Hi,
The following code yields a peculiar result: the match stops when it encounters a lowercase ID. The RECODE statements restore the expected behavior. I am using SPSS 22FP1 on Win 7 Enterprise, 32 bit. Is this a bug? dataset close all. data list free /id (a10). begin data "A" "B" "c" "D" end data. compute t1origin = 1. *recode id ("c"="C") (else=copy). execute. dataset name one. data list free /id (a10). begin data "A" "B" "c" "D" end data. compute t2origin = 2. *recode id ("c"="C") (else=copy). execute. dataset name two. dataset declare merged. star join /select * /from one as t1 /join two as t2 on t1.id=t2.id /outfile file=merged. dataset activate merged. list all. Regards, Albert-Jan ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ All right, but apart from the sanitation, the medicine, education, wine, public order, irrigation, roads, a fresh water system, and public health, what have the Romans ever done for us? ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ===================== 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 |
Looks like a bug in v22, I tested on v20.0.01 and it failed as you indicated, stopping on the lower case C but in v23 it works fine. results from v23: List [merged] id t1origin t2origin A 1.00 2.00 B 1.00 2.00 c 1.00 2.00 D 1.00 2.00 Number of cases read: 4 Number of cases listed: 4 On 29 April 2015 at 14:32, Albert-Jan Roskam <[hidden email]> wrote:
|
..That should have read..."tested on v20.0.0.1". On 29 April 2015 at 14:46, Jignesh Sutar <[hidden email]> wrote:
|
In reply to this post by Jignesh Sutar
Hi Jignesh,
Thanks for confirming this. I am flabbergasted, gobsmacked and upset! Regards, Albert-Jan ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ All right, but apart from the sanitation, the medicine, education, wine, public order, irrigation, roads, a fresh water system, and public health, what have the Romans ever done for us? ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ -------------------------------------------- On Wed, 4/29/15, Jignesh Sutar <[hidden email]> wrote: Subject: Re: STAR JOIN with id variables which contain strings starting with lowercase letters To: "Albert-Jan Roskam" <[hidden email]> Cc: "[hidden email]" <[hidden email]> Date: Wednesday, April 29, 2015, 3:46 PM Looks like a bug in v22, I tested on v20.0.01 and it failed as you indicated, stopping on the lower case C but in v23 it works fine. results from v23: List[merged] id t1origin t2origin A 1.00 2.00 B 1.00 2.00 c 1.00 2.00 D 1.00 2.00 Number of cases read: 4 Number of cases listed: 4 On 29 April 2015 at 14:32, Albert-Jan Roskam <[hidden email]> wrote: dataset close all. data list free /id (a10). begin data "A" "B" "c" "D" end data. compute t1origin = 1. *recode id ("c"="C") (else=copy). execute. dataset name one. data list free /id (a10). begin data "A" "B" "c" "D" end data. compute t2origin = 2. *recode id ("c"="C") (else=copy). execute. dataset name two. dataset declare merged. star join /select * /from one as t1 /join two as t2 on t1.id=t2.id /outfile file=merged. dataset activate merged. list all. ===================== 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 Jignesh Sutar
Sorry ! It should have read v22.0.0.1! I'll grab my coat... On 29 April 2015 at 14:47, Jignesh Sutar <[hidden email]> wrote:
|
In reply to this post by Albert-Jan Roskam-2
The old way, works fine...in 22.0.0.0
match files file=one/rename (t1origin=t1) /file=two/rename (t2origin=t2)/by id. Melissa -----Original Message----- From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Albert-Jan Roskam Sent: Wednesday, April 29, 2015 9:33 AM To: [hidden email] Subject: [SPSSX-L] STAR JOIN with id variables which contain strings starting with lowercase letters Hi, The following code yields a peculiar result: the match stops when it encounters a lowercase ID. The RECODE statements restore the expected behavior. I am using SPSS 22FP1 on Win 7 Enterprise, 32 bit. Is this a bug? dataset close all. data list free /id (a10). begin data "A" "B" "c" "D" end data. compute t1origin = 1. *recode id ("c"="C") (else=copy). executdataset name one. data list free /id (a10). begin data "A" "B" "c" "D" end data. compute t2origin = 2. *recode id ("c"="C") (else=copy). execute. dataset name two. dataset declare merged. star join /select * /from one as t1 /join two as t2 on t1.id=t2.id /outfile file=merged. dataset activate merged. list all. Regards, Albert-Jan ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ All right, but apart from the sanitation, the medicine, education, wine, public order, irrigation, roads, a fresh water system, and public health, what have the Romans ever done for us? ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ===================== 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 ________________________________ This correspondence contains proprietary information some or all of which may be legally privileged; it is for the intended recipient only. If you are not the intended recipient you must not use, disclose, distribute, copy, print, or rely on this correspondence and completely dispose of the correspondence immediately. Please notify the sender if you have received this email in error. NOTE: Messages to or from the State of Connecticut domain may be subject to the Freedom of Information statutes and regulations. ===================== 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 Albert-Jan Roskam-2
-----------------------------
On Wed, May 6, 2015 4:51 PM CEST Ives, Melissa L wrote: >The old way, works fine...in 22.0.0.0 > match files file=one/rename (t1origin=t1) /file=two/rename (t2origin=t2)/by id. Hi - yes, at least when the ids are strings Star Join is not an option. Too bad, because it addresses three nuisances that Match Files has: id vars must have same name; datasets must be sorted by ids; string pairs must be of equal width. I can't believe that such an obvious bug, for such a common use case, has not been fixed before. Or perhaps it has been introduced after its introduction in Spss v21, ie 21fp1, 21fp2, 22, or 22fp1. Albert-Jan ===================== 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 |
I missed the discussion that led to the
conclusion that you cannot use Star Join with string IDs. Under what circumstances
does it fail or give wrong results?
Rick Oliver Senior Information Developer IBM Business Analytics (SPSS) E-mail: [hidden email] From: Albert-Jan Roskam <[hidden email]> To: [hidden email] Date: 05/06/2015 02:46 PM Subject: Re: STAR JOIN with id variables which contain strings starting with lowercase letters Sent by: "SPSSX(r) Discussion" <[hidden email]> ----------------------------- On Wed, May 6, 2015 4:51 PM CEST Ives, Melissa L wrote: >The old way, works fine...in 22.0.0.0 > match files file=one/rename (t1origin=t1) /file=two/rename (t2origin=t2)/by id. Hi - yes, at least when the ids are strings Star Join is not an option. Too bad, because it addresses three nuisances that Match Files has: id vars must have same name; datasets must be sorted by ids; string pairs must be of equal width. I can't believe that such an obvious bug, for such a common use case, has not been fixed before. Or perhaps it has been introduced after its introduction in Spss v21, ie 21fp1, 21fp2, 22, or 22fp1. Albert-Jan ===================== 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 |
Administrator
|
The entire thread can be viewed here:
http://spssx-discussion.1045642.n5.nabble.com/STAR-JOIN-with-id-variables-which-contain-strings-starting-with-lowercase-letters-td5729426.html
--
Bruce Weaver bweaver@lakeheadu.ca http://sites.google.com/a/lakeheadu.ca/bweaver/ "When all else fails, RTFM." PLEASE NOTE THE FOLLOWING: 1. My Hotmail account is not monitored regularly. To send me an e-mail, please use the address shown above. 2. The SPSSX Discussion forum on Nabble is no longer linked to the SPSSX-L listserv administered by UGA (https://listserv.uga.edu/). |
Thanks.
At this point, I can only say that it appears to work correctly in release 22 FP2 and release 23. Rick Oliver Senior Information Developer IBM Business Analytics (SPSS) E-mail: [hidden email] From: Bruce Weaver <[hidden email]> To: [hidden email] Date: 05/06/2015 03:44 PM Subject: Re: STAR JOIN with id variables which contain strings starting with lowercase letters Sent by: "SPSSX(r) Discussion" <[hidden email]> The entire thread can be viewed here: http://spssx-discussion.1045642.n5.nabble.com/STAR-JOIN-with-id-variables-which-contain-strings-starting-with-lowercase-letters-td5729426.html Rick Oliver wrote > I missed the discussion that led to the conclusion that you cannot use > Star Join with string IDs. Under what circumstances does it fail or give > wrong results? > > > Rick Oliver > Senior Information Developer > IBM Business Analytics (SPSS) > E-mail: > oliverr@.ibm > > > > From: Albert-Jan Roskam > < > 000000359ea188a7-dmarc-request@.UGA >> > To: > SPSSX-L@.UGA > Date: 05/06/2015 02:46 PM > Subject: Re: STAR JOIN with id variables which contain strings > starting with lowercase letters > Sent by: "SPSSX(r) Discussion" < > SPSSX-L@.UGA > > > > > > ----------------------------- > On Wed, May 6, 2015 4:51 PM CEST Ives, Melissa L wrote: > >>The old way, works fine...in 22.0.0.0 >> match files file=one/rename (t1origin=t1) /file=two/rename > (t2origin=t2)/by id. > > Hi - yes, at least when the ids are strings Star Join is not an option. > Too bad, because it addresses three nuisances that Match Files has: id > vars must have same name; datasets must be sorted by ids; string pairs > must be of equal width. > > I can't believe that such an obvious bug, for such a common use case, has > not been fixed before. Or perhaps it has been introduced after its > introduction in Spss v21, ie 21fp1, 21fp2, 22, or 22fp1. > > Albert-Jan > > ===================== > To manage your subscription to SPSSX-L, send a message to > LISTSERV@.UGA > (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 > LISTSERV@.UGA > (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 ----- -- Bruce Weaver [hidden email] http://sites.google.com/a/lakeheadu.ca/bweaver/ "When all else fails, RTFM." NOTE: My Hotmail account is not monitored regularly. To send me an e-mail, please use the address shown above. -- View this message in context: http://spssx-discussion.1045642.n5.nabble.com/STAR-JOIN-with-id-variables-which-contain-strings-starting-with-lowercase-letters-tp5729426p5729511.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 |
In reply to this post by Albert-Jan Roskam-2
Hi Rick,
Thank you. I've been looking for FP2 for v22, but I missed its release. The original planning was Q4 of 2014, then it became Q1 2015, but it was probably released in Q2? I wrongly concluded there would not be a second fixpack because SPSS v23 was already released. Best wishes, Albert-Jan ------------------------------ On Wed, May 6, 2015 10:59 PM CEST Rick Oliver wrote: >Thanks. > >At this point, I can only say that it appears to work correctly in release >22 FP2 and release 23. > >Rick Oliver >Senior Information Developer >IBM Business Analytics (SPSS) >E-mail: [hidden email] > > > >From: Bruce Weaver <[hidden email]> >To: [hidden email] >Date: 05/06/2015 03:44 PM >Subject: Re: STAR JOIN with id variables which contain strings >starting with lowercase letters >Sent by: "SPSSX(r) Discussion" <[hidden email]> > > > >The entire thread can be viewed here: > >http://spssx-discussion.1045642.n5.nabble.com/STAR-JOIN-with-id-variables-which-contain-strings-starting-with-lowercase-letters-td5729426.html > > > > >Rick Oliver wrote >> I missed the discussion that led to the conclusion that you cannot use >> Star Join with string IDs. Under what circumstances does it fail or give > >> wrong results? >> >> >> Rick Oliver >> Senior Information Developer >> IBM Business Analytics (SPSS) >> E-mail: > >> oliverr@.ibm > >> >> >> >> From: Albert-Jan Roskam >> < > >> 000000359ea188a7-dmarc-request@.UGA > >> >> To: > >> SPSSX-L@.UGA > >> Date: 05/06/2015 02:46 PM >> Subject: Re: STAR JOIN with id variables which contain strings >> starting with lowercase letters >> Sent by: "SPSSX(r) Discussion" < > >> SPSSX-L@.UGA > >> > >> >> >> >> ----------------------------- >> On Wed, May 6, 2015 4:51 PM CEST Ives, Melissa L wrote: >> >>The old way, works fine...in 22.0.0.0 >> match files file=one/rename (t1origin=t1) /file=two/rename >> (t2origin=t2)/by id. >> >> Hi - yes, at least when the ids are strings Star Join is not an option. >> Too bad, because it addresses three nuisances that Match Files has: id >> vars must have same name; datasets must be sorted by ids; string pairs >> must be of equal width. >> >> I can't believe that such an obvious bug, for such a common use case, >has >> not been fixed before. Or perhaps it has been introduced after its >> introduction in Spss v21, ie 21fp1, 21fp2, 22, or 22fp1. >> >> Albert-Jan >> >> ===================== >> To manage your subscription to SPSSX-L, send a message to > >> LISTSERV@.UGA > >> (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 > >> LISTSERV@.UGA > >> (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 > > > > > >----- >-- >Bruce Weaver >[hidden email] >http://sites.google.com/a/lakeheadu.ca/bweaver/ > >"When all else fails, RTFM." > >NOTE: My Hotmail account is not monitored regularly. >To send me an e-mail, please use the address shown above. > >-- >View this message in context: >http://spssx-discussion.1045642.n5.nabble.com/STAR-JOIN-with-id-variables-which-contain-strings-starting-with-lowercase-letters-tp5729426p5729511.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 ===================== 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 |
Administrator
|
Am I imagining things, or was there an "Update" option in the Help menu in older versions that would check for any new Fix Packs? I don't see it in v22 for Windoze. And until I read this thread, I had no idea there was a second FP for v22. I suspect the IT folks at my university had no idea either--I just let them know about it.
This is one area in which Stata is vastly superior to SPSS. It periodically checks for updates, and prompts the user when updates are available.
--
Bruce Weaver bweaver@lakeheadu.ca http://sites.google.com/a/lakeheadu.ca/bweaver/ "When all else fails, RTFM." PLEASE NOTE THE FOLLOWING: 1. My Hotmail account is not monitored regularly. To send me an e-mail, please use the address shown above. 2. The SPSSX Discussion forum on Nabble is no longer linked to the SPSSX-L listserv administered by UGA (https://listserv.uga.edu/). |
In reply to this post by Albert-Jan Roskam-2
Hope this helps:
http://www-01.ibm.com/support/docview.wss?uid=swg24039501 Rick Oliver Senior Information Developer IBM Business Analytics (SPSS) E-mail: [hidden email] From: Albert-Jan Roskam <[hidden email]> To: Rick Oliver/Chicago/IBM@IBMUS, [hidden email] Date: 05/07/2015 02:28 AM Subject: Re: [SPSSX-L] STAR JOIN with id variables which contain strings starting with lowercase letters Hi Rick, Thank you. I've been looking for FP2 for v22, but I missed its release. The original planning was Q4 of 2014, then it became Q1 2015, but it was probably released in Q2? I wrongly concluded there would not be a second fixpack because SPSS v23 was already released. Best wishes, Albert-Jan ------------------------------ On Wed, May 6, 2015 10:59 PM CEST Rick Oliver wrote: >Thanks. > >At this point, I can only say that it appears to work correctly in release >22 FP2 and release 23. > >Rick Oliver >Senior Information Developer >IBM Business Analytics (SPSS) >E-mail: [hidden email] > > > >From: Bruce Weaver <[hidden email]> >To: [hidden email] >Date: 05/06/2015 03:44 PM >Subject: Re: STAR JOIN with id variables which contain strings >starting with lowercase letters >Sent by: "SPSSX(r) Discussion" <[hidden email]> > > > >The entire thread can be viewed here: > >http://spssx-discussion.1045642.n5.nabble.com/STAR-JOIN-with-id-variables-which-contain-strings-starting-with-lowercase-letters-td5729426.html > > > > >Rick Oliver wrote >> I missed the discussion that led to the conclusion that you cannot use >> Star Join with string IDs. Under what circumstances does it fail or give > >> wrong results? >> >> >> Rick Oliver >> Senior Information Developer >> IBM Business Analytics (SPSS) >> E-mail: > >> oliverr@.ibm > >> >> >> >> From: Albert-Jan Roskam >> < > >> 000000359ea188a7-dmarc-request@.UGA > >> >> To: > >> SPSSX-L@.UGA > >> Date: 05/06/2015 02:46 PM >> Subject: Re: STAR JOIN with id variables which contain strings >> starting with lowercase letters >> Sent by: "SPSSX(r) Discussion" < > >> SPSSX-L@.UGA > >> > >> >> >> >> ----------------------------- >> On Wed, May 6, 2015 4:51 PM CEST Ives, Melissa L wrote: >> >>The old way, works fine...in 22.0.0.0 >> match files file=one/rename (t1origin=t1) /file=two/rename >> (t2origin=t2)/by id. >> >> Hi - yes, at least when the ids are strings Star Join is not an option. >> Too bad, because it addresses three nuisances that Match Files has: id >> vars must have same name; datasets must be sorted by ids; string pairs >> must be of equal width. >> >> I can't believe that such an obvious bug, for such a common use case, >has >> not been fixed before. Or perhaps it has been introduced after its >> introduction in Spss v21, ie 21fp1, 21fp2, 22, or 22fp1. >> >> Albert-Jan >> >> ===================== >> To manage your subscription to SPSSX-L, send a message to > >> LISTSERV@.UGA > >> (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 > >> LISTSERV@.UGA > >> (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 > > > > > >----- >-- >Bruce Weaver >[hidden email] >http://sites.google.com/a/lakeheadu.ca/bweaver/ > >"When all else fails, RTFM." > >NOTE: My Hotmail account is not monitored regularly. >To send me an e-mail, please use the address shown above. > >-- >View this message in context: >http://spssx-discussion.1045642.n5.nabble.com/STAR-JOIN-with-id-variables-which-contain-strings-starting-with-lowercase-letters-tp5729426p5729511.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 ===================== 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 |
Do IBM send out a email message to notify an fix pack or new version release? As registered user I would expect this at a minimum, I don't believe I have opted out of such communication and I certainly didn't receive an email.
I don't recall and official communication being released other than proactively delving into the depths of the IBM website. |
In reply to this post by Albert-Jan Roskam-2
-----------------------------
On Thu, May 7, 2015 4:31 PM CEST Jignesh Sutar wrote: >Do IBM send out a email message to notify an fix pack or new version release? >As registered user I would expect this at a minimum, I don't believe I have >opted out of such communication and I certainly didn't receive an email. > >I don't recall and official communication being released other than >proactively delving into the depths of the IBM website. I do get such mails (I don't recall how I registered for them), but sometimes the important news (new FP, new version) gets lost in less significant news (ctables won't display properly in Telugu language). It would be nice if the subject line were something like "fixpack 2 released!". An auto-update notification like with antivirus software would also be great. Best wishes, Albert-Jan ===================== 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 Rick Oliver-3
We are still using SPSS v21, and STAR JOIN does indeed seem broken, now that
we are finally trying to make use of it. I tried it with a string-valued linking (key) variable. The lookup table has them all in uppercase, so I converted to uppercase in the main file. Result: cases with values that appear in the first 3/4 or so of the lookup table got matched, the rest did not receive the matched info. In this test, the lookup table had about 900 cases, and the main file about 800 cases. The lookup table was pre-sorted by the key variable (although that is not necessary?), the main file was not (which was one of the points of trying out STAR JOIN instead of MATCH FILES). The join was many-to-one. The lengths of the string variables that are joined on were different in the two files (another reason for trying out STAR JOIN instead of MATCH FILES). Is this a known bug, and was it fixed in later versions of SPSS? -- View this message in context: http://spssx-discussion.1045642.n5.nabble.com/STAR-JOIN-with-id-variables-which-contain-strings-starting-with-lowercase-letters-tp5729426p5729582.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 |
In reply to this post by Rick Oliver-3
We are still using SPSS v21, and STAR JOIN does indeed seem broken, now that we are finally trying to make use of it. I tried it with a string-valued linking (key) variable. The lookup table has them all in uppercase, so I converted to uppercase in the main file. Result: cases with values that appear in the first 3/4 or so of the lookup table got matched, the rest did not receive the matched info. In this test, the lookup table had about 900 cases, and the main file about 800 cases. The lookup table was pre-sorted by the key variable (although that is not necessary?), the main file was not (which was one of the points of trying out STAR JOIN instead of MATCH FILES). The join was many-to-one. The lengths of the string variables that are joined on were different in the two files (another reason for trying out STAR JOIN instead of MATCH FILES). Is this a known bug, and was it fixed in later versions of SPSS?
|
In reply to this post by Albert-Jan Roskam-2
-----------------------------
On Wed, May 20, 2015 9:14 PM CEST mbraner wrote: >We are still using SPSS v21, and STAR JOIN does indeed seem broken, now that >we are finally trying to make use of it. I tried it with a string-valued >linking (key) variable. The lookup table has them all in uppercase, so I >converted to uppercase in the main file. Result: cases with values that >appear in the first 3/4 or so of the lookup table got matched, the rest did >not receive the matched info. In this test, the lookup table had about 900 >cases, and the main file about 800 cases. The lookup table was pre-sorted >by the key variable (although that is not necessary?), the main file was not >(which was one of the points of trying out STAR JOIN instead of MATCH >FILES). The join was many-to-one. The lengths of the string variables that >are joined on were different in the two files (another reason for trying out >STAR JOIN instead of MATCH FILES). Agreed, these are certainly advantages. A third plus is that the ID variables may have different names. >Is this a known bug, and was it fixed in >later versions of SPSS? The bug I mentioned was no longer present in SPSS v22.0.0.2 (IOW Spss v22 fixpack 2) and SPSS v23. Not sure if this was 32- and 64-bit. I assume the checks were done in Windows (Rick Oliver of IBM kindly checked this) ===================== 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 |
Free forum by Nabble | Edit this page |