[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: PSPP-BUG: PSPP CVS Build
From: |
Ben Pfaff |
Subject: |
Re: PSPP-BUG: PSPP CVS Build |
Date: |
Tue, 20 Jun 2006 16:24:40 -0700 |
User-agent: |
Gnus/5.110004 (No Gnus v0.4) Emacs/21.4 (gnu/linux) |
John Darrington <address@hidden> writes:
> On Tue, Jun 20, 2006 at 06:54:52AM -0700, Ben Pfaff wrote:
> > DATA LIST LIST /IP (a15).
> > BEGIN DATA.
> > 192.168.000.2
> > 192.168.000.4
> > 192.168.000.5
> > 192.168.000.6
> > 127.0.0.1
> > END DATA.
> >
> > FREQUENCIES /VARIABLES=IP .
> >
> > it will report a single variable "192.168." with a frequency of 4,
> > whereas it's in fact four variables each with a frequency of 1.
>
> It's what SPSS does and is documented to do. We could support
> frequencies on full long string variables pretty easily, as an
> extension. I guess it *is* desirable, as you say.
>
> Oh well, I guess that's a candidate for something the
> --algorithm{compatible|enhanced} flag should control. In the
> meantime, let's at least make sure that our documentation explicitly
> mentions this rather unintuitive behaviour.
I'm going to try to work on this tonight.
--
Ben Pfaff
email: address@hidden
web: http://benpfaff.org