comp.soft-sys.stat.spss Google Group

free antivirus

sruthi...@googlemail.com (sruthi20f@googlemail.com) — Sun, 07 Sep 2008 07:09:28 UT

Re: SPSS 16 and Textpad

jkp...@gmail.com (JKPeck) — Sun, 07 Sep 2008 02:08:29 UT

TextPad is a third-party add-on. 16 did not support an api apparently
used by TextPad, which is why it wouldn't work AFAIK. 17 supports
that api, so it is possible that it would work again, but SPSS
Statistics 17 has an entirely new smart syntax editor. Syntax
coloring, pop-up command completion, error tracking, and bookmarks are

Re: SPSS 16 and Textpad

fred.sut...@lsc.gov.uk — Sat, 06 Sep 2008 17:45:20 UT

I don't think you can run textpad anymore, I have noticed with 16 that
previous syntax inn't 100% compatible, but only with minor characters
going a little screwy.
I have beta tested 17, and at last they've upgraded the syntax in
that, so you do have line numbers, colouring and error messages before

Re: Parsing Data?

bwea...@lakeheadu.ca (Bruce Weaver) — Fri, 05 Sep 2008 20:56:06 UT

This is in response to Damir's comment that he doesn't fully
understand the code. I've inserted some comments below.
This computes the position of the first blank space--i.e., between the
first id code or whatever it is and the name.
This computes the position of the final blank space (before the
date). The RTRIM is included to strip off any blanks at the end of

Adding n counts to columns in custom table

ba...@u.washington.edu — Fri, 05 Sep 2008 19:43:05 UT

Hi there,
I've created a table through the custom table pull-down menu and it
has everything I need except for a count of entries in each column,
i.e. the number of normal weight boys, normal weight girls, etc.
Anyone know any syntax I can use or options to select to add this to
my table? I've been playing around with adding /SLABELS but that's

Re: Parsing Data?

damirku...@bellsouth.net — Fri, 05 Sep 2008 17:41:00 UT

Thank you Bruce... while I don't fully understand the solution... it
worked like a charm. Thank you
Damir

Re: Parsing Data?

bwea...@lakeheadu.ca (Bruce Weaver) — Fri, 05 Sep 2008 17:18:35 UT

I don't see what the problem is. Here's an example.
* --- Start --- .
data list list / var(a50).
begin data
"1998CF009816AXX SMITH, JOHN 02-Jan-2007"
"1999CF123456BXYZ BUSH, GEORGE 01-Apr-2008"
end data.
compute blank1 = index(var,' ').
compute blank2 = rindex(rtrim(var,' '),' ').
format blank1 blank2 (f2.0).

Re: Parsing Data?

damirku...@bellsouth.net — Fri, 05 Sep 2008 14:54:42 UT

I like both suggestions... I am interested with the Read TXT function
especially. However, since the name has various end positions, that
means the date(11) also has different start positions... so date is
not always 10-21, it various as the names vary in length. Bruce's
suggestion to use the extract date wizard does not work as the names

SPSS 16 and Textpad

la...@augustmail.com (Lance Hoffmeyer) — Fri, 05 Sep 2008 14:37:06 UT

Hey all,
Wondering if anyone found a fix for the SPSS 16
and Textpad issue yet? Just installed SPSS 16 and
found out that Textpad will no longer run SPSS
because the runsyntx.exe file is no longer in SPSS 16.
Someone suggested copying runsyntx.exe from SPSS
15 and then one could run SPSS from Textpad.

Re: Parsing Data?

jazzg...@aol.com (Allen) — Fri, 05 Sep 2008 13:22:18 UT

Unless I don't get it, wouldn't it be easier to simply save the
existing file out as a flat ASCII and simply read it back into SPSS
using the "Read TXT function"?

Re: Remove all string characters from a string and leave numerics

bwea...@lakeheadu.ca (Bruce Weaver) — Fri, 05 Sep 2008 13:02:39 UT

Yes, good catch by Kenny & Fred. E.g., the code I posted turns
"100-300" into 100300, which is definitely not what you want.

Re: CTABLES

no...@nowhere.com (Kenny M) — Fri, 05 Sep 2008 11:47:26 UT

Answering my own question... My multiple response set was actually
called $Q5a so I was referring to it incorrectly. The capital Q
doesn't stop it from running the table but does stop the variable
label being suppressed. Got there in the end.

Re: CTABLES

no...@nowhere.com (Kenny M) — Fri, 05 Sep 2008 11:39:11 UT

To elaborate on this. I tested this same code on another dataset
(changing the reference to another categorical multiple response
variable) and it works. When I do the reverse and take working code
from the syntax file I use with the other data and use it on the
variable above, it produces the same error and the variable labels

Re: Remove all string characters from a string and leave numerics

fred.sut...@lsc.gov.uk — Fri, 05 Sep 2008 07:33:34 UT

I think I would agree with Kenny, you seem to have different answers
to the question, for example someone was saying how much they spend a
month, rather than a year. Presumably you would want to multiply this
by 12 to get an annual figure, to be consistent with the question.
You would need to do a fair bit of manipulation, before you get to the

Re: Parsing Data?

bwea...@lakeheadu.ca (Bruce Weaver) — Fri, 05 Sep 2008 01:37:54 UT

I don't have SPSS on this machine, but here are some suggestions.
1) Use the INDEX and RINDEX functions (in COMPUTE statements) to
find the positions of the blanks that separate the 3 parts of VAR.
2) Use the SUBSTR function (again, in COMPUTE statements) to
extract VAR1, VAR2 and VAR3. Use the positions of the 2 blanks