|
Hi,
**Additional problem: my new file does not contain the variable N_Break. (I got the code from http://www.ats.ucla.edu/stat/spss/faq/nummiss_spss.htm) Thanks. What I want: a new file with just the variables listed in the code below that is located where I want it, and opens at the end of the program so that I can see the results. What I tried: I used this syntax but the outfile was saved in a location I don't want. AGGREGATE /OUTFILE='AGGRTest.SAV' /BREAK= EmpStat1 MaritStat1 HousStat1 RaceEth11 RaceEth21 EdLevel1 Rx_source1 Trans_yn1 /N_BREAK=N. I tried changing the default location using edit->options->specified folder-> and change the location for both data files and other files but that didn't work so I added the following line Save OUTFILE="C:\Jan\PhD\PhD_analysis\VStudy1DT/AGGRTest.SAV". Now it saves AGGRTest.SAV over my original datafile (with all of my 150 original variables) and adds the variables listed above to the end of the file. It does open automatically when I run the program. Thanks, 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 |
|
Dear Jan,
My advice is this: look up in the command syntax reference: -dataset name -dataset close -dataset activate -dataset copy -dataset declare (you can abbreviate it to 'datas nam'/'datas dec' if you like, you can't abbreviate 'dataset' any further than 'datas') Working with datasets takes a couple of days to get used to but you'll soon wish you'd discovered this before!! I know many listers prefer file handles but I like to put a CD command in my syntax like cd 'c:/temp'. Now I don't need any paths anymore and if my project moves to another folder, I modify it to *cd 'c:/temp'. cd 'd:/temp'. As soon as you open any .sav file, give the dataset a name like d1. Like get fil'blah.sav'. dataset name d1. If you want to create new data with aggregate (assuming you've only d1 open), use something like dataset declare d2. aggregate /outfile d2 /blah blah.... Now you can inspect your new data before you save it. If you don't like it, run dataset close d2. dataset activate d1. And try the aggregate again. HTH,
Ruben van den Berg Consultant Models & Methods TNS NIPO Email: [hidden email] Mobiel: +31 6 24641435 Telefoon: +31 20 522 5738 Internet: www.tns-nipo.com > Date: Sun, 22 Aug 2010 10:31:40 -0700 > From: [hidden email] > Subject: Revised question: Syntax for outfile > To: [hidden email] > > Hi, > **Additional problem: my new file does not contain the variable N_Break. > (I got the code from > http://www.ats.ucla.edu/stat/spss/faq/nummiss_spss.htm) Thanks. > > What I want: a new file with just the variables listed in the code below > that is located where I want it, and opens at the end of the program so > that I can see the results. > What I tried: I used this syntax but the outfile was saved in a location > I don't want. > AGGREGATE > /OUTFILE='AGGRTest.SAV' > /BREAK= EmpStat1 MaritStat1 HousStat1 RaceEth11 RaceEth21 > EdLevel1 Rx_source1 Trans_yn1 > /N_BREAK=N. > I tried changing the default location using edit->options->specified > folder-> and change the location for both data files and other files but > that didn't work so I added the following line > Save OUTFILE="C:\Jan\PhD\PhD_analysis\VStudy1DT/AGGRTest.SAV". > Now it saves AGGRTest.SAV over my original datafile (with all of my 150 > original variables) and adds the variables listed above to the end of > the file. It does open automatically when I run the program. > Thanks, > 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 |
|
Sorry for my previous reply to Jan, I didn’t see the answers in
the revised topic name until now. Antoon Van: SPSSX(r) Discussion
[mailto:[hidden email]] Namens Ruben van den Berg Dear
Jan, My
advice is this: look up in the command syntax reference: -dataset
name -dataset
close -dataset
activate -dataset
copy -dataset
declare (you
can abbreviate it to 'datas nam'/'datas dec' if you like, you can't abbreviate
'dataset' any further than 'datas') Working
with datasets takes a couple of days to get used to but you'll soon wish you'd
discovered this before!! I
know many listers prefer file handles but I like to put a CD command in my
syntax like cd
'c:/temp'. Now
I don't need any paths anymore and if my project moves to another folder, I
modify it to *cd
'c:/temp'. cd
'd:/temp'. As
soon as you open any .sav file, give the dataset a name like d1. Like get
fil'blah.sav'. dataset
name d1. If
you want to create new data with aggregate (assuming you've only d1 open), use
something like dataset
declare d2. aggregate /outfile
d2 /blah
blah.... Now
you can inspect your new data before you save it. If you don't like it, run dataset
close d2. dataset
activate d1. And
try the aggregate again. HTH, |
| Free forum by Nabble | Edit this page |
