renaming variables - deleting the first part of the name
Locked
renaming variables - deleting the first part of the name
 
|
Dear List,
I have a file with over a 100 variables named Q101_ofcsp, Q102_ofceq, Q103_clean, Q107_cereq, Q108_load, Q109_oppor, Q110_descr, etc. I need to get rid of the QXXX_ part for every variable; that is to rename Q101_ofcsp to ofcsp, etc. Is there a way to do this in SPSS besides using RENAME VARIABLES? Any help would be greatly appreciated. Thank you very much, Zina |
Re: renaming variables - deleting the first part of the name
|
|
I don't know of another way through syntax, but if it is not much more
than 100 variables it shouldn't take too terribly long to just go in and delete the first 5 characters of each by hand in the variable view. Lise -----Original Message----- From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Zina Karapetyan Sent: Thursday, September 21, 2006 11:54 AM To: [hidden email] Subject: renaming variables - deleting the first part of the name Dear List, I have a file with over a 100 variables named Q101_ofcsp, Q102_ofceq, Q103_clean, Q107_cereq, Q108_load, Q109_oppor, Q110_descr, etc. I need to get rid of the QXXX_ part for every variable; that is to rename Q101_ofcsp to ofcsp, etc. Is there a way to do this in SPSS besides using RENAME VARIABLES? Any help would be greatly appreciated. Thank you very much, Zina |
Re: renaming variables - deleting the first part of the name
|
|
In reply to this post by Zina Karapetyan
Zina,
RENAME VARIABLES is your friend. This is slightly modified and tested code from Ray Levesque's marvelous site http://www.spsstools.net Variable names were shortened for backward-compatibility with earlier versions of SPSS that did not permit long variable names. HTH King Douglas American Airlines *(Q) I am about to merge three huge data sets. Every dataset contains hundreds of vars. The var names are identical except for the prefix "tn". For example, var names of dataset no.1 start with"t1" t1abc, t1xyz, t1sps.... * var names of dataset no.2 start with"t2" t2abc, t2xyz, t2sps.... * var names of dataset no.3 start with"t3" t3abc, t3xyz, t3sps.... * How can I rename the vars of datsets nos. 2 and 3, so that they will begin with "t1", e.g. "t1abc" (former "t3abc"), too? *(A) Posted to SPSSX-L by [hidden email] on 2002/01/23. DATA LIST FREE / Q101_ofX, Q102_ofY, Q103_cle, Q107_cer, Q108_loa, Q109_opp. BEGIN DATA 85 95 5 87 100 1 90 100 1 25 25 2 END DATA. LIST. SAVE OUTFILE='c:\temp\mydata.sav'. * Be sure you have a backup copy of your data file before running this. * Next lines ensures no error will occur when there are more than 32,000 variables in the data file (Thanks to H. Maletta for this suggestion). N OF CASES 1. FLIP. STRING newname(A8). COMPUTE newname=SUBSTR(case_lbl,6). WRITE OUTFILE 'c:\temp\temp.sps' /"RENAME VARIABLE ("case_lbl"="newname").". EXE. GET FILE='c:\temp\mydata.sav'. INCLUDE 'c:\temp\temp.sps'. Zina Karapetyan <[hidden email]> wrote: Dear List, I have a file with over a 100 variables named Q101_ofcsp, Q102_ofceq, Q103_clean, Q107_cereq, Q108_load, Q109_oppor, Q110_descr, etc. I need to get rid of the QXXX_ part for every variable; that is to rename Q101_ofcsp to ofcsp, etc. Is there a way to do this in SPSS besides using RENAME VARIABLES? Any help would be greatly appreciated. Thank you very much, Zina |
Re: renaming variables - deleting the first part of the name
|
In reply to this post by Zina Karapetyan
At 12:53 PM 9/21/2006, Zina Karapetyan wrote:
>I have a file with over a 100 variables named Q101_ofcsp, Q102_ofceq, >Q103_clean, Q107_cereq, Q108_load, Q109_oppor, Q110_descr, etc. I need >to get rid of the QXXX_ part for every variable; that is to rename >Q101_ofcsp to ofcsp, etc. Is there a way to do this in SPSS besides >using RENAME VARIABLES? Any help would be greatly appreciated. "Is there a way to do this in SPSS besides using RENAME VARIABLES?" No, in syntax there isn't; either RENAME VARIABLES, or RENAME= on a SAVE, GET, ADD FILES, or MATCH FILES. The question is (or, it may be), can you generate those 100-odd RENAME specifications (like "Q110_descr=descr") without writing them? See http://www.spsstools.net/Syntax/LabelsAndVariableNames/ChangeCharacterAtBeginningOfEachVarNames.txt which does almost exactly what you want. It uses FLIP to get the existing list of variables; I think that loses the string variables, though. Two other ways: if you have OMS (was that SPSS 12?), you can use it to get the data dictionary (the list of variables), and generate the RENAME specifications from it. See my posting "Data dictionary in SPSS", Wed, 6 Sep 2006 13:30:05 -0400. That keeps all variables. Or, if you have SPSS 14+, Python code (BEGIN PROGRAM/ END PROGRAM) has direct access to the data dictionary with the variable namse. It has string functions that are plenty good enough to break the name into pieces and build the RENAME statement; and can run the result directly, using its spss.submit function. Post again, if this isn't enough information. A thought about WHETHER to do it: It looks like the prefix "QXXX_" is something like a question number from a questionnaire; the suffix is mnemonic, descriptive of the question; and you want to keep just the mnemonic part. A lot of us, when we've a dataset with a lot of variables (100 is a lot), prefer non-mnemonic variable names like "Q110". First, "Q110" makes it clear where the variable is in the dataset, and in what context it arises. (What were the adjacent questions? What section of the questionnaire was it?) Second, "Q110" will often be easier to remember, because there are usually several equally plausible 'mnemonic' names: "descr"? "descrip"? "item_dsc"? (Long names makes this worse, not better.) I'd consider stripping the other way: Strip "Q110_descr" to "Q110", with "descr" in the variable label. |
Re: renaming variables - deleting the first part of the name
|
|
At 02:06 PM 9/21/2006, Richard Ristow wrote:
>See >http://www.spsstools.net/Syntax/LabelsAndVariableNames/ChangeCharacterAtBeginningOfEachVarNames.txt >which does almost exactly what you want. It uses FLIP to get the >existing list of variables; I think that loses the string variables, >though. Oh, I of little faith in Raynald Levesque! No, it doesn't lose string variables. From the Command Syntax Reference entry for FLIP, ".. String variables in the original file are assigned system-missing values after transposition." So, this is probably what you want; see King Douglas's posting, that gives it more completely. Remember the other tools too, though. And my caveat, about whether it's a good idea. |
|
|
I want to paste the text of a few cells' of several
pivots into one Excel or notepad file. Is there a way to do this using SPSS scripts (preferred) or syntax? Thank you! Peng __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com |
|
|
In reply to this post by Richard Ristow
Ok, here is the Python code. First it defines a regular expression pattern matching a string starting with Q or q followed by any number of digits followed by "_"
Then it creates a Python copy of the SPSS variable dictionary including only variables whose name fits this pattern. Next it loops over these variables creating new names deleting the pattern and submits rename commands. Implicitly this assumes that there always IS a character after the "_" and that the resulting names are all distinct. That could be dealt with easily with a little more code. begin program. import spss, spssaux, re pat="^[Qq][0-9]+_" vard = spssaux.VariableDict(pattern=pat) for v in vard: newname = re.sub(pat, "", str(v)) spss.Submit("RENAME VARIABLES (%s=%s)" % (v, newname)) end program. At Directions in November, I will show regular expression search and replace functions hooked up so that they can be used like transformations on the case data with SPSS 15. HTH, Jon Peck SPSS -----Original Message----- From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Richard Ristow Sent: Thursday, September 21, 2006 1:06 PM To: [hidden email] Subject: Re: [SPSSX-L] renaming variables - deleting the first part of the name At 12:53 PM 9/21/2006, Zina Karapetyan wrote: >I have a file with over a 100 variables named Q101_ofcsp, Q102_ofceq, >Q103_clean, Q107_cereq, Q108_load, Q109_oppor, Q110_descr, etc. I need >to get rid of the QXXX_ part for every variable; that is to rename >Q101_ofcsp to ofcsp, etc. Is there a way to do this in SPSS besides >using RENAME VARIABLES? Any help would be greatly appreciated. "Is there a way to do this in SPSS besides using RENAME VARIABLES?" No, in syntax there isn't; either RENAME VARIABLES, or RENAME= on a SAVE, GET, ADD FILES, or MATCH FILES. The question is (or, it may be), can you generate those 100-odd RENAME specifications (like "Q110_descr=descr") without writing them? See http://www.spsstools.net/Syntax/LabelsAndVariableNames/ChangeCharacterAtBeginningOfEachVarNames.txt which does almost exactly what you want. It uses FLIP to get the existing list of variables; I think that loses the string variables, though. Two other ways: if you have OMS (was that SPSS 12?), you can use it to get the data dictionary (the list of variables), and generate the RENAME specifications from it. See my posting "Data dictionary in SPSS", Wed, 6 Sep 2006 13:30:05 -0400. That keeps all variables. Or, if you have SPSS 14+, Python code (BEGIN PROGRAM/ END PROGRAM) has direct access to the data dictionary with the variable namse. It has string functions that are plenty good enough to break the name into pieces and build the RENAME statement; and can run the result directly, using its spss.submit function. Post again, if this isn't enough information. A thought about WHETHER to do it: It looks like the prefix "QXXX_" is something like a question number from a questionnaire; the suffix is mnemonic, descriptive of the question; and you want to keep just the mnemonic part. A lot of us, when we've a dataset with a lot of variables (100 is a lot), prefer non-mnemonic variable names like "Q110". First, "Q110" makes it clear where the variable is in the dataset, and in what context it arises. (What were the adjacent questions? What section of the questionnaire was it?) Second, "Q110" will often be easier to remember, because there are usually several equally plausible 'mnemonic' names: "descr"? "descrip"? "item_dsc"? (Long names makes this worse, not better.) I'd consider stripping the other way: Strip "Q110_descr" to "Q110", with "descr" in the variable label. |
Re: How to paste pivot cell text to excel using script?
|
|
In reply to this post by peng jia
This is along the lines of something I have been asking.
I am generating Tables like the below, but want to combine the elements from these tables into a single table for each of My Offices (whicha re what the below are), which means I want to write a particular cell from the table below to a very particular cell in a very particular excell SS, and then, as I update this data regularly, I will want to append existing tables. Any thoughts? ProgramTypeCode P2 P3 Total Count Count Count Count OfficeName ANCHORAGE CHILD CARE 0 0 1 1 CATHOLIC COMM. SVC. 0 25 298 323 CORDOVA CHILD CARE 0 0 5 5 FNSB CHILD CARE ASST 0 20 320 340 HAINES/SKAGWAY 0 1 14 15 HOMER CHILD CARE 0 12 47 59 KENAI CHILD CARE 0 18 237 255 KODIAK CHILD CARE 0 8 44 52 MAT SU CHILD CARE 0 59 299 358 METLAKATLA CHILD CAR 0 2 17 19 MUNICIPALITY OF ANCH 0 135 1634 1769 NOME, BARROW, KOTZ., 0 0 7 7 SEWARD CHILD CARE 0 1 26 27 SITKA CHILD CARE 0 1 48 49 VALDEZ CHILD CARE 0 1 23 24 WRANGELL CHILD CARE 0 1 21 22 Total 0 284 3041 3325 >From: peng jia <[hidden email]> >Reply-To: peng jia <[hidden email]> >To: [hidden email] >Subject: How to paste pivot cell text to excel using script? >Date: Thu, 21 Sep 2006 11:50:39 -0700 > >I want to paste the text of a few cells' of several >pivots into one Excel or notepad file. > >Is there a way to do this using SPSS scripts >(preferred) or syntax? > >Thank you! > >Peng > > >__________________________________________________ >Do You Yahoo!? >Tired of spam? Yahoo! Mail has the best spam protection around >http://mail.yahoo.com |
Re: renaming variables - deleting the first part of the name
|
|
In reply to this post by Lise Valentine
No need to do it by hand if you have Word or other editor that can edit columns. You can copy the names from variable view, paste in Word, select first 5 columns (press Alt while selecting) and delete, then copy and paste back into variable view.
Anita van der Kooij Data Theory Group Leiden University ________________________________ From: SPSSX(r) Discussion on behalf of Lise Valentine Sent: Thu 21/09/2006 19:24 To: [hidden email] Subject: Re: renaming variables - deleting the first part of the name I don't know of another way through syntax, but if it is not much more than 100 variables it shouldn't take too terribly long to just go in and delete the first 5 characters of each by hand in the variable view. Lise -----Original Message----- From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Zina Karapetyan Sent: Thursday, September 21, 2006 11:54 AM To: [hidden email] Subject: renaming variables - deleting the first part of the name Dear List, I have a file with over a 100 variables named Q101_ofcsp, Q102_ofceq, Q103_clean, Q107_cereq, Q108_load, Q109_oppor, Q110_descr, etc. I need to get rid of the QXXX_ part for every variable; that is to rename Q101_ofcsp to ofcsp, etc. Is there a way to do this in SPSS besides using RENAME VARIABLES? Any help would be greatly appreciated. Thank you very much, Zina ********************************************************************** This email and any files transmitted with it are confidential and intended solely for the use of the individual or entity to whom they are addressed. If you have received this email in error please notify the system manager. ********************************************************************** |
Re: renaming variables - deleting the first part of the name
|
|
In reply to this post by Peck, Jon
Hi Jon,
I just wanted to thank you for this 10-year old post: begin program. import spss, spssaux, re pat="^[Qq][0-9]+_" vard = spssaux.VariableDict(pattern=pat) for v in vard: newname = re.sub(pat, "", str(v)) spss.Submit("RENAME VARIABLES (%s=%s)" % (v, newname)) end program. ...you just made my life a whole lot easier (and faster). Thank you! Esther -- Esther Fujiwara, Ph.D. Associate Professor Department of Psychiatry University of Alberta Edmonton, AB Canada T6G 2V2 |
«
Return to SPSSX Discussion
|
1 view|%1 views
| Free forum by Nabble | Edit this page |