Quantcast

Set labels based on table

classic Classic list List threaded Threaded
6 messages Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Set labels based on table

Christian Buhtz
My observations has a field called "city". It is coded with ~40 numbers.
.e.g NewYork is 0, Berlin is 1, ...

Of course I want to set the lables for each value but I don't want to
do this manually. I have a another table (as a csv-file) holding the
values for the code numbers.

CODE;CITY
0;NewYork
1;Berlin
...

Is there a way in SPSS to use that csv-file to set the labels (not the
values!) automaticly?

An acceptable workaround would be when there is a way to change the
values from numbers to the city names.

=====================
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
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Set labels based on table

John F Hall

Another workaround would be to copy your list into Word (or Notepad) and make changes manually.

 

0;NewYork

1;Berlin

~ ~ ~

 

. . but it’s quicker to use Ctrl+H to change ; to [space] "

 

0 "NewYork

1 "Berlin

~ ~ ~

 

 

. . and the line ends  ^p^p to "

 

0 "NewYork"

1 "Berlin"

~ ~ ~

 

Then copy the list into a syntax file

 

value labels

    city     0 "NewYork"

1 "Berlin"

~ ~ ~ .                                  [Don’t forget the full stop!]

 

John F Hall (Mr)

[Retired academic survey researcher]

 

Email:   [hidden email] 

Website: www.surveyresearch.weebly.com

SPSS start page:  www.surveyresearch.weebly.com/1-survey-analysis-workshop

 

 

 

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Christian Buhtz
Sent: 19 September 2016 09:24
To: [hidden email]
Subject: Set labels based on table

 

My observations has a field called "city". It is coded with ~40 numbers.

.e.g NewYork is 0, Berlin is 1, ...

 

Of course I want to set the lables for each value but I don't want to do this manually. I have a another table (as a csv-file) holding the values for the code numbers.

 

CODE;CITY

0;NewYork

1;Berlin

...

 

Is there a way in SPSS to use that csv-file to set the labels (not the

values!) automaticly?

 

An acceptable workaround would be when there is a way to change the values from numbers to the city names.

 

=====================

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

===================== 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
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Set labels based on table

Christian Buhtz
On 2016-09-19 10:13 John F Hall <[hidden email]> wrote:
> Another workaround would be to copy your list into Word (or Notepad)
> and make changes manually.

Yes, using a good texteditor (instead of a word processor) is my
current "solution". If there would be more value labels and fields I
would do that with a python script.

Let's specify my question:
I am looking for a SPSS solution. Is it possible with SPSS?

=====================
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
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Set labels based on table

Jon Peck
In reply to this post by Christian Buhtz
Using the extension command, STATS VALLBLS FROMDATA, you can do this.  It would require you to match the labels dataset with a BY so as to align the label values with the main dataset values.  Excerpt from the help below.

Another way to do this would be to use the extendedTransforms.vlookup function with the SPSSINC TRANS extension command.  In that usage, you create a dataset containing the city index and the label (same as in the first approach) and apply that with TRANS.  I can spell that out further if you want to go that route.

From the STAT VALLBLS FROMDATA extension...

Example:
STATS VALLBLS FROMDATA VARIABLES = x1 TO x5 LBLVARS = label.

This command creates value labels for a set of variables using values of other variables for the labels. If x is a variable having values 1,2,3 and xlabel is a variable having values 'a''b''c', value labels for x are created as

1 'a'
2 'b'
3 'c'

On Mon, Sep 19, 2016 at 1:23 AM, Christian Buhtz <[hidden email]> wrote:
My observations has a field called "city". It is coded with ~40 numbers.
.e.g NewYork is 0, Berlin is 1, ...

Of course I want to set the lables for each value but I don't want to
do this manually. I have a another table (as a csv-file) holding the
values for the code numbers.

CODE;CITY
0;NewYork
1;Berlin
...

Is there a way in SPSS to use that csv-file to set the labels (not the
values!) automaticly?

An acceptable workaround would be when there is a way to change the
values from numbers to the city names.

=====================
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
[hidden email]

===================== 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
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Set labels based on table

Rich Ulrich
In reply to this post by Christian Buhtz
John's solution is to use a text editor in order to write a VALUE LABELS command,
which you then insert into your file.  In an old-style "hack", you could use SPSS
as the text editor to read and re-write the csv-file so that everything would be
in SPSS.

Jon's solution was terse in saying that you MATCH FILES between the data
and the csv-file as the first step, using TABLE=     for the csv-file  to add the
name as an additional variable.
Before (or without) employing his further solution, that already gives you a
variable that has the city names, and thus provides what you describe as
"an acceptable workaround" -- names to use instead of the values.

--
Rich Ulrich


> Date: Mon, 19 Sep 2016 09:23:35 +0200

> From: [hidden email]
> Subject: Set labels based on table
> To: [hidden email]
>
> My observations has a field called "city". It is coded with ~40 numbers.
> .e.g NewYork is 0, Berlin is 1, ...
>
> Of course I want to set the lables for each value but I don't want to
> do this manually. I have a another table (as a csv-file) holding the
> values for the code numbers.
>
> CODE;CITY
> 0;NewYork
> 1;Berlin
> ...
>
> Is there a way in SPSS to use that csv-file to set the labels (not the
> values!) automaticly?
>
> An acceptable workaround would be when there is a way to change the
> values from numbers to the city names.
>

===================== 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
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Set labels based on table

Jon Peck
It's true that just substituting the matched string variable for the numeric code would be a solution, but the value labels approach has some advantages.

The value label representation is smaller, although this is unlikely to matter unless the dataset is very large.

A few procedures, particularly CTABLES and GGRAPH and VALIDATE DATA can use the value labels for more than just labeling.  Also, the gui takes advantage of this.

Once the labels dataset is created, it's a single command to make the labels, so the work is about the same either way, but there may not be a compelling reason to choose between them.

On Monday, September 19, 2016, Rich Ulrich <[hidden email]> wrote:
John's solution is to use a text editor in order to write a VALUE LABELS command,
which you then insert into your file.  In an old-style "hack", you could use SPSS
as the text editor to read and re-write the csv-file so that everything would be
in SPSS.

Jon's solution was terse in saying that you MATCH FILES between the data
and the csv-file as the first step, using TABLE=     for the csv-file  to add the
name as an additional variable.
Before (or without) employing his further solution, that already gives you a
variable that has the city names, and thus provides what you describe as
"an acceptable workaround" -- names to use instead of the values.

--
Rich Ulrich


> Date: Mon, 19 Sep 2016 09:23:35 +0200

> From: <a href="javascript:_e(%7B%7D,&#39;cvml&#39;,&#39;c.buhtz@posteo.jp&#39;);" target="_blank">c.buhtz@...
> Subject: Set labels based on table
> To: <a href="javascript:_e(%7B%7D,&#39;cvml&#39;,&#39;SPSSX-L@LISTSERV.UGA.EDU&#39;);" target="_blank">SPSSX-L@...
>
> My observations has a field called "city". It is coded with ~40 numbers.
> .e.g NewYork is 0, Berlin is 1, ...
>
> Of course I want to set the lables for each value but I don't want to
> do this manually. I have a another table (as a csv-file) holding the
> values for the code numbers.
>
> CODE;CITY
> 0;NewYork
> 1;Berlin
> ...
>
> Is there a way in SPSS to use that csv-file to set the labels (not the
> values!) automaticly?
>
> An acceptable workaround would be when there is a way to change the
> values from numbers to the city names.
>

===================== To manage your subscription to SPSSX-L, send a message to <a href="javascript:_e(%7B%7D,&#39;cvml&#39;,&#39;LISTSERV@LISTSERV.UGA.EDU&#39;);" target="_blank">LISTSERV@... (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
[hidden email]


===================== 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
Loading...