question on merging datasets

[Deepa Bhat]

Hi everyone,

I have 2 datasets. One dataset is the baseline and one dataset is the
follow-up. I would like to merge them for the analysis.
The baseline had 29 questions and the follow-up has 73 questions.
There were 26 questions that were the same in both questionnaires.
There were 895 cases in baseline and 911 cases in follow-up. They are
not the same people.

In the SPSS help menu, there is an option to add cases or add
variables. But what about if you want to add both?

Any guidance would be appreciated.

Thanks so much,
Deepa

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

[Melissa Ives]

If you have an indicator variable for baseline vs. follow-up, use add cases first, then use CASESTOVARS to go from vertical (tall) to horizontal (long), using that indicator as the index variable.

If you don't have an indicator variable, create one such as wave=1 for baseline, wave=2 for follow-up. Then add cases and use CASESTOVARS.

Melissa

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Deepa Bhat
Sent: Monday, April 14, 2008 9:41 AM
To: [hidden email]
Subject: [SPSSX-L] question on merging datasets

Hi everyone,

I have 2 datasets. One dataset is the baseline and one dataset is the follow-up. I would like to merge them for the analysis.
The baseline had 29 questions and the follow-up has 73 questions.
There were 26 questions that were the same in both questionnaires.
There were 895 cases in baseline and 911 cases in follow-up. They are not the same people.

In the SPSS help menu, there is an option to add cases or add variables. But what about if you want to add both?

Any guidance would be appreciated.

Thanks so much,
Deepa

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

PRIVILEGED AND CONFIDENTIAL INFORMATION
This transmittal and any attachments may contain PRIVILEGED AND
CONFIDENTIAL information and is intended only for the use of the
addressee. If you are not the designated recipient, or an employee
or agent authorized to deliver such transmittals to the designated
recipient, you are hereby notified that any dissemination,
copying or publication of this transmittal is strictly prohibited. If
you have received this transmittal in error, please notify us
immediately by replying to the sender and delete this copy from your
system. You may also call us at [phone] for assistance.

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

[Richard Ristow]

At 10:41 AM 4/14/2008, Deepa Bhat wrote:

>I have 2 datasets. One dataset is the baseline and one dataset is
>the follow-up. I would like to merge them for the analysis. The
>baseline had 29 questions and the follow-up has 73 questions. There
>were 26 questions that were the same in both questionnaires. In the
>SPSS help menu, there is an option to add cases or add variables.
>But what about if you want to add both?

Melissa's right: use "add cases". The result will have all variables
found in either input. Variables will have value system-missing
(numeric variables) or blank (string variables) for cases from
sources where they don't appear.

The SPSS menus do about as well as is reasonable to do, but they
don't express the full extent of what the merging commands actually do.

Now, however, at 10:46 AM 4/14/2008, Melissa Ives wrote:

>If you have an indicator variable for baseline vs. follow-up, use
>add cases [...] If you don't have an indicator variable, create one
>such as wave=1 for baseline, wave=2 for follow-up.

So far so good, except it isn't necessary to create the indicator
variable first. On the panel that lists the variables, check
"Indicate case source as variable", and give name 'wave', or whatever
you choose. That'll give a variable whose value is 1 if the case came
from the active dataset, and 0 if it came from the dataset that's
merged in with it. If the active dataset is the baseline and you
merged is the followup, you can get the coding Melissa suggested by menu

Transform > Recode > Into same variables...

Select variable 'wave'. Select 'Old and New Values...';
Under "Old Value:", select "Value" and type in value 0;
under "New Value:", select "Value" and type in value 2;
select "Add".

The resulting syntax is

RECODE
   wave  (0=2)  .

Melissa also suggested,

>use CASESTOVARS to go from vertical (tall) to horizontal (long),
>using the indicator as the index variable.

No, not in your case where, as you wrote,

>The cases in baseline and in follow-up are not the same people.

(BTW, by way of terminology: Using 'long' for horizontal organization
may be misleading, as 'long' is commonly used -- by me as well as
others -- to mean  what's here called 'vertical' or 'tall.' What's
called 'horizontal' is also commonly referred to as 'wide'.)

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