Using Multiple Weights

[Valdiserri, Colin]

I have a total of 5 weight variables, but need to use them as one
variable. For example I need to weight my data by age, income, gender,
home ownership, and marital status.

 

How can I use each of the 5 weight variables at once?

 

Any help is greatly appreciated.

 

Thanks,

 

Colin

 


_______________________________________________________________________

The information transmitted is intended only for the person or entity to which it is addressed and may contain confidential 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 any computer.
_______________________________________________________________________

[Chris Graham-5]

Colin,

To do this you don't want to be using five distinct sets of weights -
rather you want to be using one set of weights defined by five
variables.  Imagine if you wanted to weight for just age (using 5
categories) and gender - then rather than come up with separate weights
for age and sex and use them together you'd want to be weighting based
on ten categories overall - five groups of men and five groups of women
of different ages.  How you compute weights then depends on what you're
weighting against; if it's a general population, for example, you'd
calculate the proportion of your sample in each of the ten categories
and weight so that the distribution across these matched the general
population.

The problem is that unless you have a very large dataset you're likely
to have too many categories based on the five variables for this to
workable.  Eg, if age is in 5 cats, income in 5, home ownership in 2,
and marital in 3, then you'll have 5*5*2*2*3 (300) categories overall,
and this is likely to produce extreme weights in some groups and skew
the data.  Most likely, you'd be better off conducting an analysis to
identify how each of the variables you've mentioned affects your outcome
measures and then weight only for the most important ones.

Regards,
Chris


-----Original Message-----
From: Valdiserri, Colin [mailto:[hidden email]]
Sent: 14 December 2006 19:42
Subject: Using Multiple Weights

I have a total of 5 weight variables, but need to use them as one
variable. For example I need to weight my data by age, income, gender,
home ownership, and marital status.



How can I use each of the 5 weight variables at once?



Any help is greatly appreciated.



Thanks,



Colin




_______________________________________________________________________

The information transmitted is intended only for the person or entity to
which it is addressed and may contain confidential 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 any computer.
_______________________________________________________________________

[João Duarte (hamilton)]

Colin, take a look at
http://www.spsstools.net/Syntax/Compute/WeightDataBasedOn2orMoreVars.txt

João Duarte
________________________
Hamilton Portugal
[hidden email]

> -----Original Message-----
> From: Valdiserri, Colin [mailto:[hidden email]]
> Sent: 14 December 2006 19:42
> Subject: Using Multiple Weights
>
> I have a total of 5 weight variables, but need to use them as one
> variable. For example I need to weight my data by age, income, gender,
> home ownership, and marital status.
>
>
>
> How can I use each of the 5 weight variables at once?
>
>
>
> Any help is greatly appreciated.
>
>
>
> Thanks,
>
>
>
> Colin
>
>
>
>
> _______________________________________________________________________
>
> The information transmitted is intended only for the person or entity to
> which it is addressed and may contain confidential 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 any computer.
> _______________________________________________________________________
>
>
> --
> No virus found in this incoming message.
> Checked by AVG Free Edition.
> Version: 7.5.432 / Virus Database: 268.15.19/587 - Release Date:
> 14-12-2006 19:28
>
>