Sort Variables
Locked
 
|
Hi,
we need to create sorted variables lists. Our data looks like this: ID c_0001 c_0002 c_0003 v_1 v_2 v_3 v_4 v_5 v_6 v_27 v_28 v_29 v_25_1 v_25_2 v_25_3 v_27_1 v_27_2 v_27_3 v_28_1 v_28_2 v_28_3 v_29_1 v_29_2 v_29_3 v_26_1 v_26_2 v_26_3 It should like: ID c_0001 c_0002 c_0003 v_1 v_2 v_3 v_4 v_5 v_6 v_25_1 v_25_2 v_25_3 v_26_1 v_26_2 v_26_3 v_27_1 v_27_2 v_27_3 v_28_1 v_28_2 v_28_3 v_29_1 v_29_2 v_29_3 v_27 v_28 v_29 In this little example only the three variables with v_26 aren't in the right order and the whole block with the _variables are at the end. Of course I can do that manually but if there are thousands of them it will be difficult. Is there a possibility to sort them in the right order? I can also do that with the menu and sort by name but then the whole block is still at the end. Many thanks for your suggestions! |
Administrator
|
Hello Emma. I am confused by the comment you made after showing the desired variable order.
You said, "only the three variables with v_26 aren't in the right order and the whole block with the _variables are at the end". Did you mean that THIS is the desired order? ID c_0001 c_0002 c_0003 v_1 v_2 v_3 v_4 v_5 v_6 v_27 v_28 v_29 v_25_1 v_25_2 v_25_3 v_26_1 v_26_2 v_26_3 v_27_1 v_27_2 v_27_3 v_28_1 v_28_2 v_28_3 v_29_1 v_29_2 v_29_3 If that *is* what you meant, try this: SORT VARIABLES BY NAME (A). ADD FILES FILE=* /KEEP=ID ALL. EXECUTE. HTH.
--
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 emma78
Hi,
I made a mistake, sorry for the confusion: ID c_0001 c_0002 c_0003 v_1 v_2 v_3 v_4 v_5 v_6 v_30 v_31 v_32 v_25_1 v_25_2 v_25_3 v_27_1 v_27_2 v_27_3 v_28_1 v_28_2 v_28_3 v_29_1 v_29_2 v_29_3 v_26_1 v_26_2 v_26_3 and it shoud like ID c_0001 c_0002 c_0003 v_1 v_2 v_3 v_4 v_5 v_6 v_25_1 v_25_2 v_25_3 v_26_1 v_26_2 v_26_3 v_27_1 v_27_2 v_27_3 v_28_1 v_28_2 v_28_3 v_29_1 v_29_2 v_29_3 v_30 v_31 v_32 |
|
|
In reply to this post by emma78
Assuming - you want to sort the names according to the numeric value of the number strings. Sorted as numbers, you would get 1,2,3,...25,26... Sorted as strings you would get 1,2,25,26,...,3, 4,... - these names are in an existing variable dictionary - all the names except ID should be alphabetized. ID should remain first Try this code. begin program. import spss, spssaux varlist = spssaux.VariableDict().variables varlist.pop(varlist.index('ID')) def f(key): key = key.split("_") for k, item in enumerate(key): try: item = int(item) except: pass key[k] = item return key s = sorted(varlist, key=f) spss.Submit(r"""ADD FILES FILE=* /KEEP=ID %s.""" % " ".join(s)) end program. On Wed, Apr 20, 2016 at 9:17 AM, emma78 <[hidden email]> wrote: Hi, |
|
|
Hi,
is it possible to sort this variable list lfdn v_20_1 v_20_2 v_20_3 v_20_4 v_20_5 v_20_6 v_20_7 v_20_8 v_20_9 v_20_10 v_20_11 v_20_12 v_20_13 v_20_14 v_20_15 v_20_16 v_21_1 v_21_2 v_21_3 v_21_4 v_21_5 v_21_6 v_21_7 v_21_8 v_21_9 v_21_10 v_21_11 v_21_12 v_21_13 v_21_14 v_21_15 v_21_16 that it looks like lfdn v_20_1 v_21_1 v_20_2 v_21_2 v_20_3 v_21_3 v_20_4 v_21_4 v_20_5 v_21_5 v_20_6 v_21_6 v_20_7 v_21_7 v_20_8 v_21_8 v_20_9 v_21_9 v_20_10 v_21_10 v_20_11 v_21_11 v_20_12 v_21_12 v_20_13 v_21_13 v_20_14 v_21_14 v_20_15 v_21_15 v_20_16 v_21_16 I already tried to adapt the Python script but it doesn't work :-( begin program. import spss, spssaux varlist = spssaux.VariableDict().variables varlist.pop(varlist.index('lfdn')) def f(key): key = key.split("_") for k, item in enumerate(key): try: item = int(item) except: pass key[k] = item return key s = sorted(varlist, key=f) spss.Submit(r"""ADD FILES FILE=* /KEEP=lfdn %s.""" % " ".join(s)) end program. |
|
Emma
In haste and untested, but supply your own <full file pathname>, then something like: save out '<full file pathname>' /keep lfdn v_20_1 v_21_1 v_20_2 v_21_2 v_20_3 v_21_3 v_20_4 v_21_4 v_20_5 v_21_5 v_20_6 v_21_6 v_20_7 v_21_7 v_20_8 v_21_8 v_20_9 v_21_9 v_20_10 v_21_10 v_20_11 v_21_11 v_20_12 v_21_12 v_20_13 v_21_13 v_20_14 v_21_14 v_20_15 v_21_15 v_20_16 v_21_16. John F Hall (Mr) [Retired academic survey researcher] Email: [hidden email] Website: www.surveyresearch.weebly.com SPSS start page: www.surveyresearch.weebly.com/1-survey-analysis-workshop -----Original Message----- From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of emma78 Sent: 07 February 2017 16:26 To: [hidden email] Subject: Re: Sort Variables Hi, is it possible to sort this variable list lfdn v_20_1 v_20_2 v_20_3 v_20_4 v_20_5 v_20_6 v_20_7 v_20_8 v_20_9 v_20_10 v_20_11 v_20_12 v_20_13 v_20_14 v_20_15 v_20_16 v_21_1 v_21_2 v_21_3 v_21_4 v_21_5 v_21_6 v_21_7 v_21_8 v_21_9 v_21_10 v_21_11 v_21_12 v_21_13 v_21_14 v_21_15 v_21_16 that it looks like I already tried to adapt the Python script but it doesn't work :-( begin program. import spss, spssaux varlist = spssaux.VariableDict().variables varlist.pop(varlist.index('lfdn')) def f(key): key = key.split("_") for k, item in enumerate(key): try: item = int(item) except: pass key[k] = item return key s = sorted(varlist, key=f) spss.Submit(r"""ADD FILES FILE=* /KEEP=lfdn %s.""" % " ".join(s)) end program. -- View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Sort-Variables-tp5731970 p5733780.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 emma78
Try this as your key function. def f(key): key = key.split("_") return (int(key[2]), int(key[1])) On Tue, Feb 7, 2017 at 8:26 AM, emma78 <[hidden email]> wrote: Hi, |
|
|
That is working thank you!
Is it posible to adajust it, just in case we have some variables with only one '_' in the name? That these two stays before the v_20_XXX? In the moment the syntax isn't working, if I keep this variables in the dataset, thank you! So lfdn v_2 v_3 v_20_1 v_20_2 v_20_3 v_20_4 v_20_5 v_20_6 v_20_7 v_20_8 v_20_9 v_20_10 v_20_11 v_20_12 v_20_13 v_20_14 v_20_15 v_20_16 v_21_1 v_21_2 v_21_3 v_21_4 v_21_5 |
|
|
This key function will do it. def f(key): key = key.split("_") if len(key) == 3: return (int(key[2]), int(key[1])) else: return (0, int(key[1])) On Fri, Feb 10, 2017 at 3:15 AM, emma78 <[hidden email]> wrote: That is working thank you! |
|
|
Hi,
another question regarding this sorting problem lfdn v_1 v_55 v_40_1 v_40_2 v_40_3 v_40_4 v_41_1 v_41_2 v_41_3 v_41_4 v_42_1 v_42_2 v_42_3 v_42_4 v_43_1 should look like lfdn v_1 v_40_1 v_41_1 v_42_1 v_43_1 v_40_2 v_41_2 v_42_2 v_40_3 v_41_3 v_42_3 v_40_4 v_41_4 v_42_4 v_55 I already got this script begin program. import spss, spssaux varlist = spssaux.VariableDict().variables varlist.pop(varlist.index('lfdn')) def f(key): key = key.split("_") if len(key) == 3: return (int(key[2]), int(key[1])) else: return (0, int(key[1])) for k, item in enumerate(key): try: item = int(item) except: pass key[k] = item return key s = sorted(varlist, key=f) spss.Submit(r"""ADD FILES FILE=* /KEEP=lfdn duration %s.""" % " ".join(s)) end program. but the v_55 stays the same and doesn`t appear after the v_42_4. Any suggestion on this? Another problem occurs in a dataset with a lot of variables, I got this message Traceback (most recent call last): File "<string>", line 18, in <module> File "<string>", line 10, in f IndexError: list index out of range Anybody knows what this means? |
|
Emma78,
I also recommend you to consider one of my old macros called !INTLACE, found on http://www.spsstools.net/en/KO-spssmacros/ in collection "Job tools". It does not sorts, but it re-arranges (interlaces) elements of blocks of names. It is handy to change the order of variables in a list when you have some blocks of names, say, X1 X2 Y1 Y2 Z1 Z2; the macro will re-arrange the order into: X1 Y1 Z1 X2 Y2 Z 2. Use tilda to separate your blocks. Blocks do not need be equal length. *Mode 1. Print out result. !intlace (PRINT) v_1 ~ v_55 ~ v_40_1 v_40_2 v_40_3 v_40_4 ~ v_41_1 v_41_2 v_41_3 v_41_4 ~ v_42_1 v_42_2 v_42_3 v_42_4 ~ v_43_1. Prints out result as this list: v_1 v_55 v_40_1 v_41_1 v_42_1 v_43_1 v_40_2 v_41_2 v_42_2 v_40_3 v_41_3 v_42_3 v_40_4 v_41_4 v_42_4 *Mode2. Insert result in a command. match files /file= * /keep= !intlace v_1 ~ v_55 ~ v_40_1 v_40_2 v_40_3 v_40_4 ~ v_41_1 v_41_2 v_41_3 v_41_4 ~ v_42_1 v_42_2 v_42_3 v_42_4 ~ v_43_1. exec. Produces ans runs the following syntax: match files /file= * /keep= v_1 v_55 v_40_1 v_41_1 v_42_1 v_43_1 v_40_2 v_41_2 v_42_2 v_40_3 v_41_3 v_42_3 v_40_4 v_41_4 v_42_4. exec. ===================== 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 |
|
|
Just change the line after "else" to read else: return (int(key[1]), 0) On Fri, Mar 3, 2017 at 4:20 AM, Kirill Orlov <[hidden email]> wrote:
-- |
|
|
Hi Jon, thank you.
It works, but only unless there isn`t another variable which doesn't start with v_ in the dataset. I try to write the duration and lfdn after /KEEP= but then the whole syntax doesn't work, is there an adjustment possible? lfdn duration v_1 v_55 v_40_1 v_40_2 v_40_3 v_40_4 v_41_1 v_41_2 v_41_3 v_41_4 v_42_1 v_42_2 v_42_3 v_42_4 v_43_1 |
|
|
Simple old me thinks, at least for this analysis, why not just use: save out <pathway/filename> /keep lfdn v_1 v_40_1 v_41_1 v_42_1 v_43_1 v_40_2 v_41_2 v_42_2 v_40_3 v_41_3 v_42_3 v_40_4 v_41_4 v_42_4 v_55. John F Hall (Mr) [Retired academic survey researcher] Email: [hidden email] Website: www.surveyresearch.weebly.com SPSS start page: www.surveyresearch.weebly.com/1-survey-analysis-workshop -----Original Message----- Hi Jon, thank you. It works, but only unless there isn`t another variable which doesn't start with v_ in the dataset. I try to write the duration and lfdn after /KEEP= but then the whole syntax doesn't work, is there an adjustment possible? lfdn duration v_1 v_55 v_40_1 v_40_2 v_40_3 v_40_4 v_41_1 v_41_2 v_41_3 v_41_4 v_42_1 v_42_2 v_42_3 v_42_4 v_43_1 -- View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Sort-Variables-tp5731970p5733943.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 |
«
Return to SPSSX Discussion
|
1 view|%1 views
| Free forum by Nabble | Edit this page |