SPSSX Discussion

Seeking Python and R Essentials for SPSS version 16

Classic

List

Threaded

14 messages Options

Patrick Burns

Seeking Python and R Essentials for SPSS version 16

Does anyone have access to the Python and R Essentials for SPSS version 16?
IBM does not distribute/support these prior to version 18, and I didn't have a
need to install them when my SPSS versions 16.0.2 was new. (I'm only just now
seeking to learn/run python scripts, and our non-profit doesn't have the
funding to buy new version of SPSS and subscribe to maintenance plans.)

Thanks in advance if you have copies of these...

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

Patrick Burns, Senior Researcher Economic Roundtable 315 W. 9th Street, Suite 502 Los Angeles, CA, 90015-4200 http://www.economicrt.org

Vlad Lopez

Re: Seeking Python and R Essentials for SPSS version 16

Hi Patrick,

I also need Python and R plug in for SPSS version 16. Did you get any help on your request? Please let me know.

Thank you,

vlad

On Mon, Jul 9, 2012 at 9:25 AM, Patrick Burns <[hidden email]> wrote:

Does anyone have access to the Python and R Essentials for SPSS version 16?
IBM does not distribute/support these prior to version 18, and I didn't have a
need to install them when my SPSS versions 16.0.2 was new. (I'm only just now
seeking to learn/run python scripts, and our non-profit doesn't have the
funding to buy new version of SPSS and subscribe to maintenance plans.)

Thanks in advance if you have copies of these...

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

Kelly Vander Ley

Automatic reply: Seeking Python and R Essentials for SPSS version 16

I will be out of the office from July 11th through July 23rd, returning on July 24th. If you need immediate assistance please call the main office number 503/223-8248 or 800/788-1887 and an office assistant will ensure that I get the message.

Kelly

Antoon Smulders

Re: Seeking Python and R Essentials for SPSS version 16

In reply to this post by Vlad Lopez

I wouldlike to express my feelings about the lack of service that IBM-PASW-SPSS gives to customers, but in order to stay decent I remain silent.

Antoon Smulders

Van: SPSSX(r) Discussion [mailto:[hidden email]] Namens Vladimir Lopez-Prado
Verzonden: woensdag 11 juli 2012 2:38
Aan: [hidden email]
Onderwerp: Re: Seeking Python and R Essentials for SPSS version 16

Hi Patrick,

I also need Python and R plug in for SPSS version 16. Did you get any help on your request? Please let me know.

Thank you,

vlad

On Mon, Jul 9, 2012 at 9:25 AM, Patrick Burns <[hidden email]> wrote:

King Douglas

Re: Seeking Python and R Essentials for SPSS version 16

Antoon,

With a few exceptions, I agree with your implied sentiments in regard to the deep and broad changes that IBM has made to the previously stellar qualities of SPSS customer service and support.

One exception I have in mind is the personal efforts of Jon Peck, who has never failed to extend himself to the benefit of all SPSS users. It is IBM policies that he did not make and cannot unilaterally change that are making the SPSS experience so frustrating for many of us.

Cheers,

King Douglas

American Airlines Customer Research

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Antoon Smulders
Sent: Wednesday, July 11, 2012 2:05 AM
To: [hidden email]
Subject: Re: Seeking Python and R Essentials for SPSS version 16

I wouldlike to express my feelings about the lack of service that IBM-PASW-SPSS gives to customers, but in order to stay decent I remain silent.

Antoon Smulders

Hi Patrick,

I also need Python and R plug in for SPSS version 16. Did you get any help on your request? Please let me know.

Thank you,

vlad

On Mon, Jul 9, 2012 at 9:25 AM, Patrick Burns <[hidden email]> wrote:

Bruce Weaver

Re: Seeking Python and R Essentials for SPSS version 16

Administrator

I bet the service is a lot better if you've forked out for the latest version with all the optional modules. ;-)

But yes, kudos to Jon, ViAnn and Rick for their contributions to this list.

King Douglas wrote

Antoon,

With a few exceptions, I agree with your implied sentiments in regard to the
deep and broad changes that IBM has made to the previously stellar qualities
of SPSS customer service and support.

One exception I have in mind is the personal efforts of Jon Peck, who has
never failed to extend himself to the benefit of all SPSS users. It is IBM
policies that he did not make and cannot unilaterally change that are making
the SPSS experience so frustrating for many of us.

Cheers,

King Douglas

American Airlines Customer Research

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Antoon Smulders
Sent: Wednesday, July 11, 2012 2:05 AM
To: [hidden email]
Subject: Re: Seeking Python and R Essentials for SPSS version 16

I wouldlike to express my feelings about the lack of service that
IBM-PASW-SPSS gives to customers, but in order to stay decent I remain
silent.

Antoon Smulders

Van: SPSSX(r) Discussion [mailto:[hidden email]] Namens Vladimir
Lopez-Prado
Verzonden: woensdag 11 juli 2012 2:38
Aan: [hidden email]
Onderwerp: Re: Seeking Python and R Essentials for SPSS version 16

Hi Patrick,

I also need Python and R plug in for SPSS version 16. Did you get any help
on your request? Please let me know.

Thank you,

vlad

On Mon, Jul 9, 2012 at 9:25 AM, Patrick Burns <[hidden email]>
wrote:

Does anyone have access to the Python and R Essentials for SPSS version 16?
IBM does not distribute/support these prior to version 18, and I didn't have
a
need to install them when my SPSS versions 16.0.2 was new. (I'm only just
now
seeking to learn/run python scripts, and our non-profit doesn't have the
funding to buy new version of SPSS and subscribe to maintenance plans.)

Thanks in advance if you have copies of these...

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

--
Bruce Weaver
bweaver@lakeheadu.ca
http://sites.google.com/a/lakeheadu.ca/bweaver/

"When all else fails, RTFM."

PLEASE NOTE THE FOLLOWING:
1. My Hotmail account is not monitored regularly. To send me an e-mail, please use the address shown above.
2. The SPSSX Discussion forum on Nabble is no longer linked to the SPSSX-L listserv administered by UGA (https://listserv.uga.edu/).

Antoon Smulders

Re: Seeking Python and R Essentials for SPSS version 16

In reply to this post by King Douglas

Hello King

I agree with you. Sorry Jon that I didn’t mention you as an exception.

I was only sympathizing with those who aren’t offered a link to the right Python version.

Cheers

Van: King Douglas [mailto:[hidden email]]
Verzonden: woensdag 11 juli 2012 12:48
Aan: Antoon Smulders; [hidden email]
Onderwerp: RE: Seeking Python and R Essentials for SPSS version 16

Antoon,

With a few exceptions, I agree with your implied sentiments in regard to the deep and broad changes that IBM has made to the previously stellar qualities of SPSS customer service and support.

Cheers,

King Douglas

American Airlines Customer Research

I wouldlike to express my feelings about the lack of service that IBM-PASW-SPSS gives to customers, but in order to stay decent I remain silent.

Antoon Smulders

Hi Patrick,

I also need Python and R plug in for SPSS version 16. Did you get any help on your request? Please let me know.

Thank you,

vlad

On Mon, Jul 9, 2012 at 9:25 AM, Patrick Burns <[hidden email]> wrote:

Art Kendall

Re: Seeking Python and R Essentials for SPSS version 16

In reply to this post by King Douglas

Definitely!

Jon goes out of his way to help.
Former SPSS employees ViAnn and David remind us of how SPSS used to relate to users.

SPSS pays for itself for large organizations that use it enough. (Lifetime cost of ownership).
However it is prohibitively costly for individual researchers/consultants, and not-for-profit (NGOs) especially in the Third World.

I was fortunate enough to buy the lifetime contract back when they were available in the mid-80s.
Otherwise, when I had a full time job I would not have had access to current SPSS for my part-time work.
Also, I retired in 2001 and do some paid consulting to offset some of my expenses for pro bono work. I spend most of my time in working with Statistics Without Borders and with the AAAS Science & Human Rights Coalition. I would not be able to use SPSS if I had not signed that contract back in the mid-80s. If I had to pay for the modules, it would cost several times my net consulting income.

SPSS seems to have forgotten that the popularity of "Business Intelligence" is a result of social scientists' and statisticians' efforts. Their big money may come from the large corporate world, but the popularity of SPSS in the academic and non-profit world is why there are people who know stat and methods well is the foundation underneath
SPSS seems to have forgotten that the software people start out with is what they tend to stick with.
SPSS seems to have forgotten that customers are critical in the continuous improvement of their products.

Art Kendall
Social Research Consultants

On 7/11/2012 6:48 AM, King Douglas wrote:

Antoon,

With a few exceptions, I agree with your implied sentiments in regard to the deep and broad changes that IBM has made to the previously stellar qualities of SPSS customer service and support.

One exception I have in mind is the personal efforts of Jon Peck, who has never failed to extend himself to the benefit of all SPSS users. It is IBM policies that he did not make and cannot unilaterally change that are making the SPSS experience so frustrating for many of us.

Cheers,

King Douglas

American Airlines Customer Research

From: SPSSX(r) Discussion [[hidden email]] On Behalf Of Antoon Smulders
Sent: Wednesday, July 11, 2012 2:05 AM
To: [hidden email]
Subject: Re: Seeking Python and R Essentials for SPSS version 16

I wouldlike to express my feelings about the lack of service that IBM-PASW-SPSS gives to customers, but in order to stay decent I remain silent.

Antoon Smulders

Van: SPSSX(r) Discussion [[hidden email]] Namens Vladimir Lopez-Prado
Verzonden: woensdag 11 juli 2012 2:38
Aan: [hidden email]
Onderwerp: Re: Seeking Python and R Essentials for SPSS version 16

Hi Patrick,

I also need Python and R plug in for SPSS version 16. Did you get any help on your request? Please let me know.

Thank you,

vlad

On Mon, Jul 9, 2012 at 9:25 AM, Patrick Burns <[hidden email]> wrote:

Does anyone have access to the Python and R Essentials for SPSS version 16?
IBM does not distribute/support these prior to version 18, and I didn't have a
need to install them when my SPSS versions 16.0.2 was new. (I'm only just now
seeking to learn/run python scripts, and our non-profit doesn't have the
funding to buy new version of SPSS and subscribe to maintenance plans.)

Thanks in advance if you have copies of these...

=====================
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
Social Research Consultants

Kurt Wilkening

SIGNOFF SPSSX-L

Thomas MacFarland

Re: Seeking Python and R Essentials for SPSS version 16

In reply to this post by Art Kendall

Everyone:

I’ve followed this thread and of course there are many different perspectives here.

SPSS is grand but unfortunately the cost is soon approaching a grand.

So, I’m now introducing R (standalone use of syntax and use of the R Commander GUI) into my graduate-level statistics classes. Have others had success with R, in reaction to the cost of SPSS and other proprietary software?

It is always great to have options.

Best wishes.

Tom

----------

Thomas W. MacFarland, Ed.D.

Senior Research Associate; Institutional Effectiveness and Associate Professor

Nova Southeastern University

Voice 954-262-5395 [hidden email]

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Art Kendall
Sent: Wednesday, July 11, 2012 9:09 AM
To: [hidden email]
Subject: Re: Seeking Python and R Essentials for SPSS version 16

Art Kendall

Social Research Consultants

On 7/11/2012 6:48 AM, King Douglas wrote:

Antoon,

With a few exceptions, I agree with your implied sentiments in regard to the deep and broad changes that IBM has made to the previously stellar qualities of SPSS customer service and support.

One exception I have in mind is the personal efforts of Jon Peck, who has never failed to extend himself to the benefit of all SPSS users. It is IBM policies that he did not make and cannot unilaterally change that are making the SPSS experience so frustrating for many of us.

Cheers,

King Douglas
American Airlines Customer Research

From: SPSSX(r) Discussion [[hidden email]] On Behalf Of Antoon Smulders
Sent: Wednesday, July 11, 2012 2:05 AM
To: [hidden email]
Subject: Re: Seeking Python and R Essentials for SPSS version 16

I wouldlike to express my feelings about the lack of service that IBM-PASW-SPSS gives to customers, but in order to stay decent I remain silent.

Antoon Smulders

Van: SPSSX(r) Discussion [[hidden email]] Namens Vladimir Lopez-Prado
Verzonden: woensdag 11 juli 2012 2:38
Aan: [hidden email]
Onderwerp: Re: Seeking Python and R Essentials for SPSS version 16

Hi Patrick,
I also need Python and R plug in for SPSS version 16. Did you get any help on your request? Please let me know.
Thank you,

vlad
On Mon, Jul 9, 2012 at 9:25 AM, Patrick Burns <[hidden email]> wrote:
Does anyone have access to the Python and R Essentials for SPSS version 16?
IBM does not distribute/support these prior to version 18, and I didn't have a
need to install them when my SPSS versions 16.0.2 was new. (I'm only just now
seeking to learn/run python scripts, and our non-profit doesn't have the
funding to buy new version of SPSS and subscribe to maintenance plans.)

Thanks in advance if you have copies of these...

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

J. R. Carroll

Re: Seeking Python and R Essentials for SPSS version 16

Tom,

I've used R (w/ R Commander) on/off for the last few years. I've also taught (albeit as just a part-time instructor via the graduate program) at CSUS for just over 3 years.

Additionally, I'm a pretty active Python user - so I'm not intimidated by scripting and getting the job done without using a GUI.

I mention all these things to qualify what I am about to say:

Our department (a few of the indus-org/psychology faculty members) tried to switch over one year, but the learning curve was so steep in R for people who have never done any scripting in their careers that they couldn't adapt. The graduate courses from my alma mater at CSU Sacramento did a lot of advanced SPSS analyses that may or may not be readily available in R. R just didn't fit with the curriculum of the programs we had. At least we couldn't get it to work easily enough for how we wanted to teach over the course of the summer break.

I decided NOT to teach my students "R". R is used quite a bit, but you can't deny that SPSS is used more. The issue I've heard (and seen) for the last two decades is that higher-education is losing ground in providing real-world skills to the students. I don't see R being used in the "general business place" (at least not easily, and not soon). Yes... I know that businesses use it - some big ones too! But, R is more the exception than the rule - the rule being SPSS/SAS. Your graduate students are more likely to leave the program and get jobs where SPSS/SAS are used, and quite frankly if I got an employee that was trained ONLY in R, while I would be impressed, I would expect them to use SPSS along with the other employees that I work with - and if they couldn't "hit the ground running" it would be hard for me to keep the pace in my department without halting things to train them on something I felt that a graduate program should have taught them (potentially specious, I know).

Another way to look at it is that even with R-Commander you still need to know quite a bit about R-scripting to unlock a lot of the features. R-commander is like the having the SPSS undergrad pack (which doesn't cost nearly a grand...) - lots of potential power but all locked away! Pro's and con's to both software packages. But R definitely requires a very apt individual that can climb that very steep learning curve. SPSS is very... "user friendly" to put it nicely. My experience has been that graduate students are more inclined to use (and pay) for the "user friendly" experience so they can learn the material of the course and not spend their time wrestling with how to use "R" for that one rogue instructor that insists on using software that other departments and instructors are not using (whatever the reason!).

I applaud the sentiment (as I think a lot of us have considered it, but few have made the transition), but in this case you would be part of the academic few where you are ignoring pragmatics of learning/teaching SPSS just to spite the cost (UC Davis in California I know actively teaches/uses it, and CSU Sacramento ----- SOME departments are using it, others were not. MIT for sure... not sure what other institutions use it, google it, I'm sure R has compiled a list of prominent users).

Now that I've said my soapbox on the topic as an SPSS fan-boy... I am trying to get my entire department to switch over to R just because our user licenses are so expensive for our department and we use SPSS infrequently enough (save myself) that it is offsetting our potential gains - which makes our department look less profitable. But this process has to be unilaterally made, my peers need to make a commitment to switch over, and we have to spend substantial money (a la man-hours) on having our employees rewrite our scripts into R.

I, on the other-hand, will remain an SPSS user just because it is my default at this stage in my career.

So...

No -> I wouldn't recommend punishing your students with R and R-Commander (especially if you are not great at using R without R-Commander yourself). Give your students skills that they can take into the work place that they can use to pay off their student debt =P. If anything, try to teach both of them! Or let them choose (but that can be a nightmare where some students are using R and others are using SPSS).

Yes -> SPSS is too expensive these days, most PC (non-mobile) software is (not trying to justify it). R is definitely an alternative that businesses and individuals should consider if the price-point is important enough to them.

HTH,

----

J. R. Carroll

Cell: (650) 776-6613

Email: [hidden email]

[hidden email]

On Wed, Jul 11, 2012 at 6:32 AM, Dr. Thomas W. MacFarland <[hidden email]> wrote:

Everyone:

I’ve followed this thread and of course there are many different perspectives here.

SPSS is grand but unfortunately the cost is soon approaching a grand.

So, I’m now introducing R (standalone use of syntax and use of the R Commander GUI) into my graduate-level statistics classes. Have others had success with R, in reaction to the cost of SPSS and other proprietary software?

It is always great to have options.

Best wishes.

Tom

----------

Thomas W. MacFarland, Ed.D.
Senior Research Associate; Institutional Effectiveness and Associate Professor

Nova Southeastern University
Voice <a href="tel:954-262-5395" value="+19542625395" target="_blank">954-262-5395 [hidden email]

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Art Kendall
Sent: Wednesday, July 11, 2012 9:09 AM

To: [hidden email]
Subject: Re: Seeking Python and R Essentials for SPSS version 16
Definitely!

Jon goes out of his way to help.
Former SPSS employees ViAnn and David remind us of how SPSS used to relate to users.

SPSS pays for itself for large organizations that use it enough. (Lifetime cost of ownership).
However it is prohibitively costly for individual researchers/consultants, and not-for-profit (NGOs) especially in the Third World.

I was fortunate enough to buy the lifetime contract back when they were available in the mid-80s.
Otherwise, when I had a full time job I would not have had access to current SPSS for my part-time work.
Also, I retired in 2001 and do some paid consulting to offset some of my expenses for pro bono work. I spend most of my time in working with Statistics Without Borders and with the AAAS Science & Human Rights Coalition. I would not be able to use SPSS if I had not signed that contract back in the mid-80s. If I had to pay for the modules, it would cost several times my net consulting income.

SPSS seems to have forgotten that the popularity of "Business Intelligence" is a result of social scientists' and statisticians' efforts. Their big money may come from the large corporate world, but the popularity of SPSS in the academic and non-profit world is why there are people who know stat and methods well is the foundation underneath
SPSS seems to have forgotten that the software people start out with is what they tend to stick with.
SPSS seems to have forgotten that customers are critical in the continuous improvement of their products.
Art Kendall
Social Research Consultants
On 7/11/2012 6:48 AM, King Douglas wrote:
Antoon,

With a few exceptions, I agree with your implied sentiments in regard to the deep and broad changes that IBM has made to the previously stellar qualities of SPSS customer service and support.

One exception I have in mind is the personal efforts of Jon Peck, who has never failed to extend himself to the benefit of all SPSS users. It is IBM policies that he did not make and cannot unilaterally change that are making the SPSS experience so frustrating for many of us.

Cheers,

King Douglas

American Airlines Customer Research

From: SPSSX(r) Discussion [[hidden email]] On Behalf Of Antoon Smulders
Sent: Wednesday, July 11, 2012 2:05 AM
To: [hidden email]
Subject: Re: Seeking Python and R Essentials for SPSS version 16

I wouldlike to express my feelings about the lack of service that IBM-PASW-SPSS gives to customers, but in order to stay decent I remain silent.

Antoon Smulders

Van: SPSSX(r) Discussion [[hidden email]] Namens Vladimir Lopez-Prado
Verzonden: woensdag 11 juli 2012 2:38
Aan: [hidden email]
Onderwerp: Re: Seeking Python and R Essentials for SPSS version 16

Hi Patrick,
I also need Python and R plug in for SPSS version 16. Did you get any help on your request? Please let me know.

Thank you,

vlad

On Mon, Jul 9, 2012 at 9:25 AM, Patrick Burns <[hidden email]> wrote:
Does anyone have access to the Python and R Essentials for SPSS version 16?
IBM does not distribute/support these prior to version 18, and I didn't have a
need to install them when my SPSS versions 16.0.2 was new. (I'm only just now
seeking to learn/run python scripts, and our non-profit doesn't have the
funding to buy new version of SPSS and subscribe to maintenance plans.)

Thanks in advance if you have copies of these...

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

Re: Seeking Python and R Essentials for SPSS version 16

I would not want to try to do all of the preparation that is required before doing analysis.
My subjective estimate is that usually more that 85% of the staff time is in getting the data ready. SPSS surpasses other packages at the nitty-gritty that requires so much of the effort. (Labeling, double entry, frequencies, descriptive, box plots, scatter diagrams, etc.)

Purchase price is only part of the total monetary costs of ownership. The lost opportunity to teach about meaningful analysis is dreadful.

With too steep a learning curve in an intro course, people get turned off methods and stat.

YMMV but few beginners need anything outside the base module. Particularly when people are only going to have one or two courses, I emphasize understanding the data, how to do some quality assurance on the analytic process, etc. My goal is to bring them to a point where they can work effectively and efficiently with consultants, not to give them a graduate education and years of experience in a course or two.

It is important to understand what is done to variables, but without understanding the meaning (constructs) analysis is seen as sterile. I have seen numerous times where failure to do all of the labeling etc has led to lost time and wrong conclusions.
Emphasizing getting the metadata into the variables view is important in introductory courses so that users can relate what they are doing to why they would do things in the first place.

What I have long suggested to clients is to have the Base module widely available, with a few seats that have other modules as needed. Once the data is fully prepared in SPSS, it is easy to pass it to a desk that has advanced modules, R, SAS, etc.

WRT only a few members in your department using SPSS, that is often a problem of stove-piping in schools and other organizations. God forbid that purchases help more than one department.

Art Kendall
Social Research Consultants

On 7/11/2012 1:47 PM, J. R. Carroll wrote:

Tom,

I've used R (w/ R Commander) on/off for the last few years. I've also taught (albeit as just a part-time instructor via the graduate program) at CSUS for just over 3 years.

Additionally, I'm a pretty active Python user - so I'm not intimidated by scripting and getting the job done without using a GUI.

I mention all these things to qualify what I am about to say:

Our department (a few of the indus-org/psychology faculty members) tried to switch over one year, but the learning curve was so steep in R for people who have never done any scripting in their careers that they couldn't adapt. The graduate courses from my alma mater at CSU Sacramento did a lot of advanced SPSS analyses that may or may not be readily available in R. R just didn't fit with the curriculum of the programs we had. At least we couldn't get it to work easily enough for how we wanted to teach over the course of the summer break.

I decided NOT to teach my students "R". R is used quite a bit, but you can't deny that SPSS is used more. The issue I've heard (and seen) for the last two decades is that higher-education is losing ground in providing real-world skills to the students. I don't see R being used in the "general business place" (at least not easily, and not soon). Yes... I know that businesses use it - some big ones too! But, R is more the exception than the rule - the rule being SPSS/SAS. Your graduate students are more likely to leave the program and get jobs where SPSS/SAS are used, and quite frankly if I got an employee that was trained ONLY in R, while I would be impressed, I would expect them to use SPSS along with the other employees that I work with - and if they couldn't "hit the ground running" it would be hard for me to keep the pace in my department without halting things to train them on something I felt that a graduate program should have taught them (potentially specious, I know).

Another way to look at it is that even with R-Commander you still need to know quite a bit about R-scripting to unlock a lot of the features. R-commander is like the having the SPSS undergrad pack (which doesn't cost nearly a grand...) - lots of potential power but all locked away! Pro's and con's to both software packages. But R definitely requires a very apt individual that can climb that very steep learning curve. SPSS is very... "user friendly" to put it nicely. My experience has been that graduate students are more inclined to use (and pay) for the "user friendly" experience so they can learn the material of the course and not spend their time wrestling with how to use "R" for that one rogue instructor that insists on using software that other departments and instructors are not using (whatever the reason!).

I applaud the sentiment (as I think a lot of us have considered it, but few have made the transition), but in this case you would be part of the academic few where you are ignoring pragmatics of learning/teaching SPSS just to spite the cost (UC Davis in California I know actively teaches/uses it, and CSU Sacramento ----- SOME departments are using it, others were not. MIT for sure... not sure what other institutions use it, google it, I'm sure R has compiled a list of prominent users).

Now that I've said my soapbox on the topic as an SPSS fan-boy... I am trying to get my entire department to switch over to R just because our user licenses are so expensive for our department and we use SPSS infrequently enough (save myself) that it is offsetting our potential gains - which makes our department look less profitable. But this process has to be unilaterally made, my peers need to make a commitment to switch over, and we have to spend substantial money (a la man-hours) on having our employees rewrite our scripts into R.

I, on the other-hand, will remain an SPSS user just because it is my default at this stage in my career.

So...

No -> I wouldn't recommend punishing your students with R and R-Commander (especially if you are not great at using R without R-Commander yourself). Give your students skills that they can take into the work place that they can use to pay off their student debt =P. If anything, try to teach both of them! Or let them choose (but that can be a nightmare where some students are using R and others are using SPSS).

Yes -> SPSS is too expensive these days, most PC (non-mobile) software is (not trying to justify it). R is definitely an alternative that businesses and individuals should consider if the price-point is important enough to them.

HTH,
----

J. R. Carroll

Cell: (650) 776-6613

Email: [hidden email]

[hidden email]

[hidden email]

[hidden email]
On Wed, Jul 11, 2012 at 6:32 AM, Dr. Thomas W. MacFarland <[hidden email]> wrote:
Everyone:

I’ve followed this thread and of course there are many different perspectives here.

SPSS is grand but unfortunately the cost is soon approaching a grand.

So, I’m now introducing R (standalone use of syntax and use of the R Commander GUI) into my graduate-level statistics classes. Have others had success with R, in reaction to the cost of SPSS and other proprietary software?

It is always great to have options.

Best wishes.

Tom

----------

Thomas W. MacFarland, Ed.D.

Senior Research Associate; Institutional Effectiveness and Associate Professor

Nova Southeastern University

Voice <a moz-do-not-send="true" href="tel:954-262-5395" value="+19542625395" target="_blank">954-262-5395 [hidden email]

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Art Kendall
Sent: Wednesday, July 11, 2012 9:09 AM

To: [hidden email]
Subject: Re: Seeking Python and R Essentials for SPSS version 16
Definitely!

Jon goes out of his way to help.
Former SPSS employees ViAnn and David remind us of how SPSS used to relate to users.

SPSS pays for itself for large organizations that use it enough. (Lifetime cost of ownership).
However it is prohibitively costly for individual researchers/consultants, and not-for-profit (NGOs) especially in the Third World.

I was fortunate enough to buy the lifetime contract back when they were available in the mid-80s.
Otherwise, when I had a full time job I would not have had access to current SPSS for my part-time work.
Also, I retired in 2001 and do some paid consulting to offset some of my expenses for pro bono work. I spend most of my time in working with Statistics Without Borders and with the AAAS Science & Human Rights Coalition. I would not be able to use SPSS if I had not signed that contract back in the mid-80s. If I had to pay for the modules, it would cost several times my net consulting income.

SPSS seems to have forgotten that the popularity of "Business Intelligence" is a result of social scientists' and statisticians' efforts. Their big money may come from the large corporate world, but the popularity of SPSS in the academic and non-profit world is why there are people who know stat and methods well is the foundation underneath
SPSS seems to have forgotten that the software people start out with is what they tend to stick with.
SPSS seems to have forgotten that customers are critical in the continuous improvement of their products.
Art Kendall
Social Research Consultants
On 7/11/2012 6:48 AM, King Douglas wrote:
Antoon,

With a few exceptions, I agree with your implied sentiments in regard to the deep and broad changes that IBM has made to the previously stellar qualities of SPSS customer service and support.

One exception I have in mind is the personal efforts of Jon Peck, who has never failed to extend himself to the benefit of all SPSS users. It is IBM policies that he did not make and cannot unilaterally change that are making the SPSS experience so frustrating for many of us.

Cheers,

King Douglas

American Airlines Customer Research

From: SPSSX(r) Discussion [[hidden email]] On Behalf Of Antoon Smulders
Sent: Wednesday, July 11, 2012 2:05 AM
To: [hidden email]
Subject: Re: Seeking Python and R Essentials for SPSS version 16

I wouldlike to express my feelings about the lack of service that IBM-PASW-SPSS gives to customers, but in order to stay decent I remain silent.

Antoon Smulders

Van: SPSSX(r) Discussion [[hidden email]] Namens Vladimir Lopez-Prado
Verzonden: woensdag 11 juli 2012 2:38
Aan: [hidden email]
Onderwerp: Re: Seeking Python and R Essentials for SPSS version 16

Hi Patrick,

I also need Python and R plug in for SPSS version 16. Did you get any help on your request? Please let me know.

Thank you,

vlad

On Mon, Jul 9, 2012 at 9:25 AM, Patrick Burns <[hidden email]> wrote:

Does anyone have access to the Python and R Essentials for SPSS version 16?
IBM does not distribute/support these prior to version 18, and I didn't have a
need to install them when my SPSS versions 16.0.2 was new. (I'm only just now
seeking to learn/run python scripts, and our non-profit doesn't have the
funding to buy new version of SPSS and subscribe to maintenance plans.)

Thanks in advance if you have copies of these...

=====================
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
Social Research Consultants

Art Kendall

Re: Seeking Python and R Essentials for SPSS version 16

I agree that it is useful to know more than one package especially when stat is partly a goal in itself. Over the years I have used SIFT, BMDP, P-stat, APL-Pack, SAS, SPSS, IMSL, ad hoc programs, and written FORTRAN.
[I still have my VAX powered up in my basement, but have not used it in years.]

When I advise clients to base their work on SPSS, I also explain that they can use R for esoteric stat procedures. I also explain that many other applications and packages can read SPSS files. I also recommend saving important files in other data formats and in other places.

I am a political/social psychologist as well as a math statistician. I concentrate mostly on methods in national and international policy and social issues. My aim is usually to make it as easy as possible to do all of the preparation work. In my experience, emphasizing the metadata and data prep helps students understand the relation of the stat methods to the subject matter they are dealing with.

Also, in the kind of work I deal with due diligence requires quality assurance review by at least one other person. It takes a lot less additional writing to show why the substantive conclusions are supported with SPSS. In a few comparisons many years ago, it took 15% more staff time to do and review the same analysis is SAS as in SPSS.

I have not worked extensively with R. However, the few times I used it there was something in the output that said that the software was developmental or "experimental". External reviewers in adversary situations would pick right up on that.

Art Kendall
Social Research Consultants

On 7/12/2012 9:29 AM, Dr. Thomas W. MacFarland wrote:

Dear Art and Justin:

Let me take this opportunity to thank both of you for your detailed responses to my prior post about R.

I work in Institutional Research and some jobs in our office are based on 20+ year-old processes of using SPSS (going back to SPSS on VAX machines) and I would not want to start over again with a new platform for these jobs. We don’t need another Y2K conversion effort.

However, whenever I have a totally project, with no existing procedures to draw from, I try to see if R can be used as an alternate. Although IBM is a great company I do not want to put all of my eggs in one basket – thinking back to what they did with Lotus 1-2-3 when they gained control of that software package.

If you can, look at my approach to the use of R by reviewing the table of contents to my book (http://www.springer.com/statistics/social+sciences+%26+law/book/978-1-4614-2133-7) on how R is used with Twoway ANOVA:

Download Table of contents (pdf, 65 kB)

Download Sample pages 1 (pdf, 181 kB)

I certainly agree that R is not for everyone, but it has a role with those students who have some degree of familiarity with programming languages and are willing to try new approaches.

Best wishes.

Tom

----------

Thomas W. MacFarland, Ed.D.

Senior Research Associate; Institutional Effectiveness and Associate Professor

Nova Southeastern University

Voice 954-262-5395 [hidden email]

From: SPSSX(r) Discussion [[hidden email]] On Behalf Of Art Kendall
Sent: Wednesday, July 11, 2012 5:55 PM
To: [hidden email]
Subject: Re: Seeking Python and R Essentials for SPSS version 16
I would not want to try to do all of the preparation that is required before doing analysis.
My subjective estimate is that usually more that 85% of the staff time is in getting the data ready. SPSS surpasses other packages at the nitty-gritty that requires so much of the effort. (Labeling, double entry, frequencies, descriptive, box plots, scatter diagrams, etc.)

Purchase price is only part of the total monetary costs of ownership. The lost opportunity to teach about meaningful analysis is dreadful.

With too steep a learning curve in an intro course, people get turned off methods and stat.

YMMV but few beginners need anything outside the base module. Particularly when people are only going to have one or two courses, I emphasize understanding the data, how to do some quality assurance on the analytic process, etc. My goal is to bring them to a point where they can work effectively and efficiently with consultants, not to give them a graduate education and years of experience in a course or two.

It is important to understand what is done to variables, but without understanding the meaning (constructs) analysis is seen as sterile. I have seen numerous times where failure to do all of the labeling etc has led to lost time and wrong conclusions.
Emphasizing getting the metadata into the variables view is important in introductory courses so that users can relate what they are doing to why they would do things in the first place.

What I have long suggested to clients is to have the Base module widely available, with a few seats that have other modules as needed. Once the data is fully prepared in SPSS, it is easy to pass it to a desk that has advanced modules, R, SAS, etc.

WRT only a few members in your department using SPSS, that is often a problem of stove-piping in schools and other organizations. God forbid that purchases help more than one department.
Art Kendall
Social Research Consultants
On 7/11/2012 1:47 PM, J. R. Carroll wrote:
Tom,

I've used R (w/ R Commander) on/off for the last few years. I've also taught (albeit as just a part-time instructor via the graduate program) at CSUS for just over 3 years.

Additionally, I'm a pretty active Python user - so I'm not intimidated by scripting and getting the job done without using a GUI.

I mention all these things to qualify what I am about to say:

Our department (a few of the indus-org/psychology faculty members) tried to switch over one year, but the learning curve was so steep in R for people who have never done any scripting in their careers that they couldn't adapt. The graduate courses from my alma mater at CSU Sacramento did a lot of advanced SPSS analyses that may or may not be readily available in R. R just didn't fit with the curriculum of the programs we had. At least we couldn't get it to work easily enough for how we wanted to teach over the course of the summer break.

I decided NOT to teach my students "R". R is used quite a bit, but you can't deny that SPSS is used more. The issue I've heard (and seen) for the last two decades is that higher-education is losing ground in providing real-world skills to the students. I don't see R being used in the "general business place" (at least not easily, and not soon). Yes... I know that businesses use it - some big ones too! But, R is more the exception than the rule - the rule being SPSS/SAS. Your graduate students are more likely to leave the program and get jobs where SPSS/SAS are used, and quite frankly if I got an employee that was trained ONLY in R, while I would be impressed, I would expect them to use SPSS along with the other employees that I work with - and if they couldn't "hit the ground running" it would be hard for me to keep the pace in my department without halting things to train them on something I felt that a graduate program should have taught them (potentially specious, I know).

Another way to look at it is that even with R-Commander you still need to know quite a bit about R-scripting to unlock a lot of the features. R-commander is like the having the SPSS undergrad pack (which doesn't cost nearly a grand...) - lots of potential power but all locked away! Pro's and con's to both software packages. But R definitely requires a very apt individual that can climb that very steep learning curve. SPSS is very... "user friendly" to put it nicely. My experience has been that graduate students are more inclined to use (and pay) for the "user friendly" experience so they can learn the material of the course and not spend their time wrestling with how to use "R" for that one rogue instructor that insists on using software that other departments and instructors are not using (whatever the reason!).

I applaud the sentiment (as I think a lot of us have considered it, but few have made the transition), but in this case you would be part of the academic few where you are ignoring pragmatics of learning/teaching SPSS just to spite the cost (UC Davis in California I know actively teaches/uses it, and CSU Sacramento ----- SOME departments are using it, others were not. MIT for sure... not sure what other institutions use it, google it, I'm sure R has compiled a list of prominent users).

Now that I've said my soapbox on the topic as an SPSS fan-boy... I am trying to get my entire department to switch over to R just because our user licenses are so expensive for our department and we use SPSS infrequently enough (save myself) that it is offsetting our potential gains - which makes our department look less profitable. But this process has to be unilaterally made, my peers need to make a commitment to switch over, and we have to spend substantial money (a la man-hours) on having our employees rewrite our scripts into R.

I, on the other-hand, will remain an SPSS user just because it is my default at this stage in my career.

So...

No -> I wouldn't recommend punishing your students with R and R-Commander (especially if you are not great at using R without R-Commander yourself). Give your students skills that they can take into the work place that they can use to pay off their student debt =P. If anything, try to teach both of them! Or let them choose (but that can be a nightmare where some students are using R and others are using SPSS).

Yes -> SPSS is too expensive these days, most PC (non-mobile) software is (not trying to justify it). R is definitely an alternative that businesses and individuals should consider if the price-point is important enough to them.

HTH,
----

J. R. Carroll

Cell: (650) 776-6613

Email: [hidden email]

[hidden email]

[hidden email]

[hidden email]
On Wed, Jul 11, 2012 at 6:32 AM, Dr. Thomas W. MacFarland <[hidden email]> wrote:
Everyone:

I’ve followed this thread and of course there are many different perspectives here.

SPSS is grand but unfortunately the cost is soon approaching a grand.

So, I’m now introducing R (standalone use of syntax and use of the R Commander GUI) into my graduate-level statistics classes. Have others had success with R, in reaction to the cost of SPSS and other proprietary software?

It is always great to have options.

Best wishes.

Tom

----------

Thomas W. MacFarland, Ed.D.

Senior Research Associate; Institutional Effectiveness and Associate Professor

Nova Southeastern University

Voice <a moz-do-not-send="true" href="tel:954-262-5395" target="_blank">954-262-5395 [hidden email]

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Art Kendall
Sent: Wednesday, July 11, 2012 9:09 AM

To: [hidden email]
Subject: Re: Seeking Python and R Essentials for SPSS version 16
Definitely!

Jon goes out of his way to help.
Former SPSS employees ViAnn and David remind us of how SPSS used to relate to users.

SPSS pays for itself for large organizations that use it enough. (Lifetime cost of ownership).
However it is prohibitively costly for individual researchers/consultants, and not-for-profit (NGOs) especially in the Third World.

I was fortunate enough to buy the lifetime contract back when they were available in the mid-80s.
Otherwise, when I had a full time job I would not have had access to current SPSS for my part-time work.
Also, I retired in 2001 and do some paid consulting to offset some of my expenses for pro bono work. I spend most of my time in working with Statistics Without Borders and with the AAAS Science & Human Rights Coalition. I would not be able to use SPSS if I had not signed that contract back in the mid-80s. If I had to pay for the modules, it would cost several times my net consulting income.

SPSS seems to have forgotten that the popularity of "Business Intelligence" is a result of social scientists' and statisticians' efforts. Their big money may come from the large corporate world, but the popularity of SPSS in the academic and non-profit world is why there are people who know stat and methods well is the foundation underneath
SPSS seems to have forgotten that the software people start out with is what they tend to stick with.
SPSS seems to have forgotten that customers are critical in the continuous improvement of their products.
Art Kendall
Social Research Consultants
On 7/11/2012 6:48 AM, King Douglas wrote:
Antoon,

With a few exceptions, I agree with your implied sentiments in regard to the deep and broad changes that IBM has made to the previously stellar qualities of SPSS customer service and support.

One exception I have in mind is the personal efforts of Jon Peck, who has never failed to extend himself to the benefit of all SPSS users. It is IBM policies that he did not make and cannot unilaterally change that are making the SPSS experience so frustrating for many of us.

Cheers,

King Douglas

American Airlines Customer Research

From: SPSSX(r) Discussion [[hidden email]] On Behalf Of Antoon Smulders
Sent: Wednesday, July 11, 2012 2:05 AM
To: [hidden email]
Subject: Re: Seeking Python and R Essentials for SPSS version 16

I wouldlike to express my feelings about the lack of service that IBM-PASW-SPSS gives to customers, but in order to stay decent I remain silent.

Antoon Smulders

Van: SPSSX(r) Discussion [[hidden email]] Namens Vladimir Lopez-Prado
Verzonden: woensdag 11 juli 2012 2:38
Aan: [hidden email]
Onderwerp: Re: Seeking Python and R Essentials for SPSS version 16

Hi Patrick,

I also need Python and R plug in for SPSS version 16. Did you get any help on your request? Please let me know.

Thank you,

vlad

On Mon, Jul 9, 2012 at 9:25 AM, Patrick Burns <[hidden email]> wrote:

Does anyone have access to the Python and R Essentials for SPSS version 16?
IBM does not distribute/support these prior to version 18, and I didn't have a
need to install them when my SPSS versions 16.0.2 was new. (I'm only just now
seeking to learn/run python scripts, and our non-profit doesn't have the
funding to buy new version of SPSS and subscribe to maintenance plans.)

Thanks in advance if you have copies of these...

=====================
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
Social Research Consultants

Art Kendall

Re: Seeking Python and R Essentials for SPSS version 16

In reply to this post by Art Kendall

I agree that it is useful to know more than one package especially when stat is partly a goal in itself. Over the years I have used SIFT, BMDP, P-stat, APL-Pack, SAS, SPSS, IMSL, ad hoc programs, and written FORTRAN.
[I still have my VAX powered up in my basement, but have not used it in years.]

When I advise clients to base their work on SPSS, I also explain that they can use R for esoteric stat procedures. I also explain that many other applications and packages can read SPSS files.

I am a political/social psychologist as well as a math statistician. I concentrate mostly on methods in national and international policy and social issues. My aim is usually to make it as easy as possible to do all of the preparation work. In my experience, emphasizing the metadata and data prep helps students understand the relation of the stat methods to the subject matter they are dealing with.

Also, in the kind of work I deal with due diligence requires quality assurance review by at least one other person. It takes a lot less additional writing to show why the substantive conclusions are supported with SPSS. In a few comparisons many years ago, it took 15% more staff time to do and review teh same analysis is SAS as in SPSS.

I have not worked extensively with R. However, the few times I used it there was something in the output that said that the software was developmental or "experimental". External reviewers in adversary situations would pick right up on that.

Art Kendall
Social Research Consultants

On 7/12/2012 9:29 AM, Dr. Thomas W. MacFarland wrote:

Dear Art and Justin:

Let me take this opportunity to thank both of you for your detailed responses to my prior post about R.

I work in Institutional Research and some jobs in our office are based on 20+ year-old processes of using SPSS (going back to SPSS on VAX machines) and I would not want to start over again with a new platform for these jobs. We don’t need another Y2K conversion effort.

However, whenever I have a totally project, with no existing procedures to draw from, I try to see if R can be used as an alternate. Although IBM is a great company I do not want to put all of my eggs in one basket – thinking back to what they did with Lotus 1-2-3 when they gained control of that software package.

If you can, look at my approach to the use of R by reviewing the table of contents to my book (http://www.springer.com/statistics/social+sciences+%26+law/book/978-1-4614-2133-7) on how R is used with Twoway ANOVA:

Download Table of contents (pdf, 65 kB)

Download Sample pages 1 (pdf, 181 kB)

I certainly agree that R is not for everyone, but it has a role with those students who have some degree of familiarity with programming languages and are willing to try new approaches.

Best wishes.

Tom

----------

Thomas W. MacFarland, Ed.D.

Senior Research Associate; Institutional Effectiveness and Associate Professor

Nova Southeastern University

Voice 954-262-5395 [hidden email]

From: SPSSX(r) Discussion [[hidden email]] On Behalf Of Art Kendall
Sent: Wednesday, July 11, 2012 5:55 PM
To: [hidden email]
Subject: Re: Seeking Python and R Essentials for SPSS version 16
I would not want to try to do all of the preparation that is required before doing analysis.
My subjective estimate is that usually more that 85% of the staff time is in getting the data ready. SPSS surpasses other packages at the nitty-gritty that requires so much of the effort. (Labeling, double entry, frequencies, descriptive, box plots, scatter diagrams, etc.)

Purchase price is only part of the total monetary costs of ownership. The lost opportunity to teach about meaningful analysis is dreadful.

With too steep a learning curve in an intro course, people get turned off methods and stat.

YMMV but few beginners need anything outside the base module. Particularly when people are only going to have one or two courses, I emphasize understanding the data, how to do some quality assurance on the analytic process, etc. My goal is to bring them to a point where they can work effectively and efficiently with consultants, not to give them a graduate education and years of experience in a course or two.

It is important to understand what is done to variables, but without understanding the meaning (constructs) analysis is seen as sterile. I have seen numerous times where failure to do all of the labeling etc has led to lost time and wrong conclusions.
Emphasizing getting the metadata into the variables view is important in introductory courses so that users can relate what they are doing to why they would do things in the first place.

What I have long suggested to clients is to have the Base module widely available, with a few seats that have other modules as needed. Once the data is fully prepared in SPSS, it is easy to pass it to a desk that has advanced modules, R, SAS, etc.

WRT only a few members in your department using SPSS, that is often a problem of stove-piping in schools and other organizations. God forbid that purchases help more than one department.
Art Kendall
Social Research Consultants
On 7/11/2012 1:47 PM, J. R. Carroll wrote:
Tom,

I've used R (w/ R Commander) on/off for the last few years. I've also taught (albeit as just a part-time instructor via the graduate program) at CSUS for just over 3 years.

Additionally, I'm a pretty active Python user - so I'm not intimidated by scripting and getting the job done without using a GUI.

I mention all these things to qualify what I am about to say:

Our department (a few of the indus-org/psychology faculty members) tried to switch over one year, but the learning curve was so steep in R for people who have never done any scripting in their careers that they couldn't adapt. The graduate courses from my alma mater at CSU Sacramento did a lot of advanced SPSS analyses that may or may not be readily available in R. R just didn't fit with the curriculum of the programs we had. At least we couldn't get it to work easily enough for how we wanted to teach over the course of the summer break.

I decided NOT to teach my students "R". R is used quite a bit, but you can't deny that SPSS is used more. The issue I've heard (and seen) for the last two decades is that higher-education is losing ground in providing real-world skills to the students. I don't see R being used in the "general business place" (at least not easily, and not soon). Yes... I know that businesses use it - some big ones too! But, R is more the exception than the rule - the rule being SPSS/SAS. Your graduate students are more likely to leave the program and get jobs where SPSS/SAS are used, and quite frankly if I got an employee that was trained ONLY in R, while I would be impressed, I would expect them to use SPSS along with the other employees that I work with - and if they couldn't "hit the ground running" it would be hard for me to keep the pace in my department without halting things to train them on something I felt that a graduate program should have taught them (potentially specious, I know).

Another way to look at it is that even with R-Commander you still need to know quite a bit about R-scripting to unlock a lot of the features. R-commander is like the having the SPSS undergrad pack (which doesn't cost nearly a grand...) - lots of potential power but all locked away! Pro's and con's to both software packages. But R definitely requires a very apt individual that can climb that very steep learning curve. SPSS is very... "user friendly" to put it nicely. My experience has been that graduate students are more inclined to use (and pay) for the "user friendly" experience so they can learn the material of the course and not spend their time wrestling with how to use "R" for that one rogue instructor that insists on using software that other departments and instructors are not using (whatever the reason!).

I applaud the sentiment (as I think a lot of us have considered it, but few have made the transition), but in this case you would be part of the academic few where you are ignoring pragmatics of learning/teaching SPSS just to spite the cost (UC Davis in California I know actively teaches/uses it, and CSU Sacramento ----- SOME departments are using it, others were not. MIT for sure... not sure what other institutions use it, google it, I'm sure R has compiled a list of prominent users).

Now that I've said my soapbox on the topic as an SPSS fan-boy... I am trying to get my entire department to switch over to R just because our user licenses are so expensive for our department and we use SPSS infrequently enough (save myself) that it is offsetting our potential gains - which makes our department look less profitable. But this process has to be unilaterally made, my peers need to make a commitment to switch over, and we have to spend substantial money (a la man-hours) on having our employees rewrite our scripts into R.

I, on the other-hand, will remain an SPSS user just because it is my default at this stage in my career.

So...

No -> I wouldn't recommend punishing your students with R and R-Commander (especially if you are not great at using R without R-Commander yourself). Give your students skills that they can take into the work place that they can use to pay off their student debt =P. If anything, try to teach both of them! Or let them choose (but that can be a nightmare where some students are using R and others are using SPSS).

Yes -> SPSS is too expensive these days, most PC (non-mobile) software is (not trying to justify it). R is definitely an alternative that businesses and individuals should consider if the price-point is important enough to them.

HTH,
----

J. R. Carroll

Cell: (650) 776-6613

Email: [hidden email]

[hidden email]

[hidden email]

[hidden email]
On Wed, Jul 11, 2012 at 6:32 AM, Dr. Thomas W. MacFarland <[hidden email]> wrote:
Everyone:

I’ve followed this thread and of course there are many different perspectives here.

SPSS is grand but unfortunately the cost is soon approaching a grand.

So, I’m now introducing R (standalone use of syntax and use of the R Commander GUI) into my graduate-level statistics classes. Have others had success with R, in reaction to the cost of SPSS and other proprietary software?

It is always great to have options.

Best wishes.

Tom

----------

Thomas W. MacFarland, Ed.D.

Senior Research Associate; Institutional Effectiveness and Associate Professor

Nova Southeastern University

Voice <a moz-do-not-send="true" href="tel:954-262-5395" target="_blank">954-262-5395 [hidden email]

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Art Kendall
Sent: Wednesday, July 11, 2012 9:09 AM

To: [hidden email]
Subject: Re: Seeking Python and R Essentials for SPSS version 16
Definitely!

Jon goes out of his way to help.
Former SPSS employees ViAnn and David remind us of how SPSS used to relate to users.

SPSS pays for itself for large organizations that use it enough. (Lifetime cost of ownership).
However it is prohibitively costly for individual researchers/consultants, and not-for-profit (NGOs) especially in the Third World.

I was fortunate enough to buy the lifetime contract back when they were available in the mid-80s.
Otherwise, when I had a full time job I would not have had access to current SPSS for my part-time work.
Also, I retired in 2001 and do some paid consulting to offset some of my expenses for pro bono work. I spend most of my time in working with Statistics Without Borders and with the AAAS Science & Human Rights Coalition. I would not be able to use SPSS if I had not signed that contract back in the mid-80s. If I had to pay for the modules, it would cost several times my net consulting income.

SPSS seems to have forgotten that the popularity of "Business Intelligence" is a result of social scientists' and statisticians' efforts. Their big money may come from the large corporate world, but the popularity of SPSS in the academic and non-profit world is why there are people who know stat and methods well is the foundation underneath
SPSS seems to have forgotten that the software people start out with is what they tend to stick with.
SPSS seems to have forgotten that customers are critical in the continuous improvement of their products.
Art Kendall
Social Research Consultants
On 7/11/2012 6:48 AM, King Douglas wrote:
Antoon,

With a few exceptions, I agree with your implied sentiments in regard to the deep and broad changes that IBM has made to the previously stellar qualities of SPSS customer service and support.

One exception I have in mind is the personal efforts of Jon Peck, who has never failed to extend himself to the benefit of all SPSS users. It is IBM policies that he did not make and cannot unilaterally change that are making the SPSS experience so frustrating for many of us.

Cheers,

King Douglas

American Airlines Customer Research

From: SPSSX(r) Discussion [[hidden email]] On Behalf Of Antoon Smulders
Sent: Wednesday, July 11, 2012 2:05 AM
To: [hidden email]
Subject: Re: Seeking Python and R Essentials for SPSS version 16

I wouldlike to express my feelings about the lack of service that IBM-PASW-SPSS gives to customers, but in order to stay decent I remain silent.

Antoon Smulders

Van: SPSSX(r) Discussion [[hidden email]] Namens Vladimir Lopez-Prado
Verzonden: woensdag 11 juli 2012 2:38
Aan: [hidden email]
Onderwerp: Re: Seeking Python and R Essentials for SPSS version 16

Hi Patrick,

I also need Python and R plug in for SPSS version 16. Did you get any help on your request? Please let me know.

Thank you,

vlad

On Mon, Jul 9, 2012 at 9:25 AM, Patrick Burns <[hidden email]> wrote:

Does anyone have access to the Python and R Essentials for SPSS version 16?
IBM does not distribute/support these prior to version 18, and I didn't have a
need to install them when my SPSS versions 16.0.2 was new. (I'm only just now
seeking to learn/run python scripts, and our non-profit doesn't have the
funding to buy new version of SPSS and subscribe to maintenance plans.)

Thanks in advance if you have copies of these...

=====================
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
Social Research Consultants