syntax to calculate means across different variables

[Shanti Ganesh]

Hi,

I would like to compute a new variable which is the mean of different
variables.

the new variable is called RespSelf1.

The mean has to be calculated from these groups of variables:

response.1 to response.5
response.51 to response.55
response.101 to response.105
response.151 to response.155
response.201 to response.205

How to write this down in the syntax?

I have tried:

COMPUTE RespSelf1 = mean((response.1 to response.5), (response.51 to 55) ).

and

COMPUTE RespSelf1 = mean(response.1 to response.5, response.51 to 55 ).

Neither of them worked.

Does anyone know how to do this?

Thanks a lot!

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

[Bruce Weaver]

Shanti Ganesh wrote

Hi,

I would like to compute a new variable which is the mean of different
variables.

the new variable is called RespSelf1.

The mean has to be calculated from these groups of variables:

response.1 to response.5
response.51 to response.55
response.101 to response.105
response.151 to response.155
response.201 to response.205

How to write this down in the syntax?

I have tried:

COMPUTE RespSelf1 = mean((response.1 to response.5), (response.51 to 55) ).

and

COMPUTE RespSelf1 = mean(response.1 to response.5, response.51 to 55 ).

Neither of them worked.

Does anyone know how to do this?

Thanks a lot!

This is not terribly elegant, but will work, I think.

compute #sum = sum(response.1 to response.5) +
               sum(response.51 to response.55) +
               sum(response.101 to response.105) +
               sum(response.151 to response.155) +
               sum(response.201 to response.205) .
compute #n =   nvalid(response.1 to response.5) +
               nvalid(response.51 to response.55) +
               nvalid(response.101 to response.105) +
               nvalid(response.151 to response.155) +
               nvalid(response.201 to response.205) .
compute RespSelf1 = #sum/#n.
exe.

[Marta Garcia-Granero]

Shanti Ganesh wrote:

This should work:

COMPUTE RespSelf1 = mean(response.1 to response.5,
                         response.51 to response.55,
                         response.101 to response.105,
                         response.151 to response.155,
                         response.201 to response.205).

Notice that you forgot to add "response." when you wrote "response.51 to
55" just before "55".

HTH,
Marta GG

> I would like to compute a new variable which is the mean of different
> variables.
>
> the new variable is called RespSelf1.
>
> The mean has to be calculated from these groups of variables:
>
> response.1 to response.5
> response.51 to response.55
> response.101 to response.105
> response.151 to response.155
> response.201 to response.205
>
> How to write this down in the syntax?
>
> I have tried:
>
> COMPUTE RespSelf1 = mean((response.1 to response.5), (response.51 to 55) ).
>
> and
>
> COMPUTE RespSelf1 = mean(response.1 to response.5, response.51 to 55 ).
>
> Neither of them worked.
>
> Does anyone know how to do this?
>
> Thanks a lot!
>
> =====================
> 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
>
>

--
For miscellaneous SPSS related statistical stuff, visit:
http://gjyp.nl/marta/

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

[Shanti Ganesh]

>Notice that you forgot to add "response." when you wrote "response.51 to
>55" just before "55".
>
>HTH,
>Marta GG
>

Thanks Marta for pointing out the mistake! *multi-embarassed* Glad it works
this way.

Thank you also Bruce and Andrew for the sum and nvalid tips. It works too!

best,
Shanti

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