SPSSX Discussion

R Commander, anyone?

Classic

List

Threaded

16 messages Options

Albert-Jan Roskam

R Commander, anyone?

Hi,

I was just checking R Commander, a GUI for R. It looks interesting, because (just like the Spss PRINTBACK) it generates the underlying R code in the background. It can also take spss data files. Does anybody know how actively this program is maintained, or what other good alternatives there are?

Cheers!!
Albert-Jan

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Before you criticize someone, walk a mile in their shoes, that way
when you do criticize them, you're a mile away and you have their shoes!
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

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

ajayohri

Re: R Commander, anyone?

Hi Albert,

R Commander is maintained very actively by Prof John Fox and even better it has some additional plugin packages ( it can extend other R packages like quality control qcc)

It takes less than 10 minutes for a non R language user to start crunching basic descriptive statistics and tests. For SPSS datasets I think you need other package to read it ( am not sure ) or better still import from csv.

You can read on the interview here with Dr Fox on the R GUI

http://decisionstats.wordpress.com/2009/09/14/interview-professor-john-fox-creator-r-commander/

Regards,

Ajay Ohri

Grad Student

University of Tennessee

On Tue, Oct 20, 2009 at 8:59 AM, Albert-Jan Roskam <[hidden email]> wrote:

Hi,

I was just checking R Commander, a GUI for R. It looks interesting, because (just like the Spss PRINTBACK) it generates the underlying R code in the background. It can also take spss data files. Does anybody know how actively this program is maintained, or what other good alternatives there are?

Cheers!!
Albert-Jan

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Before you criticize someone, walk a mile in their shoes, that way
when you do criticize them, you're a mile away and you have their shoes!
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

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

Bob Walker-2

GPL vs. ViZml

I am automating a graphical reporting system for a monthly consumer satisfaction study. Conceptually, how should I contrast the benefits of using GPL vs. ViZml? Does the latter primarily benefit web-based reporting model, vs. GPL which is more automation/syntax oriented?

TIA,

Bob Walker

Surveys & Forecasts, LLC

www.safllc.com

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Ajay ohri
Sent: Tuesday, October 20, 2009 9:32 AM
To: [hidden email]
Subject: Re: R Commander, anyone?

Hi Albert,

R Commander is maintained very actively by Prof John Fox and even better it has some additional plugin packages ( it can extend other R packages like quality control qcc)

You can read on the interview here with Dr Fox on the R GUI

http://decisionstats.wordpress.com/2009/09/14/interview-professor-john-fox-creator-r-commander/

Regards,

Ajay Ohri

Grad Student

University of Tennessee

On Tue, Oct 20, 2009 at 8:59 AM, Albert-Jan Roskam <[hidden email]> wrote:

Steve Simon, P.Mean Consulting

Re: R Commander, anyone?

In reply to this post by Albert-Jan Roskam

Albert-Jan Roskam wrote:

> I was just checking R Commander, a GUI for R. It looks interesting,
> because (just like the Spss PRINTBACK) it generates the underlying R
> code in the background. It can also take spss data files. Does
> anybody know how actively this program is maintained, or what other
> good alternatives there are?

I am a big fan of R, and R Commander looks like it will extend the
audience of R to those people who prefer a menu driven interface to a
command interface. Still, you have to realize that R is written quite
differently than most other software. It takes an object oriented
approach and while it is not inherently more difficult than other
approaches, it is sufficiently different that it takes a long time to
get comfortable with it. The object oriented perspective does not
disappear with a menu driven interface.

If you're curious what "object oriented" means from the perspective of
statistics, I tried to explain it at my old website:

* www.childrensmercy.org/stats/weblog2005/ObjectOrientedR.asp

I'd recommend R and R Commander for the more adventurous among you, but
most of you will be more comfortable with the more traditional approach
used by PASW/SPSS.

I'm sure that most people on this list know about PSPP, which is another
open source alternative to PASW/SPSS.

http://www.gnu.org/software/pspp/

I could not get PSPP to load on this particular computer (they do not
trust me at this particular location to install my own software), but
I'll take a look at it at home tonight.

As a very general rule, open source software usually has other
non-monetary costs associated with it. It often, for example, requires a
greater commitment of time or labor compared to commercial software. You
need to consider the non-monetary costs of free software very carefully
before you make a switch. It might be your best choice, depending on
your circumstances, but don't assume that free is always better.

I hesitate to offer any more advice, because choosing a statistical
software program is not unlike choosing a religion. It can often
generate heated debate with about as much passion as religion does. It
is also like religion in that it is highly influenced by a person's
individual needs and desires.

There's a lot of acrimony between proponents of commercial products and
proponents of open source products. In the arena of Statistics, both are
valuable. There is some overlap between the two, but commercial software
satisfies a different audience with different needs, for the most part,
compared to open source software.
--
Steve Simon, Standard Disclaimer
Second free statistics webinar, Wed, Nov 4, 11am-noon CST.
"The first three steps in data entry, with examples in PASW/SPSS"
Details at www.pmean.com/webinars

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

ViAnn Beadle

Re: GPL vs. VizML

In reply to this post by Bob Walker-2

That’s a really complicated question. Here’s my personal opinion on the whole thing:

GPL is a higher level language that SPSS parses to generate VizML. VizML is XML. VizML provides much more control over formatting than GPL and can do some specialized charts which are not exposed within the GPL language.

Writing your own VizML is not for the faint of heart. AFAIK, the only public documentation comes with some enterprise tools that SPSS sells and comments within the xschema. Probably the best way to learn it is to create a chart using GGRAPH, IGRAPH, or GGRAPH and then open that within the Chart Editor and save it as VizML.

In PASW 17, SPSS introduced a new chart Graphboard facility which uses style sheets and variable mapping applied to visualization templates which is a variant of VizML. Note that these VizML templates are not Chart Templates (*.sgt) files.

A limited set of templates are provided with Graphboard. Depending upon the type of charts you are producing they may be sufficient. Otherwise, you are on your own to build them using Viz Designer which is a separate enterprise-level product.

The enterprise level graph facilities are designed to support a web-based distribution system but are more at the level of embeddable tools within some higher-level enterprise system.

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Bob Walker
Sent: Tuesday, October 20, 2009 1:24 PM
To: [hidden email]
Subject: GPL vs. ViZml

TIA,

Bob Walker

Surveys & Forecasts, LLC

www.safllc.com

Bob Walker-2

Re: GPL vs. VizML

ViAnn,

Thanks very much… I figured this list would be a good place to address this question. And GPL should be fine for what I need. FYI, there is some VizML documentation in the PASW 17 help facility, though it is very hard to wade through.

Regards,

Bob Walker

Surveys & Forecasts, LLC

www.safllc.com

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of ViAnn Beadle
Sent: Tuesday, October 20, 2009 6:00 PM
To: [hidden email]
Subject: Re: GPL vs. VizML

That’s a really complicated question. Here’s my personal opinion on the whole thing:

The enterprise level graph facilities are designed to support a web-based distribution system but are more at the level of embeddable tools within some higher-level enterprise system.

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Bob Walker
Sent: Tuesday, October 20, 2009 1:24 PM
To: [hidden email]
Subject: GPL vs. ViZml

TIA,

Bob Walker

Surveys & Forecasts, LLC

www.safllc.com

Peck, Jon

Re: GPL vs. VizML

To add to ViAnn’s comments,

VizML is sort of the (low-level) assembly language of the graphics engine. Most users would work in the higher level GPL.

But VizDesigner, a separate product, gives you control over the VizML and civilizes the process. With VizML you might build very sophisticated graphics templates that could then be dropped in to SPSS and used in GraphBoard or even used in other SPSS products.

Regards,

Jon Peck

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Bob Walker
Sent: Tuesday, October 20, 2009 7:34 PM
To: [hidden email]
Subject: Re: [SPSSX-L] GPL vs. VizML

ViAnn,

Regards,

Bob Walker

Surveys & Forecasts, LLC

www.safllc.com

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of ViAnn Beadle
Sent: Tuesday, October 20, 2009 6:00 PM
To: [hidden email]
Subject: Re: GPL vs. VizML

That’s a really complicated question. Here’s my personal opinion on the whole thing:

The enterprise level graph facilities are designed to support a web-based distribution system but are more at the level of embeddable tools within some higher-level enterprise system.

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Bob Walker
Sent: Tuesday, October 20, 2009 1:24 PM
To: [hidden email]
Subject: GPL vs. ViZml

TIA,

Bob Walker

Surveys & Forecasts, LLC

www.safllc.com

Albert-Jan Roskam

Re: R Commander, anyone?

In reply to this post by ajayohri

Hi Ajay and Steve,

Thanks for the information. Interesting article, too! R Commander might just be the push in my back that I needed to start learning R. In a work environment it's hard to justify when simple things take a lot of time; R Commander might reduce that time, even though real coding activities are of course still needed.

Cheers!!
Albert-Jan

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Before you criticize someone, walk a mile in their shoes, that way
when you do criticize them, you're a mile away and you have their shoes!
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

--- On Tue, 10/20/09, Ajay ohri <[hidden email]> wrote:

> From: Ajay ohri <[hidden email]>
> Subject: Re: R Commander, anyone?
> To: "Albert-Jan Roskam" <[hidden email]>
> Cc: [hidden email]
> Date: Tuesday, October 20, 2009, 3:31 PM
> Hi Albert,
> R Commander is maintained very actively by Prof
> John Fox and even better it has some additional plugin
> packages ( it can extend other R packages like quality
> control qcc)
> It takes less than 10 minutes for a non R
> language user to start crunching
> basic descriptive statistics and tests. For SPSS datasets
> I think you need other package to read it ( am not sure ) or
> better still import from csv.
>
> You can read on the interview here with Dr Fox
> on the R GUI
>
> http://decisionstats.wordpress.com/2009/09/14/interview-professor-john-fox-creator-r-commander/
>
> Regards,
> Ajay Ohri
> Grad StudentUniversity of
> Tennessee
>
> On Tue, Oct 20, 2009 at 8:59 AM,
> Albert-Jan Roskam <[hidden email]>
> wrote:
>
> Hi,
>
>
>
> I was just checking R Commander, a GUI for R. It looks
> interesting, because (just like the Spss PRINTBACK) it
> generates the underlying R code in the background. It can
> also take spss data files. Does anybody know how actively
> this program is maintained, or what other good alternatives
> there are?
>
>
>
>
> Cheers!!
>
> Albert-Jan
>
>
>
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>
> Before you criticize someone, walk a mile in their shoes,
> that way
>
> when you do criticize them, you're a mile away and you
> have their shoes!
>
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>
>
>
> =====================
>
> 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

Bob Walker-2

Re: GPL vs. VizML

In reply to this post by Peck, Jon

Thanks Jon… is VizDesigner an enterprise product? I see it on the SPSS site, but no pricing info.

Regards,

Bob Walker

Surveys & Forecasts, LLC

www.safllc.com

From: Peck, Jon [mailto:[hidden email]]
Sent: Tuesday, October 20, 2009 9:42 PM
To: Bob Walker; [hidden email]
Subject: RE: Re: [SPSSX-L] GPL vs. VizML

To add to ViAnn’s comments,

VizML is sort of the (low-level) assembly language of the graphics engine. Most users would work in the higher level GPL.

Regards,

Jon Peck

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Bob Walker
Sent: Tuesday, October 20, 2009 7:34 PM
To: [hidden email]
Subject: Re: [SPSSX-L] GPL vs. VizML

ViAnn,

Regards,

Bob Walker

Surveys & Forecasts, LLC

www.safllc.com

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of ViAnn Beadle
Sent: Tuesday, October 20, 2009 6:00 PM
To: [hidden email]
Subject: Re: GPL vs. VizML

That’s a really complicated question. Here’s my personal opinion on the whole thing:

The enterprise level graph facilities are designed to support a web-based distribution system but are more at the level of embeddable tools within some higher-level enterprise system.

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Bob Walker
Sent: Tuesday, October 20, 2009 1:24 PM
To: [hidden email]
Subject: GPL vs. ViZml

TIA,

Bob Walker

Surveys & Forecasts, LLC

www.safllc.com

Mike Pritchard

Crosstabs - SPSS versus Wincross or ??

In reply to this post by ajayohri

Hello list.

The discussion about R Commander prompts me to ask what people are using for stat tested crosstabs.

I’m probably a dinosaur, but I still find the banner tables from Wincross very helpful for quickly spotting differences between groups. Thanks to John Hall and other help from this list and other SPSS sources, I’ve been able to generate some useful tables from SPSS. But, even allowing for the fact that I’m still learning the tricks, it doesn’t appear that I could generate anything from SPSS that is anywhere near as well formatted as Wincross, or that it could be done very efficiently.

I’m not currently using Wincross, so my perceptions are partly driven by what I used to get from a subcontractor. I’m wondering whether it is worth buying, but the cost is an issue. I looked at Ccount, but I’m nervous about investing much time because it doesn’t seem to have much of a following, hasn’t been updated for a while, and seems to have a steep learning curve.

Are there other tools out there that could generate a decent looking table efficiently and inexpensively? Are banner tables no longer valuable?

Thanks

Mike

_________________________________________________________________________
Mike Pritchard | [hidden email] | 5 Circles Research | 425-444-3410

Guerrero, Rodrigo

Re: Crosstabs - SPSS versus Wincross or ??

Mike,

I also get frustrated with SPSS on this topic. The CTABLES comand seems to have some flexibility., but I have not found a way to flag significant differences in the table. Looking at diferent tables is a hassle. Maybe other know more than me.

Rodrigo

From: SPSSX(r) Discussion
To: [hidden email]
Sent: Thu Oct 22 15:17:50 2009
Subject: Crosstabs - SPSS versus Wincross or ??

Hello list.

The discussion about R Commander prompts me to ask what people are using for stat tested crosstabs.

Are there other tools out there that could generate a decent looking table efficiently and inexpensively? Are banner tables no longer valuable?

Thanks

Mike

_________________________________________________________________________
Mike Pritchard | [hidden email] | 5 Circles Research | 425-444-3410

The information transmitted is intended only for the addressee(s) and may contain confidential or privileged material, or both. Any review, receipt, dissemination or other use of this information by non-addressees is prohibited. If you received this in error or are a non-addressee, please contact the sender and delete the transmitted information.

Mike Pritchard

Re: Crosstabs - SPSS versus Wincross or ??

Rodrigo,

You can check significant differences with the custom tables module – comparing means and percentages. I don’t know when this feature was added. I know if wasn’t there when I used SPSS while working for another company a few years ago, and I know it is there in my V16. You set it up in the Test Statistics tab of the Custom Tables dialog, or you can do it in syntax.

It isn’t as flexible as what I see from Wincross, and doesn’t produce compact output. I’m resigned to buying a copy of Wincross at this point, so hopefully I can find a used copy to keep the cost down. Don’t know where to look though.

I’m glad that I’m not the only one seeing the problem.

Mike

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Guerrero, Rodrigo
Sent: Thursday, October 22, 2009 6:16 PM
To: [hidden email]
Subject: Re: Crosstabs - SPSS versus Wincross or ??

From: SPSSX(r) Discussion
To: [hidden email]
Sent: Thu Oct 22 15:17:50 2009
Subject: Crosstabs - SPSS versus Wincross or ??

Hello list.

The discussion about R Commander prompts me to ask what people are using for stat tested crosstabs.

Are there other tools out there that could generate a decent looking table efficiently and inexpensively? Are banner tables no longer valuable?

Thanks

Mike

_________________________________________________________________________
Mike Pritchard | [hidden email] | 5 Circles Research | 425-444-3410

Peck, Jon

Re: Crosstabs - SPSS versus Wincross or ??

The significance tests were added in version 11.5 if my memory is correct. With V17 or later you can use the SPSSINC MERGE TABLES extension command to merge the significance table into the main table if you install the programmability plugin. In Version 18, Ctables can display the significance results automatically in the main table, but the representation is different from the separate table form. It’s APA style.

Version 18 Ctables can also do “post-computes”, i.e., include functions of the cells as additional rows/columns in the table.

Ctables is pretty flexible, so it would be interesting to see what you can’t do.

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Mike Pritchard
Sent: Thursday, October 22, 2009 7:33 PM
To: [hidden email]
Subject: Re: [SPSSX-L] Crosstabs - SPSS versus Wincross or ??

Rodrigo,

I’m glad that I’m not the only one seeing the problem.

Mike

From: SPSSX(r) Discussion
To: [hidden email]
Sent: Thu Oct 22 15:17:50 2009
Subject: Crosstabs - SPSS versus Wincross or ??

Hello list.

The discussion about R Commander prompts me to ask what people are using for stat tested crosstabs.

Are there other tools out there that could generate a decent looking table efficiently and inexpensively? Are banner tables no longer valuable?

Thanks

Mike

_________________________________________________________________________
Mike Pritchard | [hidden email] | 5 Circles Research | 425-444-3410

Bob Walker-2

Re: Crosstabs - SPSS versus Wincross or ??

In reply to this post by Mike Pritchard

Hi Mike,

Not sure if you’re using the word “WinCross” as a generic phrase for all crosstab packages, but I have been using WinCross for nearly a decade and have found it to be an exceptionally solid package. While SPSS is an excellent product, it doesn’t come close to what researchers need in terms of table output (unless you want to customize CTABLES, OMS, or other products (e.g., Quanvert). WinCross produces traditional banner (e.g., +/- 20-point) and stub tables that meet all of my marketing research client needs. You can stat test across multiple banner points at two levels of significance, including McNemar, and it includes an exceptionally good sample balancing routine based on Deming’s seminal 1943 work, a factor analysis module, and version de-rotation based on indexed variables. It reads SPSS files through v17 without any problem, slurping in all of the dictionary info (variable and value labels), and easily reads standard delimited ASCII, CSV, as well as column binary/ multipunch /EBCIDC. Of course, you have think ahead to how your SPSS files will be structured to make the best use of this facility. It also exports to Excel which is great for cutting/pasting and/or linking to PowerPoint files.

The downside of WC is in its file matching capabilities, basically unusable graphics, and variable/math transformations using its glossary logic. This is where SPSS shines and thus I have it do all of the heavy lifting prior to running crosstabs in WC.

HTH,

Bob Walker

Surveys & Forecasts, LLC

www.safllc.com

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Mike Pritchard
Sent: Thursday, October 22, 2009 4:18 PM
To: [hidden email]
Subject: Crosstabs - SPSS versus Wincross or ??

Hello list.

The discussion about R Commander prompts me to ask what people are using for stat tested crosstabs.

Are there other tools out there that could generate a decent looking table efficiently and inexpensively? Are banner tables no longer valuable?

Thanks

Mike

_________________________________________________________________________
Mike Pritchard | [hidden email] | 5 Circles Research | 425-444-3410

Mike Pritchard

Re: Crosstabs - SPSS versus Wincross or ??

Thanks Bob.

I’ve been exposed to WinCross and other custom packages that have similar output. Also eTabs and some of the products/services from survey tool vendors. So I guess I’m using the term WinCross both ways – as you write, it is very solid. At this point I’m mostly interested in the table output, and also to some extent WC => Powerpoint, rather than built in graphics, so the limitations there aren’t too big a concern. It sounds like the way you are using WC is pretty much the way I want to use it (but more sophisticated!)

Yes, that helps a lot.

Mike

_________________________________________________________________________
Mike Pritchard | [hidden email] | 5 Circles Research | 425-444-3410
Helping companies delight their customers to increase profits
Signup for SurveyTips and Seminar announcements

The information transmitted is intended only for the person or entity to which it is addressed and may contain confidential, proprietary, and/or privileged material. Any review, retransmission, dissemination or other use of, or taking of any action in reliance upon this information by persons or entities other than the intended recipient is prohibited. If you received this in error, please contact the sender and delete the material from all computers.

From: Bob Walker [mailto:[hidden email]]
Sent: Friday, October 23, 2009 12:27 PM
To: [hidden email]
Cc: 'Mike Pritchard'
Subject: RE: Crosstabs - SPSS versus Wincross or ??

Hi Mike,

HTH,

Bob Walker

Surveys & Forecasts, LLC

www.safllc.com

Hello list.

The discussion about R Commander prompts me to ask what people are using for stat tested crosstabs.

Are there other tools out there that could generate a decent looking table efficiently and inexpensively? Are banner tables no longer valuable?

Thanks

Mike

_________________________________________________________________________
Mike Pritchard | [hidden email] | 5 Circles Research | 425-444-3410

Luca Meyer-3

Re: GPL vs. VizML

In reply to this post by Bob Walker-2

<base href="x-msg://129/">Hi,

Sorry i take back an old thread but I am looking forward to build highly customized graphs using PASW 18. My question about VizML is Mac OS compatibility as it appears from what I have found on SPSS web site (http://www.spss.com/media/collateral/statistics/vizdesigner.pdf) that Windows XP is required, can anyone comment? What is the alternative of VizML for Mac users?

Thanks,

Luca

Il giorno 21/ott/2009, alle ore 15.28, Bob Walker ha scritto:

Thanks Jon… is VizDesigner an enterprise product? I see it on the SPSS site, but no pricing info.

Regards,

Bob Walker
Surveys & Forecasts, LLC
www.safllc.com

From: Peck, Jon [mailto:[hidden email]]
Sent: Tuesday, October 20, 2009 9:42 PM
To: Bob Walker; [hidden email]
Subject: RE: Re: [SPSSX-L] GPL vs. VizML

To add to ViAnn’s comments,

VizML is sort of the (low-level) assembly language of the graphics engine. Most users would work in the higher level GPL.

But VizDesigner, a separate product, gives you control over the VizML and civilizes the process. With VizML you might build very sophisticated graphics templates that could then be dropped in to SPSS and used in GraphBoard or even used in other SPSS products.

Regards,
Jon Peck

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Bob Walker
Sent: Tuesday, October 20, 2009 7:34 PM
To: [hidden email]
Subject: Re: [SPSSX-L] GPL vs. VizML

ViAnn,

Thanks very much… I figured this list would be a good place to address this question. And GPL should be fine for what I need. FYI, there is some VizML documentation in the PASW 17 help facility, though it is very hard to wade through.

Regards,

Bob Walker
Surveys & Forecasts, LLC
www.safllc.com

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of ViAnn Beadle
Sent: Tuesday, October 20, 2009 6:00 PM
To: [hidden email]
Subject: Re: GPL vs. VizML

That’s a really complicated question. Here’s my personal opinion on the whole thing:

GPL is a higher level language that SPSS parses to generate VizML. VizML is XML. VizML provides much more control over formatting than GPL and can do some specialized charts which are not exposed within the GPL language.

Writing your own VizML is not for the faint of heart. AFAIK, the only public documentation comes with some enterprise tools that SPSS sells and comments within the xschema. Probably the best way to learn it is to create a chart using GGRAPH, IGRAPH, or GGRAPH and then open that within the Chart Editor and save it as VizML.

In PASW 17, SPSS introduced a new chart Graphboard facility which uses style sheets and variable mapping applied to visualization templates which is a variant of VizML. Note that these VizML templates are not Chart Templates (*.sgt) files.

A limited set of templates are provided with Graphboard. Depending upon the type of charts you are producing they may be sufficient. Otherwise, you are on your own to build them using Viz Designer which is a separate enterprise-level product.

The enterprise level graph facilities are designed to support a web-based distribution system but are more at the level of embeddable tools within some higher-level enterprise system.

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Bob Walker
Sent: Tuesday, October 20, 2009 1:24 PM
To: [hidden email]
Subject: GPL vs. ViZml

I am automating a graphical reporting system for a monthly consumer satisfaction study. Conceptually, how should I contrast the benefits of using GPL vs. ViZml? Does the latter primarily benefit web-based reporting model, vs. GPL which is more automation/syntax oriented?

TIA,

Bob Walker
Surveys & Forecasts, LLC
www.safllc.com