CaseControl Matching
Locked
 
|
Hey,
I'm trying to do case control matching with SPSS in 1:2 rate. I first had problems with a warning of "only string variable are allowed" not the FUZZY command works fine but I get only one match and the rest cases (370) are not matched my syntax: FUZZY BY = Count Division Age SUPPLIERID = ID NEWDEMANDERIDVARS = ControlA ControlB GROUP = Index FUZZ = 3 0 5 EXACTPRIORITY = TRUE MATCHGROUPVAR = Cool DRAWPOOLSIZE = Poolee DEMANDERID = ID DS3 = howcome /OPTIONS SAMPLEWITHREPLACEMENT=FALSE MINIMIZEMEMORY=TRUE SHUFFLE=FALSE. Can anybody tell me why is happens and how can I fix it? Thanks in advance -- Sent from: http://spssx-discussion.1045642.n5.nabble.com/ ===================== 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 |
|
|
Side question: Where is the FUZZY command found? I don't see it in 24. Is it new in 25 or is this an extension procedure? Just curious.
Gene Maguin -----Original Message----- From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Nasreen Sent: Thursday, December 7, 2017 6:15 AM To: [hidden email] Subject: CaseControl Matching Hey, I'm trying to do case control matching with SPSS in 1:2 rate. I first had problems with a warning of "only string variable are allowed" not the FUZZY command works fine but I get only one match and the rest cases (370) are not matched my syntax: FUZZY BY = Count Division Age SUPPLIERID = ID NEWDEMANDERIDVARS = ControlA ControlB GROUP = Index FUZZ = 3 0 5 EXACTPRIORITY = TRUE MATCHGROUPVAR = Cool DRAWPOOLSIZE = Poolee DEMANDERID = ID DS3 = howcome /OPTIONS SAMPLEWITHREPLACEMENT=FALSE MINIMIZEMEMORY=TRUE SHUFFLE=FALSE. Can anybody tell me why is happens and how can I fix it? Thanks in advance -- Sent from: http://spssx-discussion.1045642.n5.nabble.com/ ===================== 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 ===================== 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 |
|
|
FUZZY (Data > Case Controls) is an old extension command written in Python. It can be installed from the Utilities or Extensions menu if not already present. FUZZY is also used by the Data > Propensity Score Matching custom dialog, which is simpler to use but does not have all the features of FUZZY. On Fri, Dec 8, 2017 at 7:54 AM, Maguin, Eugene <[hidden email]> wrote: Side question: Where is the FUZZY command found? I don't see it in 24. Is it new in 25 or is this an extension procedure? Just curious. |
|
|
In reply to this post by Nasreen
The number of variables listed as NEWDEMANDERIDVARS determines how many matches are attempted, so you should get two id's per case. Your group variable should have just 0 and 1 values to distinguish the cases and controls. Are you getting a match only for the first case, or are you just getting one match per case? The statistics on the Poolee variable should show you the count of available cases for each attempted match. Beyond that, I would have to see the data to diagnose further. You can send it to me at [hidden email] if you want. On Thu, Dec 7, 2017 at 4:14 AM, Nasreen <[hidden email]> wrote: Hey, |
«
Return to SPSSX Discussion
|
1 view|%1 views
| Free forum by Nabble | Edit this page |