SPSSX Discussion

Re: Weighting tables using More Than 2 Variables

Classic

List

Threaded

4 messages Options

Hans Chen

Re: Weighting tables using More Than 2 Variables

Dear Listers

I know how to weight the tables using more that 2 variables in Wincross(*Sample
Balancing*), Stat XP(*Marginal Weighting* using pond function). But I have
no idea whether it is possible or not and how to do it in SPSS. If it is
possible, could you please show me how to weight the tables using following
3 variables. the target proportions(ratio) are as follows.

Region
1 15%
2 25%
3 10%
4 18%
5 32%

Gender
Male 45%
Female 55%

Age
18-24 15%
25-34 25%
35-44 11%
45-54 19%
55-64 14%
65+ 16%.

Any suggestions will be highly appreciated.

Hans Chen

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

C. Fong

Re: Weighting tables using More Than 2 Variables

Have you look in to the "Weight by" command? I believe the command accepted one variable. Perhaps you can create one var from the three before using the weight command, e.g.

compute newwgt=regwgt*gendrwgt*agewgt.

weight by newwgt.
crosstabs var1 by var2.

Hans Chen wrote

Dear Listers

I know how to weight the tables using more that 2 variables in Wincross(*Sample
Balancing*), Stat XP(*Marginal Weighting* using pond function). But I have
no idea whether it is possible or not and how to do it in SPSS. If it is
possible, could you please show me how to weight the tables using following
3 variables. the target proportions(ratio) are as follows.

Region
1 15%
2 25%
3 10%
4 18%
5 32%

Gender
Male 45%
Female 55%

Age
18-24 15%
25-34 25%
35-44 11%
45-54 19%
55-64 14%
65+ 16%.

Any suggestions will be highly appreciated.

Hans Chen

=====================
To manage your subscription to SPSSX-L, send a message to
LISTSERV@LISTSERV.UGA.EDU (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: Weighting tables using More Than 2 Variables

I think the OP wants rim-weighting or raking. A python module is available
to do raking and is on www.spss.com/devcentral. It requires 15.01.

Try this url: http://www.spss.com/devcentral/index.cfm?pg=downloadDet&dId=50

And see if it points to the right module.

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of C.
Fong
Sent: Monday, April 28, 2008 2:37 PM
To: [hidden email]
Subject: Re: Weighting tables using More Than 2 Variables

Have you look in to the "Weight by" command? I believe the command accepted
one variable. Perhaps you can create one var from the three before using the
weight command, e.g.

compute newwgt=regwgt*gendrwgt*agewgt.

weight by newwgt.
crosstabs var1 by var2.

Hans Chen wrote:

>
> Dear Listers
>
> I know how to weight the tables using more that 2 variables in
> Wincross(*Sample
> Balancing*), Stat XP(*Marginal Weighting* using pond function). But I have
> no idea whether it is possible or not and how to do it in SPSS. If it is
> possible, could you please show me how to weight the tables using
> following
> 3 variables. the target proportions(ratio) are as follows.
>
> Region
> 1 15%
> 2 25%
> 3 10%
> 4 18%
> 5 32%
>
> Gender
> Male 45%
> Female 55%
>
> Age
> 18-24 15%
> 25-34 25%
> 35-44 11%
> 45-54 19%
> 55-64 14%
> 65+ 16%.
>
> Any suggestions will be highly appreciated.
>
> Hans Chen
>
> =====================
> 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
>
>

Jamie Burnett-3

Re: Weighting tables using More Than 2 Variables

In reply to this post by C. Fong

Hi Hans,

Weight by is syntax used to weight your dataset to some weighting variable. What you require is the tools to calculate this weight in the first place. There is a macro availale in SPSS that will calculate weights for you based on you providing the macro with the marginal distributions for each of your weighting variables. The weighting technique is called iterative proportional fitting (IPF) or more commonly raking. It is one of a suite of techniques under the heading 'calibration'. I think you will find the macro on Raynalds SPSS tools website otherwise I can send you it.

Thanks

Jamie

________________________________

From: SPSSX(r) Discussion on behalf of C. Fong
Sent: Mon 28/04/2008 21:36
To: [hidden email]
Subject: Re: Weighting tables using More Than 2 Variables

Have you look in to the "Weight by" command? I believe the command accepted
one variable. Perhaps you can create one var from the three before using the
weight command, e.g.

compute newwgt=regwgt*gendrwgt*agewgt.

weight by newwgt.
crosstabs var1 by var2.

Hans Chen wrote:

--
View this message in context: http://www.nabble.com/Re%3A-Weighting-tables-using-More-Than-2-Variables-tp16823904p16945895.html
Sent from the SPSSX Discussion mailing list archive at Nabble.com.

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

============================
This e-mail and all attachments it may contain is confidential and intended solely for the use of the individual to whom it is addressed. Any views or opinions presented are solely those of the author and do not necessarily represent those of Ipsos MORI and its associated companies. If you are not the intended recipient, be advised that you have received this e-mail in error and that any use, dissemination, printing, forwarding or copying of this e-mail is strictly prohibited. Please contact the sender if you have received this e-mail in error.

Market & Opinion Research International Ltd , Registered in England and Wales No. 948470 , 79-81 Borough Road , London SE1 1FY, United Kingdom, Email: [hidden email]
============================

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