Restructured data, variable names

[Josh]

Hi, I'm having trouble working with a restructured data set.

I've restructured my data from long to short (or wide) format. The
restructure command seems to have worked well.  There were 120 timepoints
in the long-form data. As expected, it renamed the variable names
according to the time points. So variable 'x' in the long-form dataset
become variables 'x.0, x.1, x.2, x.3' and on to 'x.120' in the short-form
dataset. And the same happens to variables y, z, etc.

The problem is that I'm having trouble working with these new variable
names. In particular, I want to be able to compute the maximum value of
variables x.0 to x.120. However, the command

compute xMax = Max(x.0 to x.120)

returns the error

'>A function argument list contains an invalid use of the TO convention.
>Execution of this command stops.'

This makes sense because 'x1 to x120' is not the same thing as 'x.1 to
x.120'. Using syntax, however, I can't find an easy way around this.

I could simply list out all of the variables (eg, Max(x.1, x.2, x.3...)),
but I'm restructuring many variables, I have many additional variables to
compute in the restructured dataset, and I want to make the syntax easy
and flexible for other people to use.

Is there any other way around this? Can you control how restructured
variables are named? Would it be easiest to add in a rename command, and
if so what might it look like? Or am I stuck writing out 120 variables
every time that I want to do one of these calculations?

Thanks for your help

Josh

=====================
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]

Take a look at the /SEPARATOR sub-command for CASESTOVARS.  You can change the separator from the default "." to whatever you want, including a null string ("").

Josh wrote

Hi, I'm having trouble working with a restructured data set.

I've restructured my data from long to short (or wide) format. The
restructure command seems to have worked well.  There were 120 timepoints
in the long-form data. As expected, it renamed the variable names
according to the time points. So variable 'x' in the long-form dataset
become variables 'x.0, x.1, x.2, x.3' and on to 'x.120' in the short-form
dataset. And the same happens to variables y, z, etc.

The problem is that I'm having trouble working with these new variable
names. In particular, I want to be able to compute the maximum value of
variables x.0 to x.120. However, the command

compute xMax = Max(x.0 to x.120)

returns the error

'>A function argument list contains an invalid use of the TO convention.
>Execution of this command stops.'

This makes sense because 'x1 to x120' is not the same thing as 'x.1 to
x.120'. Using syntax, however, I can't find an easy way around this.

I could simply list out all of the variables (eg, Max(x.1, x.2, x.3...)),
but I'm restructuring many variables, I have many additional variables to
compute in the restructured dataset, and I want to make the syntax easy
and flexible for other people to use.

Is there any other way around this? Can you control how restructured
variables are named? Would it be easiest to add in a rename command, and
if so what might it look like? Or am I stuck writing out 120 variables
every time that I want to do one of these calculations?

Thanks for your help

Josh

=====================
To manage your subscription to SPSSX-L, send a message to
[e-mail] (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]

the TO convention works anchored at zero.
try this.

data list list /x.0 to x.2.
begin data
1 2 3
3 2 1
2 3 1
end data.
frequencies variables = x.0 to x.2.

look at the variables view.
You say you should have 120 time points, but x.0 to x.120 is 121 variables.

Also, are you sure the variables are in contiguous positions?


Art Kendall
Social Research Consultants



On 1/24/2011 2:23 PM, Josh wrote:

> Hi, I'm having trouble working with a restructured data set.
>
> I've restructured my data from long to short (or wide) format. The
> restructure command seems to have worked well.  There were 120 timepoints
> in the long-form data. As expected, it renamed the variable names
> according to the time points. So variable 'x' in the long-form dataset
> become variables 'x.0, x.1, x.2, x.3' and on to 'x.120' in the short-form
> dataset. And the same happens to variables y, z, etc.
>
> The problem is that I'm having trouble working with these new variable
> names. In particular, I want to be able to compute the maximum value of
> variables x.0 to x.120. However, the command
>
> compute xMax = Max(x.0 to x.120)
>
> returns the error
>
> '>A function argument list contains an invalid use of the TO convention.
>> Execution of this command stops.'
> This makes sense because 'x1 to x120' is not the same thing as 'x.1 to
> x.120'. Using syntax, however, I can't find an easy way around this.
>
> I could simply list out all of the variables (eg, Max(x.1, x.2, x.3...)),
> but I'm restructuring many variables, I have many additional variables to
> compute in the restructured dataset, and I want to make the syntax easy
> and flexible for other people to use.
>
> Is there any other way around this? Can you control how restructured
> variables are named? Would it be easiest to add in a rename command, and
> if so what might it look like? Or am I stuck writing out 120 variables
> every time that I want to do one of these calculations?
>
> Thanks for your help
>
> Josh
>
> =====================
> 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

[Josh-2]

Perfect, that 'separator' subcommand was exactly what I was looking for.
Thanks!

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