Date:         Thu, 5 Oct 2006 11:06:50 -0700
Reply-To:     Nico Peruzzi <nperuzzi@gmail.com>
Sender:       "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From:         Nico Peruzzi <nperuzzi@gmail.com>
Subject:      how to merge/compare across multiple datasets with duplicate IDs
Content-Type: text/plain; charset=ISO-8859-1; format=flowed

Hi listers,

I've got a pile of data that came in 3 datasets.  They all have an ID
variable, however here's the trick.

dataset #1 has ID + some demographic variables (there is 1 case per ID)
dataset #2 has ID + some variables related to someone's account (note that
an ID can have more than 1 acocunt) (there is 1 case per account #)
dataset #3 has ID + vehicle variables (note that an ID can have more than 1
vehicle) (there is 1 case per vehicle ID)

here's an example of one deceptively simple-sounding chart I need to create:

Show frequencies of age ranges (comes from dataset #1) based on type of
account (comes from dataset #2)

here's another example:

Show frequencies for each vehicle type (comes from dataset #3) based on
account type (from #2) and ownership status (from #1)

All would be great if I could just merge on ID, but as I mentioned above the
only variable across all 3 datasets is ID, and there will be multiple
occurences of ID in datasets #2 and #3.

Any thoughts on how to work through or around this?

Thanks in advance, Nico

--
Nico Peruzzi, Ph.D.