SPSSX Discussion

identical individuals

Classic

List

Threaded

3 messages Options

Tore Viland

identical individuals

Technical info: SPSS 17.0.2

Data: longitudinal panel data

Hi

I have a problem identifying duplicate cases, and recoding these values according to my needs.

I have a short longitudinal panel data set (3 years), in which there are multiple observations for each individual. Each individual is only observed once, for each year, and a unique identity (id) makes it possible to follow the same individual for multiple periods. However, there are also individuals who are observed for periods less than 3 years in these data.

When running a 'duplicate tag’, I receive the values 1, 2, 3, if there are multiple records, and zero otherwise.

What I wish to do is to recode so that I can keep track of all the values labeled 3. This indicates that I need to recode the variable in such a manner that whenever there is a '3' it also need to be '3' where it now it either '2' or '1'. This would allow me to keep only individuals that are observed throughout the data-set.

Thus, the goal of the recode is to keep track of the unique observations that stay throughout the three year time-period.

Best regards,

Tore

(Norway)

Richard Ristow

Re: identical individuals

At 01:26 PM 5/14/2009, Tore Viland wrote:

I have a short longitudinal panel data set (3 years), in which there are multiple observations for each individual. Each individual is observed once, for each year, and a unique identity (id) makes it possible to follow the same individual for multiple periods. However, there are also individuals who are observed for periods less than 3 years in these data.

When running a 'duplicate tag, I receive the values 1, 2, 3, if there are multiple records, and zero otherwise. I wish to keep track of all the values labeled 3. This indicates that I need to recode the variable in such a manner that whenever there is a '3' it also need to be '3' where it now it either '2' or '1'.

I'm not sure what you mean by "running a 'duplicate tag". I thought, "Identify Duplicate Cases..." from the Data menu, but that does something else.

Anyway, here's code that does what I think you want:
|-----------------------------|---------------------------| |Output Created |15-MAY-2009 16:56:10 | |-----------------------------|---------------------------| id 1 2 2 3 4 4 4 4 5 Number of cases read: 9 Number of cases listed: 9 * .... To add the number of observations for the .... . * .... individual to every record: .... . AGGREGATE OUTFILE=* MODE=ADDVARIABLES /BREAK = id /NObs 'Number of observations for this id' = NU. LIST. List |-----------------------------|---------------------------| |Output Created |15-MAY-2009 16:56:11 | |-----------------------------|---------------------------| id NObs 1 1 2 2 2 2 3 1 4 4 4 4 4 4 4 4 5 1 Number of cases read: 9 Number of cases listed: 9=============================
APPENDIX: Test data, and code
=============================

DATA LIST FREE / id. BEGIN DATA 1 2 2 3 4 4 4 4 5 END DATA. FORMATS id (F4). LIST. * .... To add the number of observations for the .... . * .... individual to every record: .... . AGGREGATE OUTFILE=* MODE=ADDVARIABLES /BREAK = id /NObs 'Number of observations for this id' = NU. LIST. ===================== 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

Tore Viland

Re: identical individuals

Thank you Richard,

The code you provided functioned perfectly, and was just what I needed. Now, I am able to perform a set of analysis that I have not been able to do, due to the difficulties I had figuring out how to create the variable. Thank you for teaching me!

And yes, I was trying to use the "Identify Duplicate Cases", which apparently did not work at all.

Best,

Tore

On Fri, May 15, 2009 at 10:58 PM, Richard Ristow <[hidden email]> wrote:

At 01:26 PM 5/14/2009, Tore Viland wrote:

I have a short longitudinal panel data set (3 years), in which there are multiple observations for each individual. Each individual is observed once, for each year, and a unique identity (id) makes it possible to follow the same individual for multiple periods. However, there are also individuals who are observed for periods less than 3 years in these data.

When running a 'duplicate tag’, I receive the values 1, 2, 3, if there are multiple records, and zero otherwise. I wish to keep track of all the values labeled 3. This indicates that I need to recode the variable in such a manner that whenever there is a '3' it also need to be '3' where it now it either '2' or '1'.

I'm not sure what you mean by "running a 'duplicate tag’". I thought, "Identify Duplicate Cases..." from the Data menu, but that does something else.

Anyway, here's code that does what I think you want:
|-----------------------------|---------------------------| |Output Created |15-MAY-2009 16:56:10 | |-----------------------------|---------------------------| id 1 2 2 3 4 4 4 4 5 Number of cases read: 9 Number of cases listed: 9 * .... To add the number of observations for the .... . * .... individual to every record: .... . AGGREGATE OUTFILE=* MODE=ADDVARIABLES /BREAK = id /NObs 'Number of observations for this id' = NU. LIST. List |-----------------------------|---------------------------| |Output Created |15-MAY-2009 16:56:11 | |-----------------------------|---------------------------| id NObs 1 1 2 2 2 2 3 1 4 4 4 4 4 4 4 4 5 1 Number of cases read: 9 Number of cases listed: 9=============================
APPENDIX: Test data, and code
=============================

DATA LIST FREE / id. BEGIN DATA 1 2 2 3 4 4 4 4 5 END DATA. FORMATS id (F4). LIST. * .... To add the number of observations for the .... . * .... individual to every record: .... . AGGREGATE OUTFILE=* MODE=ADDVARIABLES /BREAK = id /NObs 'Number of observations for this id' = NU. LIST.