TURF analysis & weights

[Mark Webb-3]

I can’t get TURF to apply respondent weights. [SPSSINC TURF]

I would find this important in the type of data files I use.

Is it possible ? What am I missing/doing wrong here?

 

[P-Stat claims to allow respondent weighting in TURF.]

 

Mark Webb

 

[phone]

[phone]

Skype - webbmark

[hidden email]

 

[Peck, Jon]

I knew somebody would ask about weights J

 

The current version does not support case weights, but I plan to add support for this soon.  It's a little harder than for a typical SPSS procedure, because of the way the set algebra is done, so I left it out of version 1.

 

Regards,

Jon Peck

 

---

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Mark Webb
Sent: Wednesday, February 11, 2009 12:35 AM
To: [hidden email]
Subject: [SPSSX-L] TURF analysis & weights

 

I can’t get TURF to apply respondent weights. [SPSSINC TURF]

I would find this important in the type of data files I use.

Is it possible ? What am I missing/doing wrong here?

 

[P-Stat claims to allow respondent weighting in TURF.]

 

Mark Webb

 

[phone]

[phone]

Skype - webbmark

[hidden email]

 

[Mark Webb-3]

Thanks Jon

Please post a message when you do get weights up & running.

I will find it useful if weights can be applied.

 

Mark Webb

 

[phone]

[phone]

Skype - webbmark

[hidden email]

 

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Peck, Jon
Sent: 11 February 2009 03:37 PM
To: [hidden email]
Subject: Re: TURF analysis & weights

 

I knew somebody would ask about weights J

 

The current version does not support case weights, but I plan to add support for this soon.  It's a little harder than for a typical SPSS procedure, because of the way the set algebra is done, so I left it out of version 1.

 

Regards,

Jon Peck

 

---

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Mark Webb
Sent: Wednesday, February 11, 2009 12:35 AM
To: [hidden email]
Subject: [SPSSX-L] TURF analysis & weights

 

I can’t get TURF to apply respondent weights. [SPSSINC TURF]

I would find this important in the type of data files I use.

Is it possible ? What am I missing/doing wrong here?

 

[P-Stat claims to allow respondent weighting in TURF.]

 

Mark Webb

 

[phone]

[phone]

Skype - webbmark

[hidden email]

 

[Dave Taber]

Jon:

just checked the website and didn't see any mention of TURF analysis - what module is this in and when do you think you would have the same weighting features as p-stat?

thanks

dave taber

On Feb 11, 2009, at 8:36 AM, Peck, Jon wrote:

I knew somebody would ask about weights J

 

The current version does not support case weights, but I plan to add support for this soon.  It's a little harder than for a typical SPSS procedure, because of the way the set algebra is done, so I left it out of version 1.

 

Regards,

Jon Peck

 

---

From: SPSSX(r) Discussion [[hidden email]] On Behalf Of Mark Webb
Sent: Wednesday, February 11, 2009 12:35 AM
To: [hidden email]
Subject: [SPSSX-L] TURF analysis & weights

 

I can’t get TURF to apply respondent weights. [SPSSINC TURF]

I would find this important in the type of data files I use.

Is it possible ? What am I missing/doing wrong here?

 

[P-Stat claims to allow respondent weighting in TURF.]

 

Mark Webb

 

[phone]

[phone]

Skype - webbmark

[hidden email]

 

Dave Taber

novaQuant

listening - thinking - solving

[hidden email]

914.238.1310

www.novaQuant.com

[Peck, Jon]

The SPSSINC TURF analysis module has been updated on Developer Central (www.spss.com/devcentral) to support case weights.  If a weight variable is in effect in SPSS, it will be used automatically.  As usual with SPSS, nonpositive or system-missing weights cause the case to be ignored.  Otherwise, the weights are summed with no truncation or rounding.

 

To find this or any extension module on Developer Central,

-        go to the Downloads link

-        click Single Page View

-        select Extension Commands in the filter box and click Filter Downloads

 

The extension commands are listed in alphabetical order.

 

HTH,

Jon

 

p.s. Whether these weights are the same as P-Stat, I don't know.  There are two kinds of weights that can be used here.  Case weights are what SPSSINC TURF now honors.  There can also be importance or profit weights.  Those apply to questions, not cases, and allow you to give more importance to responses to certain questions than to other.  So there would be one weight number for each question rather than one for each case.  I have support for that on my TODO list but have not yet done it.  (Those weights are already in the syntax definition.)  I'm waiting to hear whether such support is needed, as I am not sure that they would be used.

 

 

 

---

From: Dave Taber [mailto:[hidden email]]
Sent: Wednesday, February 11, 2009 8:46 AM
To: Peck, Jon
Cc: [hidden email]
Subject: Re: TURF analysis & weights

 

Jon:

 

just checked the website and didn't see any mention of TURF analysis - what module is this in and when do you think you would have the same weighting features as p-stat?

 

thanks

dave taber

 

 

On Feb 11, 2009, at 8:36 AM, Peck, Jon wrote:

I knew somebody would ask about weights J

 

The current version does not support case weights, but I plan to add support for this soon.  It's a little harder than for a typical SPSS procedure, because of the way the set algebra is done, so I left it out of version 1.

 

Regards,

Jon Peck

 

---

From: SPSSX(r) Discussion [[hidden email]] On Behalf Of Mark Webb
Sent: Wednesday, February 11, 2009 12:35 AM
To: [hidden email]
Subject: [SPSSX-L] TURF analysis & weights

 

I can’t get TURF to apply respondent weights. [SPSSINC TURF]

I would find this important in the type of data files I use.

Is it possible ? What am I missing/doing wrong here?

 

[P-Stat claims to allow respondent weighting in TURF.]

 

Mark Webb

 

[phone]

[phone]

Skype - webbmark

[hidden email]

 

 

Dave Taber

novaQuant

listening - thinking - solving

[hidden email]

914.238.1310

www.novaQuant.com