Concatenate Numerals from three columns

[Manish Madan]

Hi - I tried finding information on the command "Concatenate" and it
is easy to do it when we have string variable. However, let us say I
have 3 columns as follows:

C1 C2 C3
1 4 5

Now, I want to make a new variable that reads '145'. I did as follows:

STRING V501_new (a9).
COMPUTE V501_new = CONCAT(qq01_new, qq02_new, qq04_new).

>Error # 4311 in column 55. Text: )
>The string argument required for the function specified was not supplied.
>Execution of this command stops.
EXECUTE.

any help with correct syntax, please? thanks in advance!

[Bruce Weaver]

On 12/08/2011 5:55 PM, Manish Madan wrote:
> Hi - I tried finding information on the command "Concatenate" and it
> is easy to do it when we have string variable. However, let us say I
> have 3 columns as follows:
>
> C1 C2 C3
> 1 4 5
>
> Now, I want to make a new variable that reads '145'. I did as follows:

Are those variables string or numeric? If numeric, and you want a
numeric result, do not use CONCAT. Just do this:

compute newvar = c1*100 + c2*10 + c3.

If that is not what you want, please provide more detail--i.e., what
types of variables are C1-C3, and what type of variable do you want the
result to be.

>
> STRING V501_new (a9).
> COMPUTE V501_new = CONCAT(qq01_new, qq02_new, qq04_new).
>
>> Error # 4311 in column 55. Text: )
>> The string argument required for the function specified was not supplied.
>> Execution of this command stops.
> EXECUTE.
>
> any help with correct syntax, please? thanks in advance!

--
Bruce Weaver
bwe...@lakeheadu.ca
http://sites.google.com/a/lakeheadu.ca/bweaver/Home
"When all else fails, RTFM."

[MM]

> bwea...@lakeheadu.cahttp://sites.google.com/a/lakeheadu.ca/bweaver/Home

> "When all else fails, RTFM."

Thanks Sir for your help! my variables are numeric. So, when I used
your newvar command, it does that but not exactly what I wanted.

For instance, if I have 1 in C1, missing in C2, C3. I'd hope, newvar
should give me 1 but its gave me 0. Anyhow, I found the fix for that

compute newvar=sum(c1*100,c2*10,c3) instead of using + sign.

thanks again for prompt help!

btw, in the meanwhile I was using another way of doing that (little
more complicated!)

VECTOR v=v1 TO v3.
COMPUTE V3temp=0.
LOOP #cnt=1 TO 3.
IF ~MISSING(v(#cnt))V3temp=V3temp*10 + v(#cnt).
END LOOP.

apparently this also did exactly as concat would do but obviously your
idea was simpler

[Rich Ulrich]

On Sun, 14 Aug 2011 19:33:50 -0700 (PDT), MM <manish...@gmail.com>
wrote:

>On Aug 12, 11:11�pm, Bruce Weaver <bwea...@lakeheadu.ca> wrote:

>> On 12/08/2011 5:55 PM, Manish Madan wrote:

>[snip, previous]

>Thanks Sir for your help! my variables are numeric. So, when I used
>your newvar command, it does that but not exactly what I wanted.
>
>For instance, if I have 1 in C1, missing in C2, C3. I'd hope, newvar
>should give me 1 but its gave me 0. Anyhow, I found the fix for that

You mean, "should give me 100"

>compute newvar=sum(c1*100,c2*10,c3) instead of using + sign.

I though you would need sum.1( ... ) to avoid the Missing problem;
but if it works, okay.

>
>thanks again for prompt help!
>
>btw, in the meanwhile I was using another way of doing that (little
>more complicated!)
>
>VECTOR v=v1 TO v3.
>COMPUTE V3temp=0.
>LOOP #cnt=1 TO 3.
>IF ~MISSING(v(#cnt))V3temp=V3temp*10 + v(#cnt).
>END LOOP.
>
>apparently this also did exactly as concat would do but obviously your
>idea was simpler

That would also yield "1" instead of "100" when the latter two are
missing. Also be careful and re-check the handling of Missing.

--
Rich Ulrich

[Rakesh Kuamr Maurya]

this is the right your function. but

i want to
Concatenating two or more string variables and numeric variables.

[Ruben Geert van den Berg]

Hi Rakesh,

Then use CONCAT and convert the numbers into strings with the STRING function.

Also see

http://www.spss-tutorials.com/concatenate/
http://www.spss-tutorials.com/changing-variable-properties-3-type/#stringfunction