pspp-users
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: PSPP - Spearman correlation really not available?


From: John Darrington
Subject: Re: PSPP - Spearman correlation really not available?
Date: Mon, 23 Jan 2017 13:54:36 +0100
User-agent: Mutt/1.5.23 (2014-03-12)

[ Moving discussion to pspp-users since this is a question about usage]

Spearman correlation can be done through the CROSSTABS command.  For example:

crosstabs /tables X by Y 
 /statistics = corr.

will give you both the Pearson and Spearman correlation between the variables X 
and Y.

It's true that it is not easy to do multiple correlations in one go.
For this we need the NONPAR CORR command which unfortunately doesn't
exist (yet).


I don't know why you are having difficulty copying from the output screen.  For 
me
it works fine: Use Edit|Select All and then Edit|Copy and this will put the 
whole 
screen into the clipboard.
(If you're running in the de_AT locale then presumably the above labels will 
read
Bearbeiten|Alles auswahlen   and  Bearbeiten|Kopieren)

J'

On Mon, Jan 23, 2017 at 10:15:26AM +0000, Diamond Lisa wrote:
     
     First, thanks so much for this program! I am really glad it is available 
and use it regularly.
     One feature I really miss though is the possibility of calculating a rank 
correlation - I need this so often as I work with a lot of data where the 
distribution is not normal and every time I start looking around if this is 
really not possible with PSPP. It seems such a small and basic thing but so far 
I have not encountered a solution. Is this really the case? I have a hard time 
believing this as there are so many much more advanced functions available for 
analysis. Is there maybe a way via the syntax, like with the Kruskal Wallis 
test and other nonparametric procedures? I see it listed in the manual under 
"CORR" but there is no information on how to make it happen anywhere and if I 
set the variable types to "ordinal" I still get a Pearson correlation according 
to the output.
     
     Another side note is that it would be fantastic if at some point it would 
be possible to copy directly from the output instead of having to save it first 
as a different file type - but that is relatively minor while the missing 
Spearman option is really a problem for me.
     
     
-- 
Avoid eavesdropping.  Send strong encrypted email.
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.

Attachment: signature.asc
Description: Digital signature


reply via email to

[Prev in Thread] Current Thread [Next in Thread]