Data updates
Locked
 
|
I have a small file (85 cases), and I want to update it from a larger file
that has the same cases, but many others as well. I don't want the other cases and their information; I just want the updates. That is, the smaller file has missing data that exists on the bigger file. I do not expect that the information in the two files would differ in any way except for the missing values. That is, a case from both files either has the same info for any given variable, or the information will be missing from the smaller file. Which of the data merging/matching/updating/adding commands is the most appropriate for this purpose? Thanks, Bob Schacht Robert M. Schacht, Ph.D. <[hidden email]> Pacific Basin Rehabilitation Research & Training Center 1268 Young Street, Suite #204 Research Center, University of Hawaii Honolulu, HI 96814 ===================== 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 |
|
|
Sounds like the UPDATE command will do it or is this a trick question?
-----Original Message----- From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Bob Schacht Sent: Wednesday, April 22, 2009 7:44 PM To: [hidden email] Subject: Data updates I have a small file (85 cases), and I want to update it from a larger file that has the same cases, but many others as well. I don't want the other cases and their information; I just want the updates. That is, the smaller file has missing data that exists on the bigger file. I do not expect that the information in the two files would differ in any way except for the missing values. That is, a case from both files either has the same info for any given variable, or the information will be missing from the smaller file. Which of the data merging/matching/updating/adding commands is the most appropriate for this purpose? Thanks, Bob Schacht Robert M. Schacht, Ph.D. <[hidden email]> Pacific Basin Rehabilitation Research & Training Center 1268 Young Street, Suite #204 Research Center, University of Hawaii Honolulu, HI 96814 ===================== 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 |
|
|
In reply to this post by Bob Schacht-3
Dear listers
I need to decide whether our team will upgrade to v17.0 (not the new PASW v17.0.2) or stay with v15 for another year. Any advice? Thank you in advance. Wincy ===================== 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 |
|
|
Hi Wincy!
I would recommend you to stay in SPSSS 15. 16 was lousy, and 17 is certainly better, but has still major shortcomings. I would like to know if there is anyone who thinks that 17 is better than 15? Wilhelm Landerholm Queue Sweden AB - www.qsweden.com STATB.COM - www.statb.com +46-735-460000 -----Ursprungligt meddelande----- Från: SPSSX(r) Discussion [mailto:[hidden email]] För Wincy Chan Skickat: den 23 april 2009 06:59 Till: [hidden email] Ämne: v17.0 Dear listers I need to decide whether our team will upgrade to v17.0 (not the new PASW v17.0.2) or stay with v15 for another year. Any advice? Thank you in advance. Wincy ===================== 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 |
|
|
Hi Wilhelm
I don't think it's better - I have been pleased at the improvements 17 brought over the ghastly version 16 and some of those in patches/bug-fixes in 17 (although NOT the one which renders output unreadable!). Version 17 is pleasingly faster than 15, but for my biggest projects, I have to keep using 15 because of bugs in 17 which just stop some programs from working. With long data-restructuring programs like the ones I'm working on, I simply can't afford the time when de-bugging these things, trying to work out if an error message is due to an error of mine or is due to one of these annoying bugs. When you add in the additional time involved in switching to print preview mode to wade through the output looking for bugs, it simply does not add up to being worth it. For me, the problems thoroughly outweigh the benefits. (For the record, once I have completed the current updates on my programs, in 15, I will report the bugs to SPSS Support, but right now I do not have the time to do all the work in isolating the behaviour and preparing a sufficiently small, de-identified set of data with which to demonstrate the problem. I'd been hoping that the first major patch would fix the bug, but, for all the huge size of the patch, the list of bugs it fixed was remarkably small. And it didn't fix my bugs :-( ) Others doing different types of work may not be impacted by the things that affect my type of work though and may find it perfectly fine. Regards, Adrian Barnett Information Analyst Educational Measurement and Analysis Data Management DECS ph 82261080 -----Original Message----- From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Wilhelm Landerholm | Queue Sent: Thursday, 23 April 2009 3:01 PM To: [hidden email] Subject: SV: v17.0 Hi Wincy! I would recommend you to stay in SPSSS 15. 16 was lousy, and 17 is certainly better, but has still major shortcomings. I would like to know if there is anyone who thinks that 17 is better than 15? Wilhelm Landerholm Queue Sweden AB - www.qsweden.com STATB.COM - www.statb.com +46-735-460000 -----Ursprungligt meddelande----- Från: SPSSX(r) Discussion [mailto:[hidden email]] För Wincy Chan Skickat: den 23 april 2009 06:59 Till: [hidden email] Ämne: v17.0 Dear listers I need to decide whether our team will upgrade to v17.0 (not the new PASW v17.0.2) or stay with v15 for another year. Any advice? Thank you in advance. Wincy ===================== 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 ===================== 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 |
|
|
Hi Wilhelm and Adrian
Thank you for your insights. I need to handle panel student data over years and from various surveys conducted each year. We are in the process of setting up a data base that will manage the data collected from both paper (input through OMR scanner) and an online survey system. That means we need to handle large number of cases for subgroup analyses across years (i.e. merging syntax used in various years). V15 is doing a good job now. My main concern is whether we will need to tremendously revise our old syntax if we upgrade to V17.0 now. We will likely have the database system set up in the next few months and by then we will definitely need to synchronize the data from previous years - so in between I wonder what will be the best solution to keep update with the technology while minimizing the hassles from rewriting the syntax. Any suggestion? Thanks Wincy -----Original Message----- From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Barnett, Adrian (DECS) Sent: Thursday, April 23, 2009 2:07 PM To: [hidden email] Subject: Re: v17.0 Hi Wilhelm I don't think it's better - I have been pleased at the improvements 17 brought over the ghastly version 16 and some of those in patches/bug-fixes in 17 (although NOT the one which renders output unreadable!). Version 17 is pleasingly faster than 15, but for my biggest projects, I have to keep using 15 because of bugs in 17 which just stop some programs from working. With long data-restructuring programs like the ones I'm working on, I simply can't afford the time when de-bugging these things, trying to work out if an error message is due to an error of mine or is due to one of these annoying bugs. When you add in the additional time involved in switching to print preview mode to wade through the output looking for bugs, it simply does not add up to being worth it. For me, the problems thoroughly outweigh the benefits. (For the record, once I have completed the current updates on my programs, in 15, I will report the bugs to SPSS Support, but right now I do not have the time to do all the work in isolating the behaviour and preparing a sufficiently small, de-identified set of data with which to demonstrate the problem. I'd been hoping that the first major patch would fix the bug, but, for all the huge size of the patch, the list of bugs it fixed was remarkably small. And it didn't fix my bugs :-( ) Others doing different types of work may not be impacted by the things that affect my type of work though and may find it perfectly fine. Regards, Adrian Barnett Information Analyst Educational Measurement and Analysis Data Management DECS ph 82261080 -----Original Message----- From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Wilhelm Landerholm | Queue Sent: Thursday, 23 April 2009 3:01 PM To: [hidden email] Subject: SV: v17.0 Hi Wincy! I would recommend you to stay in SPSSS 15. 16 was lousy, and 17 is certainly better, but has still major shortcomings. I would like to know if there is anyone who thinks that 17 is better than 15? Wilhelm Landerholm Queue Sweden AB - www.qsweden.com STATB.COM - www.statb.com +46-735-460000 -----Ursprungligt meddelande----- Från: SPSSX(r) Discussion [mailto:[hidden email]] För Wincy Chan Skickat: den 23 april 2009 06:59 Till: [hidden email] Ämne: v17.0 Dear listers I need to decide whether our team will upgrade to v17.0 (not the new PASW v17.0.2) or stay with v15 for another year. Any advice? Thank you in advance. Wincy ===================== 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 ===================== 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 |
|
|
In reply to this post by Barnett, Adrian (DECS)
Hi Adrian!
Thanks for your answer! I recognize "... work out if an error message is due to an error of mine or is due to one of these annoying bugs." SPSS 16 and 17 has actually resulted in me working more and more in R. All the best Wilhelm Landerholm Queue Sweden AB - www.qsweden.com STATB.COM - www.statb.com +46-735-460000 -----Ursprungligt meddelande----- Från: Barnett, Adrian (DECS) [mailto:[hidden email]] Skickat: den 23 april 2009 08:07 Till: Wilhelm Landerholm | Queue; [hidden email] Ämne: RE: v17.0 Hi Wilhelm I don't think it's better - I have been pleased at the improvements 17 brought over the ghastly version 16 and some of those in patches/bug-fixes in 17 (although NOT the one which renders output unreadable!). Version 17 is pleasingly faster than 15, but for my biggest projects, I have to keep using 15 because of bugs in 17 which just stop some programs from working. With long data-restructuring programs like the ones I'm working on, I simply can't afford the time when de-bugging these things, trying to work out if an error message is due to an error of mine or is due to one of these annoying bugs. When you add in the additional time involved in switching to print preview mode to wade through the output looking for bugs, it simply does not add up to being worth it. For me, the problems thoroughly outweigh the benefits. (For the record, once I have completed the current updates on my programs, in 15, I will report the bugs to SPSS Support, but right now I do not have the time to do all the work in isolating the behaviour and preparing a sufficiently small, de-identified set of data with which to demonstrate the problem. I'd been hoping that the first major patch would fix the bug, but, for all the huge size of the patch, the list of bugs it fixed was remarkably small. And it didn't fix my bugs :-( ) Others doing different types of work may not be impacted by the things that affect my type of work though and may find it perfectly fine. Regards, Adrian Barnett Information Analyst Educational Measurement and Analysis Data Management DECS ph 82261080 -----Original Message----- From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Wilhelm Landerholm | Queue Sent: Thursday, 23 April 2009 3:01 PM To: [hidden email] Subject: SV: v17.0 Hi Wincy! I would recommend you to stay in SPSSS 15. 16 was lousy, and 17 is certainly better, but has still major shortcomings. I would like to know if there is anyone who thinks that 17 is better than 15? Wilhelm Landerholm Queue Sweden AB - www.qsweden.com STATB.COM - www.statb.com +46-735-460000 -----Ursprungligt meddelande----- Från: SPSSX(r) Discussion [mailto:[hidden email]] För Wincy Chan Skickat: den 23 april 2009 06:59 Till: [hidden email] Ämne: v17.0 Dear listers I need to decide whether our team will upgrade to v17.0 (not the new PASW v17.0.2) or stay with v15 for another year. Any advice? Thank you in advance. Wincy ===================== 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 ===================== 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 Wincy Chan
Hi,
I have never heard of someone who has had major problems with syntax. Some small things only. Do you use Script? In this case, you may need to rewrite a lot of things to Python. All the best Wilhelm Landerholm Queue Sweden AB - www.qsweden.com STATB.COM - www.statb.com +46-735-460000 -----Ursprungligt meddelande----- Från: SPSSX(r) Discussion [mailto:[hidden email]] För Wincy Chan Skickat: den 23 april 2009 08:26 Till: [hidden email] Ämne: Re: v17.0 Hi Wilhelm and Adrian Thank you for your insights. I need to handle panel student data over years and from various surveys conducted each year. We are in the process of setting up a data base that will manage the data collected from both paper (input through OMR scanner) and an online survey system. That means we need to handle large number of cases for subgroup analyses across years (i.e. merging syntax used in various years). V15 is doing a good job now. My main concern is whether we will need to tremendously revise our old syntax if we upgrade to V17.0 now. We will likely have the database system set up in the next few months and by then we will definitely need to synchronize the data from previous years - so in between I wonder what will be the best solution to keep update with the technology while minimizing the hassles from rewriting the syntax. Any suggestion? Thanks Wincy -----Original Message----- From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Barnett, Adrian (DECS) Sent: Thursday, April 23, 2009 2:07 PM To: [hidden email] Subject: Re: v17.0 Hi Wilhelm I don't think it's better - I have been pleased at the improvements 17 brought over the ghastly version 16 and some of those in patches/bug-fixes in 17 (although NOT the one which renders output unreadable!). Version 17 is pleasingly faster than 15, but for my biggest projects, I have to keep using 15 because of bugs in 17 which just stop some programs from working. With long data-restructuring programs like the ones I'm working on, I simply can't afford the time when de-bugging these things, trying to work out if an error message is due to an error of mine or is due to one of these annoying bugs. When you add in the additional time involved in switching to print preview mode to wade through the output looking for bugs, it simply does not add up to being worth it. For me, the problems thoroughly outweigh the benefits. (For the record, once I have completed the current updates on my programs, in 15, I will report the bugs to SPSS Support, but right now I do not have the time to do all the work in isolating the behaviour and preparing a sufficiently small, de-identified set of data with which to demonstrate the problem. I'd been hoping that the first major patch would fix the bug, but, for all the huge size of the patch, the list of bugs it fixed was remarkably small. And it didn't fix my bugs :-( ) Others doing different types of work may not be impacted by the things that affect my type of work though and may find it perfectly fine. Regards, Adrian Barnett Information Analyst Educational Measurement and Analysis Data Management DECS ph 82261080 -----Original Message----- From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Wilhelm Landerholm | Queue Sent: Thursday, 23 April 2009 3:01 PM To: [hidden email] Subject: SV: v17.0 Hi Wincy! I would recommend you to stay in SPSSS 15. 16 was lousy, and 17 is certainly better, but has still major shortcomings. I would like to know if there is anyone who thinks that 17 is better than 15? Wilhelm Landerholm Queue Sweden AB - www.qsweden.com STATB.COM - www.statb.com +46-735-460000 -----Ursprungligt meddelande----- Från: SPSSX(r) Discussion [mailto:[hidden email]] För Wincy Chan Skickat: den 23 april 2009 06:59 Till: [hidden email] Ämne: v17.0 Dear listers I need to decide whether our team will upgrade to v17.0 (not the new PASW v17.0.2) or stay with v15 for another year. Any advice? Thank you in advance. Wincy ===================== 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 ===================== 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 ===================== 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 |
|
|
Dear Wilhelm, Adrian, and Mārtiņš
Thank you once again for the helpful advice. We will thoroughly test it before upgrading - will have to upgrade sooner or later! Cheers Wincy -----Original Message----- From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Wilhelm Landerholm | Queue Sent: Thursday, April 23, 2009 2:37 PM To: [hidden email] Subject: SV: v17.0 Hi, I have never heard of someone who has had major problems with syntax. Some small things only. Do you use Script? In this case, you may need to rewrite a lot of things to Python. All the best Wilhelm Landerholm Queue Sweden AB - www.qsweden.com STATB.COM - www.statb.com +46-735-460000 -----Ursprungligt meddelande----- Från: SPSSX(r) Discussion [mailto:[hidden email]] För Wincy Chan Skickat: den 23 april 2009 08:26 Till: [hidden email] Ämne: Re: v17.0 Hi Wilhelm and Adrian Thank you for your insights. I need to handle panel student data over years and from various surveys conducted each year. We are in the process of setting up a data base that will manage the data collected from both paper (input through OMR scanner) and an online survey system. That means we need to handle large number of cases for subgroup analyses across years (i.e. merging syntax used in various years). V15 is doing a good job now. My main concern is whether we will need to tremendously revise our old syntax if we upgrade to V17.0 now. We will likely have the database system set up in the next few months and by then we will definitely need to synchronize the data from previous years - so in between I wonder what will be the best solution to keep update with the technology while minimizing the hassles from rewriting the syntax. Any suggestion? Thanks Wincy -----Original Message----- From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Barnett, Adrian (DECS) Sent: Thursday, April 23, 2009 2:07 PM To: [hidden email] Subject: Re: v17.0 Hi Wilhelm I don't think it's better - I have been pleased at the improvements 17 brought over the ghastly version 16 and some of those in patches/bug-fixes in 17 (although NOT the one which renders output unreadable!). Version 17 is pleasingly faster than 15, but for my biggest projects, I have to keep using 15 because of bugs in 17 which just stop some programs from working. With long data-restructuring programs like the ones I'm working on, I simply can't afford the time when de-bugging these things, trying to work out if an error message is due to an error of mine or is due to one of these annoying bugs. When you add in the additional time involved in switching to print preview mode to wade through the output looking for bugs, it simply does not add up to being worth it. For me, the problems thoroughly outweigh the benefits. (For the record, once I have completed the current updates on my programs, in 15, I will report the bugs to SPSS Support, but right now I do not have the time to do all the work in isolating the behaviour and preparing a sufficiently small, de-identified set of data with which to demonstrate the problem. I'd been hoping that the first major patch would fix the bug, but, for all the huge size of the patch, the list of bugs it fixed was remarkably small. And it didn't fix my bugs :-( ) Others doing different types of work may not be impacted by the things that affect my type of work though and may find it perfectly fine. Regards, Adrian Barnett Information Analyst Educational Measurement and Analysis Data Management DECS ph 82261080 -----Original Message----- From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Wilhelm Landerholm | Queue Sent: Thursday, 23 April 2009 3:01 PM To: [hidden email] Subject: SV: v17.0 Hi Wincy! I would recommend you to stay in SPSSS 15. 16 was lousy, and 17 is certainly better, but has still major shortcomings. I would like to know if there is anyone who thinks that 17 is better than 15? Wilhelm Landerholm Queue Sweden AB - www.qsweden.com STATB.COM - www.statb.com +46-735-460000 -----Ursprungligt meddelande----- Från: SPSSX(r) Discussion [mailto:[hidden email]] För Wincy Chan Skickat: den 23 april 2009 06:59 Till: [hidden email] Ämne: v17.0 Dear listers I need to decide whether our team will upgrade to v17.0 (not the new PASW v17.0.2) or stay with v15 for another year. Any advice? Thank you in advance. Wincy ===================== 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 ===================== 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 ===================== 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 |
|
|
In reply to this post by Wincy Chan
Wincy,
I have been using both v 15.0 and 17.01. If your hardware resources are limited ( less than 2 GB of RAM and a slow or older CPU (Pentium 4 or less), I advise staying with 15.0. That being said, if you meet the hardware test, I have been quite happy with the new features of v. 17.0 and it's overall stability and would make the move. For your basic SPSS user I think v. 15.0 is the best release in a long time..... MK ===================== 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 ViAnn Beadle
At 05:41 PM 4/22/2009, ViAnn Beadle wrote:
Update doesn't add cases--it only updates cases in the master file from the Thanks. However, there are several things I don't understand. I tried using the Update command thusly: GET The results were as follows: SORT CASES BY SSN(A). I don't understand this. It seems to be refering to the A in SSN(A)-- should I just delete the (A), or is there something else wrong? And if the command was not executed, why was the MAP executed-- because that's what I get next, beginning with the following: Map of BY variables There is much more here, including a long list of variables from Input2 that I didn't want. Then, at the end, >Note # 5146 So, despite the MAP, it did not do the update, apparently.
|
|
|
In reply to this post by Bob Schacht-3
Hi Bob,
You could try something along the following lines (see below). Using FILE HANDLES makes the code so much more readable. And yes, you should delete the (A), which was probably copied from the sort (A) [Ascending] command. Furthermore, if you use the DATASET NAME command, you don't need to save the intermediate files, which saves time and disk space. If you don't need all vars, you can specify /DROP = (your var list here) in the UPDATE command. FILE HANDLE schacht /name = 'C:\Documents and Settings\R. Schacht\My Documents\1 BASIC VR File Folder\12 Working 08 projects\Satisfaction\FutureSurvey" GET FILE='schacht\Satisfaction\FutureSurvey\LongQ_with_85usable_onlyVer2.sav'. SORT CASES BY SSN (A). dataset name thisone. GET FILE='schacht\12 Working 08 projects\VRISS\3QuExt2008.sav'. SORT CASES BY SSN (A). dataset name thatone. UPDATE FILE=thisone /IN=IN_LongQ /FILE= thatone /IN=IN_3Q2008 /BY SSN /MAP. SAVE OUTFILE = 'schacht\updated.sav'. DATASET CLOSE ALL. Cheers!! Albert-Jan --- On Fri, 4/24/09, Bob Schacht <[hidden email]> wrote: > From: Bob Schacht <[hidden email]> > Subject: Re: Data updates > To: [hidden email] > Date: Friday, April 24, 2009, 3:54 AM > > > At 05:41 PM 4/22/2009, ViAnn Beadle wrote: > > Update > doesn't add cases--it > only updates cases in the master file from the > > transaction file. The master file is your 85 cases, the > transaction file > is > > a larger file. > > > > Thanks. However, there are several things I don't > understand. I > tried using the Update command thusly: > > > GET > > FILE='C:\Documents and Settings\R. > Schacht\My Documents\1 BASIC VR > File Folder\12 Working 08 > projects\Satisfaction\FutureSurvey\' + > > > 'LongQ_with_85usable_onlyVer2.sav'. > > SORT CASES BY > > SSN (A) . > > SAVE OUTFILE='C:\Documents and Settings\R. > Schacht\My Documents\1 BASIC > VR File Folder\12 Working 08 > projects\Satisfaction\' > + > > > 'FutureSurvey\LongQ_with_85usable_onlyVer2.sav' > > /COMPRESSED. > > GET > > FILE='C:\Documents and Settings\R. > Schacht\My Documents\1 BASIC VR > File Folder\12 Working 08 > projects\VRISS\3QuExt2008.sav'. > > DATASET NAME DataSet2 WINDOW=FRONT. > > SORT CASES BY SSN(A). > > > SAVE OUTFILE='C:\Documents and Settings\R. > Schacht\My Documents\1 BASIC > VR File Folder\12 Working 08 > projects\VRISS\3QuExt2008.sav' > > /COMPRESSED. > > > UPDATE FILE='C:\Documents and Settings\R. > Schacht\My Documents\1 BASIC VR > File Folder\12 Working 08 > projects\Satisfaction\FutureSurvey\' > + > > > 'LongQ_with_85usable_onlyVer2.sav' > /IN=IN_LongQ > > /FILE='C:\Documents and Settings\R. > Schacht\My Documents\1 BASIC VR File > Folder\12 Working 08 > projects\VRISS\3QuExt2008.sav'/IN=IN_3Q2008 > > /BY SSN(A) > > /MAP. > > The results were as follows: > > > SORT CASES BY > SSN(A). > > SAVE OUTFILE='C:\Documents and Settings\R. > Schacht\My Documents\1 BASIC > VR File Folder\12 Working 08 > projects\VRISS\3QuExt2008.sav' > > /COMPRESSED. > > /* Rather than Match, > use UPDATE, following Richard Ristow's suggestion > (5/1/2006) > > UPDATE FILE='C:\Documents and Settings\R. > Schacht\My Documents\1 BASIC VR > File Folder\12 Working 08 > projects\Satisfaction\FutureSurv > > ey\' + > > > 'LongQ_with_85usable_onlyVer2.sav' > /IN=IN_LongQ > > /FILE='C:\Documents and Settings\R. > Schacht\My Documents\1 BASIC VR File > Folder\12 Working 08 > projects\VRISS\3QuExt2008.sav'/IN=IN_3 > > Q2008 > > /BY SSN(A) > > /MAP. > > > >Error # 5101 in column 9. Text: A > > >Invalid subcommand for this command. Check for > missing > slashes. Syntax > > >checking begins with the next slash. > > >This command not executed. > > I don't understand this. It seems to be refering to the > A in SSN(A)-- > should I just delete the (A), or is there something else > wrong? > > And if the command was not executed, why was the MAP > executed-- because > that's what I get next, beginning with the following: > > > Map of BY > variables > > > Result > Input1 > Input2 > > ------ > ------ > ------ > > > SSN > SSN > SSN > > > Map of the result file > > > Result > Input1 > Input2 > > ------ > ------ > ------ > > > SSN > SSN > SSN > > There is much more here, including a long list of variables > from Input2 > that I didn't want. Then, at the end, > > >Note # > 5146 > > >There is no working file to restore. You must > define a working > file before > > >proceeding with your analysis. > > > So, despite the MAP, it did not do the update, apparently. > > > > > Do I fix the problem with the "A" by simply > deleting the > "(A)" following the "/BY SSN"? > How do I get the update without all the extra variables > added on? > > > > > > > > ===================== 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 |
|
|
At 05:27 AM 4/24/2009, Albert-jan Roskam wrote:
>Hi Bob, > >You could try something along the following lines (see below). Using FILE >HANDLES makes the code so much more readable. And yes, you should delete >the (A), which was probably copied from the sort (A) [Ascending] command. >Furthermore, if you use the DATASET NAME command, you don't need to save >the intermediate files, which saves time and disk space. If you don't need >all vars, you can specify /DROP = (your var list here) in the UPDATE command. > >FILE HANDLE schacht /name = 'C:\Documents and Settings\R. Schacht\My >Documents\1 BASIC VR File Folder\12 Working 08 >projects\Satisfaction\FutureSurvey" > >GET FILE='schacht\Satisfaction\FutureSurvey\LongQ_with_85usable_onlyVer2.sav'. >SORT CASES BY SSN (A). >dataset name thisone. > >GET FILE='schacht\12 Working 08 projects\VRISS\3QuExt2008.sav'. >SORT CASES BY SSN (A). >dataset name thatone. > >UPDATE FILE=thisone /IN=IN_LongQ /FILE= thatone /IN=IN_3Q2008 /BY SSN /MAP. >SAVE OUTFILE = 'schacht\updated.sav'. >DATASET CLOSE ALL. > >Cheers!! >Albert-Jan OK, thanks to you and to ViAnn Beadle I got the update to work. I'm feeling like an old mossback who has used SPSS for about 40 years, sporadically, without keeping up on all of the innovations. So with assistance from both of you, I used this update: >UPDATE FILE=LongQ_with_85 /IN=IN_LongQ /FILE= ThirdQuExt2008 /IN=IN_3Q2008 >/BY SSN /MAP. The original LongQ_with_85 dataset had only 85 cases; the resulting dataset, however, had not only these 85 cases, but 6700 more from the second dataset (ThirdQuExt2008), and also included an extra 120 variables from the second dataset that I didn't want. I was able to delete the extra cases by sorting the cases on a variable occurring only in the 85-case set, and deleting the unselected cases. And I can probably figure out a way to delete all the unwanted variables. But what I wanted was a merger, match or update process that would be limited to the cases and variables in my 85-case dataset. Is there a way to do that without the tedium of specifying all of the unwanted variables with a DROP command? Thanks, Bob Schacht Robert M. Schacht, Ph.D. <[hidden email]> Pacific Basin Rehabilitation Research & Training Center 1268 Young Street, Suite #204 Research Center, University of Hawaii Honolulu, HI 96814 ===================== 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 |
|
|
The following syntax defines two files and then updates the first with the
second. data list list / id (a4) x y. begin data x 1 2 a 3 4 end data. DATASET NAME analysis. data list list / id (a4) x y z. begin data. b 7 8 2 x 1 2 3 end data. DATASET NAME bigfile. dataset activate analysis. sort cases by id. dataset activate bigfile. sort cases by id. UPDATE FILE = analysis /IN = analysis_file /FILE = bigfile /BY id /keep id x y. DATASET NAME updatedfile. LIST. Note that the analysis file has two variables while the bigfile has three variables. The UPDATE command uses /KEEP rather than /DROP which is probably what you want if you have few variables in the analysis file. Note also that the IN subcommand is used to create a filter variable whose value is 1 for the analysis file and 0 for the cases coming from bigfile that are not matched in the analysis file. It makes for a very handy way to filter out or select out the cases not in the analysis file. Moral of the story: next time you have to do this, use IN and use KEEP. Note also that KEEP will take a usual variable list using the TO convention so you need only type three tokens. -----Original Message----- From: Bob Schacht [mailto:[hidden email]] Sent: Friday, April 24, 2009 2:46 PM To: Albert-jan Roskam; [hidden email] Cc: ViAnn Beadle Subject: Re: Data updates [Vi sez:] <snip/> OK, thanks to you and to ViAnn Beadle I got the update to work. I'm feeling like an old mossback who has used SPSS for about 40 years, sporadically, without keeping up on all of the innovations. So with assistance from both of you, I used this update: >UPDATE FILE=LongQ_with_85 /IN=IN_LongQ /FILE= ThirdQuExt2008 /IN=IN_3Q2008 >/BY SSN /MAP. The original LongQ_with_85 dataset had only 85 cases; the resulting dataset, however, had not only these 85 cases, but 6700 more from the second dataset (ThirdQuExt2008), and also included an extra 120 variables from the second dataset that I didn't want. I was able to delete the extra cases by sorting the cases on a variable occurring only in the 85-case set, and deleting the unselected cases. And I can probably figure out a way to delete all the unwanted variables. But what I wanted was a merger, match or update process that would be limited to the cases and variables in my 85-case dataset. Is there a way to do that without the tedium of specifying all of the unwanted variables with a DROP command? Thanks, Bob Schacht Robert M. Schacht, Ph.D. <[hidden email]> Pacific Basin Rehabilitation Research & Training Center 1268 Young Street, Suite #204 Research Center, University of Hawaii Honolulu, HI 96814 ===================== 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 |