Benjamin Oppermann | 2 Nov 01:02 2014

Re: problems importing csv files

Hello Friedrich,
Thank you! If you deleted the variable description and removed the line breaks before importing into PSPP, which program did you use for it? My text editor (Kate) will only open these files write protected, as the lines are too long.
How did you find the <CR><LF>? some look like line breaks in a text editor, but aren't. Also I don't know how to find the variable descriptions in the text file.
Ben
 
 
Am Sa, 1. Nov 2014, um 23:04, schrieb Friedrich Beckmann:
Hi Ben,
 
i looked in your file and I found two problems:
 
a) I think the language encoding is different in the Variable description (utf-8) and in the cases (latin-1). 
 
b) Some text variable contents contain a <CR><LF> like in 
"1 Jahr in Mexiko <CR><LF> 1/2 Jahr in Spanien, Andalusien“ or 
"Österreich 30 Jahre  <CR><LF> Frankreich 2 Jahre <CR><LF>Spanien 6 Monate "
 
I deleted the variable descriptions in the beginning and only left the case data. Then 
I replaced <CR><LF> with comma. Then I could import your file. So the problem is that pspp
does not accept  <CR><LF> in text variable contents. It tried this with version 0.8.3.
 
The modified file is attached. 
 
Friedrich
 
 
 
Am 01.11.2014 um 19:21 schrieb Benjamin Oppermann <ben.opp <at> eml.cc>:
 
Hello,
This is psppire 0.7.9 on Kubuntu 14.04. All software was installed and
kept up to date from the GNU/Linux distro repositories.

I can't get this .csv file correctly imported into pspp. I opened it in
Libreoffice Calc  (v.4.2.6.3) and saved as .osd.
To be honest, I'm not sure whetherI should choose commas or semicolons
or both as separating chars.
In Calc, semicolons only or bot seems to look neat, but PSPP jumbles
everything up.
Some variables have mixed data types, e.g. the text of the given answer
mixed with the numbers corresponding to the answer.
For reference, I attached one file that was output as a result of the
same online survey, and has fewer cases than the one I'm working with
now, but the same variables.
Thanks, Ben
<ver40-abc.20.csv>_______________________________________________
Pspp-users mailing list
Pspp-users <at> gnu.org
https://lists.gnu.org/mailman/listinfo/pspp-users
 
--
Friedrich Beckmann, Bismarckstraße 11, 86159 Augsburg
Tel: 0821-44923693, Mobil: 017648576782
 
 
 
 

Email had 1 attachment:

  • ver40-abc.20.modified.csv
      54k (text/csv)
 
_______________________________________________
Pspp-users mailing list
Pspp-users <at> gnu.org
https://lists.gnu.org/mailman/listinfo/pspp-users
Benjamin Oppermann | 1 Nov 19:21 2014

problems importing csv files

Hello,
This is psppire 0.7.9 on Kubuntu 14.04. All software was installed and
kept up to date from the GNU/Linux distro repositories.

I can't get this .csv file correctly imported into pspp. I opened it in
Libreoffice Calc  (v.4.2.6.3) and saved as .osd.
To be honest, I'm not sure whetherI should choose commas or semicolons
or both as separating chars. 
In Calc, semicolons only or bot seems to look neat, but PSPP jumbles
everything up.
Some variables have mixed data types, e.g. the text of the given answer
mixed with the numbers corresponding to the answer.
For reference, I attached one file that was output as a result of the
same online survey, and has fewer cases than the one I'm working with
now, but the same variables.
Thanks, Ben
_______________________________________________
Pspp-users mailing list
Pspp-users <at> gnu.org
https://lists.gnu.org/mailman/listinfo/pspp-users
news | 1 Nov 15:11 2014
Picon

delete selected output

Hi,
I guess it is not possible, but maybe I have not seen the trick: How can 
I delete some of the output, for instance a table ?

I ask as I want to output a single table into the LibreOffice 
spreadsheet and to allow this I have to produce a html output of all the 
results - which can produce hundreds of pages if you work for several 
hours - and then import the html table into Calc. This is cumbersome, 
but I see this as the only way to get tables into Calc (or any other 
spreadsheet).

Or did you find a more direct way ?

The best way, of course, would be to be able to export a single output 
element, such as a table, and not be obliged to use the entire output.

I use PSPP 0.8.4 with PSPPIRE.

TIA

ftr
_______________________________________________
gnumeric-list mailing list
gnumeric-list <at> gnome.org
https://mail.gnome.org/mailman/listinfo/gnumeric-list
F. Thomas | 30 Oct 23:22 2014
Picon

PSPP does not answer

Hi,
I work with PSPP latest version under windows 7 and every time I make a 
go through my 67,000 cases for instance with a Frequencies command PSPP 
does not answer for a minute.
Is this standard behaviour ?

TIA
ftr
John Darrington | 29 Oct 09:34 2014
Picon

Re: PSPP-BUG: Bug in Factor Analysis

I don't think this is a bug.  So I am replying to pspp-users <at> gnu.org

On Tue, Oct 21, 2014 at 01:47:11PM +0800, dear bharat wrote:
     I have been trying a couple of Factor Analysis using PSPP. At times, the
     initial communalities are not set to one. Is there a way to set it to 1.

As I recall, the initial communalities are unity for Principal Components Analysis
for Principal Axis Factoring they are unity only if the data are correlated.

     
     Also how can I compute factor scores from the analysis.

The factor scores are printed in the box headed "Factor Matrix" - but of course
that will only appear for Principal Axis Factoring.

J'

--

-- 
PGP Public key ID: 1024D/2DE827B3 
fingerprint = 8797 A26D 0854 2EAB 0285  A290 8A67 719C 2DE8 27B3
See http://sks-keyservers.net or any PGP keyserver for public key.

_______________________________________________
Pspp-users mailing list
Pspp-users <at> gnu.org
https://lists.gnu.org/mailman/listinfo/pspp-users
Miran Čas | 24 Oct 11:01 2014
Picon

RE: Pspp-users Digest, Vol 101, Issue 18

Dear colleagues!

Due to longer treatment at home, I ask you to stop sending me your e-mail messages, for now.
Thank you.

With best regards,
 Miran

-----Original Message-----
From: pspp-users-bounces+miran.cas=gozdis.si <at> gnu.org
[mailto:pspp-users-bounces+miran.cas=gozdis.si <at> gnu.org] On Behalf Of pspp-users-request <at> gnu.org
Sent: Friday, October 24, 2014 10:51 AM
To: pspp-users <at> gnu.org
Subject: Pspp-users Digest, Vol 101, Issue 18

Send Pspp-users mailing list submissions to
	pspp-users <at> gnu.org

To subscribe or unsubscribe via the World Wide Web, visit
	https://lists.gnu.org/mailman/listinfo/pspp-users
or, via email, send a message with subject or body 'help' to
	pspp-users-request <at> gnu.org

You can reach the person managing the list at
	pspp-users-owner <at> gnu.org

When replying, please edit your Subject line so it is more specific than "Re: Contents of Pspp-users digest..."

Today's Topics:

   1. Re: Re: scatterplot (Harry Thijssen)
   2. Re: AW: AW: scatterplot (news)
   3. Re: scatterplot (Friedrich Beckmann)
   4. oblimin rotation? (Matej Kovacic)
   5. Re: oblimin rotation? (Germano Rossi)
   6. Re: oblimin rotation? (Matej Kovacic)
   7. Re: oblimin rotation? (Joseph Saint Pierre)

----------------------------------------------------------------------

Message: 1
Date: Thu, 23 Oct 2014 18:33:19 +0200
From: Harry Thijssen <pspp4windows <at> gmail.com>
To: "pspp-users <at> gnu.org" <pspp-users <at> gnu.org>,
	friedrich.beckmann <at> gmx.de,	beat.hulliger <at> fhnw.ch,
	frank_thomas <at> orange.fr
Subject: Re: Re: scatterplot
Message-ID:
	<CAMMrg9Rprv5d8UTtUvsn7mT6_-bECdNSzXSzCpLAQO4g9c_aag <at> mail.gmail.com>
Content-Type: text/plain; charset="iso-8859-1"

>
>
> Content-Type: text/plain; charset="windows-1252"; Format="flowed"
>
> Ok, this explains that. Thank you.
> ok, I try SciDavis.
>
> ftr
>
> On 23/10/2014 14:10, Friedrich Beckmann wrote:
> > Hi,
> >
> > the scatterplot is not included in the 0.8.4 release. So you can see 
> > it only if you build pspp from the current sources. The description 
> > which commands work and which not are in the documentation.
> >
> > Friedrich
>

Hi Frank

It looks your message is coming from an MSWindows system. In that case:

The PSPP MSWindows builds on sourceforge are newer as the version number indicates.
You could see it as a 0.8.4+ version. In the help there is a complete version number which is something like
0.8.4-?????? or you can look at the file name of the installer which includes the date of the sources.

The latest release candidate is the version you need.

Have fun
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.gnu.org/archive/html/pspp-users/attachments/20141023/0548592e/attachment.html>

------------------------------

Message: 2
Date: Thu, 23 Oct 2014 21:13:50 +0200
From: news <news.ftr <at> free.fr>
To: pspp-users <at> gnu.org
Subject: Re: AW: AW: scatterplot
Message-ID: <5449536E.7080000 <at> free.fr>
Content-Type: text/plain; charset=utf-8; format=flowed

Thank you for the hint for the newer version .

But I don't find the syntax that works. Where do you find a syntax list ?

graph  /scatterplot=y with x.
graph  /scatterplot=y by x.

Neither of two worked.

- ftr

On 23/10/2014 12:57, Hulliger Beat wrote:
> You must use the syntax, e.g.:
>
> graph
>   /scatterplot=y with x.
>
> With my version the dots are yellow, a bad choice for most screens, but still: it works!
>
> Good luck!
>
> Beat
>
>
> -----Urspr?ngliche Nachricht-----
> Von: pspp-users-bounces+beat.hulliger=fhnw.ch <at> gnu.org 
> [mailto:pspp-users-bounces+beat.hulliger=fhnw.ch <at> gnu.org] Im Auftrag 
> von F. Thomas
> Gesendet: Donnerstag, 23. Oktober 2014 12:39
> An: pspp-users <at> gnu.org
> Betreff: Re: AW: scatterplot
>
> Where did you find it ?
> GRAPH, CGRAPH are among the functions not implemented and are not 
> listed in the command index :-( -ftr
>
> On 23/10/2014 11:44, Hulliger Beat wrote:
>> I just realised that the graph command is available with the syntax but not yet in the PSPPIRE.
Scatterplot and histograms work! Sorry for overlooking this!
>>
>> Still barcharts seem to be missing.
>>
>> Beat
>>
>>
>> -----Urspr?ngliche Nachricht-----
>> Von: pspp-users-bounces+beat.hulliger=fhnw.ch <at> gnu.org
>> [mailto:pspp-users-bounces+beat.hulliger=fhnw.ch <at> gnu.org] Im Auftrag 
>> von news
>> Gesendet: Mittwoch, 22. Oktober 2014 18:59
>> An: pspp-users <at> gnu.org
>> Betreff: Re: scatterplot
>>
>> A intermediate solution could be the easy exportation of results to gnumeric, or Libreoffice.
>>
>> If not I support this idea a,d add
>> - scatter plot with a non-numerical grouping variable
>> 	such as age by income gy gender.
>>
>> Regards,
>> ftr
>>
>> On 22/10/2014 10:55, Hulliger Beat wrote:
>>> Are there plans for
>>>
>>> i)implementing (grouped) boxplots, histograms and normalplot into 
>>> the user interface psppire?
>>>
>>> ii)adding a scatterplot to the correlation?
>>>
>>> iii)adding a barchart to frequencies (for nominal)?
>>>
>>> iv)adding the Tukey-Anscombe plot (residuals versus predicted) to 
>>> the regression?
>>>
>>> The plots do not need to be interactive (like GGRAPH) but just 
>>> decent static plots.
>>>
>>> Thanks
>>>
>>> Beat
>>>
>>>
>>>
>>> _______________________________________________
>>> Pspp-users mailing list
>>> Pspp-users <at> gnu.org
>>> https://lists.gnu.org/mailman/listinfo/pspp-users
>>>
>> _______________________________________________
>> Pspp-users mailing list
>> Pspp-users <at> gnu.org
>> https://lists.gnu.org/mailman/listinfo/pspp-users
>>
>> _______________________________________________
>> Pspp-users mailing list
>> Pspp-users <at> gnu.org
>> https://lists.gnu.org/mailman/listinfo/pspp-users
>>
>
>
> _______________________________________________
> Pspp-users mailing list
> Pspp-users <at> gnu.org
> https://lists.gnu.org/mailman/listinfo/pspp-users
>
> _______________________________________________
> Pspp-users mailing list
> Pspp-users <at> gnu.org
> https://lists.gnu.org/mailman/listinfo/pspp-users
>

------------------------------

Message: 3
Date: Fri, 24 Oct 2014 00:20:44 +0200
From: Friedrich Beckmann <friedrich.beckmann <at> gmx.de>
To: news <news.ftr <at> free.fr>
Cc: pspp-users <at> gnu.org
Subject: Re: scatterplot
Message-ID: <BB4D786E-BF52-465F-9029-FE5B0073C536 <at> gmx.de>
Content-Type: text/plain; charset="us-ascii"

Am 23.10.2014 um 21:13 schrieb news <news.ftr <at> free.fr>:

> Thank you for the hint for the newer version .
> 
> But I don't find the syntax that works. Where do you find a syntax list ?
> 
> graph  /scatterplot=y with x.
> graph  /scatterplot=y by x.
> 
> 
> Neither of two worked.
> 

Hi, you can find the latest documentation here:

http://pspp.benpfaff.org/~blp/pspp-master/latest/source/user-manual/pspp.html#GRAPH

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.gnu.org/archive/html/pspp-users/attachments/20141024/c1abdc9d/attachment.html>

------------------------------

Message: 4
Date: Fri, 24 Oct 2014 09:31:59 +0200
From: Matej Kovacic <matej.kovacic <at> owca.info>
To: pspp-users <at> gnu.org
Subject: oblimin rotation?
Message-ID: <544A006F.7000705 <at> owca.info>
Content-Type: text/plain; charset=iso-8859-2

Hi,

I have a question about factor analysis, specifically the rotations. Are there any plans to implement
oblimin or any other non-orthogonal rotation?

BTW, I was successfull to persuade Faculty of Health Care (http://www.fzj.si/en/ - Slovenia, Europe) to
switch from SPSS to PSPP.
They have found out PSPP currently has almost everything they need, except some non-orthogonal rotation
for factor analysis.

Thanks again for this great software.

Regards,

M.

------------------------------

Message: 5
Date: Fri, 24 Oct 2014 10:16:17 +0200
From: Germano Rossi <germano.rossi <at> gmail.com>
To: pspp-users <at> gnu.org
Subject: Re: oblimin rotation?
Message-ID:
	<CAOQDsF_Y_c4it1oPJKRcBx62k0jvooAS0AvTarT4XYNy20daeg <at> mail.gmail.com>
Content-Type: text/plain; charset="utf-8"

2014-10-24 9:31 GMT+02:00 Matej Kovacic <matej.kovacic <at> owca.info>:

> I have a question about factor analysis, specifically the rotations. 
> Are there any plans to implement oblimin or any other non-orthogonal rotation?
>
> BTW, I was successfull to persuade Faculty of Health Care 
> (http://www.fzj.si/en/ - Slovenia, Europe) to switch from SPSS to PSPP.
> They have found out PSPP currently has almost everything they need, 
> except some non-orthogonal rotation for factor analysis.
>

PSPP is at version 0.84 not at 1.x and is an "open" source program. This mean "gratis" for us (to use) and for
the programmers (to write).
In my opinion, it is premature to abandon SPSS for PSPP if you use it for research.
It's fine for teaching statistics or introduction to data analysis.
If SPSS is too expensive, I suggest you to use a combination of PSPP and R (both free).
Eg., In version 0.84, for the extraction (to date) PSPP uses principal components or principal axes, not
maximum likelihood o minimum residual.
Good for teaching, not for a serious research analysis. But also SPSS is limited, not being able to apply fit
indices (as does R with the package psych).

Germano
--
==================
Germano Rossi, Dipartimento di Psicologia, Universita' degli Studi di Milano Bicocca Piazza
dell'Ateneo Nuovo, 1- 20126 Milano - Italy
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.gnu.org/archive/html/pspp-users/attachments/20141024/7f4ad2cd/attachment.html>

------------------------------

Message: 6
Date: Fri, 24 Oct 2014 10:26:55 +0200
From: Matej Kovacic <matej.kovacic <at> owca.info>
To: pspp-users <at> gnu.org
Subject: Re: oblimin rotation?
Message-ID: <544A0D4F.8010901 <at> owca.info>
Content-Type: text/plain; charset=windows-1252

Hi,

> In my opinion, it is premature to abandon SPSS for PSPP if you use it 
> for research.
> It's fine for teaching statistics or introduction to data analysis.
> If SPSS is too expensive, I suggest you to use a combination of PSPP 
> and R (both free).

I know. That is why PSPP is used for teaching basic statistics only.

Regards,

M.

------------------------------

Message: 7
Date: Fri, 24 Oct 2014 10:51:06 +0200
From: Joseph Saint Pierre <josephsaintpierre <at> gmail.com>
To: pspp-users <at> gnu.org
Subject: Re: oblimin rotation?
Message-ID:
	<CABmjFSC99x5LFj=U3DHnNFowcyepPawF9shsDNOHT+ZB565D_g <at> mail.gmail.com>
Content-Type: text/plain; charset=ISO-8859-1

2014-10-24 10:26 UTC+02:00, Matej Kovacic <matej.kovacic <at> owca.info>:
> Hi,
>
>> In my opinion, it is premature to abandon SPSS for PSPP if you use it 
>> for research.
>> It's fine for teaching statistics or introduction to data analysis.
>> If SPSS is too expensive, I suggest you to use a combination of PSPP 
>> and R (both free).
>
> I know. That is why PSPP is used for teaching basic statistics only.

Hello.

It is exactly what I thought when I decided in 2005 to supervise writing a PSPP tutorial in french. PSPP is
really great for basic statistics. To perform complex methods R is better.
GPArotation package can be used for oblimin rotation in R. But I think PSPP is better than R to handle large
data files.

Joseph Saint Pierre
http://cict.fr/~stpierre

------------------------------

_______________________________________________
Pspp-users mailing list
Pspp-users <at> gnu.org
https://lists.gnu.org/mailman/listinfo/pspp-users

End of Pspp-users Digest, Vol 101, Issue 18
*******************************************
Matej Kovacic | 24 Oct 09:31 2014

oblimin rotation?

Hi,

I have a question about factor analysis, specifically the rotations. Are
there any plans to implement oblimin or any other non-orthogonal rotation?

BTW, I was successfull to persuade Faculty of Health Care
(http://www.fzj.si/en/ - Slovenia, Europe) to switch from SPSS to PSPP.
They have found out PSPP currently has almost everything they need,
except some non-orthogonal rotation for factor analysis.

Thanks again for this great software.

Regards,

M.
Hulliger Beat | 22 Oct 10:55 2014
Picon

scatterplot

Are there plans for

i)                 implementing (grouped) boxplots, histograms and normalplot into the user interface psppire?

ii)                adding a scatterplot to the correlation?

iii)              adding a barchart to frequencies (for nominal)?

iv)              adding the Tukey-Anscombe plot (residuals versus predicted) to the regression?

 

The plots do not need to be interactive (like GGRAPH) but just decent static plots.

 

Thanks

 

Beat

 

_______________________________________________
Pspp-users mailing list
Pspp-users <at> gnu.org
https://lists.gnu.org/mailman/listinfo/pspp-users
Ksenia 123 | 18 Oct 20:42 2014
Picon

The output window do not open in the PSPP on the Ubuntu 14.04

Hallo!

Now I try ti use the programm psppire 0.7.9 version on the Ubuntu 14.04, Intel® Core™ i7-3770 CPU <at> 3.40GHz × 8.

My problem is that: the new output window for results of analyses don`t open. On the my left pannel I see that PSPP have 2 windows, the window with the data I see, the window 'output" i do not see. When I  choose the button "PSPP" on this left pannel, I see the main window and the small square. I have the screenshot   of this. When I try go to the outputwindow in the data window ("Windows" - - "Output") nothing is changing.

I can not find answer to some problem in the internet and I can`t undersatand, what is it: the problem of my system or the programm bug.


With the best regards,
Ksenia
kotelnikovakseni <at> inbox.ru
_______________________________________________
Pspp-users mailing list
Pspp-users <at> gnu.org
https://lists.gnu.org/mailman/listinfo/pspp-users
Javier Serrano Andreu | 5 Oct 20:29 2014
Picon

Time series analysis

Hi,

I would like to know when the PSPP program will incorporate the analysis of time series, such as Kaplan-Meier, Cox regression, etc.

I´m very interested in that point.

Best regards.

Thank you.



Javier Serrano Andreu, MD, PhD.
Médico Especialista en Oncología Radioterápica. H. G. U. Gregorio Marañón.
Profesor Asociado. Facultad de Medicina. Universidad Complutense.
Madrid.





_______________________________________________
Pspp-users mailing list
Pspp-users <at> gnu.org
https://lists.gnu.org/mailman/listinfo/pspp-users
Ben Pfaff | 5 Oct 20:07 2014
Picon

progress on understanding .spv files

I want to say "thank you" to all of the helpful people who sent me
sample .spv and corresponding .pdf files last week.  I have spent
numerous hours looking through them and comparing them against the
rendered versions.  I have not yet entirely figured out the format, but
I've made a lot of progress.  After I do some more work, I guess that I
will have some new and more specific requests for sample files, but I
don't have any yet.

Thanks again!

Gmane