicago.  Quoting the Tribune article:

 

|Together they refined the software and distributed it to other universities.
|
|Soon the software business became too big for the university, and

|Nie and Hull formed a separate company called SPSS in 1975.

|Nie and Hull registered as owners of the "SPSS" trademark and

|licensed the mark to the company. The agreement provided the

|company with an exclusive, perpetual, worldwide, royalty-free license.

 

In any event, SPSS closed on Friday at $49.91 and IBM closed at $121.08

(according to finance.yahoo.com ).

 

SPSS paid no dividend while IBM pays a 1.80% dividend.  Some might say

that buying SPSS stock now would a cheap way to buy IBM and get a

dividend as a bonus.  Note that the sale of SPSS to IBM will take place

on October 2, 2009.  See:

http://news.yahoo.com/s/ap/20090828/ap_on_bi_ge/us_spss_ibm

http://www.chicagobusiness.com/cgi-bin/news.pl?id=35296&seenIt=1

 

Even if SPSS goes the way of Lotus, you'll still have the IBM stock.

 

-Mike Palij

New York University

[hidden email]

 

 

 

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

From: [hidden email]

To: [hidden email]

Sent: Monday, September 28, 2009 6:23 AM

Subject: Re: Can we "freeze" at SPSS 15 for several years? Epilogue

Thought everyone would be interested, there was a article in yeserday's Chicago Tribune, discussing SPSS's potential sale to IBM, and some background on the name change.  SPSS was started here in Chicago. 

----- Original Message -----
From: "Marta García-Granero" <[hidden email]>
To: [hidden email]
Sent: Monday, September 28, 2009 2:42:24 AM GMT -06:00 US/Canada Central
Subject: Re: Can we "freeze" at SPSS 15 for several years? Epilogue

Hi everybody:

To close this active thread, this is the result of our meeting at the
University. We all agreed (more or less) that:

- We wanted SPSS 15, for teaching purposes, for at least 3 years
- If we could not have it, then we would move to Stata
(teaching&research) and R (research). My particular vote was I would go
on using PASW for my work, but I would also learn Stata and R (since I
can benefit from the PASW+R combination)

Curiously, part of the problem was they thought Stata was superior to
SPSS/PASW, only because they did not know how to write macros. I gave
them a copy of my most used macros (multiple comparisons for non
parametrics, confidence intervals for proportions, medians and their
differences, boostrapping ROC AUC...) and they left the meeting think
that, after all, SPSS was not that bad... If we keep on using SPSS/PASW,
they want me to give them training on writing SPSS macros.

Unfortunately, when we contacted the local sales manager, we got no
answer to the question "Can we have SPSS 15 for 3 more years?". Their
reply was something like "yes, no, maybe, I don't know, could you repeat
the question?" (this is a joke of mine, a reference to one of my
favorite TV series, Malcolm in the middle). Now, seriously, their reply
was there would be no reply until January-February, until they got the
chance to talk with the new IBM bosses. I found that a bit surprising,
are we being bamboozled?

Therefore:

- We are going to get someone to give us a good R course.
- We are going to start talking to Stata people, just in case (to get a
couple a licenses to start playing with the program, get some training
from them).
- We will get some StatTransfer licenses as well (to move old SPSS
databases to Stata)
- We are contacting research centers and hospitals in Spain to find out
what they use and what they plan to use in the future (to move with the
stream)

I am afraid that we will get a NO from SPSS Spain in February (it has
simply been put off), and I will have to switch to Stata (at least all
my teaching material will).

Regards,
Marta

--
For miscellaneous SPSS related statistical stuff, visit:
http://gjyp.nl/marta/

=====================
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]



This is an interesting bit of history.  To add a dimension to it, I started

using SPSS around 1977 in grad school at SUNY-Stony Brook but the

mainframe in use was a SPERRY/UNIVAC 1100.  Checking the Appendices

of the 1975 SPSS manual (the 2nd Ed "Purple" manual by Nie, Hull,

Jenkins, Steinbreener & Bent) shows that by this time SPSS had been

ported to the IBM OS/370, CDC 6000 and Cyber 70, Univac, and Xerox.

I assume that work was going on to port SPSS to DEC machines, if

not the PDP series, certainly the VAX series.  By the late 1980s,

I would use SPSS on the VAX, IBM mainframes running WYLBUR,

SuperWYLBUR, and the CMS terminal (a colleague referred to these

as the "Checker cabs" of computer terminals because they were

so much larger than other terminals, especially the ADM 3a).

 

And then there were the PC versions.  I particularly like the OS/2 version

because it had capabilities more comparable to mainframe versions

(but with a GUI) than the SPSS/PC+ which seemed more like a mutant

offspring of SPSS.  But that version died when IBM decided to kill

the OS/2 operating system (or am I mistaken and is IBM still using

OS/2 is some other guise somewhere in the world?).

 

I assume that it is probably just a couple of years before SPSS/PASW

is only available as a cloud computing "service"and PCs will only have

a simplified client interface with limited capabilities.  It may just be me but

the whole cloud computing metaphor seems to be a throwback to the

mainframe days.  I only hope that when it come about, it won't have JCL. :-)

 

-Mike Palij

New York University

[hidden email]

 

 

 

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

From: [hidden email]

To: [hidden email]

Sent: Monday, September 28, 2009 8:06 AM

Subject: Re: Can we "freeze" at SPSS 15 for several years? Epilogue

Bit more on the history of SPSS.  The following is an extract from Appendix 1 of my 2006 paper Old Dog, Old Tricks. 

 

The full paper is on Old Dog, Old Tricks: Using SPSS Syntax to Avoid the Mouse Trap 

[Hall J F, 2006] Paper to ASSESS (SPSS Users) York Univ. (UK) Dec 2006

 

---

John Hall to Tony Coxon 12 June 2006

 

Tony

 

Tracked you down after all these years.  I'm doing a turn for the SPSS
users' group ASSESS in York in November and want to stick a bit in at the
beginning about how SPSS got to UK.  David Muxworthy says you brought it
over and it was first installed at Edinburgh in 1970.

I've been doing a bit of work here restoring old SPSS files from SSRC Survey
Unit days and teaching materials from PNL.  Had to teach myself Windows,
Word and mouse from scratch when I offered to review Julie Pallant's SPSS
Survival Manual for the Social Research Association.  Luckily I got SPSS to
let me have an evaluation copy of the Windows version and then they gave me
a free 5-year licence (expires Sep 2007).

I can send you the review and other materials if you're interested.

Also found some old QSN material from 1974 and 1978.  Interesting reading!

 

2:        SPSS arrives in Edinburgh, 1970

 

Tony Coxon to John Hall 12 June 2006-08-31

 

Excellent to hear from you, even though the route taken by the email was circuitous! In the signature below you will find a selection of addresses, though [hidden email] is the preferred one.

I retired from Essex (with relief) in 2002 and moved here to the Isle of Islay (eight whisky distilleries ...)

…………….

 

Now to the subject matter of the email. As you (and David) rightly say 'twas I that was responsible for bringing SPSS over to the UK ... though I keep quiet about it .

 

The background is in some ways more interesting than the mere fact. You especially will remember that the late '60s were characterised by governmental protection of home computer industries (EELM/ ICT/ICL of course) and it was virtually impossible for Universities to get their hands on an IBM. Indeed, it was only those with big natural-science clout that did -- Imperial, Newcastle, Edinburgh come to mind. Equally, you will remember that in those days all the interesting new social science software emanated from the US and Michigan in particular and that it a tedious expensive job to adapt for bloody non-IBM machines, so that those Universities with an IBM were in the privileged position to be able to implement/ run such software immediately.

 

My involvement was that in 1968 I moved from Leeds and took a year as Visiting lecturer at MIT/Harvard Political Science, returning to take up a new post as Lecturer at Edinburgh (now you see the link?). During that time, the first interactive survey package was being developed at MIT (PI: Ithiel de Solla Pool), called ADMINS and also at Harvard, David Armor was developing a general package for the analysis of BOTH survey AND textual data called DATA-TEXT (see my comments at: http://www.qualitative-research.net/fqs-texte/2-05/05-2-40-e.htm ). Armor was pissed off that the Chicago lot had pinched most of his structure  and evacuated it of the qualitative component. He was not a happy bunny ... and of course "the Chicago lot" were SPSS!. However, I did take a copy of SPSS back with me from the US to Edinburgh.

 

The story continues. Edinburgh University had a standing committee for creating a new Edinburgh all-singing, all-dancing Survey Package. Tom Burns sent the new poilitically-naive Tony Coxon as the departmental representative. I couldn't believe it ... Big departments with entrenched positions (Agriculture, Economics ...) were arguing furiously and were unprepared to give an inch. In desperation, I interjected with something like " In the mean time, would it be an idea to use an existing new package until the subtleties of the Edinburgh package are decided? ... it so happens I happen to have a tape with SPSS" ... and that was it. It was agreed, and PLU (with Marjory Barritt -- remember?!-- and David Mux. ) ran with it. Nothing more was heard of the Edinburgh Package!

 

So there you have it!

 

Incidentally, I'm still persevering with MDSX -- now in a Windows incarnation. Have a look: www.newmdsx.com

 

Tell me more about yourself and your developments.

 

 

John Hall to Tony Coxon and David Muxworthy, 16 August 2006

 

I've got quite a long way with a draft presentation for the November ASSESS meeting in York.  One paragraph reads as follows:

 

SPSS was originally written in Fortran for an IBM by three postgraduate students.  It came from Chicago to Edinburgh in 1970 via Tony Coxon and was implemented at ERCC (one of the few places with an IBM) by David Muxworthy and Marjorie Barritt (thereby scotching university plans to commission a survey processing facility at great expense from scratch) and when first installed was reputedly called more times than the Fortran compiler.  Conversions to ICL followed later, but those with CDC and DEC machines got SPSS sooner.

 

Is this accurate and complete?  Please amend to suit and return to me.

 

 

David Muxworthy to John Hall 16 and 19 Aug 2006:

 

As I understand it SPSS first appeared in 1968.  The first UK installation was indeed at Edinburgh RCC in 1970, brought in on Tony  Coxon's recommendation.  I'm not sure about plans for commissioning  something from scratch.  I'll poke about in my loft to see if I've  anything to jog the memory.

Norman Nie and Dale Bent were political science postgrads at Stanford  in the late 1960s and, fed up with the 'put a 1 in column 72'  type  command language of the programs at the time, they devised a language  that a political scientist would want to write to specify an analysis.  They scraped together some funds and hired Tex Hull to  help with coding the program, which was in Fortran IV for the 360.  I 
think Tex must have been finishing his first degree at Yale or maybe a Masters at Stanford, but  I'm not  sure what he did at Stanford. Norman and Tex both moved to Chicago, Norman to the  National Opinion Research Center, Tex to the Computing Center.   (Norman was originally from St Louis, Tex from Minnesota).  Dale went  back to Alberta and, apart from having his name on some of the  manuals, dropped out of SPSS.

People got to hear about the program, which was superior in user  interface to much that was available at the time, and requested  copies.  This led to Patrick Bova, a librarian at NORC, being hired  25% of his time to act as distribution agent and Karin Steinbrenner  being hired as full time programmer.  When I visited Chicago in the  summer of 1972 this was the total staff.  I thought I was going to a large software house.  It was surprising to find it not much bigger than a one man  and a dog in a bedroom outfit (at that time at least). Tex acted largely as  advisor but was busy as associate director of the computing center.

As I remember it, Jean Jenkins was hired as programmer later in 1972  or in 1973.  She was certainly around at the SCSS planning meetings  in the summer of 1973.  The program was so successful that NORC  became wary of losing their non-profit status and strongly encouraged Norman to form a company and move out.  This happened sometime  between 1974 (when I worked with them at NORC) and 1977 (when they  had moved to an office block in downtown Chicago).

In Edinburgh the program grew to be so popular there were demands to move it to the ICL system 4 and later the ICL 2980, the IBM having  been removed by higher authority.  This led to PLU organising 
conversions to some other platforms in UK universities, notably the  ICL 1900.  SPSS themselves arranged conversions to other series,  notably the CDC 6600 at Northwestern University, just up the road  from Chicago.

 

(JFH: …..thereby scotching university plans to commission a survey  processing facility at great expense from scratch.)

You were quite right.  Attached are the relevant minutes of the  Edinburgh committee looking into this.  These appear to be the only  references to SPSS - much of the committee time was spent on building  up maths and engineering software.  The names won't mean anything to  you but will revive memories (good or bad) for Tony.

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

From: [hidden email]

To: [hidden email]

Sent: Monday, September 28, 2009 1:02 PM

Subject: Re: Can we "freeze" at SPSS 15 for several years? Epilogue

If a person refers to a newspaper article, it's useful to be a little

more specific about whether one read it in its paper form or online

in electronic form.  I think that we're familiar here with how to

document citations and what makes for a useful reference.

 

That being said, I am not sure if the author below was referring to

the following article or not:

http://www.chicagotribune.com/business/chi-sun-nie-spss-sep27,0,3297617.story

 

The reason why I am uncertain is because the author below says the SPSS

started in Chicago while the article says:

 

|Nie worked with two friends to create the software in 1960s

|while he was a graduate student at Stanford University. In 1968

|Nie moved to the political science department at the University of Chicago

|and brought the software and one of his friends, C. Hadlai "Tex" Hull,

|with him.

 

The actual software was started at Stanford while the company was started

in Chicago.  Quoting the Tribune article:

 

|Together they refined the software and distributed it to other universities.
|
|Soon the software business became too big for the university, and

|Nie and Hull formed a separate company called SPSS in 1975.

|Nie and Hull registered as owners of the "SPSS" trademark and

|licensed the mark to the company. The agreement provided the

|company with an exclusive, perpetual, worldwide, royalty-free license.

 

In any event, SPSS closed on Friday at $49.91 and IBM closed at $121.08

(according to finance.yahoo.com ).

 

SPSS paid no dividend while IBM pays a 1.80% dividend.  Some might say

that buying SPSS stock now would a cheap way to buy IBM and get a

dividend as a bonus.  Note that the sale of SPSS to IBM will take place

on October 2, 2009.  See:

http://news.yahoo.com/s/ap/20090828/ap_on_bi_ge/us_spss_ibm

http://www.chicagobusiness.com/cgi-bin/news.pl?id=35296&seenIt=1

 

Even if SPSS goes the way of Lotus, you'll still have the IBM stock.

 

-Mike Palij

New York University

[hidden email]

 

 

 

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

From: [hidden email]

To: [hidden email]

Sent: Monday, September 28, 2009 6:23 AM

Subject: Re: Can we "freeze" at SPSS 15 for several years? Epilogue

Thought everyone would be interested, there was a article in yeserday's Chicago Tribune, discussing SPSS's potential sale to IBM, and some background on the name change.  SPSS was started here in Chicago. 

----- Original Message -----
From: "Marta García-Granero" <[hidden email]>
To: [hidden email]
Sent: Monday, September 28, 2009 2:42:24 AM GMT -06:00 US/Canada Central
Subject: Re: Can we "freeze" at SPSS 15 for several years? Epilogue

Hi everybody:

To close this active thread, this is the result of our meeting at the
University. We all agreed (more or less) that:

- We wanted SPSS 15, for teaching purposes, for at least 3 years
- If we could not have it, then we would move to Stata
(teaching&research) and R (research). My particular vote was I would go
on using PASW for my work, but I would also learn Stata and R (since I
can benefit from the PASW+R combination)

Curiously, part of the problem was they thought Stata was superior to
SPSS/PASW, only because they did not know how to write macros. I gave
them a copy of my most used macros (multiple comparisons for non
parametrics, confidence intervals for proportions, medians and their
differences, boostrapping ROC AUC...) and they left the meeting think
that, after all, SPSS was not that bad... If we keep on using SPSS/PASW,
they want me to give them training on writing SPSS macros.

Unfortunately, when we contacted the local sales manager, we got no
answer to the question "Can we have SPSS 15 for 3 more years?". Their
reply was something like "yes, no, maybe, I don't know, could you repeat
the question?" (this is a joke of mine, a reference to one of my
favorite TV series, Malcolm in the middle). Now, seriously, their reply
was there would be no reply until January-February, until they got the
chance to talk with the new IBM bosses. I found that a bit surprising,
are we being bamboozled?

Therefore:

- We are going to get someone to give us a good R course.
- We are going to start talking to Stata people, just in case (to get a
couple a licenses to start playing with the program, get some training
from them).
- We will get some StatTransfer licenses as well (to move old SPSS
databases to Stata)
- We are contacting research centers and hospitals in Spain to find out
what they use and what they plan to use in the future (to move with the
stream)

I am afraid that we will get a NO from SPSS Spain in February (it has
simply been put off), and I will have to switch to Stata (at least all
my teaching material will).

Regards,
Marta

--
For miscellaneous SPSS related statistical stuff, visit:
http://gjyp.nl/marta/

=====================
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]

I suspect that Robert Marshall might have read the article in the off-line (i.e., actual newspaper) edition of the Chicago Tribune. Here’s the link to the online edition:

 

http://www.chicagotribune.com/business/chi-sun-nie-spss-sep27,0,3297617.story

 

Note that you might be asked to sign-up for the online edition (it’s currently free) to read the article.

 

Now with regard to the statement that SPSS was developed in Chicago, Mr. Marshall is way-more than half-right. The original version was developed at Stanford but for 37 of the next 39 years, it was developed in Chicago. I’d bet the original code base is about 5% of the current code base.

 

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Mike Palij
Sent: Monday, September 28, 2009 5:03 AM
To: [hidden email]
Subject: Re: Can we "freeze" at SPSS 15 for several years? Epilogue

 

If a person refers to a newspaper article, it's useful to be a little

more specific about whether one read it in its paper form or online

in electronic form.  I think that we're familiar here with how to

document citations and what makes for a useful reference.

 

That being said, I am not sure if the author below was referring to

the following article or not:

http://www.chicagotribune.com/business/chi-sun-nie-spss-sep27,0,3297617.story

 

The reason why I am uncertain is because the author below says the SPSS

started in Chicago while the article says:

 

|Nie worked with two friends to create the software in 1960s

|while he was a graduate student at Stanford University. In 1968

|Nie moved to the political science department at the University of Chicago

|and brought the software and one of his friends, C. Hadlai "Tex" Hull,

|with him.

 

The actual software was started at Stanford while the company was started

in Chicago.  Quoting the Tribune article:

 

|Together they refined the software and distributed it to other universities.
|
|Soon the software business became too big for the university, and

|Nie and Hull formed a separate company called SPSS in 1975.

|Nie and Hull registered as owners of the "SPSS" trademark and

|licensed the mark to the company. The agreement provided the

|company with an exclusive, perpetual, worldwide, royalty-free license.

 

In any event, SPSS closed on Friday at $49.91 and IBM closed at $121.08

(according to finance.yahoo.com ).

 

SPSS paid no dividend while IBM pays a 1.80% dividend.  Some might say

that buying SPSS stock now would a cheap way to buy IBM and get a

dividend as a bonus.  Note that the sale of SPSS to IBM will take place

on October 2, 2009.  See:

http://news.yahoo.com/s/ap/20090828/ap_on_bi_ge/us_spss_ibm

http://www.chicagobusiness.com/cgi-bin/news.pl?id=35296&seenIt=1

 

Even if SPSS goes the way of Lotus, you'll still have the IBM stock.

 

-Mike Palij

New York University

[hidden email]

 

 

 

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

From: [hidden email]

To: [hidden email]

Sent: Monday, September 28, 2009 6:23 AM

Subject: Re: Can we "freeze" at SPSS 15 for several years? Epilogue

 

Thought everyone would be interested, there was a article in yeserday's Chicago Tribune, discussing SPSS's potential sale to IBM, and some background on the name change.  SPSS was started here in Chicago. 

----- Original Message -----
From: "Marta García-Granero" <[hidden email]>
To: [hidden email]
Sent: Monday, September 28, 2009 2:42:24 AM GMT -06:00 US/Canada Central
Subject: Re: Can we "freeze" at SPSS 15 for several years? Epilogue

Hi everybody:

To close this active thread, this is the result of our meeting at the
University. We all agreed (more or less) that:

- We wanted SPSS 15, for teaching purposes, for at least 3 years
- If we could not have it, then we would move to Stata
(teaching&research) and R (research). My particular vote was I would go
on using PASW for my work, but I would also learn Stata and R (since I
can benefit from the PASW+R combination)

Curiously, part of the problem was they thought Stata was superior to
SPSS/PASW, only because they did not know how to write macros. I gave
them a copy of my most used macros (multiple comparisons for non
parametrics, confidence intervals for proportions, medians and their
differences, boostrapping ROC AUC...) and they left the meeting think
that, after all, SPSS was not that bad... If we keep on using SPSS/PASW,
they want me to give them training on writing SPSS macros.

Unfortunately, when we contacted the local sales manager, we got no
answer to the question "Can we have SPSS 15 for 3 more years?". Their
reply was something like "yes, no, maybe, I don't know, could you repeat
the question?" (this is a joke of mine, a reference to one of my
favorite TV series, Malcolm in the middle). Now, seriously, their reply
was there would be no reply until January-February, until they got the
chance to talk with the new IBM bosses. I found that a bit surprising,
are we being bamboozled?

Therefore:

- We are going to get someone to give us a good R course.
- We are going to start talking to Stata people, just in case (to get a
couple a licenses to start playing with the program, get some training
from them).
- We will get some StatTransfer licenses as well (to move old SPSS
databases to Stata)
- We are contacting research centers and hospitals in Spain to find out
what they use and what they plan to use in the future (to move with the
stream)

I am afraid that we will get a NO from SPSS Spain in February (it has
simply been put off), and I will have to switch to Stata (at least all
my teaching material will).

Regards,
Marta

--
For miscellaneous SPSS related statistical stuff, visit:
http://gjyp.nl/marta/

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

[Gyorgy Bea]

I keep getting the following error, when I try to save a data imported from excel:

Assertion failed!
Program:C:\Program..\SPSS\spsswin.exe
File: Z:\cs_source\Datasource\src\dictnry.cpp
Line: 150

Expression: 0<=mark

For information on how your program can cause an assertion failure, see the Visual C++ documentaion on assertion failure, see the Visual C++ documentation on asserts

Pres Retry to debug the application - JIT must be enabled



and other times for the same file the following warning:

Warning # 552
>Possibly due to another error, a procedure has defined more new variables
>than it has added to the file.  All those which have been defined but not
>added will be discarded and will be unavailable for further processing.


Could you please advise what to do? Where these errors could come from?

Thanks a lot,
Beata

[John McConnell-2]

Hi Beata

 

That is an untrapped error from SPSS. That is SPSS did not present you with an error message and you ended up with a message from windows instead.

 

This is usually about a specific combination of data, your pc environment and input (syntax or something you specified in the menus)

 

Can you share more information with us about what you were trying to do at the time?

 

Thanks

 

John

 

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Gyorgy Bea
Sent: 28 September 2009 18:54
To: [hidden email]
Subject: SPSS - error or bug (?)

 

I keep getting the following error, when I try to save a data imported from excel:

Assertion failed!
Program:C:\Program..\SPSS\spsswin.exe
File: Z:\cs_source\Datasource\src\dictnry.cpp
Line: 150

Expression: 0<=mark

For information on how your program can cause an assertion failure, see the Visual C++ documentaion on assertion failure, see the Visual C++ documentation on asserts

Pres Retry to debug the application - JIT must be enabled



and other times for the same file the following warning:

Warning # 552
>Possibly due to another error, a procedure has defined more new variables
>than it has added to the file.  All those which have been defined but not
>added will be discarded and will be unavailable for further processing.


Could you please advise what to do? Where these errors could come from?

Thanks a lot,
Beata

 

[SPSS Support]

Hello Beata,

  Which version of SPSS/PASW Statistics are you using? There was a bug in SPSS 15 that triggered these error messages. The bug, which involved very long string variables, was fixed in version 16.0. I have pasted a related resolution from our web site below my signature.

 

David Matheson

SPSS Statistical Support

 

**********************************

Resolution number: 69641  Created on: Jan 22 2007  Last Reviewed on: Jan 30 2009
Problem Subject:  Warning 552 or Assertion Failed error when saving in SPSS

Problem Description:  I'm using SPSS for Windows, and when I attempt to save a data file, I get the following errors(s), and my data is altered.

>Warning # 552
>Possibly due to another error, a procedure has defined more new
>variables than it has added to the file. All those which have been
>defined but not added will be discarded and will be unavailable for further processing.

If I am first importing or opening a file, when I attempt to save it out to an SPSS .sav file, the program crashes and I get the following error in a Windows dialog box entitled "Microsoft Visual C++ Runtime Library":

Assertion Failed!
Program: c:\program files\spss\spsswin.exe
File: Z:\cs_source\Datasource\src\dictnry.cpp
Line: 150

Expression: 0 <= mark
(Press Retry to debug the application - JIT must be enabled)

In either case, the .sav file is created, but a new variable has been unexpectedly inserted into the file, and all subsequent variables contain incorrect or incomplete data.

Resolution Description: 
These problems will occur when saving a variable that has a defined length which is a multiple of 252, excluding 252 itself (504, 756, 1008, 1260, etc). The workaround is to simply change the variable's length to any number that is not a multiple of 252. If you add one to the variable length, you can avert the problem without compromising data integrity.

This is a defect in the software and has been reported to SPSS Development and is fixed in SPSS 16.0. We apologize for any inconvenience.

 

 

---

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Gyorgy Bea
Sent: Monday, September 28, 2009 12:54 PM
To: [hidden email]
Subject: SPSS - error or bug (?)

 

I keep getting the following error, when I try to save a data imported from excel:

Assertion failed!
Program:C:\Program..\SPSS\spsswin.exe
File: Z:\cs_source\Datasource\src\dictnry.cpp
Line: 150

Expression: 0<=mark

For information on how your program can cause an assertion failure, see the Visual C++ documentaion on assertion failure, see the Visual C++ documentation on asserts

Pres Retry to debug the application - JIT must be enabled



and other times for the same file the following warning:

Warning # 552
>Possibly due to another error, a procedure has defined more new variables
>than it has added to the file.  All those which have been defined but not
>added will be discarded and will be unavailable for further processing.


Could you please advise what to do? Where these errors could come from?

Thanks a lot,
Beata

 

[Gyorgy Bea]

Hello David,

I am using SPSS 14, and what you described was exactly the problem I was facing. Changing the variable length solved the problem.

Thank you very much!

Beata

---

From: SPSS Support <[hidden email]>
To: Gyorgy Bea <[hidden email]>; [hidden email]
Sent: Monday, September 28, 2009 11:54:08 PM
Subject: RE: SPSS - error or bug (?)

Hello Beata,

  Which version of SPSS/PASW Statistics are you using? There was a bug in SPSS 15 that triggered these error messages. The bug, which involved very long string variables, was fixed in version 16.0. I have pasted a related resolution from our web site below my signature.

 

David Matheson

SPSS Statistical Support

 

**********************************

Resolution number: 69641  Created on: Jan 22 2007  Last Reviewed on: Jan 30 2009
Problem Subject:  Warning 552 or Assertion Failed error when saving in SPSS

Problem Description:  I'm using SPSS for Windows, and when I attempt to save a data file, I get the following errors(s), and my data is altered.

>Warning # 552
>Possibly due to another error, a procedure has defined more new
>variables than it has added to the file. All those which have been
>defined but not added will be discarded and will be unavailable for further processing.

If I am first importing or opening a file, when I attempt to save it out to an SPSS .sav file, the program crashes and I get the following error in a Windows dialog box entitled "Microsoft Visual C++ Runtime Library":

Assertion Failed!
Program: c:\program files\spss\spsswin.exe
File: Z:\cs_source\Datasource\src\dictnry.cpp
Line: 150

Expression: 0 <= mark
(Press Retry to debug the application - JIT must be enabled)

In either case, the .sav file is created, but a new variable has been unexpectedly inserted into the file, and all subsequent variables contain incorrect or incomplete data.

Resolution Description: 
These problems will occur when saving a variable that has a defined length which is a multiple of 252, excluding 252 itself (504, 756, 1008, 1260, etc). The workaround is to simply change the variable's length to any number that is not a multiple of 252. If you add one to the variable length, you can avert the problem without compromising data integrity.

This is a defect in the software and has been reported to SPSS Development and is fixed in SPSS 16.0. We apologize for any inconvenience.

 

 

---

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Gyorgy Bea
Sent: Monday, September 28, 2009 12:54 PM
To: [hidden email]
Subject: SPSS - error or bug (?)

 

I keep getting the following error, when I try to save a data imported from excel:

Assertion failed!
Program:C:\Program..\SPSS\spsswin.exe
File: Z:\cs_source\Datasource\src\dictnry.cpp
Line: 150

Expression: 0<=mark

For information on how your program can cause an assertion failure, see the Visual C++ documentaion on assertion failure, see the Visual C++ documentation on asserts

Pres Retry to debug the application - JIT must be enabled



and other times for the same file the following warning:

Warning # 552
>Possibly due to another error, a procedure has defined more new variables
>than it has added to the file.  All those which have been defined but not
>added will be discarded and will be unavailable for further processing.


Could you please advise what to do? Where these errors could come from?

Thanks a lot,
Beata

 

[Auberth Hurtado]

Hi,

Anybody knows how order the correlation matrix in descending order, by
example:

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

CORRELATIONS
  /VARIABLES=hiperac fam eman amig escolar acts WITH comps .

Correlation
        comps
                Pearson Sig.            (bilateral)     N
Hiperac .416(**)        .000            217
Fam             .321(**)        .000            217
Eman            .457(**)        .000            217
Amig            .298(**)        .000            217
Escolar .430(**)        .000            217
Acts            .465(**)        .000            217


And I Want:

Correlation
        comps
                Pearson Sig.            (bilateral)     N
Acts            .465(**)        .000            217
Eman            .457(**)        .000            217
Escolar .430(**)        .000            217
Hiperac .416(**)        .000            217
Fam             .321(**)        .000            217
Amig            .298(**)        .000            217

Thanks.

Auberth.

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

[Marta Garcia-Granero]

Hi Auberth

Avoid replying to a message when you want to start a new thread for a
different question. It is confusing, and causes a delay in getting an
answer to your question...

Now, here's the only solution I can think of (although I am sure there
will be a more elegant one, Python-based).

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

* Create a dataset with the output *.
PRESERVE.
SET OVars Names ONumbers Labels TVars Names TNumbers Values.
DATASET DECLARE CorrData.
OMS
 /SELECT TABLES
 /IF COMMANDS = ["Correlations"]
     SUBTYPES = ["Correlations"]
 /DESTINATION FORMAT = SAV
  OUTFILE = CorrData.
CORRELATIONS
  /VARIABLES=hiperac fam eman amig escolar acts WITH comps .
OMSEND.
RESTORE.

* Format output dataset (restructure) and sort by R (descending order) *.
DATASET ACTIVATE CorrData.
PRESERVE.
SET ERRORS=NONE RESULTS=NONE PRINTBACK=NONE.
SORT CASES BY Var1 Var2 .
CASESTOVARS
 /ID = Var1
 /INDEX = Var2
 /GROUPBY = VARIABLE .
SORT CASES BY V1(D).
VAR LABEL V1 'Correlation coefficient' /N 'N' /V2'Significance'.
FORMAT N(F8).
RESTORE.

* Report sorted values *.
SUMMARIZE
  /TABLES=Var1 v1 N v2
  /FORMAT=LIST NOCASENUM TOTAL
  /TITLE='Sorted correlations'
  /CELLS=NONE.

HTH,
Marta GG

Auberth Hurtado wrote:

> Anybody knows how order the correlation matrix in descending order, by
> example:
>
> GET
>   FILE='C:\Program Files\SPSS\anorectic.sav'.
>
> CORRELATIONS
>   /VARIABLES=hiperac fam eman amig escolar acts WITH comps .
>
> Correlation
>         comps
>                 Pearson Sig.            (bilateral)     N
> Hiperac .416(**)        .000            217
> Fam             .321(**)        .000            217
> Eman            .457(**)        .000            217
> Amig            .298(**)        .000            217
> Escolar .430(**)        .000            217
> Acts            .465(**)        .000            217
>
>
> And I Want:
>
> Correlation
>         comps
>                 Pearson Sig.            (bilateral)     N
> Acts            .465(**)        .000            217
> Eman            .457(**)        .000            217
> Escolar .430(**)        .000            217
> Hiperac .416(**)        .000            217
> Fam             .321(**)        .000            217
> Amig            .298(**)        .000            217
>
>
>

--

For miscellaneous SPSS related statistical stuff, visit:
http://gjyp.nl/marta/

=====================
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]

First, I'm changing the subject title here because your question has nothing
to do with the previous thread.

I looks to me like you have already rearranged the table by moving the
statistics dimension into the columns within the pivot table editor.
Although there is no easy way to sort this output in the table editor, you
can change the order of categories within a dimension by dragging the row or
column and dropping it in the desired position. On mouse-up you get a choice
to move the category before the current category or to swap it with the
current category. This is obviously a manual process and would have to be
done for each such table you create by hand. I'm not sure if any of the
python modules on Devcentral can do this.

You don't tell us what you intend to do with the results. Are they to be
published to others and if so, in what format? If Excel is acceptable, you
can export the table to Excel and sort there--another manual process but one
that can be captured in a macro.

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Auberth Hurtado
Sent: Wednesday, September 30, 2009 8:41 AM
To: [hidden email]
Subject: Re: Compute

Hi,

Anybody knows how order the correlation matrix in descending order, by
example:

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

CORRELATIONS
  /VARIABLES=hiperac fam eman amig escolar acts WITH comps .

Correlation
        comps
                Pearson Sig.            (bilateral)     N
Hiperac .416(**)        .000            217
Fam             .321(**)        .000            217
Eman            .457(**)        .000            217
Amig            .298(**)        .000            217
Escolar .430(**)        .000            217
Acts            .465(**)        .000            217


And I Want:

Correlation
        comps
                Pearson Sig.            (bilateral)     N
Acts            .465(**)        .000            217
Eman            .457(**)        .000            217
Escolar .430(**)        .000            217
Hiperac .416(**)        .000            217
Fam             .321(**)        .000            217
Amig            .298(**)        .000            217

Thanks.

Auberth.

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

[Auberth Hurtado]

First, my apologies for the title, I used another email and overwritten.

Indeed, what I do is export to Excel so that it can sort this table. But my
desire is to stop doing this step and look sort directly in SPSS.
My goal is to order correlations for looking at which variables are more
related to a particular variable, and it becomes complicated in tables with
many variables.

Thanks.

Auberth.

-----Mensaje original-----
De: SPSSX(r) Discussion [mailto:[hidden email]] En nombre de ViAnn
Beadle
Enviado el: Miércoles, 30 de Septiembre de 2009 10:59 a.m.
Para: [hidden email]
Asunto: Sorting and rearranging contents of pivot tables (was Compute)

First, I'm changing the subject title here because your question has nothing
to do with the previous thread.

I looks to me like you have already rearranged the table by moving the
statistics dimension into the columns within the pivot table editor.
Although there is no easy way to sort this output in the table editor, you
can change the order of categories within a dimension by dragging the row or
column and dropping it in the desired position. On mouse-up you get a choice
to move the category before the current category or to swap it with the
current category. This is obviously a manual process and would have to be
done for each such table you create by hand. I'm not sure if any of the
python modules on Devcentral can do this.

You don't tell us what you intend to do with the results. Are they to be
published to others and if so, in what format? If Excel is acceptable, you
can export the table to Excel and sort there--another manual process but one
that can be captured in a macro.

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Auberth Hurtado
Sent: Wednesday, September 30, 2009 8:41 AM
To: [hidden email]
Subject: Re: Compute

Hi,

Anybody knows how order the correlation matrix in descending order, by
example:

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

CORRELATIONS
  /VARIABLES=hiperac fam eman amig escolar acts WITH comps .

Correlation
        comps
                Pearson Sig.            (bilateral)     N
Hiperac .416(**)        .000            217
Fam             .321(**)        .000            217
Eman            .457(**)        .000            217
Amig            .298(**)        .000            217
Escolar .430(**)        .000            217
Acts            .465(**)        .000            217


And I Want:

Correlation
        comps
                Pearson Sig.            (bilateral)     N
Acts            .465(**)        .000            217
Eman            .457(**)        .000            217
Escolar .430(**)        .000            217
Hiperac .416(**)        .000            217
Fam             .321(**)        .000            217
Amig            .298(**)        .000            217

Thanks.

Auberth.

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

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

[Albert-Jan Roskam]

hi!

You could use OMS to grab the correlation matrix and load it into a new dataset and then SORT it.

Albert-Jan

--- On Wed, 9/30/09, Auberth Hurtado <[hidden email]> wrote:

> From: Auberth Hurtado <[hidden email]>
> Subject: Re: [SPSSX-L] Sorting and rearranging contents of pivot tables              (was Compute)
> To: [hidden email]
> Date: Wednesday, September 30, 2009, 6:58 PM
> First, my apologies for the title, I
> used another email and overwritten.
>
> Indeed, what I do is export to Excel so that it can sort
> this table. But my
> desire is to stop doing this step and look sort directly in
> SPSS.
> My goal is to order correlations for looking at which
> variables are more
> related to a particular variable, and it becomes
> complicated in tables with
> many variables.
>
> Thanks.
>
> Auberth.
>
> -----Mensaje original-----
> De: SPSSX(r) Discussion [mailto:[hidden email]]
> En nombre de ViAnn
> Beadle
> Enviado el: Miércoles, 30 de Septiembre de 2009 10:59
> a.m.
> Para: [hidden email]
> Asunto: Sorting and rearranging contents of pivot tables
> (was Compute)
>
> First, I'm changing the subject title here because your
> question has nothing
> to do with the previous thread.
>
> I looks to me like you have already rearranged the table by
> moving the
> statistics dimension into the columns within the pivot
> table editor.
> Although there is no easy way to sort this output in the
> table editor, you
> can change the order of categories within a dimension by
> dragging the row or
> column and dropping it in the desired position. On mouse-up
> you get a choice
> to move the category before the current category or to swap
> it with the
> current category. This is obviously a manual process and
> would have to be
> done for each such table you create by hand. I'm not sure
> if any of the
> python modules on Devcentral can do this.
>
> You don't tell us what you intend to do with the results.
> Are they to be
> published to others and if so, in what format? If Excel is
> acceptable, you
> can export the table to Excel and sort there--another
> manual process but one
> that can be captured in a macro.
>
> -----Original Message-----
> From: SPSSX(r) Discussion [mailto:[hidden email]]
> On Behalf Of
> Auberth Hurtado
> Sent: Wednesday, September 30, 2009 8:41 AM
> To: [hidden email]
> Subject: Re: Compute
>
> Hi,
>
> Anybody knows how order the correlation matrix in
> descending order, by
> example:
>
> GET
>   FILE='C:\Program Files\SPSS\anorectic.sav'.
>
> CORRELATIONS
>   /VARIABLES=hiperac fam eman amig escolar acts WITH
> comps .
>
> Correlation
>         comps
>
> Pearson Sig.
> (bilateral)     N
> Hiperac .416(**)        .000
>           217
> Fam
>    .321(**)
> .000            217
> Eman
> .457(**)        .000
>       217
> Amig
> .298(**)        .000
>       217
> Escolar .430(**)        .000
>           217
> Acts
> .465(**)        .000
>       217
>
>
> And I Want:
>
> Correlation
>         comps
>
> Pearson Sig.
> (bilateral)     N
> Acts
> .465(**)        .000
>       217
> Eman
> .457(**)        .000
>       217
> Escolar .430(**)        .000
>           217
> Hiperac .416(**)        .000
>           217
> Fam
>    .321(**)
> .000            217
> Amig
> .298(**)        .000
>       217
>
> Thanks.
>
> Auberth.
>
> =====================
> 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
>
> =====================
> 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]

If this is a one time task simply rerun the syntax with the variables in the desired order.
CORRELATIONS
  /VARIABLES=  Acts  Eman Escolar Hiperac Fam  Amig WITH comps .

Art Kendall
Social Research Consultants

Auberth Hurtado wrote:

> Hi,
>
> Anybody knows how order the correlation matrix in descending order, by
> example:
>
> GET
>   FILE='C:\Program Files\SPSS\anorectic.sav'.
>
> CORRELATIONS
>   /VARIABLES=hiperac fam eman amig escolar acts WITH comps .
>
> Correlation
>         comps
>                 Pearson Sig.            (bilateral)     N
> Hiperac .416(**)        .000            217
> Fam             .321(**)        .000            217
> Eman            .457(**)        .000            217
> Amig            .298(**)        .000            217
> Escolar .430(**)        .000            217
> Acts            .465(**)        .000            217
>
>
> And I Want:
>
> Correlation
>         comps
>                 Pearson Sig.            (bilateral)     N
> Acts            .465(**)        .000            217
> Eman            .457(**)        .000            217
> Escolar .430(**)        .000            217
> Hiperac .416(**)        .000            217
> Fam             .321(**)        .000            217
> Amig            .298(**)        .000            217
>
> Thanks.
>
> Auberth.
>
> =====================
> 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