Value labels for missing values

[Carina Ortseifen]

Dear SPSS user,

 

does anybody knows how I can define a value label for a missing value in a numeric variable?

 

I'm using SPSS 16 for windows (german) and have a single variable number with type numeric which has values 1, 2, 3 and . (missing).

I want to define a value label, first interactively then with the command value label, but without success.

 

The only workaround I've seen is to recode, but I'm looking for a way more directly.

 

Any hints from you are welcome.

 

Yours sincerely
Carina
--
Dr. Carina Ortseifen
Abteilung Anwendungen

Tel. [phone], Fax [phone]

[hidden email]


Ruprecht-Karls-Universität Heidelberg

Universitätsrechenzentrum (URZ)

Im Neuenheimer Feld 293, 69120 Heidelberg

http://www.urz.uni-heidelberg.de <http://www.urz.uni-heidelberg.de>

 

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

[Spousta Jan]

Hi Carina,

Recode the system missing value into a number (eg. 9, it must not be used as a regular valid code) and declare it as missing. Then you can label it.

RECODE number (SYSMIS=9).
EXECUTE.
MISSING VALUES number (9).
ADD VALUE LABELS number 9 "A missing value".

fre number.

Have a nice weekend

Jan


-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Ortseifen, Carina
Sent: Friday, October 24, 2008 9:22 AM
To: [hidden email]
Subject: Value labels for missing values

Dear SPSS user,



does anybody knows how I can define a value label for a missing value in a numeric variable?



I'm using SPSS 16 for windows (german) and have a single variable number with type numeric which has values 1, 2, 3 and . (missing).

I want to define a value label, first interactively then with the command value label, but without success.



The only workaround I've seen is to recode, but I'm looking for a way more directly.



Any hints from you are welcome.



Yours sincerely
Carina
--
Dr. Carina Ortseifen
Abteilung Anwendungen

Tel. [phone], Fax [phone]

[hidden email]


Ruprecht-Karls-Universität Heidelberg

Universitätsrechenzentrum (URZ)

Im Neuenheimer Feld 293, 69120 Heidelberg

http://www.urz.uni-heidelberg.de <http://www.urz.uni-heidelberg.de>



=======
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



_____________
Tato zpráva a všechny připojené soubory jsou důvěrné a určené výlučně adresátovi(-ům). Jestliže nejste oprávněným adresátem, je zakázáno jakékoliv zveřejňování, zprostředkování nebo jiné použití těchto informací. Jestliže jste tento mail dostali neoprávněně, prosím, uvědomte odesilatele a smažte zprávu i přiložené soubory. Odesilatel nezodpovídá za jakékoliv chyby nebo opomenutí způsobené tímto přenosem.

Jste si jisti, že opravdu potřebujete vytisknout tuto zprávu a/nebo její přílohy? Myslete na přírodu.


This message and any attached files are confidential and intended solely for the addressee(s). Any publication, transmission or other use of the information by a person or entity other than the intended addressee is prohibited. If you receive this in error please contact the sender and delete the message as well as all attached documents. The sender does not accept liability for any errors or omissions as a result of the transmission.

Are you sure that you really need a print version of this message and/or its attachments? Think about nature.

-.- --

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

[ViAnn Beadle]

I'm curious as to why you would want to do this.

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Ortseifen, Carina
Sent: Friday, October 24, 2008 1:22 AM
To: [hidden email]
Subject: Value labels for missing values

Dear SPSS user,



does anybody knows how I can define a value label for a missing value in a
numeric variable?



I'm using SPSS 16 for windows (german) and have a single variable number
with type numeric which has values 1, 2, 3 and . (missing).

I want to define a value label, first interactively then with the command
value label, but without success.



The only workaround I've seen is to recode, but I'm looking for a way more
directly.



Any hints from you are welcome.



Yours sincerely
Carina
--
Dr. Carina Ortseifen
Abteilung Anwendungen

Tel. [phone], Fax [phone]

[hidden email]


Ruprecht-Karls-Universität Heidelberg

Universitätsrechenzentrum (URZ)

Im Neuenheimer Feld 293, 69120 Heidelberg

http://www.urz.uni-heidelberg.de <http://www.urz.uni-heidelberg.de>



=======
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

[Art Kendall]

labeling your different kinds of missing values is a good idea.  Very
often good data preparation will change system missing to a value that
says why it is missing.  Knowing why a value is missing  is often
critical to analysis and interpretation.

A system-missing value should be reserved for the situation where the
program cannot do what you are telling it to do.
System-missing values occur when the program is unable to read input the
way you tell it to read it or it cannot do a transformation that you
tell it to do.

User-missing values say that there is a known reason why there is not a
valid value there.


If the data is original raw data, the existence of a system-missing
value is a cue to go back and revisit how you are reading the data.
If nothing else can be done you can recode the system missing values
into a known code and give it a label of "input data not readable as
described".
If you have SPSS to treat blanks as system missing rather than a user
missing code. you would recode it and label that value "blank on input".
or re-read the data with a different default.

If the system-missing value results from a transformation, it is
advisable to "trap" the condition in your syntax and have it result in a
user-missing value, then run the revised syntax.

Altogether, for example on a variable measuring parity you might have
value labels parity
 0 'no births'
 1 'birth'
 -1 'refused to answer'
 -2'interviewer form misprinted not asked'
 -3 'not applicable -- pre-pubescent female'
 -4 'not applicable -- male'
 -5 'surrogate responder did not know'
 -6 'input data blank on input'
 -7 'input data not readable as described'.

Knowing why there is not not valid is needed to decide whether is
potentially a relation between what the "true" answer would be were it
known, or whether values should be considered as Missing At Random
(MAR), Missing Completely At Random (MCAR), etc.

to make a long story short
try something like
recode type (sysmis = -1)(else=copy) into newtype.
value labels newtype 1 'something' 2 'something else' 3'another type' -1
'whatever the reason is'.

Art Kendall
Social Research Consultants



Ortseifen, Carina wrote:

> Dear SPSS user,
>
>
>
> does anybody knows how I can define a value label for a missing value in a numeric variable?
>
>
>
> I'm using SPSS 16 for windows (german) and have a single variable number with type numeric which has values 1, 2, 3 and . (missing).
>
> I want to define a value label, first interactively then with the command value label, but without success.
>
>
>
> The only workaround I've seen is to recode, but I'm looking for a way more directly.
>
>
>
> Any hints from you are welcome.
>
>
>
> Yours sincerely
> Carina
> --
> Dr. Carina Ortseifen
> Abteilung Anwendungen
>
> Tel. [phone], Fax [phone]
>
> [hidden email]
>
>
> Ruprecht-Karls-Universität Heidelberg
>
> Universitätsrechenzentrum (URZ)
>
> Im Neuenheimer Feld 293, 69120 Heidelberg
>
> http://www.urz.uni-heidelberg.de <http://www.urz.uni-heidelberg.de>
>
>
>
> ===================
> 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

[Peck, Jon]

There is a sneaky way to place a value label on the system missing value, but it's probably a bug, so I'm not going to describe it.  sysmis is treated specially in many places, and the system isn't expecting it to have a value label, so behavior is erratic.

Value labels on user missing values are fine and a good idea, though.

Regards,
Jon Peck

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Ortseifen, Carina
Sent: Friday, October 24, 2008 1:22 AM
To: [hidden email]
Subject: [SPSSX-L] Value labels for missing values

Dear SPSS user,



does anybody knows how I can define a value label for a missing value in a numeric variable?



I'm using SPSS 16 for windows (german) and have a single variable number with type numeric which has values 1, 2, 3 and . (missing).

I want to define a value label, first interactively then with the command value label, but without success.



The only workaround I've seen is to recode, but I'm looking for a way more directly.



Any hints from you are welcome.



Yours sincerely
Carina
--
Dr. Carina Ortseifen
Abteilung Anwendungen

Tel. [phone], Fax [phone]

[hidden email]


Ruprecht-Karls-Universität Heidelberg

Universitätsrechenzentrum (URZ)

Im Neuenheimer Feld 293, 69120 Heidelberg

http://www.urz.uni-heidelberg.de <http://www.urz.uni-heidelberg.de>



=======
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

[Carina Ortseifen]

Thanks a lot for all of your helpful answers.
Finally all will say the same: Use user-defined missing values and then define value labels.

O.k. that's it. Thank you again and have a nice new week.

Mit freundlichen Grüßen
Carina
--
Dr. Carina Ortseifen
Abteilung Anwendungen

Tel. [phone], Fax [phone]
[hidden email]

Ruprecht-Karls-Universität Heidelberg
Universitätsrechenzentrum (URZ)
Im Neuenheimer Feld 293, 69120 Heidelberg
http://www.urz.uni-heidelberg.de

> -----Ursprüngliche Nachricht-----
> Von: Art Kendall [mailto:[hidden email]]
> Gesendet: Freitag, 24. Oktober 2008 18:49
> An: Ortseifen, Carina
> Cc: [hidden email]
> Betreff: Re: Value labels for missing values
>
> labeling your different kinds of missing values is a good idea.  Very
> often good data preparation will change system missing to a value that
> says why it is missing.  Knowing why a value is missing  is often
> critical to analysis and interpretation.
>
> A system-missing value should be reserved for the situation where the
> program cannot do what you are telling it to do.
> System-missing values occur when the program is unable to read input the
> way you tell it to read it or it cannot do a transformation that you
> tell it to do.
>
> User-missing values say that there is a known reason why there is not a
> valid value there.
>
>
> If the data is original raw data, the existence of a system-missing
> value is a cue to go back and revisit how you are reading the data.
> If nothing else can be done you can recode the system missing values
> into a known code and give it a label of "input data not readable as
> described".
> If you have SPSS to treat blanks as system missing rather than a user
> missing code. you would recode it and label that value "blank on input".
> or re-read the data with a different default.
>
> If the system-missing value results from a transformation, it is
> advisable to "trap" the condition in your syntax and have it result in a
> user-missing value, then run the revised syntax.
>
> Altogether, for example on a variable measuring parity you might have
> value labels parity
>  0 'no births'
>  1 'birth'
>  -1 'refused to answer'
>  -2'interviewer form misprinted not asked'
>  -3 'not applicable -- pre-pubescent female'
>  -4 'not applicable -- male'
>  -5 'surrogate responder did not know'
>  -6 'input data blank on input'
>  -7 'input data not readable as described'.
>
> Knowing why there is not not valid is needed to decide whether is
> potentially a relation between what the "true" answer would be were it
> known, or whether values should be considered as Missing At Random
> (MAR), Missing Completely At Random (MCAR), etc.
>
> to make a long story short
> try something like
> recode type (sysmis = -1)(else=copy) into newtype.
> value labels newtype 1 'something' 2 'something else' 3'another type' -1
> 'whatever the reason is'.
>
> Art Kendall
> Social Research Consultants
>
>
>
> Ortseifen, Carina wrote:
> > Dear SPSS user,
> >
> >
> >
> > does anybody knows how I can define a value label for a missing value in a
> numeric variable?
> >
> >
> >
> > I'm using SPSS 16 for windows (german) and have a single variable number with
> type numeric which has values 1, 2, 3 and . (missing).
> >
> > I want to define a value label, first interactively then with the command value
> label, but without success.
> >
> >
> >
> > The only workaround I've seen is to recode, but I'm looking for a way more
> directly.
> >
> >
> >
> > Any hints from you are welcome.
> >
> >
> >
> > Yours sincerely
> > Carina
> > --
> > Dr. Carina Ortseifen
> > Abteilung Anwendungen
> >
> > Tel. [phone], Fax [phone]
> >
> > [hidden email]
> >
> >
> > Ruprecht-Karls-Universität Heidelberg
> >
> > Universitätsrechenzentrum (URZ)
> >
> > Im Neuenheimer Feld 293, 69120 Heidelberg
> >
> > http://www.urz.uni-heidelberg.de <http://www.urz.uni-heidelberg.de>
> >
> >
> >
> > ===================
> > 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