Detailed assistance with FUZZY please!

[Sotiria]

Good evening!
I have been reading this forum for a while and your questions and answers helped so much you can't even imagine. I hope that I will be able to assist someone too. I'm a new member so please excuse me for any mistakes or inconvenience. Promise to be careful from now on!

My inquiries concern FUZZY matching..Unfortunately I did not get much help from what is posted online, and, added to that I'm a complete novice..
So, I have managed to successfully download Python Essentials and FUZZY extension (SPSS V22). My goal is to match two groups by age, gender and education. Case group consists of 142 participants and control sample consists of approximately 800 participants.
So, first question: Can FUZZY choose approximately 142 participants from the control group to match the case-group (or am I getting it all wrong)?
Secondly, in order to use FUZZY, do I have to use the command split file, have 2 seperate databases (control and case) or just one variable with values e.g. 0 for control and 1 for case?

Finally, I cannot understand what variables I must use to the FUZZY window!
what are the variables I must use to the requested inputs?


Thanks in advance for the help! I am desperate!
No wonder it is called FUZZY!!

[Jon K Peck]

See below.


Jon Peck (no "h") aka Kim
Senior Software Engineer, IBM
[hidden email]
phone: [phone]




From:        Sotiria <[hidden email]>
To:        [hidden email]
Date:        06/22/2015 11:50 AM
Subject:        [SPSSX-L] Detailed assistance with FUZZY please!
Sent by:        "SPSSX(r) Discussion" <[hidden email]>

---

Good evening!
I have been reading this forum for a while and your questions and answers
helped so much you can't even imagine. I hope that I will be able to assist
someone too. I'm a new member so please excuse me for any mistakes or
inconvenience. Promise to be careful from now on!

My inquiries concern FUZZY matching..Unfortunately I did not get much help
from what is posted online, and, added to that I'm a complete novice..
So, I have managed to successfully download Python Essentials and FUZZY
extension (SPSS V22). My goal is to match two groups by age, gender and
education. Case group consists of 142 participants and control sample
consists of approximately 800 participants.
So, first question: Can FUZZY choose approximately 142 participants from the
control group to match the case-group (or am I getting it all wrong)?

>>>Yes.  By default FUZZY generates one match from the control group for each record in the case group.
Secondly, in order to use FUZZY, do I have to use the command split file,
have 2 separate databases (control and case) or just one variable with
values e.g. 0 for control and 1 for case?

>>>FUZZY supports several different input structures.  If your data are all in one file, you just need a group indicator variable to indicate for each record whether it is a case or a control.  A value of 1 indicates that the record is a case (called demander in the FUZZY doc to reduce terminology confusion) and 0 indicates a control case (called supplier).

Finally, I cannot understand what variables I must use to the FUZZY window!
what are the variables I must use to the requested inputs?

>>>You want to match on age, gender, and education, so those are the variables you select in the "Variables to Match on" control.  Then enter the tolerance for each variable in the match in the Match Tolerance control.  If gender is a string or you want an exact match on it, enter 0 in its slot.  So you might enter 2 0 1 in that control to indicate that a difference of up to 2 in age, exact for gender, and 1 for education is acceptable.


<http://spssx-discussion.1045642.n5.nabble.com/file/n5729885/FUZZY.png>

Thanks in advance for the help! I am desperate!
No wonder it is called FUZZY!!




--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Detailed-assistance-with-FUZZY-please-tp5729885.html
Sent from the SPSSX Discussion mailing list archive at 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

[Sotiria]

Thank you very very much! And what about the last three inputs case id, etc...?

[Jon K Peck]

Use the help button in the dialog box to answer these questions.  If you still have trouble, please ask a specific question.


Jon Peck (no "h") aka Kim
Senior Software Engineer, IBM
[hidden email]
phone: [phone]




From:        Sotiria <[hidden email]>
To:        [hidden email]
Date:        06/22/2015 12:46 PM
Subject:        Re: [SPSSX-L] Detailed assistance with FUZZY please!
Sent by:        "SPSSX(r) Discussion" <[hidden email]>

---

Thank you very very much! And what about the last three inputs case id,
etc...?



--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Detailed-assistance-with-FUZZY-please-tp5729885p5729888.html
Sent from the SPSSX Discussion mailing list archive at 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

[Sotiria]

Ok I think I got it right! Just wanted to be sure! You just helped 5 spss noobs! Thanks a lot!