I analysed records for Mumbai residents (about 10,000 respondents) on their water sources and toilet facilities. I had created a new variable for Mumbai slum households versus non slum households from the variable "SELECTED CITIES=5, that includes only Mumbai residents). Then I ran a cross tabs with all the water and sanitation related variables. I think the CPU is too small.
On Mon, Sep 26, 2016 at 8:30 PM, Alan Mead <ame...@alanmead.org> wrote: > Can you tell is which variable(s) you analyzed using frequencies? > > -Alan > > > On 9/26/2016 9:48 AM, Cecily Ray wrote: > > 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 >>>> >>>> >>>> >>>> >>>> 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 >>>>> >>>>> >>>>> >>>> >>> >> > > -- > > Alan D. Mead, Ph.D. > President, Talent Algorithms Inc. > > science + technology = better workers > > +815.588.3846 (Office) > +267.334.4143 (Mobile) > http://www.alanmead.org > > I've... seen things you people wouldn't believe... > functions on fire in a copy of Orion. > I watched C-Sharp glitter in the dark near a programmable gate. > All those moments will be lost in time, like Ruby... on... Rails... Time for > Pi. > > --"The Register" user Alister, applying the famous > "Blade Runner" speech to software development > >
_______________________________________________ Pspp-users mailing list Pspp-users@gnu.org https://lists.gnu.org/mailman/listinfo/pspp-users
