Longitudinal data
Locked
 
|
Hi, lister,
I have 4 waves of data. the first wave has 2345 cases, 2695, 3112, and 2328 cases in the following 3 waves. I like know to among the 2345 cases in the first wave, how many are followed up in each of the following wave? how can I use ID and wave information to caculate the completed cases for each wave? Thanks -- Yawen ===================== 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 |
|
|
There are a number of ways to accomplish this.
If these are in separate data sets: - sort each file by ID - do MATCH FILES to capture all four into a single data set linked by ID If these are in the one data set that includes ID and Wave: - sort by ID and Wave - Aggregate by ID and create a variable for each wave Once linked (so you have variables Wave1, Wave2, Wave3, Wave4) I also like to have a variable that shows the pattern of waves. One way is: COMPUTE Waves = Wave1*1000 + Wave2*100 + Wave3*10 + Wave4 . If case has all waves the Waves = 1111. If case has only 1 & 2 then Waves= 1100. Dennis Deck, PhD RMC Research Corporation 111 SW Columbia Street, Suite 1200 Portland, Oregon 97201-5843 voice: 503-223-8248 x715 voice: 800-788-1887 x715 fax: 503-223-8248 [hidden email] -----Original Message----- From: Yawen LI [mailto:[hidden email]] Sent: Wednesday, March 12, 2008 8:50 PM Subject: Longitudinal data Hi, lister, I have 4 waves of data. the first wave has 2345 cases, 2695, 3112, and 2328 cases in the following 3 waves. I like know to among the 2345 cases in the first wave, how many are followed up in each of the following wave? how can I use ID and wave information to caculate the completed cases for each wave? Thanks -- Yawen ===================== 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 |
|
|
In reply to this post by Yawen LI
COMPUTE Waves = Wave1*1000 + Wave2*100 + Wave3*10 + Wave4 .
The above creates a new variable (Waves) in a format that shows the binary pattern of waves. Each part represents whether that wave exists with 1=wave found, 0=wave not found. For example, (Wave1*1000) evaluates to either 0 or 1000 depending on whether Wave1 is 0 or 1. Thus 1011 indicates that waves 1,3,4 exist but 2 was missing. There are other ways to deal with it (but seeing the whole pattern like this is very useful). Another approach would be to simply count the number of waves found: COUNT Nwaves = Wave1 Wave2 Wave3 Wave4 (1) . Dennis -----Original Message----- From: Annette Hawkins [mailto:[hidden email]] Sent: Thursday, March 13, 2008 8:23 PM To: Dennis Deck Subject: Re: Longitudinal data Hi Dr. Deck, I understand everything you wrote except COMPUTE Waves. What exactly does that equation do? I am interested because we have Cohort data that we are tracking which are similar to Waves. Thanks. Annette D. Hawkins, Ed. D. Math Instructor Wayne Community College 3000 Wayne Memorial Drive Goldsboro, NC 27534 "Nearly all men can stand adversity, but if you want to test a man's character, give him power." Abraham Lincoln >>> Dennis Deck <[hidden email]> 03/13/08 12:28 AM >>> There are a number of ways to accomplish this. If these are in separate data sets: - sort each file by ID - do MATCH FILES to capture all four into a single data set linked by ID If these are in the one data set that includes ID and Wave: - sort by ID and Wave - Aggregate by ID and create a variable for each wave Once linked (so you have variables Wave1, Wave2, Wave3, Wave4) I also like to have a variable that shows the pattern of waves. One way is: COMPUTE Waves = Wave1*1000 + Wave2*100 + Wave3*10 + Wave4 . If case has all waves the Waves = 1111. If case has only 1 & 2 then Waves= 1100. Dennis Deck, PhD RMC Research Corporation 111 SW Columbia Street, Suite 1200 Portland, Oregon 97201-5843 voice: 503-223-8248 x715 voice: 800-788-1887 x715 fax: 503-223-8248 [hidden email] -----Original Message----- From: Yawen LI [mailto:[hidden email]] Sent: Wednesday, March 12, 2008 8:50 PM Subject: Longitudinal data Hi, lister, I have 4 waves of data. the first wave has 2345 cases, 2695, 3112, and 2328 cases in the following 3 waves. I like know to among the 2345 cases in the first wave, how many are followed up in each of the following wave? how can I use ID and wave information to caculate the completed cases for each wave? Thanks -- Yawen ===================== 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 4̀„Mail correspondence to and from this sender may be subject to the North Carolina public records law and may be disclosed to third parties. ===================== 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 |
|
|
In reply to this post by Dennis Deck
Here is one approach. I assume your waves are numbered 1 to 4.
If not or they occur on different dates, a slighly different approach would be needed. SORT CASES BY ID . * Which wave is this (0=no,1=yes) . COMPUTE Wave1 = (Wave=1) . COMPUTE Wave2 = (Wave=2) . COMPUTE Wave3 = (Wave=3) . COMPUTE Wave4 = (Wave=4) . AGGREGATE Outfile=* /Break=ID /Presorted /Wave1 = MAX(Wave1) /Wave2 = MAX(Wave2) /Wave3 = MAX(Wave3) /Wave4 = MAX(Wave4) . Result file includes one record per subject with: ID Wave1 Wave2 Wave3 Wave4 ________________________________ From: [hidden email] [mailto:[hidden email]] On Behalf Of Yawen LI Sent: Wednesday, March 12, 2008 9:47 PM To: Dennis Deck Subject: Re: Longitudinal data Hi, Dennis, Thanks for the help. My data is in one dataset with ID and wave. Can you show me how to create a new variable for each wave using aggregate function? Thanks Yawen On Wed, Mar 12, 2008 at 9:28 PM, Dennis Deck <[hidden email]> wrote: There are a number of ways to accomplish this. If these are in separate data sets: - sort each file by ID - do MATCH FILES to capture all four into a single data set linked by ID If these are in the one data set that includes ID and Wave: - sort by ID and Wave - Aggregate by ID and create a variable for each wave Once linked (so you have variables Wave1, Wave2, Wave3, Wave4) I also like to have a variable that shows the pattern of waves. One way is: COMPUTE Waves = Wave1*1000 + Wave2*100 + Wave3*10 + Wave4 . If case has all waves the Waves = 1111. If case has only 1 & 2 then Waves= 1100. Dennis Deck, PhD RMC Research Corporation 111 SW Columbia Street, Suite 1200 Portland, Oregon 97201-5843 voice: 503-223-8248 x715 voice: 800-788-1887 x715 fax: 503-223-8248 [hidden email] -----Original Message----- From: Yawen LI [mailto:[hidden email]] Sent: Wednesday, March 12, 2008 8:50 PM Subject: Longitudinal data Hi, lister, I have 4 waves of data. the first wave has 2345 cases, 2695, 3112, and 2328 cases in the following 3 waves. I like know to among the 2345 cases in the first wave, how many are followed up in each of the following wave? how can I use ID and wave information to caculate the completed cases for each wave? Thanks -- Yawen -- Yawen Li Ph.D candidate Program Manager/China Program School of Social Work University of Southern California MRF 347, 669 W 34th St. Los Angeles Tel: 213-740-1391 Email: [hidden email] http://www-scf.usc.edu/~yawenli/ ====================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 |
«
Return to SPSSX Discussion
|
1 view|%1 views
| Free forum by Nabble | Edit this page |