SPSSX Discussion

appending cases/merging

Classic

List

Threaded

6 messages Options

Deepa Bhat

appending cases/merging

Hi everyone,

When you are appending cases from one dataset to another dataset, how
much does the "width" and "decimals" matter?
I made sure the variable name and the type (string and numeric) are the
same, but do I need to make sure the width and decimals are the same in
both datasets?

Thanks,
Deepa

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

Richard Ristow

Re: appending cases/merging

At 11:17 AM 4/25/2008, Deepa Bhat wrote:

>When you are appending cases from one dataset to another dataset,
>how much does the "width" and "decimals" matter?

For numeric variables, they don't matter at all; the values are
copied just fine. "If a numeric variable has the same name but
different formats (for example, F8.0 and F8.2)
in different input files, the format of the variable in the
first-named file is used." - Command Syntax Reference.

>I made sure the variable name and the type (string and numeric) are the
>same

Alas, for *strings*, the lengths must be the same in both files. I've
argued that this restriction is a severe mistake on SPSS's part, but
there it is.

-Best wishes,
Richard

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

Albert-Jan Roskam

Re: appending cases/merging

In reply to this post by Deepa Bhat

Hi Deepa,

I think the two files need to be 'union compatible'
(in sql speak), meaning that the width and decimals
should be equal (also variable type) between 'variable
pairs'. But it doesn't matter if the print or display
FORMAT differs between files.
Use DISPLAY DICTIONARY to compare the characteristics
of the two files.

Cheers!!
Albert-Jan

--- Deepa Bhat <[hidden email]> wrote:

> Hi everyone,
>
> When you are appending cases from one dataset to
> another dataset, how
> much does the "width" and "decimals" matter?
> I made sure the variable name and the type (string
> and numeric) are the
> same, but do I need to make sure the width and
> decimals are the same in
> both datasets?
>
> Thanks,
> Deepa
>
> =====================
> 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

Re: appending cases/merging

For numeric variables, formats don't have to match. All numbers are stored
as floating-point binary and formats are only used for display purposes.
When numeric formats conflict, the format used for variables in the first
file is used.

For string variables, formats do have to match in type and width (to the
annoyance of some users).

And variables do have to match in type (numeric or string).
-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Albert-jan Roskam
Sent: Saturday, April 26, 2008 5:10 AM
To: [hidden email]
Subject: Re: appending cases/merging

Hi Deepa,

I think the two files need to be 'union compatible'
(in sql speak), meaning that the width and decimals
should be equal (also variable type) between 'variable
pairs'. But it doesn't matter if the print or display
FORMAT differs between files.
Use DISPLAY DICTIONARY to compare the characteristics
of the two files.

Cheers!!
Albert-Jan

--- Deepa Bhat <[hidden email]> wrote:

____________________________________________________________________________
________
Be a better friend, newshound, and
know-it-all with Yahoo! Mobile. Try it now.
http://mobile.yahoo.com/;_ylt=Ahu06i62sR8HDtDypao8Wcj9tAcJ

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

Albert-Jan Roskam

Re: appending cases/merging

Really? I didn't know that, thanks! I always thought
that if a variable is declared as eg. NUMERIC (9.0),
it is stored as more bits (i.e. it occupies more
memory, because of the greater precision) than eg.
NUMERIC (2.0). So what's the difference between
FORMATS and NUMERIC if both only deal with display
purposes?

Cheers!
Albert-Jan

--- ViAnn Beadle <[hidden email]> wrote:

> For numeric variables, formats don't have to match.
> All numbers are stored
> as floating-point binary and formats are only used
> for display purposes.
> When numeric formats conflict, the format used for
> variables in the first
> file is used.
>
> For string variables, formats do have to match in
> type and width (to the
> annoyance of some users).
>
> And variables do have to match in type (numeric or
> string).
> -----Original Message-----
> From: SPSSX(r) Discussion
> [mailto:[hidden email]] On Behalf Of
> Albert-jan Roskam
> Sent: Saturday, April 26, 2008 5:10 AM
> To: [hidden email]
> Subject: Re: appending cases/merging
>
> Hi Deepa,
>
> I think the two files need to be 'union compatible'
> (in sql speak), meaning that the width and decimals
> should be equal (also variable type) between
> 'variable
> pairs'. But it doesn't matter if the print or
> display
> FORMAT differs between files.
> Use DISPLAY DICTIONARY to compare the
> characteristics
> of the two files.
>
> Cheers!!
> Albert-Jan
>
>
> --- Deepa Bhat <[hidden email]> wrote:
>
> > Hi everyone,
> >
> > When you are appending cases from one dataset to
> > another dataset, how
> > much does the "width" and "decimals" matter?
> > I made sure the variable name and the type (string
> > and numeric) are the
> > same, but do I need to make sure the width and
> > decimals are the same in
> > both datasets?
> >
> > Thanks,
> > Deepa
> >
> > =====================
> > 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
> >
>
>
>
>
>

____________________________________________________________________________
> ________
> Be a better friend, newshound, and
> know-it-all with Yahoo! Mobile. Try it now.
>
http://mobile.yahoo.com/;_ylt=Ahu06i62sR8HDtDypao8Wcj9tAcJ

>
> =====================
> 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

Re: appending cases/merging

Numbers in SPSS are always double precision floating point, which means 53 bits of precision, regardless of the format, which, as ViAnn said, is just a display setting. Numeric declares a variable and is rarely necessary, because a new variable is numeric by default. Format can change a variables format within its fundamental string or numeric type and is applied to an existing variable.

ALTER TYPE, new in SPSS 16, can change the string size or the fundamental type and does some other tricks as well.

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Albert-jan Roskam
Sent: Saturday, April 26, 2008 11:34 AM
To: [hidden email]
Subject: Re: [SPSSX-L] appending cases/merging

Really? I didn't know that, thanks! I always thought
that if a variable is declared as eg. NUMERIC (9.0),
it is stored as more bits (i.e. it occupies more
memory, because of the greater precision) than eg.
NUMERIC (2.0). So what's the difference between
FORMATS and NUMERIC if both only deal with display
purposes?

Cheers!
Albert-Jan

--- ViAnn Beadle <[hidden email]> wrote:

____________________________________________________________________________________
Be a better friend, newshound, and
know-it-all with Yahoo! Mobile. Try it now. http://mobile.yahoo.com/;_ylt=Ahu06i62sR8HDtDypao8Wcj9tAcJ

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