completely eradicating value labels

[TinyTina]

Hi everybody,

I have a data set with 1000s of fields with a constant numeric code ("0") and
a corresponding contant value label, e.g. "No data". I have to remove code
and label completely for display by CTABLES. Otherwise CTABLES will still
display a "0". And we dont want that.

In Oct 2008, Jon Peck mentioned a  ValueLabels property of an
spssaux.VariableDict object as a way to to completely eradicate the value
labels, or you yo have sth more handy up your sleeve?

Any help greatly appreciated!

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[Jon K Peck]

If the value label sets are actually identical for all these variables, you could do this most easily by editing one to remove the zero and then using Data> Copy Data Properties selecting only the value labels

Otherwise, assuming that you have the Python plugin installed (Python Essentials from the SPSS Community site, which is
www.ibm.com/developerworks/spssdevcentral)
this program will remove any labels for the value 0.

Indentation is important, so I've used dashes  to protect it from the email demons.  Replace each dash with a space.

begin program.
import spss, spssaux
for v in spssaux.VariableDict():
--vl = v.ValueLabels
--if '0' in vl:
----del(vl['0'])
----v.ValueLabels = vl
end program.

HTH,
Jon Peck (no "h") aka Kim
Senior Software Engineer, IBM
[hidden email]
new phone: [phone]




From:        TinyTina <[hidden email]>
To:        [hidden email]
Date:        05/21/2012 01:08 PM
Subject:        [SPSSX-L] completely eradicating value labels
Sent by:        "SPSSX(r) Discussion" <[hidden email]>

---

Hi everybody,

I have a data set with 1000s of fields with a constant numeric code ("0") and
a corresponding contant value label, e.g. "No data". I have to remove code
and label completely for display by CTABLES. Otherwise CTABLES will still
display a "0". And we dont want that.

In Oct 2008, Jon Peck mentioned a  ValueLabels property of an
spssaux.VariableDict object as a way to to completely eradicate the value
labels, or you yo have sth more handy up your sleeve?

Any help greatly appreciated!

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[TinyTina]

Rick,

MISSING VALUES ALL (0) helped,
VALUE LABELS ALL didnt.
CTABLES progs tricked me for a while to realize that
MISSING VALUES approach actually took effect
Thank you !

TinaTiny

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[Vlad Lopez]

Hi Tina,

To erase the value labels of consecutive variables (e.g. v1 to v100) try

value labels v1 to v100 ''.

 

and for all the variables try

value labels all ''.

 

Notice the single quotes ''. Both lines were tested and work.

 

vlad

 

 

On Mon, May 21, 2012 at 10:25 PM, SUBSCRIBE SPSSX-L Anonymous <[hidden email]> wrote:

Rick,

MISSING VALUES ALL (0) helped,
VALUE LABELS ALL didnt.
CTABLES progs tricked me for a while to realize that
MISSING VALUES approach actually took effect
Thank you !

TinaTiny

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD