Thanks to Harry, Ann and John. There are 109041 respondents (records) in the dataset.
I am interested in analyzing 1175 of these records having a certain response (recoded as a particular variable. I believe no variable has more than about 5 or 6 distinct responses (most are not continuous variables, certainly not the few of interest). Ann, are you saying you have analysed Demographic and Health Survey (or National Family Health Survey) data for India? What kind of machine did you use? I feel this should run on an ordinary laptop. I closed Chrome and like I said, the performance improved but was still slow (got "non response"), but got results. Now we shall try some further calculations. Regards, Cecily On Mon, Sep 26, 2016 at 6:58 PM, Harry Thijssen <pspp4wind...@gmail.com> wrote: > That leaves a few unpleasant options: > > a) the installation is not correct. (Maybe mixed build generations by > installing versions from different build generations without removing the > old version first). Unlikely this is the problem, but you can give it a > try be removing the directory where PSPP is installed, and install PSPP > again (maybe a newer version but I doubt a newer version will make any > difference) > b) your hardware, especially the harddisk or its driver has a problem. Is > it possible to have your hardware checked? However you wrote that you > already tried on another laptop (of your daughter) so this doens't look as > a likely case. > c) you and your daughter have some program on your computer which greatly > slows down disk IO. You could check it by testing on a clean installed PC. > > Have fun > > >I have run the same analyses with the same data set with no problems. So > I > > don't think it's the data set. > > >Ann E. Dunlop, M.A. > >Doctoral Candidate,HDFS > >Michigan State University > > 2016-09-26 10:37 GMT+02:00 Cecily Ray <cecily....@gmail.com>: > >> Dear Alan, Ann and Harry, >> >> Many thanks for your inputs on my issues of slow PSPP. >> I have used the Task Manager as suggested by Alan and >> found that often there is an alternation between PSPP >> and System Idle Manager using a high percentage of memory. >> This happens when PSPP is saying (Not Responding) >> >> My laptop has 4 GB of memory in two main memory slots, according >> to the product specification sheet. >> >> If I can get a chance later I will send a screenshot. >> >> Regards, >> Cecily Ray >> >> >> On Sat, Sep 24, 2016 at 7:47 PM, Harry Thijssen <pspp4wind...@gmail.com> >> wrote: >> >>> >>> Hi Cecily >>> >>> I don't think PSPP is to slow with big datasets. 500 Mb is not that big. >>> I just think you are not working with an optimal configuration. >>> >>> The tests Alan suggested would have given more info about the problem so >>> people could give you an advice. May guess at the moment is that you should >>> use a bigger workspace or have a problem with your tmp files on a slow >>> disk. The later is unlikely as I expect just 1 disk in your laptop. >>> >>> Assuming you have a laptop with at least 4 Gb memory I would recommend >>> setting your workspace to 500 Mb or more. You can do this in the >>> syntaxeditor window with: >>> >>> set WORKSPACE=524288. >>> >>> You can check this with: >>> >>> show ALL. >>> >>> You find the syntax editor by going in PSPPIRE to the file tab and then >>> syntax. >>> >>> If you could tell the exact file you downloaded for your work people >>> could try to find out what is going on and help you. It would also be >>> usefull if you could post what happens if you change your workspace >>> settings as suggested above. >>> >>> Have fun >>> >>> >>> >>> >>> <https://lists.gnu.org/mailman/listinfo/pspp-users> Date: Sat, 24 Sep >>>> 2016 00:08:10 +0530 >>>> From: Cecily Ray <cecily....@gmail.com> >>>> To: Alan Mead <ame...@alanmead.org> >>>> Cc: "pspp-users@gnu.org" <pspp-users@gnu.org> >>>> Subject: Re: PSPP not working >>>> Message-ID: >>>> <cafeo8o4pcfgntaocxf0+4mj8itjful7yj0np-myi1qpwy3z...@mail.gm >>>> ail.com> >>>> Content-Type: text/plain; charset="utf-8" >>>> >>>> Alan, >>>> >>>> Thank you for your helpful replies. I now understand that PSPP works >>>> fine >>>> with smaller datasets than with the large one I am using. I think I will >>>> have >>>> to go for a trial version of SPSS. >>>> >>>> Cecily >>>> >>>> >>>> >>> >> >
_______________________________________________ Pspp-users mailing list Pspp-users@gnu.org https://lists.gnu.org/mailman/listinfo/pspp-users
